Master Class

It is fairly well appreciated that more than 1.7 million individuals living in the United States are now infected with HIV. What may be less appreciated by patients and physicians is that the impact of the epidemic on women has grown significantly over time.

In 2008, using a new national surveillance system, the Centers for Disease Control and Prevention reported that women comprised more than one-fourth (27%) of the 56,300 people estimated to have been newly infected with HIV in 2006. That is, an estimated 15,000 women were newly infected with HIV in 2006 alone (JAMA 2008;300:520-9).

As ob.gyns., we care for women of all ages, from adolescents through women in their senior years. Therefore, we are in the unique position to be able to identify HIV-infected individuals who could benefit significantly from early monitoring and treatment and to decrease the risk of transmission on a broader level. We can protect our patients, and they in turn can protect their partners and future children. This is a tremendous opportunity that we should not miss.

In 2006, the CDC moved away from HIV testing recommendations that were risk based and advised routine screening for HIV infection for all patients aged 13–64 years. It recommended that physicians notify their patients that testing will be performed unless they decline.

The American College of Obstetricians and Gynecologists weighed in the next year, saying that universal testing with patient notification is more effective in identifying infected patients than is targeted risk-based testing, largely because many women found to be infected with HIV did not consider themselves to have been at risk. Opt-out testing is less cumbersome, ACOG pointed out, because it removes the requirement for lengthy pretest counseling and for detailed, testing-related informed consent.

A 2007 survey of almost 400 medical staff at San Francisco General Hospital, which serves a population with a high HIV prevalence, showed that clinicians in obstetrics/gynecology and HIV infectious disease specialists were significantly more likely to routinely test their patients than were physicians in other specialties (J. Acquir. Immune Defic. Syndr. 2009;50:114-6).

As the authors say, however, this isn't surprising given the fact that ACOG, the CDC, and other national organizations have long called for universal prenatal HIV testing.

Other studies have suggested that compliance with the CDC recommendations is low – that many ob.gyns. as well as other primary care physicians – do not routinely offer HIV testing outside of the prenatal context. Even in high prevalence areas such as New Jersey, we have documented that many providers miss the opportunity to offer the test to all their patients, deeming them unlikely to be infected. This approach hinders early access to care for potentially infected patients and fails to address an unknowing risk of transmission to partners.

We need to think differently.

We do not ask patients: Do you want a chlamydia test? A Pap smear? A cholesterol test? We tell them, instead, that we're going to check their cholesterol levels, or that it's time for their Pap smear, or that we'd recommend a chlamydia test. We need to present HIV testing in the same way – as part of the routine battery of tests that will be performed unless the patient declines. Studies have shown, interestingly, that more patients accept recommended HIV testing when they know it's routinely offered to everyone, without an assessment of risk.

Furthermore, we need to do so in spite of race and in patients of all ages, with all the generations of women whom we see.

As ob.gyns., we must be cognizant of the changing face of HIV infection in women in the last 30 years and realize the unique challenge that we now face as the infection encompasses patients from a broader age spectrum than just those in their reproductive years. We have witnessed both an increased number of HIV infections in the older population and increased survival among individuals on highly active antiretroviral therapy (HAART), including those who were perinatally infected.

We also must appreciate that women who are engaged in a relationship with a known HIV-infected partner, regardless of their age, can prevent acquisition with barrier methods such as condom use or postexposure prophylaxis as delineated by the CDC for serodiscordant couples.

Women Older Than 50

Studies have consistently demonstrated that many women are sexually active into their 60s. Even when we appreciate this, we tend to dismiss the possibility that any of our older patients might have HIV infection. We tend to assume that our patients are in stable relationships and presume there is no value to HIV testing.

It's important to appreciate, however, that the number of people aged 50 years and older who are living with HIV/AIDS has been increasing in recent years. It is estimated that almost one-fourth of all people with HIV/AIDS in the United States are age 50 years and older. While this is partly because HAART has extended the lives of many HIV-infected people, it is also attributed to newly diagnosed infections in people over 50 years.

The 2008 CDC analysis that showed that women comprised more than one-fourth of the individuals newly infected with HIV in 2006 also found that 25% of the new infections were in individuals (men and women) aged 40–49 years (13,900 out of 56,300). Another 10% were in individuals aged 50 years and over (5,800 of 56,300). Approximately 30% of the infections were a result of heterosexual contact. Earlier data from the 1990s similarly showed over 10% of new AIDS cases occurring in people older than age 50 years.

Women of all ages can wrongly believe they are not at risk of contracting HIV. In one recent survey examining patient attitudes about HIV testing and knowledge about their own risk status, only 2% of approximately 850 women of various ages considered themselves at high risk for HIV infection despite the fact that almost half of them reported having had unprotected sex at some point with more than one partner. The women were patients of ob.gyn. members of ACOG's Collaborative Ambulatory Research Network (Matern. Child Health J. 2009;13:355-63).

Older women are generally even less knowledgeable about HIV transmission and how to protect themselves than are younger women, and they are not concerned about undesired pregnancy. For these reasons, many older women may not be practicing safer sex, increasing their risk for HIV and other sexually transmitted infections.

In a nine-question survey of 514 urban women aged 50 years and older (mean age of 62), the majority of women scored poorly, answering four or fewer of the questions correctly. Eighty-four percent correctly identified unprotected heterosexual sex as a moderate- to high-risk activity, but only 13% identified condoms as being very effective in preventing HIV, and 18% said condoms are not at all effective (J. Amer. Geriatr. Soc. 2004;52:1549-53).

In another study aimed at assessing differences in the characteristics of individuals (both men and women) who refuse testing and those who accept it, investigators found that HIV test refusal was associated with female gender, white race, older age, and higher educational level (AIDS Patient Care STDS 2006;20:84-92).

Older women must be educated about their risk of heterosexual transmission and the fact that the risk for HIV acquisition has been increasing since 1994 in the United States. They need to understand that normal physiologic changes in the menopausal period such as thinning of the vaginal mucosa, increased susceptibility to vaginal abrasions during intercourse, and changes in their immune response can make them more vulnerable to disease acquisition or progression.

It also is important to educate them about the effectiveness of condoms and the importance of knowing the HIV status of their partners, because it is estimated that approximately 21% of infected individuals in the United States do not know their HIV status.

Most of all, our patients should understand that 30 years into the epidemic, we have demonstrated excellent survival in individuals on treatment, particularly among those who were diagnosed early and who are receiving HAART. A recent report from the CDC shows that average life expectancy after HIV diagnosis in the general population increased from 10.5 to 22.5 years from 1996 to 2005 (J. Acquir. Immune Defic. Syndr. 2010;53:124-130).

Interestingly, as the report points out, studies have shown that although HIV-infected women had a greater life expectancy to begin with, they showed a lesser magnitude of improvement than did men, particularly white men. (Women's life expectancy changed from 12.6 to 23.6 years.) This observation highlights the importance of earlier diagnosis and link to care.

Knowing about the successes of HAART is important because women are less likely to opt out of HIV testing when they perceive the benefits. We can explain to patients – especially those who are apprehensive about the test – that the test is integrated into the annual health care panel (along with cholesterol and triglyceride testing, and genital cytology), and that, contrary to decades ago, we can treat and control HIV disease once it is diagnosed, which is more than we can do for certain types of cancer.

Our patients need to understand that it can be a manageable chronic disease as long as it is detected and effectively addressed early in the course of their infection. Today, we have access to a wide array of pamphlets and videos that we can offer in the waiting area to help patients understand this and appreciate the value of HIV testing.

Serodiscordant Couples

Thus far, there is limited standardization or consensus on how and when to provide counseling, testing, and prevention strategies for women who are involved in HIV serodiscordant relationships. However, most experts recommend that patients whose partners are HIV positive should be tested for HIV infection annually and encouraged to use effective prevention strategies such as the male or female condom.

Screening and treatment for sexually transmitted diseases should be done annually as coinfection can increase the risk of HIV transmission. In one of the studies demonstrating an impact of STD treatment – a randomized trial conducted more than 15 years ago in rural Tanzania – improved STD education and treatment reduced HIV incidence by about 40% (Lancet 1995;346:530-6).

In cases in which an unplanned sexual encounter with an HIV-positive partner occurs without protection, postexposure prophylaxis should be considered and given as soon as the event is identified, preferably within 48 hours. The CDC's recommendations for the use of antiretroviral postexposure prophylaxis, issued in 2005, call for a 28-day course of HAART (MMWR 2005;54[RR02]:1-20)

Decisions about the optimal postexposure therapy involve various factors, including the partner's antiretroviral history, adherence to the regimen, and most recent viral load. We may need to counsel patients, however, that having undetectable virus in the blood does not necessarily mean there will not be any virus in the genital tract. Discrepancies between serum and genital viral load have been reported among HIV-infected men and women on HAART.

If a woman engages in unprotected sex with a male of unknown serostatus, she can request postexposure prophylaxis. In this case, she should be counseled about the risks and benefits of postexposure HAART, as she may expose herself to unnecessary toxicities.

When faced with these situations we can obtain guidance from, or work in partnership with, the infectious disease provider who is managing the HIV-infected partner, or we can contact state or national phone lines for linkage to immediate care. Some health departments have established nonoccupational postexposure prophylaxis programs in their jurisdictions. Overall, it is important that we be aware of the availability of postexposure HAART and its possible risks and benefits.

In the near future, a woman whose partner is HIV positive should be able to benefit from antiretroviral microbicides used before or after intercourse. In the double-blind, randomized, and well-publicized CAPRISA (Centre for the AIDS Programme of Research in South Africa) 004 trial, a 1% vaginal gel formulation of tenofovir reduced HIV acquisition by approximately 39% overall and by 54% in women with high adherence to the protocol for gel application (Science 2010;329:1168-74). Another phase III trial of tenofovir is ongoing.

The field of safe reproduction for HIV serodiscordant couples also is advancing, such that women and their partners have various options for conceiving with minimal risk of transmitting the infection.

A large body of evidence suggests that reproductive technology – that is, sperm washing and artificial insemination – can help HIV-affected couples safely conceive, and the results of further CDC-sponsored research aimed at evaluating outcomes in couples who have used these techniques to conceive are expected soon. For many couples, however, such technologies are economically inaccessible.

Experts are looking at periconception preexposure prophylaxis as a potential strategy for preventing HIV transmission in couples trying to conceive. Under this approach, the seronegative partner would take antiretroviral drugs during periods of attempted conception, with the goal of preventing initial viral replication. Clinical trials evaluating its safety and efficacy are ongoing.

Other components of a risk-reduction program should include suppressive antiretroviral therapy for the infected partner (who may yet be eligible for such therapy under current recommendations for CD4 cell count), screening and pretreatment for other sexually transmitted infections, and unprotected sexual intercourse that is limited to times of peak fertility (AIDS 2010;24:1975-82).

Perinatally Infected Women

An increasing number of perinatally infected adolescents has been identified and engaged in care throughout the United States. This population is unique in that many were initially exposed to monotherapy or dual antiretroviral regimens and thus have developed resistance to several antiretroviral regimens. In addition, their adherence to HAART is lower than required to avert failure on current regimens.

They are a population that presents a challenge to ob.gyns. because many have high-risk sexual behaviors, and when they become pregnant, they are at high risk for operative delivery due to inadequate viral suppression.

We recently reported on the pregnancy outcomes of 10 perinatally infected adolescents referred to the high-risk obstetric clinic at our institution and delivered between 1997 and 2007. Neonatal outcomes were generally favorable, but these young women had a high rate of operative delivery (62%, compared with our institutional rate of 33%) due to the failure to achieve undetectable viral load (Am. J. Obstet. Gynecol. 2009;200:149.e1-5).

Our goals for these women should therefore include supporting and counseling on treatment adherence and on the selection of well-tolerated HAART regimens that can rapidly suppress viral load. We should also minimize the use of operative deliveries when feasible to avert compromise of their future reproductive health, and provide adequate contraceptive counseling to prevent unplanned pregnancies. We also must engage the partners of these young women in HIV prevention strategies and HIV testing (in our cohort, most of the sexual partners were seronegative males) and work with them in preventing the acquisition of other sexually transmitted infections. Abnormal cervical cytology and STIs affected 80% of the patients in our cohort, and high rates of STIs have been reported in other cohorts of HIV-infected adolescents.

Engaging the patients in care and viral suppression prior to conception and educating their partners to avert HIV acquisition will be among the highest priorities in years to come, especially since this group is more disenfranchised from the health care system and less likely to engage in pregnancy prevention and planning.

Dr. Bardequez said she had no relevant financial disclosures.

Arlene D. Bardeguez, M.D., M.P.H.

Source Courtesy Dr. Arlene D. Bardequez

Source Elsevier Global Medical News

Source Elsevier Global Medical News

HIV and Nonpregnant Women

[email protected]

Since the beginning of the HIV/AIDS epidemic in the 1980s, women have been affected with this dreaded disease. Many have been at risk of acquiring the virus, many have become infected, and some consequently have been at risk of transmitting the virus to their offspring.

Ob.gyns. have played a major role in the dramatic decrease in mother-to-child transmission since its peak in 1992; prenatal screening followed by the use of antiretroviral therapy in women found to be HIV infected has reduced the risk of HIV transmission from a woman to her child to less than 2%.

Among nonpregnant women, our frequency of HIV testing has been variable, depending, for many of us, on the prevalence of HIV/AIDS in our communities and on our knowledge and/or perception of each patient's risk level.

In recent years, this selective approach to testing has been deemed faulty. It is an approach that has been subject to our own biases of risk and to misperceptions of many of the women we care for. Moreover, a risk-based approach has increasingly conflicted with the changing face of AIDS/HIV infection – most notably, the rise in heterosexual transmission and the fact that 1 in 5 infected individuals (including our patient's sexual partners) are estimated to be unaware of their infection.

Studies have shown that many women who are found to be HIV positive did not consider themselves to be at risk. Had we seen these women, we might not have considered them to be at risk either. According to the most recent HIV Surveillance Report from the Centers for Disease Control and Prevention, 30% of HIV-positive women were tested for HIV late in their illness (that is, diagnosed with AIDS within 1 year of testing positive). Had they been diagnosed earlier, these women could have had years added to their lives with the early and ongoing use of highly active antiretroviral therapy.

For these and other reasons, there are significant public health advantages to the recommendation issued by the CDC 5 years ago that health care providers routinely test (with patient notification and an opportunity to decline) all patients aged 13–64 years.

As providers for women of all ages, it is important that we are aware of changing trends and issues in the HIV epidemic and that we are attentive to the CDC's recommendations. It is for this reason that I have invited Dr. Arlene D. Bardeguez to address the role that the ob.gyn. plays in HIV prevention and testing in nonpregnant women.

Dr. Bardeguez serves as professor in the department of obstetrics, gynecology and women's health and director of HIV services at the New Jersey Medical School, Newark, N.J. Her special interest and expertise in HIV-infected women is evident in her clinical work and patient care, her research and writing, her teaching, and her work in the policy arena.

Here, she explains why HIV prevention strategies, including HIV pre- or postexposure prophylaxis, should become part of our routine clinical care. She also details how we can care for perinatally infected adolescents and how we must address the risks faced by our older patients. As providers of women's health through the age spectrum, she explains, we have an important role to play in the prevention of HIV acquisition and transmission.

It is fairly well appreciated that more than 1.7 million individuals living in the United States are now infected with HIV. What may be less appreciated by patients and physicians is that the impact of the epidemic on women has grown significantly over time.

In 2008, using a new national surveillance system, the Centers for Disease Control and Prevention reported that women comprised more than one-fourth (27%) of the 56,300 people estimated to have been newly infected with HIV in 2006. That is, an estimated 15,000 women were newly infected with HIV in 2006 alone (JAMA 2008;300:520-9).

As ob.gyns., we care for women of all ages, from adolescents through women in their senior years. Therefore, we are in the unique position to be able to identify HIV-infected individuals who could benefit significantly from early monitoring and treatment and to decrease the risk of transmission on a broader level. We can protect our patients, and they in turn can protect their partners and future children. This is a tremendous opportunity that we should not miss.

In 2006, the CDC moved away from HIV testing recommendations that were risk based and advised routine screening for HIV infection for all patients aged 13–64 years. It recommended that physicians notify their patients that testing will be performed unless they decline.

The American College of Obstetricians and Gynecologists weighed in the next year, saying that universal testing with patient notification is more effective in identifying infected patients than is targeted risk-based testing, largely because many women found to be infected with HIV did not consider themselves to have been at risk. Opt-out testing is less cumbersome, ACOG pointed out, because it removes the requirement for lengthy pretest counseling and for detailed, testing-related informed consent.

A 2007 survey of almost 400 medical staff at San Francisco General Hospital, which serves a population with a high HIV prevalence, showed that clinicians in obstetrics/gynecology and HIV infectious disease specialists were significantly more likely to routinely test their patients than were physicians in other specialties (J. Acquir. Immune Defic. Syndr. 2009;50:114-6).

As the authors say, however, this isn't surprising given the fact that ACOG, the CDC, and other national organizations have long called for universal prenatal HIV testing.

Other studies have suggested that compliance with the CDC recommendations is low – that many ob.gyns. as well as other primary care physicians – do not routinely offer HIV testing outside of the prenatal context. Even in high prevalence areas such as New Jersey, we have documented that many providers miss the opportunity to offer the test to all their patients, deeming them unlikely to be infected. This approach hinders early access to care for potentially infected patients and fails to address an unknowing risk of transmission to partners.

We need to think differently.

We do not ask patients: Do you want a chlamydia test? A Pap smear? A cholesterol test? We tell them, instead, that we're going to check their cholesterol levels, or that it's time for their Pap smear, or that we'd recommend a chlamydia test. We need to present HIV testing in the same way – as part of the routine battery of tests that will be performed unless the patient declines. Studies have shown, interestingly, that more patients accept recommended HIV testing when they know it's routinely offered to everyone, without an assessment of risk.

Furthermore, we need to do so in spite of race and in patients of all ages, with all the generations of women whom we see.

As ob.gyns., we must be cognizant of the changing face of HIV infection in women in the last 30 years and realize the unique challenge that we now face as the infection encompasses patients from a broader age spectrum than just those in their reproductive years. We have witnessed both an increased number of HIV infections in the older population and increased survival among individuals on highly active antiretroviral therapy (HAART), including those who were perinatally infected.

We also must appreciate that women who are engaged in a relationship with a known HIV-infected partner, regardless of their age, can prevent acquisition with barrier methods such as condom use or postexposure prophylaxis as delineated by the CDC for serodiscordant couples.

Women Older Than 50

Studies have consistently demonstrated that many women are sexually active into their 60s. Even when we appreciate this, we tend to dismiss the possibility that any of our older patients might have HIV infection. We tend to assume that our patients are in stable relationships and presume there is no value to HIV testing.

It's important to appreciate, however, that the number of people aged 50 years and older who are living with HIV/AIDS has been increasing in recent years. It is estimated that almost one-fourth of all people with HIV/AIDS in the United States are age 50 years and older. While this is partly because HAART has extended the lives of many HIV-infected people, it is also attributed to newly diagnosed infections in people over 50 years.

The 2008 CDC analysis that showed that women comprised more than one-fourth of the individuals newly infected with HIV in 2006 also found that 25% of the new infections were in individuals (men and women) aged 40–49 years (13,900 out of 56,300). Another 10% were in individuals aged 50 years and over (5,800 of 56,300). Approximately 30% of the infections were a result of heterosexual contact. Earlier data from the 1990s similarly showed over 10% of new AIDS cases occurring in people older than age 50 years.

Women of all ages can wrongly believe they are not at risk of contracting HIV. In one recent survey examining patient attitudes about HIV testing and knowledge about their own risk status, only 2% of approximately 850 women of various ages considered themselves at high risk for HIV infection despite the fact that almost half of them reported having had unprotected sex at some point with more than one partner. The women were patients of ob.gyn. members of ACOG's Collaborative Ambulatory Research Network (Matern. Child Health J. 2009;13:355-63).

Older women are generally even less knowledgeable about HIV transmission and how to protect themselves than are younger women, and they are not concerned about undesired pregnancy. For these reasons, many older women may not be practicing safer sex, increasing their risk for HIV and other sexually transmitted infections.

In a nine-question survey of 514 urban women aged 50 years and older (mean age of 62), the majority of women scored poorly, answering four or fewer of the questions correctly. Eighty-four percent correctly identified unprotected heterosexual sex as a moderate- to high-risk activity, but only 13% identified condoms as being very effective in preventing HIV, and 18% said condoms are not at all effective (J. Amer. Geriatr. Soc. 2004;52:1549-53).

In another study aimed at assessing differences in the characteristics of individuals (both men and women) who refuse testing and those who accept it, investigators found that HIV test refusal was associated with female gender, white race, older age, and higher educational level (AIDS Patient Care STDS 2006;20:84-92).

Older women must be educated about their risk of heterosexual transmission and the fact that the risk for HIV acquisition has been increasing since 1994 in the United States. They need to understand that normal physiologic changes in the menopausal period such as thinning of the vaginal mucosa, increased susceptibility to vaginal abrasions during intercourse, and changes in their immune response can make them more vulnerable to disease acquisition or progression.

It also is important to educate them about the effectiveness of condoms and the importance of knowing the HIV status of their partners, because it is estimated that approximately 21% of infected individuals in the United States do not know their HIV status.

Most of all, our patients should understand that 30 years into the epidemic, we have demonstrated excellent survival in individuals on treatment, particularly among those who were diagnosed early and who are receiving HAART. A recent report from the CDC shows that average life expectancy after HIV diagnosis in the general population increased from 10.5 to 22.5 years from 1996 to 2005 (J. Acquir. Immune Defic. Syndr. 2010;53:124-130).

Interestingly, as the report points out, studies have shown that although HIV-infected women had a greater life expectancy to begin with, they showed a lesser magnitude of improvement than did men, particularly white men. (Women's life expectancy changed from 12.6 to 23.6 years.) This observation highlights the importance of earlier diagnosis and link to care.

Knowing about the successes of HAART is important because women are less likely to opt out of HIV testing when they perceive the benefits. We can explain to patients – especially those who are apprehensive about the test – that the test is integrated into the annual health care panel (along with cholesterol and triglyceride testing, and genital cytology), and that, contrary to decades ago, we can treat and control HIV disease once it is diagnosed, which is more than we can do for certain types of cancer.

Our patients need to understand that it can be a manageable chronic disease as long as it is detected and effectively addressed early in the course of their infection. Today, we have access to a wide array of pamphlets and videos that we can offer in the waiting area to help patients understand this and appreciate the value of HIV testing.

Serodiscordant Couples

Thus far, there is limited standardization or consensus on how and when to provide counseling, testing, and prevention strategies for women who are involved in HIV serodiscordant relationships. However, most experts recommend that patients whose partners are HIV positive should be tested for HIV infection annually and encouraged to use effective prevention strategies such as the male or female condom.

Screening and treatment for sexually transmitted diseases should be done annually as coinfection can increase the risk of HIV transmission. In one of the studies demonstrating an impact of STD treatment – a randomized trial conducted more than 15 years ago in rural Tanzania – improved STD education and treatment reduced HIV incidence by about 40% (Lancet 1995;346:530-6).

In cases in which an unplanned sexual encounter with an HIV-positive partner occurs without protection, postexposure prophylaxis should be considered and given as soon as the event is identified, preferably within 48 hours. The CDC's recommendations for the use of antiretroviral postexposure prophylaxis, issued in 2005, call for a 28-day course of HAART (MMWR 2005;54[RR02]:1-20)

Decisions about the optimal postexposure therapy involve various factors, including the partner's antiretroviral history, adherence to the regimen, and most recent viral load. We may need to counsel patients, however, that having undetectable virus in the blood does not necessarily mean there will not be any virus in the genital tract. Discrepancies between serum and genital viral load have been reported among HIV-infected men and women on HAART.

If a woman engages in unprotected sex with a male of unknown serostatus, she can request postexposure prophylaxis. In this case, she should be counseled about the risks and benefits of postexposure HAART, as she may expose herself to unnecessary toxicities.

When faced with these situations we can obtain guidance from, or work in partnership with, the infectious disease provider who is managing the HIV-infected partner, or we can contact state or national phone lines for linkage to immediate care. Some health departments have established nonoccupational postexposure prophylaxis programs in their jurisdictions. Overall, it is important that we be aware of the availability of postexposure HAART and its possible risks and benefits.

In the near future, a woman whose partner is HIV positive should be able to benefit from antiretroviral microbicides used before or after intercourse. In the double-blind, randomized, and well-publicized CAPRISA (Centre for the AIDS Programme of Research in South Africa) 004 trial, a 1% vaginal gel formulation of tenofovir reduced HIV acquisition by approximately 39% overall and by 54% in women with high adherence to the protocol for gel application (Science 2010;329:1168-74). Another phase III trial of tenofovir is ongoing.

The field of safe reproduction for HIV serodiscordant couples also is advancing, such that women and their partners have various options for conceiving with minimal risk of transmitting the infection.

A large body of evidence suggests that reproductive technology – that is, sperm washing and artificial insemination – can help HIV-affected couples safely conceive, and the results of further CDC-sponsored research aimed at evaluating outcomes in couples who have used these techniques to conceive are expected soon. For many couples, however, such technologies are economically inaccessible.

Experts are looking at periconception preexposure prophylaxis as a potential strategy for preventing HIV transmission in couples trying to conceive. Under this approach, the seronegative partner would take antiretroviral drugs during periods of attempted conception, with the goal of preventing initial viral replication. Clinical trials evaluating its safety and efficacy are ongoing.

Other components of a risk-reduction program should include suppressive antiretroviral therapy for the infected partner (who may yet be eligible for such therapy under current recommendations for CD4 cell count), screening and pretreatment for other sexually transmitted infections, and unprotected sexual intercourse that is limited to times of peak fertility (AIDS 2010;24:1975-82).

Perinatally Infected Women

An increasing number of perinatally infected adolescents has been identified and engaged in care throughout the United States. This population is unique in that many were initially exposed to monotherapy or dual antiretroviral regimens and thus have developed resistance to several antiretroviral regimens. In addition, their adherence to HAART is lower than required to avert failure on current regimens.

They are a population that presents a challenge to ob.gyns. because many have high-risk sexual behaviors, and when they become pregnant, they are at high risk for operative delivery due to inadequate viral suppression.

We recently reported on the pregnancy outcomes of 10 perinatally infected adolescents referred to the high-risk obstetric clinic at our institution and delivered between 1997 and 2007. Neonatal outcomes were generally favorable, but these young women had a high rate of operative delivery (62%, compared with our institutional rate of 33%) due to the failure to achieve undetectable viral load (Am. J. Obstet. Gynecol. 2009;200:149.e1-5).

Our goals for these women should therefore include supporting and counseling on treatment adherence and on the selection of well-tolerated HAART regimens that can rapidly suppress viral load. We should also minimize the use of operative deliveries when feasible to avert compromise of their future reproductive health, and provide adequate contraceptive counseling to prevent unplanned pregnancies. We also must engage the partners of these young women in HIV prevention strategies and HIV testing (in our cohort, most of the sexual partners were seronegative males) and work with them in preventing the acquisition of other sexually transmitted infections. Abnormal cervical cytology and STIs affected 80% of the patients in our cohort, and high rates of STIs have been reported in other cohorts of HIV-infected adolescents.

Engaging the patients in care and viral suppression prior to conception and educating their partners to avert HIV acquisition will be among the highest priorities in years to come, especially since this group is more disenfranchised from the health care system and less likely to engage in pregnancy prevention and planning.

Dr. Bardequez said she had no relevant financial disclosures.

Arlene D. Bardeguez, M.D., M.P.H.

Source Courtesy Dr. Arlene D. Bardequez

Source Elsevier Global Medical News

Source Elsevier Global Medical News

HIV and Nonpregnant Women

[email protected]

Since the beginning of the HIV/AIDS epidemic in the 1980s, women have been affected with this dreaded disease. Many have been at risk of acquiring the virus, many have become infected, and some consequently have been at risk of transmitting the virus to their offspring.

Ob.gyns. have played a major role in the dramatic decrease in mother-to-child transmission since its peak in 1992; prenatal screening followed by the use of antiretroviral therapy in women found to be HIV infected has reduced the risk of HIV transmission from a woman to her child to less than 2%.

Among nonpregnant women, our frequency of HIV testing has been variable, depending, for many of us, on the prevalence of HIV/AIDS in our communities and on our knowledge and/or perception of each patient's risk level.

In recent years, this selective approach to testing has been deemed faulty. It is an approach that has been subject to our own biases of risk and to misperceptions of many of the women we care for. Moreover, a risk-based approach has increasingly conflicted with the changing face of AIDS/HIV infection – most notably, the rise in heterosexual transmission and the fact that 1 in 5 infected individuals (including our patient's sexual partners) are estimated to be unaware of their infection.

Studies have shown that many women who are found to be HIV positive did not consider themselves to be at risk. Had we seen these women, we might not have considered them to be at risk either. According to the most recent HIV Surveillance Report from the Centers for Disease Control and Prevention, 30% of HIV-positive women were tested for HIV late in their illness (that is, diagnosed with AIDS within 1 year of testing positive). Had they been diagnosed earlier, these women could have had years added to their lives with the early and ongoing use of highly active antiretroviral therapy.

For these and other reasons, there are significant public health advantages to the recommendation issued by the CDC 5 years ago that health care providers routinely test (with patient notification and an opportunity to decline) all patients aged 13–64 years.

As providers for women of all ages, it is important that we are aware of changing trends and issues in the HIV epidemic and that we are attentive to the CDC's recommendations. It is for this reason that I have invited Dr. Arlene D. Bardeguez to address the role that the ob.gyn. plays in HIV prevention and testing in nonpregnant women.

Dr. Bardeguez serves as professor in the department of obstetrics, gynecology and women's health and director of HIV services at the New Jersey Medical School, Newark, N.J. Her special interest and expertise in HIV-infected women is evident in her clinical work and patient care, her research and writing, her teaching, and her work in the policy arena.

Here, she explains why HIV prevention strategies, including HIV pre- or postexposure prophylaxis, should become part of our routine clinical care. She also details how we can care for perinatally infected adolescents and how we must address the risks faced by our older patients. As providers of women's health through the age spectrum, she explains, we have an important role to play in the prevention of HIV acquisition and transmission.

It is fairly well appreciated that more than 1.7 million individuals living in the United States are now infected with HIV. What may be less appreciated by patients and physicians is that the impact of the epidemic on women has grown significantly over time.

In 2008, using a new national surveillance system, the Centers for Disease Control and Prevention reported that women comprised more than one-fourth (27%) of the 56,300 people estimated to have been newly infected with HIV in 2006. That is, an estimated 15,000 women were newly infected with HIV in 2006 alone (JAMA 2008;300:520-9).

As ob.gyns., we care for women of all ages, from adolescents through women in their senior years. Therefore, we are in the unique position to be able to identify HIV-infected individuals who could benefit significantly from early monitoring and treatment and to decrease the risk of transmission on a broader level. We can protect our patients, and they in turn can protect their partners and future children. This is a tremendous opportunity that we should not miss.

In 2006, the CDC moved away from HIV testing recommendations that were risk based and advised routine screening for HIV infection for all patients aged 13–64 years. It recommended that physicians notify their patients that testing will be performed unless they decline.

The American College of Obstetricians and Gynecologists weighed in the next year, saying that universal testing with patient notification is more effective in identifying infected patients than is targeted risk-based testing, largely because many women found to be infected with HIV did not consider themselves to have been at risk. Opt-out testing is less cumbersome, ACOG pointed out, because it removes the requirement for lengthy pretest counseling and for detailed, testing-related informed consent.

A 2007 survey of almost 400 medical staff at San Francisco General Hospital, which serves a population with a high HIV prevalence, showed that clinicians in obstetrics/gynecology and HIV infectious disease specialists were significantly more likely to routinely test their patients than were physicians in other specialties (J. Acquir. Immune Defic. Syndr. 2009;50:114-6).

As the authors say, however, this isn't surprising given the fact that ACOG, the CDC, and other national organizations have long called for universal prenatal HIV testing.

Other studies have suggested that compliance with the CDC recommendations is low – that many ob.gyns. as well as other primary care physicians – do not routinely offer HIV testing outside of the prenatal context. Even in high prevalence areas such as New Jersey, we have documented that many providers miss the opportunity to offer the test to all their patients, deeming them unlikely to be infected. This approach hinders early access to care for potentially infected patients and fails to address an unknowing risk of transmission to partners.

We need to think differently.

We do not ask patients: Do you want a chlamydia test? A Pap smear? A cholesterol test? We tell them, instead, that we're going to check their cholesterol levels, or that it's time for their Pap smear, or that we'd recommend a chlamydia test. We need to present HIV testing in the same way – as part of the routine battery of tests that will be performed unless the patient declines. Studies have shown, interestingly, that more patients accept recommended HIV testing when they know it's routinely offered to everyone, without an assessment of risk.

Furthermore, we need to do so in spite of race and in patients of all ages, with all the generations of women whom we see.

As ob.gyns., we must be cognizant of the changing face of HIV infection in women in the last 30 years and realize the unique challenge that we now face as the infection encompasses patients from a broader age spectrum than just those in their reproductive years. We have witnessed both an increased number of HIV infections in the older population and increased survival among individuals on highly active antiretroviral therapy (HAART), including those who were perinatally infected.

We also must appreciate that women who are engaged in a relationship with a known HIV-infected partner, regardless of their age, can prevent acquisition with barrier methods such as condom use or postexposure prophylaxis as delineated by the CDC for serodiscordant couples.

Women Older Than 50

Studies have consistently demonstrated that many women are sexually active into their 60s. Even when we appreciate this, we tend to dismiss the possibility that any of our older patients might have HIV infection. We tend to assume that our patients are in stable relationships and presume there is no value to HIV testing.

It's important to appreciate, however, that the number of people aged 50 years and older who are living with HIV/AIDS has been increasing in recent years. It is estimated that almost one-fourth of all people with HIV/AIDS in the United States are age 50 years and older. While this is partly because HAART has extended the lives of many HIV-infected people, it is also attributed to newly diagnosed infections in people over 50 years.

The 2008 CDC analysis that showed that women comprised more than one-fourth of the individuals newly infected with HIV in 2006 also found that 25% of the new infections were in individuals (men and women) aged 40–49 years (13,900 out of 56,300). Another 10% were in individuals aged 50 years and over (5,800 of 56,300). Approximately 30% of the infections were a result of heterosexual contact. Earlier data from the 1990s similarly showed over 10% of new AIDS cases occurring in people older than age 50 years.

Women of all ages can wrongly believe they are not at risk of contracting HIV. In one recent survey examining patient attitudes about HIV testing and knowledge about their own risk status, only 2% of approximately 850 women of various ages considered themselves at high risk for HIV infection despite the fact that almost half of them reported having had unprotected sex at some point with more than one partner. The women were patients of ob.gyn. members of ACOG's Collaborative Ambulatory Research Network (Matern. Child Health J. 2009;13:355-63).

Older women are generally even less knowledgeable about HIV transmission and how to protect themselves than are younger women, and they are not concerned about undesired pregnancy. For these reasons, many older women may not be practicing safer sex, increasing their risk for HIV and other sexually transmitted infections.

In a nine-question survey of 514 urban women aged 50 years and older (mean age of 62), the majority of women scored poorly, answering four or fewer of the questions correctly. Eighty-four percent correctly identified unprotected heterosexual sex as a moderate- to high-risk activity, but only 13% identified condoms as being very effective in preventing HIV, and 18% said condoms are not at all effective (J. Amer. Geriatr. Soc. 2004;52:1549-53).

In another study aimed at assessing differences in the characteristics of individuals (both men and women) who refuse testing and those who accept it, investigators found that HIV test refusal was associated with female gender, white race, older age, and higher educational level (AIDS Patient Care STDS 2006;20:84-92).

Older women must be educated about their risk of heterosexual transmission and the fact that the risk for HIV acquisition has been increasing since 1994 in the United States. They need to understand that normal physiologic changes in the menopausal period such as thinning of the vaginal mucosa, increased susceptibility to vaginal abrasions during intercourse, and changes in their immune response can make them more vulnerable to disease acquisition or progression.

It also is important to educate them about the effectiveness of condoms and the importance of knowing the HIV status of their partners, because it is estimated that approximately 21% of infected individuals in the United States do not know their HIV status.

Most of all, our patients should understand that 30 years into the epidemic, we have demonstrated excellent survival in individuals on treatment, particularly among those who were diagnosed early and who are receiving HAART. A recent report from the CDC shows that average life expectancy after HIV diagnosis in the general population increased from 10.5 to 22.5 years from 1996 to 2005 (J. Acquir. Immune Defic. Syndr. 2010;53:124-130).

Interestingly, as the report points out, studies have shown that although HIV-infected women had a greater life expectancy to begin with, they showed a lesser magnitude of improvement than did men, particularly white men. (Women's life expectancy changed from 12.6 to 23.6 years.) This observation highlights the importance of earlier diagnosis and link to care.

Knowing about the successes of HAART is important because women are less likely to opt out of HIV testing when they perceive the benefits. We can explain to patients – especially those who are apprehensive about the test – that the test is integrated into the annual health care panel (along with cholesterol and triglyceride testing, and genital cytology), and that, contrary to decades ago, we can treat and control HIV disease once it is diagnosed, which is more than we can do for certain types of cancer.

Our patients need to understand that it can be a manageable chronic disease as long as it is detected and effectively addressed early in the course of their infection. Today, we have access to a wide array of pamphlets and videos that we can offer in the waiting area to help patients understand this and appreciate the value of HIV testing.

Serodiscordant Couples

Thus far, there is limited standardization or consensus on how and when to provide counseling, testing, and prevention strategies for women who are involved in HIV serodiscordant relationships. However, most experts recommend that patients whose partners are HIV positive should be tested for HIV infection annually and encouraged to use effective prevention strategies such as the male or female condom.

Screening and treatment for sexually transmitted diseases should be done annually as coinfection can increase the risk of HIV transmission. In one of the studies demonstrating an impact of STD treatment – a randomized trial conducted more than 15 years ago in rural Tanzania – improved STD education and treatment reduced HIV incidence by about 40% (Lancet 1995;346:530-6).

In cases in which an unplanned sexual encounter with an HIV-positive partner occurs without protection, postexposure prophylaxis should be considered and given as soon as the event is identified, preferably within 48 hours. The CDC's recommendations for the use of antiretroviral postexposure prophylaxis, issued in 2005, call for a 28-day course of HAART (MMWR 2005;54[RR02]:1-20)

Decisions about the optimal postexposure therapy involve various factors, including the partner's antiretroviral history, adherence to the regimen, and most recent viral load. We may need to counsel patients, however, that having undetectable virus in the blood does not necessarily mean there will not be any virus in the genital tract. Discrepancies between serum and genital viral load have been reported among HIV-infected men and women on HAART.

If a woman engages in unprotected sex with a male of unknown serostatus, she can request postexposure prophylaxis. In this case, she should be counseled about the risks and benefits of postexposure HAART, as she may expose herself to unnecessary toxicities.

When faced with these situations we can obtain guidance from, or work in partnership with, the infectious disease provider who is managing the HIV-infected partner, or we can contact state or national phone lines for linkage to immediate care. Some health departments have established nonoccupational postexposure prophylaxis programs in their jurisdictions. Overall, it is important that we be aware of the availability of postexposure HAART and its possible risks and benefits.

In the near future, a woman whose partner is HIV positive should be able to benefit from antiretroviral microbicides used before or after intercourse. In the double-blind, randomized, and well-publicized CAPRISA (Centre for the AIDS Programme of Research in South Africa) 004 trial, a 1% vaginal gel formulation of tenofovir reduced HIV acquisition by approximately 39% overall and by 54% in women with high adherence to the protocol for gel application (Science 2010;329:1168-74). Another phase III trial of tenofovir is ongoing.

The field of safe reproduction for HIV serodiscordant couples also is advancing, such that women and their partners have various options for conceiving with minimal risk of transmitting the infection.

A large body of evidence suggests that reproductive technology – that is, sperm washing and artificial insemination – can help HIV-affected couples safely conceive, and the results of further CDC-sponsored research aimed at evaluating outcomes in couples who have used these techniques to conceive are expected soon. For many couples, however, such technologies are economically inaccessible.

Experts are looking at periconception preexposure prophylaxis as a potential strategy for preventing HIV transmission in couples trying to conceive. Under this approach, the seronegative partner would take antiretroviral drugs during periods of attempted conception, with the goal of preventing initial viral replication. Clinical trials evaluating its safety and efficacy are ongoing.

