Surgical Techniques

There is a crisis of confidence in vaginal delivery. Women are aware of the potential for devastating consequences, and many ask for elective cesarean solely to avoid any possibility of incontinence or other problems linked to vaginal delivery.

Many obstetricians also have misgivings, though they are well aware that a cesarean is far more likely to cause maternal morbidity.¹ In a survey of female obstetricians, 31% chose elective cesarean as their preferred mode of delivery—80% of whom gave fear of perineal trauma as their reason.²

We cannot dispute the risks. The incidence of anal incontinence following recognized obstetric anal sphincter injury (OASI) is estimated at over 60%,³ and the true incidence may be much higher,⁴ particularly when injury goes unrecognized at the time of delivery.

OASI—any 3rd- or 4th-degree perineal tear—causes far more morbidity than episiotomy alone or 1st- or 2nd-degree tears ( FIGURE 1). It is the most common cause of postpartum anal incontinence. Anal incontinence is defined by the International Continence Society as involuntary loss of flatus or feces that becomes a social or hygienic problem.⁵ What’s more, incontinence due to OASI causes very high cumulative health service costs.¹³

Lack of uniform classification, insufficient training, and limited evidence from randomized controlled trials all contribute to the notoriously poor outcomes of obstetric anal sphincter injury.

To improve the outcome and reestablish confidence in vaginal delivery, more training is needed, as is more research directed toward identifying how to prevent, identify, and manage anal sphincter injury following vaginal delivery.

Taboos, embarrassment, and mistaken thinking

Even though anal incontinence may be both physically and psychologically devastating, many women do not seek medical attention due to embarrassment.^6-10 One study, for instance, found that only a third of women with fecal incontinence had ever discussed the problem with a physician.¹¹

Wood et al¹⁰ reported that most women with anal sphincter injury were either unaware that they had the injury, or felt they did not receive an adequate explanation of their injury.

Some women chose not to speak with their doctors because they believed that anal incontinence was a normal consequence of childbirth.^6,12

The scope of life-disrupting morbidities

Perineal pain and dyspareunia may persist for years

Perineal pain can be so distressing for the new mother that it may interfere with her ability to breast feed and cope with the daily tasks of motherhood.¹⁴ Short-term perineal pain is associated with reactionary edema, bruising, tight sutures, infection, and wound dehiscence. Persistent pain and discomfort from perineal trauma may also cause urinary retention and defecation problems.

Perineal pain and dyspareunia, which greatly impair sexual and social life, may last for many years after childbirth.^6,15-17 Wheeless,¹⁸ for instance, reported that some women refrained from sexual intercourse for up to 14 years because of dyspareunia following sphincter injury.

Abscess formation, wound breakdown, rectovaginal fistulae

Following primary repair of OASI, Venkatesh et al¹⁹ noted a 10% wound disruption rate.

Price of missed injury could be colostomy. Most rectovaginal fistulae occur when the physician fails to recognize the true extent of sphincter injury at the time of repair, resulting in inadequate sphincter reconstruction and wound breakdown.¹⁷ Once rectovaginal fistulae have occurred, treatment is difficult and may ultimately require permanent colostomy.^17,20

6 Risk factors for perineal trauma

1. Nulliparity

Because nulliparous women have a relatively inelastic perineum,²¹ time for perineal stretching during the second stage of labor is often inadequate, and perineal trauma is therefore more likely. Further, compared to the multipara, nulliparous women undergo more episiotomies to prevent perineal trauma, and are more likely to have instrumental delivery. This combination of factors increases their risk of OASI.

2. Macrosomia

Birth weight of more than 4 kg imposes risk of perineal injury, especially 3rd- and 4th-degree tears,^8,22,23 due to larger head circumference, prolonged labor, and difficult delivery, especially if instrumental delivery is used. Even after safe delivery of the head, shoulder dystocia—more common in macrosomic infants—may contribute to perineal and anal sphincter trauma. A large baby is also likely to disrupt the fascial supports of the pelvic floor and cause a stretch injury to the pelvic and pudendal nerves.

 

3. Malposition, malpresentation

Occipito-posterior position incurs increased incidence of sphincter injury, for these reasons:^8,22,24

• Incomplete flexion of fetal head increases the presenting diameter.
• Prolonged second stage of labor results in persistent pressure on the perineum, leading to edematous and friable tissues, which are more vulnerable to laceration, than during occipito-anterior labor.
• Instrumental delivery is more likely than with occipito-anterior position.

Malpresentations such as face and brow presentations are also reported as risk factors for anal sphincter injury.²²

Breech delivery does not appear to increase risk, but this may be due to stringent selection criteria and a low threshold for cesarean section during labor.

4. Precipitate labor

Cervical, perineal, labial, and urethral injury, all notable complications of precipitate labor, are largely due to inadequate time for maternal tissues to adjust to delivery forces. And delivery in unfavorable circumstances such as in transit to the hospital or in a standing position, without experienced assistance, allows no opportunity for management.

5. Prolonged second stage

Several studies have reported that a second stage of more than 60 minutes increases the incidence of anal sphincter injury.^22,25,26 Evidence suggests that a prolonged active second stage causes pudendal nerve damage; however, if damage occurs in the first stage, as one report indicates, then a cesarean performed after onset of labor during which the cervix dilates more than 8 cm would not avert pudendal nerve damage.²⁷

It has been suggested that a passive second stage, particularly with an epidural, should be accelerated with oxytocics, rather than resorting to instrumental delivery, which itself may cause trauma.

6. Operative delivery

Though operative delivery is integral to obstetrics and reduces the cesarean rate, maternal morbidity is more likely, compared to unassisted delivery. Injuries caused by instrumental delivery include cervical laceration, as well as anal sphincter injury.

Forceps delivery. The operator needs to be skilled in use of both forceps and vacuum extraction, since some circumstances preclude use of the vacuum extractor (prematurity, face presentation, potential fetal bleeding tendency, delivery of the aftercoming head at breech presentation, lift out at cesarean section, and equipment failure). However, it is well established that maternal injury is more likely with forceps than vacuum extraction. The reasons:

• The forceps occupy almost 10% more space in the pelvis.
• The shanks of the forceps stretch the perineum and can cause injury. The anal sphincter is particularly vulnerable when the physician pulls in the posterolateral direction to encourage flexion of the head.
• Unlike the vacuum extractor, which can detach, the forceps has no fail-safe mechanism, and therefore excessive force can be applied, particularly under epidural anaesthesia.
• Forceps delivery always requires an episiotomy, but it is not an absolute necessity with the vacuum extractor.

Vacuum delivery. A Cochrane review²⁸ of 10 trials concluded that vacuum-assisted vaginal delivery had significantly less maternal trauma (odds ratio [OR] 0.41; 95% confidence interval [CI], 0.33 to 0.50) and less general and regional anesthesia than forceps delivery.

A reduction in cephalhematoma and retinal hemorrhages with forceps might be considered a compensatory benefit; however, a 5-year follow-up of a randomized controlled trial comparing forceps with vacuum extraction found no significant differences in visual problems or child development.

 

Which cup for which position? Metal cups appear to be more suitable for occipitoposterior, transverse, and difficult occipitoanterior position deliveries.²⁸

Soft cups seem appropriate for straightforward deliveries, as they are significantly more likely to fail to achieve vaginal delivery (OR 1.65; 95% CI, 1.19 to 2.29). Though scalp injury was less likely with soft cups (OR 0.45; 95% CI, 0.15 to 0.60), the 2 groups did not differ in maternal injury.

Let mother choose position—it’s not critical

Women should be encouraged to deliver in whichever position is most comfortable. Though some evidence suggests that perineal injury is more likely with a standing position delivery, a Cochrane review found that, with the possible exception of increased blood loss, there were no deleterious effects to the mother or fetus.²⁹

The current evidence on various delivery positions is inconclusive.

Tactics for management of anal sphincter injury

Recognition and proper classification. Examination of perineal injury under adequate analgesia and light, and a combined vaginal and rectal examination are essential to assess the degree of anal sphincter injury.

If any doubt exists about the extent of the injury, a second opinion must be sought. It has been reported that the presence of an experienced person at the time of perineal assessment has increased the detection rate of anal sphincter injury.

Immediate repair of the perineal injury is advisable compared to delayed repair, as the immediate repair will reduce the bleeding and pain associated with the injury, which may in turn affect early breastfeeding and bonding. Immediate repair also prevents the development of edema (which may hinder subsequent recognition of structures involved) and reduces the possibility of infection.

Careful examination of the labia, clitoris, and urethra is essential to identify any injury. These structures need repair prior to the perineal repair.

Only a doctor experienced in anal sphincter repair or a trainee under supervision should perform a repair.

I prefer to repair the injury in the operating theater, where there is access to good lighting, appropriate equipment, and aseptic conditions.

General or regional (spinal, epidural, caudal) anesthesia is an important prerequisite—particularly for overlap repair, as the inherent tone in the sphincter muscle can cause the torn muscle ends to retract within the sheath. Muscle relaxation is necessary to retrieve the ends and overlap without tension.

The woman is placed in the lithotomy position and the full extent of the injury is evaluated by careful vaginal and rectal examination.

In the presence of a 4th-degree tear, the torn anal epithelium is repaired with interrupted 3/0 polyglactin (Vicryl, Ethicon, Somerville, NJ) sutures, with the knots tied in the anal lumen. Another option: A subcuticular repair of the anal epithelium using 3/0 polyglactin via the transvaginal approach has been used with equal success.

The sphincter muscles are repaired with 3/0 polydioxanone sulphate (PDS) clear sutures. Compared to a braided suture, these monofilamentous sutures are less likely to precipitate infection.

The internal anal sphincter should be identified and any tear repaired separately from the external sphincter, with interrupted 3/0 PDS. I advocate primary surgical repair of the internal sphincter, which has been shown to be beneficial in patients with established anal incontinence.

The external anal sphincter should be repaired with 3/0 PDS sutures, with either end-to-end or overlapping technique. No published randomized studies at present suggest that primary overlap technique is better than primary end-to-end technique. However the secondary overlapping techniques carried out by coloproctologists have shown better continence rates compared to secondary end-to-end technique.

Extra attention should be directed to reconstructing the perineal muscles, to provide support to the sphincter repair and maintain the vaginoanal distance. This may offer some protection in subsequent vaginal delivery and may prevent suture migration.

A vaginal and rectal examination must be performed and swabs and needles should be checked.

Intravenous antibiotics should be commenced intraoperatively and continued orally for 1 week.

A stool softener (lactulose 10 mL, 3 times daily) and a bulking agent should be prescribed for at least 2 weeks post-operatively, as passage of a large bolus of hard stool may disrupt the repair.

A comprehensive record should be documented, together with a diagram to demonstrate the injury.

The woman should be informed of the injury and the possible sequelae.

It is usual to ensure that a bowel action has occurred prior to discharge.

A hospital follow-up by an experienced doctor is essential.

Obstetric anal sphincter injury by the numbers

0.5%–5%	Incidence in centers performing mediolateral episiotomy15,34
Up to 50%	Incidence for forceps delivery with midline episiotomy35
At least 1 in 20	Number of women with anal incontinence up to 1 year after childbirth36,37
Over 60%	Incidence of anal incontinence following recognized anal sphincter injury3
One third	Number of women with anal incontinence who have discussed the problem with a doctor11

 

Future pregnancies: Set course by symptoms

Consider subsequent vaginal delivery only under these circumstances (FIGURE 2):

• The woman is asymptomatic.
• She has no evidence of anal sphincter defects detected by endoanal scan or low pressures on manometry.
• Delivery will be carried out by an experienced midwife or doctor.

Since no evidence suggests that an elective prophylactic episiotomy will prevent another tear, perform episiotomy only if clinically indicated (ie, if the perineum is thick and inelastic, and an episiotomy will prevent multiple radial tears).

Asymptomatic women with low squeeze pressures and a defect greater than 1 quadrant are at increased risk of developing anal incontinence following another vaginal delivery; therefore, counseling should include the option of cesarean section.

Symptomatic women with severe injuries. Offer a secondary sphincter repair, and deliver future pregnancies by cesarean.

Women with mild symptoms can be managed conservatively with:

• dietary advice to avoid gas-producing foods,
• regulation of bowel action,
• bulking agents,
• constipating agents such as loperamide and codeine phosphate,
• pelvic floor exercises, and
• biofeedback.

This group of women is at risk of deterioration with a subsequent vaginal delivery, and should therefore be offered cesarean section. The risk of developing a repeat 3rd-degree tear is low, but no randomized studies have been performed to evaluate the benefit of routine cesarean section.

The author reports no financial relationships relevant to this article.

FIGURE 2 Pregnancy after sphincter injury: How to manage delivery³⁰

There is a crisis of confidence in vaginal delivery. Women are aware of the potential for devastating consequences, and many ask for elective cesarean solely to avoid any possibility of incontinence or other problems linked to vaginal delivery.

Many obstetricians also have misgivings, though they are well aware that a cesarean is far more likely to cause maternal morbidity.¹ In a survey of female obstetricians, 31% chose elective cesarean as their preferred mode of delivery—80% of whom gave fear of perineal trauma as their reason.²

We cannot dispute the risks. The incidence of anal incontinence following recognized obstetric anal sphincter injury (OASI) is estimated at over 60%,³ and the true incidence may be much higher,⁴ particularly when injury goes unrecognized at the time of delivery.

OASI—any 3rd- or 4th-degree perineal tear—causes far more morbidity than episiotomy alone or 1st- or 2nd-degree tears ( FIGURE 1). It is the most common cause of postpartum anal incontinence. Anal incontinence is defined by the International Continence Society as involuntary loss of flatus or feces that becomes a social or hygienic problem.⁵ What’s more, incontinence due to OASI causes very high cumulative health service costs.¹³

Lack of uniform classification, insufficient training, and limited evidence from randomized controlled trials all contribute to the notoriously poor outcomes of obstetric anal sphincter injury.

To improve the outcome and reestablish confidence in vaginal delivery, more training is needed, as is more research directed toward identifying how to prevent, identify, and manage anal sphincter injury following vaginal delivery.

Taboos, embarrassment, and mistaken thinking

Even though anal incontinence may be both physically and psychologically devastating, many women do not seek medical attention due to embarrassment.^6-10 One study, for instance, found that only a third of women with fecal incontinence had ever discussed the problem with a physician.¹¹

Wood et al¹⁰ reported that most women with anal sphincter injury were either unaware that they had the injury, or felt they did not receive an adequate explanation of their injury.

Some women chose not to speak with their doctors because they believed that anal incontinence was a normal consequence of childbirth.^6,12

The scope of life-disrupting morbidities

Perineal pain and dyspareunia may persist for years

Perineal pain can be so distressing for the new mother that it may interfere with her ability to breast feed and cope with the daily tasks of motherhood.¹⁴ Short-term perineal pain is associated with reactionary edema, bruising, tight sutures, infection, and wound dehiscence. Persistent pain and discomfort from perineal trauma may also cause urinary retention and defecation problems.

Perineal pain and dyspareunia, which greatly impair sexual and social life, may last for many years after childbirth.^6,15-17 Wheeless,¹⁸ for instance, reported that some women refrained from sexual intercourse for up to 14 years because of dyspareunia following sphincter injury.

Abscess formation, wound breakdown, rectovaginal fistulae

Following primary repair of OASI, Venkatesh et al¹⁹ noted a 10% wound disruption rate.

Price of missed injury could be colostomy. Most rectovaginal fistulae occur when the physician fails to recognize the true extent of sphincter injury at the time of repair, resulting in inadequate sphincter reconstruction and wound breakdown.¹⁷ Once rectovaginal fistulae have occurred, treatment is difficult and may ultimately require permanent colostomy.^17,20

6 Risk factors for perineal trauma

1. Nulliparity

Because nulliparous women have a relatively inelastic perineum,²¹ time for perineal stretching during the second stage of labor is often inadequate, and perineal trauma is therefore more likely. Further, compared to the multipara, nulliparous women undergo more episiotomies to prevent perineal trauma, and are more likely to have instrumental delivery. This combination of factors increases their risk of OASI.

2. Macrosomia

Birth weight of more than 4 kg imposes risk of perineal injury, especially 3rd- and 4th-degree tears,^8,22,23 due to larger head circumference, prolonged labor, and difficult delivery, especially if instrumental delivery is used. Even after safe delivery of the head, shoulder dystocia—more common in macrosomic infants—may contribute to perineal and anal sphincter trauma. A large baby is also likely to disrupt the fascial supports of the pelvic floor and cause a stretch injury to the pelvic and pudendal nerves.

 

3. Malposition, malpresentation

Occipito-posterior position incurs increased incidence of sphincter injury, for these reasons:^8,22,24

• Incomplete flexion of fetal head increases the presenting diameter.
• Prolonged second stage of labor results in persistent pressure on the perineum, leading to edematous and friable tissues, which are more vulnerable to laceration, than during occipito-anterior labor.
• Instrumental delivery is more likely than with occipito-anterior position.

Malpresentations such as face and brow presentations are also reported as risk factors for anal sphincter injury.²²

Breech delivery does not appear to increase risk, but this may be due to stringent selection criteria and a low threshold for cesarean section during labor.

4. Precipitate labor

Cervical, perineal, labial, and urethral injury, all notable complications of precipitate labor, are largely due to inadequate time for maternal tissues to adjust to delivery forces. And delivery in unfavorable circumstances such as in transit to the hospital or in a standing position, without experienced assistance, allows no opportunity for management.

5. Prolonged second stage

Several studies have reported that a second stage of more than 60 minutes increases the incidence of anal sphincter injury.^22,25,26 Evidence suggests that a prolonged active second stage causes pudendal nerve damage; however, if damage occurs in the first stage, as one report indicates, then a cesarean performed after onset of labor during which the cervix dilates more than 8 cm would not avert pudendal nerve damage.²⁷

It has been suggested that a passive second stage, particularly with an epidural, should be accelerated with oxytocics, rather than resorting to instrumental delivery, which itself may cause trauma.

6. Operative delivery

Though operative delivery is integral to obstetrics and reduces the cesarean rate, maternal morbidity is more likely, compared to unassisted delivery. Injuries caused by instrumental delivery include cervical laceration, as well as anal sphincter injury.

Forceps delivery. The operator needs to be skilled in use of both forceps and vacuum extraction, since some circumstances preclude use of the vacuum extractor (prematurity, face presentation, potential fetal bleeding tendency, delivery of the aftercoming head at breech presentation, lift out at cesarean section, and equipment failure). However, it is well established that maternal injury is more likely with forceps than vacuum extraction. The reasons:

• The forceps occupy almost 10% more space in the pelvis.
• The shanks of the forceps stretch the perineum and can cause injury. The anal sphincter is particularly vulnerable when the physician pulls in the posterolateral direction to encourage flexion of the head.
• Unlike the vacuum extractor, which can detach, the forceps has no fail-safe mechanism, and therefore excessive force can be applied, particularly under epidural anaesthesia.
• Forceps delivery always requires an episiotomy, but it is not an absolute necessity with the vacuum extractor.

Vacuum delivery. A Cochrane review²⁸ of 10 trials concluded that vacuum-assisted vaginal delivery had significantly less maternal trauma (odds ratio [OR] 0.41; 95% confidence interval [CI], 0.33 to 0.50) and less general and regional anesthesia than forceps delivery.

A reduction in cephalhematoma and retinal hemorrhages with forceps might be considered a compensatory benefit; however, a 5-year follow-up of a randomized controlled trial comparing forceps with vacuum extraction found no significant differences in visual problems or child development.

 

Which cup for which position? Metal cups appear to be more suitable for occipitoposterior, transverse, and difficult occipitoanterior position deliveries.²⁸

Soft cups seem appropriate for straightforward deliveries, as they are significantly more likely to fail to achieve vaginal delivery (OR 1.65; 95% CI, 1.19 to 2.29). Though scalp injury was less likely with soft cups (OR 0.45; 95% CI, 0.15 to 0.60), the 2 groups did not differ in maternal injury.

Let mother choose position—it’s not critical

Women should be encouraged to deliver in whichever position is most comfortable. Though some evidence suggests that perineal injury is more likely with a standing position delivery, a Cochrane review found that, with the possible exception of increased blood loss, there were no deleterious effects to the mother or fetus.²⁹

The current evidence on various delivery positions is inconclusive.

Tactics for management of anal sphincter injury

Recognition and proper classification. Examination of perineal injury under adequate analgesia and light, and a combined vaginal and rectal examination are essential to assess the degree of anal sphincter injury.

If any doubt exists about the extent of the injury, a second opinion must be sought. It has been reported that the presence of an experienced person at the time of perineal assessment has increased the detection rate of anal sphincter injury.

Immediate repair of the perineal injury is advisable compared to delayed repair, as the immediate repair will reduce the bleeding and pain associated with the injury, which may in turn affect early breastfeeding and bonding. Immediate repair also prevents the development of edema (which may hinder subsequent recognition of structures involved) and reduces the possibility of infection.

Careful examination of the labia, clitoris, and urethra is essential to identify any injury. These structures need repair prior to the perineal repair.

Only a doctor experienced in anal sphincter repair or a trainee under supervision should perform a repair.

I prefer to repair the injury in the operating theater, where there is access to good lighting, appropriate equipment, and aseptic conditions.

General or regional (spinal, epidural, caudal) anesthesia is an important prerequisite—particularly for overlap repair, as the inherent tone in the sphincter muscle can cause the torn muscle ends to retract within the sheath. Muscle relaxation is necessary to retrieve the ends and overlap without tension.

The woman is placed in the lithotomy position and the full extent of the injury is evaluated by careful vaginal and rectal examination.

In the presence of a 4th-degree tear, the torn anal epithelium is repaired with interrupted 3/0 polyglactin (Vicryl, Ethicon, Somerville, NJ) sutures, with the knots tied in the anal lumen. Another option: A subcuticular repair of the anal epithelium using 3/0 polyglactin via the transvaginal approach has been used with equal success.

The sphincter muscles are repaired with 3/0 polydioxanone sulphate (PDS) clear sutures. Compared to a braided suture, these monofilamentous sutures are less likely to precipitate infection.

The internal anal sphincter should be identified and any tear repaired separately from the external sphincter, with interrupted 3/0 PDS. I advocate primary surgical repair of the internal sphincter, which has been shown to be beneficial in patients with established anal incontinence.

The external anal sphincter should be repaired with 3/0 PDS sutures, with either end-to-end or overlapping technique. No published randomized studies at present suggest that primary overlap technique is better than primary end-to-end technique. However the secondary overlapping techniques carried out by coloproctologists have shown better continence rates compared to secondary end-to-end technique.

Extra attention should be directed to reconstructing the perineal muscles, to provide support to the sphincter repair and maintain the vaginoanal distance. This may offer some protection in subsequent vaginal delivery and may prevent suture migration.

A vaginal and rectal examination must be performed and swabs and needles should be checked.

Intravenous antibiotics should be commenced intraoperatively and continued orally for 1 week.

A stool softener (lactulose 10 mL, 3 times daily) and a bulking agent should be prescribed for at least 2 weeks post-operatively, as passage of a large bolus of hard stool may disrupt the repair.

A comprehensive record should be documented, together with a diagram to demonstrate the injury.

The woman should be informed of the injury and the possible sequelae.

It is usual to ensure that a bowel action has occurred prior to discharge.

A hospital follow-up by an experienced doctor is essential.

Obstetric anal sphincter injury by the numbers

0.5%–5%	Incidence in centers performing mediolateral episiotomy15,34
Up to 50%	Incidence for forceps delivery with midline episiotomy35
At least 1 in 20	Number of women with anal incontinence up to 1 year after childbirth36,37
Over 60%	Incidence of anal incontinence following recognized anal sphincter injury3
One third	Number of women with anal incontinence who have discussed the problem with a doctor11

 

Future pregnancies: Set course by symptoms

Consider subsequent vaginal delivery only under these circumstances (FIGURE 2):

• The woman is asymptomatic.
• She has no evidence of anal sphincter defects detected by endoanal scan or low pressures on manometry.
• Delivery will be carried out by an experienced midwife or doctor.

Since no evidence suggests that an elective prophylactic episiotomy will prevent another tear, perform episiotomy only if clinically indicated (ie, if the perineum is thick and inelastic, and an episiotomy will prevent multiple radial tears).

Asymptomatic women with low squeeze pressures and a defect greater than 1 quadrant are at increased risk of developing anal incontinence following another vaginal delivery; therefore, counseling should include the option of cesarean section.

Symptomatic women with severe injuries. Offer a secondary sphincter repair, and deliver future pregnancies by cesarean.

Women with mild symptoms can be managed conservatively with:

• dietary advice to avoid gas-producing foods,
• regulation of bowel action,
• bulking agents,
• constipating agents such as loperamide and codeine phosphate,
• pelvic floor exercises, and
• biofeedback.

This group of women is at risk of deterioration with a subsequent vaginal delivery, and should therefore be offered cesarean section. The risk of developing a repeat 3rd-degree tear is low, but no randomized studies have been performed to evaluate the benefit of routine cesarean section.

The author reports no financial relationships relevant to this article.

FIGURE 2 Pregnancy after sphincter injury: How to manage delivery³⁰

There is a crisis of confidence in vaginal delivery. Women are aware of the potential for devastating consequences, and many ask for elective cesarean solely to avoid any possibility of incontinence or other problems linked to vaginal delivery.

Many obstetricians also have misgivings, though they are well aware that a cesarean is far more likely to cause maternal morbidity.¹ In a survey of female obstetricians, 31% chose elective cesarean as their preferred mode of delivery—80% of whom gave fear of perineal trauma as their reason.²

We cannot dispute the risks. The incidence of anal incontinence following recognized obstetric anal sphincter injury (OASI) is estimated at over 60%,³ and the true incidence may be much higher,⁴ particularly when injury goes unrecognized at the time of delivery.

OASI—any 3rd- or 4th-degree perineal tear—causes far more morbidity than episiotomy alone or 1st- or 2nd-degree tears ( FIGURE 1). It is the most common cause of postpartum anal incontinence. Anal incontinence is defined by the International Continence Society as involuntary loss of flatus or feces that becomes a social or hygienic problem.⁵ What’s more, incontinence due to OASI causes very high cumulative health service costs.¹³

Lack of uniform classification, insufficient training, and limited evidence from randomized controlled trials all contribute to the notoriously poor outcomes of obstetric anal sphincter injury.

To improve the outcome and reestablish confidence in vaginal delivery, more training is needed, as is more research directed toward identifying how to prevent, identify, and manage anal sphincter injury following vaginal delivery.

Taboos, embarrassment, and mistaken thinking

Even though anal incontinence may be both physically and psychologically devastating, many women do not seek medical attention due to embarrassment.^6-10 One study, for instance, found that only a third of women with fecal incontinence had ever discussed the problem with a physician.¹¹

Wood et al¹⁰ reported that most women with anal sphincter injury were either unaware that they had the injury, or felt they did not receive an adequate explanation of their injury.

Some women chose not to speak with their doctors because they believed that anal incontinence was a normal consequence of childbirth.^6,12

The scope of life-disrupting morbidities

Perineal pain and dyspareunia may persist for years

Perineal pain can be so distressing for the new mother that it may interfere with her ability to breast feed and cope with the daily tasks of motherhood.¹⁴ Short-term perineal pain is associated with reactionary edema, bruising, tight sutures, infection, and wound dehiscence. Persistent pain and discomfort from perineal trauma may also cause urinary retention and defecation problems.

Perineal pain and dyspareunia, which greatly impair sexual and social life, may last for many years after childbirth.^6,15-17 Wheeless,¹⁸ for instance, reported that some women refrained from sexual intercourse for up to 14 years because of dyspareunia following sphincter injury.

Abscess formation, wound breakdown, rectovaginal fistulae

Following primary repair of OASI, Venkatesh et al¹⁹ noted a 10% wound disruption rate.

Price of missed injury could be colostomy. Most rectovaginal fistulae occur when the physician fails to recognize the true extent of sphincter injury at the time of repair, resulting in inadequate sphincter reconstruction and wound breakdown.¹⁷ Once rectovaginal fistulae have occurred, treatment is difficult and may ultimately require permanent colostomy.^17,20

6 Risk factors for perineal trauma

1. Nulliparity

Because nulliparous women have a relatively inelastic perineum,²¹ time for perineal stretching during the second stage of labor is often inadequate, and perineal trauma is therefore more likely. Further, compared to the multipara, nulliparous women undergo more episiotomies to prevent perineal trauma, and are more likely to have instrumental delivery. This combination of factors increases their risk of OASI.

2. Macrosomia

Birth weight of more than 4 kg imposes risk of perineal injury, especially 3rd- and 4th-degree tears,^8,22,23 due to larger head circumference, prolonged labor, and difficult delivery, especially if instrumental delivery is used. Even after safe delivery of the head, shoulder dystocia—more common in macrosomic infants—may contribute to perineal and anal sphincter trauma. A large baby is also likely to disrupt the fascial supports of the pelvic floor and cause a stretch injury to the pelvic and pudendal nerves.

 

3. Malposition, malpresentation

Occipito-posterior position incurs increased incidence of sphincter injury, for these reasons:^8,22,24

• Incomplete flexion of fetal head increases the presenting diameter.
• Prolonged second stage of labor results in persistent pressure on the perineum, leading to edematous and friable tissues, which are more vulnerable to laceration, than during occipito-anterior labor.
• Instrumental delivery is more likely than with occipito-anterior position.

Malpresentations such as face and brow presentations are also reported as risk factors for anal sphincter injury.²²

Breech delivery does not appear to increase risk, but this may be due to stringent selection criteria and a low threshold for cesarean section during labor.

4. Precipitate labor

Cervical, perineal, labial, and urethral injury, all notable complications of precipitate labor, are largely due to inadequate time for maternal tissues to adjust to delivery forces. And delivery in unfavorable circumstances such as in transit to the hospital or in a standing position, without experienced assistance, allows no opportunity for management.

5. Prolonged second stage

Several studies have reported that a second stage of more than 60 minutes increases the incidence of anal sphincter injury.^22,25,26 Evidence suggests that a prolonged active second stage causes pudendal nerve damage; however, if damage occurs in the first stage, as one report indicates, then a cesarean performed after onset of labor during which the cervix dilates more than 8 cm would not avert pudendal nerve damage.²⁷

It has been suggested that a passive second stage, particularly with an epidural, should be accelerated with oxytocics, rather than resorting to instrumental delivery, which itself may cause trauma.

6. Operative delivery

Though operative delivery is integral to obstetrics and reduces the cesarean rate, maternal morbidity is more likely, compared to unassisted delivery. Injuries caused by instrumental delivery include cervical laceration, as well as anal sphincter injury.

Forceps delivery. The operator needs to be skilled in use of both forceps and vacuum extraction, since some circumstances preclude use of the vacuum extractor (prematurity, face presentation, potential fetal bleeding tendency, delivery of the aftercoming head at breech presentation, lift out at cesarean section, and equipment failure). However, it is well established that maternal injury is more likely with forceps than vacuum extraction. The reasons:

• The forceps occupy almost 10% more space in the pelvis.
• The shanks of the forceps stretch the perineum and can cause injury. The anal sphincter is particularly vulnerable when the physician pulls in the posterolateral direction to encourage flexion of the head.
• Unlike the vacuum extractor, which can detach, the forceps has no fail-safe mechanism, and therefore excessive force can be applied, particularly under epidural anaesthesia.
• Forceps delivery always requires an episiotomy, but it is not an absolute necessity with the vacuum extractor.

Vacuum delivery. A Cochrane review²⁸ of 10 trials concluded that vacuum-assisted vaginal delivery had significantly less maternal trauma (odds ratio [OR] 0.41; 95% confidence interval [CI], 0.33 to 0.50) and less general and regional anesthesia than forceps delivery.

A reduction in cephalhematoma and retinal hemorrhages with forceps might be considered a compensatory benefit; however, a 5-year follow-up of a randomized controlled trial comparing forceps with vacuum extraction found no significant differences in visual problems or child development.

