PHILADELPHIA—Preclinical research suggests a cyclin-dependent kinase (CDK) inhibitor is active against acute leukemias, particularly those with mixed-lineage leukemia rearrangements (MLL-r).

CYC065 selectively inhibits CDK2, which drives cell-cycle transition and activates major DNA double-strand break repair pathways; CDK5, which drives metastatic spread; and CDK9, which regulates the transcription of genes important for the proliferation and survival of malignant cells.

Experiments have shown that CYC065 exhibits activity against acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL), with and without MLL-r.

Daniella Zheleva, PhD, and her colleagues described these experiments in a poster at the AACR Annual Meeting 2015 (abstract 1650). All of the researchers are employees of Cyclacel Ltd., the company developing CYC065.

The researchers tested CYC065 in a panel of AML cell lines with wild-type MLL (HEL, HL60, Kasumi-1, KG-1, OCI-AML5, and PL21) and MLL-r (EOL-1, ML-2, MOLM-13, MV4-11, Nomo-1, OCI-AML2, and THP-1).

They found that MLL-r cell lines were “highly sensitive” to CYC065, but the sensitivity of cell lines with wild-type MLL correlated with the level of Bcl-2 family proteins. In the wild-type cell lines, IC_50/70/90 values were correlated with Bcl_XL and inversely correlated with Bak.

Six-hour pulse treatment of CYC065 at 0.5 µM to 1 µM was sufficient to cause 90% or greater cell death in sensitive cell lines. And cell lines with reduced sensitivity to the drug could be targeted by exposure to 10-hour pulse treatments of CYC065, or to CYC065 in combination with short pulses of Bcl-2 inhibitors.

The researchers observed “potent antitumor activity” when they administered CYC065 in AML xenograft models.

In an EOL-1 model, the median tumor growth inhibition on day 19 was 97% for mice that received CYC065 at 40 mg/kg (every day on days 1-5 and 8-12), 95% for mice that received CYC065 at 20 mg/kg every day on days 1-5 and 8-12), and 41% for mice that received cytarabine at 100 mg/kg (every day on days 1-5).

In the HL60 model, the median tumor growth inhibition on day 11 was 90% for mice that received CYC065 at 70 mg/kg (every day on days 1-5 and 8-12). And 2 mice achieved a complete response to treatment.

The researchers also found that CYC065 synergizes with cytarabine, particularly when CYC065 is given first. In fact, the combination could overcome the cytarabine resistance observed in the MV4-11 cell line.

CYC065 was “strongly synergistic” with Bcl2/Bcl_XL inhibitors as well, the researchers said. CYC065 synergized with ABT-199, ABT-263, and ABT-737 in both AML cell lines (THP-1 and HEL) and ALL cell lines (Jurkat and SEM).

The researchers said the potent in vitro and in vivo activity of CYC065 and the ability to combine the drug with other antileukemic agents suggest that it may have therapeutic potential in AML and ALL.

PHILADELPHIA—Preclinical research suggests a cyclin-dependent kinase (CDK) inhibitor is active against acute leukemias, particularly those with mixed-lineage leukemia rearrangements (MLL-r).

CYC065 selectively inhibits CDK2, which drives cell-cycle transition and activates major DNA double-strand break repair pathways; CDK5, which drives metastatic spread; and CDK9, which regulates the transcription of genes important for the proliferation and survival of malignant cells.

Experiments have shown that CYC065 exhibits activity against acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL), with and without MLL-r.

Daniella Zheleva, PhD, and her colleagues described these experiments in a poster at the AACR Annual Meeting 2015 (abstract 1650). All of the researchers are employees of Cyclacel Ltd., the company developing CYC065.

The researchers tested CYC065 in a panel of AML cell lines with wild-type MLL (HEL, HL60, Kasumi-1, KG-1, OCI-AML5, and PL21) and MLL-r (EOL-1, ML-2, MOLM-13, MV4-11, Nomo-1, OCI-AML2, and THP-1).