Other components of a risk-reduction program should include suppressive antiretroviral therapy for the infected partner (who may yet be eligible for such therapy under current recommendations for CD4 cell count), screening and pretreatment for other sexually transmitted infections, and unprotected sexual intercourse that is limited to times of peak fertility (AIDS 2010;24:1975-82).

Perinatally Infected Women

An increasing number of perinatally infected adolescents has been identified and engaged in care throughout the United States. This population is unique in that many were initially exposed to monotherapy or dual antiretroviral regimens and thus have developed resistance to several antiretroviral regimens. In addition, their adherence to HAART is lower than required to avert failure on current regimens.

They are a population that presents a challenge to ob.gyns. because many have high-risk sexual behaviors, and when they become pregnant, they are at high risk for operative delivery due to inadequate viral suppression.

We recently reported on the pregnancy outcomes of 10 perinatally infected adolescents referred to the high-risk obstetric clinic at our institution and delivered between 1997 and 2007. Neonatal outcomes were generally favorable, but these young women had a high rate of operative delivery (62%, compared with our institutional rate of 33%) due to the failure to achieve undetectable viral load (Am. J. Obstet. Gynecol. 2009;200:149.e1-5).

Our goals for these women should therefore include supporting and counseling on treatment adherence and on the selection of well-tolerated HAART regimens that can rapidly suppress viral load. We should also minimize the use of operative deliveries when feasible to avert compromise of their future reproductive health, and provide adequate contraceptive counseling to prevent unplanned pregnancies. We also must engage the partners of these young women in HIV prevention strategies and HIV testing (in our cohort, most of the sexual partners were seronegative males) and work with them in preventing the acquisition of other sexually transmitted infections. Abnormal cervical cytology and STIs affected 80% of the patients in our cohort, and high rates of STIs have been reported in other cohorts of HIV-infected adolescents.

Engaging the patients in care and viral suppression prior to conception and educating their partners to avert HIV acquisition will be among the highest priorities in years to come, especially since this group is more disenfranchised from the health care system and less likely to engage in pregnancy prevention and planning.

Dr. Bardequez said she had no relevant financial disclosures.

Arlene D. Bardeguez, M.D., M.P.H.

Source Courtesy Dr. Arlene D. Bardequez

Source Elsevier Global Medical News

Source Elsevier Global Medical News

HIV and Nonpregnant Women

[email protected]

Since the beginning of the HIV/AIDS epidemic in the 1980s, women have been affected with this dreaded disease. Many have been at risk of acquiring the virus, many have become infected, and some consequently have been at risk of transmitting the virus to their offspring.

Ob.gyns. have played a major role in the dramatic decrease in mother-to-child transmission since its peak in 1992; prenatal screening followed by the use of antiretroviral therapy in women found to be HIV infected has reduced the risk of HIV transmission from a woman to her child to less than 2%.

Among nonpregnant women, our frequency of HIV testing has been variable, depending, for many of us, on the prevalence of HIV/AIDS in our communities and on our knowledge and/or perception of each patient's risk level.

In recent years, this selective approach to testing has been deemed faulty. It is an approach that has been subject to our own biases of risk and to misperceptions of many of the women we care for. Moreover, a risk-based approach has increasingly conflicted with the changing face of AIDS/HIV infection – most notably, the rise in heterosexual transmission and the fact that 1 in 5 infected individuals (including our patient's sexual partners) are estimated to be unaware of their infection.

Studies have shown that many women who are found to be HIV positive did not consider themselves to be at risk. Had we seen these women, we might not have considered them to be at risk either. According to the most recent HIV Surveillance Report from the Centers for Disease Control and Prevention, 30% of HIV-positive women were tested for HIV late in their illness (that is, diagnosed with AIDS within 1 year of testing positive). Had they been diagnosed earlier, these women could have had years added to their lives with the early and ongoing use of highly active antiretroviral therapy.

For these and other reasons, there are significant public health advantages to the recommendation issued by the CDC 5 years ago that health care providers routinely test (with patient notification and an opportunity to decline) all patients aged 13–64 years.

As providers for women of all ages, it is important that we are aware of changing trends and issues in the HIV epidemic and that we are attentive to the CDC's recommendations. It is for this reason that I have invited Dr. Arlene D. Bardeguez to address the role that the ob.gyn. plays in HIV prevention and testing in nonpregnant women.

Dr. Bardeguez serves as professor in the department of obstetrics, gynecology and women's health and director of HIV services at the New Jersey Medical School, Newark, N.J. Her special interest and expertise in HIV-infected women is evident in her clinical work and patient care, her research and writing, her teaching, and her work in the policy arena.

Here, she explains why HIV prevention strategies, including HIV pre- or postexposure prophylaxis, should become part of our routine clinical care. She also details how we can care for perinatally infected adolescents and how we must address the risks faced by our older patients. As providers of women's health through the age spectrum, she explains, we have an important role to play in the prevention of HIV acquisition and transmission.

Deep endometriosis involving the bowel is uncommon, but not rare, among patients with pelvic endometriosis. There is a growing body of literature describing minimally invasive colorectal resection for invasive endometriosis, and a growing feeling among gynecologic surgeons that endometriosis involving the rectal wall is better treated with rectal resection than with a shaving, or scraping, technique.

In our experience of working within a multispecialty surgical team, addressing endometriosis in a systematic fashion has led to a higher rate of patient satisfaction and quality of life, a lower rate of recurrent symptoms, and less surgical morbidity than has been seen with other approaches. While still infrequent, the greatest impact of this multispecialty approach has been the more liberal inclusion of bowel resection as part of the treatment for deep pelvic endometriosis.

It is our belief that patients who have evidence of bowel involvement and have a colorectal resection as part of their endometriosis treatment ultimately have a better surgical outcome. The opposite also appears to hold true. Those patients who have endometriosis involving the bowel wall and who do not receive a bowel resection have been more prone to recurrent disease and symptoms resulting in subsequent reoperations that are often more difficult to perform than a planned colorectal resection would have been.

Furthermore, surgery involving premeditated bowel resection is often safer than treatment employing a scraping technique when endometriosis is deep seated. In cases in which endometriosis has infiltrated the rectal wall, scraping techniques often result in unidentified bowel perforation, which leads to operations that are much more difficult to perform than planned elective resections. In addition, delayed missed injuries often will result in multiple trips to the operating room that frequently include the use of a colostomy in an otherwise young, healthy female patient.

Diagnosing Rectal Involvement

The full extent of disease, including the precise involvement of the bowel, sometimes can be difficult to determine prior to surgery. In many cases, however, physical examination combined with endoscopy and ultrasonography is enough to diagnose bowel involvement that would best be treated with segmental resection.

Pain associated with defecation, rectal bleeding around menses, and constipation often are found on history. On physical exam, endometriosis involving the rectal wall is most commonly associated with pain on both vaginal and rectal examination. Nodularity of the rectal wall, as well as distortion of the rectal folds, often is palpated.

The presence of either nodularity or rectal wall distortion raises the likelihood that there is significant rectal wall involvement. Conversely, if there is minimal or no nodularity of the rectal wall, the likelihood of rectal involvement is low.

When physical examination is abnormal, we proceed with colonoscopy, which enables us to visualize external compression on the rectal wall, nodularity, or other signs of endometriosis infiltrating the bowel wall. Colonoscopy is preferred because endometriosis is often found in the cecum, which would be missed on flexible sigmoidoscopy. Rarely will endometriosis actually penetrate the bowel wall. Most of the time, wall distortion with what is often seen as a submucosal mass is the only positive finding. Biopsy of the abnormality is typically unsatisfying, with normal colonic mucosa being the most common finding on microscopy.

When there are positive endoscopic findings, we prefer endorectal ultrasonography (ERUS) over transvaginal ultrasound to further evaluate rectal wall involvement of endometriosis. Using a 10-megahertz, 3-D ultrasound, we are able to visualize if endometriosis is attached to the bowel wall and to what degree.

The most important finding is whether or not endometriosis is invading the muscularis propria by the presence or absence of a space between the lesion(s) and the muscularis propria. If a space is visualized, there is a fairly good chance that the endometriosis may be safely scraped off the rectum.

If we see, on the other hand, that the endometriosis is either invading or firmly attached to the rectal muscularis propria, we know that chances of successfully scraping the lesion(s) off the rectum will be very low. In that case, a segmental resection of the rectum can be scheduled in conjunction with the rest of the endometriosis removal. Previous endometriosis surgery, it must be noted, leads to scar tissue which will often distort ERUS images and make the exam less accurate.

Surgical Planning

When the preoperative work-up is confirmatory, a combined surgical approach is scheduled. When the work-up is negative and the patient is scheduled for removal of endometriosis from the nongastrointestinal organs, we remain on surgical standby because bowel involvement of endometriosis is occasionally discovered in symptomatic patients despite a negative work-up.

Standby status also allows the gynecologist to be more aggressive because rectal wall injury can be corrected at the time of surgery. Flexible sigmoidoscopy always should be performed at the end of any endometriosis operation in which bowel proximity is encountered. By submerging the bowel under water and inflating air via the sigmoidoscope, the presence of air bubbles often will identify a missed bowel injury.

Segmental resection usually involves no more than 5-6 cm of the rectum. We do, however, extend the resection a bit proximally if the patient has a history of chronic constipation. Resecting more of the rectum and sigmoid colon to straighten out the left side of the large bowel and shorten the overall length will better alleviate the patient's constipation symptoms. Combined with improvement in the patient's defecation-related symptoms associated with the endometriosis, patient satisfaction regarding the elimination of constipation symptoms is often quite high.

Basic Surgical Technique

Rectal resection for deep endometriosis is comparable to resection of a T4 rectal cancer (one that has invaded outside the rectal wall), except that in the case of endometriosis, we typically are treating young, otherwise healthy patients. In these patients, the risk of complications – mainly, the risk of a permanent colostomy – is all the more concerning. It is important that patients understand the risk and benefits of the surgery and that the colorectal surgeon has the proper expertise for such a technically demanding, risky operation.

The operation is performed in a modified lithotomy position. A laparoscopic or hand-assisted laparoscopic approach can be used. We have performed both techniques, but find a hand-assisted laparoscopic approach faster. A robotic-assisted approach also is being developed.

Depending upon the type of camera used, a 5- or 10-mm port is placed in the umbilicus. The only other ancillary ports needed are 5-mm dissecting ports placed in the right and left lower quadrants. A mini-Pfannenstiel incision is needed to remove the rectal specimen. By extending this incision 3 centimeters, a hand port for hand-assisted laparoscopic surgery can be placed.

Surgery is initiated by the gynecologic surgeon, who resects endometriosis off all nongastrointestinal organs. Endometriosis involving the colon and rectum is left intact. If indicated, hysterectomy with salpingo-oophorectomy is performed at this time.

Next is the laparoscopic colorectal portion of the surgery. First, the inferior mesenteric artery is ligated at the root of the aorta so that various collateral vessels within the marginal branches and Riolan's arch are not sacrificed. Usually, this ligation alone will adequately free up the sigmoid colon enough for a tension-free anastomosis. If the sigmoid colon still cannot be lowered into the rectum without undue tension, we also will ligate the rectal tributary of the inferior mesenteric vein, one of the two main tributaries of the mesenteric vein.

The remainder of the mobilization involves dissecting along the White line of Toldt until the colon falls freely into the rectum. Rarely will we need to mobilize the splenic flexure of the colon to achieve adequate length.

With the left side of the colon freely mobilized, we turn our attention to the pelvis and subsequent rectal dissection. We do not remove the lesion from the rectum, since we have already confirmed that the lesion is firmly attached to the rectal wall. The endometriosis is removed en bloc with the rectum, similar to what is done for rectal cancer.

While the lateral and posterior dissection of the mesorectum can be easily done laparoscopically or robotically, we believe the anterior dissection of the rectum – removal of the endometriosis off the posterior aspect of the vagina while leaving it attached to the rectum – is more easily performed using a hand-assisted laparoscopic approach or even a hybrid open approach through the mini-Pfannenstiel incision.

Dissection is carried out distally until a soft, normal section of rectum is identified. At least 2 cm of normal rectum is needed for a safe anastomosis.

Endometriosis involving the bowel usually appears as a white fibrotic, submucosal mass and feels similar to invasive rectal cancer. The difference, of course, is that rectal cancer is mostly intraluminal, whereas endometriosis usually originates outside the bowel wall and invades inward. Occasionally, one will find “chocolate”-filled cysts within the endometriotic mass, but this is rare.

Endometriosis with bowel involvement is typically anterior to the rectum and posterior to the vagina, but lesions posterior to the rectum have been found, which would denote a nonanatomical spreading distribution.

There are two techniques for a low colorectal anastomosis: The hand-sewn anastomosis technique and the end-to-end anastomosis (EEA) stapler technique. The hand-sewn anastomosis has largely been replaced by the EEA stapler because stapling the proximal colon to the lower rectum is easier to perform, faster, and results in a similar anastomotic leak rate when compared to a hand-sewn anastomosis.

An anastomosis in the upper or mid-rectum has a low risk of leaking (less than 2%). Sometimes, in patients with a deep cul-de-sac, the lesion is attached to the mid-rectum and the anastomosis must be performed in the lower rectum, within a few centimeters of the anus. Low rectal anastomoses have leak rates as high as 10%. A flexible sigmoidoscopy must be performed to check for an air leak. If one is found, the anastomosis should be reconstructed or repaired. Temporary diverting ileostomy should be considered if the anastomosis is suboptimal.

An endorectal ultrasound shows endometriosis invading the muscularis propria (1 o'clock position).

Source Courtesy Dr. John J. Park

Endometriosis invades the rectal wall. The mucosa is distorted, but intact.

The inferior mesenteric artery (left) is ligated at the root of the aorta so that various collateral vessels within the marginal branches and Riolan's arch are not sacrificed. This ligation alone often will adequately free up the sigmoid colon enough for a tension-free anastomosis. If the sigmoid colon still cannot be lowered into the rectum without undue tension, the surgeon also will ligate the rectal tributary of the inferior mesenteric vein (right), one of the two main tributaries of the mesenteric vein.

The colon is being attached to the distal rectum using an end-to-end anastomosis stapler.

Source Images Courtesy Dr. John J. Park

Deeply Invasive Rectosigmoid Endometriosis

www.isge.org

www.aagl.org

Deeply invasive rectosigmoid endometriosis can be associated with a severe – and at times incapacitating – symptom complex. This includes dysmenorrhea – both premenstrual and menstrual – deep dyspareunia, dyschezia, and rectal bleeding at time of menses. There also can be an impact on fertility as well, which can be rectified with bowel resection. In the accompanying graphic (right), a number of studies revealing pregnancy post bowel resection for rectosigmoid endometriosis are noted.

As bowel resection is generally not in the armamentarium of the gynecologic surgeon treating benign disease, the proper treatment of deep infiltrated rectosigmoid endometriosis must involve a cooperative effort with a colorectal surgeon who is capable of performing advanced minimally invasive surgery. This collaboration permits the minimally invasive gynecologist to laparoscopically excise endometriosis, lyse pelvic adhesions, resect ovarian endometriomata, and where indicated, perform ureterolysis and total laparoscopic hysterectomy. The colorectal or general surgeon can then proceed with the bowel resection via a minimally invasive approach.

For this current Master Class in Gynecologic Surgery, I have solicited the expertise of Dr. John J. Park. Dr. Park is a clinical assistant professor of surgery in the division of colorectal surgery at the University of Illinois at Chicago, as well as attending surgeon at Advocate Lutheran General Hospital, Park Ridge, Ill. Dr. Park completed his residency in general surgery at the University of Illinois and his colorectal surgery residency at Mayo Clinic, Rochester, Minn. Dr. Park is board certified in general surgery and colon and rectal surgery.

Deep endometriosis involving the bowel is uncommon, but not rare, among patients with pelvic endometriosis. There is a growing body of literature describing minimally invasive colorectal resection for invasive endometriosis, and a growing feeling among gynecologic surgeons that endometriosis involving the rectal wall is better treated with rectal resection than with a shaving, or scraping, technique.

In our experience of working within a multispecialty surgical team, addressing endometriosis in a systematic fashion has led to a higher rate of patient satisfaction and quality of life, a lower rate of recurrent symptoms, and less surgical morbidity than has been seen with other approaches. While still infrequent, the greatest impact of this multispecialty approach has been the more liberal inclusion of bowel resection as part of the treatment for deep pelvic endometriosis.

It is our belief that patients who have evidence of bowel involvement and have a colorectal resection as part of their endometriosis treatment ultimately have a better surgical outcome. The opposite also appears to hold true. Those patients who have endometriosis involving the bowel wall and who do not receive a bowel resection have been more prone to recurrent disease and symptoms resulting in subsequent reoperations that are often more difficult to perform than a planned colorectal resection would have been.

Furthermore, surgery involving premeditated bowel resection is often safer than treatment employing a scraping technique when endometriosis is deep seated. In cases in which endometriosis has infiltrated the rectal wall, scraping techniques often result in unidentified bowel perforation, which leads to operations that are much more difficult to perform than planned elective resections. In addition, delayed missed injuries often will result in multiple trips to the operating room that frequently include the use of a colostomy in an otherwise young, healthy female patient.

Diagnosing Rectal Involvement

The full extent of disease, including the precise involvement of the bowel, sometimes can be difficult to determine prior to surgery. In many cases, however, physical examination combined with endoscopy and ultrasonography is enough to diagnose bowel involvement that would best be treated with segmental resection.

Pain associated with defecation, rectal bleeding around menses, and constipation often are found on history. On physical exam, endometriosis involving the rectal wall is most commonly associated with pain on both vaginal and rectal examination. Nodularity of the rectal wall, as well as distortion of the rectal folds, often is palpated.

The presence of either nodularity or rectal wall distortion raises the likelihood that there is significant rectal wall involvement. Conversely, if there is minimal or no nodularity of the rectal wall, the likelihood of rectal involvement is low.

When physical examination is abnormal, we proceed with colonoscopy, which enables us to visualize external compression on the rectal wall, nodularity, or other signs of endometriosis infiltrating the bowel wall. Colonoscopy is preferred because endometriosis is often found in the cecum, which would be missed on flexible sigmoidoscopy. Rarely will endometriosis actually penetrate the bowel wall. Most of the time, wall distortion with what is often seen as a submucosal mass is the only positive finding. Biopsy of the abnormality is typically unsatisfying, with normal colonic mucosa being the most common finding on microscopy.

When there are positive endoscopic findings, we prefer endorectal ultrasonography (ERUS) over transvaginal ultrasound to further evaluate rectal wall involvement of endometriosis. Using a 10-megahertz, 3-D ultrasound, we are able to visualize if endometriosis is attached to the bowel wall and to what degree.

The most important finding is whether or not endometriosis is invading the muscularis propria by the presence or absence of a space between the lesion(s) and the muscularis propria. If a space is visualized, there is a fairly good chance that the endometriosis may be safely scraped off the rectum.

If we see, on the other hand, that the endometriosis is either invading or firmly attached to the rectal muscularis propria, we know that chances of successfully scraping the lesion(s) off the rectum will be very low. In that case, a segmental resection of the rectum can be scheduled in conjunction with the rest of the endometriosis removal. Previous endometriosis surgery, it must be noted, leads to scar tissue which will often distort ERUS images and make the exam less accurate.

Surgical Planning

When the preoperative work-up is confirmatory, a combined surgical approach is scheduled. When the work-up is negative and the patient is scheduled for removal of endometriosis from the nongastrointestinal organs, we remain on surgical standby because bowel involvement of endometriosis is occasionally discovered in symptomatic patients despite a negative work-up.

Standby status also allows the gynecologist to be more aggressive because rectal wall injury can be corrected at the time of surgery. Flexible sigmoidoscopy always should be performed at the end of any endometriosis operation in which bowel proximity is encountered. By submerging the bowel under water and inflating air via the sigmoidoscope, the presence of air bubbles often will identify a missed bowel injury.

Segmental resection usually involves no more than 5-6 cm of the rectum. We do, however, extend the resection a bit proximally if the patient has a history of chronic constipation. Resecting more of the rectum and sigmoid colon to straighten out the left side of the large bowel and shorten the overall length will better alleviate the patient's constipation symptoms. Combined with improvement in the patient's defecation-related symptoms associated with the endometriosis, patient satisfaction regarding the elimination of constipation symptoms is often quite high.

Basic Surgical Technique

Rectal resection for deep endometriosis is comparable to resection of a T4 rectal cancer (one that has invaded outside the rectal wall), except that in the case of endometriosis, we typically are treating young, otherwise healthy patients. In these patients, the risk of complications – mainly, the risk of a permanent colostomy – is all the more concerning. It is important that patients understand the risk and benefits of the surgery and that the colorectal surgeon has the proper expertise for such a technically demanding, risky operation.

The operation is performed in a modified lithotomy position. A laparoscopic or hand-assisted laparoscopic approach can be used. We have performed both techniques, but find a hand-assisted laparoscopic approach faster. A robotic-assisted approach also is being developed.

Depending upon the type of camera used, a 5- or 10-mm port is placed in the umbilicus. The only other ancillary ports needed are 5-mm dissecting ports placed in the right and left lower quadrants. A mini-Pfannenstiel incision is needed to remove the rectal specimen. By extending this incision 3 centimeters, a hand port for hand-assisted laparoscopic surgery can be placed.

Surgery is initiated by the gynecologic surgeon, who resects endometriosis off all nongastrointestinal organs. Endometriosis involving the colon and rectum is left intact. If indicated, hysterectomy with salpingo-oophorectomy is performed at this time.

Next is the laparoscopic colorectal portion of the surgery. First, the inferior mesenteric artery is ligated at the root of the aorta so that various collateral vessels within the marginal branches and Riolan's arch are not sacrificed. Usually, this ligation alone will adequately free up the sigmoid colon enough for a tension-free anastomosis. If the sigmoid colon still cannot be lowered into the rectum without undue tension, we also will ligate the rectal tributary of the inferior mesenteric vein, one of the two main tributaries of the mesenteric vein.

The remainder of the mobilization involves dissecting along the White line of Toldt until the colon falls freely into the rectum. Rarely will we need to mobilize the splenic flexure of the colon to achieve adequate length.

With the left side of the colon freely mobilized, we turn our attention to the pelvis and subsequent rectal dissection. We do not remove the lesion from the rectum, since we have already confirmed that the lesion is firmly attached to the rectal wall. The endometriosis is removed en bloc with the rectum, similar to what is done for rectal cancer.

While the lateral and posterior dissection of the mesorectum can be easily done laparoscopically or robotically, we believe the anterior dissection of the rectum – removal of the endometriosis off the posterior aspect of the vagina while leaving it attached to the rectum – is more easily performed using a hand-assisted laparoscopic approach or even a hybrid open approach through the mini-Pfannenstiel incision.

Dissection is carried out distally until a soft, normal section of rectum is identified. At least 2 cm of normal rectum is needed for a safe anastomosis.

Endometriosis involving the bowel usually appears as a white fibrotic, submucosal mass and feels similar to invasive rectal cancer. The difference, of course, is that rectal cancer is mostly intraluminal, whereas endometriosis usually originates outside the bowel wall and invades inward. Occasionally, one will find “chocolate”-filled cysts within the endometriotic mass, but this is rare.

Endometriosis with bowel involvement is typically anterior to the rectum and posterior to the vagina, but lesions posterior to the rectum have been found, which would denote a nonanatomical spreading distribution.

There are two techniques for a low colorectal anastomosis: The hand-sewn anastomosis technique and the end-to-end anastomosis (EEA) stapler technique. The hand-sewn anastomosis has largely been replaced by the EEA stapler because stapling the proximal colon to the lower rectum is easier to perform, faster, and results in a similar anastomotic leak rate when compared to a hand-sewn anastomosis.

An anastomosis in the upper or mid-rectum has a low risk of leaking (less than 2%). Sometimes, in patients with a deep cul-de-sac, the lesion is attached to the mid-rectum and the anastomosis must be performed in the lower rectum, within a few centimeters of the anus. Low rectal anastomoses have leak rates as high as 10%. A flexible sigmoidoscopy must be performed to check for an air leak. If one is found, the anastomosis should be reconstructed or repaired. Temporary diverting ileostomy should be considered if the anastomosis is suboptimal.

An endorectal ultrasound shows endometriosis invading the muscularis propria (1 o'clock position).

Source Courtesy Dr. John J. Park

Endometriosis invades the rectal wall. The mucosa is distorted, but intact.

The inferior mesenteric artery (left) is ligated at the root of the aorta so that various collateral vessels within the marginal branches and Riolan's arch are not sacrificed. This ligation alone often will adequately free up the sigmoid colon enough for a tension-free anastomosis. If the sigmoid colon still cannot be lowered into the rectum without undue tension, the surgeon also will ligate the rectal tributary of the inferior mesenteric vein (right), one of the two main tributaries of the mesenteric vein.

The colon is being attached to the distal rectum using an end-to-end anastomosis stapler.

Source Images Courtesy Dr. John J. Park

Deeply Invasive Rectosigmoid Endometriosis

www.isge.org

www.aagl.org

Deeply invasive rectosigmoid endometriosis can be associated with a severe – and at times incapacitating – symptom complex. This includes dysmenorrhea – both premenstrual and menstrual – deep dyspareunia, dyschezia, and rectal bleeding at time of menses. There also can be an impact on fertility as well, which can be rectified with bowel resection. In the accompanying graphic (right), a number of studies revealing pregnancy post bowel resection for rectosigmoid endometriosis are noted.

As bowel resection is generally not in the armamentarium of the gynecologic surgeon treating benign disease, the proper treatment of deep infiltrated rectosigmoid endometriosis must involve a cooperative effort with a colorectal surgeon who is capable of performing advanced minimally invasive surgery. This collaboration permits the minimally invasive gynecologist to laparoscopically excise endometriosis, lyse pelvic adhesions, resect ovarian endometriomata, and where indicated, perform ureterolysis and total laparoscopic hysterectomy. The colorectal or general surgeon can then proceed with the bowel resection via a minimally invasive approach.

For this current Master Class in Gynecologic Surgery, I have solicited the expertise of Dr. John J. Park. Dr. Park is a clinical assistant professor of surgery in the division of colorectal surgery at the University of Illinois at Chicago, as well as attending surgeon at Advocate Lutheran General Hospital, Park Ridge, Ill. Dr. Park completed his residency in general surgery at the University of Illinois and his colorectal surgery residency at Mayo Clinic, Rochester, Minn. Dr. Park is board certified in general surgery and colon and rectal surgery.

Deep endometriosis involving the bowel is uncommon, but not rare, among patients with pelvic endometriosis. There is a growing body of literature describing minimally invasive colorectal resection for invasive endometriosis, and a growing feeling among gynecologic surgeons that endometriosis involving the rectal wall is better treated with rectal resection than with a shaving, or scraping, technique.

In our experience of working within a multispecialty surgical team, addressing endometriosis in a systematic fashion has led to a higher rate of patient satisfaction and quality of life, a lower rate of recurrent symptoms, and less surgical morbidity than has been seen with other approaches. While still infrequent, the greatest impact of this multispecialty approach has been the more liberal inclusion of bowel resection as part of the treatment for deep pelvic endometriosis.

It is our belief that patients who have evidence of bowel involvement and have a colorectal resection as part of their endometriosis treatment ultimately have a better surgical outcome. The opposite also appears to hold true. Those patients who have endometriosis involving the bowel wall and who do not receive a bowel resection have been more prone to recurrent disease and symptoms resulting in subsequent reoperations that are often more difficult to perform than a planned colorectal resection would have been.

Furthermore, surgery involving premeditated bowel resection is often safer than treatment employing a scraping technique when endometriosis is deep seated. In cases in which endometriosis has infiltrated the rectal wall, scraping techniques often result in unidentified bowel perforation, which leads to operations that are much more difficult to perform than planned elective resections. In addition, delayed missed injuries often will result in multiple trips to the operating room that frequently include the use of a colostomy in an otherwise young, healthy female patient.

Diagnosing Rectal Involvement

The full extent of disease, including the precise involvement of the bowel, sometimes can be difficult to determine prior to surgery. In many cases, however, physical examination combined with endoscopy and ultrasonography is enough to diagnose bowel involvement that would best be treated with segmental resection.

Pain associated with defecation, rectal bleeding around menses, and constipation often are found on history. On physical exam, endometriosis involving the rectal wall is most commonly associated with pain on both vaginal and rectal examination. Nodularity of the rectal wall, as well as distortion of the rectal folds, often is palpated.

The presence of either nodularity or rectal wall distortion raises the likelihood that there is significant rectal wall involvement. Conversely, if there is minimal or no nodularity of the rectal wall, the likelihood of rectal involvement is low.

When physical examination is abnormal, we proceed with colonoscopy, which enables us to visualize external compression on the rectal wall, nodularity, or other signs of endometriosis infiltrating the bowel wall. Colonoscopy is preferred because endometriosis is often found in the cecum, which would be missed on flexible sigmoidoscopy. Rarely will endometriosis actually penetrate the bowel wall. Most of the time, wall distortion with what is often seen as a submucosal mass is the only positive finding. Biopsy of the abnormality is typically unsatisfying, with normal colonic mucosa being the most common finding on microscopy.

When there are positive endoscopic findings, we prefer endorectal ultrasonography (ERUS) over transvaginal ultrasound to further evaluate rectal wall involvement of endometriosis. Using a 10-megahertz, 3-D ultrasound, we are able to visualize if endometriosis is attached to the bowel wall and to what degree.

The most important finding is whether or not endometriosis is invading the muscularis propria by the presence or absence of a space between the lesion(s) and the muscularis propria. If a space is visualized, there is a fairly good chance that the endometriosis may be safely scraped off the rectum.

If we see, on the other hand, that the endometriosis is either invading or firmly attached to the rectal muscularis propria, we know that chances of successfully scraping the lesion(s) off the rectum will be very low. In that case, a segmental resection of the rectum can be scheduled in conjunction with the rest of the endometriosis removal. Previous endometriosis surgery, it must be noted, leads to scar tissue which will often distort ERUS images and make the exam less accurate.

Surgical Planning

When the preoperative work-up is confirmatory, a combined surgical approach is scheduled. When the work-up is negative and the patient is scheduled for removal of endometriosis from the nongastrointestinal organs, we remain on surgical standby because bowel involvement of endometriosis is occasionally discovered in symptomatic patients despite a negative work-up.

Standby status also allows the gynecologist to be more aggressive because rectal wall injury can be corrected at the time of surgery. Flexible sigmoidoscopy always should be performed at the end of any endometriosis operation in which bowel proximity is encountered. By submerging the bowel under water and inflating air via the sigmoidoscope, the presence of air bubbles often will identify a missed bowel injury.

Segmental resection usually involves no more than 5-6 cm of the rectum. We do, however, extend the resection a bit proximally if the patient has a history of chronic constipation. Resecting more of the rectum and sigmoid colon to straighten out the left side of the large bowel and shorten the overall length will better alleviate the patient's constipation symptoms. Combined with improvement in the patient's defecation-related symptoms associated with the endometriosis, patient satisfaction regarding the elimination of constipation symptoms is often quite high.

Basic Surgical Technique

Rectal resection for deep endometriosis is comparable to resection of a T4 rectal cancer (one that has invaded outside the rectal wall), except that in the case of endometriosis, we typically are treating young, otherwise healthy patients. In these patients, the risk of complications – mainly, the risk of a permanent colostomy – is all the more concerning. It is important that patients understand the risk and benefits of the surgery and that the colorectal surgeon has the proper expertise for such a technically demanding, risky operation.

The operation is performed in a modified lithotomy position. A laparoscopic or hand-assisted laparoscopic approach can be used. We have performed both techniques, but find a hand-assisted laparoscopic approach faster. A robotic-assisted approach also is being developed.

Depending upon the type of camera used, a 5- or 10-mm port is placed in the umbilicus. The only other ancillary ports needed are 5-mm dissecting ports placed in the right and left lower quadrants. A mini-Pfannenstiel incision is needed to remove the rectal specimen. By extending this incision 3 centimeters, a hand port for hand-assisted laparoscopic surgery can be placed.

Surgery is initiated by the gynecologic surgeon, who resects endometriosis off all nongastrointestinal organs. Endometriosis involving the colon and rectum is left intact. If indicated, hysterectomy with salpingo-oophorectomy is performed at this time.

Next is the laparoscopic colorectal portion of the surgery. First, the inferior mesenteric artery is ligated at the root of the aorta so that various collateral vessels within the marginal branches and Riolan's arch are not sacrificed. Usually, this ligation alone will adequately free up the sigmoid colon enough for a tension-free anastomosis. If the sigmoid colon still cannot be lowered into the rectum without undue tension, we also will ligate the rectal tributary of the inferior mesenteric vein, one of the two main tributaries of the mesenteric vein.

The remainder of the mobilization involves dissecting along the White line of Toldt until the colon falls freely into the rectum. Rarely will we need to mobilize the splenic flexure of the colon to achieve adequate length.

With the left side of the colon freely mobilized, we turn our attention to the pelvis and subsequent rectal dissection. We do not remove the lesion from the rectum, since we have already confirmed that the lesion is firmly attached to the rectal wall. The endometriosis is removed en bloc with the rectum, similar to what is done for rectal cancer.

While the lateral and posterior dissection of the mesorectum can be easily done laparoscopically or robotically, we believe the anterior dissection of the rectum – removal of the endometriosis off the posterior aspect of the vagina while leaving it attached to the rectum – is more easily performed using a hand-assisted laparoscopic approach or even a hybrid open approach through the mini-Pfannenstiel incision.

Dissection is carried out distally until a soft, normal section of rectum is identified. At least 2 cm of normal rectum is needed for a safe anastomosis.

Endometriosis involving the bowel usually appears as a white fibrotic, submucosal mass and feels similar to invasive rectal cancer. The difference, of course, is that rectal cancer is mostly intraluminal, whereas endometriosis usually originates outside the bowel wall and invades inward. Occasionally, one will find “chocolate”-filled cysts within the endometriotic mass, but this is rare.

Endometriosis with bowel involvement is typically anterior to the rectum and posterior to the vagina, but lesions posterior to the rectum have been found, which would denote a nonanatomical spreading distribution.

There are two techniques for a low colorectal anastomosis: The hand-sewn anastomosis technique and the end-to-end anastomosis (EEA) stapler technique. The hand-sewn anastomosis has largely been replaced by the EEA stapler because stapling the proximal colon to the lower rectum is easier to perform, faster, and results in a similar anastomotic leak rate when compared to a hand-sewn anastomosis.

An anastomosis in the upper or mid-rectum has a low risk of leaking (less than 2%). Sometimes, in patients with a deep cul-de-sac, the lesion is attached to the mid-rectum and the anastomosis must be performed in the lower rectum, within a few centimeters of the anus. Low rectal anastomoses have leak rates as high as 10%. A flexible sigmoidoscopy must be performed to check for an air leak. If one is found, the anastomosis should be reconstructed or repaired. Temporary diverting ileostomy should be considered if the anastomosis is suboptimal.

An endorectal ultrasound shows endometriosis invading the muscularis propria (1 o'clock position).

Source Courtesy Dr. John J. Park

Endometriosis invades the rectal wall. The mucosa is distorted, but intact.

The inferior mesenteric artery (left) is ligated at the root of the aorta so that various collateral vessels within the marginal branches and Riolan's arch are not sacrificed. This ligation alone often will adequately free up the sigmoid colon enough for a tension-free anastomosis. If the sigmoid colon still cannot be lowered into the rectum without undue tension, the surgeon also will ligate the rectal tributary of the inferior mesenteric vein (right), one of the two main tributaries of the mesenteric vein.

The colon is being attached to the distal rectum using an end-to-end anastomosis stapler.

Source Images Courtesy Dr. John J. Park

Deeply Invasive Rectosigmoid Endometriosis

www.isge.org

www.aagl.org

Deeply invasive rectosigmoid endometriosis can be associated with a severe – and at times incapacitating – symptom complex. This includes dysmenorrhea – both premenstrual and menstrual – deep dyspareunia, dyschezia, and rectal bleeding at time of menses. There also can be an impact on fertility as well, which can be rectified with bowel resection. In the accompanying graphic (right), a number of studies revealing pregnancy post bowel resection for rectosigmoid endometriosis are noted.

As bowel resection is generally not in the armamentarium of the gynecologic surgeon treating benign disease, the proper treatment of deep infiltrated rectosigmoid endometriosis must involve a cooperative effort with a colorectal surgeon who is capable of performing advanced minimally invasive surgery. This collaboration permits the minimally invasive gynecologist to laparoscopically excise endometriosis, lyse pelvic adhesions, resect ovarian endometriomata, and where indicated, perform ureterolysis and total laparoscopic hysterectomy. The colorectal or general surgeon can then proceed with the bowel resection via a minimally invasive approach.

For this current Master Class in Gynecologic Surgery, I have solicited the expertise of Dr. John J. Park. Dr. Park is a clinical assistant professor of surgery in the division of colorectal surgery at the University of Illinois at Chicago, as well as attending surgeon at Advocate Lutheran General Hospital, Park Ridge, Ill. Dr. Park completed his residency in general surgery at the University of Illinois and his colorectal surgery residency at Mayo Clinic, Rochester, Minn. Dr. Park is board certified in general surgery and colon and rectal surgery.

While infant deaths associated with congenital heart defects have declined substantially over the past 2 decades, congenital heart disease remains the most common fatal congenital anomaly in the first year after birth.

Prematurity is the most significant cause of death in the first week of life, but after that point, birth defects take over as the leading cause of infant mortality (and overall, birth defects are the leading cause of infant mortality). Cardiovascular defects, in turn, are the single largest contributor to infant mortality attributable to birth defects – severe congenital heart disease (CHD) affects approximately 0.5% of all neonates and is responsible for one-third of deaths between birth and 1 year of life.

Prenatal diagnosis of CHD is important because early detection can improve the planning of services and provision of coordinated multidisciplinary care. While most fetal therapy for CHD is investigational and still evolving, studies from all over the world have shown that if a baby is known to have a heart problem and is delivered at a facility that provides definitive care, the baby will likely fare better.

Unfortunately, most infants born with CHD do not have defined risk factors. As obstetricians we must always be alert to the possibility that, even without a clear risk factor, there could be a cardiovascular problem.

It is important to know, on the other hand, who is at increased risk and should be evaluated further, and who is not at increased risk. We know now that certain infants whom we haven't traditionally thought of as being at risk for CHD are indeed at higher risk. Our knowledge of familial contributions to CHD has grown, for instance, giving obstetricians the responsibility to be alert for potential familial genetic patterns so that the proper counseling can be provided.

There are also noninherited risk factors that can be identified and potentially modified. It is unclear what proportion of CHD can be prevented, but at the least, obstetricians should be aware of such risk factors so they can provide guidance to parents and future parents that could reduce the risk of their children having a major cardiovascular malformation, and so they can ensure proper surveillance in any pregnancy.

Familial Risks

Over the past 15 years or so, our understanding of inherited causes of congenital heart defects has increased significantly, and while there is much more to learn, it is now appreciated that genetics plays a greater role in CHD than previously estimated.

Molecular genetics studies in families with multiple affected individuals have even led to the identification of specific genetic abnormalities for several forms of CHD, such as the single gene mutation sometimes seen in tetralogy of Fallot; others are related to mutations in more than one gene.

While most chromosome defects are not inherited, some anomalies or syndromes with cardiac phenotypes – for instance, those involving microdeletions or gonadal mosaicism – can be inherited and play a small but increasingly appreciated role in CHD. The William-Beuren syndrome and the 22q11.2 deletion syndrome, for instance, are microdeletion syndromes that show autosomal dominant inheritance.

Overall, parents of a child with CHD that is not associated with a typical chromosomal aberration have a 2%–3% chance of having another child with CHD; it is estimated that half of affected siblings will be diagnosed with the same lesion, the other half with a different lesion.

Classic Mendelian transmission is occasionally responsible for inherited CHD in families, but recurrence risk is significant only when the family history of CHD involves first- or second-degree relatives.

Fetal echocardiography is definitely warranted when the mother or father – or a sibling – of the fetus has CHD, as well as when CHD has affected a parent's own mother, father, sister, or brother.

Once you get further out in the family history, to cousins and other third-degree and more distant relatives, the risk is not high enough to warrant a more detailed fetal examination. This is important for counseling; parents who are worried about a history of CHD in third-degree relatives should be reassured.