 

Which cup for which position? Metal cups appear to be more suitable for occipitoposterior, transverse, and difficult occipitoanterior position deliveries.²⁸

Soft cups seem appropriate for straightforward deliveries, as they are significantly more likely to fail to achieve vaginal delivery (OR 1.65; 95% CI, 1.19 to 2.29). Though scalp injury was less likely with soft cups (OR 0.45; 95% CI, 0.15 to 0.60), the 2 groups did not differ in maternal injury.

Let mother choose position—it’s not critical

Women should be encouraged to deliver in whichever position is most comfortable. Though some evidence suggests that perineal injury is more likely with a standing position delivery, a Cochrane review found that, with the possible exception of increased blood loss, there were no deleterious effects to the mother or fetus.²⁹

The current evidence on various delivery positions is inconclusive.

Tactics for management of anal sphincter injury

Recognition and proper classification. Examination of perineal injury under adequate analgesia and light, and a combined vaginal and rectal examination are essential to assess the degree of anal sphincter injury.

If any doubt exists about the extent of the injury, a second opinion must be sought. It has been reported that the presence of an experienced person at the time of perineal assessment has increased the detection rate of anal sphincter injury.

Immediate repair of the perineal injury is advisable compared to delayed repair, as the immediate repair will reduce the bleeding and pain associated with the injury, which may in turn affect early breastfeeding and bonding. Immediate repair also prevents the development of edema (which may hinder subsequent recognition of structures involved) and reduces the possibility of infection.

Careful examination of the labia, clitoris, and urethra is essential to identify any injury. These structures need repair prior to the perineal repair.

Only a doctor experienced in anal sphincter repair or a trainee under supervision should perform a repair.

I prefer to repair the injury in the operating theater, where there is access to good lighting, appropriate equipment, and aseptic conditions.

General or regional (spinal, epidural, caudal) anesthesia is an important prerequisite—particularly for overlap repair, as the inherent tone in the sphincter muscle can cause the torn muscle ends to retract within the sheath. Muscle relaxation is necessary to retrieve the ends and overlap without tension.

The woman is placed in the lithotomy position and the full extent of the injury is evaluated by careful vaginal and rectal examination.

In the presence of a 4th-degree tear, the torn anal epithelium is repaired with interrupted 3/0 polyglactin (Vicryl, Ethicon, Somerville, NJ) sutures, with the knots tied in the anal lumen. Another option: A subcuticular repair of the anal epithelium using 3/0 polyglactin via the transvaginal approach has been used with equal success.

The sphincter muscles are repaired with 3/0 polydioxanone sulphate (PDS) clear sutures. Compared to a braided suture, these monofilamentous sutures are less likely to precipitate infection.

The internal anal sphincter should be identified and any tear repaired separately from the external sphincter, with interrupted 3/0 PDS. I advocate primary surgical repair of the internal sphincter, which has been shown to be beneficial in patients with established anal incontinence.

The external anal sphincter should be repaired with 3/0 PDS sutures, with either end-to-end or overlapping technique. No published randomized studies at present suggest that primary overlap technique is better than primary end-to-end technique. However the secondary overlapping techniques carried out by coloproctologists have shown better continence rates compared to secondary end-to-end technique.

Extra attention should be directed to reconstructing the perineal muscles, to provide support to the sphincter repair and maintain the vaginoanal distance. This may offer some protection in subsequent vaginal delivery and may prevent suture migration.

A vaginal and rectal examination must be performed and swabs and needles should be checked.

Intravenous antibiotics should be commenced intraoperatively and continued orally for 1 week.

A stool softener (lactulose 10 mL, 3 times daily) and a bulking agent should be prescribed for at least 2 weeks post-operatively, as passage of a large bolus of hard stool may disrupt the repair.

A comprehensive record should be documented, together with a diagram to demonstrate the injury.

The woman should be informed of the injury and the possible sequelae.

It is usual to ensure that a bowel action has occurred prior to discharge.

A hospital follow-up by an experienced doctor is essential.

Obstetric anal sphincter injury by the numbers

0.5%–5%	Incidence in centers performing mediolateral episiotomy15,34
Up to 50%	Incidence for forceps delivery with midline episiotomy35
At least 1 in 20	Number of women with anal incontinence up to 1 year after childbirth36,37
Over 60%	Incidence of anal incontinence following recognized anal sphincter injury3
One third	Number of women with anal incontinence who have discussed the problem with a doctor11

 

Future pregnancies: Set course by symptoms

Consider subsequent vaginal delivery only under these circumstances (FIGURE 2):

• The woman is asymptomatic.
• She has no evidence of anal sphincter defects detected by endoanal scan or low pressures on manometry.
• Delivery will be carried out by an experienced midwife or doctor.

Since no evidence suggests that an elective prophylactic episiotomy will prevent another tear, perform episiotomy only if clinically indicated (ie, if the perineum is thick and inelastic, and an episiotomy will prevent multiple radial tears).

Asymptomatic women with low squeeze pressures and a defect greater than 1 quadrant are at increased risk of developing anal incontinence following another vaginal delivery; therefore, counseling should include the option of cesarean section.

Symptomatic women with severe injuries. Offer a secondary sphincter repair, and deliver future pregnancies by cesarean.

Women with mild symptoms can be managed conservatively with:

• dietary advice to avoid gas-producing foods,
• regulation of bowel action,
• bulking agents,
• constipating agents such as loperamide and codeine phosphate,
• pelvic floor exercises, and
• biofeedback.

This group of women is at risk of deterioration with a subsequent vaginal delivery, and should therefore be offered cesarean section. The risk of developing a repeat 3rd-degree tear is low, but no randomized studies have been performed to evaluate the benefit of routine cesarean section.

The author reports no financial relationships relevant to this article.

FIGURE 2 Pregnancy after sphincter injury: How to manage delivery³⁰

Gynecologic laparoscopy in the obese? What was once the purview of the very talented or the foolhardy may now be the preferred surgical approach.

Obese women who undergo laparoscopy recover faster, with less pain, fewer wound infections, and shorter hospital stays than with laparotomy. Though it is true that obesity increases operative time and the risk for conversion to laparotomy, little evidence supports the theory that a body mass index (BMI) of 30 kg/m² or higher should exclude laparoscopy.

Preoperative management

Unique elements of the physical

It is important to identify central obesity, which is more difficult to accommodate than distribution around the hips. Unfortunately, the roughly 40 million obese Americans tend to have central fat distribution.^1,2

In central obesity, the subcutaneous tissue is thick, often requiring extra long ports to attain peritoneal access.

The relationship of the umbilicus to the underlying aortic bifurcation also shifts more caudally. This relationship should be noted and planned for before going to the operating room (FIGURE).

Abdominal obesity in particular confers additional risks during all types of surgery: higher rates of atelectasis, thromboembolism, cardiovascular dysfunction, and wound infection.

Closely inspect the skin and panniculus after a routine examination. Obesity predisposes patients to dark, moist, anoxic spaces beneath folds of skin that need to be identified and inspected for evidence of fungal or bacterial infection. To optimize postoperative wound healing, treat any preexisting infections before surgery.

Cigarette smoking further burdens pulmonary mechanics and oxygenation during surgery, so it is important to encourage smokers to kick the habit at least 8 weeks before elective surgery.³

In general, use the history and physical examination to focus on the recognized risk factors of obesity, with specific emphasis on hypertension, coronary artery disease, arrhythmia, pulmonary obstructive disease, peripheral vascular disease, diabetes, gastric reflux, and arthritis.⁴

Special tests and laboratory studies

EKG and chest x-ray. In morbidly obese patients (BMI >40), preoperative evaluation includes an electrocardiogram (EKG) and chest x-ray to identify any cardiomegaly, arrhythmias, and occult ischemia or conduction blockage.

Arterial blood gas sampling. Given the higher risk of postoperative thrombotic events in obese patients, it can be helpful to assess preoperative oxygenation and ventilation/perfusion status via arterial blood gas sampling. The obese may have elevated baseline Aa gradients, which, if not noted prior to surgery, can confuse later management of suspected pulmonary emboli.

During testing, assess venous access and counsel the patient if central venous line placement may be possible at surgery. Though central line placement is not routinely recommended, it may be warranted in patients with particularly difficult peripheral venous access.

Skip pulmonary function testing because the results rarely change surgical management. We consider its routine use to be wasteful.

Laboratory evaluation should include betahuman chorionic gonadotropin (in premenopausal patients), complete blood count, electrolytes, glucose, renal function, and type and screen.

Spell out risks at informed consent

The preoperative appointment is your chance to answer questions the patient may have and clearly delineate the risks and benefits of surgery. During this discussion, spell out the increased risks of conversion to laparotomy, prolonged anesthesia, postoperative thrombosis, wound infection, and pulmonary complications, and make sure all are listed on the written consent form.

Operative management

Prophylactic measures

Complete bowel preparation is recommended the evening prior to surgery, since intraabdominal visualization can be difficult and conversion to laparotomy may be necessary. Bowel prep decompresses the lumen, improving visualization and the outcome of any bowel injury.

Preoperative histamine receptor blockade is recommended for optimal results, since higher body mass can lead to increases in low pH gastric volume and difficulties with intubation.⁵ A typical regimen is 50 mg intravenous (IV) ranitidine 20 minutes prior to surgery.

Beta blockade. All patients with hypertension or a history of coronary artery disease should receive preoperative beta blockade, assuming there are no contraindications such as reactive airway disease or cardiac conduction block. Atenolol 10 mg IV 20 minutes prior to surgery is a standard initial dose. All patients already taking beta blockers should simply continue their home regimen through the day of surgery with small sips of water.

 

Antibiotics. The high rate of postoperative wound infection in obese patients makes preoperative antibiotic treatment logical. Although we were unable to find any studies demonstrating the benefit of routine antibiotic prophylaxis in obese patients undergoing laparoscopy, data do show a benefit when laparotomy is performed.

If no allergies or contraindications exist, give 1 to 2 g of a first-or second-generation cephalosporin intravenously 20 to 30 minutes prior to anesthesia induction.

Sequential compression devices. Since both obesity and gynecologic surgery are risk factors for deep venous thrombosis, use large sequential compression devices on the lower extremities, beginning before induction of anesthesia.

Position the patient for optimal access

Only 1 recent publication explores this issue in obese laparoscopy patients. Lamvu et al⁵ advocate the armstucked (“military”), low lithotomy position, with liberal padding on the legs and arms and a gel pad under the lower back. They also recommend stationary shoulder blocks to help maintain positioning in the Trendelenburg (head down) position, and they use clamps, gauze, weights, and tape to maintain the panniculus in its caudad position.

Novel technique realigns umbilical axis. We, too, use padding liberally on all pressure points, but do not weight the panniculus. In fact, we prefer its cephalad migration in the Trendelenburg position. Pelosi and Pelosi⁶ describe a useful technique to realign the umbilical axis cephalad before placing the first trocar (FIGURE). Once the Trendelenburg position is attained (after initial trocar placement), this cephalad position eases ancillary port placement.

Tucking 1 arm facilitates surgery, anesthesia access. Tucking both arms is ideal but not always feasible. It is especially problematic when adipose tissue surrounding the biceps makes the military position impossible. Further, anesthesiologists may be unwilling to abandon access to the peripheral intravenous site, since placement and emergency replacement can be difficult.

Central venous access is always an option but is not without risk and should be avoided, if possible. A creative alternative: Tuck the nonaccessed arm at the patient’s side and place the other arm over the chest. Maintain this position by tucking a sheet over the chest. This gives the anesthesia team access to 1 arm while facilitating ideal surgeon positioning.

Do not use shoulder blocks when the patient’s arms are extended, as this increases the risk of brachial plexus injury should the patient slide.

Success hinges on port placement, pneumoperitoneum

The success or failure of most laparoscopic surgeries is determined in the initial minutes during placement of the operative ports. This is especially true in obese patients. No single variable is more important to successful laparoscopy in obese patients than the establishment of pneumoperitoneum.

 

Entry variables of 3 body types. Obesity increases the distance between skin and fascia, and can increase the distance between fascia and peritoneum. The difficulty of placing the Veress needle or trocar into the peritoneal cavity increases with this distance. Preperitoneal insufflation of gas exacerbates the problem. In addition, dissection to the level of the fascia for an open (Hasson) approach sometimes requires incision extension and increases the risk of postoperative wound infection.

Obesity also changes the relationship of the umbilicus to the aortic bifurcation. Utilizing computed tomography, Hurd et al⁷ demonstrated that the umbilicus migrates caudally in relation to the aortic bifurcation as the BMI increases. In nonobese patients (BMI 30) patients, the umbilicus had a median location 2.4 and 2.9 cm caudal to the aortic bifurcation, respectively. However, in both groups, the umbilicus was directly over the aortic bifurcation in 30% of patients.

The same group of researchers, again using computed tomography, demonstrated that the distance between the umbilicus and peritoneum at a 45° angle from the umbilicus into the pelvis, in both nonobese and overweight patients, was only 2 cm. In obese patients, this distance increased to a median of 12 cm. Hurd et al⁸ also noted that the distance between the umbilicus and the underlying vessels was only 6 cm at a 90° angle in nonobese patients, but averaged 13 cm in obese patients.

To optimize intraperitoneal Veress needle and trocar placement while minimizing risk to the underlying vasculature, Hurd and colleagues recommend a 45° angle from the umbilicus toward the pelvis in nonobese patients and a 90° approach in obese patients. In overweight patients, the approach should range between 45° and 90° (TABLE 1).

TABLE 1

Instrument placement in laparoscopy: Anatomic distances and suggested angles

	DISTANCE FROM THE UMBILICUS (CM)
GROUP	TO BIFURCATION	TO PERITONEUM	TO VESSELS AT 90°	RECOMMENDED PLACEMENT ANGLE
Nonobese (BMI	0.4 ± 1.6	2 ± 2	6 ± 3	45°
Overweight (BMI 25–30)	2.4 ± 1.9	2 ± 1	10 ± 2	45–90°
Obese (BMI >30)	2.9 ± 2.5	12 (median)	13 ± 4	90°
Data are presented as mean ± standard deviation, median, or degrees from horizontal
Source: Hurd WW, et al 7

Gaining intraperitoneal access: Which approach is best?

A number of studies and case series have explored the fundamental difficulty of gaining intraperitoneal access. Pasic et al⁹ retrospectively analyzed outcomes in separate cohorts of obese and nonobese patients, focusing on 4 entry approaches:

• transumbilical open,
  
• transumbilical Veress needle placement,
  
• subcostal Veress needle placement in the midclavicular line of the left upper quadrant, and
  
• transuterine Veress needle placement.
  

The only group that demonstrated a significantly higher failure rate for obese patients was the open approach. Ultimately, the authors recommended using the Veress needle in the left upper quadrant or via the uterus for obese patients.

In contrast, the Pelosi case series of 67 consecutive obese patients⁶ reported no failures with a transumbilical open approach after realignment of the umbilical access. This entailed assessing the position of the umbilicus in relation to a line drawn between the 2 anterior superior iliac spines. The umbilicus then was repositioned 8 cm above this line in its “anatomical” position prior to initiating open dissection (FIGURE).

After the open trocar was inserted through the fascia and peritoneum and the patient was placed in the Trendelenburg position, the panniculus maintained its orientation. Pelosi and Pelosi concluded that this realignment of the umbilical axis decreases the depth of open dissection and avoids inadvertent placement of a trocar through both sides of the panniculus.

A prospective, randomized study¹⁰ comparing transumbilical and transuterine Veress needle placement in obese patients found the latter approach useful, but recorded a single case of postoperative chlamydial pelvic inflammatory disease. Thus, preoperative testing for sexually transmitted disease is recommended for this approach.

Avoid dogmatic reliance on a single approach

These studies demonstrate a fundamental surgical truism: Sound physiologic and anatomic knowledge, combined with versatility and a grasp of multiple approaches to any problem, are ultimately more successful than unyielding reliance on a single approach. Aim for prudent use of open or closed laparoscopy in a variety of locations, taking into account the patient’s surgical history, distribution of fat, and umbilical displacement.

After achieving pneumoperitoneum

Place a salinefilled spinal needle into the peritoneal cavity on suction to establish abdominal wall thickness. In this way, trocars of appropriate length can be selected.

Some authorities advocate insufflation to a high intraperitoneal pressure (25 to 30mm Hg) prior to placing the initial umbilical trocar if a closed technique is being used.¹¹ This further elevates the abdominal wall and decreases the risk of preperitoneal trocar placement. After successful trocar placement, immediately reduce intraabdominal pressure to 15 mm Hg to avoid pulmonary compromise, excessive catecholamine release, and subcutaneous emphysema.

 

Techniques to enhance visualization

Excess adipose tissue occupies the pericolic, omental, mesenteric, and retroperitoneal spaces in obese patients, obscuring visualization of intraperitoneal and retroperitoneal structures.

Preoperative mechanical bowel preparation can deflate the bowel and enhance visualization (TABLE 2). At times, an extra ancillary trocar for placement of a bowel retractor also can improve visualization.

In the morbidly obese, insufflation pressure of 15 mm Hg will sometimes produce poor visualization. Obese patients generally tolerate this pressure reasonably well, but increasing it to improve visualization can make adequate oxygenation impossible.

Gasless laparoscopy—in which a mechanical retractor is attached from the table to the patient’s anterior abdominal wall—may help improve pulmonary mechanical parameters. Unfortunately, this technique often produces poorer visualization than insufflation at normal pressure.

A new technique that combines approaches may help avoid the need to convert to laparotomy.¹² In this “Foley lap lift,” a 14-French Foley catheter is passed through the anterior abdominal wall, and the balloon is inflated. The catheter then is elevated and clamped to a retractor holder attached to the angled foot of the bed. This upward traction with continuous gas flow at normal pressure improves visualization without pulmonary compromise.

TABLE 2

Techniques to enhance visualization

Preoperative mechanical bowel preparation Ancillary trocar for placement of bowel retractor Gasless laparoscopy* Foley lap lift*
*See page 70 for details

Close port sites at the fascial level

The risk of bowel herniation through a trocar site is higher in obese patients than the general population because of the greater intraabdominal pressures. Increases in atelectasis from diminished functional residual capacity also predispose the obese patient to postoperative pulmonary complications and can lead to recurrent Valsalva (cough) and subsequent bowel herniation.

Given these risks, it is imperative that all port sites 10 mm or larger be closed at the fascial level. Unfortunately, the distance from the anterior abdominal wall to the fascia underlying these sites makes direct visualization and closure almost impossible.

Fortunately, several fascial closure devices are available and are reasonably inexpensive and easy to use. When using them, be sure to maintain the other port sites, as closure requires direct visualization and a second instrument.

Postoperative strategies

Successful postoperative care builds on preoperative and intraoperative tactics.

Perform aggressive pulmonary toilet

With intraoperative decreases in functional residual capacity, postoperative atelectasis is likely to be profound, with a potential for ventilation/perfusion mismatch and hypoxemia.

Aggressive pulmonary toilet including regular incentive spirometry and deep breathing and coughing exercises is important to reinflate dependent lung regions. Pulse oximetry with sufficient supplementary oxygen also is important to maintain adequate saturation.

Encourage early ambulation

This requires adequate but not oversedating analgesia, early catheter removal, and a motivated nursing staff.

Early ambulation is associated with fewer episodes of deep venous thrombosis, pulmonary complications, and ileus, and also eases pain management.

Continue thrombosis prophylaxis with sequential compression devices, subcutaneous heparin, or both, until the patient is spending most of her time out of bed.

Dr. Robinson reports no relevant financial relationships.

Dr. Isaacson serves on the speakers bureau for Karl Storz Endoscopy.

Gynecologic laparoscopy in the obese? What was once the purview of the very talented or the foolhardy may now be the preferred surgical approach.

Obese women who undergo laparoscopy recover faster, with less pain, fewer wound infections, and shorter hospital stays than with laparotomy. Though it is true that obesity increases operative time and the risk for conversion to laparotomy, little evidence supports the theory that a body mass index (BMI) of 30 kg/m² or higher should exclude laparoscopy.

Preoperative management

Unique elements of the physical

It is important to identify central obesity, which is more difficult to accommodate than distribution around the hips. Unfortunately, the roughly 40 million obese Americans tend to have central fat distribution.^1,2

In central obesity, the subcutaneous tissue is thick, often requiring extra long ports to attain peritoneal access.

The relationship of the umbilicus to the underlying aortic bifurcation also shifts more caudally. This relationship should be noted and planned for before going to the operating room (FIGURE).

Abdominal obesity in particular confers additional risks during all types of surgery: higher rates of atelectasis, thromboembolism, cardiovascular dysfunction, and wound infection.

Closely inspect the skin and panniculus after a routine examination. Obesity predisposes patients to dark, moist, anoxic spaces beneath folds of skin that need to be identified and inspected for evidence of fungal or bacterial infection. To optimize postoperative wound healing, treat any preexisting infections before surgery.

Cigarette smoking further burdens pulmonary mechanics and oxygenation during surgery, so it is important to encourage smokers to kick the habit at least 8 weeks before elective surgery.³

In general, use the history and physical examination to focus on the recognized risk factors of obesity, with specific emphasis on hypertension, coronary artery disease, arrhythmia, pulmonary obstructive disease, peripheral vascular disease, diabetes, gastric reflux, and arthritis.⁴

Special tests and laboratory studies

EKG and chest x-ray. In morbidly obese patients (BMI >40), preoperative evaluation includes an electrocardiogram (EKG) and chest x-ray to identify any cardiomegaly, arrhythmias, and occult ischemia or conduction blockage.

Arterial blood gas sampling. Given the higher risk of postoperative thrombotic events in obese patients, it can be helpful to assess preoperative oxygenation and ventilation/perfusion status via arterial blood gas sampling. The obese may have elevated baseline Aa gradients, which, if not noted prior to surgery, can confuse later management of suspected pulmonary emboli.

During testing, assess venous access and counsel the patient if central venous line placement may be possible at surgery. Though central line placement is not routinely recommended, it may be warranted in patients with particularly difficult peripheral venous access.

Skip pulmonary function testing because the results rarely change surgical management. We consider its routine use to be wasteful.

Laboratory evaluation should include betahuman chorionic gonadotropin (in premenopausal patients), complete blood count, electrolytes, glucose, renal function, and type and screen.

Spell out risks at informed consent

The preoperative appointment is your chance to answer questions the patient may have and clearly delineate the risks and benefits of surgery. During this discussion, spell out the increased risks of conversion to laparotomy, prolonged anesthesia, postoperative thrombosis, wound infection, and pulmonary complications, and make sure all are listed on the written consent form.

Operative management

Prophylactic measures

Complete bowel preparation is recommended the evening prior to surgery, since intraabdominal visualization can be difficult and conversion to laparotomy may be necessary. Bowel prep decompresses the lumen, improving visualization and the outcome of any bowel injury.

Preoperative histamine receptor blockade is recommended for optimal results, since higher body mass can lead to increases in low pH gastric volume and difficulties with intubation.⁵ A typical regimen is 50 mg intravenous (IV) ranitidine 20 minutes prior to surgery.

Beta blockade. All patients with hypertension or a history of coronary artery disease should receive preoperative beta blockade, assuming there are no contraindications such as reactive airway disease or cardiac conduction block. Atenolol 10 mg IV 20 minutes prior to surgery is a standard initial dose. All patients already taking beta blockers should simply continue their home regimen through the day of surgery with small sips of water.

 

Antibiotics. The high rate of postoperative wound infection in obese patients makes preoperative antibiotic treatment logical. Although we were unable to find any studies demonstrating the benefit of routine antibiotic prophylaxis in obese patients undergoing laparoscopy, data do show a benefit when laparotomy is performed.

If no allergies or contraindications exist, give 1 to 2 g of a first-or second-generation cephalosporin intravenously 20 to 30 minutes prior to anesthesia induction.

Sequential compression devices. Since both obesity and gynecologic surgery are risk factors for deep venous thrombosis, use large sequential compression devices on the lower extremities, beginning before induction of anesthesia.

Position the patient for optimal access

Only 1 recent publication explores this issue in obese laparoscopy patients. Lamvu et al⁵ advocate the armstucked (“military”), low lithotomy position, with liberal padding on the legs and arms and a gel pad under the lower back. They also recommend stationary shoulder blocks to help maintain positioning in the Trendelenburg (head down) position, and they use clamps, gauze, weights, and tape to maintain the panniculus in its caudad position.

Novel technique realigns umbilical axis. We, too, use padding liberally on all pressure points, but do not weight the panniculus. In fact, we prefer its cephalad migration in the Trendelenburg position. Pelosi and Pelosi⁶ describe a useful technique to realign the umbilical axis cephalad before placing the first trocar (FIGURE). Once the Trendelenburg position is attained (after initial trocar placement), this cephalad position eases ancillary port placement.

Tucking 1 arm facilitates surgery, anesthesia access. Tucking both arms is ideal but not always feasible. It is especially problematic when adipose tissue surrounding the biceps makes the military position impossible. Further, anesthesiologists may be unwilling to abandon access to the peripheral intravenous site, since placement and emergency replacement can be difficult.

Central venous access is always an option but is not without risk and should be avoided, if possible. A creative alternative: Tuck the nonaccessed arm at the patient’s side and place the other arm over the chest. Maintain this position by tucking a sheet over the chest. This gives the anesthesia team access to 1 arm while facilitating ideal surgeon positioning.

Do not use shoulder blocks when the patient’s arms are extended, as this increases the risk of brachial plexus injury should the patient slide.

Success hinges on port placement, pneumoperitoneum

The success or failure of most laparoscopic surgeries is determined in the initial minutes during placement of the operative ports. This is especially true in obese patients. No single variable is more important to successful laparoscopy in obese patients than the establishment of pneumoperitoneum.

 

Entry variables of 3 body types. Obesity increases the distance between skin and fascia, and can increase the distance between fascia and peritoneum. The difficulty of placing the Veress needle or trocar into the peritoneal cavity increases with this distance. Preperitoneal insufflation of gas exacerbates the problem. In addition, dissection to the level of the fascia for an open (Hasson) approach sometimes requires incision extension and increases the risk of postoperative wound infection.

Obesity also changes the relationship of the umbilicus to the aortic bifurcation. Utilizing computed tomography, Hurd et al⁷ demonstrated that the umbilicus migrates caudally in relation to the aortic bifurcation as the BMI increases. In nonobese patients (BMI 30) patients, the umbilicus had a median location 2.4 and 2.9 cm caudal to the aortic bifurcation, respectively. However, in both groups, the umbilicus was directly over the aortic bifurcation in 30% of patients.

The same group of researchers, again using computed tomography, demonstrated that the distance between the umbilicus and peritoneum at a 45° angle from the umbilicus into the pelvis, in both nonobese and overweight patients, was only 2 cm. In obese patients, this distance increased to a median of 12 cm. Hurd et al⁸ also noted that the distance between the umbilicus and the underlying vessels was only 6 cm at a 90° angle in nonobese patients, but averaged 13 cm in obese patients.

To optimize intraperitoneal Veress needle and trocar placement while minimizing risk to the underlying vasculature, Hurd and colleagues recommend a 45° angle from the umbilicus toward the pelvis in nonobese patients and a 90° approach in obese patients. In overweight patients, the approach should range between 45° and 90° (TABLE 1).

TABLE 1

Instrument placement in laparoscopy: Anatomic distances and suggested angles

	DISTANCE FROM THE UMBILICUS (CM)
GROUP	TO BIFURCATION	TO PERITONEUM	TO VESSELS AT 90°	RECOMMENDED PLACEMENT ANGLE
Nonobese (BMI	0.4 ± 1.6	2 ± 2	6 ± 3	45°
Overweight (BMI 25–30)	2.4 ± 1.9	2 ± 1	10 ± 2	45–90°
Obese (BMI >30)	2.9 ± 2.5	12 (median)	13 ± 4	90°
Data are presented as mean ± standard deviation, median, or degrees from horizontal
Source: Hurd WW, et al 7

Gaining intraperitoneal access: Which approach is best?

A number of studies and case series have explored the fundamental difficulty of gaining intraperitoneal access. Pasic et al⁹ retrospectively analyzed outcomes in separate cohorts of obese and nonobese patients, focusing on 4 entry approaches:

• transumbilical open,
  
• transumbilical Veress needle placement,
  
• subcostal Veress needle placement in the midclavicular line of the left upper quadrant, and
  
• transuterine Veress needle placement.
  

The only group that demonstrated a significantly higher failure rate for obese patients was the open approach. Ultimately, the authors recommended using the Veress needle in the left upper quadrant or via the uterus for obese patients.

In contrast, the Pelosi case series of 67 consecutive obese patients⁶ reported no failures with a transumbilical open approach after realignment of the umbilical access. This entailed assessing the position of the umbilicus in relation to a line drawn between the 2 anterior superior iliac spines. The umbilicus then was repositioned 8 cm above this line in its “anatomical” position prior to initiating open dissection (FIGURE).

After the open trocar was inserted through the fascia and peritoneum and the patient was placed in the Trendelenburg position, the panniculus maintained its orientation. Pelosi and Pelosi concluded that this realignment of the umbilical axis decreases the depth of open dissection and avoids inadvertent placement of a trocar through both sides of the panniculus.

A prospective, randomized study¹⁰ comparing transumbilical and transuterine Veress needle placement in obese patients found the latter approach useful, but recorded a single case of postoperative chlamydial pelvic inflammatory disease. Thus, preoperative testing for sexually transmitted disease is recommended for this approach.

Avoid dogmatic reliance on a single approach

These studies demonstrate a fundamental surgical truism: Sound physiologic and anatomic knowledge, combined with versatility and a grasp of multiple approaches to any problem, are ultimately more successful than unyielding reliance on a single approach. Aim for prudent use of open or closed laparoscopy in a variety of locations, taking into account the patient’s surgical history, distribution of fat, and umbilical displacement.

After achieving pneumoperitoneum

Place a salinefilled spinal needle into the peritoneal cavity on suction to establish abdominal wall thickness. In this way, trocars of appropriate length can be selected.

Some authorities advocate insufflation to a high intraperitoneal pressure (25 to 30mm Hg) prior to placing the initial umbilical trocar if a closed technique is being used.¹¹ This further elevates the abdominal wall and decreases the risk of preperitoneal trocar placement. After successful trocar placement, immediately reduce intraabdominal pressure to 15 mm Hg to avoid pulmonary compromise, excessive catecholamine release, and subcutaneous emphysema.

 

Techniques to enhance visualization

Excess adipose tissue occupies the pericolic, omental, mesenteric, and retroperitoneal spaces in obese patients, obscuring visualization of intraperitoneal and retroperitoneal structures.

Preoperative mechanical bowel preparation can deflate the bowel and enhance visualization (TABLE 2). At times, an extra ancillary trocar for placement of a bowel retractor also can improve visualization.

In the morbidly obese, insufflation pressure of 15 mm Hg will sometimes produce poor visualization. Obese patients generally tolerate this pressure reasonably well, but increasing it to improve visualization can make adequate oxygenation impossible.

Gasless laparoscopy—in which a mechanical retractor is attached from the table to the patient’s anterior abdominal wall—may help improve pulmonary mechanical parameters. Unfortunately, this technique often produces poorer visualization than insufflation at normal pressure.

A new technique that combines approaches may help avoid the need to convert to laparotomy.¹² In this “Foley lap lift,” a 14-French Foley catheter is passed through the anterior abdominal wall, and the balloon is inflated. The catheter then is elevated and clamped to a retractor holder attached to the angled foot of the bed. This upward traction with continuous gas flow at normal pressure improves visualization without pulmonary compromise.

TABLE 2

Techniques to enhance visualization

Preoperative mechanical bowel preparation Ancillary trocar for placement of bowel retractor Gasless laparoscopy* Foley lap lift*
*See page 70 for details

Close port sites at the fascial level

The risk of bowel herniation through a trocar site is higher in obese patients than the general population because of the greater intraabdominal pressures. Increases in atelectasis from diminished functional residual capacity also predispose the obese patient to postoperative pulmonary complications and can lead to recurrent Valsalva (cough) and subsequent bowel herniation.

Given these risks, it is imperative that all port sites 10 mm or larger be closed at the fascial level. Unfortunately, the distance from the anterior abdominal wall to the fascia underlying these sites makes direct visualization and closure almost impossible.