They found that MLL-r cell lines were “highly sensitive” to CYC065, but the sensitivity of cell lines with wild-type MLL correlated with the level of Bcl-2 family proteins. In the wild-type cell lines, IC_50/70/90 values were correlated with Bcl_XL and inversely correlated with Bak.

Six-hour pulse treatment of CYC065 at 0.5 µM to 1 µM was sufficient to cause 90% or greater cell death in sensitive cell lines. And cell lines with reduced sensitivity to the drug could be targeted by exposure to 10-hour pulse treatments of CYC065, or to CYC065 in combination with short pulses of Bcl-2 inhibitors.

The researchers observed “potent antitumor activity” when they administered CYC065 in AML xenograft models.

In an EOL-1 model, the median tumor growth inhibition on day 19 was 97% for mice that received CYC065 at 40 mg/kg (every day on days 1-5 and 8-12), 95% for mice that received CYC065 at 20 mg/kg every day on days 1-5 and 8-12), and 41% for mice that received cytarabine at 100 mg/kg (every day on days 1-5).

In the HL60 model, the median tumor growth inhibition on day 11 was 90% for mice that received CYC065 at 70 mg/kg (every day on days 1-5 and 8-12). And 2 mice achieved a complete response to treatment.

The researchers also found that CYC065 synergizes with cytarabine, particularly when CYC065 is given first. In fact, the combination could overcome the cytarabine resistance observed in the MV4-11 cell line.

CYC065 was “strongly synergistic” with Bcl2/Bcl_XL inhibitors as well, the researchers said. CYC065 synergized with ABT-199, ABT-263, and ABT-737 in both AML cell lines (THP-1 and HEL) and ALL cell lines (Jurkat and SEM).

The researchers said the potent in vitro and in vivo activity of CYC065 and the ability to combine the drug with other antileukemic agents suggest that it may have therapeutic potential in AML and ALL.

PHILADELPHIA—Preclinical research suggests a cyclin-dependent kinase (CDK) inhibitor is active against acute leukemias, particularly those with mixed-lineage leukemia rearrangements (MLL-r).

CYC065 selectively inhibits CDK2, which drives cell-cycle transition and activates major DNA double-strand break repair pathways; CDK5, which drives metastatic spread; and CDK9, which regulates the transcription of genes important for the proliferation and survival of malignant cells.

Experiments have shown that CYC065 exhibits activity against acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL), with and without MLL-r.

Daniella Zheleva, PhD, and her colleagues described these experiments in a poster at the AACR Annual Meeting 2015 (abstract 1650). All of the researchers are employees of Cyclacel Ltd., the company developing CYC065.

The researchers tested CYC065 in a panel of AML cell lines with wild-type MLL (HEL, HL60, Kasumi-1, KG-1, OCI-AML5, and PL21) and MLL-r (EOL-1, ML-2, MOLM-13, MV4-11, Nomo-1, OCI-AML2, and THP-1).

They found that MLL-r cell lines were “highly sensitive” to CYC065, but the sensitivity of cell lines with wild-type MLL correlated with the level of Bcl-2 family proteins. In the wild-type cell lines, IC_50/70/90 values were correlated with Bcl_XL and inversely correlated with Bak.

Six-hour pulse treatment of CYC065 at 0.5 µM to 1 µM was sufficient to cause 90% or greater cell death in sensitive cell lines. And cell lines with reduced sensitivity to the drug could be targeted by exposure to 10-hour pulse treatments of CYC065, or to CYC065 in combination with short pulses of Bcl-2 inhibitors.

The researchers observed “potent antitumor activity” when they administered CYC065 in AML xenograft models.

In an EOL-1 model, the median tumor growth inhibition on day 19 was 97% for mice that received CYC065 at 40 mg/kg (every day on days 1-5 and 8-12), 95% for mice that received CYC065 at 20 mg/kg every day on days 1-5 and 8-12), and 41% for mice that received cytarabine at 100 mg/kg (every day on days 1-5).