Among the changes in patterns of referral for fetal echocardiography that we detected at Yale-New Haven Hospital from 1985 to 2003 was an 18% increase in referrals for a family history of CHD, including family history in more distant relatives.

This increase was not accompanied by any change in the percentage of structural cardiovascular heart defects consequently detected (J. Ultrasound Med. 2006;25:197-202).

As an increasing number of patients with major congenital cardiac defects have been surviving to adulthood and parenthood, numerous investigators have attempted to identify specific recurrence risks.

One study done in the United Kingdom, for instance, identified 727 adults with surgically modified major heart defects and their 393 live offspring. Of these infants, 16 were born with cardiac malformations, representing a total recurrence risk of 4.1%. Recurrence risk in offspring ranged from 3.1% for tetralogy of Fallot to 7.8% for atrioventricular septal defect (Lancet 1998;351:311-6).

CHD occurred more often in offspring of affected mothers (5.7%) than affected fathers (2.2%). Compared with offspring, sibling risk was significantly lower: 2.1% overall.

A much larger and more recent study done in Denmark again showed strong familial clustering in first-degree relatives for CHDs, particularly for recurrences of the same heart defect. (Very few families experience a second heart defect, the study found.)

The study – a national cohort study of more than 1.7 million people born during a 29-year period – also is one of the largest studies, if not the largest study, to document a decreasing risk as family history gets more distant.

The relative risks of any CHD in singletons were 3.21 for a family history of any CHD in first-degree relatives, 1.78 for a family history involving second-degree relatives, and 1.10 for a family history in third-degree relatives (Circulation 2009;120:295-301).

Only with a history of affected first- and second-degree relatives was there a statistically greater chance of having an affected fetus.

In twins, the relative risks of any CHD were 12.5 for same-sex twins and 6.93 for twins of both sexes, the Danish investigators noted.

And in looking at the contribution of CHD family history to the total number of CHD cases in the population, they found that 2.2% of heart defect cases in the population were attributed to CHD family history in first-degree relatives.

Another notable finding from other studies is that women with cyanotic heart disease have a higher risk of having a baby with CHD than do women with noncyanotic heart disease.

Maternal Risks

Just as we've learned much about inherited causes of congenital heart disease over the past 15 years, there is a growing body of epidemiologic literature on potential fetal exposures – from maternal illnesses to maternal drug exposures – that can alter the risk of CHD.

The risk factors for CHD maternal teratogen exposure are numerous. They include lithium, alcohol, isotretinoin, and various anticonvulsant drugs, and many are well-appreciated by ob.gyns.

Other factors for which risk has been well determined, and can be better appreciated, include:

▸ High vitamin A intake. Findings are not completely consistent, but we have enough data now to suggest that women who take extra-large doses of vitamin A may actually be putting their fetuses at risk of birth defects.

One study worth noting found that among more than 22,000 pregnant women, those who took more than 10,000 IU of vitamin A from supplements were 4.8 times more likely to have babies with birth defects associated with cranial-neural-crest tissue than were women who consumed 5,000 IU or less per day (N. Engl. J. Med. 1995;333:1369-73).

Typical prenatal vitamins have 5,000 IU in each dose. This is one reason that women with twin pregnancies can take extra folic acid, but should not double up on their prenatal multivitamins.

▸ Folate antagonists. Common drugs such as trimethoprim, triamterene, sulfasalazine, phenytoin, phenobarbital, primidone, carbamazepine, and cholestyramine may increase the risk not only of neural-tube defects, but of cardiovascular defects as well, in addition to oral clefts and urinary tract defects.

Fortunately, studies such as one published in 2000 involving thousands of infants show that the folic acid component of prenatal multivitamin supplements can reduce the risks of these defects, just as it reduces the risk of neural-tube defects (N. Engl. J. Med. 2000;343:1608-14).

▸ Paxil (paroxetine). This is the only antidepressant that has been shown in some studies to increase the risk of CHD. Its manufacturer, GlaxoSmithKline, changed the label's pregnancy precaution in 2005 from a Pregnancy Category C to Category D. If a patient becomes pregnant while taking the drug, she should be advised of potential harm to the fetus.

One epidemiologic study showed that women taking Paxil were two times more likely to have an infant with CHD, and 2.2 times more likely to have an infant with any congenital malformation, than were women taking other antidepressants.

▸ Diabetes. The risk of fetal anomalies with maternal diabetes and elevated hemoglobin A_1c in early pregnancy has been known for some time.

In a study published in 1981, for instance, the risk of CHD and other fetal anomalies rose from 5% to 22% as maternal HbA_1c rose from a range of 7%–8.5% to greater than 8.5% (N. Engl. J. Med. 1981;304:1331-4).

We've also known for some time that differences in CHD may exist even with good metabolic control. Studies have documented mild cardiac hypertrophy involving the interventricular septum and the ventricular free walls, for instance, in diabetic mothers with good metabolic control (J. Pediatr. 1991;118:103-7 and Am J. Obstet. Gynecol. 1991;164:837-43). Such growth affects cardiac diastolic function.

With the epidemic of obesity and the increasing prevalence of early type 2 diabetes and glucose intolerance among women of childbearing age, however, this is an increasingly important risk factor to appreciate and counsel about.

The most important message, we've learned, is that there's no such thing as perfect control – that good metabolic control will not necessarily protect diabetic mothers from the higher risk of CHD.

Just as detection and appropriate management of diabetes before and during pregnancy are of utmost importance, so is fetal echocardiography for every pregnant woman who has pregestational diabetes – even diabetes that is well controlled.

Indeed, the same review of all fetal echocardiography performed between 1985 and 2003 at Yale-New Haven Hospital that showed an increase in referrals for family history also showed a 9% increase in the proportion of studies done for pregestational diabetes as the indication. The increase was most striking when it came to women who had recently been diagnosed, compared with long-standing diabetes – a finding that likely reflects the increase in obesity.

▸ Phenylketonuria. Fortunately, strict dietary control before conception and during pregnancy can reduce the increased risk of heart defects faced by women with this disorder. We need to remember that aspartame (NutraSweet) can cause phenylalanine levels to increase in women with PKU, but not in normal women. Women without PKU can be reassured that there is no evidence linking aspartame with birth defects.

Fetal Risks

Among the fetal risk factors important to consider are:

▸ Extracardiac anomalies. The identification of any extracardiac anomaly should raise our level of suspicion for other anomalies, including congenital heart defects. If we see one anomaly – anywhere in the fetus – there often are really two. And if we see two anomalies, there frequently are really three.

▸ Nonimmune hydrops. All fetuses found to have NIH should be evaluated with fetal echocardiography. Structural heart disease in fetuses with NIH is usually indicative of a poor prognosis for survival, but when rhythm disturbances/arrhythmias are detected in association with NIH, there is sometimes an option for prenatal treatment.

▸ Fetal arrhythmias. An irregular heartbeat is usually not a problem, but tachycardia and especially bradycardia are associated with an increased risk of CHD. There may be structural heart defects in as many as half of fetuses with fixed bradycardia (i.e., baseline heart rate less than 100). In general, it is best that all arrhythmias are examined; it is just too hard to tell them apart by auscultation alone.

▸ Nuchal translucency. Numerous studies have shown that elevated first-trimester nuchal translucency (NT) increases the risk of major congenital heart defects in chromosomally normal fetuses, and that risk increases with increasing NT measurement.

In a large prospective multicenter study conducted by the National Institute of Child Health and Human Development, for instance, investigators identified 21 cases of major congenital heart defects in 8,167 chromosomally normal pregnancies. They reported that the incidence of CHD per 1,000 pregnancies rose from 1.9 with an NT measurement of less than 2.0 mm, to 4.8 with an NT measurement of 2.0–2.4 mm, to 6.0 with an NT measurement of 2.5–3.4 mm, to 23 of every 1,000 pregnancies with an NT measurement of 3.5 mm or greater (Am. J. Obstet. Gynecol. 2005;192:1357–61).

If the NT is greater than 3.5 mm, measured by a qualified sonographer or sonologist at 11–14 weeks as part of an aneuploidy risk assessment scan, the patient should be referred for fetal echocardiography.

▸ In vitro fertilization. We recently investigated the prevalence of congenital heart defects among IVF pregnancies at our referral program at Yale, and found that children conceived through IVF were 3–12 times as likely to have CHD as was the general population (J. Ultrasound Med. 2010;29:917-22).

Similar data have come from Australia and Europe, with reported odds ratios for IVF versus natural conception of 3–4. I tell patients, therefore, that it's not just one place or one study suggesting risk. Indeed, it's a meaningful risk factor.

▸ Monochorionic twins. In a systemic literature review we conducted several years ago that included 40 fetuses with CHDs among 830 fetuses from monochorionic/diamniotic twin gestations, the rate of CHDs in these twin gestations was significantly higher than the prevalence rate of CHDs in the general population (J. Ultrasound Med. 2007;26:1491-8).

Congenital heart defects were almost three times as likely to complicate the monochorionic/diamniotic twin gestations affected by twin-to-twin transfusion syndrome (TTTS), compared with those without TTTS, but an increase occurred regardless of the presence of TTTS. Ventricular septal defects were among the most frequent heart defects. Fetal echocardiography may be considered for all monochorionic/diamniotic twin gestations.

Dr. Copel disclosed that he has received research support from Philips Healthcare and Siemens Healthcare. Both companies manufacture echocardiography and other ultrasound systems.

A two-dimensional four-chamber view of a normal fetal heart (left). Fetal image of a complete atrioventricular septal defect with large atrial (**) and ventricular (*) septal defects (right).

Source Images: © Elsevier Inc.

Diagnosing Birth Defects

Birth defects continue to account for the majority of infant deaths, and their biologic basis continues to present a mixed picture, with the majority of causes still unknown. Cardiac defects – the most common type of birth defect – result in varying types of morbidity, but remain the most severe and disabling of all birth defects. As our guest author points out below, cardiovascular defects are the single largest contributor to birth defect–attributable infant mortality.

What is clear is the fact that when birth defects are identified prenatally, decisions can be made regarding the timing and route of delivery and even the facility where delivery occurs. We know that such decision making can be highly influential on the ultimate outcome of the infant.

Fortunately, there has been improvement in recent years in diagnostic technology that enables more prenatal diagnosis of congenital heart disease, and certain conditions that in the past went unknown or undiagnosed are now being identified early so that specialists can intervene in a timely manner.

While certain pregnancies are clearly at higher risk – those involving mothers who have pregestational diabetes, for instance, or mothers with exposure to particular toxins – there are other scenarios and factors that increase risk of which we should be aware.

It's a tricky evaluative process, for, as our guest author points out, most infants born with congenital heart disease do not have defined risk factors. At the least, however, we can be aware of the familial, maternal, and fetal factors that are known to increase risk and then ensure that all at-risk pregnancies are properly evaluated – often with fetal echocardiography – to determine if a cardiac defect is present and, if so, to plan the delivery-related issues of timing, mode, and facility.

In light of the importance of this subject and the role that ultrasound scanning, genetic counseling, and early decision making and planning can play in the ultimate outcome of the fetus, we decided to do a Master Class on congenital heart defects. We have invited Dr. Joshua A. Copel, professor of obstetrics, gynecology, and reproductive services, and of pediatrics, and vice chair of obstetrics at Yale University, New Haven, Conn., to serve as our guest professor.

Dr. Copel has written and lectured extensively on fetal arrhythmias, fetal cardiac anomalies and congenital heart disease, and sonographic monitoring and fetal echocardiography. Here he discusses what we should know about both familial contributions to congenital heart disease and various noninherited risk factors.

While infant deaths associated with congenital heart defects have declined substantially over the past 2 decades, congenital heart disease remains the most common fatal congenital anomaly in the first year after birth.

Prematurity is the most significant cause of death in the first week of life, but after that point, birth defects take over as the leading cause of infant mortality (and overall, birth defects are the leading cause of infant mortality). Cardiovascular defects, in turn, are the single largest contributor to infant mortality attributable to birth defects – severe congenital heart disease (CHD) affects approximately 0.5% of all neonates and is responsible for one-third of deaths between birth and 1 year of life.

Prenatal diagnosis of CHD is important because early detection can improve the planning of services and provision of coordinated multidisciplinary care. While most fetal therapy for CHD is investigational and still evolving, studies from all over the world have shown that if a baby is known to have a heart problem and is delivered at a facility that provides definitive care, the baby will likely fare better.

Unfortunately, most infants born with CHD do not have defined risk factors. As obstetricians we must always be alert to the possibility that, even without a clear risk factor, there could be a cardiovascular problem.

It is important to know, on the other hand, who is at increased risk and should be evaluated further, and who is not at increased risk. We know now that certain infants whom we haven't traditionally thought of as being at risk for CHD are indeed at higher risk. Our knowledge of familial contributions to CHD has grown, for instance, giving obstetricians the responsibility to be alert for potential familial genetic patterns so that the proper counseling can be provided.

There are also noninherited risk factors that can be identified and potentially modified. It is unclear what proportion of CHD can be prevented, but at the least, obstetricians should be aware of such risk factors so they can provide guidance to parents and future parents that could reduce the risk of their children having a major cardiovascular malformation, and so they can ensure proper surveillance in any pregnancy.

Familial Risks

Over the past 15 years or so, our understanding of inherited causes of congenital heart defects has increased significantly, and while there is much more to learn, it is now appreciated that genetics plays a greater role in CHD than previously estimated.

Molecular genetics studies in families with multiple affected individuals have even led to the identification of specific genetic abnormalities for several forms of CHD, such as the single gene mutation sometimes seen in tetralogy of Fallot; others are related to mutations in more than one gene.

While most chromosome defects are not inherited, some anomalies or syndromes with cardiac phenotypes – for instance, those involving microdeletions or gonadal mosaicism – can be inherited and play a small but increasingly appreciated role in CHD. The William-Beuren syndrome and the 22q11.2 deletion syndrome, for instance, are microdeletion syndromes that show autosomal dominant inheritance.

Overall, parents of a child with CHD that is not associated with a typical chromosomal aberration have a 2%–3% chance of having another child with CHD; it is estimated that half of affected siblings will be diagnosed with the same lesion, the other half with a different lesion.

Classic Mendelian transmission is occasionally responsible for inherited CHD in families, but recurrence risk is significant only when the family history of CHD involves first- or second-degree relatives.

Fetal echocardiography is definitely warranted when the mother or father – or a sibling – of the fetus has CHD, as well as when CHD has affected a parent's own mother, father, sister, or brother.

Once you get further out in the family history, to cousins and other third-degree and more distant relatives, the risk is not high enough to warrant a more detailed fetal examination. This is important for counseling; parents who are worried about a history of CHD in third-degree relatives should be reassured.

Among the changes in patterns of referral for fetal echocardiography that we detected at Yale-New Haven Hospital from 1985 to 2003 was an 18% increase in referrals for a family history of CHD, including family history in more distant relatives.

This increase was not accompanied by any change in the percentage of structural cardiovascular heart defects consequently detected (J. Ultrasound Med. 2006;25:197-202).

As an increasing number of patients with major congenital cardiac defects have been surviving to adulthood and parenthood, numerous investigators have attempted to identify specific recurrence risks.

One study done in the United Kingdom, for instance, identified 727 adults with surgically modified major heart defects and their 393 live offspring. Of these infants, 16 were born with cardiac malformations, representing a total recurrence risk of 4.1%. Recurrence risk in offspring ranged from 3.1% for tetralogy of Fallot to 7.8% for atrioventricular septal defect (Lancet 1998;351:311-6).

CHD occurred more often in offspring of affected mothers (5.7%) than affected fathers (2.2%). Compared with offspring, sibling risk was significantly lower: 2.1% overall.

A much larger and more recent study done in Denmark again showed strong familial clustering in first-degree relatives for CHDs, particularly for recurrences of the same heart defect. (Very few families experience a second heart defect, the study found.)

The study – a national cohort study of more than 1.7 million people born during a 29-year period – also is one of the largest studies, if not the largest study, to document a decreasing risk as family history gets more distant.

The relative risks of any CHD in singletons were 3.21 for a family history of any CHD in first-degree relatives, 1.78 for a family history involving second-degree relatives, and 1.10 for a family history in third-degree relatives (Circulation 2009;120:295-301).

Only with a history of affected first- and second-degree relatives was there a statistically greater chance of having an affected fetus.

In twins, the relative risks of any CHD were 12.5 for same-sex twins and 6.93 for twins of both sexes, the Danish investigators noted.

And in looking at the contribution of CHD family history to the total number of CHD cases in the population, they found that 2.2% of heart defect cases in the population were attributed to CHD family history in first-degree relatives.

Another notable finding from other studies is that women with cyanotic heart disease have a higher risk of having a baby with CHD than do women with noncyanotic heart disease.

Maternal Risks

Just as we've learned much about inherited causes of congenital heart disease over the past 15 years, there is a growing body of epidemiologic literature on potential fetal exposures – from maternal illnesses to maternal drug exposures – that can alter the risk of CHD.

The risk factors for CHD maternal teratogen exposure are numerous. They include lithium, alcohol, isotretinoin, and various anticonvulsant drugs, and many are well-appreciated by ob.gyns.

Other factors for which risk has been well determined, and can be better appreciated, include:

▸ High vitamin A intake. Findings are not completely consistent, but we have enough data now to suggest that women who take extra-large doses of vitamin A may actually be putting their fetuses at risk of birth defects.

One study worth noting found that among more than 22,000 pregnant women, those who took more than 10,000 IU of vitamin A from supplements were 4.8 times more likely to have babies with birth defects associated with cranial-neural-crest tissue than were women who consumed 5,000 IU or less per day (N. Engl. J. Med. 1995;333:1369-73).

Typical prenatal vitamins have 5,000 IU in each dose. This is one reason that women with twin pregnancies can take extra folic acid, but should not double up on their prenatal multivitamins.

▸ Folate antagonists. Common drugs such as trimethoprim, triamterene, sulfasalazine, phenytoin, phenobarbital, primidone, carbamazepine, and cholestyramine may increase the risk not only of neural-tube defects, but of cardiovascular defects as well, in addition to oral clefts and urinary tract defects.

Fortunately, studies such as one published in 2000 involving thousands of infants show that the folic acid component of prenatal multivitamin supplements can reduce the risks of these defects, just as it reduces the risk of neural-tube defects (N. Engl. J. Med. 2000;343:1608-14).

▸ Paxil (paroxetine). This is the only antidepressant that has been shown in some studies to increase the risk of CHD. Its manufacturer, GlaxoSmithKline, changed the label's pregnancy precaution in 2005 from a Pregnancy Category C to Category D. If a patient becomes pregnant while taking the drug, she should be advised of potential harm to the fetus.

One epidemiologic study showed that women taking Paxil were two times more likely to have an infant with CHD, and 2.2 times more likely to have an infant with any congenital malformation, than were women taking other antidepressants.

▸ Diabetes. The risk of fetal anomalies with maternal diabetes and elevated hemoglobin A_1c in early pregnancy has been known for some time.

In a study published in 1981, for instance, the risk of CHD and other fetal anomalies rose from 5% to 22% as maternal HbA_1c rose from a range of 7%–8.5% to greater than 8.5% (N. Engl. J. Med. 1981;304:1331-4).

We've also known for some time that differences in CHD may exist even with good metabolic control. Studies have documented mild cardiac hypertrophy involving the interventricular septum and the ventricular free walls, for instance, in diabetic mothers with good metabolic control (J. Pediatr. 1991;118:103-7 and Am J. Obstet. Gynecol. 1991;164:837-43). Such growth affects cardiac diastolic function.

With the epidemic of obesity and the increasing prevalence of early type 2 diabetes and glucose intolerance among women of childbearing age, however, this is an increasingly important risk factor to appreciate and counsel about.

The most important message, we've learned, is that there's no such thing as perfect control – that good metabolic control will not necessarily protect diabetic mothers from the higher risk of CHD.

Just as detection and appropriate management of diabetes before and during pregnancy are of utmost importance, so is fetal echocardiography for every pregnant woman who has pregestational diabetes – even diabetes that is well controlled.

Indeed, the same review of all fetal echocardiography performed between 1985 and 2003 at Yale-New Haven Hospital that showed an increase in referrals for family history also showed a 9% increase in the proportion of studies done for pregestational diabetes as the indication. The increase was most striking when it came to women who had recently been diagnosed, compared with long-standing diabetes – a finding that likely reflects the increase in obesity.

▸ Phenylketonuria. Fortunately, strict dietary control before conception and during pregnancy can reduce the increased risk of heart defects faced by women with this disorder. We need to remember that aspartame (NutraSweet) can cause phenylalanine levels to increase in women with PKU, but not in normal women. Women without PKU can be reassured that there is no evidence linking aspartame with birth defects.

Fetal Risks

Among the fetal risk factors important to consider are:

▸ Extracardiac anomalies. The identification of any extracardiac anomaly should raise our level of suspicion for other anomalies, including congenital heart defects. If we see one anomaly – anywhere in the fetus – there often are really two. And if we see two anomalies, there frequently are really three.

▸ Nonimmune hydrops. All fetuses found to have NIH should be evaluated with fetal echocardiography. Structural heart disease in fetuses with NIH is usually indicative of a poor prognosis for survival, but when rhythm disturbances/arrhythmias are detected in association with NIH, there is sometimes an option for prenatal treatment.

▸ Fetal arrhythmias. An irregular heartbeat is usually not a problem, but tachycardia and especially bradycardia are associated with an increased risk of CHD. There may be structural heart defects in as many as half of fetuses with fixed bradycardia (i.e., baseline heart rate less than 100). In general, it is best that all arrhythmias are examined; it is just too hard to tell them apart by auscultation alone.

▸ Nuchal translucency. Numerous studies have shown that elevated first-trimester nuchal translucency (NT) increases the risk of major congenital heart defects in chromosomally normal fetuses, and that risk increases with increasing NT measurement.

In a large prospective multicenter study conducted by the National Institute of Child Health and Human Development, for instance, investigators identified 21 cases of major congenital heart defects in 8,167 chromosomally normal pregnancies. They reported that the incidence of CHD per 1,000 pregnancies rose from 1.9 with an NT measurement of less than 2.0 mm, to 4.8 with an NT measurement of 2.0–2.4 mm, to 6.0 with an NT measurement of 2.5–3.4 mm, to 23 of every 1,000 pregnancies with an NT measurement of 3.5 mm or greater (Am. J. Obstet. Gynecol. 2005;192:1357–61).

If the NT is greater than 3.5 mm, measured by a qualified sonographer or sonologist at 11–14 weeks as part of an aneuploidy risk assessment scan, the patient should be referred for fetal echocardiography.

▸ In vitro fertilization. We recently investigated the prevalence of congenital heart defects among IVF pregnancies at our referral program at Yale, and found that children conceived through IVF were 3–12 times as likely to have CHD as was the general population (J. Ultrasound Med. 2010;29:917-22).

Similar data have come from Australia and Europe, with reported odds ratios for IVF versus natural conception of 3–4. I tell patients, therefore, that it's not just one place or one study suggesting risk. Indeed, it's a meaningful risk factor.

▸ Monochorionic twins. In a systemic literature review we conducted several years ago that included 40 fetuses with CHDs among 830 fetuses from monochorionic/diamniotic twin gestations, the rate of CHDs in these twin gestations was significantly higher than the prevalence rate of CHDs in the general population (J. Ultrasound Med. 2007;26:1491-8).

Congenital heart defects were almost three times as likely to complicate the monochorionic/diamniotic twin gestations affected by twin-to-twin transfusion syndrome (TTTS), compared with those without TTTS, but an increase occurred regardless of the presence of TTTS. Ventricular septal defects were among the most frequent heart defects. Fetal echocardiography may be considered for all monochorionic/diamniotic twin gestations.

Dr. Copel disclosed that he has received research support from Philips Healthcare and Siemens Healthcare. Both companies manufacture echocardiography and other ultrasound systems.

A two-dimensional four-chamber view of a normal fetal heart (left). Fetal image of a complete atrioventricular septal defect with large atrial (**) and ventricular (*) septal defects (right).

Source Images: © Elsevier Inc.

Diagnosing Birth Defects

Birth defects continue to account for the majority of infant deaths, and their biologic basis continues to present a mixed picture, with the majority of causes still unknown. Cardiac defects – the most common type of birth defect – result in varying types of morbidity, but remain the most severe and disabling of all birth defects. As our guest author points out below, cardiovascular defects are the single largest contributor to birth defect–attributable infant mortality.

What is clear is the fact that when birth defects are identified prenatally, decisions can be made regarding the timing and route of delivery and even the facility where delivery occurs. We know that such decision making can be highly influential on the ultimate outcome of the infant.

Fortunately, there has been improvement in recent years in diagnostic technology that enables more prenatal diagnosis of congenital heart disease, and certain conditions that in the past went unknown or undiagnosed are now being identified early so that specialists can intervene in a timely manner.

While certain pregnancies are clearly at higher risk – those involving mothers who have pregestational diabetes, for instance, or mothers with exposure to particular toxins – there are other scenarios and factors that increase risk of which we should be aware.

It's a tricky evaluative process, for, as our guest author points out, most infants born with congenital heart disease do not have defined risk factors. At the least, however, we can be aware of the familial, maternal, and fetal factors that are known to increase risk and then ensure that all at-risk pregnancies are properly evaluated – often with fetal echocardiography – to determine if a cardiac defect is present and, if so, to plan the delivery-related issues of timing, mode, and facility.

In light of the importance of this subject and the role that ultrasound scanning, genetic counseling, and early decision making and planning can play in the ultimate outcome of the fetus, we decided to do a Master Class on congenital heart defects. We have invited Dr. Joshua A. Copel, professor of obstetrics, gynecology, and reproductive services, and of pediatrics, and vice chair of obstetrics at Yale University, New Haven, Conn., to serve as our guest professor.

Dr. Copel has written and lectured extensively on fetal arrhythmias, fetal cardiac anomalies and congenital heart disease, and sonographic monitoring and fetal echocardiography. Here he discusses what we should know about both familial contributions to congenital heart disease and various noninherited risk factors.

While infant deaths associated with congenital heart defects have declined substantially over the past 2 decades, congenital heart disease remains the most common fatal congenital anomaly in the first year after birth.

Prematurity is the most significant cause of death in the first week of life, but after that point, birth defects take over as the leading cause of infant mortality (and overall, birth defects are the leading cause of infant mortality). Cardiovascular defects, in turn, are the single largest contributor to infant mortality attributable to birth defects – severe congenital heart disease (CHD) affects approximately 0.5% of all neonates and is responsible for one-third of deaths between birth and 1 year of life.

Prenatal diagnosis of CHD is important because early detection can improve the planning of services and provision of coordinated multidisciplinary care. While most fetal therapy for CHD is investigational and still evolving, studies from all over the world have shown that if a baby is known to have a heart problem and is delivered at a facility that provides definitive care, the baby will likely fare better.

Unfortunately, most infants born with CHD do not have defined risk factors. As obstetricians we must always be alert to the possibility that, even without a clear risk factor, there could be a cardiovascular problem.

It is important to know, on the other hand, who is at increased risk and should be evaluated further, and who is not at increased risk. We know now that certain infants whom we haven't traditionally thought of as being at risk for CHD are indeed at higher risk. Our knowledge of familial contributions to CHD has grown, for instance, giving obstetricians the responsibility to be alert for potential familial genetic patterns so that the proper counseling can be provided.

There are also noninherited risk factors that can be identified and potentially modified. It is unclear what proportion of CHD can be prevented, but at the least, obstetricians should be aware of such risk factors so they can provide guidance to parents and future parents that could reduce the risk of their children having a major cardiovascular malformation, and so they can ensure proper surveillance in any pregnancy.

Familial Risks

Over the past 15 years or so, our understanding of inherited causes of congenital heart defects has increased significantly, and while there is much more to learn, it is now appreciated that genetics plays a greater role in CHD than previously estimated.

Molecular genetics studies in families with multiple affected individuals have even led to the identification of specific genetic abnormalities for several forms of CHD, such as the single gene mutation sometimes seen in tetralogy of Fallot; others are related to mutations in more than one gene.

While most chromosome defects are not inherited, some anomalies or syndromes with cardiac phenotypes – for instance, those involving microdeletions or gonadal mosaicism – can be inherited and play a small but increasingly appreciated role in CHD. The William-Beuren syndrome and the 22q11.2 deletion syndrome, for instance, are microdeletion syndromes that show autosomal dominant inheritance.

Overall, parents of a child with CHD that is not associated with a typical chromosomal aberration have a 2%–3% chance of having another child with CHD; it is estimated that half of affected siblings will be diagnosed with the same lesion, the other half with a different lesion.

Classic Mendelian transmission is occasionally responsible for inherited CHD in families, but recurrence risk is significant only when the family history of CHD involves first- or second-degree relatives.

Fetal echocardiography is definitely warranted when the mother or father – or a sibling – of the fetus has CHD, as well as when CHD has affected a parent's own mother, father, sister, or brother.

Once you get further out in the family history, to cousins and other third-degree and more distant relatives, the risk is not high enough to warrant a more detailed fetal examination. This is important for counseling; parents who are worried about a history of CHD in third-degree relatives should be reassured.

Among the changes in patterns of referral for fetal echocardiography that we detected at Yale-New Haven Hospital from 1985 to 2003 was an 18% increase in referrals for a family history of CHD, including family history in more distant relatives.

This increase was not accompanied by any change in the percentage of structural cardiovascular heart defects consequently detected (J. Ultrasound Med. 2006;25:197-202).

As an increasing number of patients with major congenital cardiac defects have been surviving to adulthood and parenthood, numerous investigators have attempted to identify specific recurrence risks.

One study done in the United Kingdom, for instance, identified 727 adults with surgically modified major heart defects and their 393 live offspring. Of these infants, 16 were born with cardiac malformations, representing a total recurrence risk of 4.1%. Recurrence risk in offspring ranged from 3.1% for tetralogy of Fallot to 7.8% for atrioventricular septal defect (Lancet 1998;351:311-6).

CHD occurred more often in offspring of affected mothers (5.7%) than affected fathers (2.2%). Compared with offspring, sibling risk was significantly lower: 2.1% overall.

A much larger and more recent study done in Denmark again showed strong familial clustering in first-degree relatives for CHDs, particularly for recurrences of the same heart defect. (Very few families experience a second heart defect, the study found.)

The study – a national cohort study of more than 1.7 million people born during a 29-year period – also is one of the largest studies, if not the largest study, to document a decreasing risk as family history gets more distant.

The relative risks of any CHD in singletons were 3.21 for a family history of any CHD in first-degree relatives, 1.78 for a family history involving second-degree relatives, and 1.10 for a family history in third-degree relatives (Circulation 2009;120:295-301).

Only with a history of affected first- and second-degree relatives was there a statistically greater chance of having an affected fetus.

In twins, the relative risks of any CHD were 12.5 for same-sex twins and 6.93 for twins of both sexes, the Danish investigators noted.

And in looking at the contribution of CHD family history to the total number of CHD cases in the population, they found that 2.2% of heart defect cases in the population were attributed to CHD family history in first-degree relatives.

Another notable finding from other studies is that women with cyanotic heart disease have a higher risk of having a baby with CHD than do women with noncyanotic heart disease.

Maternal Risks

Just as we've learned much about inherited causes of congenital heart disease over the past 15 years, there is a growing body of epidemiologic literature on potential fetal exposures – from maternal illnesses to maternal drug exposures – that can alter the risk of CHD.

The risk factors for CHD maternal teratogen exposure are numerous. They include lithium, alcohol, isotretinoin, and various anticonvulsant drugs, and many are well-appreciated by ob.gyns.

Other factors for which risk has been well determined, and can be better appreciated, include:

▸ High vitamin A intake. Findings are not completely consistent, but we have enough data now to suggest that women who take extra-large doses of vitamin A may actually be putting their fetuses at risk of birth defects.

One study worth noting found that among more than 22,000 pregnant women, those who took more than 10,000 IU of vitamin A from supplements were 4.8 times more likely to have babies with birth defects associated with cranial-neural-crest tissue than were women who consumed 5,000 IU or less per day (N. Engl. J. Med. 1995;333:1369-73).

Typical prenatal vitamins have 5,000 IU in each dose. This is one reason that women with twin pregnancies can take extra folic acid, but should not double up on their prenatal multivitamins.

▸ Folate antagonists. Common drugs such as trimethoprim, triamterene, sulfasalazine, phenytoin, phenobarbital, primidone, carbamazepine, and cholestyramine may increase the risk not only of neural-tube defects, but of cardiovascular defects as well, in addition to oral clefts and urinary tract defects.

Fortunately, studies such as one published in 2000 involving thousands of infants show that the folic acid component of prenatal multivitamin supplements can reduce the risks of these defects, just as it reduces the risk of neural-tube defects (N. Engl. J. Med. 2000;343:1608-14).

▸ Paxil (paroxetine). This is the only antidepressant that has been shown in some studies to increase the risk of CHD. Its manufacturer, GlaxoSmithKline, changed the label's pregnancy precaution in 2005 from a Pregnancy Category C to Category D. If a patient becomes pregnant while taking the drug, she should be advised of potential harm to the fetus.

One epidemiologic study showed that women taking Paxil were two times more likely to have an infant with CHD, and 2.2 times more likely to have an infant with any congenital malformation, than were women taking other antidepressants.

▸ Diabetes. The risk of fetal anomalies with maternal diabetes and elevated hemoglobin A_1c in early pregnancy has been known for some time.

In a study published in 1981, for instance, the risk of CHD and other fetal anomalies rose from 5% to 22% as maternal HbA_1c rose from a range of 7%–8.5% to greater than 8.5% (N. Engl. J. Med. 1981;304:1331-4).

We've also known for some time that differences in CHD may exist even with good metabolic control. Studies have documented mild cardiac hypertrophy involving the interventricular septum and the ventricular free walls, for instance, in diabetic mothers with good metabolic control (J. Pediatr. 1991;118:103-7 and Am J. Obstet. Gynecol. 1991;164:837-43). Such growth affects cardiac diastolic function.

With the epidemic of obesity and the increasing prevalence of early type 2 diabetes and glucose intolerance among women of childbearing age, however, this is an increasingly important risk factor to appreciate and counsel about.

The most important message, we've learned, is that there's no such thing as perfect control – that good metabolic control will not necessarily protect diabetic mothers from the higher risk of CHD.

Just as detection and appropriate management of diabetes before and during pregnancy are of utmost importance, so is fetal echocardiography for every pregnant woman who has pregestational diabetes – even diabetes that is well controlled.

Indeed, the same review of all fetal echocardiography performed between 1985 and 2003 at Yale-New Haven Hospital that showed an increase in referrals for family history also showed a 9% increase in the proportion of studies done for pregestational diabetes as the indication. The increase was most striking when it came to women who had recently been diagnosed, compared with long-standing diabetes – a finding that likely reflects the increase in obesity.

▸ Phenylketonuria. Fortunately, strict dietary control before conception and during pregnancy can reduce the increased risk of heart defects faced by women with this disorder. We need to remember that aspartame (NutraSweet) can cause phenylalanine levels to increase in women with PKU, but not in normal women. Women without PKU can be reassured that there is no evidence linking aspartame with birth defects.

Fetal Risks

Among the fetal risk factors important to consider are:

▸ Extracardiac anomalies. The identification of any extracardiac anomaly should raise our level of suspicion for other anomalies, including congenital heart defects. If we see one anomaly – anywhere in the fetus – there often are really two. And if we see two anomalies, there frequently are really three.

▸ Nonimmune hydrops. All fetuses found to have NIH should be evaluated with fetal echocardiography. Structural heart disease in fetuses with NIH is usually indicative of a poor prognosis for survival, but when rhythm disturbances/arrhythmias are detected in association with NIH, there is sometimes an option for prenatal treatment.

▸ Fetal arrhythmias. An irregular heartbeat is usually not a problem, but tachycardia and especially bradycardia are associated with an increased risk of CHD. There may be structural heart defects in as many as half of fetuses with fixed bradycardia (i.e., baseline heart rate less than 100). In general, it is best that all arrhythmias are examined; it is just too hard to tell them apart by auscultation alone.

▸ Nuchal translucency. Numerous studies have shown that elevated first-trimester nuchal translucency (NT) increases the risk of major congenital heart defects in chromosomally normal fetuses, and that risk increases with increasing NT measurement.

In a large prospective multicenter study conducted by the National Institute of Child Health and Human Development, for instance, investigators identified 21 cases of major congenital heart defects in 8,167 chromosomally normal pregnancies. They reported that the incidence of CHD per 1,000 pregnancies rose from 1.9 with an NT measurement of less than 2.0 mm, to 4.8 with an NT measurement of 2.0–2.4 mm, to 6.0 with an NT measurement of 2.5–3.4 mm, to 23 of every 1,000 pregnancies with an NT measurement of 3.5 mm or greater (Am. J. Obstet. Gynecol. 2005;192:1357–61).

If the NT is greater than 3.5 mm, measured by a qualified sonographer or sonologist at 11–14 weeks as part of an aneuploidy risk assessment scan, the patient should be referred for fetal echocardiography.

▸ In vitro fertilization. We recently investigated the prevalence of congenital heart defects among IVF pregnancies at our referral program at Yale, and found that children conceived through IVF were 3–12 times as likely to have CHD as was the general population (J. Ultrasound Med. 2010;29:917-22).

Similar data have come from Australia and Europe, with reported odds ratios for IVF versus natural conception of 3–4. I tell patients, therefore, that it's not just one place or one study suggesting risk. Indeed, it's a meaningful risk factor.

▸ Monochorionic twins. In a systemic literature review we conducted several years ago that included 40 fetuses with CHDs among 830 fetuses from monochorionic/diamniotic twin gestations, the rate of CHDs in these twin gestations was significantly higher than the prevalence rate of CHDs in the general population (J. Ultrasound Med. 2007;26:1491-8).

Congenital heart defects were almost three times as likely to complicate the monochorionic/diamniotic twin gestations affected by twin-to-twin transfusion syndrome (TTTS), compared with those without TTTS, but an increase occurred regardless of the presence of TTTS. Ventricular septal defects were among the most frequent heart defects. Fetal echocardiography may be considered for all monochorionic/diamniotic twin gestations.

Dr. Copel disclosed that he has received research support from Philips Healthcare and Siemens Healthcare. Both companies manufacture echocardiography and other ultrasound systems.

A two-dimensional four-chamber view of a normal fetal heart (left). Fetal image of a complete atrioventricular septal defect with large atrial (**) and ventricular (*) septal defects (right).

Source Images: © Elsevier Inc.

Diagnosing Birth Defects

Birth defects continue to account for the majority of infant deaths, and their biologic basis continues to present a mixed picture, with the majority of causes still unknown. Cardiac defects – the most common type of birth defect – result in varying types of morbidity, but remain the most severe and disabling of all birth defects. As our guest author points out below, cardiovascular defects are the single largest contributor to birth defect–attributable infant mortality.

What is clear is the fact that when birth defects are identified prenatally, decisions can be made regarding the timing and route of delivery and even the facility where delivery occurs. We know that such decision making can be highly influential on the ultimate outcome of the infant.

Fortunately, there has been improvement in recent years in diagnostic technology that enables more prenatal diagnosis of congenital heart disease, and certain conditions that in the past went unknown or undiagnosed are now being identified early so that specialists can intervene in a timely manner.

While certain pregnancies are clearly at higher risk – those involving mothers who have pregestational diabetes, for instance, or mothers with exposure to particular toxins – there are other scenarios and factors that increase risk of which we should be aware.

It's a tricky evaluative process, for, as our guest author points out, most infants born with congenital heart disease do not have defined risk factors. At the least, however, we can be aware of the familial, maternal, and fetal factors that are known to increase risk and then ensure that all at-risk pregnancies are properly evaluated – often with fetal echocardiography – to determine if a cardiac defect is present and, if so, to plan the delivery-related issues of timing, mode, and facility.

In light of the importance of this subject and the role that ultrasound scanning, genetic counseling, and early decision making and planning can play in the ultimate outcome of the fetus, we decided to do a Master Class on congenital heart defects. We have invited Dr. Joshua A. Copel, professor of obstetrics, gynecology, and reproductive services, and of pediatrics, and vice chair of obstetrics at Yale University, New Haven, Conn., to serve as our guest professor.

Dr. Copel has written and lectured extensively on fetal arrhythmias, fetal cardiac anomalies and congenital heart disease, and sonographic monitoring and fetal echocardiography. Here he discusses what we should know about both familial contributions to congenital heart disease and various noninherited risk factors.