Fortunately, several fascial closure devices are available and are reasonably inexpensive and easy to use. When using them, be sure to maintain the other port sites, as closure requires direct visualization and a second instrument.

Postoperative strategies

Successful postoperative care builds on preoperative and intraoperative tactics.

Perform aggressive pulmonary toilet

With intraoperative decreases in functional residual capacity, postoperative atelectasis is likely to be profound, with a potential for ventilation/perfusion mismatch and hypoxemia.

Aggressive pulmonary toilet including regular incentive spirometry and deep breathing and coughing exercises is important to reinflate dependent lung regions. Pulse oximetry with sufficient supplementary oxygen also is important to maintain adequate saturation.

Encourage early ambulation

This requires adequate but not oversedating analgesia, early catheter removal, and a motivated nursing staff.

Early ambulation is associated with fewer episodes of deep venous thrombosis, pulmonary complications, and ileus, and also eases pain management.

Continue thrombosis prophylaxis with sequential compression devices, subcutaneous heparin, or both, until the patient is spending most of her time out of bed.

Dr. Robinson reports no relevant financial relationships.

Dr. Isaacson serves on the speakers bureau for Karl Storz Endoscopy.

Gynecologic laparoscopy in the obese? What was once the purview of the very talented or the foolhardy may now be the preferred surgical approach.

Obese women who undergo laparoscopy recover faster, with less pain, fewer wound infections, and shorter hospital stays than with laparotomy. Though it is true that obesity increases operative time and the risk for conversion to laparotomy, little evidence supports the theory that a body mass index (BMI) of 30 kg/m² or higher should exclude laparoscopy.

Preoperative management

Unique elements of the physical

It is important to identify central obesity, which is more difficult to accommodate than distribution around the hips. Unfortunately, the roughly 40 million obese Americans tend to have central fat distribution.^1,2

In central obesity, the subcutaneous tissue is thick, often requiring extra long ports to attain peritoneal access.

The relationship of the umbilicus to the underlying aortic bifurcation also shifts more caudally. This relationship should be noted and planned for before going to the operating room (FIGURE).

Abdominal obesity in particular confers additional risks during all types of surgery: higher rates of atelectasis, thromboembolism, cardiovascular dysfunction, and wound infection.

Closely inspect the skin and panniculus after a routine examination. Obesity predisposes patients to dark, moist, anoxic spaces beneath folds of skin that need to be identified and inspected for evidence of fungal or bacterial infection. To optimize postoperative wound healing, treat any preexisting infections before surgery.

Cigarette smoking further burdens pulmonary mechanics and oxygenation during surgery, so it is important to encourage smokers to kick the habit at least 8 weeks before elective surgery.³

In general, use the history and physical examination to focus on the recognized risk factors of obesity, with specific emphasis on hypertension, coronary artery disease, arrhythmia, pulmonary obstructive disease, peripheral vascular disease, diabetes, gastric reflux, and arthritis.⁴

Special tests and laboratory studies

EKG and chest x-ray. In morbidly obese patients (BMI >40), preoperative evaluation includes an electrocardiogram (EKG) and chest x-ray to identify any cardiomegaly, arrhythmias, and occult ischemia or conduction blockage.

Arterial blood gas sampling. Given the higher risk of postoperative thrombotic events in obese patients, it can be helpful to assess preoperative oxygenation and ventilation/perfusion status via arterial blood gas sampling. The obese may have elevated baseline Aa gradients, which, if not noted prior to surgery, can confuse later management of suspected pulmonary emboli.

During testing, assess venous access and counsel the patient if central venous line placement may be possible at surgery. Though central line placement is not routinely recommended, it may be warranted in patients with particularly difficult peripheral venous access.

Skip pulmonary function testing because the results rarely change surgical management. We consider its routine use to be wasteful.

Laboratory evaluation should include betahuman chorionic gonadotropin (in premenopausal patients), complete blood count, electrolytes, glucose, renal function, and type and screen.

Spell out risks at informed consent

The preoperative appointment is your chance to answer questions the patient may have and clearly delineate the risks and benefits of surgery. During this discussion, spell out the increased risks of conversion to laparotomy, prolonged anesthesia, postoperative thrombosis, wound infection, and pulmonary complications, and make sure all are listed on the written consent form.

Operative management

Prophylactic measures

Complete bowel preparation is recommended the evening prior to surgery, since intraabdominal visualization can be difficult and conversion to laparotomy may be necessary. Bowel prep decompresses the lumen, improving visualization and the outcome of any bowel injury.

Preoperative histamine receptor blockade is recommended for optimal results, since higher body mass can lead to increases in low pH gastric volume and difficulties with intubation.⁵ A typical regimen is 50 mg intravenous (IV) ranitidine 20 minutes prior to surgery.

Beta blockade. All patients with hypertension or a history of coronary artery disease should receive preoperative beta blockade, assuming there are no contraindications such as reactive airway disease or cardiac conduction block. Atenolol 10 mg IV 20 minutes prior to surgery is a standard initial dose. All patients already taking beta blockers should simply continue their home regimen through the day of surgery with small sips of water.

 

Antibiotics. The high rate of postoperative wound infection in obese patients makes preoperative antibiotic treatment logical. Although we were unable to find any studies demonstrating the benefit of routine antibiotic prophylaxis in obese patients undergoing laparoscopy, data do show a benefit when laparotomy is performed.

If no allergies or contraindications exist, give 1 to 2 g of a first-or second-generation cephalosporin intravenously 20 to 30 minutes prior to anesthesia induction.

Sequential compression devices. Since both obesity and gynecologic surgery are risk factors for deep venous thrombosis, use large sequential compression devices on the lower extremities, beginning before induction of anesthesia.

Position the patient for optimal access

Only 1 recent publication explores this issue in obese laparoscopy patients. Lamvu et al⁵ advocate the armstucked (“military”), low lithotomy position, with liberal padding on the legs and arms and a gel pad under the lower back. They also recommend stationary shoulder blocks to help maintain positioning in the Trendelenburg (head down) position, and they use clamps, gauze, weights, and tape to maintain the panniculus in its caudad position.

Novel technique realigns umbilical axis. We, too, use padding liberally on all pressure points, but do not weight the panniculus. In fact, we prefer its cephalad migration in the Trendelenburg position. Pelosi and Pelosi⁶ describe a useful technique to realign the umbilical axis cephalad before placing the first trocar (FIGURE). Once the Trendelenburg position is attained (after initial trocar placement), this cephalad position eases ancillary port placement.

Tucking 1 arm facilitates surgery, anesthesia access. Tucking both arms is ideal but not always feasible. It is especially problematic when adipose tissue surrounding the biceps makes the military position impossible. Further, anesthesiologists may be unwilling to abandon access to the peripheral intravenous site, since placement and emergency replacement can be difficult.

Central venous access is always an option but is not without risk and should be avoided, if possible. A creative alternative: Tuck the nonaccessed arm at the patient’s side and place the other arm over the chest. Maintain this position by tucking a sheet over the chest. This gives the anesthesia team access to 1 arm while facilitating ideal surgeon positioning.

Do not use shoulder blocks when the patient’s arms are extended, as this increases the risk of brachial plexus injury should the patient slide.

Success hinges on port placement, pneumoperitoneum

The success or failure of most laparoscopic surgeries is determined in the initial minutes during placement of the operative ports. This is especially true in obese patients. No single variable is more important to successful laparoscopy in obese patients than the establishment of pneumoperitoneum.

 

Entry variables of 3 body types. Obesity increases the distance between skin and fascia, and can increase the distance between fascia and peritoneum. The difficulty of placing the Veress needle or trocar into the peritoneal cavity increases with this distance. Preperitoneal insufflation of gas exacerbates the problem. In addition, dissection to the level of the fascia for an open (Hasson) approach sometimes requires incision extension and increases the risk of postoperative wound infection.

Obesity also changes the relationship of the umbilicus to the aortic bifurcation. Utilizing computed tomography, Hurd et al⁷ demonstrated that the umbilicus migrates caudally in relation to the aortic bifurcation as the BMI increases. In nonobese patients (BMI 30) patients, the umbilicus had a median location 2.4 and 2.9 cm caudal to the aortic bifurcation, respectively. However, in both groups, the umbilicus was directly over the aortic bifurcation in 30% of patients.

The same group of researchers, again using computed tomography, demonstrated that the distance between the umbilicus and peritoneum at a 45° angle from the umbilicus into the pelvis, in both nonobese and overweight patients, was only 2 cm. In obese patients, this distance increased to a median of 12 cm. Hurd et al⁸ also noted that the distance between the umbilicus and the underlying vessels was only 6 cm at a 90° angle in nonobese patients, but averaged 13 cm in obese patients.

To optimize intraperitoneal Veress needle and trocar placement while minimizing risk to the underlying vasculature, Hurd and colleagues recommend a 45° angle from the umbilicus toward the pelvis in nonobese patients and a 90° approach in obese patients. In overweight patients, the approach should range between 45° and 90° (TABLE 1).

TABLE 1

Instrument placement in laparoscopy: Anatomic distances and suggested angles

	DISTANCE FROM THE UMBILICUS (CM)
GROUP	TO BIFURCATION	TO PERITONEUM	TO VESSELS AT 90°	RECOMMENDED PLACEMENT ANGLE
Nonobese (BMI	0.4 ± 1.6	2 ± 2	6 ± 3	45°
Overweight (BMI 25–30)	2.4 ± 1.9	2 ± 1	10 ± 2	45–90°
Obese (BMI >30)	2.9 ± 2.5	12 (median)	13 ± 4	90°
Data are presented as mean ± standard deviation, median, or degrees from horizontal
Source: Hurd WW, et al 7

Gaining intraperitoneal access: Which approach is best?

A number of studies and case series have explored the fundamental difficulty of gaining intraperitoneal access. Pasic et al⁹ retrospectively analyzed outcomes in separate cohorts of obese and nonobese patients, focusing on 4 entry approaches:

• transumbilical open,
  
• transumbilical Veress needle placement,
  
• subcostal Veress needle placement in the midclavicular line of the left upper quadrant, and
  
• transuterine Veress needle placement.
  

The only group that demonstrated a significantly higher failure rate for obese patients was the open approach. Ultimately, the authors recommended using the Veress needle in the left upper quadrant or via the uterus for obese patients.

In contrast, the Pelosi case series of 67 consecutive obese patients⁶ reported no failures with a transumbilical open approach after realignment of the umbilical access. This entailed assessing the position of the umbilicus in relation to a line drawn between the 2 anterior superior iliac spines. The umbilicus then was repositioned 8 cm above this line in its “anatomical” position prior to initiating open dissection (FIGURE).

After the open trocar was inserted through the fascia and peritoneum and the patient was placed in the Trendelenburg position, the panniculus maintained its orientation. Pelosi and Pelosi concluded that this realignment of the umbilical axis decreases the depth of open dissection and avoids inadvertent placement of a trocar through both sides of the panniculus.

A prospective, randomized study¹⁰ comparing transumbilical and transuterine Veress needle placement in obese patients found the latter approach useful, but recorded a single case of postoperative chlamydial pelvic inflammatory disease. Thus, preoperative testing for sexually transmitted disease is recommended for this approach.

Avoid dogmatic reliance on a single approach

These studies demonstrate a fundamental surgical truism: Sound physiologic and anatomic knowledge, combined with versatility and a grasp of multiple approaches to any problem, are ultimately more successful than unyielding reliance on a single approach. Aim for prudent use of open or closed laparoscopy in a variety of locations, taking into account the patient’s surgical history, distribution of fat, and umbilical displacement.

After achieving pneumoperitoneum

Place a salinefilled spinal needle into the peritoneal cavity on suction to establish abdominal wall thickness. In this way, trocars of appropriate length can be selected.

Some authorities advocate insufflation to a high intraperitoneal pressure (25 to 30mm Hg) prior to placing the initial umbilical trocar if a closed technique is being used.¹¹ This further elevates the abdominal wall and decreases the risk of preperitoneal trocar placement. After successful trocar placement, immediately reduce intraabdominal pressure to 15 mm Hg to avoid pulmonary compromise, excessive catecholamine release, and subcutaneous emphysema.

 

Techniques to enhance visualization

Excess adipose tissue occupies the pericolic, omental, mesenteric, and retroperitoneal spaces in obese patients, obscuring visualization of intraperitoneal and retroperitoneal structures.

Preoperative mechanical bowel preparation can deflate the bowel and enhance visualization (TABLE 2). At times, an extra ancillary trocar for placement of a bowel retractor also can improve visualization.

In the morbidly obese, insufflation pressure of 15 mm Hg will sometimes produce poor visualization. Obese patients generally tolerate this pressure reasonably well, but increasing it to improve visualization can make adequate oxygenation impossible.

Gasless laparoscopy—in which a mechanical retractor is attached from the table to the patient’s anterior abdominal wall—may help improve pulmonary mechanical parameters. Unfortunately, this technique often produces poorer visualization than insufflation at normal pressure.

A new technique that combines approaches may help avoid the need to convert to laparotomy.¹² In this “Foley lap lift,” a 14-French Foley catheter is passed through the anterior abdominal wall, and the balloon is inflated. The catheter then is elevated and clamped to a retractor holder attached to the angled foot of the bed. This upward traction with continuous gas flow at normal pressure improves visualization without pulmonary compromise.

TABLE 2

Techniques to enhance visualization

Preoperative mechanical bowel preparation Ancillary trocar for placement of bowel retractor Gasless laparoscopy* Foley lap lift*
*See page 70 for details

Close port sites at the fascial level

The risk of bowel herniation through a trocar site is higher in obese patients than the general population because of the greater intraabdominal pressures. Increases in atelectasis from diminished functional residual capacity also predispose the obese patient to postoperative pulmonary complications and can lead to recurrent Valsalva (cough) and subsequent bowel herniation.

Given these risks, it is imperative that all port sites 10 mm or larger be closed at the fascial level. Unfortunately, the distance from the anterior abdominal wall to the fascia underlying these sites makes direct visualization and closure almost impossible.

Fortunately, several fascial closure devices are available and are reasonably inexpensive and easy to use. When using them, be sure to maintain the other port sites, as closure requires direct visualization and a second instrument.

Postoperative strategies

Successful postoperative care builds on preoperative and intraoperative tactics.

Perform aggressive pulmonary toilet

With intraoperative decreases in functional residual capacity, postoperative atelectasis is likely to be profound, with a potential for ventilation/perfusion mismatch and hypoxemia.

Aggressive pulmonary toilet including regular incentive spirometry and deep breathing and coughing exercises is important to reinflate dependent lung regions. Pulse oximetry with sufficient supplementary oxygen also is important to maintain adequate saturation.

Encourage early ambulation

This requires adequate but not oversedating analgesia, early catheter removal, and a motivated nursing staff.

Early ambulation is associated with fewer episodes of deep venous thrombosis, pulmonary complications, and ileus, and also eases pain management.

Continue thrombosis prophylaxis with sequential compression devices, subcutaneous heparin, or both, until the patient is spending most of her time out of bed.

Dr. Robinson reports no relevant financial relationships.

Dr. Isaacson serves on the speakers bureau for Karl Storz Endoscopy.

WHAT WENT WRONG?

A 44-year-old woman undergoing resection of a submucous myoma from the left cornual region has persistent bleeding at the resection site. The surgeon continues coagulation at the bleeding site, using a rollerball electrode in an attempt to achieve hemostasis, but perforates the uterus. Immediate laparoscopy to identify collateral injury reveals some thermal damage on the posterior leaf of the broad ligament, but no bowel injury. After 24 hours of observation, she is afebrile without leukocytosis. She is discharged with explicit instructions to return if she has symptoms suggesting bowel injury. She returns in 72 hours, with abdominal pain and low-grade fever. CT reveals extravasation of contrast from the left ureter in the pelvis. Immediate laparotomy finds perforation of the left ureter secondary to a thermal injury. She undergoes ureteroneocystotomy and recovers.

This case illustrates one of the most common complications of operative hysteroscopy: uterine perforation with collateral injury. Both could have been avoided if the Ob/Gyn had stopped the procedure when bleeding first occurred, removed the instruments, and allowed the uterus to contract spontaneously.

This is just one of the strategies that can reduce the risks of hysteroscopic surgery. Numerous reports confirm that operative hysteroscopy is safe and effective, but as more gynecologists perform an increasing number of procedures, we must be aware of potential complications and do our best to minimize risk to our patients.

Complications cannot be completely avoided, and may occur when a procedure is done correctly by experienced doctors. They are far more likely if techniques or equipment are used improperly. This article describes ways to minimize risk.

When the American Association of Gynecologic Laparoscopists (AAGL) surveyed its members in 1993, it found a complication rate of 2% for operative hysteroscopy.¹ The rate of major complications—perforation, hemorrhage, fluid overload, and bowel or urinary tract injury—was less than 1%. A prospective multicenter trial² of 13,600 procedures in the Netherlands found a higher complication rate for operative (0.95%) than for diagnostic hysteroscopy (0.13%).

Preoperative precautions

We can reduce the risk of complications if contraindications are not ignored, equipment is thoroughly inspected and understood, and the surgeon goes through a mental checklist and plans each procedure. A “time out”before the operation begins, when every member of the team is briefed, is also valuable in preventing errors.

A hands-on course necessary before undertaking advanced resectoscopic surgery, to become familiar with equipment and techniques, followed by proctoring by a surgeon credentialed for the procedure.

Contraindications

Ignoring contraindications to hysteroscopic surgery increases the risk of complications and is the single greatest factor leading to patient injury and physician liability.

Contraindications include:

• Unfamiliarity with equipment, instruments, or technique
• Lack of appropriate equipment or staff familiar with the equipment
• Acute pelvic inflammatory disease
• Pregnancy
• Genital tract malignancies
• Lack of informed consent
• Inability to dilate the cervix
• Inability to distend the uterus to obtain visualization
• Poor surgical candidates who may not tolerate fluid overload because of renal disease, or radiofrequency current when a cardiac pacemaker is present
• The patient desires and expects complete amenorrhea³

Mechanical or traumatic complications

These types of complications are among the most common. Other categories include preoperative complications (ie, improper patient selection and lack of informed consent), electrosurgical and gaseous, complications related to distention media, and postoperative complications (ie, infection and late sequelae).

 

Inability to insert the hysteroscope

This may be caused by a stenotic, nulliparous cervix; menopause; GnRH agonists; previous cone biopsy, laceration, or cryosurgery; or an acutely retroflexed or anteflexed uterus.

Acute flexion problems can be corrected using a long-bladed, open-sided Graves speculum deep in the anterior or posterior fornix. The speculum pushes the fundus to the midposition and facilitates dilation. Once the hysteroscope is inserted, remove the speculum.

Placing a tenaculum on the posterior lip of the cervix of an acutely retroflexed uterus will straighten the cervical canal when traction is applied.

Inserting a laminaria tent the evening before surgery helps dilate the cervix easily and atraumatically.⁴ However, the laminaria can sometimes create a false passage, leading to perforation.

Cervical ripening agents such as intravaginal or oral misoprostol (200 μg inserted vaginally or 400 μg orally 8 to 12 hours preoperatively) also can facilitate dilation.

Intracervical injection of vasopressin solution (4 IU in 100 cc sodium chloride, injected at the 4 and 8 o’clock positions) can reduce the force needed to dilate the cervix.⁵ Half-size dilators may help; they also reduce the risk of cervical laceration.

Laceration of the cervix

Although this is a minor complication, substantial bleeding sometimes occurs when the cervix is lacerated by the tenaculum. In these cases, suture the cervix.

Occasionally, a touch of cautery from the rollerball electrode at low power (20 to 30 W) can control the bleeding.

Silver nitrate sticks or ferric subsulfate (Monsel’s) paste are also effective on superficial lacerations.

Bleeding from lower uterus or cervical canal can obscure view

In some cases, bleeding is delayed, necessitating additional surgery. Intravasation of distention fluid also can occur at these lacerations. Coagulation with the electrode may be necessary when bleeding is heavy.

Check for collateral injury when uterine perforation occurs

Perforation is a well-documented risk of operative hysteroscopy and should be discussed with the patient when obtaining informed consent. In the AAGL survey,¹ the incidence of perforation was 14 per 1,000. It was even higher during transection of lateral and fundal adhesions: 2 to 3 per 100.⁶

Although perforation is more common with thermal energy sources, it may occur mechanically when scissors are used to transect a uterine septum, synechiae, or polyps.

When the cervix is stenotic or the uterus is acutely ante- or retroflexed, sounds and dilators can perforate the uterus.

Most perforations—even those involving large dilators—usually do not require treatment, although further assessment may be necessary to rule out bowel injury. Most perforations occur in the fundal region or posterior lower segment.

A false passage can be created when entering the uterus. Occasionally the surgeon may be fooled into thinking the hysteroscope is in the uterine cavity, since the false passage distends (FIGURE 1). If muscle fibers are visible and the tubal ostea are not, assume the passage is false. Slowly remove the hysteroscope and identify the true cavity for confirmation. Discontinue the procedure—even if no perforation is detected—to prevent distention fluid from being absorbed into the circulation through the injury. Adequate distention is not possible at this time.

Delay repeat hysteroscopy for 2 to 3 months.

To avoid creating a false passage, dilate the cervix with slow, steady pressure and stop as soon as the internal os opens; do not attempt to push the dilator to the uterine fundus.

Often the external os opens, but the internal os cannot be dilated the extra 1 to 2 mm necessary to accommodate the 27-French resectoscope. Rather than exert more force and risk perforation or laceration, simply turn on the resectoscope’s inflow with the outflow shut off, and let the fluid pressure dilate the cervix.

Always insert the hysteroscope or resectoscope under direct vision rather than use an obturator. Keep the “dark circle” in the center of the field and slowly advance the hysteroscope toward it until the cavity is reached.

Avulsion of the myometrium sometimes occurs during removal of incompletely resected myomas (FIGURE 2). Keep the myoma grasper away from the fundus when removing myoma segments, and avoid excessive traction on what may be a thin segment of myometrium. Injuries can occur when the grasper perforates the uterus and bowel is inadvertently grasped. Large injuries require laparoscopic repair.

Perforation is more likely in repeat procedures. In a report of 80 repeat endometrial ablations, Townsend and colleagues⁷ noted 8 perforations that prevented completion of the procedure. In a series of 75 repeat ablations compared with 800 primary ablations by the same surgeon, the rate of serious perioperative complications was significantly higher in the repeat ablation group (9.3% versus 2.0%).⁸

 

When perforation occurs during the use of thermal energy, laparoscopy is necessary to assess the organs overlying the site.⁹ During setup for laparoscopy, bring the hysteroscope near the area of perforation to inspect the bowel beyond the uterus. Since the pelvis fills quickly with distention fluid, the hysteroscope can even be placed through the perforation to yield an excellent view of the undersurfaces of the bowel immediately adjacent to the injured area (FIGURE 3). (Disconnect the electrosurgical cord before doing this!)

Thorough laparoscopic inspection of the bowel in the pelvis often reveals thermal injury, which appears as a whitish patch on the bowel serosa. To repair bowel injuries, bring the injured segment out through a minilaparotomy and excise the damage with a 2- to 3-cm border. A general surgeon should be called in to consult.

If no injury is apparent, discharge the patient but follow her closely, including daily white blood cell counts for 4 to 5 days. Instruct her to take her temperature twice daily and return to the hospital immediately if any signs of bowel perforation develop. Delayed perforation from thermal injury can occur as late as 2 weeks following surgery, and the patient should be apprised of this possibility.

FIGURE 1 Signs of a false passage

Myometrial fibers signal that a false passage has been created. Stop the procedure even if no perforation is detected, to prevent distention fluid from being absorbed into the circulation through the injury.

FIGURE 2 Risk of myomectomy: Myometrial avulsion

Small bowel visible within the uterine cavity after avulsion of uterine wall at the time of myomectomy

FIGURE 3 Use the hysteroscope to assess perforation site

Hysteroscopic view of perforation at the fundus. The small bowel is visible beyond the perforation at left.

Intraoperative bleeding is rare

Bleeding is unlikely unless vessels are lacerated or injured in the cervical canal or lower uterine segment during dilation or deep ablation or vaporization. Bleeding is more common when endomyometrial resection is performed with the wire loop electrode or during ablation or vaporization of fibroids. Bleeding sufficient to require intervention occurs at a rate of 0.5% to 1.9% in several reported series.

To achieve hemostasis via balloon tamponade, insert a Foley catheter with a 30-cc balloon into the uterine cavity, inject 15 to 20 mL (or more for a larger cavity) of fluid into the balloon, and observe the patient.¹⁰ If there is no bleeding in 1 hour, remove half the fluid. Remove the remainder of the fluid and the catheter over the next hour if no further bleeding occurs.

Alternative method: Pack the uterus. I prefer 1/2-inch–gauge packing that has been soaked in a dilute vasopressin solution (20 U [1 mL] in 60 mL normal saline).¹¹

The benefits of vasopressin. Before balloon tamponade or packing the uterus, I inject very dilute vasopressin (4 U [0.2 mL] in 60 mL normal saline) directly into the cervix 2 cm deep, at the 4 and 8 o’clock positions. Phillips¹² demonstrated a marked decrease in blood loss during resectoscopic surgery using this approach. I also do this routinely prior to operative hysteroscopy, since the vasopressin-induced vasoconstriction reduces intravasation of distention media.

A vaporizing electrode may prevent significant blood loss during myoma resection by sealing blood vessels as the tissue is vaporized.¹³ All major manufacturers of hysteroscopic equipment produce these electrodes.

In my series of 44 endometrial ablations and hysteroscopic myomectomies performed with the vaporizing electrode, blood loss was “minimal” or less than 50 mL in 29 cases. The maximum blood loss was 300 mL in a patient with a 4-cm submucous myoma who was managed emergently for hemorrhage.¹⁴ In another case, during resection of a 5-cm submucous myoma, I encountered significant bleeding from large vessels at the base of the myoma, which required intrauterine tamponade with a vasopressin-soaked pack.

 

Preoperative measures may decrease vascularity. In their analysis of 16 randomized and nonrandomized controlled trials published in the English literature between 1990 and 1996, Parazzini and colleagues¹⁵ found that preoperative danazol or GnRH agonists decreased the thickness and vascularity of the endometrium and shrank myomata, resulting in shorter operating times, less blood loss, and less intravasation of distending fluid.

Electrosurgical and gaseous complications

Most electrosurgical complications involve activation of an electrode at the time of perforation, or current diversion to the outer sheath.

Thermal injuries also can be caused by overheating of the return pad or use of a weighted speculum that has not fully cooled after removal from the autoclave. The latter can be avoided by immersing the entire speculum in cool saline for at least 1 to 2 minutes prior to inserting it into the vagina. The blade cools much faster than the weighted ball, so be sure to check both to prevent a perineal or buttock burn.

Perforation with an active electrode

This usually occurs when current is applied as the electrode is extended or the resectoscope is moved toward the fundus. It can be avoided if the electrode is activated only when moving it toward the operator.

Perforations with intraabdominal burns also have occurred during attempts to coagulate bleeders—especially in the cornual regions.

Diversion of current can be destructive

Genital tract injuries have occurred as a result of current diversion. Vilos and colleagues¹⁶ reported 13 electrical burns during endometrial ablation, and mention many more anecdotal reports. The usual cause: electrode insulation failure, which allows current to jump to the outer sheath of the resectoscope.

To avoid this, inspect all electrodes thoroughly before surgery and use them only once.

Capacitative coupling also diverts current

Since the sheath-within-a-sheath design of the resectoscope resembles a capacitor, high-voltage current can jump to the outer sheath without direct contact from the electrode. When Munro¹⁷ bench-tested electrosurgical generators and electrodes with and without insulation defects, he found that capacitative coupling with intact electrodes occurred more frequently with high-voltage coagulation current than with lower-voltage cutting current.

One way to avoid these injuries is to activate the electrode intermittently, with short bursts, rather than rolling back and forth over an area with continuous current. Another strategy is placing a damp sponge in the posterior vagina extending out the introitus; this protects the mucosa and perineal skin—especially in obese patients.

How to avoid return-pad injuries

Keep the patient’s thigh completely dry; ensure that the pad is flat against the skin at application, with no bubbles or creases; and use only return electrode monitor (REM) dispersive pads.

Especially when using vaporization electrodes, avoid prolonged activation of the electrode at high power. To minimize risk of vaporization, use a second dispersive pad connected to the first via a “y” connector to further disperse current and heat at the return pad.

Also, limit the use of coagulation current and use a maximum generator setting of 60 to 80 W in the coagulation mode.

Take steps to avert gas embolism, but watch closely for signs

Initial reports of this potentially fatal complication came mostly from laser ablation procedures, but gas embolism can occur during all diagnostic and operative hysteroscopic procedures, especially the latter.

Sources of gas embolism: room air, carbon dioxide, carbon monoxide, and other gaseous products of vaporization or tissue combustion. The anesthesiologist is usually the first to identify the signs.

Signs of gas embolism. The surgeon should ask to be immediately alerted to any sudden fall in oxygen saturation, as well as to hypotension, hypercarbia, arrhythmias, tachypnea, or a “mill wheel” murmur. If any of these signs are detected and a gas embolism is suspected, stop the procedure and ventilate the patient with 100% oxygen.

Carbon dioxide is a soluble gas, so these emboli generally resolve rapidly. In contrast, room air emboli are more likely to be fatal.

Reduce the risk of air embolism by avoiding the Trendelenburg position and leaving the last dilator in the cervix until just before inserting the resectoscope.

Also limit repetitive removal and reinsertion of the resectoscope, as often occurs during myoma resection. By vaporizing rather than resecting myomas, it is possible to eliminate the need to continually remove fibroid chips. Preoperative GnRH agonists narrow venous sinuses and help prevent this complication.

Intracervical injection of dilute vasopressin prior to dilatation of the cervix creates vascular spasm and may help prevent gas from entering the circulation.

Complications from distention media

Excess absorption of distention media is one of the most frequent complications. Most surgeons use low-viscosity, sodium-free fluids for operative hysteroscopy, since fluids that contain electrolytes are incompatible with monopolar electrosurgical instruments. The use of 3% sorbitol, 1.5% glycine, or sorbitol-mannitol solutions can lead to dilutional hyponatremia and hypoosmolality.¹⁸ Although the vast majority of women quickly recover from these conditions, some cases of permanent morbidity and even death have been reported.¹⁹ The overall incidence of dilutional hyponatremia was 0.2% in 1993, according to the AAGL member survey.¹

 

Hyponatremia and hypoosmolality more likely in premenopausal women

These conditions may have catastrophic consequences if they are not recognized and corrected promptly. The brain swells as it attempts to become isoosmotic with the vascular system. If swelling exceeds 5%, the risk of severe neurological damage dramatically increases.

This is an important problem in premenopausal women, since estrogen and progesterone inhibit sodium-potassium adenosine triphosphatase (ATPase) activity in the brain. This sodium pump protects the brain against cerebral edema, which can cause herniation of the brain stem and death. Although men and postmenopausal women develop dilutional hyponatremia, they are less likely to suffer brain damage because the sodium pump is intact.

Taskin et al²⁰ conducted a randomized trial showing an increase in the sodium-potassium ATPase pump activity and decreased volume deficit during hysteroscopic surgery in patients pretreated with a GnRH agonist, compared with a control group. This increased pump activity in the brain and endometrium may decrease women’s susceptibility to hyponatremic complications and brain damage.

Vigilant monitoring of fluid intake/output during hysteroscopic surgery is necessary to prevent hyponatremic complications. Avoid the pitfall of erroneously attributing deficits to fluid “in the drapes” by using drapes with a fluid-collection pouch.

The standard of care is use of electronic inflow-outflow measuring systems. Manufacturers of hysteroscopic equipment offer highly accurate electronic fluid monitoring systems that measure the weight of the distending fluid infused and collected rather than relying on manual estimation of deficit. The latter method may be inaccurate since the volume of the supply bag can vary by as much as 10%.