In the HL60 model, the median tumor growth inhibition on day 11 was 90% for mice that received CYC065 at 70 mg/kg (every day on days 1-5 and 8-12). And 2 mice achieved a complete response to treatment.

The researchers also found that CYC065 synergizes with cytarabine, particularly when CYC065 is given first. In fact, the combination could overcome the cytarabine resistance observed in the MV4-11 cell line.

CYC065 was “strongly synergistic” with Bcl2/Bcl_XL inhibitors as well, the researchers said. CYC065 synergized with ABT-199, ABT-263, and ABT-737 in both AML cell lines (THP-1 and HEL) and ALL cell lines (Jurkat and SEM).

The researchers said the potent in vitro and in vivo activity of CYC065 and the ability to combine the drug with other antileukemic agents suggest that it may have therapeutic potential in AML and ALL.

Attendees at AACR 2015

PHILADELPHIA—A drug that has fallen short of expectations in clinical trials of prostate cancer may be effective for treating multiple myeloma (MM), according to research presented at the AACR Annual Meeting 2015.

The drug, tasquinimod, inhibits the function of S100A9, a pro-inflammatory protein that is elevated in MM, prostate cancer, and other malignancies.

Researchers found that tasquinimod can reduce tumor growth and improve survival in mouse models of MM. And these effects are associated with reduced angiogenesis in the bone marrow.

Cindy Lin, PhD, of the Wistar Institute in Philadelphia, Pennsylvania, and her colleagues presented these findings as abstract 1364.* The research was supported by Active Biotech and Ipsen, the companies developing tasquinimod.

Dr Lin noted that tasquinimod has already been tested in clinical trials of prostate cancer and initially appeared to be very effective. However, recent results from a phase 3 trial suggested the drug does not confer a favorable risk-benefit ratio for this population.

So Active Biotech and Ipsen decided to discontinue all trials of tasquinimod in prostate cancer. But the preclinical results observed in MM suggest tasquinimod may hold promise for treating these patients.

Activity in MM

Dr Lin said previous preclinical experiments revealed that myeloid-derived suppressor cells are involved in regulating MM progression, and these cells produce S100A9.

Tasquinimod is a quinoline-3-carboxamide derivative that binds to S100A9 and inhibits interaction with its receptors. So Dr Lin and her colleagues decided to investigate the antitumor effect of the drug in mouse models of MM.

The researchers initially tested tasquinimod in a syngeneic MM model, randomizing mice to treatment or control. Mice in the treatment group received tasquinimod at 30 mg/kg/day in their drinking water for 28 days.

Tasquinimod significantly improved survival in this model (P<0.005). All control mice died within 30 days of tumor inoculation, but about 40% of tasquinimod-treated mice were still alive more than 80 days out.

The researchers then tested tasquinimod in xenograft models of human MM. The drug significantly reduced tumor size in both H929 (P=0.0042) and RPMI-8226 models (P=0.0003). Treatment significantly improved survival in the H929 (P=0.0008) and RPMI-8226 models as well (P=0.0243).

Dr Lin said she and her colleagues did not see any side effects of treatment in any of the mice.

Investigating the mechanism

To determine if the antitumor effect of tasquinimod is, in fact, mediated through inhibition of S100A9, the researchers administered the drug to S100A9-knockout mice with MM. Tasquinimod did not improve survival in these mice, which suggests its anti-MM effects are mediated through S100A9 inhibition.

“To try and investigate some of the mechanisms of how survival is improved in tasquinimod-treated, tumor-bearing mice, we looked at a variety of different things, including angiogenesis,” Dr Lin said.

“We used CD31 immunohistochemistry to look at angiogenesis, and, in the untreated mice, we didn’t see a lot of staining. But in the tumor-bearing mice, there was a lot more staining [P=0.0231]. And when we gave the mice tasquinimod, angiogenesis was significantly decreased [P<0.0001].”

The researchers also looked at different angiogenic factors. And they found that, compared to control-treated mice with MM, tumor-bearing mice that received tasquinimod had a significant decrease in serum levels of VEGF, FGF2, tissue factor, and endoglin.