Surgery has an important role in the management of a patient's regret following tubal sterilization. While assisted reproductive technologies (ART) have made great strides in efficacy and patient acceptability, reanastomosis of the tubal segments remains an attractive option for couples who have no other fertility issues and who find the risk of multiple pregnancies unacceptable or the extensive medical treatment of ART impractical or undesirable. The re-establishment of some degree of reproductive tract function also can have important psychological and/or religious implications.

As with all gynecologic operations, there has been a trend toward the development and diffusion of minimally invasive (laparoscopic) versions of the classic microsurgical tubal reanastomosis.

The biggest problem with conventional laparoscopic tubal reanastomosis is that it is one of the most technically challenging gynecologic operations ever conceived. Before the full introduction of robotics at our institution, I have observed the struggle of skilled reproductive surgeons with every step of this operation.

It is no surprise, therefore, that laparoscopic tubal reanastomosis was among the first gynecologic operations for which robotic assistance was described. In fact, a feasibility study for tubal reanastomosis using the da Vinci surgical system (Intuitive Surgical Inc.) was published in 2000 by Dr. Michel Degueldre and colleagues in Belgium – 5 years before the Food and Drug Administration approved this surgical platform for gynecologic applications in the United States (Fertil. Steril. 2000;74:1020-3).

Two more recent case series compared robot-assisted tubal reanastomosis performed with the da Vinci surgical system to the “gold standard” of microsurgical reanastomosis by minilaparotomy. In a 2007 case-control study by Dr. Allison K. Rodgers and colleagues at the Cleveland Clinic, surgical times were significantly longer, and costs were higher, for the robot compared with open surgery. Hospitalization times were not significantly different, as patients undergoing minilaparotomy were discharged home on the day of surgery. Pregnancy rates also were similar (61% for robotic vs. 79% for minilaparotomy), as were ectopic pregnancy rates. Complications occurred less frequently in the robotic group, however, and the return to normal activity was shorter in this group by approximately 1 week (Obstet. Gynecol. 2007;109:1375-80).

A 2008 prospective cohort study by Dr. Sejal P. Dharia Patel and colleagues confirmed that surgical times are significantly longer for the robot group. This team did not practice outpatient minilaparotomy as did Dr. Rodgers' team, and patients undergoing robot-assisted laparoscopic surgery were discharged on the day of surgery. Hence, hospitalization times were significantly shorter in the robot-assisted group. Time to recovery was also significantly shorter. Pregnancy rates (62.5% for robotic vs. 50% for open) and ectopic pregnancy rates were not significantly different. Cost per delivery was similar between the two procedures (Fertil. Steril. 2008;90:1175-9).

These data indicate that robot-assisted tubal reanastomosis is safe and that its results are comparable to those obtained by classic tubal microsurgery performed by trained subspecialists. In terms of cost, it appears that even at the current high operating costs for the robot, open surgery is cost effective only if patients are sent home on the day of surgery, but not if they are admitted to the hospital.

Our robotic team performed the first successful robot-assisted tubal reanastomosis (with postoperative delivery) in New England in February 2007, and has since completely converted to the robotic approach. In our 4 years of experience, we have successfully completed approximately 350 robot-assisted reproductive surgeries, including robot-assisted tubal reanastomosis.

Setup

As in all laparoscopic procedures, patient positioning and port placement are vital. Robotic tubal reanastomosis is performed with the patient in dorsal lithotomy position on Allen stirrups. Preferably, all four robotic arms of the da Vinci patient-side cart are employed. The camera port is always placed within the umbilicus. The three 8-mm da Vinci ports are positioned as follows: Port 1 is 8-10 cm to the right of the camera port, port 2 is 8-10 cm to the left of the camera port, and port 3 is 8-10 cm to the left of port 2 (

Ports 1 and 2 are safely located in an area of the abdominal wall that is between the epigastric vessels (superficial and inferior) and the superficial circumflex vessels, making injury of any of these vessels extremely unlikely. Port 3 is located in the left lateral portion of the anterior abdominal wall. In women with a smaller abdomen, it is necessary to slide port 3 about 15-30 degrees caudal to port 2, while keeping the distance of 8-10 cm.

Optimal placement of robotic port 3 is undoubtedly the most challenging of the three 8-mm ports. Due to the obtrusive nature of the da Vinci patient-side cart, external interference between robotic arms 2 and 3 and between robotic arm 3 and the patient arm support systems (such as arm toboggans) is common during the learning curve of this operation. Moreover, internal interference between instruments in port 2 and port 3 is also possible (particularly if the degree of caudal shift of port 3 is excessive and the instrument crosses the pelvis transversely).

One also should be cognizant of the fact that placement of robotic port 3 in the lateral abdominal wall introduces the risk of a rarely observed complication of gynecologic laparoscopy: injury of the deep circumflex artery and vein (stemming from the external iliac artery and vein). The course of these vessels is significantly lateral to that of the inferior as well as superficial epigastric vessels, and is usually lateral to typical port sites in gynecologic surgery. Therefore, at the end of any robotic case employing a third instrument port in such a lateral location, all port sites (particularly the site of robotic port 3) should be re-evaluated laparoscopically for possible occult vessel injury after removal of the trocars, since the release of tamponade from the trocar may allow reactivation of bleeding.

To help prevent injury to the deep circumflex vessels, we also recommend exclusive use of blunt nondisposable robotic trocar obturators (instead of semisharp single-use obturators) for all robot-assisted procedures employing lateral placement of the third instrument arm.

Placement of the bedside surgical assistant port in robotic surgery has traditionally been high in the abdominal wall at either side of the umbilicus. However, we feel strongly that for the main reproductive surgery applications (namely, tubal reanastomosis and uterine myomectomy), the assistant port must be placed in one of the lower quadrants. Such placement is based on considerations of patient safety, assistant safety, and surgical ergonomics.

Reproductive microsurgery is suture intensive, and needle exchanges should never occur beyond the visual field of the console surgeon; loss of a 6.0 or an 8.0 needle between loops of bowel in the upper abdomen can turn an elegant procedure into a surgical nightmare.

In terms of assistant safety, placement of the assistant port as the most lateral port (instead of between the robotic camera arm and a robotic instrument arm) avoids the possibility that the assistant's hand could be caught between colliding robotic arms.

Finally, placing the assistant port in the lower quadrant allows for an overall port configuration that is compatible with any advanced conventional laparoscopic maneuvers that may be needed during the case (approximating the “ultralateral” port placement previously described for conventional laparoscopy).

Our right lower quadrant assistant port is always a 12-mm port. Even though it may be tempting to use a small-caliber assistant port in a microsurgical case like this, we have experienced problems with micro-needles becoming stuck in the plastic valve of assistant ports smaller than 12 mm in diameter. Ideally, a valve-free assistant port should be used for these cases.

Tubal reanastomosis is often performed best with the uterus in anteversion, so it is essential to employ a manipulator that allows the uterus to be fixed in any desired position between 0 and 90 degrees of anteversion. Such a manipulator should provide reliable chromopertubation. Several nondisposable devices work perfectly for this application, such as the Hayden Uterine Manipulator (Hayden Medical Inc.), the Pelosi Uterine Manipulator (Apple Medical Corp.), the Valtchev Uterine Mobilizer (Conkin Surgical Instruments Ltd.), and the RUMI Uterine Manipulator (CooperSurgical Inc.).

In our experience, the degree of anteversion provided by these devices is more than enough to complete a reanastomosis procedure without the need for an assistant to actively support the manipulator. Clearly, a vaginal delineator is never required for this procedure, and all of these manipulators can be assembled without the delineator.

As in all robot-assisted reproductive surgery techniques, we prefer lateral docking of the patient-side cart: This allows ample space for access to uterine positioning devices (

Surgical Technique

Our philosophy in transitioning from the gold-standard minilaparotomy approach to the laparoscopic approach has been that three essential aspects of the operation could not be compromised: 1) Reanastomosis should occur over a tubal stent; 2) secure orientation should be achieved by applying more than two reanastomosis sutures per tube; and 3) the thinnest safely employable suture should be chosen. At this point, our surgical protocol for robot-assisted laparoscopic tubal reanastomosis is identical to the one that we used for minilaparotomy.

The robotic microinstruments employed in this operation are shown in

The choice of whether to use the Potts scissors in port 1 (right-sided) or port 2 (left-sided) depends on how important it is for the surgeon to have the opportunity to use two graspers at the same time for tissue manipulation. With the robotic arm configuration described here (one robotic port on the patient's right, and two robotic ports on the patient's left), the surgeon has to toggle between instruments 2 and 3 on the left side. By keeping the scissors on the left side, the surgeon can have the ProGrasp and the micro-bipolar forceps at his disposal at the same time when needed. This means, however, that most surgeons would have to operate the robotic Potts scissors with the nondominant hand. In our experience, the accuracy of operation is not compromised as long as the robotic console scaling setting is at very fine (1:3) or ultrafine (1:5) (the latter is preferred for this operation but is not available at this time on the latest version of the da Vinci system).

A dilute solution of vasopressin is injected into the mesosalpinx in order to decrease blood loss during mobilization of the proximal and distal tubal segments. Potts scissors and micro-bipolar graspers are used to mobilize tubal segments and to deperitonealize the edges of the mesosalpinx. Even though we have bipolar electrocautery (micro-bipolar forceps) at our disposal, we employ it sparingly to avoid occult thermal damage to the tubal epithelium.

A graduated-tip ERCP (endoscopic retrograde cholangiopancreatography) cannula (Contour 3-4-5 Tip ERCP Cannula, Boston Scientific, Natick, Mass.) is inserted through the fimbriated end of the distal tubal segment; it exits through the newly opened proximal lumen of this distal segment and enters the newly created opening of the proximal tubal segment. This stent provides anatomic orientation and helps to identify the tubal lumen.

Preparation of the tubal edges and placement of the ERCP cannula are performed with the assistance of chromopertubation. It is particularly important to note abundant spillage of indigo carmine solution when the proximal tubal stump is opened.

From this point in the operation, the main role of the robotic ProGrasp operated through port 3 is to hold the ERCP cannula in place, thereby providing a steady and reliable stent.

The second stage of the procedure involves the suturing of the proximal and distal tubal stumps together (

Once the reapproximation is complete, the ERCP catheter is removed from the tube and immediately removed from the abdominal cavity, and prompt fill and spill of indigo carmine is observed, indicating patency. The same procedure is performed on the contralateral side, with great care taken to avoid inadvertently damaging the delicate reanastomosis line in the first tube.

This procedure invariably involves minimal or no blood loss. We gently irrigate the pelvis at regular intervals during the case to avoid desiccation, and carefully remove any small blood clots that may form. We do not employ any other adhesion-prevention strategies.

Patients leave the hospital within 3 hours of surgery and expect complete recovery within 2 weeks. Pelvic rest is recommended for the first month after surgery, and contraception is recommended until after their hysterosalpingogram 2 months after surgery.

A 'Swift Learning Curve'

Robotic assistance allows easy performance of classic microsurgical reanastomosis through laparoscopic access. Aside from shorter recovery time and a lower chance of complications, the robotic approach does not provide major clinical advantages over classic minilaparotomy. However, having performed and taught all three types of tubal reanastomosis (classic microsurgical, laparoscopic, and robot-assisted laparoscopic), I have been impressed by a unique quality of robotic reanastomosis: its eminent reproducibility and swift learning curve.

The enabling nature of robotic technology makes tubal reanastomosis a perfect example of an operation that is more safely learned and performed robotically. At our institution, we have developed a protocol for fast-track teaching of robot-assisted laparoscopic tubal reanastomosis that involves several hours of inanimate training at the console to get familiar with microrobotic instruments and sutures, as well as a chance to assist on these cases at bedside.

This is followed by an intensive use of Telestration, a technology specific to the da Vinci surgical system that greatly improves communication between the teacher and apprentice during surgery. The apprentice sits at the console while the attending surgeon remains by his/her side and communicates precise instructions by direct verbal cues and by drawing on a dedicated monitor that shows the operator's right field of vision. The drawings are transmitted in real time to the console, where they superimpose on the operator's visual field so that he/she may incorporate them into the current surgical act.

Aside from the expected differences in speed of performance, the quality and the safety of the operations performed by the teacher and the apprentice are absolutely comparable even on the first case. Coupled with a simple teaching strategy, robotic technology thus dramatically shortens the learning curve of a complex microsurgical operation. It is hard to deny the value of a procedure that can be safely taught and reliably reproduced in a single session.

The Ultimate Microsurgical Procedure

www.aagl.org

For this current excerpt of Master Class in Gynecologic Surgery, I have solicited the expertise of Dr. Antonio Rosario Gargiulo. Dr. Gargiulo is an assistant professor of obstetrics, gynecology, and reproductive biology at Harvard Medical School, Boston. He is also the director of robotic surgery at the Center for Infertility and Reproductive Surgery at Brigham and Women's Hospital in Boston. After completing his subspecialty training in reproductive endocrinology and infertility at Brigham and Women's Hospital, Dr. Gargiulo rapidly built a busy reproductive surgery practice and has become a well-known expert in minimally invasive gynecologic surgery. Since 2007, Dr. Gargiulo has worked in robot-assisted reproductive surgery, and he performed the first robotic tubal anastomosis in New England. It is a pleasure to watch such a young and gifted surgeon successfully and safely pushing the envelope in gynecologic robot-assisted microsurgery.

In 1999, Dr. Tommaso Falcone and his Cleveland Clinic team published the first report on robot-assisted tubal anastomosis. Using the Zeus robotic surgical system (Computer Motion Inc.), a far less sophisticated tool than the currently available da Vinci surgical system (Intuitive Surgical), the procedure lasted more than 5 hours.

Only a decade later, Dr. Martin Caillet presented a study on robot-assisted laparoscopic microsurgical tubal reanastomosis at the First European Symposium in Robotic Gynecological Surgery in Milan, Italy (Fertil. Steril. 2010;94:1844-7). In 97 patients who underwent robot-assisted microsurgical tubal reanastomosis, using the da Vinci surgical system, the overall pregnancy and birth rates were 71% and 62%, respectively. Not surprisingly, pregnancy and delivery rates were age related. In patients less than 35 years of age, 91% became pregnant and 88% delivered. By the time a patient reached 36-39 years of age, pregnancy and delivery rates were 75% and 66%, respectively, while patients aged 40-42 years had a 50% pregnancy rate and 43.8% delivery rate. Importantly, the success rate is similar to rates reported in the most successful studies using the open microsurgical tubal reanastomosis technique.

Dr. Gargiulo will discuss his technique in this Gynecologic Surgery Master Class.

Surgery has an important role in the management of a patient's regret following tubal sterilization. While assisted reproductive technologies (ART) have made great strides in efficacy and patient acceptability, reanastomosis of the tubal segments remains an attractive option for couples who have no other fertility issues and who find the risk of multiple pregnancies unacceptable or the extensive medical treatment of ART impractical or undesirable. The re-establishment of some degree of reproductive tract function also can have important psychological and/or religious implications.

As with all gynecologic operations, there has been a trend toward the development and diffusion of minimally invasive (laparoscopic) versions of the classic microsurgical tubal reanastomosis.

The biggest problem with conventional laparoscopic tubal reanastomosis is that it is one of the most technically challenging gynecologic operations ever conceived. Before the full introduction of robotics at our institution, I have observed the struggle of skilled reproductive surgeons with every step of this operation.

It is no surprise, therefore, that laparoscopic tubal reanastomosis was among the first gynecologic operations for which robotic assistance was described. In fact, a feasibility study for tubal reanastomosis using the da Vinci surgical system (Intuitive Surgical Inc.) was published in 2000 by Dr. Michel Degueldre and colleagues in Belgium – 5 years before the Food and Drug Administration approved this surgical platform for gynecologic applications in the United States (Fertil. Steril. 2000;74:1020-3).

Two more recent case series compared robot-assisted tubal reanastomosis performed with the da Vinci surgical system to the “gold standard” of microsurgical reanastomosis by minilaparotomy. In a 2007 case-control study by Dr. Allison K. Rodgers and colleagues at the Cleveland Clinic, surgical times were significantly longer, and costs were higher, for the robot compared with open surgery. Hospitalization times were not significantly different, as patients undergoing minilaparotomy were discharged home on the day of surgery. Pregnancy rates also were similar (61% for robotic vs. 79% for minilaparotomy), as were ectopic pregnancy rates. Complications occurred less frequently in the robotic group, however, and the return to normal activity was shorter in this group by approximately 1 week (Obstet. Gynecol. 2007;109:1375-80).

A 2008 prospective cohort study by Dr. Sejal P. Dharia Patel and colleagues confirmed that surgical times are significantly longer for the robot group. This team did not practice outpatient minilaparotomy as did Dr. Rodgers' team, and patients undergoing robot-assisted laparoscopic surgery were discharged on the day of surgery. Hence, hospitalization times were significantly shorter in the robot-assisted group. Time to recovery was also significantly shorter. Pregnancy rates (62.5% for robotic vs. 50% for open) and ectopic pregnancy rates were not significantly different. Cost per delivery was similar between the two procedures (Fertil. Steril. 2008;90:1175-9).

These data indicate that robot-assisted tubal reanastomosis is safe and that its results are comparable to those obtained by classic tubal microsurgery performed by trained subspecialists. In terms of cost, it appears that even at the current high operating costs for the robot, open surgery is cost effective only if patients are sent home on the day of surgery, but not if they are admitted to the hospital.

Our robotic team performed the first successful robot-assisted tubal reanastomosis (with postoperative delivery) in New England in February 2007, and has since completely converted to the robotic approach. In our 4 years of experience, we have successfully completed approximately 350 robot-assisted reproductive surgeries, including robot-assisted tubal reanastomosis.

Setup

As in all laparoscopic procedures, patient positioning and port placement are vital. Robotic tubal reanastomosis is performed with the patient in dorsal lithotomy position on Allen stirrups. Preferably, all four robotic arms of the da Vinci patient-side cart are employed. The camera port is always placed within the umbilicus. The three 8-mm da Vinci ports are positioned as follows: Port 1 is 8-10 cm to the right of the camera port, port 2 is 8-10 cm to the left of the camera port, and port 3 is 8-10 cm to the left of port 2 (

Ports 1 and 2 are safely located in an area of the abdominal wall that is between the epigastric vessels (superficial and inferior) and the superficial circumflex vessels, making injury of any of these vessels extremely unlikely. Port 3 is located in the left lateral portion of the anterior abdominal wall. In women with a smaller abdomen, it is necessary to slide port 3 about 15-30 degrees caudal to port 2, while keeping the distance of 8-10 cm.

Optimal placement of robotic port 3 is undoubtedly the most challenging of the three 8-mm ports. Due to the obtrusive nature of the da Vinci patient-side cart, external interference between robotic arms 2 and 3 and between robotic arm 3 and the patient arm support systems (such as arm toboggans) is common during the learning curve of this operation. Moreover, internal interference between instruments in port 2 and port 3 is also possible (particularly if the degree of caudal shift of port 3 is excessive and the instrument crosses the pelvis transversely).

One also should be cognizant of the fact that placement of robotic port 3 in the lateral abdominal wall introduces the risk of a rarely observed complication of gynecologic laparoscopy: injury of the deep circumflex artery and vein (stemming from the external iliac artery and vein). The course of these vessels is significantly lateral to that of the inferior as well as superficial epigastric vessels, and is usually lateral to typical port sites in gynecologic surgery. Therefore, at the end of any robotic case employing a third instrument port in such a lateral location, all port sites (particularly the site of robotic port 3) should be re-evaluated laparoscopically for possible occult vessel injury after removal of the trocars, since the release of tamponade from the trocar may allow reactivation of bleeding.

To help prevent injury to the deep circumflex vessels, we also recommend exclusive use of blunt nondisposable robotic trocar obturators (instead of semisharp single-use obturators) for all robot-assisted procedures employing lateral placement of the third instrument arm.

Placement of the bedside surgical assistant port in robotic surgery has traditionally been high in the abdominal wall at either side of the umbilicus. However, we feel strongly that for the main reproductive surgery applications (namely, tubal reanastomosis and uterine myomectomy), the assistant port must be placed in one of the lower quadrants. Such placement is based on considerations of patient safety, assistant safety, and surgical ergonomics.

Reproductive microsurgery is suture intensive, and needle exchanges should never occur beyond the visual field of the console surgeon; loss of a 6.0 or an 8.0 needle between loops of bowel in the upper abdomen can turn an elegant procedure into a surgical nightmare.

In terms of assistant safety, placement of the assistant port as the most lateral port (instead of between the robotic camera arm and a robotic instrument arm) avoids the possibility that the assistant's hand could be caught between colliding robotic arms.

Finally, placing the assistant port in the lower quadrant allows for an overall port configuration that is compatible with any advanced conventional laparoscopic maneuvers that may be needed during the case (approximating the “ultralateral” port placement previously described for conventional laparoscopy).

Our right lower quadrant assistant port is always a 12-mm port. Even though it may be tempting to use a small-caliber assistant port in a microsurgical case like this, we have experienced problems with micro-needles becoming stuck in the plastic valve of assistant ports smaller than 12 mm in diameter. Ideally, a valve-free assistant port should be used for these cases.

Tubal reanastomosis is often performed best with the uterus in anteversion, so it is essential to employ a manipulator that allows the uterus to be fixed in any desired position between 0 and 90 degrees of anteversion. Such a manipulator should provide reliable chromopertubation. Several nondisposable devices work perfectly for this application, such as the Hayden Uterine Manipulator (Hayden Medical Inc.), the Pelosi Uterine Manipulator (Apple Medical Corp.), the Valtchev Uterine Mobilizer (Conkin Surgical Instruments Ltd.), and the RUMI Uterine Manipulator (CooperSurgical Inc.).

In our experience, the degree of anteversion provided by these devices is more than enough to complete a reanastomosis procedure without the need for an assistant to actively support the manipulator. Clearly, a vaginal delineator is never required for this procedure, and all of these manipulators can be assembled without the delineator.

As in all robot-assisted reproductive surgery techniques, we prefer lateral docking of the patient-side cart: This allows ample space for access to uterine positioning devices (

Surgical Technique

Our philosophy in transitioning from the gold-standard minilaparotomy approach to the laparoscopic approach has been that three essential aspects of the operation could not be compromised: 1) Reanastomosis should occur over a tubal stent; 2) secure orientation should be achieved by applying more than two reanastomosis sutures per tube; and 3) the thinnest safely employable suture should be chosen. At this point, our surgical protocol for robot-assisted laparoscopic tubal reanastomosis is identical to the one that we used for minilaparotomy.

The robotic microinstruments employed in this operation are shown in

The choice of whether to use the Potts scissors in port 1 (right-sided) or port 2 (left-sided) depends on how important it is for the surgeon to have the opportunity to use two graspers at the same time for tissue manipulation. With the robotic arm configuration described here (one robotic port on the patient's right, and two robotic ports on the patient's left), the surgeon has to toggle between instruments 2 and 3 on the left side. By keeping the scissors on the left side, the surgeon can have the ProGrasp and the micro-bipolar forceps at his disposal at the same time when needed. This means, however, that most surgeons would have to operate the robotic Potts scissors with the nondominant hand. In our experience, the accuracy of operation is not compromised as long as the robotic console scaling setting is at very fine (1:3) or ultrafine (1:5) (the latter is preferred for this operation but is not available at this time on the latest version of the da Vinci system).

A dilute solution of vasopressin is injected into the mesosalpinx in order to decrease blood loss during mobilization of the proximal and distal tubal segments. Potts scissors and micro-bipolar graspers are used to mobilize tubal segments and to deperitonealize the edges of the mesosalpinx. Even though we have bipolar electrocautery (micro-bipolar forceps) at our disposal, we employ it sparingly to avoid occult thermal damage to the tubal epithelium.

A graduated-tip ERCP (endoscopic retrograde cholangiopancreatography) cannula (Contour 3-4-5 Tip ERCP Cannula, Boston Scientific, Natick, Mass.) is inserted through the fimbriated end of the distal tubal segment; it exits through the newly opened proximal lumen of this distal segment and enters the newly created opening of the proximal tubal segment. This stent provides anatomic orientation and helps to identify the tubal lumen.

Preparation of the tubal edges and placement of the ERCP cannula are performed with the assistance of chromopertubation. It is particularly important to note abundant spillage of indigo carmine solution when the proximal tubal stump is opened.

From this point in the operation, the main role of the robotic ProGrasp operated through port 3 is to hold the ERCP cannula in place, thereby providing a steady and reliable stent.

The second stage of the procedure involves the suturing of the proximal and distal tubal stumps together (

Once the reapproximation is complete, the ERCP catheter is removed from the tube and immediately removed from the abdominal cavity, and prompt fill and spill of indigo carmine is observed, indicating patency. The same procedure is performed on the contralateral side, with great care taken to avoid inadvertently damaging the delicate reanastomosis line in the first tube.

This procedure invariably involves minimal or no blood loss. We gently irrigate the pelvis at regular intervals during the case to avoid desiccation, and carefully remove any small blood clots that may form. We do not employ any other adhesion-prevention strategies.

Patients leave the hospital within 3 hours of surgery and expect complete recovery within 2 weeks. Pelvic rest is recommended for the first month after surgery, and contraception is recommended until after their hysterosalpingogram 2 months after surgery.

A 'Swift Learning Curve'

Robotic assistance allows easy performance of classic microsurgical reanastomosis through laparoscopic access. Aside from shorter recovery time and a lower chance of complications, the robotic approach does not provide major clinical advantages over classic minilaparotomy. However, having performed and taught all three types of tubal reanastomosis (classic microsurgical, laparoscopic, and robot-assisted laparoscopic), I have been impressed by a unique quality of robotic reanastomosis: its eminent reproducibility and swift learning curve.

The enabling nature of robotic technology makes tubal reanastomosis a perfect example of an operation that is more safely learned and performed robotically. At our institution, we have developed a protocol for fast-track teaching of robot-assisted laparoscopic tubal reanastomosis that involves several hours of inanimate training at the console to get familiar with microrobotic instruments and sutures, as well as a chance to assist on these cases at bedside.

This is followed by an intensive use of Telestration, a technology specific to the da Vinci surgical system that greatly improves communication between the teacher and apprentice during surgery. The apprentice sits at the console while the attending surgeon remains by his/her side and communicates precise instructions by direct verbal cues and by drawing on a dedicated monitor that shows the operator's right field of vision. The drawings are transmitted in real time to the console, where they superimpose on the operator's visual field so that he/she may incorporate them into the current surgical act.

Aside from the expected differences in speed of performance, the quality and the safety of the operations performed by the teacher and the apprentice are absolutely comparable even on the first case. Coupled with a simple teaching strategy, robotic technology thus dramatically shortens the learning curve of a complex microsurgical operation. It is hard to deny the value of a procedure that can be safely taught and reliably reproduced in a single session.

The Ultimate Microsurgical Procedure

www.aagl.org

For this current excerpt of Master Class in Gynecologic Surgery, I have solicited the expertise of Dr. Antonio Rosario Gargiulo. Dr. Gargiulo is an assistant professor of obstetrics, gynecology, and reproductive biology at Harvard Medical School, Boston. He is also the director of robotic surgery at the Center for Infertility and Reproductive Surgery at Brigham and Women's Hospital in Boston. After completing his subspecialty training in reproductive endocrinology and infertility at Brigham and Women's Hospital, Dr. Gargiulo rapidly built a busy reproductive surgery practice and has become a well-known expert in minimally invasive gynecologic surgery. Since 2007, Dr. Gargiulo has worked in robot-assisted reproductive surgery, and he performed the first robotic tubal anastomosis in New England. It is a pleasure to watch such a young and gifted surgeon successfully and safely pushing the envelope in gynecologic robot-assisted microsurgery.

In 1999, Dr. Tommaso Falcone and his Cleveland Clinic team published the first report on robot-assisted tubal anastomosis. Using the Zeus robotic surgical system (Computer Motion Inc.), a far less sophisticated tool than the currently available da Vinci surgical system (Intuitive Surgical), the procedure lasted more than 5 hours.

Only a decade later, Dr. Martin Caillet presented a study on robot-assisted laparoscopic microsurgical tubal reanastomosis at the First European Symposium in Robotic Gynecological Surgery in Milan, Italy (Fertil. Steril. 2010;94:1844-7). In 97 patients who underwent robot-assisted microsurgical tubal reanastomosis, using the da Vinci surgical system, the overall pregnancy and birth rates were 71% and 62%, respectively. Not surprisingly, pregnancy and delivery rates were age related. In patients less than 35 years of age, 91% became pregnant and 88% delivered. By the time a patient reached 36-39 years of age, pregnancy and delivery rates were 75% and 66%, respectively, while patients aged 40-42 years had a 50% pregnancy rate and 43.8% delivery rate. Importantly, the success rate is similar to rates reported in the most successful studies using the open microsurgical tubal reanastomosis technique.

Dr. Gargiulo will discuss his technique in this Gynecologic Surgery Master Class.

Surgery has an important role in the management of a patient's regret following tubal sterilization. While assisted reproductive technologies (ART) have made great strides in efficacy and patient acceptability, reanastomosis of the tubal segments remains an attractive option for couples who have no other fertility issues and who find the risk of multiple pregnancies unacceptable or the extensive medical treatment of ART impractical or undesirable. The re-establishment of some degree of reproductive tract function also can have important psychological and/or religious implications.

As with all gynecologic operations, there has been a trend toward the development and diffusion of minimally invasive (laparoscopic) versions of the classic microsurgical tubal reanastomosis.

The biggest problem with conventional laparoscopic tubal reanastomosis is that it is one of the most technically challenging gynecologic operations ever conceived. Before the full introduction of robotics at our institution, I have observed the struggle of skilled reproductive surgeons with every step of this operation.

It is no surprise, therefore, that laparoscopic tubal reanastomosis was among the first gynecologic operations for which robotic assistance was described. In fact, a feasibility study for tubal reanastomosis using the da Vinci surgical system (Intuitive Surgical Inc.) was published in 2000 by Dr. Michel Degueldre and colleagues in Belgium – 5 years before the Food and Drug Administration approved this surgical platform for gynecologic applications in the United States (Fertil. Steril. 2000;74:1020-3).

Two more recent case series compared robot-assisted tubal reanastomosis performed with the da Vinci surgical system to the “gold standard” of microsurgical reanastomosis by minilaparotomy. In a 2007 case-control study by Dr. Allison K. Rodgers and colleagues at the Cleveland Clinic, surgical times were significantly longer, and costs were higher, for the robot compared with open surgery. Hospitalization times were not significantly different, as patients undergoing minilaparotomy were discharged home on the day of surgery. Pregnancy rates also were similar (61% for robotic vs. 79% for minilaparotomy), as were ectopic pregnancy rates. Complications occurred less frequently in the robotic group, however, and the return to normal activity was shorter in this group by approximately 1 week (Obstet. Gynecol. 2007;109:1375-80).

A 2008 prospective cohort study by Dr. Sejal P. Dharia Patel and colleagues confirmed that surgical times are significantly longer for the robot group. This team did not practice outpatient minilaparotomy as did Dr. Rodgers' team, and patients undergoing robot-assisted laparoscopic surgery were discharged on the day of surgery. Hence, hospitalization times were significantly shorter in the robot-assisted group. Time to recovery was also significantly shorter. Pregnancy rates (62.5% for robotic vs. 50% for open) and ectopic pregnancy rates were not significantly different. Cost per delivery was similar between the two procedures (Fertil. Steril. 2008;90:1175-9).

These data indicate that robot-assisted tubal reanastomosis is safe and that its results are comparable to those obtained by classic tubal microsurgery performed by trained subspecialists. In terms of cost, it appears that even at the current high operating costs for the robot, open surgery is cost effective only if patients are sent home on the day of surgery, but not if they are admitted to the hospital.

Our robotic team performed the first successful robot-assisted tubal reanastomosis (with postoperative delivery) in New England in February 2007, and has since completely converted to the robotic approach. In our 4 years of experience, we have successfully completed approximately 350 robot-assisted reproductive surgeries, including robot-assisted tubal reanastomosis.

Setup

As in all laparoscopic procedures, patient positioning and port placement are vital. Robotic tubal reanastomosis is performed with the patient in dorsal lithotomy position on Allen stirrups. Preferably, all four robotic arms of the da Vinci patient-side cart are employed. The camera port is always placed within the umbilicus. The three 8-mm da Vinci ports are positioned as follows: Port 1 is 8-10 cm to the right of the camera port, port 2 is 8-10 cm to the left of the camera port, and port 3 is 8-10 cm to the left of port 2 (

Ports 1 and 2 are safely located in an area of the abdominal wall that is between the epigastric vessels (superficial and inferior) and the superficial circumflex vessels, making injury of any of these vessels extremely unlikely. Port 3 is located in the left lateral portion of the anterior abdominal wall. In women with a smaller abdomen, it is necessary to slide port 3 about 15-30 degrees caudal to port 2, while keeping the distance of 8-10 cm.

Optimal placement of robotic port 3 is undoubtedly the most challenging of the three 8-mm ports. Due to the obtrusive nature of the da Vinci patient-side cart, external interference between robotic arms 2 and 3 and between robotic arm 3 and the patient arm support systems (such as arm toboggans) is common during the learning curve of this operation. Moreover, internal interference between instruments in port 2 and port 3 is also possible (particularly if the degree of caudal shift of port 3 is excessive and the instrument crosses the pelvis transversely).

One also should be cognizant of the fact that placement of robotic port 3 in the lateral abdominal wall introduces the risk of a rarely observed complication of gynecologic laparoscopy: injury of the deep circumflex artery and vein (stemming from the external iliac artery and vein). The course of these vessels is significantly lateral to that of the inferior as well as superficial epigastric vessels, and is usually lateral to typical port sites in gynecologic surgery. Therefore, at the end of any robotic case employing a third instrument port in such a lateral location, all port sites (particularly the site of robotic port 3) should be re-evaluated laparoscopically for possible occult vessel injury after removal of the trocars, since the release of tamponade from the trocar may allow reactivation of bleeding.

To help prevent injury to the deep circumflex vessels, we also recommend exclusive use of blunt nondisposable robotic trocar obturators (instead of semisharp single-use obturators) for all robot-assisted procedures employing lateral placement of the third instrument arm.

Placement of the bedside surgical assistant port in robotic surgery has traditionally been high in the abdominal wall at either side of the umbilicus. However, we feel strongly that for the main reproductive surgery applications (namely, tubal reanastomosis and uterine myomectomy), the assistant port must be placed in one of the lower quadrants. Such placement is based on considerations of patient safety, assistant safety, and surgical ergonomics.

Reproductive microsurgery is suture intensive, and needle exchanges should never occur beyond the visual field of the console surgeon; loss of a 6.0 or an 8.0 needle between loops of bowel in the upper abdomen can turn an elegant procedure into a surgical nightmare.

In terms of assistant safety, placement of the assistant port as the most lateral port (instead of between the robotic camera arm and a robotic instrument arm) avoids the possibility that the assistant's hand could be caught between colliding robotic arms.

Finally, placing the assistant port in the lower quadrant allows for an overall port configuration that is compatible with any advanced conventional laparoscopic maneuvers that may be needed during the case (approximating the “ultralateral” port placement previously described for conventional laparoscopy).

Our right lower quadrant assistant port is always a 12-mm port. Even though it may be tempting to use a small-caliber assistant port in a microsurgical case like this, we have experienced problems with micro-needles becoming stuck in the plastic valve of assistant ports smaller than 12 mm in diameter. Ideally, a valve-free assistant port should be used for these cases.

Tubal reanastomosis is often performed best with the uterus in anteversion, so it is essential to employ a manipulator that allows the uterus to be fixed in any desired position between 0 and 90 degrees of anteversion. Such a manipulator should provide reliable chromopertubation. Several nondisposable devices work perfectly for this application, such as the Hayden Uterine Manipulator (Hayden Medical Inc.), the Pelosi Uterine Manipulator (Apple Medical Corp.), the Valtchev Uterine Mobilizer (Conkin Surgical Instruments Ltd.), and the RUMI Uterine Manipulator (CooperSurgical Inc.).

In our experience, the degree of anteversion provided by these devices is more than enough to complete a reanastomosis procedure without the need for an assistant to actively support the manipulator. Clearly, a vaginal delineator is never required for this procedure, and all of these manipulators can be assembled without the delineator.

As in all robot-assisted reproductive surgery techniques, we prefer lateral docking of the patient-side cart: This allows ample space for access to uterine positioning devices (

Surgical Technique

Our philosophy in transitioning from the gold-standard minilaparotomy approach to the laparoscopic approach has been that three essential aspects of the operation could not be compromised: 1) Reanastomosis should occur over a tubal stent; 2) secure orientation should be achieved by applying more than two reanastomosis sutures per tube; and 3) the thinnest safely employable suture should be chosen. At this point, our surgical protocol for robot-assisted laparoscopic tubal reanastomosis is identical to the one that we used for minilaparotomy.

The robotic microinstruments employed in this operation are shown in

The choice of whether to use the Potts scissors in port 1 (right-sided) or port 2 (left-sided) depends on how important it is for the surgeon to have the opportunity to use two graspers at the same time for tissue manipulation. With the robotic arm configuration described here (one robotic port on the patient's right, and two robotic ports on the patient's left), the surgeon has to toggle between instruments 2 and 3 on the left side. By keeping the scissors on the left side, the surgeon can have the ProGrasp and the micro-bipolar forceps at his disposal at the same time when needed. This means, however, that most surgeons would have to operate the robotic Potts scissors with the nondominant hand. In our experience, the accuracy of operation is not compromised as long as the robotic console scaling setting is at very fine (1:3) or ultrafine (1:5) (the latter is preferred for this operation but is not available at this time on the latest version of the da Vinci system).

A dilute solution of vasopressin is injected into the mesosalpinx in order to decrease blood loss during mobilization of the proximal and distal tubal segments. Potts scissors and micro-bipolar graspers are used to mobilize tubal segments and to deperitonealize the edges of the mesosalpinx. Even though we have bipolar electrocautery (micro-bipolar forceps) at our disposal, we employ it sparingly to avoid occult thermal damage to the tubal epithelium.

A graduated-tip ERCP (endoscopic retrograde cholangiopancreatography) cannula (Contour 3-4-5 Tip ERCP Cannula, Boston Scientific, Natick, Mass.) is inserted through the fimbriated end of the distal tubal segment; it exits through the newly opened proximal lumen of this distal segment and enters the newly created opening of the proximal tubal segment. This stent provides anatomic orientation and helps to identify the tubal lumen.

Preparation of the tubal edges and placement of the ERCP cannula are performed with the assistance of chromopertubation. It is particularly important to note abundant spillage of indigo carmine solution when the proximal tubal stump is opened.

From this point in the operation, the main role of the robotic ProGrasp operated through port 3 is to hold the ERCP cannula in place, thereby providing a steady and reliable stent.

The second stage of the procedure involves the suturing of the proximal and distal tubal stumps together (

Once the reapproximation is complete, the ERCP catheter is removed from the tube and immediately removed from the abdominal cavity, and prompt fill and spill of indigo carmine is observed, indicating patency. The same procedure is performed on the contralateral side, with great care taken to avoid inadvertently damaging the delicate reanastomosis line in the first tube.

This procedure invariably involves minimal or no blood loss. We gently irrigate the pelvis at regular intervals during the case to avoid desiccation, and carefully remove any small blood clots that may form. We do not employ any other adhesion-prevention strategies.

Patients leave the hospital within 3 hours of surgery and expect complete recovery within 2 weeks. Pelvic rest is recommended for the first month after surgery, and contraception is recommended until after their hysterosalpingogram 2 months after surgery.

A 'Swift Learning Curve'

Robotic assistance allows easy performance of classic microsurgical reanastomosis through laparoscopic access. Aside from shorter recovery time and a lower chance of complications, the robotic approach does not provide major clinical advantages over classic minilaparotomy. However, having performed and taught all three types of tubal reanastomosis (classic microsurgical, laparoscopic, and robot-assisted laparoscopic), I have been impressed by a unique quality of robotic reanastomosis: its eminent reproducibility and swift learning curve.

The enabling nature of robotic technology makes tubal reanastomosis a perfect example of an operation that is more safely learned and performed robotically. At our institution, we have developed a protocol for fast-track teaching of robot-assisted laparoscopic tubal reanastomosis that involves several hours of inanimate training at the console to get familiar with microrobotic instruments and sutures, as well as a chance to assist on these cases at bedside.