Adjust intrauterine pressure to reduce the likelihood of intravasation. High intrauterine pressure may be desirable for visualization, but it greatly increases the risk of intravasation.

I adjust pressure and flow rates by opening or closing the inflow and outflow valves of the resectoscope until slight amounts of bleeding from resected tissue can be visualized. Since intrauterine pressure is extremely difficult to monitor accurately during operative hysteroscopy, this practice ensures that it remains below the patient’s mean arterial pressure, thus minimizing the risk of intravasation.

Use a dilute vasopressin injection to constrict blood vessels and decrease the chance of intravasation.

Vaporizing electrodes for myoma resection and ablation seal blood vessels and reduce fluid absorption.

Guidelines for distention media

To reduce the likelihood of these complications, I recommend that surgeons:

• Draw preoperative serum electrolytes for a baseline in all patients.
• Give all patients with myomas 2 monthly injections of depot leuprolide acetate (3.75 mg intramuscularly). Give patients without fibroids a single injection 4 to 6 weeks prior to the procedure.
• Place a fluid-collection drape or a larger, plastic Mayo stand cover with the bottom cut off under the patient’s buttocks so that fluid drains into a “kick” bucket. Also adjust the resectoscope’s outflow tubing to drain into the collection bag, which should be kept on constant suction to the flow-stat electronic fluid monitor.
• Continuously record inflow and outflow using the electronic monitor with the deficit alarm set to 500 mL.
• Keep distention fluid at room temperature and monitor the patient’s core temperature continuously. Significant fluid intravasation will lower the patient’s temperature, and this may be the first sign of fluid overload.
• Perform operative hysteroscopy under spinal or epidural anesthesia so the anesthesiologist can continually assess the patient’s sensorium. Confusion and irritability are early signs of dilutional hyponatremia.
• If the fluid deficit reaches 750 mL, immediately give 20 to 40 mg of intravenous furosemide and draw a serum sodium. Do not wait for the result of the sodium level before treatment, since a 5- to 20-minute delay can be catastrophic.
• Interrupt the procedure for 5 to 10 minutes to allow the uterus to contract and to seal off small blood vessels.
• Discontinue the procedure if the fluid deficit reaches 1,500 mL or if the serum sodium level is below 125 mEq/L.

I do not limit the duration of resectoscopic procedures as long as fluid deficits are below 750 mL, as measured by electronic fluid monitor. I also ensure that the operating room staff is well educated in the use of the monitor and able to troubleshoot intraoperatively.

If the machine fails during the procedure, reset it with the alarm limit lowered to reflect the deficit recorded before failure.

Monitor the color of the outflow fluid. Excessive blood loss counted as part of the outflow can occasionally mask a distention fluid deficit.

Choosing a distention medium

There is no ideal distention medium for monopolar operative hysteroscopy. Several authors have suggested that 5% mannitol is advantageous since it is isosmolar and acts as an osmotic diuretic. However, it does not prevent hyponatremia. The main disadvantage of 5% mannitol is its high cost and limited availability in 3-L bags or 4-L bottles.

 

The use of bipolar devices in normal saline prevents dilutional hyponatremia, but fluid deficits must still be monitored electronically so they do not exceed 2,000 mL. The false sense of security that may occur when normal saline is used for distention may lead to inaction when a large deficit occurs. This can lead to pulmonary edema and death.

Postoperative and late complications

These include infection, endometrial cancer, iatrogenic adenomyosis, hematometria, post-ablation tubal ligation syndrome, and pregnancy.

Infection rate is 0.3% to 2%

Infection is relatively rare following endometrial ablation, with a rate of 0.3% to 0.5% reported in most series. Endometritis, parametritis, and pyometra are more common following resection of submucous myomas, with rates as high as 2% reported.

Infection is more likely after prolonged procedures, especially when the hysteroscope is repeatedly inserted and removed. It also is more likely if the patient has a history of pelvic inflammatory disease. I generally administer prophylactic antibiotics (1 dose of intravenous ceftizoxime, 1 g, approximately 30 to 60 minutes prior to surgery).

I also insert a laminaria tent the evening prior to surgery. Patients with a history of pelvic inflammatory disease are discharged on doxycycline (100 mg twice daily for 7 days).

Be alert for endometrial cancer

This malignancy has been diagnosed at the time of endometrial ablation and reported in patients who have undergone prior endometrial ablations or fibroid resections. Thus, endometrial sampling should be part of the workup of abnormal uterine bleeding before the patient is scheduled for operative hysteroscopy. In women at high risk for endometrial cancer, perform office diagnostic hysteroscopy, with directed biopsy of any suspicious areas.

When viable endometrial glands are “buried” during ablation, or synechiae develop, preventing the egress of blood, there is a chance that diagnosis of endometrial cancer will be delayed. However, this theoretical fear has not been proven clinically.

Patients whose abnormal bleeding recurs after ablation should undergo sampling and office hysteroscopy, just as if they had not undergone a previous ablation. Theoretically, women who undergo endomyometrial resection or vaporization should have a lower incidence of endometrial cancer, since the tissue most susceptible to malignancy is removed. This has not yet been proven scientifically.

In their comprehensive review of late complications of operative hysteroscopy, Cooper and Brady²¹ suggest that patients at high risk for endometrial cancer who present with abnormal uterine bleeding not controlled by hormones might be better served by hysterectomy. Unfortunately, these patients tend to be high-risk surgical candidates.

If atypia is present, do not perform endometrial ablation or resection. I do perform ablation and resection in patients with complex hyperplasia without atypia if it has been reversed with progestin and does not recur for at least 6 months without progestin therapy. These patients undergo office hysteroscopy and sampling of the endometrium before operative hysteroscopy is scheduled.

Iatrogenic adenomyosis

Two theories suggest this is a late complication of operative hysteroscopy. According to the first, when the endometrium is incompletely resected, scarring over this tissue causes the viable glands to grow into the myometrium. The other theory suggests that viable endometrial debris is transported into the myometrium by vessels opened at resection.

I have found that using the vaporization electrode followed by application of a rollerball over the surface of the cavity most effectively reaches maximal tissue depth and, theoretically, prevents adenomyosis. Since most adenomyosis occurs on the posterior wall, I take a strip from this area for pathologic analysis to determine whether adenomyosis preceded the ablation or developed subsequent to it.

Hematometria

This can occur following operative hysteroscopy if viable glands are left in the fundal or cornual region and synechiae develop in the lower segment, preventing egress of blood. It also can occur if the upper endocervix is ablated and subsequently scars, causing stenosis. To avoid this, ablate only to the level of the internal os.

Diagnosis and treatment. Hematometria can be diagnosed easily by ultrasound and treated with office hysteroscopy using a narrow-diameter, rigid, continuous-flow hysteroscope with an operating channel to pass small instruments.

Post-ablation tubal ligation syndrome

This is cornual hematometria that develops when viable endometrial cells are left in the cornua when the cavity also contains synechiae, causing cyclic bleeding. Since there is no egress from the cervix or tubes, blood gradually builds up, leading to hematosalpinx and pain. One way to avoid this is to ensure complete ablation of the cornual endometrium.

Prevention. Some experts recommend that the small rollerball electrode be placed in the cornua, with slightly reduced intrauterine pressure, to allow the cornual endometrium to collapse around the rollerball. A short burst of current is then applied to ablate the tissue.

 

This complication is less likely after hydrothermablation, since the free-flowing saline ablates the cornua completely. After more than 50 hydrothermablations performed in patients with prior tubal sterilizations, I have not seen any cases of post-ablation tubal ligation syndrome.

Treatment consists of bilateral salpingectomy or tubal fulguration at the cornual region and repeat ablation or resection of viable endometrial tissue. Another option is hysterectomy.

Post-ablation pregnancy can be very complicated

Pregnancy after endometrial ablation occurs at a rate of 0.2% to 1.6%. Counsel patients that this procedure does not prevent pregnancy and that contraception is vital. Uterine rupture after fibroid resection has been reported.

In a review of 37 post-ablation pregnancies, only 11 of 17 women who chose not to terminate carried the gestation beyond 28 weeks. In addition, there was a high incidence of intrauterine growth restriction, prematurity, and placenta accreta.²²

Hysteroscopic tubal sterilization with the Essure system (Conceptus, San Carlos, Calif), or laparoscopic tubal fulguration performed at the time of ablation averts these complications.

Complications of global ablation

Global ablation technologies were developed to enable gynecologists with limited operative hysteroscopy skills to perform endometrial ablation and to make ablation safer for the patient. These technologies completely eliminate the risk of distention-media complications, but widespread use has resulted in other complications that have been reported in the literature to only a limited extent.

Most published articles on global endometrial ablation are from the original US Food and Drug Administration (FDA) trials, in which the complication rates were extraordinarily low. Widespread commercial use of these technologies since FDA approval, especially by practitioners with limited skills, has increased complications.

Do not override safety systems

Complications are more frequent when devices are misused or safety systems overridden. And, fear of litigation makes physicians unwilling to report complications.

In the FDA Manufacturer and User Facility Device Experience (MAUDE) database (www.fda.gov.cdrh/maude.html), complications include bowel burns after unrecognized perforation, and bowel burns associated with electrosurgical, microwave energy, or heat transferred through intact myometrium.²³ Vaginal burns, uterine necrosis myometritis requiring hysterectomy, and death from unrecognized bowel burn also have been reported.

Most global procedures are performed blindly, and some doctors fail to perform diagnostic hysteroscopy before and after surgery, which I feel is mandatory with any endometrial ablation. Hydrothermablation is the only global technique that has the advantage of direct observation. In more than 150 procedures done in my office under local anesthesia, the only complications were 2 false passages. Both were promptly identified during diagnostic hysteroscopy, and the surgery was rescheduled 2 to 3 months later.

The author has served on the speakers’ bureau for Boston Scientific.

WHAT WENT WRONG?

A 44-year-old woman undergoing resection of a submucous myoma from the left cornual region has persistent bleeding at the resection site. The surgeon continues coagulation at the bleeding site, using a rollerball electrode in an attempt to achieve hemostasis, but perforates the uterus. Immediate laparoscopy to identify collateral injury reveals some thermal damage on the posterior leaf of the broad ligament, but no bowel injury. After 24 hours of observation, she is afebrile without leukocytosis. She is discharged with explicit instructions to return if she has symptoms suggesting bowel injury. She returns in 72 hours, with abdominal pain and low-grade fever. CT reveals extravasation of contrast from the left ureter in the pelvis. Immediate laparotomy finds perforation of the left ureter secondary to a thermal injury. She undergoes ureteroneocystotomy and recovers.

This case illustrates one of the most common complications of operative hysteroscopy: uterine perforation with collateral injury. Both could have been avoided if the Ob/Gyn had stopped the procedure when bleeding first occurred, removed the instruments, and allowed the uterus to contract spontaneously.

This is just one of the strategies that can reduce the risks of hysteroscopic surgery. Numerous reports confirm that operative hysteroscopy is safe and effective, but as more gynecologists perform an increasing number of procedures, we must be aware of potential complications and do our best to minimize risk to our patients.

Complications cannot be completely avoided, and may occur when a procedure is done correctly by experienced doctors. They are far more likely if techniques or equipment are used improperly. This article describes ways to minimize risk.

When the American Association of Gynecologic Laparoscopists (AAGL) surveyed its members in 1993, it found a complication rate of 2% for operative hysteroscopy.¹ The rate of major complications—perforation, hemorrhage, fluid overload, and bowel or urinary tract injury—was less than 1%. A prospective multicenter trial² of 13,600 procedures in the Netherlands found a higher complication rate for operative (0.95%) than for diagnostic hysteroscopy (0.13%).

Preoperative precautions

We can reduce the risk of complications if contraindications are not ignored, equipment is thoroughly inspected and understood, and the surgeon goes through a mental checklist and plans each procedure. A “time out”before the operation begins, when every member of the team is briefed, is also valuable in preventing errors.

A hands-on course necessary before undertaking advanced resectoscopic surgery, to become familiar with equipment and techniques, followed by proctoring by a surgeon credentialed for the procedure.

Contraindications

Ignoring contraindications to hysteroscopic surgery increases the risk of complications and is the single greatest factor leading to patient injury and physician liability.

Contraindications include:

• Unfamiliarity with equipment, instruments, or technique
• Lack of appropriate equipment or staff familiar with the equipment
• Acute pelvic inflammatory disease
• Pregnancy
• Genital tract malignancies
• Lack of informed consent
• Inability to dilate the cervix
• Inability to distend the uterus to obtain visualization
• Poor surgical candidates who may not tolerate fluid overload because of renal disease, or radiofrequency current when a cardiac pacemaker is present
• The patient desires and expects complete amenorrhea³

Mechanical or traumatic complications

These types of complications are among the most common. Other categories include preoperative complications (ie, improper patient selection and lack of informed consent), electrosurgical and gaseous, complications related to distention media, and postoperative complications (ie, infection and late sequelae).

 

Inability to insert the hysteroscope

This may be caused by a stenotic, nulliparous cervix; menopause; GnRH agonists; previous cone biopsy, laceration, or cryosurgery; or an acutely retroflexed or anteflexed uterus.

Acute flexion problems can be corrected using a long-bladed, open-sided Graves speculum deep in the anterior or posterior fornix. The speculum pushes the fundus to the midposition and facilitates dilation. Once the hysteroscope is inserted, remove the speculum.

Placing a tenaculum on the posterior lip of the cervix of an acutely retroflexed uterus will straighten the cervical canal when traction is applied.

Inserting a laminaria tent the evening before surgery helps dilate the cervix easily and atraumatically.⁴ However, the laminaria can sometimes create a false passage, leading to perforation.

Cervical ripening agents such as intravaginal or oral misoprostol (200 μg inserted vaginally or 400 μg orally 8 to 12 hours preoperatively) also can facilitate dilation.

Intracervical injection of vasopressin solution (4 IU in 100 cc sodium chloride, injected at the 4 and 8 o’clock positions) can reduce the force needed to dilate the cervix.⁵ Half-size dilators may help; they also reduce the risk of cervical laceration.

Laceration of the cervix

Although this is a minor complication, substantial bleeding sometimes occurs when the cervix is lacerated by the tenaculum. In these cases, suture the cervix.

Occasionally, a touch of cautery from the rollerball electrode at low power (20 to 30 W) can control the bleeding.

Silver nitrate sticks or ferric subsulfate (Monsel’s) paste are also effective on superficial lacerations.

Bleeding from lower uterus or cervical canal can obscure view

In some cases, bleeding is delayed, necessitating additional surgery. Intravasation of distention fluid also can occur at these lacerations. Coagulation with the electrode may be necessary when bleeding is heavy.

Check for collateral injury when uterine perforation occurs

Perforation is a well-documented risk of operative hysteroscopy and should be discussed with the patient when obtaining informed consent. In the AAGL survey,¹ the incidence of perforation was 14 per 1,000. It was even higher during transection of lateral and fundal adhesions: 2 to 3 per 100.⁶

Although perforation is more common with thermal energy sources, it may occur mechanically when scissors are used to transect a uterine septum, synechiae, or polyps.

When the cervix is stenotic or the uterus is acutely ante- or retroflexed, sounds and dilators can perforate the uterus.

Most perforations—even those involving large dilators—usually do not require treatment, although further assessment may be necessary to rule out bowel injury. Most perforations occur in the fundal region or posterior lower segment.

A false passage can be created when entering the uterus. Occasionally the surgeon may be fooled into thinking the hysteroscope is in the uterine cavity, since the false passage distends (FIGURE 1). If muscle fibers are visible and the tubal ostea are not, assume the passage is false. Slowly remove the hysteroscope and identify the true cavity for confirmation. Discontinue the procedure—even if no perforation is detected—to prevent distention fluid from being absorbed into the circulation through the injury. Adequate distention is not possible at this time.

Delay repeat hysteroscopy for 2 to 3 months.

To avoid creating a false passage, dilate the cervix with slow, steady pressure and stop as soon as the internal os opens; do not attempt to push the dilator to the uterine fundus.

Often the external os opens, but the internal os cannot be dilated the extra 1 to 2 mm necessary to accommodate the 27-French resectoscope. Rather than exert more force and risk perforation or laceration, simply turn on the resectoscope’s inflow with the outflow shut off, and let the fluid pressure dilate the cervix.

Always insert the hysteroscope or resectoscope under direct vision rather than use an obturator. Keep the “dark circle” in the center of the field and slowly advance the hysteroscope toward it until the cavity is reached.

Avulsion of the myometrium sometimes occurs during removal of incompletely resected myomas (FIGURE 2). Keep the myoma grasper away from the fundus when removing myoma segments, and avoid excessive traction on what may be a thin segment of myometrium. Injuries can occur when the grasper perforates the uterus and bowel is inadvertently grasped. Large injuries require laparoscopic repair.

Perforation is more likely in repeat procedures. In a report of 80 repeat endometrial ablations, Townsend and colleagues⁷ noted 8 perforations that prevented completion of the procedure. In a series of 75 repeat ablations compared with 800 primary ablations by the same surgeon, the rate of serious perioperative complications was significantly higher in the repeat ablation group (9.3% versus 2.0%).⁸

 

When perforation occurs during the use of thermal energy, laparoscopy is necessary to assess the organs overlying the site.⁹ During setup for laparoscopy, bring the hysteroscope near the area of perforation to inspect the bowel beyond the uterus. Since the pelvis fills quickly with distention fluid, the hysteroscope can even be placed through the perforation to yield an excellent view of the undersurfaces of the bowel immediately adjacent to the injured area (FIGURE 3). (Disconnect the electrosurgical cord before doing this!)

Thorough laparoscopic inspection of the bowel in the pelvis often reveals thermal injury, which appears as a whitish patch on the bowel serosa. To repair bowel injuries, bring the injured segment out through a minilaparotomy and excise the damage with a 2- to 3-cm border. A general surgeon should be called in to consult.

If no injury is apparent, discharge the patient but follow her closely, including daily white blood cell counts for 4 to 5 days. Instruct her to take her temperature twice daily and return to the hospital immediately if any signs of bowel perforation develop. Delayed perforation from thermal injury can occur as late as 2 weeks following surgery, and the patient should be apprised of this possibility.

FIGURE 1 Signs of a false passage

Myometrial fibers signal that a false passage has been created. Stop the procedure even if no perforation is detected, to prevent distention fluid from being absorbed into the circulation through the injury.

FIGURE 2 Risk of myomectomy: Myometrial avulsion

Small bowel visible within the uterine cavity after avulsion of uterine wall at the time of myomectomy

FIGURE 3 Use the hysteroscope to assess perforation site

Hysteroscopic view of perforation at the fundus. The small bowel is visible beyond the perforation at left.

Intraoperative bleeding is rare

Bleeding is unlikely unless vessels are lacerated or injured in the cervical canal or lower uterine segment during dilation or deep ablation or vaporization. Bleeding is more common when endomyometrial resection is performed with the wire loop electrode or during ablation or vaporization of fibroids. Bleeding sufficient to require intervention occurs at a rate of 0.5% to 1.9% in several reported series.

To achieve hemostasis via balloon tamponade, insert a Foley catheter with a 30-cc balloon into the uterine cavity, inject 15 to 20 mL (or more for a larger cavity) of fluid into the balloon, and observe the patient.¹⁰ If there is no bleeding in 1 hour, remove half the fluid. Remove the remainder of the fluid and the catheter over the next hour if no further bleeding occurs.

Alternative method: Pack the uterus. I prefer 1/2-inch–gauge packing that has been soaked in a dilute vasopressin solution (20 U [1 mL] in 60 mL normal saline).¹¹

The benefits of vasopressin. Before balloon tamponade or packing the uterus, I inject very dilute vasopressin (4 U [0.2 mL] in 60 mL normal saline) directly into the cervix 2 cm deep, at the 4 and 8 o’clock positions. Phillips¹² demonstrated a marked decrease in blood loss during resectoscopic surgery using this approach. I also do this routinely prior to operative hysteroscopy, since the vasopressin-induced vasoconstriction reduces intravasation of distention media.

A vaporizing electrode may prevent significant blood loss during myoma resection by sealing blood vessels as the tissue is vaporized.¹³ All major manufacturers of hysteroscopic equipment produce these electrodes.

In my series of 44 endometrial ablations and hysteroscopic myomectomies performed with the vaporizing electrode, blood loss was “minimal” or less than 50 mL in 29 cases. The maximum blood loss was 300 mL in a patient with a 4-cm submucous myoma who was managed emergently for hemorrhage.¹⁴ In another case, during resection of a 5-cm submucous myoma, I encountered significant bleeding from large vessels at the base of the myoma, which required intrauterine tamponade with a vasopressin-soaked pack.

 

Preoperative measures may decrease vascularity. In their analysis of 16 randomized and nonrandomized controlled trials published in the English literature between 1990 and 1996, Parazzini and colleagues¹⁵ found that preoperative danazol or GnRH agonists decreased the thickness and vascularity of the endometrium and shrank myomata, resulting in shorter operating times, less blood loss, and less intravasation of distending fluid.

Electrosurgical and gaseous complications

Most electrosurgical complications involve activation of an electrode at the time of perforation, or current diversion to the outer sheath.

Thermal injuries also can be caused by overheating of the return pad or use of a weighted speculum that has not fully cooled after removal from the autoclave. The latter can be avoided by immersing the entire speculum in cool saline for at least 1 to 2 minutes prior to inserting it into the vagina. The blade cools much faster than the weighted ball, so be sure to check both to prevent a perineal or buttock burn.

Perforation with an active electrode

This usually occurs when current is applied as the electrode is extended or the resectoscope is moved toward the fundus. It can be avoided if the electrode is activated only when moving it toward the operator.

Perforations with intraabdominal burns also have occurred during attempts to coagulate bleeders—especially in the cornual regions.

Diversion of current can be destructive

Genital tract injuries have occurred as a result of current diversion. Vilos and colleagues¹⁶ reported 13 electrical burns during endometrial ablation, and mention many more anecdotal reports. The usual cause: electrode insulation failure, which allows current to jump to the outer sheath of the resectoscope.

To avoid this, inspect all electrodes thoroughly before surgery and use them only once.

Capacitative coupling also diverts current

Since the sheath-within-a-sheath design of the resectoscope resembles a capacitor, high-voltage current can jump to the outer sheath without direct contact from the electrode. When Munro¹⁷ bench-tested electrosurgical generators and electrodes with and without insulation defects, he found that capacitative coupling with intact electrodes occurred more frequently with high-voltage coagulation current than with lower-voltage cutting current.

One way to avoid these injuries is to activate the electrode intermittently, with short bursts, rather than rolling back and forth over an area with continuous current. Another strategy is placing a damp sponge in the posterior vagina extending out the introitus; this protects the mucosa and perineal skin—especially in obese patients.

How to avoid return-pad injuries

Keep the patient’s thigh completely dry; ensure that the pad is flat against the skin at application, with no bubbles or creases; and use only return electrode monitor (REM) dispersive pads.

Especially when using vaporization electrodes, avoid prolonged activation of the electrode at high power. To minimize risk of vaporization, use a second dispersive pad connected to the first via a “y” connector to further disperse current and heat at the return pad.

Also, limit the use of coagulation current and use a maximum generator setting of 60 to 80 W in the coagulation mode.

Take steps to avert gas embolism, but watch closely for signs

Initial reports of this potentially fatal complication came mostly from laser ablation procedures, but gas embolism can occur during all diagnostic and operative hysteroscopic procedures, especially the latter.

Sources of gas embolism: room air, carbon dioxide, carbon monoxide, and other gaseous products of vaporization or tissue combustion. The anesthesiologist is usually the first to identify the signs.

Signs of gas embolism. The surgeon should ask to be immediately alerted to any sudden fall in oxygen saturation, as well as to hypotension, hypercarbia, arrhythmias, tachypnea, or a “mill wheel” murmur. If any of these signs are detected and a gas embolism is suspected, stop the procedure and ventilate the patient with 100% oxygen.

Carbon dioxide is a soluble gas, so these emboli generally resolve rapidly. In contrast, room air emboli are more likely to be fatal.

Reduce the risk of air embolism by avoiding the Trendelenburg position and leaving the last dilator in the cervix until just before inserting the resectoscope.

Also limit repetitive removal and reinsertion of the resectoscope, as often occurs during myoma resection. By vaporizing rather than resecting myomas, it is possible to eliminate the need to continually remove fibroid chips. Preoperative GnRH agonists narrow venous sinuses and help prevent this complication.

Intracervical injection of dilute vasopressin prior to dilatation of the cervix creates vascular spasm and may help prevent gas from entering the circulation.

Complications from distention media

Excess absorption of distention media is one of the most frequent complications. Most surgeons use low-viscosity, sodium-free fluids for operative hysteroscopy, since fluids that contain electrolytes are incompatible with monopolar electrosurgical instruments. The use of 3% sorbitol, 1.5% glycine, or sorbitol-mannitol solutions can lead to dilutional hyponatremia and hypoosmolality.¹⁸ Although the vast majority of women quickly recover from these conditions, some cases of permanent morbidity and even death have been reported.¹⁹ The overall incidence of dilutional hyponatremia was 0.2% in 1993, according to the AAGL member survey.¹

 

Hyponatremia and hypoosmolality more likely in premenopausal women

These conditions may have catastrophic consequences if they are not recognized and corrected promptly. The brain swells as it attempts to become isoosmotic with the vascular system. If swelling exceeds 5%, the risk of severe neurological damage dramatically increases.

This is an important problem in premenopausal women, since estrogen and progesterone inhibit sodium-potassium adenosine triphosphatase (ATPase) activity in the brain. This sodium pump protects the brain against cerebral edema, which can cause herniation of the brain stem and death. Although men and postmenopausal women develop dilutional hyponatremia, they are less likely to suffer brain damage because the sodium pump is intact.

Taskin et al²⁰ conducted a randomized trial showing an increase in the sodium-potassium ATPase pump activity and decreased volume deficit during hysteroscopic surgery in patients pretreated with a GnRH agonist, compared with a control group. This increased pump activity in the brain and endometrium may decrease women’s susceptibility to hyponatremic complications and brain damage.

Vigilant monitoring of fluid intake/output during hysteroscopic surgery is necessary to prevent hyponatremic complications. Avoid the pitfall of erroneously attributing deficits to fluid “in the drapes” by using drapes with a fluid-collection pouch.

The standard of care is use of electronic inflow-outflow measuring systems. Manufacturers of hysteroscopic equipment offer highly accurate electronic fluid monitoring systems that measure the weight of the distending fluid infused and collected rather than relying on manual estimation of deficit. The latter method may be inaccurate since the volume of the supply bag can vary by as much as 10%.

Adjust intrauterine pressure to reduce the likelihood of intravasation. High intrauterine pressure may be desirable for visualization, but it greatly increases the risk of intravasation.

I adjust pressure and flow rates by opening or closing the inflow and outflow valves of the resectoscope until slight amounts of bleeding from resected tissue can be visualized. Since intrauterine pressure is extremely difficult to monitor accurately during operative hysteroscopy, this practice ensures that it remains below the patient’s mean arterial pressure, thus minimizing the risk of intravasation.

Use a dilute vasopressin injection to constrict blood vessels and decrease the chance of intravasation.

Vaporizing electrodes for myoma resection and ablation seal blood vessels and reduce fluid absorption.

Guidelines for distention media

To reduce the likelihood of these complications, I recommend that surgeons:

• Draw preoperative serum electrolytes for a baseline in all patients.
• Give all patients with myomas 2 monthly injections of depot leuprolide acetate (3.75 mg intramuscularly). Give patients without fibroids a single injection 4 to 6 weeks prior to the procedure.
• Place a fluid-collection drape or a larger, plastic Mayo stand cover with the bottom cut off under the patient’s buttocks so that fluid drains into a “kick” bucket. Also adjust the resectoscope’s outflow tubing to drain into the collection bag, which should be kept on constant suction to the flow-stat electronic fluid monitor.
• Continuously record inflow and outflow using the electronic monitor with the deficit alarm set to 500 mL.
• Keep distention fluid at room temperature and monitor the patient’s core temperature continuously. Significant fluid intravasation will lower the patient’s temperature, and this may be the first sign of fluid overload.
• Perform operative hysteroscopy under spinal or epidural anesthesia so the anesthesiologist can continually assess the patient’s sensorium. Confusion and irritability are early signs of dilutional hyponatremia.
• If the fluid deficit reaches 750 mL, immediately give 20 to 40 mg of intravenous furosemide and draw a serum sodium. Do not wait for the result of the sodium level before treatment, since a 5- to 20-minute delay can be catastrophic.
• Interrupt the procedure for 5 to 10 minutes to allow the uterus to contract and to seal off small blood vessels.
• Discontinue the procedure if the fluid deficit reaches 1,500 mL or if the serum sodium level is below 125 mEq/L.

I do not limit the duration of resectoscopic procedures as long as fluid deficits are below 750 mL, as measured by electronic fluid monitor. I also ensure that the operating room staff is well educated in the use of the monitor and able to troubleshoot intraoperatively.

If the machine fails during the procedure, reset it with the alarm limit lowered to reflect the deficit recorded before failure.

Monitor the color of the outflow fluid. Excessive blood loss counted as part of the outflow can occasionally mask a distention fluid deficit.

Choosing a distention medium

There is no ideal distention medium for monopolar operative hysteroscopy. Several authors have suggested that 5% mannitol is advantageous since it is isosmolar and acts as an osmotic diuretic. However, it does not prevent hyponatremia. The main disadvantage of 5% mannitol is its high cost and limited availability in 3-L bags or 4-L bottles.

 

The use of bipolar devices in normal saline prevents dilutional hyponatremia, but fluid deficits must still be monitored electronically so they do not exceed 2,000 mL. The false sense of security that may occur when normal saline is used for distention may lead to inaction when a large deficit occurs. This can lead to pulmonary edema and death.

Postoperative and late complications

These include infection, endometrial cancer, iatrogenic adenomyosis, hematometria, post-ablation tubal ligation syndrome, and pregnancy.

Infection rate is 0.3% to 2%

Infection is relatively rare following endometrial ablation, with a rate of 0.3% to 0.5% reported in most series. Endometritis, parametritis, and pyometra are more common following resection of submucous myomas, with rates as high as 2% reported.

Infection is more likely after prolonged procedures, especially when the hysteroscope is repeatedly inserted and removed. It also is more likely if the patient has a history of pelvic inflammatory disease. I generally administer prophylactic antibiotics (1 dose of intravenous ceftizoxime, 1 g, approximately 30 to 60 minutes prior to surgery).

I also insert a laminaria tent the evening prior to surgery. Patients with a history of pelvic inflammatory disease are discharged on doxycycline (100 mg twice daily for 7 days).

Be alert for endometrial cancer

This malignancy has been diagnosed at the time of endometrial ablation and reported in patients who have undergone prior endometrial ablations or fibroid resections. Thus, endometrial sampling should be part of the workup of abnormal uterine bleeding before the patient is scheduled for operative hysteroscopy. In women at high risk for endometrial cancer, perform office diagnostic hysteroscopy, with directed biopsy of any suspicious areas.

When viable endometrial glands are “buried” during ablation, or synechiae develop, preventing the egress of blood, there is a chance that diagnosis of endometrial cancer will be delayed. However, this theoretical fear has not been proven clinically.

Patients whose abnormal bleeding recurs after ablation should undergo sampling and office hysteroscopy, just as if they had not undergone a previous ablation. Theoretically, women who undergo endomyometrial resection or vaporization should have a lower incidence of endometrial cancer, since the tissue most susceptible to malignancy is removed. This has not yet been proven scientifically.

In their comprehensive review of late complications of operative hysteroscopy, Cooper and Brady²¹ suggest that patients at high risk for endometrial cancer who present with abnormal uterine bleeding not controlled by hormones might be better served by hysterectomy. Unfortunately, these patients tend to be high-risk surgical candidates.

If atypia is present, do not perform endometrial ablation or resection. I do perform ablation and resection in patients with complex hyperplasia without atypia if it has been reversed with progestin and does not recur for at least 6 months without progestin therapy. These patients undergo office hysteroscopy and sampling of the endometrium before operative hysteroscopy is scheduled.