This is followed by an intensive use of Telestration, a technology specific to the da Vinci surgical system that greatly improves communication between the teacher and apprentice during surgery. The apprentice sits at the console while the attending surgeon remains by his/her side and communicates precise instructions by direct verbal cues and by drawing on a dedicated monitor that shows the operator's right field of vision. The drawings are transmitted in real time to the console, where they superimpose on the operator's visual field so that he/she may incorporate them into the current surgical act.

Aside from the expected differences in speed of performance, the quality and the safety of the operations performed by the teacher and the apprentice are absolutely comparable even on the first case. Coupled with a simple teaching strategy, robotic technology thus dramatically shortens the learning curve of a complex microsurgical operation. It is hard to deny the value of a procedure that can be safely taught and reliably reproduced in a single session.

The Ultimate Microsurgical Procedure

www.aagl.org

For this current excerpt of Master Class in Gynecologic Surgery, I have solicited the expertise of Dr. Antonio Rosario Gargiulo. Dr. Gargiulo is an assistant professor of obstetrics, gynecology, and reproductive biology at Harvard Medical School, Boston. He is also the director of robotic surgery at the Center for Infertility and Reproductive Surgery at Brigham and Women's Hospital in Boston. After completing his subspecialty training in reproductive endocrinology and infertility at Brigham and Women's Hospital, Dr. Gargiulo rapidly built a busy reproductive surgery practice and has become a well-known expert in minimally invasive gynecologic surgery. Since 2007, Dr. Gargiulo has worked in robot-assisted reproductive surgery, and he performed the first robotic tubal anastomosis in New England. It is a pleasure to watch such a young and gifted surgeon successfully and safely pushing the envelope in gynecologic robot-assisted microsurgery.

In 1999, Dr. Tommaso Falcone and his Cleveland Clinic team published the first report on robot-assisted tubal anastomosis. Using the Zeus robotic surgical system (Computer Motion Inc.), a far less sophisticated tool than the currently available da Vinci surgical system (Intuitive Surgical), the procedure lasted more than 5 hours.

Only a decade later, Dr. Martin Caillet presented a study on robot-assisted laparoscopic microsurgical tubal reanastomosis at the First European Symposium in Robotic Gynecological Surgery in Milan, Italy (Fertil. Steril. 2010;94:1844-7). In 97 patients who underwent robot-assisted microsurgical tubal reanastomosis, using the da Vinci surgical system, the overall pregnancy and birth rates were 71% and 62%, respectively. Not surprisingly, pregnancy and delivery rates were age related. In patients less than 35 years of age, 91% became pregnant and 88% delivered. By the time a patient reached 36-39 years of age, pregnancy and delivery rates were 75% and 66%, respectively, while patients aged 40-42 years had a 50% pregnancy rate and 43.8% delivery rate. Importantly, the success rate is similar to rates reported in the most successful studies using the open microsurgical tubal reanastomosis technique.

Dr. Gargiulo will discuss his technique in this Gynecologic Surgery Master Class.

During the last 10–15 years, in an effort to improve troubling rates of spontaneous preterm delivery and other adverse pregnancy outcomes, investigators have looked at many kinds of clinical and subclinical infections and explored their possible associations to preterm birth.

Bacterial vaginosis is one infection that has been associated in numerous studies with a higher risk of preterm birth. Periodontal disease is another. While not all studies have found an association, there is substantial evidence – mainly from observational and epidemiologic studies – linking periodontal disease to spontaneous preterm birth and identifying the disease as a probable risk factor for preterm delivery.

One of the larger studies was a prospective cohort study involving more than 1,300 pregnant women who were enrolled at 21–24 weeks' gestation and provided information on various possible risk factors for preterm birth. Later analyses showed that women with moderate to severe periodontal disease were 4.5 times as likely to deliver spontaneously before 37 weeks' gestation, 5.3 times as likely to deliver before 35 weeks' gestation, and 7.1 times as likely to deliver before 32 weeks (J. Am. Dent. Assoc. 2001;132:875–80).

Other published studies report lower levels of risk, and a more recent meta-analysis that included 17 studies and more than 7,000 women suggested a 2.8-fold increased risk of preterm birth in women with periodontal disease (Am J. Obstet. Gynecol. 2007;196:135.e1–7).

Today, interestingly, we know that bacterial vaginosis and periodontal disease each present our patients with a similar magnitude of increased risk for preterm delivery: a two- to threefold increased risk.

Unfortunately, hopes that identifying and treating the conditions could reduce risk and improve pregnancy outcomes have been dashed – in both cases. In the case of periodontal disease, three major randomized controlled trials in the United States – including the Periodontal Infections and Prematurity Study (PIPS) published in February of this year – have provided evidence that screening and treating periodontal disease during pregnancy are not likely to reduce rates of preterm birth.

This does not mean, however, that we should ignore the problem of periodontal disease. It is a huge problem, affecting up to 40% of pregnant women according to most reports, and there is no evidence to suggest that dental examinations or treatment are deleterious during pregnancy. In all the studies that have been done over the last decade or so, there is nothing to suggest that we shouldn't look for periodontal disease and treat it.

Periodontal disease is clearly associated with other poor health outcomes, in addition to its association with preterm birth, and study after study has shown that good oral health is important for good overall health.

Despite our inability to reduce preterm birth rates with periodontal treatment, it is important to recognize the value of good oral health for all adults, including pregnant women.

The Disease and Its Effects

Periodontal disease often evolves from untreated gingivitis, which causes the gums to redden, swell, and bleed more easily. Bacterial plaque on the surface of the teeth spreads and grows below the gum line (dentistry speaks of a subgingival biofilm), adding to progressive gram-negative anaerobic infection of the mouth and inflammatory responses that ultimately lead to the destruction of tissue and bone.

As Dr. Kim A. Boggess has described in numerous articles on periodontal disease in pregnancy, damage occurs both directly from bacteria in plaque and indirectly through bacterial stimulation of local and systemic inflammatory and immune responses.

Interestingly, there is no single validated definition of periodontal disease. Instead, the clinical criteria used to define periodontal disease have varied among studies, which can make all the data difficult to interpret. Some investigators have focused on the magnitude and extent of attachment loss or other clinical measures of periodontal disease, whereas others hone in on measures of infection and host response to oral bacteria. There are commonly agreed upon clinical markers, however, including gingival recession, tooth attachment loss, and bleeding on gingival probing.

Much of the research into the role of maternal oral health in pregnancy outcomes has been driven by appreciation of the importance that oral health plays in overall general health, and by a growing recognition that periodontal disease can trigger chronic, systemic inflammation, which in turn can drive various disease processes.

The conditions most often associated with periodontal disease are cardiovascular disease and diabetes. Some studies published in the last decade have shown, for instance, that individuals with periodontal disease have at least a 1.5-fold increased risk of developing cardiovascular disease. There also is some evidence that treating periodontal disease can improve various measures of cardiovascular function – such as blood pressure and levels of inflammatory cytokines. In addition, some data suggest that periodontal treatment results in better diabetic control.

Maternal periodontal disease also has been associated with other adverse pregnancy outcomes such as preeclampsia, gestational diabetes, fetal loss, and low birth weight. In a “clinical expert series” on maternal oral health in pregnancy published in 2008, Dr. Boggess provides a comprehensive summary of the literature on these associations, and details why good oral health should be a goal for all individuals, including pregnant women (Obstet. Gynecol. 2008;111:976–86).

Treatment and Preterm Birth

While some of the initial studies of periodontal treatment in pregnancy were promising, suggesting that treatment may reduce the risk for preterm birth, we now have three large studies in the United States that have been negative. Each has involved randomization to active treatment with scaling and root planing or placebo treatment, and each has shown no significant difference in preterm birth between the two groups.

In the multicenter Periodontal Infections and Prematurity Study (PIPS) trial reported early this year, we screened more than 3,500 women between 6 and 20 weeks' gestation and found a prevalence of periodontal disease of 50%. (We defined periodontal disease as attachment loss of at least 3 mm on at least three teeth. Moderate to severe disease was defined as attachment loss of 5 mm or more on three or more teeth.)

The 756 women with periodontal disease who returned for the scheduled treatment visit were then randomly assigned in a 1:1 ratio to active treatment or placebo (superficial cleaning). The mean gestational age at screening was 13.1 weeks, and the mean gestational age at treatment was 16.5 weeks. The groups were balanced with respect to gestational age, periodontal disease severity, and history of preterm delivery (Am. J. Obstet. Gynecol. 2010;202:147.e1–8).

There was no significant difference between the two treatment groups in the incidence of spontaneous preterm birth at less than 35 weeks' gestation (our primary end point) or at less than 37 weeks' gestation. We also saw no difference in mean birth weight or the proportion of low-birth-weight or very-low-birth-weight newborns. There also was no difference in composite neonatal morbidity/mortality between the groups.

These findings are largely concordant with those of two other recent studies. In one study published in 2006, more than 800 women were randomly assigned to receive either antepartum periodontal treatment (before 21 weeks' gestation) or postpartum treatment (control). Periodontal treatment improved measures of periodontitis but did not significantly alter the risk of preterm delivery at less than 37 weeks' gestation (N. Engl. J. Med. 2006;355: 1885–94).

The other study – coined the MOTOR study (Maternal Oral Therapy to Reduce Obstetric Risk) – randomized more than 1,800 patients at three sites to periodontal treatment early in the second trimester or delayed treatment after delivery. Again, investigators demonstrated improvements in oral health after treatment, but found no significant reduction in preterm birth at less than 37 weeks of gestation (Obstet. Gynecol. 2009;114:551–9).

Current Thinking

What should we do in the wake of these negative findings?

First, we must realize that periodontal treatment in these trials improved the oral health of pregnant women, and that the benefits of good oral health cannot be disputed. Secondly, we must still appreciate – and share with our patients – that periodontal disease is very common and does appear to be associated with preterm birth (and possibly other adverse pregnancy outcomes), as well as with other negative health outcomes such as cardiovascular disease and diabetes.

We should be careful, however, and be sure to tell patients that treatment of periodontal disease alone does not appear to reduce the risk of preterm birth.

We need to study these associations further and better understand the mechanisms of periodontal disease–associated preterm birth. There also are unanswered questions about treatment. For example, is it possible that treatment prior to pregnancy may reduce the risk of preterm birth? Is it possible that using adjuvant antibiotic mouthwash may improve pregnancy outcomes? Questions such as these should be answered with additional clinical trials.

We also must better understand and delineate reported disparities in oral health. Periodontal disease disproportionately affects racial and ethnic minorities and those of low socioeconomic status. While differences in access to care and other behaviors and practices likely play a role in these disparities, experts believe that there also may be population differences in oral microbiology or inflammatory responses to bacterial colonization.

As we wait for more information, we can tell our patients about the importance of good oral health, and we can reassure them that periodontal disease treatment in pregnancy appears to be safe. We are not ready, however, to recommend routine screening and treatment of periodontal disease in pregnancy to improve pregnancy outcomes.

DR. MACONES said he has no disclosures relevant to this article. E-mail him at

[email protected]

Infection's Relationship to Prematurity

The United States spends almost 18% of its gross domestic product on health care, yet its infant mortality rate is higher than that in most other developed countries. The latest available data show the United States ranking 29th in the world in infant mortality.

One may ask why the United States continues to have this asynchrony between its investments and such an adverse health outcome. One way to assess this is to examine the factors that contribute most significantly to infant mortality: prematurity and birth defects. Prematurity remains a vexing problem in the United States – one for which the mechanism and the treatment remain, at best, elusive.

Infection or inflammation is considered to play a dominant role in the pathogenesis of prematurity. Data to support this role have been generated from a number of controlled, uncontrolled, and even laboratory studies. Most recently, additional studies have shown that inflammation or infection occurring within body cavities, including the vagina (bacterial vaginosis) or the oral cavity (periodontal disease) are associated with increased rates of prematurity.

The conundrum that we find ourselves in at this point is that there does not appear to be an effective means of altering the status of infection or inflammation in order to have a direct impact on prematurity rates. The studies so far have been controversial, leaving obstetricians very confused as to how they can best intervene and improve the perinatal outcome.

It is because of this very difficult situation that we believe it is important to have a Master Class that examines the relationship between infection – most significantly, periodontal infection – and the outcome of prematurity, and the options that can be exercised at this time with regard to oral health, prenatal care, and management pending definitive answers.

We have invited Dr. George A. Macones, an expert in maternal-fetal medicine who has extensively studied the prediction and prevention of prematurity, to serve as our guest author. Dr. Macones is the Mitchell and Elaine Yanow Professor and chair of the department of obstetrics and gynecology at Washington University, St. Louis. In this column, Dr. Macones details the value of counseling our patients about good oral health.

During the last 10–15 years, in an effort to improve troubling rates of spontaneous preterm delivery and other adverse pregnancy outcomes, investigators have looked at many kinds of clinical and subclinical infections and explored their possible associations to preterm birth.

Bacterial vaginosis is one infection that has been associated in numerous studies with a higher risk of preterm birth. Periodontal disease is another. While not all studies have found an association, there is substantial evidence – mainly from observational and epidemiologic studies – linking periodontal disease to spontaneous preterm birth and identifying the disease as a probable risk factor for preterm delivery.

One of the larger studies was a prospective cohort study involving more than 1,300 pregnant women who were enrolled at 21–24 weeks' gestation and provided information on various possible risk factors for preterm birth. Later analyses showed that women with moderate to severe periodontal disease were 4.5 times as likely to deliver spontaneously before 37 weeks' gestation, 5.3 times as likely to deliver before 35 weeks' gestation, and 7.1 times as likely to deliver before 32 weeks (J. Am. Dent. Assoc. 2001;132:875–80).

Other published studies report lower levels of risk, and a more recent meta-analysis that included 17 studies and more than 7,000 women suggested a 2.8-fold increased risk of preterm birth in women with periodontal disease (Am J. Obstet. Gynecol. 2007;196:135.e1–7).

Today, interestingly, we know that bacterial vaginosis and periodontal disease each present our patients with a similar magnitude of increased risk for preterm delivery: a two- to threefold increased risk.

Unfortunately, hopes that identifying and treating the conditions could reduce risk and improve pregnancy outcomes have been dashed – in both cases. In the case of periodontal disease, three major randomized controlled trials in the United States – including the Periodontal Infections and Prematurity Study (PIPS) published in February of this year – have provided evidence that screening and treating periodontal disease during pregnancy are not likely to reduce rates of preterm birth.

This does not mean, however, that we should ignore the problem of periodontal disease. It is a huge problem, affecting up to 40% of pregnant women according to most reports, and there is no evidence to suggest that dental examinations or treatment are deleterious during pregnancy. In all the studies that have been done over the last decade or so, there is nothing to suggest that we shouldn't look for periodontal disease and treat it.

Periodontal disease is clearly associated with other poor health outcomes, in addition to its association with preterm birth, and study after study has shown that good oral health is important for good overall health.

Despite our inability to reduce preterm birth rates with periodontal treatment, it is important to recognize the value of good oral health for all adults, including pregnant women.

The Disease and Its Effects

Periodontal disease often evolves from untreated gingivitis, which causes the gums to redden, swell, and bleed more easily. Bacterial plaque on the surface of the teeth spreads and grows below the gum line (dentistry speaks of a subgingival biofilm), adding to progressive gram-negative anaerobic infection of the mouth and inflammatory responses that ultimately lead to the destruction of tissue and bone.

As Dr. Kim A. Boggess has described in numerous articles on periodontal disease in pregnancy, damage occurs both directly from bacteria in plaque and indirectly through bacterial stimulation of local and systemic inflammatory and immune responses.

Interestingly, there is no single validated definition of periodontal disease. Instead, the clinical criteria used to define periodontal disease have varied among studies, which can make all the data difficult to interpret. Some investigators have focused on the magnitude and extent of attachment loss or other clinical measures of periodontal disease, whereas others hone in on measures of infection and host response to oral bacteria. There are commonly agreed upon clinical markers, however, including gingival recession, tooth attachment loss, and bleeding on gingival probing.

Much of the research into the role of maternal oral health in pregnancy outcomes has been driven by appreciation of the importance that oral health plays in overall general health, and by a growing recognition that periodontal disease can trigger chronic, systemic inflammation, which in turn can drive various disease processes.

The conditions most often associated with periodontal disease are cardiovascular disease and diabetes. Some studies published in the last decade have shown, for instance, that individuals with periodontal disease have at least a 1.5-fold increased risk of developing cardiovascular disease. There also is some evidence that treating periodontal disease can improve various measures of cardiovascular function – such as blood pressure and levels of inflammatory cytokines. In addition, some data suggest that periodontal treatment results in better diabetic control.

Maternal periodontal disease also has been associated with other adverse pregnancy outcomes such as preeclampsia, gestational diabetes, fetal loss, and low birth weight. In a “clinical expert series” on maternal oral health in pregnancy published in 2008, Dr. Boggess provides a comprehensive summary of the literature on these associations, and details why good oral health should be a goal for all individuals, including pregnant women (Obstet. Gynecol. 2008;111:976–86).

Treatment and Preterm Birth

While some of the initial studies of periodontal treatment in pregnancy were promising, suggesting that treatment may reduce the risk for preterm birth, we now have three large studies in the United States that have been negative. Each has involved randomization to active treatment with scaling and root planing or placebo treatment, and each has shown no significant difference in preterm birth between the two groups.

In the multicenter Periodontal Infections and Prematurity Study (PIPS) trial reported early this year, we screened more than 3,500 women between 6 and 20 weeks' gestation and found a prevalence of periodontal disease of 50%. (We defined periodontal disease as attachment loss of at least 3 mm on at least three teeth. Moderate to severe disease was defined as attachment loss of 5 mm or more on three or more teeth.)

The 756 women with periodontal disease who returned for the scheduled treatment visit were then randomly assigned in a 1:1 ratio to active treatment or placebo (superficial cleaning). The mean gestational age at screening was 13.1 weeks, and the mean gestational age at treatment was 16.5 weeks. The groups were balanced with respect to gestational age, periodontal disease severity, and history of preterm delivery (Am. J. Obstet. Gynecol. 2010;202:147.e1–8).

There was no significant difference between the two treatment groups in the incidence of spontaneous preterm birth at less than 35 weeks' gestation (our primary end point) or at less than 37 weeks' gestation. We also saw no difference in mean birth weight or the proportion of low-birth-weight or very-low-birth-weight newborns. There also was no difference in composite neonatal morbidity/mortality between the groups.

These findings are largely concordant with those of two other recent studies. In one study published in 2006, more than 800 women were randomly assigned to receive either antepartum periodontal treatment (before 21 weeks' gestation) or postpartum treatment (control). Periodontal treatment improved measures of periodontitis but did not significantly alter the risk of preterm delivery at less than 37 weeks' gestation (N. Engl. J. Med. 2006;355: 1885–94).

The other study – coined the MOTOR study (Maternal Oral Therapy to Reduce Obstetric Risk) – randomized more than 1,800 patients at three sites to periodontal treatment early in the second trimester or delayed treatment after delivery. Again, investigators demonstrated improvements in oral health after treatment, but found no significant reduction in preterm birth at less than 37 weeks of gestation (Obstet. Gynecol. 2009;114:551–9).

Current Thinking

What should we do in the wake of these negative findings?

First, we must realize that periodontal treatment in these trials improved the oral health of pregnant women, and that the benefits of good oral health cannot be disputed. Secondly, we must still appreciate – and share with our patients – that periodontal disease is very common and does appear to be associated with preterm birth (and possibly other adverse pregnancy outcomes), as well as with other negative health outcomes such as cardiovascular disease and diabetes.

We should be careful, however, and be sure to tell patients that treatment of periodontal disease alone does not appear to reduce the risk of preterm birth.

We need to study these associations further and better understand the mechanisms of periodontal disease–associated preterm birth. There also are unanswered questions about treatment. For example, is it possible that treatment prior to pregnancy may reduce the risk of preterm birth? Is it possible that using adjuvant antibiotic mouthwash may improve pregnancy outcomes? Questions such as these should be answered with additional clinical trials.

We also must better understand and delineate reported disparities in oral health. Periodontal disease disproportionately affects racial and ethnic minorities and those of low socioeconomic status. While differences in access to care and other behaviors and practices likely play a role in these disparities, experts believe that there also may be population differences in oral microbiology or inflammatory responses to bacterial colonization.

As we wait for more information, we can tell our patients about the importance of good oral health, and we can reassure them that periodontal disease treatment in pregnancy appears to be safe. We are not ready, however, to recommend routine screening and treatment of periodontal disease in pregnancy to improve pregnancy outcomes.

DR. MACONES said he has no disclosures relevant to this article. E-mail him at

[email protected]

Infection's Relationship to Prematurity

The United States spends almost 18% of its gross domestic product on health care, yet its infant mortality rate is higher than that in most other developed countries. The latest available data show the United States ranking 29th in the world in infant mortality.

One may ask why the United States continues to have this asynchrony between its investments and such an adverse health outcome. One way to assess this is to examine the factors that contribute most significantly to infant mortality: prematurity and birth defects. Prematurity remains a vexing problem in the United States – one for which the mechanism and the treatment remain, at best, elusive.

Infection or inflammation is considered to play a dominant role in the pathogenesis of prematurity. Data to support this role have been generated from a number of controlled, uncontrolled, and even laboratory studies. Most recently, additional studies have shown that inflammation or infection occurring within body cavities, including the vagina (bacterial vaginosis) or the oral cavity (periodontal disease) are associated with increased rates of prematurity.

The conundrum that we find ourselves in at this point is that there does not appear to be an effective means of altering the status of infection or inflammation in order to have a direct impact on prematurity rates. The studies so far have been controversial, leaving obstetricians very confused as to how they can best intervene and improve the perinatal outcome.

It is because of this very difficult situation that we believe it is important to have a Master Class that examines the relationship between infection – most significantly, periodontal infection – and the outcome of prematurity, and the options that can be exercised at this time with regard to oral health, prenatal care, and management pending definitive answers.

We have invited Dr. George A. Macones, an expert in maternal-fetal medicine who has extensively studied the prediction and prevention of prematurity, to serve as our guest author. Dr. Macones is the Mitchell and Elaine Yanow Professor and chair of the department of obstetrics and gynecology at Washington University, St. Louis. In this column, Dr. Macones details the value of counseling our patients about good oral health.

During the last 10–15 years, in an effort to improve troubling rates of spontaneous preterm delivery and other adverse pregnancy outcomes, investigators have looked at many kinds of clinical and subclinical infections and explored their possible associations to preterm birth.

Bacterial vaginosis is one infection that has been associated in numerous studies with a higher risk of preterm birth. Periodontal disease is another. While not all studies have found an association, there is substantial evidence – mainly from observational and epidemiologic studies – linking periodontal disease to spontaneous preterm birth and identifying the disease as a probable risk factor for preterm delivery.

One of the larger studies was a prospective cohort study involving more than 1,300 pregnant women who were enrolled at 21–24 weeks' gestation and provided information on various possible risk factors for preterm birth. Later analyses showed that women with moderate to severe periodontal disease were 4.5 times as likely to deliver spontaneously before 37 weeks' gestation, 5.3 times as likely to deliver before 35 weeks' gestation, and 7.1 times as likely to deliver before 32 weeks (J. Am. Dent. Assoc. 2001;132:875–80).

Other published studies report lower levels of risk, and a more recent meta-analysis that included 17 studies and more than 7,000 women suggested a 2.8-fold increased risk of preterm birth in women with periodontal disease (Am J. Obstet. Gynecol. 2007;196:135.e1–7).

Today, interestingly, we know that bacterial vaginosis and periodontal disease each present our patients with a similar magnitude of increased risk for preterm delivery: a two- to threefold increased risk.

Unfortunately, hopes that identifying and treating the conditions could reduce risk and improve pregnancy outcomes have been dashed – in both cases. In the case of periodontal disease, three major randomized controlled trials in the United States – including the Periodontal Infections and Prematurity Study (PIPS) published in February of this year – have provided evidence that screening and treating periodontal disease during pregnancy are not likely to reduce rates of preterm birth.

This does not mean, however, that we should ignore the problem of periodontal disease. It is a huge problem, affecting up to 40% of pregnant women according to most reports, and there is no evidence to suggest that dental examinations or treatment are deleterious during pregnancy. In all the studies that have been done over the last decade or so, there is nothing to suggest that we shouldn't look for periodontal disease and treat it.

Periodontal disease is clearly associated with other poor health outcomes, in addition to its association with preterm birth, and study after study has shown that good oral health is important for good overall health.

Despite our inability to reduce preterm birth rates with periodontal treatment, it is important to recognize the value of good oral health for all adults, including pregnant women.

The Disease and Its Effects

Periodontal disease often evolves from untreated gingivitis, which causes the gums to redden, swell, and bleed more easily. Bacterial plaque on the surface of the teeth spreads and grows below the gum line (dentistry speaks of a subgingival biofilm), adding to progressive gram-negative anaerobic infection of the mouth and inflammatory responses that ultimately lead to the destruction of tissue and bone.

As Dr. Kim A. Boggess has described in numerous articles on periodontal disease in pregnancy, damage occurs both directly from bacteria in plaque and indirectly through bacterial stimulation of local and systemic inflammatory and immune responses.

Interestingly, there is no single validated definition of periodontal disease. Instead, the clinical criteria used to define periodontal disease have varied among studies, which can make all the data difficult to interpret. Some investigators have focused on the magnitude and extent of attachment loss or other clinical measures of periodontal disease, whereas others hone in on measures of infection and host response to oral bacteria. There are commonly agreed upon clinical markers, however, including gingival recession, tooth attachment loss, and bleeding on gingival probing.

Much of the research into the role of maternal oral health in pregnancy outcomes has been driven by appreciation of the importance that oral health plays in overall general health, and by a growing recognition that periodontal disease can trigger chronic, systemic inflammation, which in turn can drive various disease processes.

The conditions most often associated with periodontal disease are cardiovascular disease and diabetes. Some studies published in the last decade have shown, for instance, that individuals with periodontal disease have at least a 1.5-fold increased risk of developing cardiovascular disease. There also is some evidence that treating periodontal disease can improve various measures of cardiovascular function – such as blood pressure and levels of inflammatory cytokines. In addition, some data suggest that periodontal treatment results in better diabetic control.

Maternal periodontal disease also has been associated with other adverse pregnancy outcomes such as preeclampsia, gestational diabetes, fetal loss, and low birth weight. In a “clinical expert series” on maternal oral health in pregnancy published in 2008, Dr. Boggess provides a comprehensive summary of the literature on these associations, and details why good oral health should be a goal for all individuals, including pregnant women (Obstet. Gynecol. 2008;111:976–86).

Treatment and Preterm Birth

While some of the initial studies of periodontal treatment in pregnancy were promising, suggesting that treatment may reduce the risk for preterm birth, we now have three large studies in the United States that have been negative. Each has involved randomization to active treatment with scaling and root planing or placebo treatment, and each has shown no significant difference in preterm birth between the two groups.

In the multicenter Periodontal Infections and Prematurity Study (PIPS) trial reported early this year, we screened more than 3,500 women between 6 and 20 weeks' gestation and found a prevalence of periodontal disease of 50%. (We defined periodontal disease as attachment loss of at least 3 mm on at least three teeth. Moderate to severe disease was defined as attachment loss of 5 mm or more on three or more teeth.)

The 756 women with periodontal disease who returned for the scheduled treatment visit were then randomly assigned in a 1:1 ratio to active treatment or placebo (superficial cleaning). The mean gestational age at screening was 13.1 weeks, and the mean gestational age at treatment was 16.5 weeks. The groups were balanced with respect to gestational age, periodontal disease severity, and history of preterm delivery (Am. J. Obstet. Gynecol. 2010;202:147.e1–8).

There was no significant difference between the two treatment groups in the incidence of spontaneous preterm birth at less than 35 weeks' gestation (our primary end point) or at less than 37 weeks' gestation. We also saw no difference in mean birth weight or the proportion of low-birth-weight or very-low-birth-weight newborns. There also was no difference in composite neonatal morbidity/mortality between the groups.

These findings are largely concordant with those of two other recent studies. In one study published in 2006, more than 800 women were randomly assigned to receive either antepartum periodontal treatment (before 21 weeks' gestation) or postpartum treatment (control). Periodontal treatment improved measures of periodontitis but did not significantly alter the risk of preterm delivery at less than 37 weeks' gestation (N. Engl. J. Med. 2006;355: 1885–94).

The other study – coined the MOTOR study (Maternal Oral Therapy to Reduce Obstetric Risk) – randomized more than 1,800 patients at three sites to periodontal treatment early in the second trimester or delayed treatment after delivery. Again, investigators demonstrated improvements in oral health after treatment, but found no significant reduction in preterm birth at less than 37 weeks of gestation (Obstet. Gynecol. 2009;114:551–9).

Current Thinking

What should we do in the wake of these negative findings?

First, we must realize that periodontal treatment in these trials improved the oral health of pregnant women, and that the benefits of good oral health cannot be disputed. Secondly, we must still appreciate – and share with our patients – that periodontal disease is very common and does appear to be associated with preterm birth (and possibly other adverse pregnancy outcomes), as well as with other negative health outcomes such as cardiovascular disease and diabetes.

We should be careful, however, and be sure to tell patients that treatment of periodontal disease alone does not appear to reduce the risk of preterm birth.

We need to study these associations further and better understand the mechanisms of periodontal disease–associated preterm birth. There also are unanswered questions about treatment. For example, is it possible that treatment prior to pregnancy may reduce the risk of preterm birth? Is it possible that using adjuvant antibiotic mouthwash may improve pregnancy outcomes? Questions such as these should be answered with additional clinical trials.

We also must better understand and delineate reported disparities in oral health. Periodontal disease disproportionately affects racial and ethnic minorities and those of low socioeconomic status. While differences in access to care and other behaviors and practices likely play a role in these disparities, experts believe that there also may be population differences in oral microbiology or inflammatory responses to bacterial colonization.

As we wait for more information, we can tell our patients about the importance of good oral health, and we can reassure them that periodontal disease treatment in pregnancy appears to be safe. We are not ready, however, to recommend routine screening and treatment of periodontal disease in pregnancy to improve pregnancy outcomes.

DR. MACONES said he has no disclosures relevant to this article. E-mail him at

[email protected]

Infection's Relationship to Prematurity

The United States spends almost 18% of its gross domestic product on health care, yet its infant mortality rate is higher than that in most other developed countries. The latest available data show the United States ranking 29th in the world in infant mortality.

One may ask why the United States continues to have this asynchrony between its investments and such an adverse health outcome. One way to assess this is to examine the factors that contribute most significantly to infant mortality: prematurity and birth defects. Prematurity remains a vexing problem in the United States – one for which the mechanism and the treatment remain, at best, elusive.

Infection or inflammation is considered to play a dominant role in the pathogenesis of prematurity. Data to support this role have been generated from a number of controlled, uncontrolled, and even laboratory studies. Most recently, additional studies have shown that inflammation or infection occurring within body cavities, including the vagina (bacterial vaginosis) or the oral cavity (periodontal disease) are associated with increased rates of prematurity.

The conundrum that we find ourselves in at this point is that there does not appear to be an effective means of altering the status of infection or inflammation in order to have a direct impact on prematurity rates. The studies so far have been controversial, leaving obstetricians very confused as to how they can best intervene and improve the perinatal outcome.

It is because of this very difficult situation that we believe it is important to have a Master Class that examines the relationship between infection – most significantly, periodontal infection – and the outcome of prematurity, and the options that can be exercised at this time with regard to oral health, prenatal care, and management pending definitive answers.

We have invited Dr. George A. Macones, an expert in maternal-fetal medicine who has extensively studied the prediction and prevention of prematurity, to serve as our guest author. Dr. Macones is the Mitchell and Elaine Yanow Professor and chair of the department of obstetrics and gynecology at Washington University, St. Louis. In this column, Dr. Macones details the value of counseling our patients about good oral health.

The sacrospinous ligament suspension technique was first described by Karl Richter in 1968 and later introduced into the United States by David H. Nichols and Clyde L. Randall in 1971. It has been and continues to be an effective technique for apical suspension via the vaginal route, and is a valuable addition to the surgical armamentarium of the gynecologic surgeon.

In the 1990s, the procedure was done less frequently because of the popularity of uterosacral ligament suspension. Recently, however, sacrospinous ligament suspension has regained popularity for various reasons. The uterosacral ligament technique, for one, requires peritoneal entry, and the ligament is often of variable strength and also can sometimes be difficult to identify.

In addition, new tools and variations in technique, such as use of the Capio needle driver, have made sacrospinous ligament suspension easier and safer. Finally, the popularity of vaginal mesh procedures has created renewed interest in sacrospinous suspension as a direct visualization attachment technique for apical mesh, compared with trocar/needle-based techniques, which involve blind passage and possible injury to the bowel and bladder.

With proper technique, the procedure is safe, effective, and durable and has few complications related to future sexual function. Long-term success rates have been excellent in properly selected patients.

Indications

Various techniques of apical suspension are available to the gynecologic surgeon. Sacrospinous suspension is indicated in patients with adequate vaginal length who desire a vaginal procedure. An office-based exam should be performed to assess vaginal length and location/severity of prolapse.

Oftentimes, the procedure can be performed using the traditional technique with attachment of the vaginal mucosa or with mesh augmentation using the sacrospinous sutures as the apical mesh attachment points. In my practice, the procedure is contraindicated in patients with a short vagina, chronic pelvic pain, or any history of sciatica.

Prior to Surgery

Vaginal exam prior to initial dissection is helpful in ensuring that the vagina is of adequate length to reach the sacrospinous ligament. Marking of the vaginal apex for placement of suspension sutures sites also is helpful. The vagina is reapproximated to either or both sacrospinous ligaments using an Allis clamp, which is then adjusted in order to maximize vaginal length and reapproximation to the corresponding ligament. The location of the Allis clamp is then tagged with a full thickness marking suture.

Surgical Dissection

The procedure begins with entry into the sacrospinous space. Traditionally, this dissection has been described through a posterior vaginal mucosal incision associated with rectocele repair. A midline incision is made from the perineal body to the vaginal apex. The vaginal mucosa is then dissected off the underlying rectovaginal septum distally and any enterocele proximally. In the upper third of the vagina, lateral dissection is extended in the pararectal space until areolar tissue is encountered. Blunt dissection is then performed toward the ischial spine in a back-and-forth manner.

The relevant anatomy including the ischial spine, the sacrospinous ligament, the coccygeus muscle, and the lateral side wall with White's line, is identified (Fig. A). An identical dissection is performed on the contralateral side.

An anterior approach to sacrospinous suspension was described by Peter K. Sands et al. in 2000. This is especially helpful if the patient has only anterior and apical defects without the need for rectocele dissection or is undergoing an anterior mesh augmentation procedure. The anterior vaginal wall is opened, and the endopelvic connective tissue is separated from the pubic ramus at the level of the bladder neck to the ischial spine, exposing the paravesical and pararectal space. The sacrospinous ligament is identified and isolated through this defect.

Perhaps the easiest method of entering the sacrospinous space is through a midcompartment approach just lateral to the enterocele. This is often described with isolated apical/enterocele defects. The vaginal mucosa over the apex/enterocele is incised in the midline. The edges of the incision are grasped using Allis clamps, and lateral dissection is performed between the vaginal mucosa and enterocele sac until loose areolar tissue is noted. Blunt finger dissection in a back-and-forth motion is performed to the ischial spine.

An identical procedure is then performed on the contralateral side. Such midcompartment dissection is associated with very little bleeding and quick access to the sacrospinous space.

Suture Placement

A variety of tools and techniques have been described to place the sacrospinous suspension sutures. Traditionally, suture placement has been described using a standard needle holder, Miya hook, Des-Champs ligature, Shutt punch, or Nichols-Veronikis ligature carrier. Each device is loaded with the suture/needle.

Vaginal retractors (I prefer the Breisky-Navratril retractors) are used to gain exposure to the sacrospinous space until the ligament is visualized. The suture is placed around the sacrospinous ligament approximately two fingerbreadths medial to the ischial spine, with care given to avoid injury to the pudendal neurovascular bundle (Fig. B).

A permanent suture (Ethibond or Gore-Tex) is used for a pulley stitch attachment, while a delayed absorbable suture such as polydioxanone (PDS) is used for a full thickness vaginal attachment. A second suture may be placed just slightly medial to the first at the surgeon's discretion. Bilateral sacrospinous sutures also could be placed. Bilateral suspension sutures are especially useful when considering mesh augmentation of the anterior and/or posterior segment.

More recently, traditional devices have been replaced with the Capio needle driver. This is a disposable multiuse suture retrieval device which makes sacrospinous ligament suspension significantly easier, faster, and safer. The device has a medium caliber shaft with a plunger for suture application. The end has a hook which allows push-catch retrieval of a small needle-based suture. Various permanent and delayed absorbable sutures are available. Under direct finger guidance, the device is used to hook the sacrospinous ligament at the appropriate location. Depression of the plunger passes the needle through the ligament, and the needle is then retrieved by fins on the other side. Removal of the device completes placement of the suture.

One of the true benefits of the Capio needle driver is the ability to perform suture placement under direct finger guidance without the need for visualization using retractors – a benefit that minimizes the extent of dissection and the time involved. In my opinion, this device has revolutionized sacrospinous suspension by allowing more physicians to perform the procedure safely and effectively.

The next evolution in sacrospinous suspension will include anchor-based single-point attachment – an approach that has recently become available and may supplant traditional suture placement, which can potentially strangulate tissue and result in postoperative pain. Additional clinical experience is required before this technique can be supported, but initial results are encouraging, especially with respect to postoperative sacrospinous pain.

Completion of Procedure

Once the suture(s) are in place, a rectal exam is recommended to exclude unintentional rectal injury or suture placement. Once confirmed, tie-down of the sutures can be completed. In cases of simple sacrospinous suspension, the suture is taken through the vagina at the apex marked prior to initial incision. Two techniques are available for this.

The traditional pulley technique with permanent suture is taken through full thickness vagina excluding the epithelium, and then tied down prior to closure of the mucosal incision, thereby burying the knot under the mucosa. This sometimes can be technically confusing and difficult, and may reduce the strength of vaginal attachment. The benefit of this technique is use of a permanent suspension suture.

An alternative technique utilizes delayed absorbable suture and involves both arms of the suture being taken through the full thickness vaginal mucosa at the apex (Fig. C, D). The mucosal incision is then closed followed by suture(s) tie-down (Fig. E, F)F I prefer this technique as it is technically easier and allows full thickness attachment of the vagina. More importantly, it gives the surgeon easy access to the suspension sutures if the sutures need to be removed in the postoperative period in cases of persistent postoperative pain.

Regardless of technique, it is important to tie down the suture securely, but not tightly, as strangulation of the sacrospinous tissue may increase the chances of postoperative pain.

I prefer bilateral sacrospinous suspension sutures with only one attachment on each side in order to minimize deviation of the vagina to one side as well as to maximize support. A single suture on each side also removes any confusion over which suture may be involved in uncommon intraoperative complications such as rectal penetration or ureteral kinking.

If mesh/graft augmentation is being performed, the bilateral sacrospinous suspension sutures are taken through the apical lateral extensions of the trapezoid-shaped mesh and tied down. Although permanent suture can be used for this technique, I would recommend the use of delayed absorbable suture; permanent suture can sometimes strangulate the sacrospinous ligament with contraction of the mesh/graft over time.

After initial tie-down, these suspension sutures then can be taken through the vaginal apex as described above for further apical support. The distal ends of the mesh are then attached to the pubocervical fascia lateral to the bladder neck or the perineal body, depending on whether the procedure involves anterior or posterior mesh placement, respectively.

If mesh is used, hemostasis should be ensured when the vagina is closed using interrupted suture with little or no mucosal excision, thereby minimizing tension at the suture line and hopefully reducing the risk of postoperative mesh exposure.

A rectal exam should be done to exclude rectal injury/stitch penetration. In addition, it's important to assess the tissue bridge spanning the sacrospinous ligaments to make sure this is not too tight. A tight bridge may cause significant postoperative pain as well as defecatory dysfunction by partially obstructing the rectosigmoid. When using a mesh, this bridge can be minimized by cutting the apical transverse distance to at least 10 cm.

Cystoscopy with IV indigo carmine is performed at the end of the procedure to exclude unintentional urethral extraction/kinking. The vagina is packed at the surgeon's discretion.

Preventing, Managing Complications

With proper technique, complications associated with the sacrospinous suspension are relatively uncommon.

They can be broadly categorized as occurring intraoperatively and postoperatively, and can be largely avoided by minimizing wide dissection, by placing sutures at least 2 cm medial to the ischial spine to avoid injury to the pudendal neurovascular bundle, and by always performing a rectal exam as well as cystoscopy with IV indigo carmine following the surgical procedure.

Intraoperative Complications

Intraoperative complications can be associated with dissection into the sacrospinous space and placement of the suspension suture. Dissection-related complications include injury to the rectosigmoid as well as bleeding during dissection. It is important to make sure that dissection of the endopelvic fascia is performed sharply until a relatively avascular and areolar space is created; at this time, blunt dissection with the surgeon's finger can be easily accomplished.

Hugging the lateral side wall on each side should minimize risk of injury to the bowel. Rectal exam after placement of the suture is essential to the diagnosis of any unintentional bowel injury or suture penetration. Any confirmed rectal injury would need repair at the time of surgery.

Use of finger dissection in a back-and-forth motion rather than a sweeping up-down or side-side motion will minimize injury to the surrounding vasculature while still creating a tract large enough to place the suture. Placement of the suture can occasionally be associated with bleeding if there is any injury to the pudendal neurovascular bundle or its associated branches.

Oftentimes, tie-down of the suspension suture will control the bleeder. If there is persistent and uncontrollable bleeding, it is best not to be overly aggressive with hemostatic sutures or surgical clips, as these may result in increasing injury to the pudendal neurovascular bundle. Adequate exposure and suction are essential. Initial control of the hemostasis with pressure and tapenade for several minutes is usually successful. Placement of hemostatic agents such as Surgicel or Flo-Seal is often effective, followed by suture/clip placement if needed. Postoperative embolization for persistent bleeders has also been reported.

Placement of the suture also can sometimes be associated with ureteral kinking/obstruction. Following tie-down of the suspension sutures, cystoscopy with IV indigo carmine is recommended. If the ureter fails to spill on either side, repeat IV indigo carmine followed by ureteral stent placement is suggested. Stent placement will allow one to determine the relative site of obstruction based on how far the stent can be inserted. Typically, obstruction associated with sacrospinous sutures allows the stent to be passed 5-9 cm.

Removal of the suspension suture almost always results in resolution of the obstruction with resulting ureteral spill. A repeat suspension suture could then be placed slightly more medial at the surgeon's discretion. Repeat cystoscopy should be performed to confirm continued ureteral patency.

Postoperative Complications

Postoperative complications include hematoma/bleeding and complaints of buttock pain secondary to the involvement of the pudendal nerve branches. Bleeding should be banished accordingly. If bleeding is significant, reoperation or embolization is generally the best option. Small self-limited hematomas can be expectantly managed or drained via vaginal access as needed. It may be best to drain hematomas in cases in which mesh was placed at the time of sacrospinous suspension so as to prevent significant abscess and postoperative infection.

Mild buttock discomfort following sacrospinous suspension is not uncommon, and it is usually managed conservatively with observation, nonsteroidals, and muscle relaxants such as baclofen. The patient should be monitored on a weekly basis to ensure continued improvement.

For severe or persistent pain, removal of the suture should be considered; this is easiest if the suture was tied transvaginally rather than with the traditional pulley stitch technique. (In the latter case, suture removal involves opening the vagina.) Transvaginal excision of the suspension sutures can often be performed in the office or at the bedside with a lighted speculum and long scissors. Most patients report almost immediate relief after removal of the suture.

A. Here Briesky-Navratil retractors are used to retract the rectum medially and the bladder superiorly. B. Here is the technique of passage of a Miya hook through the ligament. Inset is the technique of retrieval of the suture.

Source Atlas of Pelvic Anatomy and Gynecologic Surgery (Philadelphia: Elsevier Health Sciences, 2006)

C. Here two sutures have been passed through the complex. D. Technique of fixing the vaginal apex to the coccygeus-sacrospinous ligament complex (C-SSL). If a pulley stitch is performed, then permanent sutures should be used. If the sutures are passed through the vaginal epithelium and tied in the vaginal lumen, then delayed absorbable sutures should be used. E. The vagina is closed prior to tying the suspension sutures. F. Tied sacrospinous sutures.

Source Atlas of Pelvic Anatomy and Gynecologic Surgery (Philadelphia: Elsevier Health Sciences, 2006)

Treating Advanced Pelvic Prolapse

www.aagl.org

It is estimated that 50% of parous women have evidence of loss of pelvic support.

Today, women with advanced pelvic organ prolapse have a number of surgical options. In a 1997 study by Olsen et al., the authors included 13 different procedures to treat advanced pelvic organ prolapse (Obstet. Gynecol. 1997;89:501-6). Since the date of that article, many more procedures – laparoscopic, robot-assisted, and vaginal – have been described and utilized to treat advanced organ prolapse. Even with the ability to use minimally invasive abdominal techniques, many physicians continue to prefer a vaginal route to correct advanced pelvic organ prolapse. They cite the fact that a vaginal approach is associated with reduced hospitalization, less postoperative pain, a faster return to normal activity, and a superior cosmetic result.

Over the past 40 years, one of the most popular procedures in the vaginal surgeon's armamentarium has been the sacrospinous ligament suspension. More recently, the procedure has been described with mesh placement as well.

We have asked an expert on this surgical technique, Dr. Neeraj Kohli, to write this Gynecologic Surgery Master Class. Dr. Kohli is a leader in the field of minimally invasive pelvic surgery and the treatment of pelvic prolapse and urinary incontinence. He is director of the division of urogynecology at Brigham and Women's Hospital and assistant professor of obstetrics and gynecology at Harvard Medical School, both in Boston.

The sacrospinous ligament suspension technique was first described by Karl Richter in 1968 and later introduced into the United States by David H. Nichols and Clyde L. Randall in 1971. It has been and continues to be an effective technique for apical suspension via the vaginal route, and is a valuable addition to the surgical armamentarium of the gynecologic surgeon.

In the 1990s, the procedure was done less frequently because of the popularity of uterosacral ligament suspension. Recently, however, sacrospinous ligament suspension has regained popularity for various reasons. The uterosacral ligament technique, for one, requires peritoneal entry, and the ligament is often of variable strength and also can sometimes be difficult to identify.

In addition, new tools and variations in technique, such as use of the Capio needle driver, have made sacrospinous ligament suspension easier and safer. Finally, the popularity of vaginal mesh procedures has created renewed interest in sacrospinous suspension as a direct visualization attachment technique for apical mesh, compared with trocar/needle-based techniques, which involve blind passage and possible injury to the bowel and bladder.

With proper technique, the procedure is safe, effective, and durable and has few complications related to future sexual function. Long-term success rates have been excellent in properly selected patients.

Indications

Various techniques of apical suspension are available to the gynecologic surgeon. Sacrospinous suspension is indicated in patients with adequate vaginal length who desire a vaginal procedure. An office-based exam should be performed to assess vaginal length and location/severity of prolapse.

Oftentimes, the procedure can be performed using the traditional technique with attachment of the vaginal mucosa or with mesh augmentation using the sacrospinous sutures as the apical mesh attachment points. In my practice, the procedure is contraindicated in patients with a short vagina, chronic pelvic pain, or any history of sciatica.

Prior to Surgery

Vaginal exam prior to initial dissection is helpful in ensuring that the vagina is of adequate length to reach the sacrospinous ligament. Marking of the vaginal apex for placement of suspension sutures sites also is helpful. The vagina is reapproximated to either or both sacrospinous ligaments using an Allis clamp, which is then adjusted in order to maximize vaginal length and reapproximation to the corresponding ligament. The location of the Allis clamp is then tagged with a full thickness marking suture.

Surgical Dissection

The procedure begins with entry into the sacrospinous space. Traditionally, this dissection has been described through a posterior vaginal mucosal incision associated with rectocele repair. A midline incision is made from the perineal body to the vaginal apex. The vaginal mucosa is then dissected off the underlying rectovaginal septum distally and any enterocele proximally. In the upper third of the vagina, lateral dissection is extended in the pararectal space until areolar tissue is encountered. Blunt dissection is then performed toward the ischial spine in a back-and-forth manner.

The relevant anatomy including the ischial spine, the sacrospinous ligament, the coccygeus muscle, and the lateral side wall with White's line, is identified (Fig. A). An identical dissection is performed on the contralateral side.

An anterior approach to sacrospinous suspension was described by Peter K. Sands et al. in 2000. This is especially helpful if the patient has only anterior and apical defects without the need for rectocele dissection or is undergoing an anterior mesh augmentation procedure. The anterior vaginal wall is opened, and the endopelvic connective tissue is separated from the pubic ramus at the level of the bladder neck to the ischial spine, exposing the paravesical and pararectal space. The sacrospinous ligament is identified and isolated through this defect.

Perhaps the easiest method of entering the sacrospinous space is through a midcompartment approach just lateral to the enterocele. This is often described with isolated apical/enterocele defects. The vaginal mucosa over the apex/enterocele is incised in the midline. The edges of the incision are grasped using Allis clamps, and lateral dissection is performed between the vaginal mucosa and enterocele sac until loose areolar tissue is noted. Blunt finger dissection in a back-and-forth motion is performed to the ischial spine.

An identical procedure is then performed on the contralateral side. Such midcompartment dissection is associated with very little bleeding and quick access to the sacrospinous space.

Suture Placement

A variety of tools and techniques have been described to place the sacrospinous suspension sutures. Traditionally, suture placement has been described using a standard needle holder, Miya hook, Des-Champs ligature, Shutt punch, or Nichols-Veronikis ligature carrier. Each device is loaded with the suture/needle.

Vaginal retractors (I prefer the Breisky-Navratril retractors) are used to gain exposure to the sacrospinous space until the ligament is visualized. The suture is placed around the sacrospinous ligament approximately two fingerbreadths medial to the ischial spine, with care given to avoid injury to the pudendal neurovascular bundle (Fig. B).

A permanent suture (Ethibond or Gore-Tex) is used for a pulley stitch attachment, while a delayed absorbable suture such as polydioxanone (PDS) is used for a full thickness vaginal attachment. A second suture may be placed just slightly medial to the first at the surgeon's discretion. Bilateral sacrospinous sutures also could be placed. Bilateral suspension sutures are especially useful when considering mesh augmentation of the anterior and/or posterior segment.

More recently, traditional devices have been replaced with the Capio needle driver. This is a disposable multiuse suture retrieval device which makes sacrospinous ligament suspension significantly easier, faster, and safer. The device has a medium caliber shaft with a plunger for suture application. The end has a hook which allows push-catch retrieval of a small needle-based suture. Various permanent and delayed absorbable sutures are available. Under direct finger guidance, the device is used to hook the sacrospinous ligament at the appropriate location. Depression of the plunger passes the needle through the ligament, and the needle is then retrieved by fins on the other side. Removal of the device completes placement of the suture.

One of the true benefits of the Capio needle driver is the ability to perform suture placement under direct finger guidance without the need for visualization using retractors – a benefit that minimizes the extent of dissection and the time involved. In my opinion, this device has revolutionized sacrospinous suspension by allowing more physicians to perform the procedure safely and effectively.

The next evolution in sacrospinous suspension will include anchor-based single-point attachment – an approach that has recently become available and may supplant traditional suture placement, which can potentially strangulate tissue and result in postoperative pain. Additional clinical experience is required before this technique can be supported, but initial results are encouraging, especially with respect to postoperative sacrospinous pain.

Completion of Procedure

Once the suture(s) are in place, a rectal exam is recommended to exclude unintentional rectal injury or suture placement. Once confirmed, tie-down of the sutures can be completed. In cases of simple sacrospinous suspension, the suture is taken through the vagina at the apex marked prior to initial incision. Two techniques are available for this.

The traditional pulley technique with permanent suture is taken through full thickness vagina excluding the epithelium, and then tied down prior to closure of the mucosal incision, thereby burying the knot under the mucosa. This sometimes can be technically confusing and difficult, and may reduce the strength of vaginal attachment. The benefit of this technique is use of a permanent suspension suture.

An alternative technique utilizes delayed absorbable suture and involves both arms of the suture being taken through the full thickness vaginal mucosa at the apex (Fig. C, D). The mucosal incision is then closed followed by suture(s) tie-down (Fig. E, F)F I prefer this technique as it is technically easier and allows full thickness attachment of the vagina. More importantly, it gives the surgeon easy access to the suspension sutures if the sutures need to be removed in the postoperative period in cases of persistent postoperative pain.

Regardless of technique, it is important to tie down the suture securely, but not tightly, as strangulation of the sacrospinous tissue may increase the chances of postoperative pain.

I prefer bilateral sacrospinous suspension sutures with only one attachment on each side in order to minimize deviation of the vagina to one side as well as to maximize support. A single suture on each side also removes any confusion over which suture may be involved in uncommon intraoperative complications such as rectal penetration or ureteral kinking.

If mesh/graft augmentation is being performed, the bilateral sacrospinous suspension sutures are taken through the apical lateral extensions of the trapezoid-shaped mesh and tied down. Although permanent suture can be used for this technique, I would recommend the use of delayed absorbable suture; permanent suture can sometimes strangulate the sacrospinous ligament with contraction of the mesh/graft over time.

After initial tie-down, these suspension sutures then can be taken through the vaginal apex as described above for further apical support. The distal ends of the mesh are then attached to the pubocervical fascia lateral to the bladder neck or the perineal body, depending on whether the procedure involves anterior or posterior mesh placement, respectively.

If mesh is used, hemostasis should be ensured when the vagina is closed using interrupted suture with little or no mucosal excision, thereby minimizing tension at the suture line and hopefully reducing the risk of postoperative mesh exposure.

A rectal exam should be done to exclude rectal injury/stitch penetration. In addition, it's important to assess the tissue bridge spanning the sacrospinous ligaments to make sure this is not too tight. A tight bridge may cause significant postoperative pain as well as defecatory dysfunction by partially obstructing the rectosigmoid. When using a mesh, this bridge can be minimized by cutting the apical transverse distance to at least 10 cm.

Cystoscopy with IV indigo carmine is performed at the end of the procedure to exclude unintentional urethral extraction/kinking. The vagina is packed at the surgeon's discretion.

Preventing, Managing Complications

With proper technique, complications associated with the sacrospinous suspension are relatively uncommon.

They can be broadly categorized as occurring intraoperatively and postoperatively, and can be largely avoided by minimizing wide dissection, by placing sutures at least 2 cm medial to the ischial spine to avoid injury to the pudendal neurovascular bundle, and by always performing a rectal exam as well as cystoscopy with IV indigo carmine following the surgical procedure.

Intraoperative Complications

Intraoperative complications can be associated with dissection into the sacrospinous space and placement of the suspension suture. Dissection-related complications include injury to the rectosigmoid as well as bleeding during dissection. It is important to make sure that dissection of the endopelvic fascia is performed sharply until a relatively avascular and areolar space is created; at this time, blunt dissection with the surgeon's finger can be easily accomplished.

Hugging the lateral side wall on each side should minimize risk of injury to the bowel. Rectal exam after placement of the suture is essential to the diagnosis of any unintentional bowel injury or suture penetration. Any confirmed rectal injury would need repair at the time of surgery.

Use of finger dissection in a back-and-forth motion rather than a sweeping up-down or side-side motion will minimize injury to the surrounding vasculature while still creating a tract large enough to place the suture. Placement of the suture can occasionally be associated with bleeding if there is any injury to the pudendal neurovascular bundle or its associated branches.

Oftentimes, tie-down of the suspension suture will control the bleeder. If there is persistent and uncontrollable bleeding, it is best not to be overly aggressive with hemostatic sutures or surgical clips, as these may result in increasing injury to the pudendal neurovascular bundle. Adequate exposure and suction are essential. Initial control of the hemostasis with pressure and tapenade for several minutes is usually successful. Placement of hemostatic agents such as Surgicel or Flo-Seal is often effective, followed by suture/clip placement if needed. Postoperative embolization for persistent bleeders has also been reported.

Placement of the suture also can sometimes be associated with ureteral kinking/obstruction. Following tie-down of the suspension sutures, cystoscopy with IV indigo carmine is recommended. If the ureter fails to spill on either side, repeat IV indigo carmine followed by ureteral stent placement is suggested. Stent placement will allow one to determine the relative site of obstruction based on how far the stent can be inserted. Typically, obstruction associated with sacrospinous sutures allows the stent to be passed 5-9 cm.

Removal of the suspension suture almost always results in resolution of the obstruction with resulting ureteral spill. A repeat suspension suture could then be placed slightly more medial at the surgeon's discretion. Repeat cystoscopy should be performed to confirm continued ureteral patency.

Postoperative Complications

Postoperative complications include hematoma/bleeding and complaints of buttock pain secondary to the involvement of the pudendal nerve branches. Bleeding should be banished accordingly. If bleeding is significant, reoperation or embolization is generally the best option. Small self-limited hematomas can be expectantly managed or drained via vaginal access as needed. It may be best to drain hematomas in cases in which mesh was placed at the time of sacrospinous suspension so as to prevent significant abscess and postoperative infection.

Mild buttock discomfort following sacrospinous suspension is not uncommon, and it is usually managed conservatively with observation, nonsteroidals, and muscle relaxants such as baclofen. The patient should be monitored on a weekly basis to ensure continued improvement.

For severe or persistent pain, removal of the suture should be considered; this is easiest if the suture was tied transvaginally rather than with the traditional pulley stitch technique. (In the latter case, suture removal involves opening the vagina.) Transvaginal excision of the suspension sutures can often be performed in the office or at the bedside with a lighted speculum and long scissors. Most patients report almost immediate relief after removal of the suture.

A. Here Briesky-Navratil retractors are used to retract the rectum medially and the bladder superiorly. B. Here is the technique of passage of a Miya hook through the ligament. Inset is the technique of retrieval of the suture.

Source Atlas of Pelvic Anatomy and Gynecologic Surgery (Philadelphia: Elsevier Health Sciences, 2006)

C. Here two sutures have been passed through the complex. D. Technique of fixing the vaginal apex to the coccygeus-sacrospinous ligament complex (C-SSL). If a pulley stitch is performed, then permanent sutures should be used. If the sutures are passed through the vaginal epithelium and tied in the vaginal lumen, then delayed absorbable sutures should be used. E. The vagina is closed prior to tying the suspension sutures. F. Tied sacrospinous sutures.

Source Atlas of Pelvic Anatomy and Gynecologic Surgery (Philadelphia: Elsevier Health Sciences, 2006)

Treating Advanced Pelvic Prolapse

www.aagl.org

It is estimated that 50% of parous women have evidence of loss of pelvic support.

Today, women with advanced pelvic organ prolapse have a number of surgical options. In a 1997 study by Olsen et al., the authors included 13 different procedures to treat advanced pelvic organ prolapse (Obstet. Gynecol. 1997;89:501-6). Since the date of that article, many more procedures – laparoscopic, robot-assisted, and vaginal – have been described and utilized to treat advanced organ prolapse. Even with the ability to use minimally invasive abdominal techniques, many physicians continue to prefer a vaginal route to correct advanced pelvic organ prolapse. They cite the fact that a vaginal approach is associated with reduced hospitalization, less postoperative pain, a faster return to normal activity, and a superior cosmetic result.

Over the past 40 years, one of the most popular procedures in the vaginal surgeon's armamentarium has been the sacrospinous ligament suspension. More recently, the procedure has been described with mesh placement as well.

We have asked an expert on this surgical technique, Dr. Neeraj Kohli, to write this Gynecologic Surgery Master Class. Dr. Kohli is a leader in the field of minimally invasive pelvic surgery and the treatment of pelvic prolapse and urinary incontinence. He is director of the division of urogynecology at Brigham and Women's Hospital and assistant professor of obstetrics and gynecology at Harvard Medical School, both in Boston.

The sacrospinous ligament suspension technique was first described by Karl Richter in 1968 and later introduced into the United States by David H. Nichols and Clyde L. Randall in 1971. It has been and continues to be an effective technique for apical suspension via the vaginal route, and is a valuable addition to the surgical armamentarium of the gynecologic surgeon.

In the 1990s, the procedure was done less frequently because of the popularity of uterosacral ligament suspension. Recently, however, sacrospinous ligament suspension has regained popularity for various reasons. The uterosacral ligament technique, for one, requires peritoneal entry, and the ligament is often of variable strength and also can sometimes be difficult to identify.

In addition, new tools and variations in technique, such as use of the Capio needle driver, have made sacrospinous ligament suspension easier and safer. Finally, the popularity of vaginal mesh procedures has created renewed interest in sacrospinous suspension as a direct visualization attachment technique for apical mesh, compared with trocar/needle-based techniques, which involve blind passage and possible injury to the bowel and bladder.

With proper technique, the procedure is safe, effective, and durable and has few complications related to future sexual function. Long-term success rates have been excellent in properly selected patients.

Indications

Various techniques of apical suspension are available to the gynecologic surgeon. Sacrospinous suspension is indicated in patients with adequate vaginal length who desire a vaginal procedure. An office-based exam should be performed to assess vaginal length and location/severity of prolapse.

Oftentimes, the procedure can be performed using the traditional technique with attachment of the vaginal mucosa or with mesh augmentation using the sacrospinous sutures as the apical mesh attachment points. In my practice, the procedure is contraindicated in patients with a short vagina, chronic pelvic pain, or any history of sciatica.

Prior to Surgery

Vaginal exam prior to initial dissection is helpful in ensuring that the vagina is of adequate length to reach the sacrospinous ligament. Marking of the vaginal apex for placement of suspension sutures sites also is helpful. The vagina is reapproximated to either or both sacrospinous ligaments using an Allis clamp, which is then adjusted in order to maximize vaginal length and reapproximation to the corresponding ligament. The location of the Allis clamp is then tagged with a full thickness marking suture.

Surgical Dissection

The procedure begins with entry into the sacrospinous space. Traditionally, this dissection has been described through a posterior vaginal mucosal incision associated with rectocele repair. A midline incision is made from the perineal body to the vaginal apex. The vaginal mucosa is then dissected off the underlying rectovaginal septum distally and any enterocele proximally. In the upper third of the vagina, lateral dissection is extended in the pararectal space until areolar tissue is encountered. Blunt dissection is then performed toward the ischial spine in a back-and-forth manner.

The relevant anatomy including the ischial spine, the sacrospinous ligament, the coccygeus muscle, and the lateral side wall with White's line, is identified (Fig. A). An identical dissection is performed on the contralateral side.

An anterior approach to sacrospinous suspension was described by Peter K. Sands et al. in 2000. This is especially helpful if the patient has only anterior and apical defects without the need for rectocele dissection or is undergoing an anterior mesh augmentation procedure. The anterior vaginal wall is opened, and the endopelvic connective tissue is separated from the pubic ramus at the level of the bladder neck to the ischial spine, exposing the paravesical and pararectal space. The sacrospinous ligament is identified and isolated through this defect.

Perhaps the easiest method of entering the sacrospinous space is through a midcompartment approach just lateral to the enterocele. This is often described with isolated apical/enterocele defects. The vaginal mucosa over the apex/enterocele is incised in the midline. The edges of the incision are grasped using Allis clamps, and lateral dissection is performed between the vaginal mucosa and enterocele sac until loose areolar tissue is noted. Blunt finger dissection in a back-and-forth motion is performed to the ischial spine.

An identical procedure is then performed on the contralateral side. Such midcompartment dissection is associated with very little bleeding and quick access to the sacrospinous space.

Suture Placement

A variety of tools and techniques have been described to place the sacrospinous suspension sutures. Traditionally, suture placement has been described using a standard needle holder, Miya hook, Des-Champs ligature, Shutt punch, or Nichols-Veronikis ligature carrier. Each device is loaded with the suture/needle.

Vaginal retractors (I prefer the Breisky-Navratril retractors) are used to gain exposure to the sacrospinous space until the ligament is visualized. The suture is placed around the sacrospinous ligament approximately two fingerbreadths medial to the ischial spine, with care given to avoid injury to the pudendal neurovascular bundle (Fig. B).

A permanent suture (Ethibond or Gore-Tex) is used for a pulley stitch attachment, while a delayed absorbable suture such as polydioxanone (PDS) is used for a full thickness vaginal attachment. A second suture may be placed just slightly medial to the first at the surgeon's discretion. Bilateral sacrospinous sutures also could be placed. Bilateral suspension sutures are especially useful when considering mesh augmentation of the anterior and/or posterior segment.

More recently, traditional devices have been replaced with the Capio needle driver. This is a disposable multiuse suture retrieval device which makes sacrospinous ligament suspension significantly easier, faster, and safer. The device has a medium caliber shaft with a plunger for suture application. The end has a hook which allows push-catch retrieval of a small needle-based suture. Various permanent and delayed absorbable sutures are available. Under direct finger guidance, the device is used to hook the sacrospinous ligament at the appropriate location. Depression of the plunger passes the needle through the ligament, and the needle is then retrieved by fins on the other side. Removal of the device completes placement of the suture.

One of the true benefits of the Capio needle driver is the ability to perform suture placement under direct finger guidance without the need for visualization using retractors – a benefit that minimizes the extent of dissection and the time involved. In my opinion, this device has revolutionized sacrospinous suspension by allowing more physicians to perform the procedure safely and effectively.

The next evolution in sacrospinous suspension will include anchor-based single-point attachment – an approach that has recently become available and may supplant traditional suture placement, which can potentially strangulate tissue and result in postoperative pain. Additional clinical experience is required before this technique can be supported, but initial results are encouraging, especially with respect to postoperative sacrospinous pain.

Completion of Procedure

Once the suture(s) are in place, a rectal exam is recommended to exclude unintentional rectal injury or suture placement. Once confirmed, tie-down of the sutures can be completed. In cases of simple sacrospinous suspension, the suture is taken through the vagina at the apex marked prior to initial incision. Two techniques are available for this.

The traditional pulley technique with permanent suture is taken through full thickness vagina excluding the epithelium, and then tied down prior to closure of the mucosal incision, thereby burying the knot under the mucosa. This sometimes can be technically confusing and difficult, and may reduce the strength of vaginal attachment. The benefit of this technique is use of a permanent suspension suture.

An alternative technique utilizes delayed absorbable suture and involves both arms of the suture being taken through the full thickness vaginal mucosa at the apex (Fig. C, D). The mucosal incision is then closed followed by suture(s) tie-down (Fig. E, F)F I prefer this technique as it is technically easier and allows full thickness attachment of the vagina. More importantly, it gives the surgeon easy access to the suspension sutures if the sutures need to be removed in the postoperative period in cases of persistent postoperative pain.

Regardless of technique, it is important to tie down the suture securely, but not tightly, as strangulation of the sacrospinous tissue may increase the chances of postoperative pain.

I prefer bilateral sacrospinous suspension sutures with only one attachment on each side in order to minimize deviation of the vagina to one side as well as to maximize support. A single suture on each side also removes any confusion over which suture may be involved in uncommon intraoperative complications such as rectal penetration or ureteral kinking.

If mesh/graft augmentation is being performed, the bilateral sacrospinous suspension sutures are taken through the apical lateral extensions of the trapezoid-shaped mesh and tied down. Although permanent suture can be used for this technique, I would recommend the use of delayed absorbable suture; permanent suture can sometimes strangulate the sacrospinous ligament with contraction of the mesh/graft over time.

After initial tie-down, these suspension sutures then can be taken through the vaginal apex as described above for further apical support. The distal ends of the mesh are then attached to the pubocervical fascia lateral to the bladder neck or the perineal body, depending on whether the procedure involves anterior or posterior mesh placement, respectively.

If mesh is used, hemostasis should be ensured when the vagina is closed using interrupted suture with little or no mucosal excision, thereby minimizing tension at the suture line and hopefully reducing the risk of postoperative mesh exposure.

A rectal exam should be done to exclude rectal injury/stitch penetration. In addition, it's important to assess the tissue bridge spanning the sacrospinous ligaments to make sure this is not too tight. A tight bridge may cause significant postoperative pain as well as defecatory dysfunction by partially obstructing the rectosigmoid. When using a mesh, this bridge can be minimized by cutting the apical transverse distance to at least 10 cm.

Cystoscopy with IV indigo carmine is performed at the end of the procedure to exclude unintentional urethral extraction/kinking. The vagina is packed at the surgeon's discretion.

Preventing, Managing Complications

With proper technique, complications associated with the sacrospinous suspension are relatively uncommon.

They can be broadly categorized as occurring intraoperatively and postoperatively, and can be largely avoided by minimizing wide dissection, by placing sutures at least 2 cm medial to the ischial spine to avoid injury to the pudendal neurovascular bundle, and by always performing a rectal exam as well as cystoscopy with IV indigo carmine following the surgical procedure.

Intraoperative Complications

Intraoperative complications can be associated with dissection into the sacrospinous space and placement of the suspension suture. Dissection-related complications include injury to the rectosigmoid as well as bleeding during dissection. It is important to make sure that dissection of the endopelvic fascia is performed sharply until a relatively avascular and areolar space is created; at this time, blunt dissection with the surgeon's finger can be easily accomplished.

Hugging the lateral side wall on each side should minimize risk of injury to the bowel. Rectal exam after placement of the suture is essential to the diagnosis of any unintentional bowel injury or suture penetration. Any confirmed rectal injury would need repair at the time of surgery.

Use of finger dissection in a back-and-forth motion rather than a sweeping up-down or side-side motion will minimize injury to the surrounding vasculature while still creating a tract large enough to place the suture. Placement of the suture can occasionally be associated with bleeding if there is any injury to the pudendal neurovascular bundle or its associated branches.

Oftentimes, tie-down of the suspension suture will control the bleeder. If there is persistent and uncontrollable bleeding, it is best not to be overly aggressive with hemostatic sutures or surgical clips, as these may result in increasing injury to the pudendal neurovascular bundle. Adequate exposure and suction are essential. Initial control of the hemostasis with pressure and tapenade for several minutes is usually successful. Placement of hemostatic agents such as Surgicel or Flo-Seal is often effective, followed by suture/clip placement if needed. Postoperative embolization for persistent bleeders has also been reported.

Placement of the suture also can sometimes be associated with ureteral kinking/obstruction. Following tie-down of the suspension sutures, cystoscopy with IV indigo carmine is recommended. If the ureter fails to spill on either side, repeat IV indigo carmine followed by ureteral stent placement is suggested. Stent placement will allow one to determine the relative site of obstruction based on how far the stent can be inserted. Typically, obstruction associated with sacrospinous sutures allows the stent to be passed 5-9 cm.

Removal of the suspension suture almost always results in resolution of the obstruction with resulting ureteral spill. A repeat suspension suture could then be placed slightly more medial at the surgeon's discretion. Repeat cystoscopy should be performed to confirm continued ureteral patency.

Postoperative Complications

Postoperative complications include hematoma/bleeding and complaints of buttock pain secondary to the involvement of the pudendal nerve branches. Bleeding should be banished accordingly. If bleeding is significant, reoperation or embolization is generally the best option. Small self-limited hematomas can be expectantly managed or drained via vaginal access as needed. It may be best to drain hematomas in cases in which mesh was placed at the time of sacrospinous suspension so as to prevent significant abscess and postoperative infection.

Mild buttock discomfort following sacrospinous suspension is not uncommon, and it is usually managed conservatively with observation, nonsteroidals, and muscle relaxants such as baclofen. The patient should be monitored on a weekly basis to ensure continued improvement.

For severe or persistent pain, removal of the suture should be considered; this is easiest if the suture was tied transvaginally rather than with the traditional pulley stitch technique. (In the latter case, suture removal involves opening the vagina.) Transvaginal excision of the suspension sutures can often be performed in the office or at the bedside with a lighted speculum and long scissors. Most patients report almost immediate relief after removal of the suture.

A. Here Briesky-Navratil retractors are used to retract the rectum medially and the bladder superiorly. B. Here is the technique of passage of a Miya hook through the ligament. Inset is the technique of retrieval of the suture.

Source Atlas of Pelvic Anatomy and Gynecologic Surgery (Philadelphia: Elsevier Health Sciences, 2006)

C. Here two sutures have been passed through the complex. D. Technique of fixing the vaginal apex to the coccygeus-sacrospinous ligament complex (C-SSL). If a pulley stitch is performed, then permanent sutures should be used. If the sutures are passed through the vaginal epithelium and tied in the vaginal lumen, then delayed absorbable sutures should be used. E. The vagina is closed prior to tying the suspension sutures. F. Tied sacrospinous sutures.

Source Atlas of Pelvic Anatomy and Gynecologic Surgery (Philadelphia: Elsevier Health Sciences, 2006)

Treating Advanced Pelvic Prolapse

www.aagl.org

It is estimated that 50% of parous women have evidence of loss of pelvic support.

Today, women with advanced pelvic organ prolapse have a number of surgical options. In a 1997 study by Olsen et al., the authors included 13 different procedures to treat advanced pelvic organ prolapse (Obstet. Gynecol. 1997;89:501-6). Since the date of that article, many more procedures – laparoscopic, robot-assisted, and vaginal – have been described and utilized to treat advanced organ prolapse. Even with the ability to use minimally invasive abdominal techniques, many physicians continue to prefer a vaginal route to correct advanced pelvic organ prolapse. They cite the fact that a vaginal approach is associated with reduced hospitalization, less postoperative pain, a faster return to normal activity, and a superior cosmetic result.

Over the past 40 years, one of the most popular procedures in the vaginal surgeon's armamentarium has been the sacrospinous ligament suspension. More recently, the procedure has been described with mesh placement as well.

We have asked an expert on this surgical technique, Dr. Neeraj Kohli, to write this Gynecologic Surgery Master Class. Dr. Kohli is a leader in the field of minimally invasive pelvic surgery and the treatment of pelvic prolapse and urinary incontinence. He is director of the division of urogynecology at Brigham and Women's Hospital and assistant professor of obstetrics and gynecology at Harvard Medical School, both in Boston.

Inherited thrombophilia and its association with both maternal thrombosis and adverse pregnancy outcomes is an issue that has come to the forefront over the past few years.

The association between inherited thrombophilia and maternal thrombosis appears to be fairly robust. Collectively, these thrombophilias account for 50%-70% of all maternal venous thrombotic events in pregnancy. Knowledge of the thrombophilic status of a patient can, therefore, have a significant impact on her clinical care. We understand better today, however, that personal and family history plays a critical role in assessing maternal thrombotic risk.

By contrast, the precise nature of the link between inherited thrombophilia and adverse pregnancy outcomes is still unclear. Over the past decade, the number of negative reports—those showing a lack of association—has increased significantly, and multiple prospective cohort studies have failed to consistently demonstrate the associations suggested by prior case-control studies that were smaller and mainly retrospective.

Collectively, this new landscape of research findings suggests that we should stop screening for inherited thrombophilia in patients with adverse pregnancy outcomes except in the setting of institutional review board–approved studies, and that we should better focus our approach to preventing maternal thrombosis through more careful, individualized risk assessment and through targeted use of antithrombotic therapy.

A New Evidence Base

Initial reports of associations between inherited thrombophilia and adverse pregnancy outcomes such as fetal loss, preeclampsia, fetal growth restriction, and abruptio placentae made some biological sense, but were based largely on small retrospective case-control studies with often inconsistent or contradictory findings.

In the case of fetal loss, numerous studies published in the 1990s and into the next decade showed a moderate association between inherited thrombophilia and stillbirth in particular.

A European retrospective cohort study published in 1996, for instance, found that the increased risk of loss among women with thrombophilia was greater after 28 weeks (odds ratio 3.6) than at or before 28 weeks (OR 1.4), and that the highest risk for stillbirth was associated with combined thrombophilic defects and antithrombin and protein C deficiencies (Lancet 1996;348:913-6).

This confusingly named study—the European Prospective Cohort on Thrombophilia (EPCOT)—involved 571 women with thrombophilia having 1,524 pregnancies, and 395 controls having 1,019 pregnancies.

In 2005, investigators of a larger case-control study nested within the 32,683-patient Nimes Obstetricians and Haematologist cohort reported an association between the factor V Leiden (FVL) mutation and pregnancy loss after 10 weeks (OR 3.5) but not between 3 and 9 weeks (J. Thromb. Haemost. 2005;3:2178-84)

A retrospective cohort study published in 2004 of 491 patients with a history of adverse pregnancy outcomes suggested, moreover, that one or more thrombophilia were actually protective of recurrent fetal losses at less than 10 weeks (Thromb. Haemost. 2004;91:290-5). However, the association of any one thrombophilia with later fetal losses was less significant in this study than in other studies (OR 1.76).

And an earlier meta-analysis of 31 studies looking at fetal loss and various thrombophilic disorders (most of them small case-control studies) concluded that FVL was associated with first-trimester pregnancy loss (OR 2.0) as well as later loss, although the association was much stronger (OR 3.3) with late, nonrecurrent fetal loss (Lancet 2003;361:901-8).

Although these and other studies suggested a link between FVL and stillbirth (and perhaps other thrombophilias and stillbirth), the absolute magnitude of the association (i.e., the absolute risk) was still very small. Moreover, over the past decade, the number of negative reports, especially amongst prospective studies, has increased—a temporal dichotomy that strongly suggests an initial bias toward positive studies and a growing comfort in reporting negative studies.

The larger prospective cohort studies reported over the last 5 years or so enerally have not found an association between inherited thrombophilia and stillbirth—or other adverse pregnancy outcomes for that matter.

For example, a 2005 study conducted by the Eunice Kennedy Shriver National Institute of Child Health and Human Development's Maternal-Fetal Medicine Units Network identified 134 FVL mutation carriers among nearly 4,900 gravidas in their first trimester of pregnancy and found no increase in fetal loss, preeclampsia, abruption, or intrauterine growth retardation (IUGR). A secondary analysis these data published earlier this year similarly found no association between the prothrombin gene G20210A mutation (PGM) and adverse pregnancy outcomes (Obstet. Gynecol. 2005;106:517-24 and Obstet. Gynecol. 2010;115:14-20).

Another prospective study of 4,250 unselected pregnancies also found no significant associations between FVL and preeclampsia, IUGR, and pregnancy loss (Br. J. Haematol. 2008;140:236-40).

Some of these findings are similar to previous reports from smaller prospective cohort studies. Investigators reported in 1999, for instance, no association between activated protein C resistance and fetal loss, preeclampsia, and IUGR. And in 2000, investigators had similarly reported a lack of association between FVL and methylenetetrahydrofolate reductase (MTHFR) polymorphism and preeclampsia or IUGR.

In general, the reported linkage between inherited thrombophilia and adverse outcomes other than stillbirth was always more tenuous. In the case of preeclampsia, however, meta-analyses of studies done before 2000 showed a fairly strong association between thrombophilia and preeclampsia, while studies published in and after 2001 found no such association.

There are a few exceptions to the lack of association found in larger prospective cohort studies. Most notably, an Australian study published this year of nulliparous women was suggestive of a weak association between the PGM and a composite index of adverse pregnancy outcomes (Obstet. Gynecol. 2010;115:5-13).

However, when investigators analyzed individual outcomes, they found that the only statistically significant associations were between the PGM and placental abruption, and between FVL and stillbirth. These associations, moreover, were based on very small sample size (nine and six patients, respectively). The investigators concluded that “the majority of asymptomatic women who carry an inherited thrombophilia polymorphism have a successful pregnancy outcome.”

There also is, at best, conflicting evidence in the literature of any benefit to heparin therapy for recurrent fetal loss.

A New Outlook on Screening

Given the evolving body of literature, it now seems wholly unjustified to screen low-risk populations. Knowing whether or not the patient has inherited thrombophilia, particularly in the nulliparous state, does not appear to be important for predicting outcomes.

There are questions that remain, however—most notably the question of whether women who have repetitive fetal losses or repetitive preeclampsia or abruptions should be screened and treated for inherited thrombophilia. Certainly, the failure of large prospective cohort studies to demonstrate any consistent association dampens our enthusiasm for the idea that inherited thrombophilia are to blame.

My opinion on this topic has evolved considerably over the last 10 years. I now believe that while screening for antiphospholipid syndrome is still warranted, screening for inherited thrombophilia in women having recurrent adverse pregnancy outcomes should occur only in the setting of an institutional review board–approved study in which ascertainment is done before a subsequent pregnancy and the patient's thrombophilia status is correlated with subsequent outcome (i.e., live birth, miscarriage, stillbirth, fetal growth restriction, preeclampsia, or abruption).

Furthermore, until we have established a definitive link between inherited thrombophilia and adverse pregnancy outcomes, we shouldn't even begin to think about clinical trials of thromboprophylaxis for affected women.

A particularly thorny question that has been raised concerns the issue of early fetal loss. Some have argued that the latest prospective cohort studies involved blood collection at or after 10 weeks' gestation and, therefore, are not relevant to conclusions drawn about the association (or lack thereof) between inherited thrombophilia and embryonic fetal loss.