Iatrogenic adenomyosis

Two theories suggest this is a late complication of operative hysteroscopy. According to the first, when the endometrium is incompletely resected, scarring over this tissue causes the viable glands to grow into the myometrium. The other theory suggests that viable endometrial debris is transported into the myometrium by vessels opened at resection.

I have found that using the vaporization electrode followed by application of a rollerball over the surface of the cavity most effectively reaches maximal tissue depth and, theoretically, prevents adenomyosis. Since most adenomyosis occurs on the posterior wall, I take a strip from this area for pathologic analysis to determine whether adenomyosis preceded the ablation or developed subsequent to it.

Hematometria

This can occur following operative hysteroscopy if viable glands are left in the fundal or cornual region and synechiae develop in the lower segment, preventing egress of blood. It also can occur if the upper endocervix is ablated and subsequently scars, causing stenosis. To avoid this, ablate only to the level of the internal os.

Diagnosis and treatment. Hematometria can be diagnosed easily by ultrasound and treated with office hysteroscopy using a narrow-diameter, rigid, continuous-flow hysteroscope with an operating channel to pass small instruments.

Post-ablation tubal ligation syndrome

This is cornual hematometria that develops when viable endometrial cells are left in the cornua when the cavity also contains synechiae, causing cyclic bleeding. Since there is no egress from the cervix or tubes, blood gradually builds up, leading to hematosalpinx and pain. One way to avoid this is to ensure complete ablation of the cornual endometrium.

Prevention. Some experts recommend that the small rollerball electrode be placed in the cornua, with slightly reduced intrauterine pressure, to allow the cornual endometrium to collapse around the rollerball. A short burst of current is then applied to ablate the tissue.

 

This complication is less likely after hydrothermablation, since the free-flowing saline ablates the cornua completely. After more than 50 hydrothermablations performed in patients with prior tubal sterilizations, I have not seen any cases of post-ablation tubal ligation syndrome.

Treatment consists of bilateral salpingectomy or tubal fulguration at the cornual region and repeat ablation or resection of viable endometrial tissue. Another option is hysterectomy.

Post-ablation pregnancy can be very complicated

Pregnancy after endometrial ablation occurs at a rate of 0.2% to 1.6%. Counsel patients that this procedure does not prevent pregnancy and that contraception is vital. Uterine rupture after fibroid resection has been reported.

In a review of 37 post-ablation pregnancies, only 11 of 17 women who chose not to terminate carried the gestation beyond 28 weeks. In addition, there was a high incidence of intrauterine growth restriction, prematurity, and placenta accreta.²²

Hysteroscopic tubal sterilization with the Essure system (Conceptus, San Carlos, Calif), or laparoscopic tubal fulguration performed at the time of ablation averts these complications.

Complications of global ablation

Global ablation technologies were developed to enable gynecologists with limited operative hysteroscopy skills to perform endometrial ablation and to make ablation safer for the patient. These technologies completely eliminate the risk of distention-media complications, but widespread use has resulted in other complications that have been reported in the literature to only a limited extent.

Most published articles on global endometrial ablation are from the original US Food and Drug Administration (FDA) trials, in which the complication rates were extraordinarily low. Widespread commercial use of these technologies since FDA approval, especially by practitioners with limited skills, has increased complications.

Do not override safety systems

Complications are more frequent when devices are misused or safety systems overridden. And, fear of litigation makes physicians unwilling to report complications.

In the FDA Manufacturer and User Facility Device Experience (MAUDE) database (www.fda.gov.cdrh/maude.html), complications include bowel burns after unrecognized perforation, and bowel burns associated with electrosurgical, microwave energy, or heat transferred through intact myometrium.²³ Vaginal burns, uterine necrosis myometritis requiring hysterectomy, and death from unrecognized bowel burn also have been reported.

Most global procedures are performed blindly, and some doctors fail to perform diagnostic hysteroscopy before and after surgery, which I feel is mandatory with any endometrial ablation. Hydrothermablation is the only global technique that has the advantage of direct observation. In more than 150 procedures done in my office under local anesthesia, the only complications were 2 false passages. Both were promptly identified during diagnostic hysteroscopy, and the surgery was rescheduled 2 to 3 months later.

The author has served on the speakers’ bureau for Boston Scientific.

WHAT WENT WRONG?

A 44-year-old woman undergoing resection of a submucous myoma from the left cornual region has persistent bleeding at the resection site. The surgeon continues coagulation at the bleeding site, using a rollerball electrode in an attempt to achieve hemostasis, but perforates the uterus. Immediate laparoscopy to identify collateral injury reveals some thermal damage on the posterior leaf of the broad ligament, but no bowel injury. After 24 hours of observation, she is afebrile without leukocytosis. She is discharged with explicit instructions to return if she has symptoms suggesting bowel injury. She returns in 72 hours, with abdominal pain and low-grade fever. CT reveals extravasation of contrast from the left ureter in the pelvis. Immediate laparotomy finds perforation of the left ureter secondary to a thermal injury. She undergoes ureteroneocystotomy and recovers.

This case illustrates one of the most common complications of operative hysteroscopy: uterine perforation with collateral injury. Both could have been avoided if the Ob/Gyn had stopped the procedure when bleeding first occurred, removed the instruments, and allowed the uterus to contract spontaneously.

This is just one of the strategies that can reduce the risks of hysteroscopic surgery. Numerous reports confirm that operative hysteroscopy is safe and effective, but as more gynecologists perform an increasing number of procedures, we must be aware of potential complications and do our best to minimize risk to our patients.

Complications cannot be completely avoided, and may occur when a procedure is done correctly by experienced doctors. They are far more likely if techniques or equipment are used improperly. This article describes ways to minimize risk.

When the American Association of Gynecologic Laparoscopists (AAGL) surveyed its members in 1993, it found a complication rate of 2% for operative hysteroscopy.¹ The rate of major complications—perforation, hemorrhage, fluid overload, and bowel or urinary tract injury—was less than 1%. A prospective multicenter trial² of 13,600 procedures in the Netherlands found a higher complication rate for operative (0.95%) than for diagnostic hysteroscopy (0.13%).

Preoperative precautions

We can reduce the risk of complications if contraindications are not ignored, equipment is thoroughly inspected and understood, and the surgeon goes through a mental checklist and plans each procedure. A “time out”before the operation begins, when every member of the team is briefed, is also valuable in preventing errors.

A hands-on course necessary before undertaking advanced resectoscopic surgery, to become familiar with equipment and techniques, followed by proctoring by a surgeon credentialed for the procedure.

Contraindications

Ignoring contraindications to hysteroscopic surgery increases the risk of complications and is the single greatest factor leading to patient injury and physician liability.

Contraindications include:

• Unfamiliarity with equipment, instruments, or technique
• Lack of appropriate equipment or staff familiar with the equipment
• Acute pelvic inflammatory disease
• Pregnancy
• Genital tract malignancies
• Lack of informed consent
• Inability to dilate the cervix
• Inability to distend the uterus to obtain visualization
• Poor surgical candidates who may not tolerate fluid overload because of renal disease, or radiofrequency current when a cardiac pacemaker is present
• The patient desires and expects complete amenorrhea³

Mechanical or traumatic complications

These types of complications are among the most common. Other categories include preoperative complications (ie, improper patient selection and lack of informed consent), electrosurgical and gaseous, complications related to distention media, and postoperative complications (ie, infection and late sequelae).

 

Inability to insert the hysteroscope

This may be caused by a stenotic, nulliparous cervix; menopause; GnRH agonists; previous cone biopsy, laceration, or cryosurgery; or an acutely retroflexed or anteflexed uterus.

Acute flexion problems can be corrected using a long-bladed, open-sided Graves speculum deep in the anterior or posterior fornix. The speculum pushes the fundus to the midposition and facilitates dilation. Once the hysteroscope is inserted, remove the speculum.

Placing a tenaculum on the posterior lip of the cervix of an acutely retroflexed uterus will straighten the cervical canal when traction is applied.

Inserting a laminaria tent the evening before surgery helps dilate the cervix easily and atraumatically.⁴ However, the laminaria can sometimes create a false passage, leading to perforation.

Cervical ripening agents such as intravaginal or oral misoprostol (200 μg inserted vaginally or 400 μg orally 8 to 12 hours preoperatively) also can facilitate dilation.

Intracervical injection of vasopressin solution (4 IU in 100 cc sodium chloride, injected at the 4 and 8 o’clock positions) can reduce the force needed to dilate the cervix.⁵ Half-size dilators may help; they also reduce the risk of cervical laceration.

Laceration of the cervix

Although this is a minor complication, substantial bleeding sometimes occurs when the cervix is lacerated by the tenaculum. In these cases, suture the cervix.

Occasionally, a touch of cautery from the rollerball electrode at low power (20 to 30 W) can control the bleeding.

Silver nitrate sticks or ferric subsulfate (Monsel’s) paste are also effective on superficial lacerations.

Bleeding from lower uterus or cervical canal can obscure view

In some cases, bleeding is delayed, necessitating additional surgery. Intravasation of distention fluid also can occur at these lacerations. Coagulation with the electrode may be necessary when bleeding is heavy.

Check for collateral injury when uterine perforation occurs

Perforation is a well-documented risk of operative hysteroscopy and should be discussed with the patient when obtaining informed consent. In the AAGL survey,¹ the incidence of perforation was 14 per 1,000. It was even higher during transection of lateral and fundal adhesions: 2 to 3 per 100.⁶

Although perforation is more common with thermal energy sources, it may occur mechanically when scissors are used to transect a uterine septum, synechiae, or polyps.

When the cervix is stenotic or the uterus is acutely ante- or retroflexed, sounds and dilators can perforate the uterus.

Most perforations—even those involving large dilators—usually do not require treatment, although further assessment may be necessary to rule out bowel injury. Most perforations occur in the fundal region or posterior lower segment.

A false passage can be created when entering the uterus. Occasionally the surgeon may be fooled into thinking the hysteroscope is in the uterine cavity, since the false passage distends (FIGURE 1). If muscle fibers are visible and the tubal ostea are not, assume the passage is false. Slowly remove the hysteroscope and identify the true cavity for confirmation. Discontinue the procedure—even if no perforation is detected—to prevent distention fluid from being absorbed into the circulation through the injury. Adequate distention is not possible at this time.

Delay repeat hysteroscopy for 2 to 3 months.

To avoid creating a false passage, dilate the cervix with slow, steady pressure and stop as soon as the internal os opens; do not attempt to push the dilator to the uterine fundus.

Often the external os opens, but the internal os cannot be dilated the extra 1 to 2 mm necessary to accommodate the 27-French resectoscope. Rather than exert more force and risk perforation or laceration, simply turn on the resectoscope’s inflow with the outflow shut off, and let the fluid pressure dilate the cervix.

Always insert the hysteroscope or resectoscope under direct vision rather than use an obturator. Keep the “dark circle” in the center of the field and slowly advance the hysteroscope toward it until the cavity is reached.

Avulsion of the myometrium sometimes occurs during removal of incompletely resected myomas (FIGURE 2). Keep the myoma grasper away from the fundus when removing myoma segments, and avoid excessive traction on what may be a thin segment of myometrium. Injuries can occur when the grasper perforates the uterus and bowel is inadvertently grasped. Large injuries require laparoscopic repair.

Perforation is more likely in repeat procedures. In a report of 80 repeat endometrial ablations, Townsend and colleagues⁷ noted 8 perforations that prevented completion of the procedure. In a series of 75 repeat ablations compared with 800 primary ablations by the same surgeon, the rate of serious perioperative complications was significantly higher in the repeat ablation group (9.3% versus 2.0%).⁸

 

When perforation occurs during the use of thermal energy, laparoscopy is necessary to assess the organs overlying the site.⁹ During setup for laparoscopy, bring the hysteroscope near the area of perforation to inspect the bowel beyond the uterus. Since the pelvis fills quickly with distention fluid, the hysteroscope can even be placed through the perforation to yield an excellent view of the undersurfaces of the bowel immediately adjacent to the injured area (FIGURE 3). (Disconnect the electrosurgical cord before doing this!)

Thorough laparoscopic inspection of the bowel in the pelvis often reveals thermal injury, which appears as a whitish patch on the bowel serosa. To repair bowel injuries, bring the injured segment out through a minilaparotomy and excise the damage with a 2- to 3-cm border. A general surgeon should be called in to consult.

If no injury is apparent, discharge the patient but follow her closely, including daily white blood cell counts for 4 to 5 days. Instruct her to take her temperature twice daily and return to the hospital immediately if any signs of bowel perforation develop. Delayed perforation from thermal injury can occur as late as 2 weeks following surgery, and the patient should be apprised of this possibility.

FIGURE 1 Signs of a false passage

Myometrial fibers signal that a false passage has been created. Stop the procedure even if no perforation is detected, to prevent distention fluid from being absorbed into the circulation through the injury.

FIGURE 2 Risk of myomectomy: Myometrial avulsion

Small bowel visible within the uterine cavity after avulsion of uterine wall at the time of myomectomy

FIGURE 3 Use the hysteroscope to assess perforation site

Hysteroscopic view of perforation at the fundus. The small bowel is visible beyond the perforation at left.

Intraoperative bleeding is rare

Bleeding is unlikely unless vessels are lacerated or injured in the cervical canal or lower uterine segment during dilation or deep ablation or vaporization. Bleeding is more common when endomyometrial resection is performed with the wire loop electrode or during ablation or vaporization of fibroids. Bleeding sufficient to require intervention occurs at a rate of 0.5% to 1.9% in several reported series.

To achieve hemostasis via balloon tamponade, insert a Foley catheter with a 30-cc balloon into the uterine cavity, inject 15 to 20 mL (or more for a larger cavity) of fluid into the balloon, and observe the patient.¹⁰ If there is no bleeding in 1 hour, remove half the fluid. Remove the remainder of the fluid and the catheter over the next hour if no further bleeding occurs.

Alternative method: Pack the uterus. I prefer 1/2-inch–gauge packing that has been soaked in a dilute vasopressin solution (20 U [1 mL] in 60 mL normal saline).¹¹

The benefits of vasopressin. Before balloon tamponade or packing the uterus, I inject very dilute vasopressin (4 U [0.2 mL] in 60 mL normal saline) directly into the cervix 2 cm deep, at the 4 and 8 o’clock positions. Phillips¹² demonstrated a marked decrease in blood loss during resectoscopic surgery using this approach. I also do this routinely prior to operative hysteroscopy, since the vasopressin-induced vasoconstriction reduces intravasation of distention media.

A vaporizing electrode may prevent significant blood loss during myoma resection by sealing blood vessels as the tissue is vaporized.¹³ All major manufacturers of hysteroscopic equipment produce these electrodes.

In my series of 44 endometrial ablations and hysteroscopic myomectomies performed with the vaporizing electrode, blood loss was “minimal” or less than 50 mL in 29 cases. The maximum blood loss was 300 mL in a patient with a 4-cm submucous myoma who was managed emergently for hemorrhage.¹⁴ In another case, during resection of a 5-cm submucous myoma, I encountered significant bleeding from large vessels at the base of the myoma, which required intrauterine tamponade with a vasopressin-soaked pack.

 

Preoperative measures may decrease vascularity. In their analysis of 16 randomized and nonrandomized controlled trials published in the English literature between 1990 and 1996, Parazzini and colleagues¹⁵ found that preoperative danazol or GnRH agonists decreased the thickness and vascularity of the endometrium and shrank myomata, resulting in shorter operating times, less blood loss, and less intravasation of distending fluid.

Electrosurgical and gaseous complications

Most electrosurgical complications involve activation of an electrode at the time of perforation, or current diversion to the outer sheath.

Thermal injuries also can be caused by overheating of the return pad or use of a weighted speculum that has not fully cooled after removal from the autoclave. The latter can be avoided by immersing the entire speculum in cool saline for at least 1 to 2 minutes prior to inserting it into the vagina. The blade cools much faster than the weighted ball, so be sure to check both to prevent a perineal or buttock burn.

Perforation with an active electrode

This usually occurs when current is applied as the electrode is extended or the resectoscope is moved toward the fundus. It can be avoided if the electrode is activated only when moving it toward the operator.

Perforations with intraabdominal burns also have occurred during attempts to coagulate bleeders—especially in the cornual regions.

Diversion of current can be destructive

Genital tract injuries have occurred as a result of current diversion. Vilos and colleagues¹⁶ reported 13 electrical burns during endometrial ablation, and mention many more anecdotal reports. The usual cause: electrode insulation failure, which allows current to jump to the outer sheath of the resectoscope.

To avoid this, inspect all electrodes thoroughly before surgery and use them only once.

Capacitative coupling also diverts current

Since the sheath-within-a-sheath design of the resectoscope resembles a capacitor, high-voltage current can jump to the outer sheath without direct contact from the electrode. When Munro¹⁷ bench-tested electrosurgical generators and electrodes with and without insulation defects, he found that capacitative coupling with intact electrodes occurred more frequently with high-voltage coagulation current than with lower-voltage cutting current.

One way to avoid these injuries is to activate the electrode intermittently, with short bursts, rather than rolling back and forth over an area with continuous current. Another strategy is placing a damp sponge in the posterior vagina extending out the introitus; this protects the mucosa and perineal skin—especially in obese patients.

How to avoid return-pad injuries

Keep the patient’s thigh completely dry; ensure that the pad is flat against the skin at application, with no bubbles or creases; and use only return electrode monitor (REM) dispersive pads.

Especially when using vaporization electrodes, avoid prolonged activation of the electrode at high power. To minimize risk of vaporization, use a second dispersive pad connected to the first via a “y” connector to further disperse current and heat at the return pad.

Also, limit the use of coagulation current and use a maximum generator setting of 60 to 80 W in the coagulation mode.

Take steps to avert gas embolism, but watch closely for signs

Initial reports of this potentially fatal complication came mostly from laser ablation procedures, but gas embolism can occur during all diagnostic and operative hysteroscopic procedures, especially the latter.

Sources of gas embolism: room air, carbon dioxide, carbon monoxide, and other gaseous products of vaporization or tissue combustion. The anesthesiologist is usually the first to identify the signs.

Signs of gas embolism. The surgeon should ask to be immediately alerted to any sudden fall in oxygen saturation, as well as to hypotension, hypercarbia, arrhythmias, tachypnea, or a “mill wheel” murmur. If any of these signs are detected and a gas embolism is suspected, stop the procedure and ventilate the patient with 100% oxygen.

Carbon dioxide is a soluble gas, so these emboli generally resolve rapidly. In contrast, room air emboli are more likely to be fatal.

Reduce the risk of air embolism by avoiding the Trendelenburg position and leaving the last dilator in the cervix until just before inserting the resectoscope.

Also limit repetitive removal and reinsertion of the resectoscope, as often occurs during myoma resection. By vaporizing rather than resecting myomas, it is possible to eliminate the need to continually remove fibroid chips. Preoperative GnRH agonists narrow venous sinuses and help prevent this complication.

Intracervical injection of dilute vasopressin prior to dilatation of the cervix creates vascular spasm and may help prevent gas from entering the circulation.

Complications from distention media

Excess absorption of distention media is one of the most frequent complications. Most surgeons use low-viscosity, sodium-free fluids for operative hysteroscopy, since fluids that contain electrolytes are incompatible with monopolar electrosurgical instruments. The use of 3% sorbitol, 1.5% glycine, or sorbitol-mannitol solutions can lead to dilutional hyponatremia and hypoosmolality.¹⁸ Although the vast majority of women quickly recover from these conditions, some cases of permanent morbidity and even death have been reported.¹⁹ The overall incidence of dilutional hyponatremia was 0.2% in 1993, according to the AAGL member survey.¹

 

Hyponatremia and hypoosmolality more likely in premenopausal women

These conditions may have catastrophic consequences if they are not recognized and corrected promptly. The brain swells as it attempts to become isoosmotic with the vascular system. If swelling exceeds 5%, the risk of severe neurological damage dramatically increases.

This is an important problem in premenopausal women, since estrogen and progesterone inhibit sodium-potassium adenosine triphosphatase (ATPase) activity in the brain. This sodium pump protects the brain against cerebral edema, which can cause herniation of the brain stem and death. Although men and postmenopausal women develop dilutional hyponatremia, they are less likely to suffer brain damage because the sodium pump is intact.

Taskin et al²⁰ conducted a randomized trial showing an increase in the sodium-potassium ATPase pump activity and decreased volume deficit during hysteroscopic surgery in patients pretreated with a GnRH agonist, compared with a control group. This increased pump activity in the brain and endometrium may decrease women’s susceptibility to hyponatremic complications and brain damage.

Vigilant monitoring of fluid intake/output during hysteroscopic surgery is necessary to prevent hyponatremic complications. Avoid the pitfall of erroneously attributing deficits to fluid “in the drapes” by using drapes with a fluid-collection pouch.

The standard of care is use of electronic inflow-outflow measuring systems. Manufacturers of hysteroscopic equipment offer highly accurate electronic fluid monitoring systems that measure the weight of the distending fluid infused and collected rather than relying on manual estimation of deficit. The latter method may be inaccurate since the volume of the supply bag can vary by as much as 10%.

Adjust intrauterine pressure to reduce the likelihood of intravasation. High intrauterine pressure may be desirable for visualization, but it greatly increases the risk of intravasation.

I adjust pressure and flow rates by opening or closing the inflow and outflow valves of the resectoscope until slight amounts of bleeding from resected tissue can be visualized. Since intrauterine pressure is extremely difficult to monitor accurately during operative hysteroscopy, this practice ensures that it remains below the patient’s mean arterial pressure, thus minimizing the risk of intravasation.

Use a dilute vasopressin injection to constrict blood vessels and decrease the chance of intravasation.

Vaporizing electrodes for myoma resection and ablation seal blood vessels and reduce fluid absorption.

Guidelines for distention media

To reduce the likelihood of these complications, I recommend that surgeons:

• Draw preoperative serum electrolytes for a baseline in all patients.
• Give all patients with myomas 2 monthly injections of depot leuprolide acetate (3.75 mg intramuscularly). Give patients without fibroids a single injection 4 to 6 weeks prior to the procedure.
• Place a fluid-collection drape or a larger, plastic Mayo stand cover with the bottom cut off under the patient’s buttocks so that fluid drains into a “kick” bucket. Also adjust the resectoscope’s outflow tubing to drain into the collection bag, which should be kept on constant suction to the flow-stat electronic fluid monitor.
• Continuously record inflow and outflow using the electronic monitor with the deficit alarm set to 500 mL.
• Keep distention fluid at room temperature and monitor the patient’s core temperature continuously. Significant fluid intravasation will lower the patient’s temperature, and this may be the first sign of fluid overload.
• Perform operative hysteroscopy under spinal or epidural anesthesia so the anesthesiologist can continually assess the patient’s sensorium. Confusion and irritability are early signs of dilutional hyponatremia.
• If the fluid deficit reaches 750 mL, immediately give 20 to 40 mg of intravenous furosemide and draw a serum sodium. Do not wait for the result of the sodium level before treatment, since a 5- to 20-minute delay can be catastrophic.
• Interrupt the procedure for 5 to 10 minutes to allow the uterus to contract and to seal off small blood vessels.
• Discontinue the procedure if the fluid deficit reaches 1,500 mL or if the serum sodium level is below 125 mEq/L.

I do not limit the duration of resectoscopic procedures as long as fluid deficits are below 750 mL, as measured by electronic fluid monitor. I also ensure that the operating room staff is well educated in the use of the monitor and able to troubleshoot intraoperatively.

If the machine fails during the procedure, reset it with the alarm limit lowered to reflect the deficit recorded before failure.

Monitor the color of the outflow fluid. Excessive blood loss counted as part of the outflow can occasionally mask a distention fluid deficit.

Choosing a distention medium

There is no ideal distention medium for monopolar operative hysteroscopy. Several authors have suggested that 5% mannitol is advantageous since it is isosmolar and acts as an osmotic diuretic. However, it does not prevent hyponatremia. The main disadvantage of 5% mannitol is its high cost and limited availability in 3-L bags or 4-L bottles.

 

The use of bipolar devices in normal saline prevents dilutional hyponatremia, but fluid deficits must still be monitored electronically so they do not exceed 2,000 mL. The false sense of security that may occur when normal saline is used for distention may lead to inaction when a large deficit occurs. This can lead to pulmonary edema and death.

Postoperative and late complications

These include infection, endometrial cancer, iatrogenic adenomyosis, hematometria, post-ablation tubal ligation syndrome, and pregnancy.

Infection rate is 0.3% to 2%

Infection is relatively rare following endometrial ablation, with a rate of 0.3% to 0.5% reported in most series. Endometritis, parametritis, and pyometra are more common following resection of submucous myomas, with rates as high as 2% reported.

Infection is more likely after prolonged procedures, especially when the hysteroscope is repeatedly inserted and removed. It also is more likely if the patient has a history of pelvic inflammatory disease. I generally administer prophylactic antibiotics (1 dose of intravenous ceftizoxime, 1 g, approximately 30 to 60 minutes prior to surgery).

I also insert a laminaria tent the evening prior to surgery. Patients with a history of pelvic inflammatory disease are discharged on doxycycline (100 mg twice daily for 7 days).

Be alert for endometrial cancer

This malignancy has been diagnosed at the time of endometrial ablation and reported in patients who have undergone prior endometrial ablations or fibroid resections. Thus, endometrial sampling should be part of the workup of abnormal uterine bleeding before the patient is scheduled for operative hysteroscopy. In women at high risk for endometrial cancer, perform office diagnostic hysteroscopy, with directed biopsy of any suspicious areas.

When viable endometrial glands are “buried” during ablation, or synechiae develop, preventing the egress of blood, there is a chance that diagnosis of endometrial cancer will be delayed. However, this theoretical fear has not been proven clinically.

Patients whose abnormal bleeding recurs after ablation should undergo sampling and office hysteroscopy, just as if they had not undergone a previous ablation. Theoretically, women who undergo endomyometrial resection or vaporization should have a lower incidence of endometrial cancer, since the tissue most susceptible to malignancy is removed. This has not yet been proven scientifically.

In their comprehensive review of late complications of operative hysteroscopy, Cooper and Brady²¹ suggest that patients at high risk for endometrial cancer who present with abnormal uterine bleeding not controlled by hormones might be better served by hysterectomy. Unfortunately, these patients tend to be high-risk surgical candidates.

If atypia is present, do not perform endometrial ablation or resection. I do perform ablation and resection in patients with complex hyperplasia without atypia if it has been reversed with progestin and does not recur for at least 6 months without progestin therapy. These patients undergo office hysteroscopy and sampling of the endometrium before operative hysteroscopy is scheduled.

Iatrogenic adenomyosis

Two theories suggest this is a late complication of operative hysteroscopy. According to the first, when the endometrium is incompletely resected, scarring over this tissue causes the viable glands to grow into the myometrium. The other theory suggests that viable endometrial debris is transported into the myometrium by vessels opened at resection.

I have found that using the vaporization electrode followed by application of a rollerball over the surface of the cavity most effectively reaches maximal tissue depth and, theoretically, prevents adenomyosis. Since most adenomyosis occurs on the posterior wall, I take a strip from this area for pathologic analysis to determine whether adenomyosis preceded the ablation or developed subsequent to it.

Hematometria

This can occur following operative hysteroscopy if viable glands are left in the fundal or cornual region and synechiae develop in the lower segment, preventing egress of blood. It also can occur if the upper endocervix is ablated and subsequently scars, causing stenosis. To avoid this, ablate only to the level of the internal os.

Diagnosis and treatment. Hematometria can be diagnosed easily by ultrasound and treated with office hysteroscopy using a narrow-diameter, rigid, continuous-flow hysteroscope with an operating channel to pass small instruments.

Post-ablation tubal ligation syndrome

This is cornual hematometria that develops when viable endometrial cells are left in the cornua when the cavity also contains synechiae, causing cyclic bleeding. Since there is no egress from the cervix or tubes, blood gradually builds up, leading to hematosalpinx and pain. One way to avoid this is to ensure complete ablation of the cornual endometrium.

Prevention. Some experts recommend that the small rollerball electrode be placed in the cornua, with slightly reduced intrauterine pressure, to allow the cornual endometrium to collapse around the rollerball. A short burst of current is then applied to ablate the tissue.

 

This complication is less likely after hydrothermablation, since the free-flowing saline ablates the cornua completely. After more than 50 hydrothermablations performed in patients with prior tubal sterilizations, I have not seen any cases of post-ablation tubal ligation syndrome.

Treatment consists of bilateral salpingectomy or tubal fulguration at the cornual region and repeat ablation or resection of viable endometrial tissue. Another option is hysterectomy.

Post-ablation pregnancy can be very complicated

Pregnancy after endometrial ablation occurs at a rate of 0.2% to 1.6%. Counsel patients that this procedure does not prevent pregnancy and that contraception is vital. Uterine rupture after fibroid resection has been reported.

In a review of 37 post-ablation pregnancies, only 11 of 17 women who chose not to terminate carried the gestation beyond 28 weeks. In addition, there was a high incidence of intrauterine growth restriction, prematurity, and placenta accreta.²²

Hysteroscopic tubal sterilization with the Essure system (Conceptus, San Carlos, Calif), or laparoscopic tubal fulguration performed at the time of ablation averts these complications.

Complications of global ablation

Global ablation technologies were developed to enable gynecologists with limited operative hysteroscopy skills to perform endometrial ablation and to make ablation safer for the patient. These technologies completely eliminate the risk of distention-media complications, but widespread use has resulted in other complications that have been reported in the literature to only a limited extent.

Most published articles on global endometrial ablation are from the original US Food and Drug Administration (FDA) trials, in which the complication rates were extraordinarily low. Widespread commercial use of these technologies since FDA approval, especially by practitioners with limited skills, has increased complications.

Do not override safety systems

Complications are more frequent when devices are misused or safety systems overridden. And, fear of litigation makes physicians unwilling to report complications.

In the FDA Manufacturer and User Facility Device Experience (MAUDE) database (www.fda.gov.cdrh/maude.html), complications include bowel burns after unrecognized perforation, and bowel burns associated with electrosurgical, microwave energy, or heat transferred through intact myometrium.²³ Vaginal burns, uterine necrosis myometritis requiring hysterectomy, and death from unrecognized bowel burn also have been reported.

Most global procedures are performed blindly, and some doctors fail to perform diagnostic hysteroscopy before and after surgery, which I feel is mandatory with any endometrial ablation. Hydrothermablation is the only global technique that has the advantage of direct observation. In more than 150 procedures done in my office under local anesthesia, the only complications were 2 false passages. Both were promptly identified during diagnostic hysteroscopy, and the surgery was rescheduled 2 to 3 months later.

The author has served on the speakers’ bureau for Boston Scientific.

Identifying vault prolapse can be difficult in a woman with extensive vaginal prolapse, and operative failure is likely if support to the apex is not restored.

Because this condition is so challenging to identify, many women undergoing anterior and/or posterior colporrhaphy likely have undiagnosed vault prolapse. This may contribute to the 29.2% rate of reoperation in women who undergo pelvic floor reconstructive procedures.¹

This article reviews the anatomy of apical support, tells how to identify vaginal vault prolapse during the physical exam, and outlines effective surgical options—both vaginal and abdominal—for its correction. We focus on accurate pelvic assessment as the basis for planning the surgery.

Vaginal stability is fragile

The stability of vaginal anatomy is precarious, since it depends on a series of interrelationships between both dynamic and static structures. When the relationships between the ligaments and fascia at the vaginal apex or vault are impaired, vault prolapse ensues.

Thanks to cadaveric and radiographic studies, our understanding of the complexities of vaginal anatomy has improved considerably; still, the area of vaginal support we least understand is the coalescence of ligaments and fascia at the vaginal apex or vault.

Grade II prolapse, at least, in 64.8%

An analysis of Women’s Health Initiative enrollees with an intact uterus found that 64.8% had at least grade II prolapse (ie, leading edge of prolapse at –1 to +1 cm from the hymen) according to the Pelvic Organ Prolapse Quantification System (POP-Q).² Approximately 8% of enrollees had a point D (vaginal apex) of greater than –6 cm, suggesting some degree of vault prolapse.

Hysterectomy appears to contribute. The incidence is about 1% at 3 years; 5% at 17 years.³

In the United States, approximately 30,000 vaginal vault repairs were performed in 1999.

Normal support structure

Several support structures coalesce at the vaginal apex. If the cervix is present, it serves as an obvious strong attachment site (FIGURE 1). In hysterectomized women, the structures may lack a strong attachment site, resulting in weakness and prolapse.

FIGURE 1 Vaginal support system

The coalescence of both sets of ligaments forms the uterosacral-cardinal ligament complex at the vaginal apex, which is likely crucial to vault support. Reprinted with permission of The Cleveland Clinic Foundation.

2 sets of ligaments determine support

Uterosacral ligaments—peritoneal and fibromuscular tissue bands extending from the vaginal apex to the sacrum—are the principal support for the vaginal apex, despite their apparent lack of strength.

The role of the cardinal ligaments—which extend laterally from the apex to the pelvic sidewall, adjacent to the ischial spine—is less clear. Since they lie proximal to the ureters, restoring vault support by shortening or reattaching them to the apex is a less attractive option.

The coalescence of these 2 sets of ligaments forms the complex that likely maintains vault support.

In hysterectomized women, locating the attachment of this complex to the vaginal cuff (seen on the exam as apical “dimples”) is key to identifying vault prolapse.

New view of cystoceles, rectoceles

The fibromuscular tissue layer underlying the vaginal epithelium envelops the entire vaginal canal, extending from apex to perineum and from arcus tendineus to arcus tendineus.

As the aponeurosis does for the abdominal wall, the endopelvic fascia maintains integrity of the anterior and posterior vaginal walls. If the fascial layer detaches from the vaginal apex, a true hernia can develop in the form of an enterocele—anterior or posterior—further weakening vault integrity (FIGURE 2).

Reconstructive surgeons are beginning to view cystoceles and rectoceles as a detachment of the endopelvic fascia from the vaginal apex. Thus, it is critical to restore anterior and posterior vaginal wall fascial integrity from apex to perineum by reattaching the endogenous fascia to the vaginal apex, or by placing a biologic or synthetic graft.

FIGURE 2 Apical defects contribute to vault prolapse

 

Vault prolapse is often associated with defects of the apical fascia, represented here by dark lines, which must be addressed during vault reconstruction. Reprinted with permission of The Cleveland Clinic Foundation.

Specific technique, tools to help identify prolapse

Any patient with an advanced degree of vaginal prolapse should be assessed for vault prolapse using a careful, structured pelvic exam. In many cases, this can be difficult, even if the uterus is present.

Necessary tools include a bivalved speculum and a right-angle retractor, or the posterior blade of another gynecologic speculum.

When the uterus is present

An exteriorized cervix does not necessarily mean vault prolapse; this may occur with substantial cervical hypertrophy, while the apex remains well supported (FIGURE 3).

Exam technique. Place the right-angle speculum blade in the posterior fornix, inserting it to its full extent, and ask the patient to perform a Valsalva maneuver. If vault prolapse is present, the uterus will descend further as the speculum is slowly removed; reinsertion of the speculum will resuspend the uterus. If the vault is well supported, the cervix will remain in place despite Valsalva efforts.

Assess the degree of vault prolapse during this examination, to determine whether a McCall culdoplasty will restore vault support.

If uterine suspension is performed in a woman with substantial cervical hypertrophy, cervical prolapse may persist, necessitating partial amputation (Manchester procedure).

FIGURE 3 Exteriorized cervix does not necessarily mean vault prolapse

Cervical prolapse may be associated with vault prolapse (left) or simply represent cervical hypertrophy without vault prolapse (right). Reprinted with permission of The Cleveland Clinic Foundation.

In the hysterectomized patient

The goal of physical exam is to identify the apical scar tissue (cuff) resultant from the hysterectomy. In most women, the cuff is visible as a transverse band of tissue firmer than the adjacent vaginal walls. If the woman has extensive prolapse, the tissue is stretched and thus not as obvious.

Exam technique. Use a bivalved speculum to visualize the apex. In women with extensive prolapse, redundant vaginal tissue may impede visualization. Fortunately, the sites of previous attachment of the uterosacral-cardinal ligament complex can usually be identified as “dimples” on either side of the midline at the cuff (FIGURE 4).

Use both right-angle speculum blades, or 1 blade along the anterior vaginal wall and the index and middle fingers of your other hand along the posterior vaginal wall, to identify the dimples. Then place the tip of the speculum between the dimples, elevate the vault while the patient performs a Valsalva effort, and determine the degree of vault prolapse. This can be confirmed by digital exam by identifying the dimples by tact and elevating them to their ipsilateral ischial spines.

FIGURE 4 Identifying the vault in the hysterectomized patient

Posthysterectomy vault prolapse can be identified by looking for “dimples” at the apex, which represent sites of previous uterosacral-cardinal ligament complex attachment. Reprinted with permission of The Cleveland Clinic Foundation.

Which exam findings point to which technique?

The importance of accurate pelvic assessment is impossible to overemphasize. Besides determining the degree and type of prolapse present, the exam enhances surgical planning. Fascial tears or defects are usually identifiable during careful vaginal exam as areas of sudden change in the thickness of the vaginal wall.

By the end of the pelvic exam, we usually have developed a surgical plan for the prolapse repair, pending urodynamic assessment to determine the best anti-incontinence procedure, if necessary.

What are the surgical goals?

Objectives are to normalize support of all anatomic compartments; alleviate clinical symptoms; and optimize sexual, bowel, and bladder function—without precipitating new support or functional problems.

Abdominal versus vaginal approach

Most surgeons prefer a vaginal approach to pelvic reconstruction. However, this decision should be based on the patient’s individual variables.

If sexual function is critical to the patient, a sacrocolpopexy should be the primary option. Note that age does not always predict the importance of sexual function.

Vaginal length. If the vaginal apex (dimples) reaches the ischial spines with ease, a vaginal procedure should suffice. If it does not reach the spines, or extends far above, an abdominal sacrocolpopexy or obliterative procedure may more be appropriate.

Previous reconstructive procedures. Keep in mind that the area around the sacral promontory, or sacrospinous ligaments, may be difficult or risky to reach due to scarring and fibrosis. This is doubly important in this age of commonplace graft use.

Large paravaginal defects. Vaginal repairs can be technically difficult, and long-term outcomes have not been reported. An abdominal approach is probably better if substantial paravaginal defects are present.

 

Medical comorbidities. Use a vaginal or obliterative procedure under regional anesthesia if the patient is medically delicate or elderly.

Tissue quality usually improves with preoperative local estrogen, but large fascial defects adjacent to the cuff or perineum may require graft reinforcement.

Colorectal dysfunction frequently coexists in women with vault prolapse. Thus, a woman with extensive rectal prolapse should probably undergo concomitant Ripstein rectopexy and sacrocolpopexy, or a perineal proctosigmoidectomy and vaginal-approach vault suspension.

Careful and consistent preparation

Surgical success depends in great part on developing a clear understanding of anatomic defects and urodynamic dysfunction during the preoperative evaluation, to determine the most appropriate procedures.

Tissue preparation with low-dose estrogen

cream (1 g, two nights per week) is crucial for most postmenopausal women.

Obtain medical clearance, and optimize

perioperative safety by using spinal anesthesia, antiembolism stockings, and prophylactic intravenous antibiotics.

Retain vaginal packing at least 24 hours to prevent stress on sutures due to coughing or vomiting.

Advise patients in advance that, for 6 weeks after surgery, they must avoid overexertion and lifting more than 5 lb.

After 6 weeks, we restart estrogen cream and prescribe routine, daily Kegel exercise.

Vaginal procedures

McCall/Mayo culdoplasty

This involves plicating the uterosacral ligaments in the midline while reefing the peritoneum in the cul-de-sac, resulting in posterior culdoplasty. It usually is performed at the time of vaginal hysterectomy using nonabsorbable sutures to incorporate both uterosacral ligaments, intervening cul-desac peritoneum, and full-thickness apical vaginal mucosa. Multiple sutures may be required if prolapse is extensive.

Generally, we try to place our uppermost suture on the uterosacral ligaments at a distance from the cuff equal to the amount of vault prolapse (POP-Q: TVL minus point D [point C if uterus is absent]).

Be careful not to injure or kink the ureters when placing the suture through the uterosacral ligaments, as the ureters lie 1 to 2 cm lateral at the level of the cervix. We recommend cystoscopy with visualization of ureteral patency.

Success rates are high, but objective long-term data is scant.^4,5

Uterosacral ligament suspension

Excellent anatomic outcomes have been described when the uterosacral ligaments are reattached to the vaginal apex (similar to the McCall technique).^6,7 The physiologic nature of this technique makes it very attractive. It involves opening the vaginal wall from anterior to posterior over the apical defect, and identifying the pubocervical fascia, rectovaginal fascia, and uterosacral ligaments.

Technique. Place 1 permanent 1-0 suture and 1 delayed absorbable 1-0 suture in the posteromedial aspect of each uterosacral ligament 1 to 2 cm proximal and medial to each ischial spine. Then place 1 arm of each permanent suture through the pubocervical and rectovaginal fascia, and 1 arm of each delayed absorbable suture through the same tissue, also incorporating the vaginal epithelium. After repairing all additional defects, tie the sutures to suspend the vault.

When prolapse is extensive, redundant peritoneum can hinder identification of the uterosacral ligaments.

Success rates are 87% to 90%, but ureteral injury is a limiting factor, with rates as high as 11%. Therefore, cystoscopy is essential. Long-term data are lacking.

Iliococcygeus suspension

This safe and simple procedure involves elevating the vaginal apex to the iliococcygeus muscles along the lateral pelvic sidewall. This can be done without a vaginal incision by placing a monofilament permanent suture (polypropylene) full thickness through the vaginal wall into the muscle uni-or bilaterally.

Candidates should not be sexually active, as there will be a suture knot in the vagina. The procedure may be useful in elderly patients for whom complete restoration of vaginal anatomy is not a goal. It also can be performed as a salvage operation in women with suboptimal vault support and good distal vaginal anatomy. In addition, it can be performed following posterior vaginal wall dissection with entry into the pararectal space.

Technique. Place the sutures into the fascia overlying the iliococcygeus muscle, anterior to the ischial spine and inferior to the arcus tendineus fascia pelvis, and incorporate the pubocervical fascia anteriorly and the rectovaginal fascia posteriorly.

Success rates. Shull reported a 95% cure rate of the apical compartment among 42 women, at 6 weeks to 5 years.⁸ However, the prolapse at other sites was 14%. A randomized trial comparing this procedure to sacrospinous fixation demonstrated similar satisfactory outcomes.⁹

Sacrospinous ligament fixation

Probably the most commonly performed apical suspension procedure from the vaginal approach is fixation of the apex to the sacrospinous ligaments. Although many describe unilateral fixation, we advocate bilateral fixation to avoid lateral deviation of the vaginal axis (FIGURE 5).

Technique. After entering the pararectal space through a posterior vaginal wall dissection, identify the sacrospinous ligaments and place 2 nonabsorbable sutures through each ligament, rather than around it, as the pudendal vessels pass behind it.

 

Place the first suture 2 cm medial to the ischial spine, and the second suture 1 cm medial to the first. Then pass each suture through the underside of the vaginal apex—in the midline if the procedure is done unilaterally and under each apex if it is bilateral. When tied, the sutures suspend the vaginal apex by approximating it to the ligament, ideally without a suture bridge.

We use CV-2 GoreTex (WL Gore and Associates, Flagstaff, Ariz) sutures passed through the ligaments with a Miya hook, and we reinforce the underside of the vaginal apex with a rectangular piece of Prolene mesh (Ethicon, Somerville, NJ) if the mucosa is thinned.

Success rates are 70% to 97%.^10,11 A significant concern is the nonanatomic posterior axial deflection of the vagina. Many investigators have reported an anterior compartment prolapse rate of up to 20% after fixation, likely secondary to increased force on the anterior compartment with increases in abdominal pressure. This is especially likely if a concomitant anti-incontinence procedure is performed.

Other complications include hemorrhage, vaginal shortening, sexual dysfunction, and buttock pain.

FIGURE 5 Bilateral sacrospinous fixation avoids lateral vaginal deviation

With bilateral fixation of the vault to the sacrospinous ligaments, the vaginal axis is more horizontal. It may be reinforced to enhance longevity. Reprinted with permission of The Cleveland Clinic Foundation.

Posterior IVS vault suspension

This novel, minimally invasive technique uses the posterior intravaginal slingplasty (Posterior IVS; Tyco/US Surgical, Norwalk, Conn). First described as infracoccygeal sacropexy, it was introduced as an outpatient procedure in Australia. Concerns about postoperative vaginal length and risk of rectal injury led to poor acceptance. The procedure was modified by a few US surgeons to enhance safety and vaginal length.

Technique. Enter the pararectal space in a fashion similar to that of sacrospinous fixation. A specially designed tunneler device delivers a multifilament polypropylene tape through bilateral perianal incisions. Secure the tape to the vaginal apex, and adjust it to provide vault support.

We modified this procedure to create neoligaments analogous to cardinal ligaments, by directing the tunneler through the iliococcygeus muscles in close proximity to the ischial spines and arcus tendineus. The resultant vaginal axis is physiologic, and vaginal length is normalized.

By combining this technique with perineoplasty and attaching the rectovaginal and pubocervical fascia to the tape, all levels of pelvic support are repaired once the vault is positioned by pulling on the perianal tape ends.

The new Apogee technique (American Medical Systems, Minnetonka, Minn) uses a similar perianal approach with monofilament polypropylene mesh.

Success rates. Preliminary success rates are 88% to 100%, and complication rates are minimal.¹² Vaginal length averages 7 to 8 cm. Most initially reported complications involved graft erosion or rejection; shifting from nylon to polypropylene graft material reduced this problem.

Abdominal procedures

Sacral colpopexy

Considered the gold standard, the sacral colpopexy vaginal vault suspension technique has a consistent cure rate above 90%.¹³ It may be the ideal procedure for pelvic floor muscle weakness and/or attenuated fascia with multiple defects, for women for whom optimal sexual function is critical, and for those with other indications for abdominal surgery.

A graft is placed between the vagina and the sacral promontory to restore vaginal support (FIGURE 6). Materials have included autologous and synthetic materials. We use polypropylene mesh because of its high tensile strength, biocompatibility, low infection rate, and low incidence of erosion. Biologic grafts such as cadaveric fascia lata have increased failure rates due to graft breakdown.

The resultant vaginal axis is the most physiologic of all vault reconstructive procedures. This procedure appears to have the best longevity of all vault suspension procedures. It can be performed laparo-scopically at selected centers.

Technique. First, access the presacral space overlying the sacral promontory, taking care not to disturb the presacral and middle sacral vessels. We perform this step first to avoid potential periosteal tissue contamination. We routinely use 2 bone anchors to secure the mesh—making sterility imperative. Bone anchors reduce periosteal tissue trauma and decrease risk of potentially life-threatening hemorrhage.

Mobilize the bladder from the anterior vaginal apex. Repair any apical fascial defects, restoring continuity of the pubocervical and rectovaginal fascia, which often detach from the apex. Using 2-0 Prolene sutures, suture the y-shaped graft to both the anterior and posterior vaginal walls, incorporating all fascial edges.

Culdoplasty follows; this obliterates the cul-de-sac to prevent subsequent enterocele formation.

Next, place the graft in a tension-free manner, creating a suspensory bridge from the apex to the sacral promontory. Irrigate copiously. Close the peritoneum over the graft along its entire length.

 

Follow with any anti-incontinence and paravaginal support procedures as well as posterior colporrhaphy as needed.^14,15

Major complications include hemorrhage, usually involving periosteal perforators along the sacrum. Graft erosion may affect up to 5% to 7% of sacral colpopexies.

FIGURE 6 Mesh bridge aids vault suspension

Abdominal sacrocolpopexy with a mesh bridge from the vaginal apex to the sacral promontory. Reprinted with permission of The Cleveland Clinic Foundation.

Uterosacral ligament suspension

In this procedure, which can be performed open or laparoscopically, the remnants of the uterosacral ligaments suspend the vaginal apex. The laparoscopic procedure is simple, especially if the uterus is in place.

Technique. Identify the course of the ureters in relation to the ligaments, and use nonabsorbable sutures to incorporate both of the uterosacral ligaments, peritoneum, and the vaginal apex—including the pubocervical and rectovaginal fascia (FIGURE 7).

Place multiple sutures (include the posterior vaginal wall) to obliterate the cul-de-sac and prevent enterocele development.

Success rates. Long-term data are minimal, but outcomes should be similar to the vaginal-approach culdoplasty.

FIGURE 7 Suspension from uterosacral ligaments

Laparoscopic uterosacral ligament suspension incorporating both uterosacral ligaments and cervix or vaginal cuff.

Reprinted with permission of The Cleveland Clinic Foundation.

Obliterative procedures

LeForte colpocleisis or colpectomy/vaginectomy are the simplest treatments for advanced prolapse in elderly women who are not—and will not be—sexually active.¹⁶

We prefer the LeForte colpocleisis, in which rectangular segments of the anterior and posterior vaginal walls are denuded of their epithelium, followed by approximation of the rectangles to one another.

Success rates exceed 95%, and safety is maintained if spinal anesthesia is used in conjunction with a high perineoplasty.

Dr. Biller reports no relevant financial relationships. Dr. Davila reports research support from AMS and Tyco/US Surgical. He also serves as a consultant to AMS, and as a speaker for AMS and Tyco/US Surgical.

Identifying vault prolapse can be difficult in a woman with extensive vaginal prolapse, and operative failure is likely if support to the apex is not restored.

Because this condition is so challenging to identify, many women undergoing anterior and/or posterior colporrhaphy likely have undiagnosed vault prolapse. This may contribute to the 29.2% rate of reoperation in women who undergo pelvic floor reconstructive procedures.¹

This article reviews the anatomy of apical support, tells how to identify vaginal vault prolapse during the physical exam, and outlines effective surgical options—both vaginal and abdominal—for its correction. We focus on accurate pelvic assessment as the basis for planning the surgery.

Vaginal stability is fragile

The stability of vaginal anatomy is precarious, since it depends on a series of interrelationships between both dynamic and static structures. When the relationships between the ligaments and fascia at the vaginal apex or vault are impaired, vault prolapse ensues.

Thanks to cadaveric and radiographic studies, our understanding of the complexities of vaginal anatomy has improved considerably; still, the area of vaginal support we least understand is the coalescence of ligaments and fascia at the vaginal apex or vault.

Grade II prolapse, at least, in 64.8%

An analysis of Women’s Health Initiative enrollees with an intact uterus found that 64.8% had at least grade II prolapse (ie, leading edge of prolapse at –1 to +1 cm from the hymen) according to the Pelvic Organ Prolapse Quantification System (POP-Q).² Approximately 8% of enrollees had a point D (vaginal apex) of greater than –6 cm, suggesting some degree of vault prolapse.

Hysterectomy appears to contribute. The incidence is about 1% at 3 years; 5% at 17 years.³

In the United States, approximately 30,000 vaginal vault repairs were performed in 1999.

Normal support structure

Several support structures coalesce at the vaginal apex. If the cervix is present, it serves as an obvious strong attachment site (FIGURE 1). In hysterectomized women, the structures may lack a strong attachment site, resulting in weakness and prolapse.

FIGURE 1 Vaginal support system

The coalescence of both sets of ligaments forms the uterosacral-cardinal ligament complex at the vaginal apex, which is likely crucial to vault support. Reprinted with permission of The Cleveland Clinic Foundation.

2 sets of ligaments determine support

Uterosacral ligaments—peritoneal and fibromuscular tissue bands extending from the vaginal apex to the sacrum—are the principal support for the vaginal apex, despite their apparent lack of strength.

The role of the cardinal ligaments—which extend laterally from the apex to the pelvic sidewall, adjacent to the ischial spine—is less clear. Since they lie proximal to the ureters, restoring vault support by shortening or reattaching them to the apex is a less attractive option.

The coalescence of these 2 sets of ligaments forms the complex that likely maintains vault support.

In hysterectomized women, locating the attachment of this complex to the vaginal cuff (seen on the exam as apical “dimples”) is key to identifying vault prolapse.

New view of cystoceles, rectoceles

The fibromuscular tissue layer underlying the vaginal epithelium envelops the entire vaginal canal, extending from apex to perineum and from arcus tendineus to arcus tendineus.

As the aponeurosis does for the abdominal wall, the endopelvic fascia maintains integrity of the anterior and posterior vaginal walls. If the fascial layer detaches from the vaginal apex, a true hernia can develop in the form of an enterocele—anterior or posterior—further weakening vault integrity (FIGURE 2).

Reconstructive surgeons are beginning to view cystoceles and rectoceles as a detachment of the endopelvic fascia from the vaginal apex. Thus, it is critical to restore anterior and posterior vaginal wall fascial integrity from apex to perineum by reattaching the endogenous fascia to the vaginal apex, or by placing a biologic or synthetic graft.

FIGURE 2 Apical defects contribute to vault prolapse

 

Vault prolapse is often associated with defects of the apical fascia, represented here by dark lines, which must be addressed during vault reconstruction. Reprinted with permission of The Cleveland Clinic Foundation.

Specific technique, tools to help identify prolapse

Any patient with an advanced degree of vaginal prolapse should be assessed for vault prolapse using a careful, structured pelvic exam. In many cases, this can be difficult, even if the uterus is present.

Necessary tools include a bivalved speculum and a right-angle retractor, or the posterior blade of another gynecologic speculum.

When the uterus is present

An exteriorized cervix does not necessarily mean vault prolapse; this may occur with substantial cervical hypertrophy, while the apex remains well supported (FIGURE 3).

Exam technique. Place the right-angle speculum blade in the posterior fornix, inserting it to its full extent, and ask the patient to perform a Valsalva maneuver. If vault prolapse is present, the uterus will descend further as the speculum is slowly removed; reinsertion of the speculum will resuspend the uterus. If the vault is well supported, the cervix will remain in place despite Valsalva efforts.

Assess the degree of vault prolapse during this examination, to determine whether a McCall culdoplasty will restore vault support.

If uterine suspension is performed in a woman with substantial cervical hypertrophy, cervical prolapse may persist, necessitating partial amputation (Manchester procedure).

FIGURE 3 Exteriorized cervix does not necessarily mean vault prolapse

Cervical prolapse may be associated with vault prolapse (left) or simply represent cervical hypertrophy without vault prolapse (right). Reprinted with permission of The Cleveland Clinic Foundation.

In the hysterectomized patient

The goal of physical exam is to identify the apical scar tissue (cuff) resultant from the hysterectomy. In most women, the cuff is visible as a transverse band of tissue firmer than the adjacent vaginal walls. If the woman has extensive prolapse, the tissue is stretched and thus not as obvious.

Exam technique. Use a bivalved speculum to visualize the apex. In women with extensive prolapse, redundant vaginal tissue may impede visualization. Fortunately, the sites of previous attachment of the uterosacral-cardinal ligament complex can usually be identified as “dimples” on either side of the midline at the cuff (FIGURE 4).

Use both right-angle speculum blades, or 1 blade along the anterior vaginal wall and the index and middle fingers of your other hand along the posterior vaginal wall, to identify the dimples. Then place the tip of the speculum between the dimples, elevate the vault while the patient performs a Valsalva effort, and determine the degree of vault prolapse. This can be confirmed by digital exam by identifying the dimples by tact and elevating them to their ipsilateral ischial spines.

FIGURE 4 Identifying the vault in the hysterectomized patient

Posthysterectomy vault prolapse can be identified by looking for “dimples” at the apex, which represent sites of previous uterosacral-cardinal ligament complex attachment. Reprinted with permission of The Cleveland Clinic Foundation.

Which exam findings point to which technique?

The importance of accurate pelvic assessment is impossible to overemphasize. Besides determining the degree and type of prolapse present, the exam enhances surgical planning. Fascial tears or defects are usually identifiable during careful vaginal exam as areas of sudden change in the thickness of the vaginal wall.

By the end of the pelvic exam, we usually have developed a surgical plan for the prolapse repair, pending urodynamic assessment to determine the best anti-incontinence procedure, if necessary.

What are the surgical goals?

Objectives are to normalize support of all anatomic compartments; alleviate clinical symptoms; and optimize sexual, bowel, and bladder function—without precipitating new support or functional problems.

Abdominal versus vaginal approach

Most surgeons prefer a vaginal approach to pelvic reconstruction. However, this decision should be based on the patient’s individual variables.

If sexual function is critical to the patient, a sacrocolpopexy should be the primary option. Note that age does not always predict the importance of sexual function.

Vaginal length. If the vaginal apex (dimples) reaches the ischial spines with ease, a vaginal procedure should suffice. If it does not reach the spines, or extends far above, an abdominal sacrocolpopexy or obliterative procedure may more be appropriate.

Previous reconstructive procedures. Keep in mind that the area around the sacral promontory, or sacrospinous ligaments, may be difficult or risky to reach due to scarring and fibrosis. This is doubly important in this age of commonplace graft use.

Large paravaginal defects. Vaginal repairs can be technically difficult, and long-term outcomes have not been reported. An abdominal approach is probably better if substantial paravaginal defects are present.

 

Medical comorbidities. Use a vaginal or obliterative procedure under regional anesthesia if the patient is medically delicate or elderly.

Tissue quality usually improves with preoperative local estrogen, but large fascial defects adjacent to the cuff or perineum may require graft reinforcement.

Colorectal dysfunction frequently coexists in women with vault prolapse. Thus, a woman with extensive rectal prolapse should probably undergo concomitant Ripstein rectopexy and sacrocolpopexy, or a perineal proctosigmoidectomy and vaginal-approach vault suspension.

Careful and consistent preparation

Surgical success depends in great part on developing a clear understanding of anatomic defects and urodynamic dysfunction during the preoperative evaluation, to determine the most appropriate procedures.

Tissue preparation with low-dose estrogen

cream (1 g, two nights per week) is crucial for most postmenopausal women.

Obtain medical clearance, and optimize

perioperative safety by using spinal anesthesia, antiembolism stockings, and prophylactic intravenous antibiotics.

Retain vaginal packing at least 24 hours to prevent stress on sutures due to coughing or vomiting.

Advise patients in advance that, for 6 weeks after surgery, they must avoid overexertion and lifting more than 5 lb.

After 6 weeks, we restart estrogen cream and prescribe routine, daily Kegel exercise.

Vaginal procedures

McCall/Mayo culdoplasty

This involves plicating the uterosacral ligaments in the midline while reefing the peritoneum in the cul-de-sac, resulting in posterior culdoplasty. It usually is performed at the time of vaginal hysterectomy using nonabsorbable sutures to incorporate both uterosacral ligaments, intervening cul-desac peritoneum, and full-thickness apical vaginal mucosa. Multiple sutures may be required if prolapse is extensive.

Generally, we try to place our uppermost suture on the uterosacral ligaments at a distance from the cuff equal to the amount of vault prolapse (POP-Q: TVL minus point D [point C if uterus is absent]).

Be careful not to injure or kink the ureters when placing the suture through the uterosacral ligaments, as the ureters lie 1 to 2 cm lateral at the level of the cervix. We recommend cystoscopy with visualization of ureteral patency.

Success rates are high, but objective long-term data is scant.^4,5

Uterosacral ligament suspension

Excellent anatomic outcomes have been described when the uterosacral ligaments are reattached to the vaginal apex (similar to the McCall technique).^6,7 The physiologic nature of this technique makes it very attractive. It involves opening the vaginal wall from anterior to posterior over the apical defect, and identifying the pubocervical fascia, rectovaginal fascia, and uterosacral ligaments.

Technique. Place 1 permanent 1-0 suture and 1 delayed absorbable 1-0 suture in the posteromedial aspect of each uterosacral ligament 1 to 2 cm proximal and medial to each ischial spine. Then place 1 arm of each permanent suture through the pubocervical and rectovaginal fascia, and 1 arm of each delayed absorbable suture through the same tissue, also incorporating the vaginal epithelium. After repairing all additional defects, tie the sutures to suspend the vault.

When prolapse is extensive, redundant peritoneum can hinder identification of the uterosacral ligaments.

Success rates are 87% to 90%, but ureteral injury is a limiting factor, with rates as high as 11%. Therefore, cystoscopy is essential. Long-term data are lacking.

Iliococcygeus suspension

This safe and simple procedure involves elevating the vaginal apex to the iliococcygeus muscles along the lateral pelvic sidewall. This can be done without a vaginal incision by placing a monofilament permanent suture (polypropylene) full thickness through the vaginal wall into the muscle uni-or bilaterally.

Candidates should not be sexually active, as there will be a suture knot in the vagina. The procedure may be useful in elderly patients for whom complete restoration of vaginal anatomy is not a goal. It also can be performed as a salvage operation in women with suboptimal vault support and good distal vaginal anatomy. In addition, it can be performed following posterior vaginal wall dissection with entry into the pararectal space.

Technique. Place the sutures into the fascia overlying the iliococcygeus muscle, anterior to the ischial spine and inferior to the arcus tendineus fascia pelvis, and incorporate the pubocervical fascia anteriorly and the rectovaginal fascia posteriorly.

Success rates. Shull reported a 95% cure rate of the apical compartment among 42 women, at 6 weeks to 5 years.⁸ However, the prolapse at other sites was 14%. A randomized trial comparing this procedure to sacrospinous fixation demonstrated similar satisfactory outcomes.⁹

Sacrospinous ligament fixation

Probably the most commonly performed apical suspension procedure from the vaginal approach is fixation of the apex to the sacrospinous ligaments. Although many describe unilateral fixation, we advocate bilateral fixation to avoid lateral deviation of the vaginal axis (FIGURE 5).

Technique. After entering the pararectal space through a posterior vaginal wall dissection, identify the sacrospinous ligaments and place 2 nonabsorbable sutures through each ligament, rather than around it, as the pudendal vessels pass behind it.

 

Place the first suture 2 cm medial to the ischial spine, and the second suture 1 cm medial to the first. Then pass each suture through the underside of the vaginal apex—in the midline if the procedure is done unilaterally and under each apex if it is bilateral. When tied, the sutures suspend the vaginal apex by approximating it to the ligament, ideally without a suture bridge.

We use CV-2 GoreTex (WL Gore and Associates, Flagstaff, Ariz) sutures passed through the ligaments with a Miya hook, and we reinforce the underside of the vaginal apex with a rectangular piece of Prolene mesh (Ethicon, Somerville, NJ) if the mucosa is thinned.

Success rates are 70% to 97%.^10,11 A significant concern is the nonanatomic posterior axial deflection of the vagina. Many investigators have reported an anterior compartment prolapse rate of up to 20% after fixation, likely secondary to increased force on the anterior compartment with increases in abdominal pressure. This is especially likely if a concomitant anti-incontinence procedure is performed.

Other complications include hemorrhage, vaginal shortening, sexual dysfunction, and buttock pain.

FIGURE 5 Bilateral sacrospinous fixation avoids lateral vaginal deviation

With bilateral fixation of the vault to the sacrospinous ligaments, the vaginal axis is more horizontal. It may be reinforced to enhance longevity. Reprinted with permission of The Cleveland Clinic Foundation.

Posterior IVS vault suspension

This novel, minimally invasive technique uses the posterior intravaginal slingplasty (Posterior IVS; Tyco/US Surgical, Norwalk, Conn). First described as infracoccygeal sacropexy, it was introduced as an outpatient procedure in Australia. Concerns about postoperative vaginal length and risk of rectal injury led to poor acceptance. The procedure was modified by a few US surgeons to enhance safety and vaginal length.

Technique. Enter the pararectal space in a fashion similar to that of sacrospinous fixation. A specially designed tunneler device delivers a multifilament polypropylene tape through bilateral perianal incisions. Secure the tape to the vaginal apex, and adjust it to provide vault support.

We modified this procedure to create neoligaments analogous to cardinal ligaments, by directing the tunneler through the iliococcygeus muscles in close proximity to the ischial spines and arcus tendineus. The resultant vaginal axis is physiologic, and vaginal length is normalized.