However, I believe there are several reasons why we can conclude that thrombophilia and embryonic fetal losses are not linked. For one, there are enough data available from negative retrospective studies in which blood was obtained right after the pregnancy was completed. Secondly, there is no correlation between inherited thrombophilia and subsequent in vitro fertilization (IVF) failures in almost a dozen published studies. In fact, there is actually some evidence that FVL is associated with IVF success.

Lastly, we now know there is very little blood flow to the placenta before 10 weeks' gestation. There is some evidence, in fact, that hypoxia is the normal state of the embryo and may even be the preferred condition for culturing embryos in IVF.

Again, this issue requires prospective studies amongst patients with recurrent loss in which ascertainment occurs before the pregnancy commences.

Maternal Thrombotic Risk

While it's fair to say that, in general, inherited thrombophilia modestly increases the risk of maternal venous thrombotic events (VTE), it is critical to appreciate the role that a personal or strong family history of thrombosis (i.e., an affected first-degree relative) plays in determining a mother's risk.

Most women (greater than 93%) without a personal or strong family history of VTE will have uneventful pregnancies even when highly thrombogenic mutations are present. Once a personal or family history is factored in, however, the risk of VTE increases dramatically.

In the absence of a personal history of VTE or such an episode in a first-degree relative, heterozygosity for FVL or PGM is associated with a risk of thrombosis in pregnancy of well under 1% (0.2% and 0.5%, respectively). Similarly, protein C and protein S deficiencies are associated with a VTE risk under 1% in the absence of a personal or close family history.

In contrast, with a positive personal or family history, the risk of VTE in pregnancy increases to 10% in women who have heterozygosity for FVL, greater than 10% for women who have heterozygosity for PGM, 4%-17% in cases of protein C deficiency, and potentially up to 22% in cases of protein S deficiency.

Without a personal or family history, therefore, women with these lower-risk thrombophilia do not require anticoagulation during pregnancy unless they have other risk factors for thrombosis, such as significant obesity or orders for bed rest.

Patients with known inherited thrombophilia and a positive history, on the other hand, should receive antepartum thromboprophylaxis followed by postpartum anticoagulation. (Women who have a cesarean delivery should receive postpartum anticoagulation whether they have a personal or family history or not.)

Anticoagulation during pregnancy is also warranted—regardless of personal or family history—in the rare cases in which a patient is known to have homozygosity for FVL or homozygosity for PGM, or if a patient is known to have “double heterozygosity” for both FVL and PGM. Antithrombin deficiency, the most thrombogenic of all the inherited thrombophilias, also warrants antepartum anticoagulation as well as antithrombin infusions during labor and delivery.

To date, two studies have attempted to determine the value of screening for inherited thrombophilia based on a family history of prior VTE, and neither has shown that widespread screening would be particularly useful or cost effective in this setting. One certainly can argue, on the other hand, in favor of screening for thrombophilia in women who have a strong family history of VTE coupled with other risk factors.

Individualized risk assessment is always valuable. A woman with multiple risk factors—one who is obese, smokes, and is being put on bed rest, for instance—is a candidate for low-molecular-weight heparin (LMWH) therapy, for instance, even without a history of thrombosis and regardless of her thrombophilia status. If such a patient also has hypertension or preeclampsia, however, I'd be reluctant to give her either heparin or LMWH, for fear of abruption or even intracranial hemorrhage.

In what other circumstances is screening for thrombophilia warranted?

It can be justified when there is a personal history of VTE associated with a risk factor that is not recurrent. In this case, the absence of a thrombophilia reduces the risk of occurrence/recurrence of VTE during pregnancy to a very low level, while the presence of a thrombophilia would mandate antepartum anticoagulation. In any case, she should receive postpartum prophylaxis since 75%-80% of fatal pulmonary emboli in pregnancy occur after cesarean delivery.

For instance, screening is valuable in a woman who had a VTE earlier in her life when she was on oral contraception and was put in a cast after a skiing accident. If she does not have a documented thrombophilia, you will not need to give her anticoagulation during the pregnancy—only post partum.

The Work-Up

When screening for inherited thrombophilia is warranted, I recommend limiting it to FVL, PGM, protein C deficiency, antithrombin deficiency, and protein S deficiency. (See table, p. 20.)

Screening for FVL, even during pregnancy, can be done with a second-generation screening test for active protein C resistance, or by polymerase chain reaction (PCR).

Screening for the PGM should be done by PCR, and I recommend getting an antithrombin activity level and a protein C activity level to screen for antithrombin deficiency and protein C deficiency, respectively.

Screening for protein S deficiency is trickier, since circulating protein S activity levels can vary dramatically in pregnancy (i.e., various conditions from infections to surgery to hormonal status can affect activity levels of protein S).

I recommend first assessing the protein S free antigen level. In nonpregnant patients, a free antigen level less than 55% indicates risk for deficiency. Free antigen levels drop significantly in pregnancy, however, making a level at or below 29% in the first and second trimesters, and a level at or below 24% in the third trimester, indicative of risk. Such levels can be accepted as indicating protein S deficiency, or deficiency can be confirmed by then measuring the protein S activity level.

I do not recommend screening for MTHFR mutations or hyperhomocysteinemia. There does not appear to be any association between MTHFR mutations and adverse pregnancy outcomes, and the probable association between hyperhomocysteinemia and maternal venous thrombotic events that exists in general is of far less concern in the United States since grains are fortified with folate. If there is any concern, extra folic acid supplementation should be protective.

Source Elsevier Global Medical News

Key Points

▸ Most positive associations between inherited thrombophilia and adverse pregnancy outcomes were derived from small case-control studies. Many studies are contradictory.

▸ Large prospective cohort studies have failed to demonstrate any consistent association between inherited thrombophilia and adverse pregnancy outcomes.

▸ There appears to be a modest association between thrombophilia and fetal loss after 10 weeks in retrospective, but not most prospective, studies.

▸ There is no current support for screening for inherited thrombophilia in women experiencing recurrent unexplained fetal loss or other adverse pregnancy outcomes. Diagnosis and treatment regimens should occur only in the context of an institutional review board–approved research protocol.

▸ Patients with known inherited thrombophilia and a personal or family history of prior VTE should receive antepartum thromboprophylaxis followed by postpartum anticoagulation.

▸ Unless they have additional, significant risk factors, women with lower-risk thrombophilias (i.e., heterozygotes for FVL, PGM, protein C deficiency, or protein S deficiency) and no history of prior VTE or an affected first-degree relative do not require antepartum thromboprophylaxis.

▸ Women who have a personal history of VTE associated with a nonrecurrent risk factor should be screened.

Source: Dr. Lockwood

Thrombophilia and Adverse Outcomes

Adverse pregnancy outcomes are among the most perplexing pregnancy-related problems because we still have little precise knowledge about the etiology—and often, the mechanisms—associated with them. Over time, a number of causes have been identified and suggested, and some potential therapeutic agents have been proposed.

The relationship between thrombophilia and adverse outcomes has been a long-term association. A number of experiential reports and uncontrolled trials have endorsed this relationship. In fact, experimental therapeutic trials with heparin and other agents have attempted to improve outcomes and have reported incremental benefits when these agents have been used.

This has further galvanized the belief that thrombophilia may in fact be strongly etiologic in the pathophysiology of some adverse pregnancy outcomes. Thus, interventions based on a presumed mechanistic basis have been supported. However, newer data have seemed not to bear out this long-held association between thrombophilia and adverse outcomes, and the implied treatment.

It is in light of this controversy and the conflicting positions that we have decided to do a Master Class to thoroughly review the subject, to look at what data exist that can help unravel this relationship, and to examine whether screening patients for thrombophilia and treating it as a basis for improving pregnancy outcomes is warranted.

We have invited Dr. Charles J. Lockwood to address the topic. Dr. Lockwood is the Anita O'Keeffe Young Professor of Women's Health and chair of the department of obstetrics, gynecology, and reproductive sciences at Yale University, New Haven, Conn., and chief of obstetrics and gynecology at Yale–New Haven Hospital.

Dr. Lockwood has studied and thought a great deal about the association between inherited thrombophilia and adverse pregnancy outcomes, as well as the association between thrombophilia and maternal thrombosis. He urges us to step back and, in light of a “new landscape of research findings,” take a more careful approach to assessment and screening.

Inherited thrombophilia and its association with both maternal thrombosis and adverse pregnancy outcomes is an issue that has come to the forefront over the past few years.

The association between inherited thrombophilia and maternal thrombosis appears to be fairly robust. Collectively, these thrombophilias account for 50%-70% of all maternal venous thrombotic events in pregnancy. Knowledge of the thrombophilic status of a patient can, therefore, have a significant impact on her clinical care. We understand better today, however, that personal and family history plays a critical role in assessing maternal thrombotic risk.

By contrast, the precise nature of the link between inherited thrombophilia and adverse pregnancy outcomes is still unclear. Over the past decade, the number of negative reports—those showing a lack of association—has increased significantly, and multiple prospective cohort studies have failed to consistently demonstrate the associations suggested by prior case-control studies that were smaller and mainly retrospective.

Collectively, this new landscape of research findings suggests that we should stop screening for inherited thrombophilia in patients with adverse pregnancy outcomes except in the setting of institutional review board–approved studies, and that we should better focus our approach to preventing maternal thrombosis through more careful, individualized risk assessment and through targeted use of antithrombotic therapy.

A New Evidence Base

Initial reports of associations between inherited thrombophilia and adverse pregnancy outcomes such as fetal loss, preeclampsia, fetal growth restriction, and abruptio placentae made some biological sense, but were based largely on small retrospective case-control studies with often inconsistent or contradictory findings.

In the case of fetal loss, numerous studies published in the 1990s and into the next decade showed a moderate association between inherited thrombophilia and stillbirth in particular.

A European retrospective cohort study published in 1996, for instance, found that the increased risk of loss among women with thrombophilia was greater after 28 weeks (odds ratio 3.6) than at or before 28 weeks (OR 1.4), and that the highest risk for stillbirth was associated with combined thrombophilic defects and antithrombin and protein C deficiencies (Lancet 1996;348:913-6).

This confusingly named study—the European Prospective Cohort on Thrombophilia (EPCOT)—involved 571 women with thrombophilia having 1,524 pregnancies, and 395 controls having 1,019 pregnancies.

In 2005, investigators of a larger case-control study nested within the 32,683-patient Nimes Obstetricians and Haematologist cohort reported an association between the factor V Leiden (FVL) mutation and pregnancy loss after 10 weeks (OR 3.5) but not between 3 and 9 weeks (J. Thromb. Haemost. 2005;3:2178-84)

A retrospective cohort study published in 2004 of 491 patients with a history of adverse pregnancy outcomes suggested, moreover, that one or more thrombophilia were actually protective of recurrent fetal losses at less than 10 weeks (Thromb. Haemost. 2004;91:290-5). However, the association of any one thrombophilia with later fetal losses was less significant in this study than in other studies (OR 1.76).

And an earlier meta-analysis of 31 studies looking at fetal loss and various thrombophilic disorders (most of them small case-control studies) concluded that FVL was associated with first-trimester pregnancy loss (OR 2.0) as well as later loss, although the association was much stronger (OR 3.3) with late, nonrecurrent fetal loss (Lancet 2003;361:901-8).

Although these and other studies suggested a link between FVL and stillbirth (and perhaps other thrombophilias and stillbirth), the absolute magnitude of the association (i.e., the absolute risk) was still very small. Moreover, over the past decade, the number of negative reports, especially amongst prospective studies, has increased—a temporal dichotomy that strongly suggests an initial bias toward positive studies and a growing comfort in reporting negative studies.

The larger prospective cohort studies reported over the last 5 years or so enerally have not found an association between inherited thrombophilia and stillbirth—or other adverse pregnancy outcomes for that matter.

For example, a 2005 study conducted by the Eunice Kennedy Shriver National Institute of Child Health and Human Development's Maternal-Fetal Medicine Units Network identified 134 FVL mutation carriers among nearly 4,900 gravidas in their first trimester of pregnancy and found no increase in fetal loss, preeclampsia, abruption, or intrauterine growth retardation (IUGR). A secondary analysis these data published earlier this year similarly found no association between the prothrombin gene G20210A mutation (PGM) and adverse pregnancy outcomes (Obstet. Gynecol. 2005;106:517-24 and Obstet. Gynecol. 2010;115:14-20).

Another prospective study of 4,250 unselected pregnancies also found no significant associations between FVL and preeclampsia, IUGR, and pregnancy loss (Br. J. Haematol. 2008;140:236-40).

Some of these findings are similar to previous reports from smaller prospective cohort studies. Investigators reported in 1999, for instance, no association between activated protein C resistance and fetal loss, preeclampsia, and IUGR. And in 2000, investigators had similarly reported a lack of association between FVL and methylenetetrahydrofolate reductase (MTHFR) polymorphism and preeclampsia or IUGR.

In general, the reported linkage between inherited thrombophilia and adverse outcomes other than stillbirth was always more tenuous. In the case of preeclampsia, however, meta-analyses of studies done before 2000 showed a fairly strong association between thrombophilia and preeclampsia, while studies published in and after 2001 found no such association.

There are a few exceptions to the lack of association found in larger prospective cohort studies. Most notably, an Australian study published this year of nulliparous women was suggestive of a weak association between the PGM and a composite index of adverse pregnancy outcomes (Obstet. Gynecol. 2010;115:5-13).

However, when investigators analyzed individual outcomes, they found that the only statistically significant associations were between the PGM and placental abruption, and between FVL and stillbirth. These associations, moreover, were based on very small sample size (nine and six patients, respectively). The investigators concluded that “the majority of asymptomatic women who carry an inherited thrombophilia polymorphism have a successful pregnancy outcome.”

There also is, at best, conflicting evidence in the literature of any benefit to heparin therapy for recurrent fetal loss.

A New Outlook on Screening

Given the evolving body of literature, it now seems wholly unjustified to screen low-risk populations. Knowing whether or not the patient has inherited thrombophilia, particularly in the nulliparous state, does not appear to be important for predicting outcomes.

There are questions that remain, however—most notably the question of whether women who have repetitive fetal losses or repetitive preeclampsia or abruptions should be screened and treated for inherited thrombophilia. Certainly, the failure of large prospective cohort studies to demonstrate any consistent association dampens our enthusiasm for the idea that inherited thrombophilia are to blame.

My opinion on this topic has evolved considerably over the last 10 years. I now believe that while screening for antiphospholipid syndrome is still warranted, screening for inherited thrombophilia in women having recurrent adverse pregnancy outcomes should occur only in the setting of an institutional review board–approved study in which ascertainment is done before a subsequent pregnancy and the patient's thrombophilia status is correlated with subsequent outcome (i.e., live birth, miscarriage, stillbirth, fetal growth restriction, preeclampsia, or abruption).

Furthermore, until we have established a definitive link between inherited thrombophilia and adverse pregnancy outcomes, we shouldn't even begin to think about clinical trials of thromboprophylaxis for affected women.

A particularly thorny question that has been raised concerns the issue of early fetal loss. Some have argued that the latest prospective cohort studies involved blood collection at or after 10 weeks' gestation and, therefore, are not relevant to conclusions drawn about the association (or lack thereof) between inherited thrombophilia and embryonic fetal loss.

However, I believe there are several reasons why we can conclude that thrombophilia and embryonic fetal losses are not linked. For one, there are enough data available from negative retrospective studies in which blood was obtained right after the pregnancy was completed. Secondly, there is no correlation between inherited thrombophilia and subsequent in vitro fertilization (IVF) failures in almost a dozen published studies. In fact, there is actually some evidence that FVL is associated with IVF success.

Lastly, we now know there is very little blood flow to the placenta before 10 weeks' gestation. There is some evidence, in fact, that hypoxia is the normal state of the embryo and may even be the preferred condition for culturing embryos in IVF.

Again, this issue requires prospective studies amongst patients with recurrent loss in which ascertainment occurs before the pregnancy commences.

Maternal Thrombotic Risk

While it's fair to say that, in general, inherited thrombophilia modestly increases the risk of maternal venous thrombotic events (VTE), it is critical to appreciate the role that a personal or strong family history of thrombosis (i.e., an affected first-degree relative) plays in determining a mother's risk.

Most women (greater than 93%) without a personal or strong family history of VTE will have uneventful pregnancies even when highly thrombogenic mutations are present. Once a personal or family history is factored in, however, the risk of VTE increases dramatically.

In the absence of a personal history of VTE or such an episode in a first-degree relative, heterozygosity for FVL or PGM is associated with a risk of thrombosis in pregnancy of well under 1% (0.2% and 0.5%, respectively). Similarly, protein C and protein S deficiencies are associated with a VTE risk under 1% in the absence of a personal or close family history.

In contrast, with a positive personal or family history, the risk of VTE in pregnancy increases to 10% in women who have heterozygosity for FVL, greater than 10% for women who have heterozygosity for PGM, 4%-17% in cases of protein C deficiency, and potentially up to 22% in cases of protein S deficiency.

Without a personal or family history, therefore, women with these lower-risk thrombophilia do not require anticoagulation during pregnancy unless they have other risk factors for thrombosis, such as significant obesity or orders for bed rest.

Patients with known inherited thrombophilia and a positive history, on the other hand, should receive antepartum thromboprophylaxis followed by postpartum anticoagulation. (Women who have a cesarean delivery should receive postpartum anticoagulation whether they have a personal or family history or not.)

Anticoagulation during pregnancy is also warranted—regardless of personal or family history—in the rare cases in which a patient is known to have homozygosity for FVL or homozygosity for PGM, or if a patient is known to have “double heterozygosity” for both FVL and PGM. Antithrombin deficiency, the most thrombogenic of all the inherited thrombophilias, also warrants antepartum anticoagulation as well as antithrombin infusions during labor and delivery.

To date, two studies have attempted to determine the value of screening for inherited thrombophilia based on a family history of prior VTE, and neither has shown that widespread screening would be particularly useful or cost effective in this setting. One certainly can argue, on the other hand, in favor of screening for thrombophilia in women who have a strong family history of VTE coupled with other risk factors.

Individualized risk assessment is always valuable. A woman with multiple risk factors—one who is obese, smokes, and is being put on bed rest, for instance—is a candidate for low-molecular-weight heparin (LMWH) therapy, for instance, even without a history of thrombosis and regardless of her thrombophilia status. If such a patient also has hypertension or preeclampsia, however, I'd be reluctant to give her either heparin or LMWH, for fear of abruption or even intracranial hemorrhage.

In what other circumstances is screening for thrombophilia warranted?

It can be justified when there is a personal history of VTE associated with a risk factor that is not recurrent. In this case, the absence of a thrombophilia reduces the risk of occurrence/recurrence of VTE during pregnancy to a very low level, while the presence of a thrombophilia would mandate antepartum anticoagulation. In any case, she should receive postpartum prophylaxis since 75%-80% of fatal pulmonary emboli in pregnancy occur after cesarean delivery.

For instance, screening is valuable in a woman who had a VTE earlier in her life when she was on oral contraception and was put in a cast after a skiing accident. If she does not have a documented thrombophilia, you will not need to give her anticoagulation during the pregnancy—only post partum.

The Work-Up

When screening for inherited thrombophilia is warranted, I recommend limiting it to FVL, PGM, protein C deficiency, antithrombin deficiency, and protein S deficiency. (See table, p. 20.)

Screening for FVL, even during pregnancy, can be done with a second-generation screening test for active protein C resistance, or by polymerase chain reaction (PCR).

Screening for the PGM should be done by PCR, and I recommend getting an antithrombin activity level and a protein C activity level to screen for antithrombin deficiency and protein C deficiency, respectively.

Screening for protein S deficiency is trickier, since circulating protein S activity levels can vary dramatically in pregnancy (i.e., various conditions from infections to surgery to hormonal status can affect activity levels of protein S).

I recommend first assessing the protein S free antigen level. In nonpregnant patients, a free antigen level less than 55% indicates risk for deficiency. Free antigen levels drop significantly in pregnancy, however, making a level at or below 29% in the first and second trimesters, and a level at or below 24% in the third trimester, indicative of risk. Such levels can be accepted as indicating protein S deficiency, or deficiency can be confirmed by then measuring the protein S activity level.

I do not recommend screening for MTHFR mutations or hyperhomocysteinemia. There does not appear to be any association between MTHFR mutations and adverse pregnancy outcomes, and the probable association between hyperhomocysteinemia and maternal venous thrombotic events that exists in general is of far less concern in the United States since grains are fortified with folate. If there is any concern, extra folic acid supplementation should be protective.

Source Elsevier Global Medical News

Key Points

▸ Most positive associations between inherited thrombophilia and adverse pregnancy outcomes were derived from small case-control studies. Many studies are contradictory.

▸ Large prospective cohort studies have failed to demonstrate any consistent association between inherited thrombophilia and adverse pregnancy outcomes.

▸ There appears to be a modest association between thrombophilia and fetal loss after 10 weeks in retrospective, but not most prospective, studies.

▸ There is no current support for screening for inherited thrombophilia in women experiencing recurrent unexplained fetal loss or other adverse pregnancy outcomes. Diagnosis and treatment regimens should occur only in the context of an institutional review board–approved research protocol.

▸ Patients with known inherited thrombophilia and a personal or family history of prior VTE should receive antepartum thromboprophylaxis followed by postpartum anticoagulation.

▸ Unless they have additional, significant risk factors, women with lower-risk thrombophilias (i.e., heterozygotes for FVL, PGM, protein C deficiency, or protein S deficiency) and no history of prior VTE or an affected first-degree relative do not require antepartum thromboprophylaxis.

▸ Women who have a personal history of VTE associated with a nonrecurrent risk factor should be screened.

Source: Dr. Lockwood

Thrombophilia and Adverse Outcomes

Adverse pregnancy outcomes are among the most perplexing pregnancy-related problems because we still have little precise knowledge about the etiology—and often, the mechanisms—associated with them. Over time, a number of causes have been identified and suggested, and some potential therapeutic agents have been proposed.

The relationship between thrombophilia and adverse outcomes has been a long-term association. A number of experiential reports and uncontrolled trials have endorsed this relationship. In fact, experimental therapeutic trials with heparin and other agents have attempted to improve outcomes and have reported incremental benefits when these agents have been used.

This has further galvanized the belief that thrombophilia may in fact be strongly etiologic in the pathophysiology of some adverse pregnancy outcomes. Thus, interventions based on a presumed mechanistic basis have been supported. However, newer data have seemed not to bear out this long-held association between thrombophilia and adverse outcomes, and the implied treatment.

It is in light of this controversy and the conflicting positions that we have decided to do a Master Class to thoroughly review the subject, to look at what data exist that can help unravel this relationship, and to examine whether screening patients for thrombophilia and treating it as a basis for improving pregnancy outcomes is warranted.

We have invited Dr. Charles J. Lockwood to address the topic. Dr. Lockwood is the Anita O'Keeffe Young Professor of Women's Health and chair of the department of obstetrics, gynecology, and reproductive sciences at Yale University, New Haven, Conn., and chief of obstetrics and gynecology at Yale–New Haven Hospital.

Dr. Lockwood has studied and thought a great deal about the association between inherited thrombophilia and adverse pregnancy outcomes, as well as the association between thrombophilia and maternal thrombosis. He urges us to step back and, in light of a “new landscape of research findings,” take a more careful approach to assessment and screening.

Inherited thrombophilia and its association with both maternal thrombosis and adverse pregnancy outcomes is an issue that has come to the forefront over the past few years.

The association between inherited thrombophilia and maternal thrombosis appears to be fairly robust. Collectively, these thrombophilias account for 50%-70% of all maternal venous thrombotic events in pregnancy. Knowledge of the thrombophilic status of a patient can, therefore, have a significant impact on her clinical care. We understand better today, however, that personal and family history plays a critical role in assessing maternal thrombotic risk.

By contrast, the precise nature of the link between inherited thrombophilia and adverse pregnancy outcomes is still unclear. Over the past decade, the number of negative reports—those showing a lack of association—has increased significantly, and multiple prospective cohort studies have failed to consistently demonstrate the associations suggested by prior case-control studies that were smaller and mainly retrospective.

Collectively, this new landscape of research findings suggests that we should stop screening for inherited thrombophilia in patients with adverse pregnancy outcomes except in the setting of institutional review board–approved studies, and that we should better focus our approach to preventing maternal thrombosis through more careful, individualized risk assessment and through targeted use of antithrombotic therapy.

A New Evidence Base

Initial reports of associations between inherited thrombophilia and adverse pregnancy outcomes such as fetal loss, preeclampsia, fetal growth restriction, and abruptio placentae made some biological sense, but were based largely on small retrospective case-control studies with often inconsistent or contradictory findings.

In the case of fetal loss, numerous studies published in the 1990s and into the next decade showed a moderate association between inherited thrombophilia and stillbirth in particular.

A European retrospective cohort study published in 1996, for instance, found that the increased risk of loss among women with thrombophilia was greater after 28 weeks (odds ratio 3.6) than at or before 28 weeks (OR 1.4), and that the highest risk for stillbirth was associated with combined thrombophilic defects and antithrombin and protein C deficiencies (Lancet 1996;348:913-6).

This confusingly named study—the European Prospective Cohort on Thrombophilia (EPCOT)—involved 571 women with thrombophilia having 1,524 pregnancies, and 395 controls having 1,019 pregnancies.

In 2005, investigators of a larger case-control study nested within the 32,683-patient Nimes Obstetricians and Haematologist cohort reported an association between the factor V Leiden (FVL) mutation and pregnancy loss after 10 weeks (OR 3.5) but not between 3 and 9 weeks (J. Thromb. Haemost. 2005;3:2178-84)

A retrospective cohort study published in 2004 of 491 patients with a history of adverse pregnancy outcomes suggested, moreover, that one or more thrombophilia were actually protective of recurrent fetal losses at less than 10 weeks (Thromb. Haemost. 2004;91:290-5). However, the association of any one thrombophilia with later fetal losses was less significant in this study than in other studies (OR 1.76).

And an earlier meta-analysis of 31 studies looking at fetal loss and various thrombophilic disorders (most of them small case-control studies) concluded that FVL was associated with first-trimester pregnancy loss (OR 2.0) as well as later loss, although the association was much stronger (OR 3.3) with late, nonrecurrent fetal loss (Lancet 2003;361:901-8).

Although these and other studies suggested a link between FVL and stillbirth (and perhaps other thrombophilias and stillbirth), the absolute magnitude of the association (i.e., the absolute risk) was still very small. Moreover, over the past decade, the number of negative reports, especially amongst prospective studies, has increased—a temporal dichotomy that strongly suggests an initial bias toward positive studies and a growing comfort in reporting negative studies.

The larger prospective cohort studies reported over the last 5 years or so enerally have not found an association between inherited thrombophilia and stillbirth—or other adverse pregnancy outcomes for that matter.

For example, a 2005 study conducted by the Eunice Kennedy Shriver National Institute of Child Health and Human Development's Maternal-Fetal Medicine Units Network identified 134 FVL mutation carriers among nearly 4,900 gravidas in their first trimester of pregnancy and found no increase in fetal loss, preeclampsia, abruption, or intrauterine growth retardation (IUGR). A secondary analysis these data published earlier this year similarly found no association between the prothrombin gene G20210A mutation (PGM) and adverse pregnancy outcomes (Obstet. Gynecol. 2005;106:517-24 and Obstet. Gynecol. 2010;115:14-20).

Another prospective study of 4,250 unselected pregnancies also found no significant associations between FVL and preeclampsia, IUGR, and pregnancy loss (Br. J. Haematol. 2008;140:236-40).

Some of these findings are similar to previous reports from smaller prospective cohort studies. Investigators reported in 1999, for instance, no association between activated protein C resistance and fetal loss, preeclampsia, and IUGR. And in 2000, investigators had similarly reported a lack of association between FVL and methylenetetrahydrofolate reductase (MTHFR) polymorphism and preeclampsia or IUGR.

In general, the reported linkage between inherited thrombophilia and adverse outcomes other than stillbirth was always more tenuous. In the case of preeclampsia, however, meta-analyses of studies done before 2000 showed a fairly strong association between thrombophilia and preeclampsia, while studies published in and after 2001 found no such association.

There are a few exceptions to the lack of association found in larger prospective cohort studies. Most notably, an Australian study published this year of nulliparous women was suggestive of a weak association between the PGM and a composite index of adverse pregnancy outcomes (Obstet. Gynecol. 2010;115:5-13).

However, when investigators analyzed individual outcomes, they found that the only statistically significant associations were between the PGM and placental abruption, and between FVL and stillbirth. These associations, moreover, were based on very small sample size (nine and six patients, respectively). The investigators concluded that “the majority of asymptomatic women who carry an inherited thrombophilia polymorphism have a successful pregnancy outcome.”

There also is, at best, conflicting evidence in the literature of any benefit to heparin therapy for recurrent fetal loss.

A New Outlook on Screening

Given the evolving body of literature, it now seems wholly unjustified to screen low-risk populations. Knowing whether or not the patient has inherited thrombophilia, particularly in the nulliparous state, does not appear to be important for predicting outcomes.

There are questions that remain, however—most notably the question of whether women who have repetitive fetal losses or repetitive preeclampsia or abruptions should be screened and treated for inherited thrombophilia. Certainly, the failure of large prospective cohort studies to demonstrate any consistent association dampens our enthusiasm for the idea that inherited thrombophilia are to blame.

My opinion on this topic has evolved considerably over the last 10 years. I now believe that while screening for antiphospholipid syndrome is still warranted, screening for inherited thrombophilia in women having recurrent adverse pregnancy outcomes should occur only in the setting of an institutional review board–approved study in which ascertainment is done before a subsequent pregnancy and the patient's thrombophilia status is correlated with subsequent outcome (i.e., live birth, miscarriage, stillbirth, fetal growth restriction, preeclampsia, or abruption).

Furthermore, until we have established a definitive link between inherited thrombophilia and adverse pregnancy outcomes, we shouldn't even begin to think about clinical trials of thromboprophylaxis for affected women.

A particularly thorny question that has been raised concerns the issue of early fetal loss. Some have argued that the latest prospective cohort studies involved blood collection at or after 10 weeks' gestation and, therefore, are not relevant to conclusions drawn about the association (or lack thereof) between inherited thrombophilia and embryonic fetal loss.

However, I believe there are several reasons why we can conclude that thrombophilia and embryonic fetal losses are not linked. For one, there are enough data available from negative retrospective studies in which blood was obtained right after the pregnancy was completed. Secondly, there is no correlation between inherited thrombophilia and subsequent in vitro fertilization (IVF) failures in almost a dozen published studies. In fact, there is actually some evidence that FVL is associated with IVF success.

Lastly, we now know there is very little blood flow to the placenta before 10 weeks' gestation. There is some evidence, in fact, that hypoxia is the normal state of the embryo and may even be the preferred condition for culturing embryos in IVF.

Again, this issue requires prospective studies amongst patients with recurrent loss in which ascertainment occurs before the pregnancy commences.

Maternal Thrombotic Risk

While it's fair to say that, in general, inherited thrombophilia modestly increases the risk of maternal venous thrombotic events (VTE), it is critical to appreciate the role that a personal or strong family history of thrombosis (i.e., an affected first-degree relative) plays in determining a mother's risk.

Most women (greater than 93%) without a personal or strong family history of VTE will have uneventful pregnancies even when highly thrombogenic mutations are present. Once a personal or family history is factored in, however, the risk of VTE increases dramatically.

In the absence of a personal history of VTE or such an episode in a first-degree relative, heterozygosity for FVL or PGM is associated with a risk of thrombosis in pregnancy of well under 1% (0.2% and 0.5%, respectively). Similarly, protein C and protein S deficiencies are associated with a VTE risk under 1% in the absence of a personal or close family history.

In contrast, with a positive personal or family history, the risk of VTE in pregnancy increases to 10% in women who have heterozygosity for FVL, greater than 10% for women who have heterozygosity for PGM, 4%-17% in cases of protein C deficiency, and potentially up to 22% in cases of protein S deficiency.

Without a personal or family history, therefore, women with these lower-risk thrombophilia do not require anticoagulation during pregnancy unless they have other risk factors for thrombosis, such as significant obesity or orders for bed rest.

Patients with known inherited thrombophilia and a positive history, on the other hand, should receive antepartum thromboprophylaxis followed by postpartum anticoagulation. (Women who have a cesarean delivery should receive postpartum anticoagulation whether they have a personal or family history or not.)

Anticoagulation during pregnancy is also warranted—regardless of personal or family history—in the rare cases in which a patient is known to have homozygosity for FVL or homozygosity for PGM, or if a patient is known to have “double heterozygosity” for both FVL and PGM. Antithrombin deficiency, the most thrombogenic of all the inherited thrombophilias, also warrants antepartum anticoagulation as well as antithrombin infusions during labor and delivery.

To date, two studies have attempted to determine the value of screening for inherited thrombophilia based on a family history of prior VTE, and neither has shown that widespread screening would be particularly useful or cost effective in this setting. One certainly can argue, on the other hand, in favor of screening for thrombophilia in women who have a strong family history of VTE coupled with other risk factors.

Individualized risk assessment is always valuable. A woman with multiple risk factors—one who is obese, smokes, and is being put on bed rest, for instance—is a candidate for low-molecular-weight heparin (LMWH) therapy, for instance, even without a history of thrombosis and regardless of her thrombophilia status. If such a patient also has hypertension or preeclampsia, however, I'd be reluctant to give her either heparin or LMWH, for fear of abruption or even intracranial hemorrhage.

In what other circumstances is screening for thrombophilia warranted?

It can be justified when there is a personal history of VTE associated with a risk factor that is not recurrent. In this case, the absence of a thrombophilia reduces the risk of occurrence/recurrence of VTE during pregnancy to a very low level, while the presence of a thrombophilia would mandate antepartum anticoagulation. In any case, she should receive postpartum prophylaxis since 75%-80% of fatal pulmonary emboli in pregnancy occur after cesarean delivery.

For instance, screening is valuable in a woman who had a VTE earlier in her life when she was on oral contraception and was put in a cast after a skiing accident. If she does not have a documented thrombophilia, you will not need to give her anticoagulation during the pregnancy—only post partum.

The Work-Up

When screening for inherited thrombophilia is warranted, I recommend limiting it to FVL, PGM, protein C deficiency, antithrombin deficiency, and protein S deficiency. (See table, p. 20.)

Screening for FVL, even during pregnancy, can be done with a second-generation screening test for active protein C resistance, or by polymerase chain reaction (PCR).

Screening for the PGM should be done by PCR, and I recommend getting an antithrombin activity level and a protein C activity level to screen for antithrombin deficiency and protein C deficiency, respectively.

Screening for protein S deficiency is trickier, since circulating protein S activity levels can vary dramatically in pregnancy (i.e., various conditions from infections to surgery to hormonal status can affect activity levels of protein S).

I recommend first assessing the protein S free antigen level. In nonpregnant patients, a free antigen level less than 55% indicates risk for deficiency. Free antigen levels drop significantly in pregnancy, however, making a level at or below 29% in the first and second trimesters, and a level at or below 24% in the third trimester, indicative of risk. Such levels can be accepted as indicating protein S deficiency, or deficiency can be confirmed by then measuring the protein S activity level.

I do not recommend screening for MTHFR mutations or hyperhomocysteinemia. There does not appear to be any association between MTHFR mutations and adverse pregnancy outcomes, and the probable association between hyperhomocysteinemia and maternal venous thrombotic events that exists in general is of far less concern in the United States since grains are fortified with folate. If there is any concern, extra folic acid supplementation should be protective.

Source Elsevier Global Medical News

Key Points

▸ Most positive associations between inherited thrombophilia and adverse pregnancy outcomes were derived from small case-control studies. Many studies are contradictory.

▸ Large prospective cohort studies have failed to demonstrate any consistent association between inherited thrombophilia and adverse pregnancy outcomes.

▸ There appears to be a modest association between thrombophilia and fetal loss after 10 weeks in retrospective, but not most prospective, studies.

▸ There is no current support for screening for inherited thrombophilia in women experiencing recurrent unexplained fetal loss or other adverse pregnancy outcomes. Diagnosis and treatment regimens should occur only in the context of an institutional review board–approved research protocol.

▸ Patients with known inherited thrombophilia and a personal or family history of prior VTE should receive antepartum thromboprophylaxis followed by postpartum anticoagulation.

▸ Unless they have additional, significant risk factors, women with lower-risk thrombophilias (i.e., heterozygotes for FVL, PGM, protein C deficiency, or protein S deficiency) and no history of prior VTE or an affected first-degree relative do not require antepartum thromboprophylaxis.

▸ Women who have a personal history of VTE associated with a nonrecurrent risk factor should be screened.

Source: Dr. Lockwood

Thrombophilia and Adverse Outcomes

Adverse pregnancy outcomes are among the most perplexing pregnancy-related problems because we still have little precise knowledge about the etiology—and often, the mechanisms—associated with them. Over time, a number of causes have been identified and suggested, and some potential therapeutic agents have been proposed.

The relationship between thrombophilia and adverse outcomes has been a long-term association. A number of experiential reports and uncontrolled trials have endorsed this relationship. In fact, experimental therapeutic trials with heparin and other agents have attempted to improve outcomes and have reported incremental benefits when these agents have been used.

This has further galvanized the belief that thrombophilia may in fact be strongly etiologic in the pathophysiology of some adverse pregnancy outcomes. Thus, interventions based on a presumed mechanistic basis have been supported. However, newer data have seemed not to bear out this long-held association between thrombophilia and adverse outcomes, and the implied treatment.

It is in light of this controversy and the conflicting positions that we have decided to do a Master Class to thoroughly review the subject, to look at what data exist that can help unravel this relationship, and to examine whether screening patients for thrombophilia and treating it as a basis for improving pregnancy outcomes is warranted.

We have invited Dr. Charles J. Lockwood to address the topic. Dr. Lockwood is the Anita O'Keeffe Young Professor of Women's Health and chair of the department of obstetrics, gynecology, and reproductive sciences at Yale University, New Haven, Conn., and chief of obstetrics and gynecology at Yale–New Haven Hospital.

Dr. Lockwood has studied and thought a great deal about the association between inherited thrombophilia and adverse pregnancy outcomes, as well as the association between thrombophilia and maternal thrombosis. He urges us to step back and, in light of a “new landscape of research findings,” take a more careful approach to assessment and screening.

Endometriosis affects about 2.5%-3.3% of reproductive-aged women and is characterized by extrauterine growth of endometrial tissue consisting of endometrial glands and stroma. Its depth ranges from superficial to deep infiltration.

When described as deep infiltrating endometriosis (DIE), the disease involves lesions that are invasive and that extend into areas or organs that are in direct contact with already affected areas. Although the theory of migration of endometrial glands may help to explain endometriosis in general and superficial disease in particular, the prevailing opinion currently is that DIE may result from metaplasia of remnants of Müllerian tissue.

DIE lesions penetrate greater than 5 mm under the peritoneal surface and can cause severe pelvic pain. The lesions are mainly composed of smooth muscle with active glandular epithelium, which causes fibrosis and eventually their nodular characteristics.

In cases in which the DIE lesions affect areas such as the rectovaginal space (as well as the uterosacral ligaments and the rectocervical area), treatment is more successful with a surgical approach.

Although rectovaginal lesions are eventually diagnosed through surgical methods, clinical examinations and imaging should be performed to support the surgical approach.

On medical history, many of these patients present with pelvic pain, dysmenorrhea, dyspareunia, or infertility.

Clinical examination must be done to help appreciate the location of the nodular lesion. The size of the nodule may be palpated on bimanual exam and can involve uterosacral ligaments, the posterior vaginal wall, the anterior rectal wall, and the posterior fornix.

Preoperative imaging, especially transvaginal and transrectal ultrasound and MRI, will establish the distribution and depth of the deep lesions.

The presence of deeply infiltrative endometriosis and the resulting alteration of normal anatomical planes make this one of the most challenging surgical cases that gynecologists encounter. Laparoscopic treatment requires an intimate understanding and application of pelvic anatomy, and the surgical fundamentals of visualization and traction-counter-traction, as well as electrosurgery.

Pathophysiology

The presence of deep endometriotic lesions in the posterior cul-de-sac is, again, likely a consequence of metaplasia of Müllerian rests, and the nodules are composed of smooth muscle proliferation and fibrosis, which is a result of infiltration.