By combining this technique with perineoplasty and attaching the rectovaginal and pubocervical fascia to the tape, all levels of pelvic support are repaired once the vault is positioned by pulling on the perianal tape ends.

The new Apogee technique (American Medical Systems, Minnetonka, Minn) uses a similar perianal approach with monofilament polypropylene mesh.

Success rates. Preliminary success rates are 88% to 100%, and complication rates are minimal.¹² Vaginal length averages 7 to 8 cm. Most initially reported complications involved graft erosion or rejection; shifting from nylon to polypropylene graft material reduced this problem.

Abdominal procedures

Sacral colpopexy

Considered the gold standard, the sacral colpopexy vaginal vault suspension technique has a consistent cure rate above 90%.¹³ It may be the ideal procedure for pelvic floor muscle weakness and/or attenuated fascia with multiple defects, for women for whom optimal sexual function is critical, and for those with other indications for abdominal surgery.

A graft is placed between the vagina and the sacral promontory to restore vaginal support (FIGURE 6). Materials have included autologous and synthetic materials. We use polypropylene mesh because of its high tensile strength, biocompatibility, low infection rate, and low incidence of erosion. Biologic grafts such as cadaveric fascia lata have increased failure rates due to graft breakdown.

The resultant vaginal axis is the most physiologic of all vault reconstructive procedures. This procedure appears to have the best longevity of all vault suspension procedures. It can be performed laparo-scopically at selected centers.

Technique. First, access the presacral space overlying the sacral promontory, taking care not to disturb the presacral and middle sacral vessels. We perform this step first to avoid potential periosteal tissue contamination. We routinely use 2 bone anchors to secure the mesh—making sterility imperative. Bone anchors reduce periosteal tissue trauma and decrease risk of potentially life-threatening hemorrhage.

Mobilize the bladder from the anterior vaginal apex. Repair any apical fascial defects, restoring continuity of the pubocervical and rectovaginal fascia, which often detach from the apex. Using 2-0 Prolene sutures, suture the y-shaped graft to both the anterior and posterior vaginal walls, incorporating all fascial edges.

Culdoplasty follows; this obliterates the cul-de-sac to prevent subsequent enterocele formation.

Next, place the graft in a tension-free manner, creating a suspensory bridge from the apex to the sacral promontory. Irrigate copiously. Close the peritoneum over the graft along its entire length.

 

Follow with any anti-incontinence and paravaginal support procedures as well as posterior colporrhaphy as needed.^14,15

Major complications include hemorrhage, usually involving periosteal perforators along the sacrum. Graft erosion may affect up to 5% to 7% of sacral colpopexies.

FIGURE 6 Mesh bridge aids vault suspension

Abdominal sacrocolpopexy with a mesh bridge from the vaginal apex to the sacral promontory. Reprinted with permission of The Cleveland Clinic Foundation.

Uterosacral ligament suspension

In this procedure, which can be performed open or laparoscopically, the remnants of the uterosacral ligaments suspend the vaginal apex. The laparoscopic procedure is simple, especially if the uterus is in place.

Technique. Identify the course of the ureters in relation to the ligaments, and use nonabsorbable sutures to incorporate both of the uterosacral ligaments, peritoneum, and the vaginal apex—including the pubocervical and rectovaginal fascia (FIGURE 7).

Place multiple sutures (include the posterior vaginal wall) to obliterate the cul-de-sac and prevent enterocele development.

Success rates. Long-term data are minimal, but outcomes should be similar to the vaginal-approach culdoplasty.

FIGURE 7 Suspension from uterosacral ligaments

Laparoscopic uterosacral ligament suspension incorporating both uterosacral ligaments and cervix or vaginal cuff.

Reprinted with permission of The Cleveland Clinic Foundation.

Obliterative procedures

LeForte colpocleisis or colpectomy/vaginectomy are the simplest treatments for advanced prolapse in elderly women who are not—and will not be—sexually active.¹⁶

We prefer the LeForte colpocleisis, in which rectangular segments of the anterior and posterior vaginal walls are denuded of their epithelium, followed by approximation of the rectangles to one another.

Success rates exceed 95%, and safety is maintained if spinal anesthesia is used in conjunction with a high perineoplasty.

Dr. Biller reports no relevant financial relationships. Dr. Davila reports research support from AMS and Tyco/US Surgical. He also serves as a consultant to AMS, and as a speaker for AMS and Tyco/US Surgical.

Identifying vault prolapse can be difficult in a woman with extensive vaginal prolapse, and operative failure is likely if support to the apex is not restored.

Because this condition is so challenging to identify, many women undergoing anterior and/or posterior colporrhaphy likely have undiagnosed vault prolapse. This may contribute to the 29.2% rate of reoperation in women who undergo pelvic floor reconstructive procedures.¹

This article reviews the anatomy of apical support, tells how to identify vaginal vault prolapse during the physical exam, and outlines effective surgical options—both vaginal and abdominal—for its correction. We focus on accurate pelvic assessment as the basis for planning the surgery.

Vaginal stability is fragile

The stability of vaginal anatomy is precarious, since it depends on a series of interrelationships between both dynamic and static structures. When the relationships between the ligaments and fascia at the vaginal apex or vault are impaired, vault prolapse ensues.

Thanks to cadaveric and radiographic studies, our understanding of the complexities of vaginal anatomy has improved considerably; still, the area of vaginal support we least understand is the coalescence of ligaments and fascia at the vaginal apex or vault.

Grade II prolapse, at least, in 64.8%

An analysis of Women’s Health Initiative enrollees with an intact uterus found that 64.8% had at least grade II prolapse (ie, leading edge of prolapse at –1 to +1 cm from the hymen) according to the Pelvic Organ Prolapse Quantification System (POP-Q).² Approximately 8% of enrollees had a point D (vaginal apex) of greater than –6 cm, suggesting some degree of vault prolapse.

Hysterectomy appears to contribute. The incidence is about 1% at 3 years; 5% at 17 years.³

In the United States, approximately 30,000 vaginal vault repairs were performed in 1999.

Normal support structure

Several support structures coalesce at the vaginal apex. If the cervix is present, it serves as an obvious strong attachment site (FIGURE 1). In hysterectomized women, the structures may lack a strong attachment site, resulting in weakness and prolapse.

FIGURE 1 Vaginal support system

The coalescence of both sets of ligaments forms the uterosacral-cardinal ligament complex at the vaginal apex, which is likely crucial to vault support. Reprinted with permission of The Cleveland Clinic Foundation.

2 sets of ligaments determine support

Uterosacral ligaments—peritoneal and fibromuscular tissue bands extending from the vaginal apex to the sacrum—are the principal support for the vaginal apex, despite their apparent lack of strength.

The role of the cardinal ligaments—which extend laterally from the apex to the pelvic sidewall, adjacent to the ischial spine—is less clear. Since they lie proximal to the ureters, restoring vault support by shortening or reattaching them to the apex is a less attractive option.

The coalescence of these 2 sets of ligaments forms the complex that likely maintains vault support.

In hysterectomized women, locating the attachment of this complex to the vaginal cuff (seen on the exam as apical “dimples”) is key to identifying vault prolapse.

New view of cystoceles, rectoceles

The fibromuscular tissue layer underlying the vaginal epithelium envelops the entire vaginal canal, extending from apex to perineum and from arcus tendineus to arcus tendineus.

As the aponeurosis does for the abdominal wall, the endopelvic fascia maintains integrity of the anterior and posterior vaginal walls. If the fascial layer detaches from the vaginal apex, a true hernia can develop in the form of an enterocele—anterior or posterior—further weakening vault integrity (FIGURE 2).

Reconstructive surgeons are beginning to view cystoceles and rectoceles as a detachment of the endopelvic fascia from the vaginal apex. Thus, it is critical to restore anterior and posterior vaginal wall fascial integrity from apex to perineum by reattaching the endogenous fascia to the vaginal apex, or by placing a biologic or synthetic graft.

FIGURE 2 Apical defects contribute to vault prolapse

 

Vault prolapse is often associated with defects of the apical fascia, represented here by dark lines, which must be addressed during vault reconstruction. Reprinted with permission of The Cleveland Clinic Foundation.

Specific technique, tools to help identify prolapse

Any patient with an advanced degree of vaginal prolapse should be assessed for vault prolapse using a careful, structured pelvic exam. In many cases, this can be difficult, even if the uterus is present.

Necessary tools include a bivalved speculum and a right-angle retractor, or the posterior blade of another gynecologic speculum.

When the uterus is present

An exteriorized cervix does not necessarily mean vault prolapse; this may occur with substantial cervical hypertrophy, while the apex remains well supported (FIGURE 3).

Exam technique. Place the right-angle speculum blade in the posterior fornix, inserting it to its full extent, and ask the patient to perform a Valsalva maneuver. If vault prolapse is present, the uterus will descend further as the speculum is slowly removed; reinsertion of the speculum will resuspend the uterus. If the vault is well supported, the cervix will remain in place despite Valsalva efforts.

Assess the degree of vault prolapse during this examination, to determine whether a McCall culdoplasty will restore vault support.

If uterine suspension is performed in a woman with substantial cervical hypertrophy, cervical prolapse may persist, necessitating partial amputation (Manchester procedure).

FIGURE 3 Exteriorized cervix does not necessarily mean vault prolapse

Cervical prolapse may be associated with vault prolapse (left) or simply represent cervical hypertrophy without vault prolapse (right). Reprinted with permission of The Cleveland Clinic Foundation.

In the hysterectomized patient

The goal of physical exam is to identify the apical scar tissue (cuff) resultant from the hysterectomy. In most women, the cuff is visible as a transverse band of tissue firmer than the adjacent vaginal walls. If the woman has extensive prolapse, the tissue is stretched and thus not as obvious.

Exam technique. Use a bivalved speculum to visualize the apex. In women with extensive prolapse, redundant vaginal tissue may impede visualization. Fortunately, the sites of previous attachment of the uterosacral-cardinal ligament complex can usually be identified as “dimples” on either side of the midline at the cuff (FIGURE 4).

Use both right-angle speculum blades, or 1 blade along the anterior vaginal wall and the index and middle fingers of your other hand along the posterior vaginal wall, to identify the dimples. Then place the tip of the speculum between the dimples, elevate the vault while the patient performs a Valsalva effort, and determine the degree of vault prolapse. This can be confirmed by digital exam by identifying the dimples by tact and elevating them to their ipsilateral ischial spines.

FIGURE 4 Identifying the vault in the hysterectomized patient

Posthysterectomy vault prolapse can be identified by looking for “dimples” at the apex, which represent sites of previous uterosacral-cardinal ligament complex attachment. Reprinted with permission of The Cleveland Clinic Foundation.

Which exam findings point to which technique?

The importance of accurate pelvic assessment is impossible to overemphasize. Besides determining the degree and type of prolapse present, the exam enhances surgical planning. Fascial tears or defects are usually identifiable during careful vaginal exam as areas of sudden change in the thickness of the vaginal wall.

By the end of the pelvic exam, we usually have developed a surgical plan for the prolapse repair, pending urodynamic assessment to determine the best anti-incontinence procedure, if necessary.

What are the surgical goals?

Objectives are to normalize support of all anatomic compartments; alleviate clinical symptoms; and optimize sexual, bowel, and bladder function—without precipitating new support or functional problems.

Abdominal versus vaginal approach

Most surgeons prefer a vaginal approach to pelvic reconstruction. However, this decision should be based on the patient’s individual variables.

If sexual function is critical to the patient, a sacrocolpopexy should be the primary option. Note that age does not always predict the importance of sexual function.

Vaginal length. If the vaginal apex (dimples) reaches the ischial spines with ease, a vaginal procedure should suffice. If it does not reach the spines, or extends far above, an abdominal sacrocolpopexy or obliterative procedure may more be appropriate.

Previous reconstructive procedures. Keep in mind that the area around the sacral promontory, or sacrospinous ligaments, may be difficult or risky to reach due to scarring and fibrosis. This is doubly important in this age of commonplace graft use.

Large paravaginal defects. Vaginal repairs can be technically difficult, and long-term outcomes have not been reported. An abdominal approach is probably better if substantial paravaginal defects are present.

 

Medical comorbidities. Use a vaginal or obliterative procedure under regional anesthesia if the patient is medically delicate or elderly.

Tissue quality usually improves with preoperative local estrogen, but large fascial defects adjacent to the cuff or perineum may require graft reinforcement.

Colorectal dysfunction frequently coexists in women with vault prolapse. Thus, a woman with extensive rectal prolapse should probably undergo concomitant Ripstein rectopexy and sacrocolpopexy, or a perineal proctosigmoidectomy and vaginal-approach vault suspension.

Careful and consistent preparation

Surgical success depends in great part on developing a clear understanding of anatomic defects and urodynamic dysfunction during the preoperative evaluation, to determine the most appropriate procedures.

Tissue preparation with low-dose estrogen

cream (1 g, two nights per week) is crucial for most postmenopausal women.

Obtain medical clearance, and optimize

perioperative safety by using spinal anesthesia, antiembolism stockings, and prophylactic intravenous antibiotics.

Retain vaginal packing at least 24 hours to prevent stress on sutures due to coughing or vomiting.

Advise patients in advance that, for 6 weeks after surgery, they must avoid overexertion and lifting more than 5 lb.

After 6 weeks, we restart estrogen cream and prescribe routine, daily Kegel exercise.

Vaginal procedures

McCall/Mayo culdoplasty

This involves plicating the uterosacral ligaments in the midline while reefing the peritoneum in the cul-de-sac, resulting in posterior culdoplasty. It usually is performed at the time of vaginal hysterectomy using nonabsorbable sutures to incorporate both uterosacral ligaments, intervening cul-desac peritoneum, and full-thickness apical vaginal mucosa. Multiple sutures may be required if prolapse is extensive.

Generally, we try to place our uppermost suture on the uterosacral ligaments at a distance from the cuff equal to the amount of vault prolapse (POP-Q: TVL minus point D [point C if uterus is absent]).

Be careful not to injure or kink the ureters when placing the suture through the uterosacral ligaments, as the ureters lie 1 to 2 cm lateral at the level of the cervix. We recommend cystoscopy with visualization of ureteral patency.

Success rates are high, but objective long-term data is scant.^4,5

Uterosacral ligament suspension

Excellent anatomic outcomes have been described when the uterosacral ligaments are reattached to the vaginal apex (similar to the McCall technique).^6,7 The physiologic nature of this technique makes it very attractive. It involves opening the vaginal wall from anterior to posterior over the apical defect, and identifying the pubocervical fascia, rectovaginal fascia, and uterosacral ligaments.

Technique. Place 1 permanent 1-0 suture and 1 delayed absorbable 1-0 suture in the posteromedial aspect of each uterosacral ligament 1 to 2 cm proximal and medial to each ischial spine. Then place 1 arm of each permanent suture through the pubocervical and rectovaginal fascia, and 1 arm of each delayed absorbable suture through the same tissue, also incorporating the vaginal epithelium. After repairing all additional defects, tie the sutures to suspend the vault.

When prolapse is extensive, redundant peritoneum can hinder identification of the uterosacral ligaments.

Success rates are 87% to 90%, but ureteral injury is a limiting factor, with rates as high as 11%. Therefore, cystoscopy is essential. Long-term data are lacking.

Iliococcygeus suspension

This safe and simple procedure involves elevating the vaginal apex to the iliococcygeus muscles along the lateral pelvic sidewall. This can be done without a vaginal incision by placing a monofilament permanent suture (polypropylene) full thickness through the vaginal wall into the muscle uni-or bilaterally.

Candidates should not be sexually active, as there will be a suture knot in the vagina. The procedure may be useful in elderly patients for whom complete restoration of vaginal anatomy is not a goal. It also can be performed as a salvage operation in women with suboptimal vault support and good distal vaginal anatomy. In addition, it can be performed following posterior vaginal wall dissection with entry into the pararectal space.

Technique. Place the sutures into the fascia overlying the iliococcygeus muscle, anterior to the ischial spine and inferior to the arcus tendineus fascia pelvis, and incorporate the pubocervical fascia anteriorly and the rectovaginal fascia posteriorly.

Success rates. Shull reported a 95% cure rate of the apical compartment among 42 women, at 6 weeks to 5 years.⁸ However, the prolapse at other sites was 14%. A randomized trial comparing this procedure to sacrospinous fixation demonstrated similar satisfactory outcomes.⁹

Sacrospinous ligament fixation

Probably the most commonly performed apical suspension procedure from the vaginal approach is fixation of the apex to the sacrospinous ligaments. Although many describe unilateral fixation, we advocate bilateral fixation to avoid lateral deviation of the vaginal axis (FIGURE 5).

Technique. After entering the pararectal space through a posterior vaginal wall dissection, identify the sacrospinous ligaments and place 2 nonabsorbable sutures through each ligament, rather than around it, as the pudendal vessels pass behind it.

 

Place the first suture 2 cm medial to the ischial spine, and the second suture 1 cm medial to the first. Then pass each suture through the underside of the vaginal apex—in the midline if the procedure is done unilaterally and under each apex if it is bilateral. When tied, the sutures suspend the vaginal apex by approximating it to the ligament, ideally without a suture bridge.

We use CV-2 GoreTex (WL Gore and Associates, Flagstaff, Ariz) sutures passed through the ligaments with a Miya hook, and we reinforce the underside of the vaginal apex with a rectangular piece of Prolene mesh (Ethicon, Somerville, NJ) if the mucosa is thinned.

Success rates are 70% to 97%.^10,11 A significant concern is the nonanatomic posterior axial deflection of the vagina. Many investigators have reported an anterior compartment prolapse rate of up to 20% after fixation, likely secondary to increased force on the anterior compartment with increases in abdominal pressure. This is especially likely if a concomitant anti-incontinence procedure is performed.

Other complications include hemorrhage, vaginal shortening, sexual dysfunction, and buttock pain.

FIGURE 5 Bilateral sacrospinous fixation avoids lateral vaginal deviation

With bilateral fixation of the vault to the sacrospinous ligaments, the vaginal axis is more horizontal. It may be reinforced to enhance longevity. Reprinted with permission of The Cleveland Clinic Foundation.

Posterior IVS vault suspension

This novel, minimally invasive technique uses the posterior intravaginal slingplasty (Posterior IVS; Tyco/US Surgical, Norwalk, Conn). First described as infracoccygeal sacropexy, it was introduced as an outpatient procedure in Australia. Concerns about postoperative vaginal length and risk of rectal injury led to poor acceptance. The procedure was modified by a few US surgeons to enhance safety and vaginal length.

Technique. Enter the pararectal space in a fashion similar to that of sacrospinous fixation. A specially designed tunneler device delivers a multifilament polypropylene tape through bilateral perianal incisions. Secure the tape to the vaginal apex, and adjust it to provide vault support.

We modified this procedure to create neoligaments analogous to cardinal ligaments, by directing the tunneler through the iliococcygeus muscles in close proximity to the ischial spines and arcus tendineus. The resultant vaginal axis is physiologic, and vaginal length is normalized.

By combining this technique with perineoplasty and attaching the rectovaginal and pubocervical fascia to the tape, all levels of pelvic support are repaired once the vault is positioned by pulling on the perianal tape ends.

The new Apogee technique (American Medical Systems, Minnetonka, Minn) uses a similar perianal approach with monofilament polypropylene mesh.

Success rates. Preliminary success rates are 88% to 100%, and complication rates are minimal.¹² Vaginal length averages 7 to 8 cm. Most initially reported complications involved graft erosion or rejection; shifting from nylon to polypropylene graft material reduced this problem.

Abdominal procedures

Sacral colpopexy

Considered the gold standard, the sacral colpopexy vaginal vault suspension technique has a consistent cure rate above 90%.¹³ It may be the ideal procedure for pelvic floor muscle weakness and/or attenuated fascia with multiple defects, for women for whom optimal sexual function is critical, and for those with other indications for abdominal surgery.

A graft is placed between the vagina and the sacral promontory to restore vaginal support (FIGURE 6). Materials have included autologous and synthetic materials. We use polypropylene mesh because of its high tensile strength, biocompatibility, low infection rate, and low incidence of erosion. Biologic grafts such as cadaveric fascia lata have increased failure rates due to graft breakdown.

The resultant vaginal axis is the most physiologic of all vault reconstructive procedures. This procedure appears to have the best longevity of all vault suspension procedures. It can be performed laparo-scopically at selected centers.

Technique. First, access the presacral space overlying the sacral promontory, taking care not to disturb the presacral and middle sacral vessels. We perform this step first to avoid potential periosteal tissue contamination. We routinely use 2 bone anchors to secure the mesh—making sterility imperative. Bone anchors reduce periosteal tissue trauma and decrease risk of potentially life-threatening hemorrhage.

Mobilize the bladder from the anterior vaginal apex. Repair any apical fascial defects, restoring continuity of the pubocervical and rectovaginal fascia, which often detach from the apex. Using 2-0 Prolene sutures, suture the y-shaped graft to both the anterior and posterior vaginal walls, incorporating all fascial edges.

Culdoplasty follows; this obliterates the cul-de-sac to prevent subsequent enterocele formation.

Next, place the graft in a tension-free manner, creating a suspensory bridge from the apex to the sacral promontory. Irrigate copiously. Close the peritoneum over the graft along its entire length.

 

Follow with any anti-incontinence and paravaginal support procedures as well as posterior colporrhaphy as needed.^14,15

Major complications include hemorrhage, usually involving periosteal perforators along the sacrum. Graft erosion may affect up to 5% to 7% of sacral colpopexies.

FIGURE 6 Mesh bridge aids vault suspension

Abdominal sacrocolpopexy with a mesh bridge from the vaginal apex to the sacral promontory. Reprinted with permission of The Cleveland Clinic Foundation.

Uterosacral ligament suspension

In this procedure, which can be performed open or laparoscopically, the remnants of the uterosacral ligaments suspend the vaginal apex. The laparoscopic procedure is simple, especially if the uterus is in place.

Technique. Identify the course of the ureters in relation to the ligaments, and use nonabsorbable sutures to incorporate both of the uterosacral ligaments, peritoneum, and the vaginal apex—including the pubocervical and rectovaginal fascia (FIGURE 7).

Place multiple sutures (include the posterior vaginal wall) to obliterate the cul-de-sac and prevent enterocele development.

Success rates. Long-term data are minimal, but outcomes should be similar to the vaginal-approach culdoplasty.

FIGURE 7 Suspension from uterosacral ligaments

Laparoscopic uterosacral ligament suspension incorporating both uterosacral ligaments and cervix or vaginal cuff.

Reprinted with permission of The Cleveland Clinic Foundation.

Obliterative procedures

LeForte colpocleisis or colpectomy/vaginectomy are the simplest treatments for advanced prolapse in elderly women who are not—and will not be—sexually active.¹⁶

We prefer the LeForte colpocleisis, in which rectangular segments of the anterior and posterior vaginal walls are denuded of their epithelium, followed by approximation of the rectangles to one another.

Success rates exceed 95%, and safety is maintained if spinal anesthesia is used in conjunction with a high perineoplasty.

Dr. Biller reports no relevant financial relationships. Dr. Davila reports research support from AMS and Tyco/US Surgical. He also serves as a consultant to AMS, and as a speaker for AMS and Tyco/US Surgical.

The disturbing fact that epithelial ovarian cancer often recurs after clinical remission poses this challenge: How do we identify the women with subclinical disease who may benefit from additional consolidation therapy?

Given the inability of noninvasive studies such as computed tomography, magnetic resonance imaging, and positron emission tomography to reliably detect small-volume and microscopic disease, second-look laparotomy (SLL) is the only technique capable of confirming a complete pathologic response to therapy.

Ob/Gyns involved in care of women with advanced ovarian cancer face the challenge of weighing the benefits of SLL against the potential morbidities of invasive surgery. This article describes those benefits, surgical technique, the prognostic significance of findings, and the status of salvage and consolidation therapies.

What SLL conveys

“Second look laparotomy” has rather loosely described many secondary surgeries for ovarian cancer, but we adopt the more rigorous definition: “a systematic surgical reexploration in asymptomatic patients who have no clinical evidence of tumor following initial surgery and completion of a planned program of chemotherapy.”¹

Procedures to debulk recurrent or residual disease, relieve symptomatic tumor, or accomplish interval cytoreduction cannot be deemed second-look laparotomy.

Prognostic, therapeutic limitations complicate the decision

Although negative findings at SLL confer an improved prognosis, disease ultimately recurs in up to 60% of patients.^2,3 Moreover, despite intensive research, consistently effective consolidation and salvage regimens remain elusive.

SLL may provide some information about prognosis, but that information is far from certain. Because of the cost and morbidity inherent in SLL, routine use has largely been limited to patients in clinical trials, where findings may serve as a surrogate endpoint for investigational therapies.

For these reasons, we strongly recommend careful discussion of this complex decision with patients prior to surgery, in consultation with a gynecologic oncologist.

Which patients are and are not candidates?

Candidates should be in clinical remission as determined by physical examination, abdominopelvic imaging, and serum CA-125 determination. Although SLL will detect residual disease in up to 50% of patients undergoing the procedure after primary chemotherapy, SLL is an imperfect method of determining the true response to therapy. Thus, it should be offered only to patients for whom results will influence clinical decision-making.

Patients with stage I disease treated with appropriate chemotherapy should not undergo SLL because of the low incidence of positive findings.⁴

Residual disease: 30% to 50%

Second-look laparotomy requires thorough inspection of the peritoneal cavity and retroperitoneum, but when properly performed on appropriate candidates, SLL detects residual disease in 30% to 50% of patients.^2,5

Generally, stage and volume of residual disease at initial surgery are most closely correlated with findings. In a review of 31 series, patients with stage III and IV disease undergoing surgery had fewer negative SLLs (39% and 33%, respectively) than patients with stage I and II disease (81% and 69%, respectively).⁶ Similarly, in pooled data on 1,797 patients, 72% of those with no gross residual disease at the conclusion of primary surgery had negative findings at SLL, compared with 50% of those with optimal residual, and only 23% of those with suboptimal residual.⁶

Surgical technique

Surgery begins with a large vertical incision and involves the components listed in the TABLE.

If gross disease is apparent:

Consider surgical cytoreduction, which is typically performed at the surgeon’s discretion.

In the absence of gross tumor:

Use a 5-point strategy to search thoroughly for occult disease.

• Take washings for cytology from the abdomen and pelvis;
  
• lyse any adhesions to allow adequate examination of all peritoneal surfaces;
  
• obtain random biopsies from the pelvis, bladder serosa, vaginal cuff, culde-sac, paracolic gutters, and hemidiaphragms, as well as adhesions, sites of prior documented tumor, infundibulopelvic ligament pedicles, and areas suspicious for tumor recurrence;
  
• consider removing the uterus, adnexae, omentum, and appendix, if not done at the primary surgery; and
  
• sample any remaining pelvic and paraaortic lymph nodes.
  

Meticulous sampling is crucial

 

Although the number of biopsies performed at SLL varies widely by surgeon, meticulous sampling of peritoneal surfaces may be necessary to detect occult tumor. In cases of microscopic disease, fewer than 5% of biopsies may be positive for tumor.⁷ Not surprisingly, some studies have noted a significantly worse survival rate among patients deemed to be in complete pathologic remission who underwent fewer biopsies at the time of SLL.⁸

TABLE

Components of second-look laparotomy

Vertical incision
Thorough inspection of abdomen and pelvis
Abdominopelvic washings for cytologic analysis
Complete adhesiolysis
Systematic biopsy of:
Undersurfaces of bilateral hemidiaphragms Paracolic gutters Pelvic peritoneum Pedicles of ovarian vessels Any suspicious lesions Areas of known prior tumor Adhesions
If necessary: Complete hysterectomy, salpingo-oophorectomy, omentectomy, and appendectomy
Pelvic and paraaortic lymph node sampling

What do SLL findings predict?

Survival rates

Women who achieve a complete pathologic response after primary chemotherapy have the greatest survival. Rubin et al³ noted a 10-year survival rate of 51% for 91 patients after negative SLL. Median survivals for patients with negative findings have been reported in excess of 70 months.^7,9 Tuxen et al⁹ examined 242 patients after SLL and reported a median survival of 149 months for those with negative findings, versus 39 and 24 months for those with microscopic and gross disease, respectively (P<.0005>

Survival rates among women with negative findings were substantially higher than among patients with positive findings, even though the latter group received salvage chemotherapy.

Recurrence rates

In a review of 38 studies encompassing 1,511 patients, Barter and Barnes⁶ noted a 23% rate of recurrence among women with negative findings at SLL. Other studies from single institutions document recurrence rates approximating 50%.^3,10 Patients experiencing recurrence after negative SLL have median survivals of 11 to 45 months.^3,8,10-12

Gross versus microscopic disease

Studies comparing outcomes based on volume of disease detected at SLL have found statistically improved survival for patients without evidence of gross tumor.^9,13-15 Podratz et al¹⁴ reported 4-year survival of 55% for women with microscopic findings versus 19% for those with gross disease (P<.01>

The presence of gross tumor at SLL indicates a grave prognosis; median survival ranges between 13 and 24 months.^9,16,17 Nevertheless, several studies have shown that patients able to undergo debulking of all visible disease derive a survival benefit.^4,15,18,19

Given the potential complications of extensive debulking surgery and lack of a proven survival benefit for patients unable to achieve complete cytoreduction, debulking should only be attempted if persistent disease is judged to be completely resectable.

When SLL is positive: Salvage therapy regimens

Many different salvage regimens for epithelial ovarian cancer have been investigated for use after positive second-look laparotomy, including intraperitoneal radioactive phosphorus (³²P), systemic chemotherapy, whole abdominal radiation (WAR), hormonal therapy, and biologic response modifiers. Unfortunately, studies of salvage therapy tend to be retrospective, nonrandomized, and uncontrolled, and no proven regimen has yet been found.

Whole abdominal radiation

This modality appears to confer no definitive survival benefit and does produce toxicity. MacGibbon et al²⁰ treated 51 patients with WAR for both salvage and consolidation. Of these, 27% could not complete treatment because of progressive disease, bowel perforation, myelosuppression, and bowel toxicity. An additional 24% required treatment delays because of hematologic and gastrointestinal toxicity. Among those completing the prescribed course of radiation, 6 developed late bowel symptoms, and 2 of these required surgical intervention to relieve bowel obstruction.

Other salvage therapies

Recently, Dowdy et al¹⁸ reported long-term follow-up for 145 patients with positive findings at SLL. Neither intraperitoneal ³²P nor WAR provided a survival benefit. Multivariate analysis indicated that only grade and volume of residual disease following cytoreduction were associated with improved survival.

Other trials involving intraperitoneal interferon-alpha and carboplatin,²¹ and high-dose chemotherapy with autologous stem cell rescue²² have also failed to demonstrate any significant advantages in survival or rate of progression. An early phase II study of intraperitoneal paclitaxel showed promise: Markman et al²³ noted a complete pathologic response in 61% of patients with microscopic disease at SLL, but only 4% of those with gross disease achieved a complete response.

Need for effective consolidation therapy

A critical component of cancer care is targeting patients at highest risk of recurrence for effective consolidation therapy. The factors most strongly correlated with disease progression are stage at diagnosis, tumor grade, and volume of residual disease after initial cytoreduction.

Many consolidation therapies have been described, including systemic and intraperitoneal chemotherapy, WAR, intraperitoneal ³²P, and biologic response modifiers.

Significant risk of distant recurrences

Although most tumors recur in the abdomen and pelvis, approximately 30% recur at distant locations. For this reason, consider systemic treatment when planning the consolidation regimen. Bertucci et al²² studied systemic melphalan-based, high-dose chemotherapy with autologous stem cell rescue and noted a 5-year progression-free survival of 43% and overall survival of 75%.

 

Most therapies are localized. Most other studies have focused on therapies localized to the peritoneal cavity. Results for intraperitoneal ³²P and WAR are mixed. Several different regimens of intraperitoneal chemotherapy have produced median disease-free survival rates of 18 to 41 months.^24-27 In 1998, Barakat et al²⁷ examined the use of intraperitoneal cisplatin and etoposide, reporting a statistically improved median disease-free survival for patients receiving intraperitoneal consolidation (median disease-free survival not yet reached), compared with patients treated by observation (28.5 months).

Bottom line

Second-look laparotomy reveals that approximately 50% of patients with a complete clinical response still harbor residual disease after primary chemotherapy. Even women who achieve a complete pathologic response have recurrence rates as high as 60%. While SLL can be a useful tool, the information it yields must be weighed against the potential morbidities of invasive surgery. Given its limited prognostic value, SLL should be offered only when results will influence clinical decision-making, or as part of a clinical trial.

The authors report no financial relationships relevant to this article.

The disturbing fact that epithelial ovarian cancer often recurs after clinical remission poses this challenge: How do we identify the women with subclinical disease who may benefit from additional consolidation therapy?

Given the inability of noninvasive studies such as computed tomography, magnetic resonance imaging, and positron emission tomography to reliably detect small-volume and microscopic disease, second-look laparotomy (SLL) is the only technique capable of confirming a complete pathologic response to therapy.

Ob/Gyns involved in care of women with advanced ovarian cancer face the challenge of weighing the benefits of SLL against the potential morbidities of invasive surgery. This article describes those benefits, surgical technique, the prognostic significance of findings, and the status of salvage and consolidation therapies.

What SLL conveys

“Second look laparotomy” has rather loosely described many secondary surgeries for ovarian cancer, but we adopt the more rigorous definition: “a systematic surgical reexploration in asymptomatic patients who have no clinical evidence of tumor following initial surgery and completion of a planned program of chemotherapy.”¹

Procedures to debulk recurrent or residual disease, relieve symptomatic tumor, or accomplish interval cytoreduction cannot be deemed second-look laparotomy.

Prognostic, therapeutic limitations complicate the decision

Although negative findings at SLL confer an improved prognosis, disease ultimately recurs in up to 60% of patients.^2,3 Moreover, despite intensive research, consistently effective consolidation and salvage regimens remain elusive.

SLL may provide some information about prognosis, but that information is far from certain. Because of the cost and morbidity inherent in SLL, routine use has largely been limited to patients in clinical trials, where findings may serve as a surrogate endpoint for investigational therapies.

For these reasons, we strongly recommend careful discussion of this complex decision with patients prior to surgery, in consultation with a gynecologic oncologist.

Which patients are and are not candidates?

Candidates should be in clinical remission as determined by physical examination, abdominopelvic imaging, and serum CA-125 determination. Although SLL will detect residual disease in up to 50% of patients undergoing the procedure after primary chemotherapy, SLL is an imperfect method of determining the true response to therapy. Thus, it should be offered only to patients for whom results will influence clinical decision-making.

Patients with stage I disease treated with appropriate chemotherapy should not undergo SLL because of the low incidence of positive findings.⁴

Residual disease: 30% to 50%

Second-look laparotomy requires thorough inspection of the peritoneal cavity and retroperitoneum, but when properly performed on appropriate candidates, SLL detects residual disease in 30% to 50% of patients.^2,5

Generally, stage and volume of residual disease at initial surgery are most closely correlated with findings. In a review of 31 series, patients with stage III and IV disease undergoing surgery had fewer negative SLLs (39% and 33%, respectively) than patients with stage I and II disease (81% and 69%, respectively).⁶ Similarly, in pooled data on 1,797 patients, 72% of those with no gross residual disease at the conclusion of primary surgery had negative findings at SLL, compared with 50% of those with optimal residual, and only 23% of those with suboptimal residual.⁶

Surgical technique

Surgery begins with a large vertical incision and involves the components listed in the TABLE.

If gross disease is apparent:

Consider surgical cytoreduction, which is typically performed at the surgeon’s discretion.

In the absence of gross tumor:

Use a 5-point strategy to search thoroughly for occult disease.

• Take washings for cytology from the abdomen and pelvis;
  
• lyse any adhesions to allow adequate examination of all peritoneal surfaces;
  
• obtain random biopsies from the pelvis, bladder serosa, vaginal cuff, culde-sac, paracolic gutters, and hemidiaphragms, as well as adhesions, sites of prior documented tumor, infundibulopelvic ligament pedicles, and areas suspicious for tumor recurrence;
  
• consider removing the uterus, adnexae, omentum, and appendix, if not done at the primary surgery; and
  
• sample any remaining pelvic and paraaortic lymph nodes.
  

Meticulous sampling is crucial

 

Although the number of biopsies performed at SLL varies widely by surgeon, meticulous sampling of peritoneal surfaces may be necessary to detect occult tumor. In cases of microscopic disease, fewer than 5% of biopsies may be positive for tumor.⁷ Not surprisingly, some studies have noted a significantly worse survival rate among patients deemed to be in complete pathologic remission who underwent fewer biopsies at the time of SLL.⁸

TABLE

Components of second-look laparotomy

Vertical incision
Thorough inspection of abdomen and pelvis
Abdominopelvic washings for cytologic analysis
Complete adhesiolysis
Systematic biopsy of:
Undersurfaces of bilateral hemidiaphragms Paracolic gutters Pelvic peritoneum Pedicles of ovarian vessels Any suspicious lesions Areas of known prior tumor Adhesions
If necessary: Complete hysterectomy, salpingo-oophorectomy, omentectomy, and appendectomy
Pelvic and paraaortic lymph node sampling

What do SLL findings predict?

Survival rates

Women who achieve a complete pathologic response after primary chemotherapy have the greatest survival. Rubin et al³ noted a 10-year survival rate of 51% for 91 patients after negative SLL. Median survivals for patients with negative findings have been reported in excess of 70 months.^7,9 Tuxen et al⁹ examined 242 patients after SLL and reported a median survival of 149 months for those with negative findings, versus 39 and 24 months for those with microscopic and gross disease, respectively (P<.0005>

Survival rates among women with negative findings were substantially higher than among patients with positive findings, even though the latter group received salvage chemotherapy.

Recurrence rates

In a review of 38 studies encompassing 1,511 patients, Barter and Barnes⁶ noted a 23% rate of recurrence among women with negative findings at SLL. Other studies from single institutions document recurrence rates approximating 50%.^3,10 Patients experiencing recurrence after negative SLL have median survivals of 11 to 45 months.^3,8,10-12

Gross versus microscopic disease

Studies comparing outcomes based on volume of disease detected at SLL have found statistically improved survival for patients without evidence of gross tumor.^9,13-15 Podratz et al¹⁴ reported 4-year survival of 55% for women with microscopic findings versus 19% for those with gross disease (P<.01>

The presence of gross tumor at SLL indicates a grave prognosis; median survival ranges between 13 and 24 months.^9,16,17 Nevertheless, several studies have shown that patients able to undergo debulking of all visible disease derive a survival benefit.^4,15,18,19

Given the potential complications of extensive debulking surgery and lack of a proven survival benefit for patients unable to achieve complete cytoreduction, debulking should only be attempted if persistent disease is judged to be completely resectable.

When SLL is positive: Salvage therapy regimens

Many different salvage regimens for epithelial ovarian cancer have been investigated for use after positive second-look laparotomy, including intraperitoneal radioactive phosphorus (³²P), systemic chemotherapy, whole abdominal radiation (WAR), hormonal therapy, and biologic response modifiers. Unfortunately, studies of salvage therapy tend to be retrospective, nonrandomized, and uncontrolled, and no proven regimen has yet been found.

Whole abdominal radiation

This modality appears to confer no definitive survival benefit and does produce toxicity. MacGibbon et al²⁰ treated 51 patients with WAR for both salvage and consolidation. Of these, 27% could not complete treatment because of progressive disease, bowel perforation, myelosuppression, and bowel toxicity. An additional 24% required treatment delays because of hematologic and gastrointestinal toxicity. Among those completing the prescribed course of radiation, 6 developed late bowel symptoms, and 2 of these required surgical intervention to relieve bowel obstruction.

Other salvage therapies

Recently, Dowdy et al¹⁸ reported long-term follow-up for 145 patients with positive findings at SLL. Neither intraperitoneal ³²P nor WAR provided a survival benefit. Multivariate analysis indicated that only grade and volume of residual disease following cytoreduction were associated with improved survival.

Other trials involving intraperitoneal interferon-alpha and carboplatin,²¹ and high-dose chemotherapy with autologous stem cell rescue²² have also failed to demonstrate any significant advantages in survival or rate of progression. An early phase II study of intraperitoneal paclitaxel showed promise: Markman et al²³ noted a complete pathologic response in 61% of patients with microscopic disease at SLL, but only 4% of those with gross disease achieved a complete response.

Need for effective consolidation therapy

A critical component of cancer care is targeting patients at highest risk of recurrence for effective consolidation therapy. The factors most strongly correlated with disease progression are stage at diagnosis, tumor grade, and volume of residual disease after initial cytoreduction.

Many consolidation therapies have been described, including systemic and intraperitoneal chemotherapy, WAR, intraperitoneal ³²P, and biologic response modifiers.

Significant risk of distant recurrences

Although most tumors recur in the abdomen and pelvis, approximately 30% recur at distant locations. For this reason, consider systemic treatment when planning the consolidation regimen. Bertucci et al²² studied systemic melphalan-based, high-dose chemotherapy with autologous stem cell rescue and noted a 5-year progression-free survival of 43% and overall survival of 75%.

 

Most therapies are localized. Most other studies have focused on therapies localized to the peritoneal cavity. Results for intraperitoneal ³²P and WAR are mixed. Several different regimens of intraperitoneal chemotherapy have produced median disease-free survival rates of 18 to 41 months.^24-27 In 1998, Barakat et al²⁷ examined the use of intraperitoneal cisplatin and etoposide, reporting a statistically improved median disease-free survival for patients receiving intraperitoneal consolidation (median disease-free survival not yet reached), compared with patients treated by observation (28.5 months).

Bottom line

Second-look laparotomy reveals that approximately 50% of patients with a complete clinical response still harbor residual disease after primary chemotherapy. Even women who achieve a complete pathologic response have recurrence rates as high as 60%. While SLL can be a useful tool, the information it yields must be weighed against the potential morbidities of invasive surgery. Given its limited prognostic value, SLL should be offered only when results will influence clinical decision-making, or as part of a clinical trial.

The authors report no financial relationships relevant to this article.

The disturbing fact that epithelial ovarian cancer often recurs after clinical remission poses this challenge: How do we identify the women with subclinical disease who may benefit from additional consolidation therapy?

Given the inability of noninvasive studies such as computed tomography, magnetic resonance imaging, and positron emission tomography to reliably detect small-volume and microscopic disease, second-look laparotomy (SLL) is the only technique capable of confirming a complete pathologic response to therapy.

Ob/Gyns involved in care of women with advanced ovarian cancer face the challenge of weighing the benefits of SLL against the potential morbidities of invasive surgery. This article describes those benefits, surgical technique, the prognostic significance of findings, and the status of salvage and consolidation therapies.

What SLL conveys

“Second look laparotomy” has rather loosely described many secondary surgeries for ovarian cancer, but we adopt the more rigorous definition: “a systematic surgical reexploration in asymptomatic patients who have no clinical evidence of tumor following initial surgery and completion of a planned program of chemotherapy.”¹

Procedures to debulk recurrent or residual disease, relieve symptomatic tumor, or accomplish interval cytoreduction cannot be deemed second-look laparotomy.

Prognostic, therapeutic limitations complicate the decision

Although negative findings at SLL confer an improved prognosis, disease ultimately recurs in up to 60% of patients.^2,3 Moreover, despite intensive research, consistently effective consolidation and salvage regimens remain elusive.

SLL may provide some information about prognosis, but that information is far from certain. Because of the cost and morbidity inherent in SLL, routine use has largely been limited to patients in clinical trials, where findings may serve as a surrogate endpoint for investigational therapies.

For these reasons, we strongly recommend careful discussion of this complex decision with patients prior to surgery, in consultation with a gynecologic oncologist.

Which patients are and are not candidates?

Candidates should be in clinical remission as determined by physical examination, abdominopelvic imaging, and serum CA-125 determination. Although SLL will detect residual disease in up to 50% of patients undergoing the procedure after primary chemotherapy, SLL is an imperfect method of determining the true response to therapy. Thus, it should be offered only to patients for whom results will influence clinical decision-making.

Patients with stage I disease treated with appropriate chemotherapy should not undergo SLL because of the low incidence of positive findings.⁴

Residual disease: 30% to 50%

Second-look laparotomy requires thorough inspection of the peritoneal cavity and retroperitoneum, but when properly performed on appropriate candidates, SLL detects residual disease in 30% to 50% of patients.^2,5

Generally, stage and volume of residual disease at initial surgery are most closely correlated with findings. In a review of 31 series, patients with stage III and IV disease undergoing surgery had fewer negative SLLs (39% and 33%, respectively) than patients with stage I and II disease (81% and 69%, respectively).⁶ Similarly, in pooled data on 1,797 patients, 72% of those with no gross residual disease at the conclusion of primary surgery had negative findings at SLL, compared with 50% of those with optimal residual, and only 23% of those with suboptimal residual.⁶

Surgical technique

Surgery begins with a large vertical incision and involves the components listed in the TABLE.

If gross disease is apparent:

Consider surgical cytoreduction, which is typically performed at the surgeon’s discretion.

In the absence of gross tumor:

Use a 5-point strategy to search thoroughly for occult disease.

• Take washings for cytology from the abdomen and pelvis;
  
• lyse any adhesions to allow adequate examination of all peritoneal surfaces;
  
• obtain random biopsies from the pelvis, bladder serosa, vaginal cuff, culde-sac, paracolic gutters, and hemidiaphragms, as well as adhesions, sites of prior documented tumor, infundibulopelvic ligament pedicles, and areas suspicious for tumor recurrence;
  
• consider removing the uterus, adnexae, omentum, and appendix, if not done at the primary surgery; and
  
• sample any remaining pelvic and paraaortic lymph nodes.
  

Meticulous sampling is crucial

 

Although the number of biopsies performed at SLL varies widely by surgeon, meticulous sampling of peritoneal surfaces may be necessary to detect occult tumor. In cases of microscopic disease, fewer than 5% of biopsies may be positive for tumor.⁷ Not surprisingly, some studies have noted a significantly worse survival rate among patients deemed to be in complete pathologic remission who underwent fewer biopsies at the time of SLL.⁸

TABLE

Components of second-look laparotomy

Vertical incision
Thorough inspection of abdomen and pelvis
Abdominopelvic washings for cytologic analysis
Complete adhesiolysis
Systematic biopsy of:
Undersurfaces of bilateral hemidiaphragms Paracolic gutters Pelvic peritoneum Pedicles of ovarian vessels Any suspicious lesions Areas of known prior tumor Adhesions
If necessary: Complete hysterectomy, salpingo-oophorectomy, omentectomy, and appendectomy
Pelvic and paraaortic lymph node sampling

What do SLL findings predict?

Survival rates

Women who achieve a complete pathologic response after primary chemotherapy have the greatest survival. Rubin et al³ noted a 10-year survival rate of 51% for 91 patients after negative SLL. Median survivals for patients with negative findings have been reported in excess of 70 months.^7,9 Tuxen et al⁹ examined 242 patients after SLL and reported a median survival of 149 months for those with negative findings, versus 39 and 24 months for those with microscopic and gross disease, respectively (P<.0005>

Survival rates among women with negative findings were substantially higher than among patients with positive findings, even though the latter group received salvage chemotherapy.

Recurrence rates

In a review of 38 studies encompassing 1,511 patients, Barter and Barnes⁶ noted a 23% rate of recurrence among women with negative findings at SLL. Other studies from single institutions document recurrence rates approximating 50%.^3,10 Patients experiencing recurrence after negative SLL have median survivals of 11 to 45 months.^3,8,10-12

Gross versus microscopic disease

Studies comparing outcomes based on volume of disease detected at SLL have found statistically improved survival for patients without evidence of gross tumor.^9,13-15 Podratz et al¹⁴ reported 4-year survival of 55% for women with microscopic findings versus 19% for those with gross disease (P<.01>

The presence of gross tumor at SLL indicates a grave prognosis; median survival ranges between 13 and 24 months.^9,16,17 Nevertheless, several studies have shown that patients able to undergo debulking of all visible disease derive a survival benefit.^4,15,18,19

Given the potential complications of extensive debulking surgery and lack of a proven survival benefit for patients unable to achieve complete cytoreduction, debulking should only be attempted if persistent disease is judged to be completely resectable.

When SLL is positive: Salvage therapy regimens

Many different salvage regimens for epithelial ovarian cancer have been investigated for use after positive second-look laparotomy, including intraperitoneal radioactive phosphorus (³²P), systemic chemotherapy, whole abdominal radiation (WAR), hormonal therapy, and biologic response modifiers. Unfortunately, studies of salvage therapy tend to be retrospective, nonrandomized, and uncontrolled, and no proven regimen has yet been found.

Whole abdominal radiation

This modality appears to confer no definitive survival benefit and does produce toxicity. MacGibbon et al²⁰ treated 51 patients with WAR for both salvage and consolidation. Of these, 27% could not complete treatment because of progressive disease, bowel perforation, myelosuppression, and bowel toxicity. An additional 24% required treatment delays because of hematologic and gastrointestinal toxicity. Among those completing the prescribed course of radiation, 6 developed late bowel symptoms, and 2 of these required surgical intervention to relieve bowel obstruction.

Other salvage therapies

Recently, Dowdy et al¹⁸ reported long-term follow-up for 145 patients with positive findings at SLL. Neither intraperitoneal ³²P nor WAR provided a survival benefit. Multivariate analysis indicated that only grade and volume of residual disease following cytoreduction were associated with improved survival.

Other trials involving intraperitoneal interferon-alpha and carboplatin,²¹ and high-dose chemotherapy with autologous stem cell rescue²² have also failed to demonstrate any significant advantages in survival or rate of progression. An early phase II study of intraperitoneal paclitaxel showed promise: Markman et al²³ noted a complete pathologic response in 61% of patients with microscopic disease at SLL, but only 4% of those with gross disease achieved a complete response.

Need for effective consolidation therapy

A critical component of cancer care is targeting patients at highest risk of recurrence for effective consolidation therapy. The factors most strongly correlated with disease progression are stage at diagnosis, tumor grade, and volume of residual disease after initial cytoreduction.

Many consolidation therapies have been described, including systemic and intraperitoneal chemotherapy, WAR, intraperitoneal ³²P, and biologic response modifiers.

Significant risk of distant recurrences

Although most tumors recur in the abdomen and pelvis, approximately 30% recur at distant locations. For this reason, consider systemic treatment when planning the consolidation regimen. Bertucci et al²² studied systemic melphalan-based, high-dose chemotherapy with autologous stem cell rescue and noted a 5-year progression-free survival of 43% and overall survival of 75%.

 

Most therapies are localized. Most other studies have focused on therapies localized to the peritoneal cavity. Results for intraperitoneal ³²P and WAR are mixed. Several different regimens of intraperitoneal chemotherapy have produced median disease-free survival rates of 18 to 41 months.^24-27 In 1998, Barakat et al²⁷ examined the use of intraperitoneal cisplatin and etoposide, reporting a statistically improved median disease-free survival for patients receiving intraperitoneal consolidation (median disease-free survival not yet reached), compared with patients treated by observation (28.5 months).

Bottom line

Second-look laparotomy reveals that approximately 50% of patients with a complete clinical response still harbor residual disease after primary chemotherapy. Even women who achieve a complete pathologic response have recurrence rates as high as 60%. While SLL can be a useful tool, the information it yields must be weighed against the potential morbidities of invasive surgery. Given its limited prognostic value, SLL should be offered only when results will influence clinical decision-making, or as part of a clinical trial.

The authors report no financial relationships relevant to this article.

After several decades of slow progress, the field of urogynecology is experiencing dynamic change, including:

• new minimally invasive, tension-free, midurethral sling procedures, especially the transobturator approach,
• correction of size-specific defects to repair prolapse, and
• use of mesh/graft augmentation in pro lapse repair.

These developments are some of the most important since Kelly and Dunn first described suburethral fascial plication for stress incontinence and cystocele in 1914.¹ They have come about through increased understanding of the pathophysiology of incontinence and prolapse, innovative technology and techniques, and improved communication and coordination among physicians worldwide.

New sling procedures, promising outcomes

The minimally invasive midurethral sling procedure spawned notable new approaches and is a mainstay of surgical treatment for stress urinary incontinence.

First described in Sweden by Ulmsten in 1995,² the tension-free vaginal tape procedure is a revolutionary change in the suburethral sling procedure and is now the most widely performed surgery for stress incontinence worldwide.

In it, a tension-free vaginal tape (TVT) (Gynecare, Somerville, NJ) of synthetic polypropylene mesh is attached to 2 needles and passed through a vaginal incision and the retropubic space, exiting to small incisions in the suprapubic region to create a suburethral sling or hammock and provide urethral support during increased abdominal pressure. The sling remains fixed by friction and subsequent adhesions.

Although the traditional suburethral sling was less invasive than other abdominal incontinence procedures, it was associated with a steep learning curve and a high incidence of postoperative irritative bladder symptoms and voiding dysfunction.³

Tension-free vaginal tape: Excellent long-term cure

It can be performed routinely in under 30 minutes using local anesthesia, with minimal postoperative complications. Five-year cure rates approach 95%,⁴ and data presented at the 2003 International Urogynecology Association clinical meeting describe an objective 7-year cure rate of 82%.⁵

The incidence and severity of postoperative voiding dysfunction following the TVT procedure is significantly lower than that reported after traditional suburethral sling (2%–40%) or transvaginal needle suspension (2%–50%) procedures.^6-9 Although bladder perforation has occurred in up to 10% of patients, reports of complications, including major hemorrhage, tape erosion, and bowel and nerve injury, are rare.^10,11

Modifications

These products have a modified approach, materials, or refinements in technique to address various needs. For example, American Medical Systems (Minnetonka, Minn) introduced the SPARC procedure, which allows abdominal placement of the midurethral sling, similar to a needle suspension technique.

CR Bard (Murray Hill, NJ) introduced a midurethral sling of porcine dermis (Pelvicol) to address concerns physicians may have about using synthetic materials.

Minimally invasive midurethral slings include the Advantage (Boston Scientific, Natick, Mass), Centrasorb (Caldera Medical, Thousand Oaks, Calif), Stratasis TF (Cook, West Lafayette, Ind), and Uretex (CR Bard).

Newest approach: Transobturator sling

In this technique, the sling is placed in the midurethral position, but the insertion points are in the genital area lateral to the vagina, and the needle passes through the obturator membrane and paraurethral space. Because it avoids passage of the needles through the retropubic space, the transobturator approach theoretically should reduce the risk of bowel, bladder, and major blood vessel injury.

The procedure was initially described in Europe and introduced in the United States in 2003. Current product offerings include the outside-in approach of ObTape (Mentor, Santa Barbara, Calif), Monarc (American Medical Systems), and Uretex (CR Bard). Gynecare offers a variation of its TVT—the TVT-Obturator—which involves an inside-out approach to further minimize risk of vascular injury.

Shortage of long-term data, but good early results

In a 1-year follow-up of patients undergoing a sling procedure with the UraTape transobturator sling (Mentor), Delorme and colleagues¹² reported that 29 of 32 patients (90.6%) were cured and 3 (9.4%) improved. De Leval¹³ described his inside-out approach with equally good results: no bladder or urethral injuries and no vascular (hematoma or bleeding) or neurological complications. A transobturator technique using porcine dermis has also been described.

Tension-free vaginal tape versus transobturator sling

In 2004, a small randomized, prospective trial of TVT (n = 29) versus transobturator tape (n = 27) with 1-year follow-up found the transobturator approach to be safer and easier to place with equivalent short-term results.¹⁴ Mean operative time was significantly shorter in the transobturator group (15±4 minutes versus 27±8 minutes, P <.001).

No bladder injury occurred in the transobturator group versus 9.7% (n = 3) in the TVT group (P >.05). The rate of postoperative urinary retention was 25.8% (n = 8) in the TVT group versus 13.3% (n = 4) in the transobturator group (P >.05). Cure rates were similar for the TVT and transobturator groups: 83.9% versus 90%, respectively), improvement (9.7% versus 3.3%), and failure (6.5% versus 6.7%). No vaginal erosion occurred in either group.

 

Additional investigations of the transobturator tape procedure are underway.

Correcting site-specific defects in prolapse repair

This repair rationale should become the standard, although it has yet to be widely adopted and procedural refinement and research are continuing.

As early as 1908, site-specific defects in the endopelvic fascia were identified as the likely cause of anterior vaginal segment prolapse. Like hernia repair, which requires closure of the fascial defect, the “cystocele hernia” repair advocated by George White involved reattaching the endopelvic fascia to the arcus tendineus fascia pelvis using a series of interrupted sutures through an abdominal retropubic approach.

Although this view was later abandoned, it resurfaced in the 1980s, when Cullen Richardson described midline, lateral, and transverse defects (FIGURE 1) in the endopelvic fascia as the cause of cystocele and rectocele.¹⁵ Richardson advocated diagnosis that identified these fascial defects, along with treatment with site-specific repair.

(In the intervening decades, prolapse was thought to result from generalized weakening or attenuation of the endopelvic fascia that supports the bladder, rectum, or vagina, leading to cystocele, rectocele, or uterine prolapse, respectively. Traditional repairs, still widely performed by most gynecologists, consist of the anterior and posterior colporrhaphy, which involve midline plication of the endopelvic fascia to reduce the prolapse and recreate support by strengthening the weakened fascial layer.)

FIGURE 1 Identify the site of defect

Whether the defect that causes a rectocele or cystocele is transverse, central, or lateral determines the type of repair to be performed. Reprinted with permission from Richardson AC. The rectovaginal septum revisited: its relationship to rectocele and its importance in rectocele repair. Clin Obstet Gynecol.1993;36(4):976–983.

Excellent cure rates with fewer complications

Site-specific repair (FIGURE 2) has been adopted by many pelvic surgeons, who report excellent cure rates with fewer complications such as dyspareunia, vaginal narrowing, and increased blood loss—all of which are more common with traditional anterior and posterior colporrhaphy.

Traditional colporrhaphy does correct the underlying fascial defect when the “hernia” is in the midline. However, for lateral and transverse defects, traditional colporrhaphy leaves the defect uncorrected and may even create additional tension, resulting in recurrence.

Site-specific repair also is useful in the treatment of enterocele, which has been described as a herniation of the peritoneum through a defect of the anterior and posterior fascial planes at the vaginal apex.¹⁶

Widening but not broad acceptance

Although cure rates for site-specific prolapse repair range from 75% to 85%,^17,18 the concept has yet to be widely adopted and continues to undergo procedural refinement and research, as well as physician education. With increasing experience and clinical data, it may become the standard for pelvic prolapse repair.

FIGURE 2 Repair the defect to correct the rectocele

Surgical technique for a transverse tear in the rectovaginal fascia: After identifying the defect, place a series of interrupted sutures to reapproximate the fascial edges (copyright Miklos/Kohli).

Mesh augmentation: Useful in selected patients

Although augmentation with mesh or biomaterials is easy to perform, it remains unclear which technique and materials are optimal.

The use of mesh and biomaterials to augment repair of a cystocele, rectocele, or enterocele is slowly increasing, although general surgeons have been utilizing these materials for many years in hernia repair.

Augmentation has been advocated for “pelvic hernias” because of poor long-term cure rates for traditional prolapse surgeries (which range from 40% to 80%—well behind rates of 90% or more for incontinence procedures).

A recent informal survey of the members of the American Urogynecologic Society and International Urogynecologic Association revealed that most pelvic surgeons are using some type of graft or mesh to augment repairs in selected patients.

Synthetic mesh versus autologous and heterologous grafts

Synthetic materials have the advantage of being readily available, cost-effective, and consistent in quality, but they may cause significant complications, including infection, stricture, and erosion. This is especially important in the vagina, which needs to stretch during intercourse and which alters in thickness and other properties during a woman’s lifetime.

In contrast, autologous and heterologous donor grafts provide naturally occurring biomaterials capable of remodeling. Unfortunately, the in vivo tissue response is not yet fully understood. Other disadvantages: Biomaterials may lack consistent tissue properties and can be expensive.

For these reasons, graft materials remain in an early period of evaluation, although their use is expected to rise steadily with increasing experience and new product development.

Limited data on safety and efficacy

Although many pelvic surgeons use graft/mesh materials in prolapse repair, data on their safety and efficacy are limited, partly due to the variety of surgical techniques and materials available. Another factor is the difficulty of obtaining good long-term data with large patient numbers. Most of the literature is comprised of case reports, with few prospective, randomized trials.

 

That said, initial data suggest a significant improvement in cure rates, compared with traditional techniques, with minimal short-term complications. Long-term results and complication rates are not yet available.

What existing studies show

A variety of synthetic materials have been used in surgical correction of cystocele. In the largest series to date, Flood and colleagues¹⁹ reported their 12-year experience with 142 women undergoing a modified anterior colporrhaphy reinforced with Marlex (Davol, Cranston, RI) mesh: 100% success in correcting cystocele and a 74% success rate for urinary stress incontinence, with a mean follow-up of 3.2 years and no significant intraoperative complications.

In a prospective randomized trial of 125 patients utilizing absorbable polyglactin 910 mesh (Vicryl) (Ethicon, Somerville, NJ) to augment standard anterior colporrhaphy, Koduri et al²⁰ reported a failure rate of 13% in the colporrhaphy-alone group compared to 1% in the colporrhaphy-mesh group at the 1-year follow-up. Subjectively, both groups improved equally.

Clemons et al²¹ used a human dermis graft to treat advanced recurrent cystocele in 33 women, with a follow-up of 18 months. They noted 13 (41%) objective failures and 1 (3%) subjective failure. Complications included 1 case of febrile morbidity, 1 cystotomy, and 1 anterior wall breakdown secondary to hematoma formation caused by heparin therapy. No other erosions or rejections were seen.

Most effective applications

Graft/mesh augmentation may be most effective in posterior vaginal segment reconstruction and rectocele repair, as it obviates the need for levatorplasty in patients with poor rectovaginal fascia. A variety of synthetic materials have been used for posterior wall reconstruction in small series.

Mersilene mesh

Mersilene mesh (Ethicon) was used by Fox and Stanton²² to augment traditional rectocele repair in 29 women followed for 14 months. Most of the women had undergone previous rectocele repair with recurrence of their prolapse. All women with stage II and stage III vault prolapse were corrected, with an increase in stage I prolapse from 20% to 27%. All women with stage II and stage III rectocele were corrected, with a decrease in stage I prolapse from 36% to 7%. The only significant intraoperative complication was a cystotomy. One mesh became infected postoperatively, requiring removal.

Cadaveric dermal graft

Recently, Kohli and Miklos²³ reported their experience with 57 patients undergoing augmented rectocele repair using a cadaveric dermal graft (FIGURE 3) over a 2-year period. Average follow-up was 11 months. Average patient age in the follow-up group was 63.6±10.9 years (range: 33–79 years) and average parity was 2.8±1.5 (range: 0–7). No major intraoperative complications (hollow viscous injury, blood loss greater than 500 cc, or transfusion) or post-operative complications (infection, abscess, or hematoma) were noted. No graft-related complications such as rejection, erosion, infection, and fistula formation were noted during the follow-up period.

Using previously accepted Pelvic Organ Prolapse Quantification (POP-Q) parameters for success, Kohli and Miklos found 54 of 57 women (95%) to have surgical cure at follow-up. These authors have also described the use of a dermal graft in the repair of a complicated rectovaginal fistula.²⁴

FIGURE 3 Placement of mesh augmentation

Augment enterocele/rectocele repair by attaching the mesh apically, laterally, and distally (copyright Miklos/Kohli).

Future outlook

Further data on the efficacy and safety of graft/mesh augmentation—including identification of the optimal technique and ideal material—are necessary before this approach can be widely adopted. However, in selected patients, it represents a significant advance in the surgical treatment of pelvic prolapse.

The author is a preceptor for American Medical Systems, Gynecare, CR Bard, and Mentor and a consultant for Boston Scientific.