The endometriotic foci migrate to the rectovaginal area, where hyperplasia of smooth muscle incites an inflammatory response; this evolves into retraction, which then leads to pelvic fibrosis and a subsequent reduction in uterine mobility and distortion of the pelvic anatomy.

DIE lesions have been classified through studies by Dr. Philippe Koninckx and his colleagues into three types based on their depth of invasion and location.

Type I lesions are conically shaped rectovaginal septum nodules and are located between the posterior and anterior walls of the vaginal mucosa and rectal muscularis, respectively. Lesions categorized as type II are deeply located and form from the posterior fornix to the rectovaginal region. They are typically covered by extensive adhesions causing retraction. The most severe lesions—type III—are composed of spherical nodules. The largest dimension of these lesions is located under the peritoneal fold of the rectouterine pouch of Douglas. The cranial movement of these posterior fornix lesions eventually causes the nodules to join the anterior rectal wall and creates an “hourglass”-like appearance.

Anatomy

It is important to appreciate the relationships of the avascular spaces and their relevance to the dissection of the obliterated cul-de-sac and excision of rectovaginal endometriosis. Also remember that the ureter enters the pelvis over the bifurcation of the common iliac and medial to the infundibulopelvic ligament (see photo below).

The pararectal space is bordered laterally by the pelvic sidewall, anterolaterally by the cardinal ligament, and medially by the rectal pillars. The ureter courses beneath the peritoneum and through the rectal pillar.

The rectovaginal space is bordered laterally by the uterosacral ligaments, anteriorly by the vaginal fascia, and posteriorly by the rectal fascia.

The potential surgical space between the ureter and uterosacral ligament is utilized to transect the uterosacral ligament, which provides a means of access from the pararectal space into the rectovaginal space.

Instrumentation and Process

Surgical management of DIE is essential for restoring pelvic anatomy, relieving debilitating pelvic pain, and eliminating endometriotic nodular foci. Laparoscopic surgery provides magnified views of the posterior cul-de-sac and its pathology, and results in less postoperative pain and decreased recurrence of adhesions.

Instruments typically utilized during these procedures include monopolar scissors, bipolar coagulation, and the Harmonic scalpel.

Energy sources that provide the least amount of lateral spread are key in these procedures as the relationships of the pelvic organs, ureters, and rectum are exceptionally close.

The systematic approach of resecting these nodules entails restoring the normal anatomic relationships of the adnexa and sigmoid, then dissecting the obliterated cul-de-sac and performing a thorough excision of the rectovaginal endometriotic lesions.

We have divided the procedure into four parts: ureterolysis, dissection of the rectovaginal septum, excision of the rectovaginal nodule, and reconstruction.

When the ureters and bowel are involved with the disease process, the surgical approach should take into account the importance of restoring normal anatomy. The use of uterine manipulators with a colpotomy cup can help delineate the posterior vaginal fornix and the rectum. The introduction of rectal and vaginal probes during the surgery will improve the exposition and excision of the lesions (see figure above).

After pneumoperitoneum is established and maintained at 15 mm Hg, we have used a standard technique of placing a 10-mm trocar in the umbilicus for the laparoscope and three 5-mm ancillary trocars. One 5-mm trocar is placed to the right at 10 cm lateral to the umbilicus, and two 5-mm trocars are placed to the left (10 cm lateral to the umbilicus and in the left lower quadrant).

A thorough examination of the abdomen and pelvis should be performed to assess the disease and degree of dissection needed to successfully access the rectovaginal space. To excise the lesions, we have used the bipolar RoBi forceps (Karl Storz), monopolar scissors, and the Harmonic scalpel.

Ureterolysis

In all of our cases thus far, ureterolysis was performed before resection of any DIE nodules. The ureter crosses the pelvic brim close to the bifurcation of the common iliac artery, at which point it becomes the pelvic ureter. It continues on the pelvic sidewall medial to the infundibulopelvic ligament as it crosses the external iliac artery. Branches of the internal iliac artery supply the descending portion of the ureter and move along the course of the ureter from the lateral aspect.

The dissection of the ureter begins at the pelvic brim where the anatomy is normal. The peritoneum superior to the ureter is grasped and entered, and the incision is extended. Medial traction is placed on the inferior edge of the peritoneal incision, and dissection is continued in the fat/nonfat interface until the ureter is identified.

The ureter is surrounded by a layer of loose areolar tissue; this layer is entered by using a blunt dissector to dissect parallel to the ureter. Small vessels should be coagulated in the process to ensure visibility. The ureterolysis is directed toward the uterosacral ligaments and continued until the ureter enters the cardinal ligament (see photo above).

When complete obliteration of the cul- de-sac is present and the uterosacral ligaments are obscured bilaterally, ureterolysis is carried out on the opposite side to improve pelvic anatomic restoration. The dissection is sufficient when both ureters are mobilized completely and when each can be traced from the pelvic brim to its insertion into the bladder (see photo above).

Dissection of Rectovaginal Septum

The next step is to enter and dissect the rectovaginal septum. Prior to excision of the nodule, the pouch of Douglas is first accessed by freeing the area from any adhesions or ovarian endometriomas.

After the successful bilateral ureterolysis, the ureters can be identified and the posterior fornix can be delineated with the rectal probe and colpotomizer.

The posterior fornix is then pushed upward, and a transverse incision is made over the posterior cervix superior to the rectum—an area also known as the prerectal fascia (see photo above).

The rectovaginal septum extends laterally from the posterior cervix and uterosacral ligaments to the pelvic sidewall, where it then courses caudally to insert in the perineal body. This area is dissected inferiorly until the uterosacral ligaments (also known as the medial rectal pillars) are identified. At this stage, the ureter, the pararectal space, the lateral rectal pillar with associated nerves, and the medial rectal pillar are seen. The rectal probe will then help identify the rectum and the dissected uterosacral ligaments in their respective planes.

The potential space medial to the ureter is used to transect the uterosacral ligament and thus enter the rectovaginal space from the lateral aspect to medial. The uterosacral ligament is again transected, and the rectovaginal space entered (see photo above).

If the dissection is continued too superiorly, the prerectal fascia is transected and the vagina may be entered. If this occurs, then the fascia is retracted superiorly and the rectum is retracted inferiorly to help identify the correct dissection plane.

By understanding the anatomic principle that fat belongs to the rectum, one can identify the fat/nonfat interface; this facilitates dissection superior to this plane. It is also important to maintain the integrity of the vaginal and rectal spaces as much as possible to decrease the risk of bowel perforation. The dissection is continued until the space is fully developed.

Excision of the Nodule

When the uterosacral ligament is transected, this permits a closer inspection of the posterior vaginal wall and the ability to assess the need for further rectal dissection. The posterior fornix is then incised along the rectovaginal margin, allowing the space to be opened. The nodule is then excised by continuing along the original shape (triangular) of the pouch of Douglas (see photo at bottom of column).

One might find that the nodule extends into the pararectal fascia, the muscularis layer of the rectum, the posterior vagina, or the rectal wall. The rectal probe is used to help delineate the rectum from the remaining lesion. Such lesions can often be dissected with sharp scissors or may require excision with a Harmonic scalpel.

Reconstruction

The last step is the reconstruction of any structures that were compromised during the dissection. The patient is given IV indigo carmine to ensure the ureters are not compromised, and a cystoscopy is performed at the conclusion of the case to confirm function. The rectal wall integrity is confirmed with the injection of dilute indigo carmine through an 18 french foley catheter placed in the rectum or via an air leak test performed with the aide of a proctoscope.

Because the fibers of rectovaginal septum run vertically and blend with the muscular wall of the vagina, some deep-infiltrating lesions are part of the vaginal wall, and in these cases excision of the affected area of the vagina is necessary. Once these lesions are fully resected, the vagina is reattached to the cervix by means of an interrupted figure of eight suture, and the anterior rectal wall is also reinforced with sutures. The pneumoperitoneum can be maintained by using a blue suction bulb in the vagina.

Once the reconstruction is completed, the restoration of the pelvic anatomy should be apparent and additional attention should be paid to defects to ensure proper closure.

The surgical management of rectovaginal endometriosis nodules can be technically demanding as it can include the repair of the vagina, bowel, bladder, and ureters. A systematic approach and adequate endoscopic experience, however, can significantly decrease the risk of injury, Taking the time to perform the ureterolysis before the beginning of the case, moreover, is beneficial in providing landmarks and protecting the integrity of the ureters. Although long-term experience is forthcoming, the surgical intervention of DIE has proven to be beneficial in the short term by decreasing patients' pain and improving their quality of lifestyle.

The yellow-shaded region represents the ureter, the blue represents the infundibulopelvic ligament, and the red represents the bifurcation of the right common iliac artery.

The pararectal space is bordered by the pelvic sidewall, the cardinal ligament, and the rectal pillars.

The rectovaginal space is bordered by the uterosacral ligaments, the vaginal fascia, and the rectal fascia.

The space between the ureter and uterosacral ligament is utilized to transect the uterosacral ligament.

Source Images provided by Dr. Resad Pasic

Use of a rectal probe and vaginal tenaculum allows for proper plane appreciation when dissecting lesions in the rectovaginal space.

The right ureter is seen here on the medial aspect of the peritoneum. Ureterolysis can be done bluntly with the graspers and scissors, or with the Harmonic scalpel as shown here.

Bilateral ureterolysis is done before dissection of the rectovaginal septum.

To transect the rectovaginal septum, place the uterus on stretch in the anteverted position. The rectal probe is placed in the rectum and the uterosacral septum is identified and transected with the harmonic scalpel.

Transection of the uterosacral ligament is done after ureterolysis is completed. (Ureter is visible in the left upper portion; rectum is in lower right corner.)

The rectal nodule is grasped and placed on tension. The rectal probe is used to help delineate the rectal borders.

Source Images provided by Dr. Resad Pasic

Keys to the Obliterated Cul-de-Sac

Although the best approach to treatment of the obliterated cul-de-sac and excision of rectovaginal endometriosis is surgical, this laparoscopic procedure can be a daunting task for even the most experienced minimally invasive gynecologic surgeon. The potential risk to the rectum and ureter must be immediately recognized. It is for this reason that Dr. Harry Reich, one of the legendary pioneers in minimally invasive gynecologic surgery, stated over 20 years ago that dissection of the obliterated cul-de-sac and excision of deep rectovaginal endometriosis was the most difficult procedure in the gynecologist's armamentarium.

Obviously, the anatomy has remained unchanged, but safety has been enhanced through the creation of a strategic approach for dealing with this unique surgical dilemma. For this current issue of the Master Class in Gynecologic Surgery, I have called upon Dr. Resad Pasic, who was my vice president during my tenure as president of the AAGL and who is now the immediate past president of that organization. Dr. Pasic is professor of obstetrics and gynecology, the director of the section of operative gynecologic endoscopy, and a codirector of the fellowship in laparoscopy and minimally invasive surgery at the department of ob.gyn. and women's health at the University of Louisville (Ky.).

As the director of the postgraduate course on advanced laparoscopic anatomy, dissection, and reparative pelvic surgery on unembalmed female cadavers, Dr. Pasic's expertise in anatomy and minimally invasive gynecologic surgery makes him an excellent candidate to lead the discussion of the systematic approach to the obliterated cul-de-sac and excision of rectovaginal endometriosis.

Joining Dr. Pasic as coauthors for this Master Class are Dr. Jessica A. Shepherd and Dr. Joseph L. Hudgens, the current clinical fellows in the division of gynecologic endoscopy at the University of Louisville.

Dr. Shepherd obtained her medical degree from Ross University in Roseau, Dominica, in 2005 and then completed her internship and residency at Drexel University in Philadelphia. In addition to her current involvement as a fellow in gynecologic endoscopy, Dr. Shepherd is also completing her MBA at the University of Louisville. She is currently a member of the AAGL and serves on the ad hoc review committee for the Journal of Minimally Invasive Gynecology.

Dr. Hudgens obtained his medical degree from the University of Arkansas, Little Rock, in 2005. He completed his residency training in ob.gyn. at the University of Louisville in 2009. Dr. Hudgens recently completed a fellowship in minimally invasive gynecology in 2010 and has joined the Center for Women's Health at Owensboro Medical Health System in Owensboro, Ky. Dr. Hudgens also is currently an active member of the AAGL and serves as a member of the ad hoc review committee for the Journal of Minimally Invasive Gynecology.

Endometriosis affects about 2.5%-3.3% of reproductive-aged women and is characterized by extrauterine growth of endometrial tissue consisting of endometrial glands and stroma. Its depth ranges from superficial to deep infiltration.

When described as deep infiltrating endometriosis (DIE), the disease involves lesions that are invasive and that extend into areas or organs that are in direct contact with already affected areas. Although the theory of migration of endometrial glands may help to explain endometriosis in general and superficial disease in particular, the prevailing opinion currently is that DIE may result from metaplasia of remnants of Müllerian tissue.

DIE lesions penetrate greater than 5 mm under the peritoneal surface and can cause severe pelvic pain. The lesions are mainly composed of smooth muscle with active glandular epithelium, which causes fibrosis and eventually their nodular characteristics.

In cases in which the DIE lesions affect areas such as the rectovaginal space (as well as the uterosacral ligaments and the rectocervical area), treatment is more successful with a surgical approach.

Although rectovaginal lesions are eventually diagnosed through surgical methods, clinical examinations and imaging should be performed to support the surgical approach.

On medical history, many of these patients present with pelvic pain, dysmenorrhea, dyspareunia, or infertility.

Clinical examination must be done to help appreciate the location of the nodular lesion. The size of the nodule may be palpated on bimanual exam and can involve uterosacral ligaments, the posterior vaginal wall, the anterior rectal wall, and the posterior fornix.

Preoperative imaging, especially transvaginal and transrectal ultrasound and MRI, will establish the distribution and depth of the deep lesions.

The presence of deeply infiltrative endometriosis and the resulting alteration of normal anatomical planes make this one of the most challenging surgical cases that gynecologists encounter. Laparoscopic treatment requires an intimate understanding and application of pelvic anatomy, and the surgical fundamentals of visualization and traction-counter-traction, as well as electrosurgery.

Pathophysiology

The presence of deep endometriotic lesions in the posterior cul-de-sac is, again, likely a consequence of metaplasia of Müllerian rests, and the nodules are composed of smooth muscle proliferation and fibrosis, which is a result of infiltration.

The endometriotic foci migrate to the rectovaginal area, where hyperplasia of smooth muscle incites an inflammatory response; this evolves into retraction, which then leads to pelvic fibrosis and a subsequent reduction in uterine mobility and distortion of the pelvic anatomy.

DIE lesions have been classified through studies by Dr. Philippe Koninckx and his colleagues into three types based on their depth of invasion and location.

Type I lesions are conically shaped rectovaginal septum nodules and are located between the posterior and anterior walls of the vaginal mucosa and rectal muscularis, respectively. Lesions categorized as type II are deeply located and form from the posterior fornix to the rectovaginal region. They are typically covered by extensive adhesions causing retraction. The most severe lesions—type III—are composed of spherical nodules. The largest dimension of these lesions is located under the peritoneal fold of the rectouterine pouch of Douglas. The cranial movement of these posterior fornix lesions eventually causes the nodules to join the anterior rectal wall and creates an “hourglass”-like appearance.

Anatomy

It is important to appreciate the relationships of the avascular spaces and their relevance to the dissection of the obliterated cul-de-sac and excision of rectovaginal endometriosis. Also remember that the ureter enters the pelvis over the bifurcation of the common iliac and medial to the infundibulopelvic ligament (see photo below).

The pararectal space is bordered laterally by the pelvic sidewall, anterolaterally by the cardinal ligament, and medially by the rectal pillars. The ureter courses beneath the peritoneum and through the rectal pillar.

The rectovaginal space is bordered laterally by the uterosacral ligaments, anteriorly by the vaginal fascia, and posteriorly by the rectal fascia.

The potential surgical space between the ureter and uterosacral ligament is utilized to transect the uterosacral ligament, which provides a means of access from the pararectal space into the rectovaginal space.

Instrumentation and Process

Surgical management of DIE is essential for restoring pelvic anatomy, relieving debilitating pelvic pain, and eliminating endometriotic nodular foci. Laparoscopic surgery provides magnified views of the posterior cul-de-sac and its pathology, and results in less postoperative pain and decreased recurrence of adhesions.

Instruments typically utilized during these procedures include monopolar scissors, bipolar coagulation, and the Harmonic scalpel.

Energy sources that provide the least amount of lateral spread are key in these procedures as the relationships of the pelvic organs, ureters, and rectum are exceptionally close.

The systematic approach of resecting these nodules entails restoring the normal anatomic relationships of the adnexa and sigmoid, then dissecting the obliterated cul-de-sac and performing a thorough excision of the rectovaginal endometriotic lesions.

We have divided the procedure into four parts: ureterolysis, dissection of the rectovaginal septum, excision of the rectovaginal nodule, and reconstruction.

When the ureters and bowel are involved with the disease process, the surgical approach should take into account the importance of restoring normal anatomy. The use of uterine manipulators with a colpotomy cup can help delineate the posterior vaginal fornix and the rectum. The introduction of rectal and vaginal probes during the surgery will improve the exposition and excision of the lesions (see figure above).

After pneumoperitoneum is established and maintained at 15 mm Hg, we have used a standard technique of placing a 10-mm trocar in the umbilicus for the laparoscope and three 5-mm ancillary trocars. One 5-mm trocar is placed to the right at 10 cm lateral to the umbilicus, and two 5-mm trocars are placed to the left (10 cm lateral to the umbilicus and in the left lower quadrant).

A thorough examination of the abdomen and pelvis should be performed to assess the disease and degree of dissection needed to successfully access the rectovaginal space. To excise the lesions, we have used the bipolar RoBi forceps (Karl Storz), monopolar scissors, and the Harmonic scalpel.

Ureterolysis

In all of our cases thus far, ureterolysis was performed before resection of any DIE nodules. The ureter crosses the pelvic brim close to the bifurcation of the common iliac artery, at which point it becomes the pelvic ureter. It continues on the pelvic sidewall medial to the infundibulopelvic ligament as it crosses the external iliac artery. Branches of the internal iliac artery supply the descending portion of the ureter and move along the course of the ureter from the lateral aspect.

The dissection of the ureter begins at the pelvic brim where the anatomy is normal. The peritoneum superior to the ureter is grasped and entered, and the incision is extended. Medial traction is placed on the inferior edge of the peritoneal incision, and dissection is continued in the fat/nonfat interface until the ureter is identified.

The ureter is surrounded by a layer of loose areolar tissue; this layer is entered by using a blunt dissector to dissect parallel to the ureter. Small vessels should be coagulated in the process to ensure visibility. The ureterolysis is directed toward the uterosacral ligaments and continued until the ureter enters the cardinal ligament (see photo above).

When complete obliteration of the cul- de-sac is present and the uterosacral ligaments are obscured bilaterally, ureterolysis is carried out on the opposite side to improve pelvic anatomic restoration. The dissection is sufficient when both ureters are mobilized completely and when each can be traced from the pelvic brim to its insertion into the bladder (see photo above).

Dissection of Rectovaginal Septum

The next step is to enter and dissect the rectovaginal septum. Prior to excision of the nodule, the pouch of Douglas is first accessed by freeing the area from any adhesions or ovarian endometriomas.

After the successful bilateral ureterolysis, the ureters can be identified and the posterior fornix can be delineated with the rectal probe and colpotomizer.

The posterior fornix is then pushed upward, and a transverse incision is made over the posterior cervix superior to the rectum—an area also known as the prerectal fascia (see photo above).

The rectovaginal septum extends laterally from the posterior cervix and uterosacral ligaments to the pelvic sidewall, where it then courses caudally to insert in the perineal body. This area is dissected inferiorly until the uterosacral ligaments (also known as the medial rectal pillars) are identified. At this stage, the ureter, the pararectal space, the lateral rectal pillar with associated nerves, and the medial rectal pillar are seen. The rectal probe will then help identify the rectum and the dissected uterosacral ligaments in their respective planes.

The potential space medial to the ureter is used to transect the uterosacral ligament and thus enter the rectovaginal space from the lateral aspect to medial. The uterosacral ligament is again transected, and the rectovaginal space entered (see photo above).

If the dissection is continued too superiorly, the prerectal fascia is transected and the vagina may be entered. If this occurs, then the fascia is retracted superiorly and the rectum is retracted inferiorly to help identify the correct dissection plane.

By understanding the anatomic principle that fat belongs to the rectum, one can identify the fat/nonfat interface; this facilitates dissection superior to this plane. It is also important to maintain the integrity of the vaginal and rectal spaces as much as possible to decrease the risk of bowel perforation. The dissection is continued until the space is fully developed.

Excision of the Nodule

When the uterosacral ligament is transected, this permits a closer inspection of the posterior vaginal wall and the ability to assess the need for further rectal dissection. The posterior fornix is then incised along the rectovaginal margin, allowing the space to be opened. The nodule is then excised by continuing along the original shape (triangular) of the pouch of Douglas (see photo at bottom of column).

One might find that the nodule extends into the pararectal fascia, the muscularis layer of the rectum, the posterior vagina, or the rectal wall. The rectal probe is used to help delineate the rectum from the remaining lesion. Such lesions can often be dissected with sharp scissors or may require excision with a Harmonic scalpel.

Reconstruction

The last step is the reconstruction of any structures that were compromised during the dissection. The patient is given IV indigo carmine to ensure the ureters are not compromised, and a cystoscopy is performed at the conclusion of the case to confirm function. The rectal wall integrity is confirmed with the injection of dilute indigo carmine through an 18 french foley catheter placed in the rectum or via an air leak test performed with the aide of a proctoscope.

Because the fibers of rectovaginal septum run vertically and blend with the muscular wall of the vagina, some deep-infiltrating lesions are part of the vaginal wall, and in these cases excision of the affected area of the vagina is necessary. Once these lesions are fully resected, the vagina is reattached to the cervix by means of an interrupted figure of eight suture, and the anterior rectal wall is also reinforced with sutures. The pneumoperitoneum can be maintained by using a blue suction bulb in the vagina.

Once the reconstruction is completed, the restoration of the pelvic anatomy should be apparent and additional attention should be paid to defects to ensure proper closure.

The surgical management of rectovaginal endometriosis nodules can be technically demanding as it can include the repair of the vagina, bowel, bladder, and ureters. A systematic approach and adequate endoscopic experience, however, can significantly decrease the risk of injury, Taking the time to perform the ureterolysis before the beginning of the case, moreover, is beneficial in providing landmarks and protecting the integrity of the ureters. Although long-term experience is forthcoming, the surgical intervention of DIE has proven to be beneficial in the short term by decreasing patients' pain and improving their quality of lifestyle.

The yellow-shaded region represents the ureter, the blue represents the infundibulopelvic ligament, and the red represents the bifurcation of the right common iliac artery.

The pararectal space is bordered by the pelvic sidewall, the cardinal ligament, and the rectal pillars.

The rectovaginal space is bordered by the uterosacral ligaments, the vaginal fascia, and the rectal fascia.

The space between the ureter and uterosacral ligament is utilized to transect the uterosacral ligament.

Source Images provided by Dr. Resad Pasic

Use of a rectal probe and vaginal tenaculum allows for proper plane appreciation when dissecting lesions in the rectovaginal space.

The right ureter is seen here on the medial aspect of the peritoneum. Ureterolysis can be done bluntly with the graspers and scissors, or with the Harmonic scalpel as shown here.

Bilateral ureterolysis is done before dissection of the rectovaginal septum.

To transect the rectovaginal septum, place the uterus on stretch in the anteverted position. The rectal probe is placed in the rectum and the uterosacral septum is identified and transected with the harmonic scalpel.

Transection of the uterosacral ligament is done after ureterolysis is completed. (Ureter is visible in the left upper portion; rectum is in lower right corner.)

The rectal nodule is grasped and placed on tension. The rectal probe is used to help delineate the rectal borders.

Source Images provided by Dr. Resad Pasic

Keys to the Obliterated Cul-de-Sac

Although the best approach to treatment of the obliterated cul-de-sac and excision of rectovaginal endometriosis is surgical, this laparoscopic procedure can be a daunting task for even the most experienced minimally invasive gynecologic surgeon. The potential risk to the rectum and ureter must be immediately recognized. It is for this reason that Dr. Harry Reich, one of the legendary pioneers in minimally invasive gynecologic surgery, stated over 20 years ago that dissection of the obliterated cul-de-sac and excision of deep rectovaginal endometriosis was the most difficult procedure in the gynecologist's armamentarium.

Obviously, the anatomy has remained unchanged, but safety has been enhanced through the creation of a strategic approach for dealing with this unique surgical dilemma. For this current issue of the Master Class in Gynecologic Surgery, I have called upon Dr. Resad Pasic, who was my vice president during my tenure as president of the AAGL and who is now the immediate past president of that organization. Dr. Pasic is professor of obstetrics and gynecology, the director of the section of operative gynecologic endoscopy, and a codirector of the fellowship in laparoscopy and minimally invasive surgery at the department of ob.gyn. and women's health at the University of Louisville (Ky.).

As the director of the postgraduate course on advanced laparoscopic anatomy, dissection, and reparative pelvic surgery on unembalmed female cadavers, Dr. Pasic's expertise in anatomy and minimally invasive gynecologic surgery makes him an excellent candidate to lead the discussion of the systematic approach to the obliterated cul-de-sac and excision of rectovaginal endometriosis.

Joining Dr. Pasic as coauthors for this Master Class are Dr. Jessica A. Shepherd and Dr. Joseph L. Hudgens, the current clinical fellows in the division of gynecologic endoscopy at the University of Louisville.

Dr. Shepherd obtained her medical degree from Ross University in Roseau, Dominica, in 2005 and then completed her internship and residency at Drexel University in Philadelphia. In addition to her current involvement as a fellow in gynecologic endoscopy, Dr. Shepherd is also completing her MBA at the University of Louisville. She is currently a member of the AAGL and serves on the ad hoc review committee for the Journal of Minimally Invasive Gynecology.

Dr. Hudgens obtained his medical degree from the University of Arkansas, Little Rock, in 2005. He completed his residency training in ob.gyn. at the University of Louisville in 2009. Dr. Hudgens recently completed a fellowship in minimally invasive gynecology in 2010 and has joined the Center for Women's Health at Owensboro Medical Health System in Owensboro, Ky. Dr. Hudgens also is currently an active member of the AAGL and serves as a member of the ad hoc review committee for the Journal of Minimally Invasive Gynecology.

Endometriosis affects about 2.5%-3.3% of reproductive-aged women and is characterized by extrauterine growth of endometrial tissue consisting of endometrial glands and stroma. Its depth ranges from superficial to deep infiltration.

When described as deep infiltrating endometriosis (DIE), the disease involves lesions that are invasive and that extend into areas or organs that are in direct contact with already affected areas. Although the theory of migration of endometrial glands may help to explain endometriosis in general and superficial disease in particular, the prevailing opinion currently is that DIE may result from metaplasia of remnants of Müllerian tissue.

DIE lesions penetrate greater than 5 mm under the peritoneal surface and can cause severe pelvic pain. The lesions are mainly composed of smooth muscle with active glandular epithelium, which causes fibrosis and eventually their nodular characteristics.

In cases in which the DIE lesions affect areas such as the rectovaginal space (as well as the uterosacral ligaments and the rectocervical area), treatment is more successful with a surgical approach.

Although rectovaginal lesions are eventually diagnosed through surgical methods, clinical examinations and imaging should be performed to support the surgical approach.

On medical history, many of these patients present with pelvic pain, dysmenorrhea, dyspareunia, or infertility.

Clinical examination must be done to help appreciate the location of the nodular lesion. The size of the nodule may be palpated on bimanual exam and can involve uterosacral ligaments, the posterior vaginal wall, the anterior rectal wall, and the posterior fornix.

Preoperative imaging, especially transvaginal and transrectal ultrasound and MRI, will establish the distribution and depth of the deep lesions.

The presence of deeply infiltrative endometriosis and the resulting alteration of normal anatomical planes make this one of the most challenging surgical cases that gynecologists encounter. Laparoscopic treatment requires an intimate understanding and application of pelvic anatomy, and the surgical fundamentals of visualization and traction-counter-traction, as well as electrosurgery.

Pathophysiology

The presence of deep endometriotic lesions in the posterior cul-de-sac is, again, likely a consequence of metaplasia of Müllerian rests, and the nodules are composed of smooth muscle proliferation and fibrosis, which is a result of infiltration.

The endometriotic foci migrate to the rectovaginal area, where hyperplasia of smooth muscle incites an inflammatory response; this evolves into retraction, which then leads to pelvic fibrosis and a subsequent reduction in uterine mobility and distortion of the pelvic anatomy.

DIE lesions have been classified through studies by Dr. Philippe Koninckx and his colleagues into three types based on their depth of invasion and location.

Type I lesions are conically shaped rectovaginal septum nodules and are located between the posterior and anterior walls of the vaginal mucosa and rectal muscularis, respectively. Lesions categorized as type II are deeply located and form from the posterior fornix to the rectovaginal region. They are typically covered by extensive adhesions causing retraction. The most severe lesions—type III—are composed of spherical nodules. The largest dimension of these lesions is located under the peritoneal fold of the rectouterine pouch of Douglas. The cranial movement of these posterior fornix lesions eventually causes the nodules to join the anterior rectal wall and creates an “hourglass”-like appearance.

Anatomy

It is important to appreciate the relationships of the avascular spaces and their relevance to the dissection of the obliterated cul-de-sac and excision of rectovaginal endometriosis. Also remember that the ureter enters the pelvis over the bifurcation of the common iliac and medial to the infundibulopelvic ligament (see photo below).

The pararectal space is bordered laterally by the pelvic sidewall, anterolaterally by the cardinal ligament, and medially by the rectal pillars. The ureter courses beneath the peritoneum and through the rectal pillar.

The rectovaginal space is bordered laterally by the uterosacral ligaments, anteriorly by the vaginal fascia, and posteriorly by the rectal fascia.

The potential surgical space between the ureter and uterosacral ligament is utilized to transect the uterosacral ligament, which provides a means of access from the pararectal space into the rectovaginal space.

Instrumentation and Process

Surgical management of DIE is essential for restoring pelvic anatomy, relieving debilitating pelvic pain, and eliminating endometriotic nodular foci. Laparoscopic surgery provides magnified views of the posterior cul-de-sac and its pathology, and results in less postoperative pain and decreased recurrence of adhesions.

Instruments typically utilized during these procedures include monopolar scissors, bipolar coagulation, and the Harmonic scalpel.

Energy sources that provide the least amount of lateral spread are key in these procedures as the relationships of the pelvic organs, ureters, and rectum are exceptionally close.

The systematic approach of resecting these nodules entails restoring the normal anatomic relationships of the adnexa and sigmoid, then dissecting the obliterated cul-de-sac and performing a thorough excision of the rectovaginal endometriotic lesions.

We have divided the procedure into four parts: ureterolysis, dissection of the rectovaginal septum, excision of the rectovaginal nodule, and reconstruction.

When the ureters and bowel are involved with the disease process, the surgical approach should take into account the importance of restoring normal anatomy. The use of uterine manipulators with a colpotomy cup can help delineate the posterior vaginal fornix and the rectum. The introduction of rectal and vaginal probes during the surgery will improve the exposition and excision of the lesions (see figure above).

After pneumoperitoneum is established and maintained at 15 mm Hg, we have used a standard technique of placing a 10-mm trocar in the umbilicus for the laparoscope and three 5-mm ancillary trocars. One 5-mm trocar is placed to the right at 10 cm lateral to the umbilicus, and two 5-mm trocars are placed to the left (10 cm lateral to the umbilicus and in the left lower quadrant).

A thorough examination of the abdomen and pelvis should be performed to assess the disease and degree of dissection needed to successfully access the rectovaginal space. To excise the lesions, we have used the bipolar RoBi forceps (Karl Storz), monopolar scissors, and the Harmonic scalpel.

Ureterolysis

In all of our cases thus far, ureterolysis was performed before resection of any DIE nodules. The ureter crosses the pelvic brim close to the bifurcation of the common iliac artery, at which point it becomes the pelvic ureter. It continues on the pelvic sidewall medial to the infundibulopelvic ligament as it crosses the external iliac artery. Branches of the internal iliac artery supply the descending portion of the ureter and move along the course of the ureter from the lateral aspect.

The dissection of the ureter begins at the pelvic brim where the anatomy is normal. The peritoneum superior to the ureter is grasped and entered, and the incision is extended. Medial traction is placed on the inferior edge of the peritoneal incision, and dissection is continued in the fat/nonfat interface until the ureter is identified.

The ureter is surrounded by a layer of loose areolar tissue; this layer is entered by using a blunt dissector to dissect parallel to the ureter. Small vessels should be coagulated in the process to ensure visibility. The ureterolysis is directed toward the uterosacral ligaments and continued until the ureter enters the cardinal ligament (see photo above).

When complete obliteration of the cul- de-sac is present and the uterosacral ligaments are obscured bilaterally, ureterolysis is carried out on the opposite side to improve pelvic anatomic restoration. The dissection is sufficient when both ureters are mobilized completely and when each can be traced from the pelvic brim to its insertion into the bladder (see photo above).

Dissection of Rectovaginal Septum

The next step is to enter and dissect the rectovaginal septum. Prior to excision of the nodule, the pouch of Douglas is first accessed by freeing the area from any adhesions or ovarian endometriomas.

After the successful bilateral ureterolysis, the ureters can be identified and the posterior fornix can be delineated with the rectal probe and colpotomizer.

The posterior fornix is then pushed upward, and a transverse incision is made over the posterior cervix superior to the rectum—an area also known as the prerectal fascia (see photo above).

The rectovaginal septum extends laterally from the posterior cervix and uterosacral ligaments to the pelvic sidewall, where it then courses caudally to insert in the perineal body. This area is dissected inferiorly until the uterosacral ligaments (also known as the medial rectal pillars) are identified. At this stage, the ureter, the pararectal space, the lateral rectal pillar with associated nerves, and the medial rectal pillar are seen. The rectal probe will then help identify the rectum and the dissected uterosacral ligaments in their respective planes.

The potential space medial to the ureter is used to transect the uterosacral ligament and thus enter the rectovaginal space from the lateral aspect to medial. The uterosacral ligament is again transected, and the rectovaginal space entered (see photo above).

If the dissection is continued too superiorly, the prerectal fascia is transected and the vagina may be entered. If this occurs, then the fascia is retracted superiorly and the rectum is retracted inferiorly to help identify the correct dissection plane.

By understanding the anatomic principle that fat belongs to the rectum, one can identify the fat/nonfat interface; this facilitates dissection superior to this plane. It is also important to maintain the integrity of the vaginal and rectal spaces as much as possible to decrease the risk of bowel perforation. The dissection is continued until the space is fully developed.

Excision of the Nodule

When the uterosacral ligament is transected, this permits a closer inspection of the posterior vaginal wall and the ability to assess the need for further rectal dissection. The posterior fornix is then incised along the rectovaginal margin, allowing the space to be opened. The nodule is then excised by continuing along the original shape (triangular) of the pouch of Douglas (see photo at bottom of column).

One might find that the nodule extends into the pararectal fascia, the muscularis layer of the rectum, the posterior vagina, or the rectal wall. The rectal probe is used to help delineate the rectum from the remaining lesion. Such lesions can often be dissected with sharp scissors or may require excision with a Harmonic scalpel.

Reconstruction

The last step is the reconstruction of any structures that were compromised during the dissection. The patient is given IV indigo carmine to ensure the ureters are not compromised, and a cystoscopy is performed at the conclusion of the case to confirm function. The rectal wall integrity is confirmed with the injection of dilute indigo carmine through an 18 french foley catheter placed in the rectum or via an air leak test performed with the aide of a proctoscope.

Because the fibers of rectovaginal septum run vertically and blend with the muscular wall of the vagina, some deep-infiltrating lesions are part of the vaginal wall, and in these cases excision of the affected area of the vagina is necessary. Once these lesions are fully resected, the vagina is reattached to the cervix by means of an interrupted figure of eight suture, and the anterior rectal wall is also reinforced with sutures. The pneumoperitoneum can be maintained by using a blue suction bulb in the vagina.

Once the reconstruction is completed, the restoration of the pelvic anatomy should be apparent and additional attention should be paid to defects to ensure proper closure.

The surgical management of rectovaginal endometriosis nodules can be technically demanding as it can include the repair of the vagina, bowel, bladder, and ureters. A systematic approach and adequate endoscopic experience, however, can significantly decrease the risk of injury, Taking the time to perform the ureterolysis before the beginning of the case, moreover, is beneficial in providing landmarks and protecting the integrity of the ureters. Although long-term experience is forthcoming, the surgical intervention of DIE has proven to be beneficial in the short term by decreasing patients' pain and improving their quality of lifestyle.

The yellow-shaded region represents the ureter, the blue represents the infundibulopelvic ligament, and the red represents the bifurcation of the right common iliac artery.

The pararectal space is bordered by the pelvic sidewall, the cardinal ligament, and the rectal pillars.

The rectovaginal space is bordered by the uterosacral ligaments, the vaginal fascia, and the rectal fascia.

The space between the ureter and uterosacral ligament is utilized to transect the uterosacral ligament.

Source Images provided by Dr. Resad Pasic

Use of a rectal probe and vaginal tenaculum allows for proper plane appreciation when dissecting lesions in the rectovaginal space.

The right ureter is seen here on the medial aspect of the peritoneum. Ureterolysis can be done bluntly with the graspers and scissors, or with the Harmonic scalpel as shown here.

Bilateral ureterolysis is done before dissection of the rectovaginal septum.

To transect the rectovaginal septum, place the uterus on stretch in the anteverted position. The rectal probe is placed in the rectum and the uterosacral septum is identified and transected with the harmonic scalpel.

Transection of the uterosacral ligament is done after ureterolysis is completed. (Ureter is visible in the left upper portion; rectum is in lower right corner.)

The rectal nodule is grasped and placed on tension. The rectal probe is used to help delineate the rectal borders.

Source Images provided by Dr. Resad Pasic

Keys to the Obliterated Cul-de-Sac

Although the best approach to treatment of the obliterated cul-de-sac and excision of rectovaginal endometriosis is surgical, this laparoscopic procedure can be a daunting task for even the most experienced minimally invasive gynecologic surgeon. The potential risk to the rectum and ureter must be immediately recognized. It is for this reason that Dr. Harry Reich, one of the legendary pioneers in minimally invasive gynecologic surgery, stated over 20 years ago that dissection of the obliterated cul-de-sac and excision of deep rectovaginal endometriosis was the most difficult procedure in the gynecologist's armamentarium.

Obviously, the anatomy has remained unchanged, but safety has been enhanced through the creation of a strategic approach for dealing with this unique surgical dilemma. For this current issue of the Master Class in Gynecologic Surgery, I have called upon Dr. Resad Pasic, who was my vice president during my tenure as president of the AAGL and who is now the immediate past president of that organization. Dr. Pasic is professor of obstetrics and gynecology, the director of the section of operative gynecologic endoscopy, and a codirector of the fellowship in laparoscopy and minimally invasive surgery at the department of ob.gyn. and women's health at the University of Louisville (Ky.).

As the director of the postgraduate course on advanced laparoscopic anatomy, dissection, and reparative pelvic surgery on unembalmed female cadavers, Dr. Pasic's expertise in anatomy and minimally invasive gynecologic surgery makes him an excellent candidate to lead the discussion of the systematic approach to the obliterated cul-de-sac and excision of rectovaginal endometriosis.

Joining Dr. Pasic as coauthors for this Master Class are Dr. Jessica A. Shepherd and Dr. Joseph L. Hudgens, the current clinical fellows in the division of gynecologic endoscopy at the University of Louisville.

Dr. Shepherd obtained her medical degree from Ross University in Roseau, Dominica, in 2005 and then completed her internship and residency at Drexel University in Philadelphia. In addition to her current involvement as a fellow in gynecologic endoscopy, Dr. Shepherd is also completing her MBA at the University of Louisville. She is currently a member of the AAGL and serves on the ad hoc review committee for the Journal of Minimally Invasive Gynecology.

Dr. Hudgens obtained his medical degree from the University of Arkansas, Little Rock, in 2005. He completed his residency training in ob.gyn. at the University of Louisville in 2009. Dr. Hudgens recently completed a fellowship in minimally invasive gynecology in 2010 and has joined the Center for Women's Health at Owensboro Medical Health System in Owensboro, Ky. Dr. Hudgens also is currently an active member of the AAGL and serves as a member of the ad hoc review committee for the Journal of Minimally Invasive Gynecology.