The team is now assessing the effects of S100A9 and tasquinimod on megakaryocytes and platelets, 2 of the major cell populations that promote angiogenesis.

*Information in the abstract differs from that presented at the meeting.

Attendees at AACR 2015

PHILADELPHIA—A drug that has fallen short of expectations in clinical trials of prostate cancer may be effective for treating multiple myeloma (MM), according to research presented at the AACR Annual Meeting 2015.

The drug, tasquinimod, inhibits the function of S100A9, a pro-inflammatory protein that is elevated in MM, prostate cancer, and other malignancies.

Researchers found that tasquinimod can reduce tumor growth and improve survival in mouse models of MM. And these effects are associated with reduced angiogenesis in the bone marrow.

Cindy Lin, PhD, of the Wistar Institute in Philadelphia, Pennsylvania, and her colleagues presented these findings as abstract 1364.* The research was supported by Active Biotech and Ipsen, the companies developing tasquinimod.

Dr Lin noted that tasquinimod has already been tested in clinical trials of prostate cancer and initially appeared to be very effective. However, recent results from a phase 3 trial suggested the drug does not confer a favorable risk-benefit ratio for this population.

So Active Biotech and Ipsen decided to discontinue all trials of tasquinimod in prostate cancer. But the preclinical results observed in MM suggest tasquinimod may hold promise for treating these patients.

Activity in MM

Dr Lin said previous preclinical experiments revealed that myeloid-derived suppressor cells are involved in regulating MM progression, and these cells produce S100A9.

Tasquinimod is a quinoline-3-carboxamide derivative that binds to S100A9 and inhibits interaction with its receptors. So Dr Lin and her colleagues decided to investigate the antitumor effect of the drug in mouse models of MM.

The researchers initially tested tasquinimod in a syngeneic MM model, randomizing mice to treatment or control. Mice in the treatment group received tasquinimod at 30 mg/kg/day in their drinking water for 28 days.

Tasquinimod significantly improved survival in this model (P<0.005). All control mice died within 30 days of tumor inoculation, but about 40% of tasquinimod-treated mice were still alive more than 80 days out.

The researchers then tested tasquinimod in xenograft models of human MM. The drug significantly reduced tumor size in both H929 (P=0.0042) and RPMI-8226 models (P=0.0003). Treatment significantly improved survival in the H929 (P=0.0008) and RPMI-8226 models as well (P=0.0243).

Dr Lin said she and her colleagues did not see any side effects of treatment in any of the mice.

Investigating the mechanism

To determine if the antitumor effect of tasquinimod is, in fact, mediated through inhibition of S100A9, the researchers administered the drug to S100A9-knockout mice with MM. Tasquinimod did not improve survival in these mice, which suggests its anti-MM effects are mediated through S100A9 inhibition.

“To try and investigate some of the mechanisms of how survival is improved in tasquinimod-treated, tumor-bearing mice, we looked at a variety of different things, including angiogenesis,” Dr Lin said.

“We used CD31 immunohistochemistry to look at angiogenesis, and, in the untreated mice, we didn’t see a lot of staining. But in the tumor-bearing mice, there was a lot more staining [P=0.0231]. And when we gave the mice tasquinimod, angiogenesis was significantly decreased [P<0.0001].”

The researchers also looked at different angiogenic factors. And they found that, compared to control-treated mice with MM, tumor-bearing mice that received tasquinimod had a significant decrease in serum levels of VEGF, FGF2, tissue factor, and endoglin.

The team is now assessing the effects of S100A9 and tasquinimod on megakaryocytes and platelets, 2 of the major cell populations that promote angiogenesis.

*Information in the abstract differs from that presented at the meeting.

Attendees at AACR 2015

PHILADELPHIA—A drug that has fallen short of expectations in clinical trials of prostate cancer may be effective for treating multiple myeloma (MM), according to research presented at the AACR Annual Meeting 2015.

The drug, tasquinimod, inhibits the function of S100A9, a pro-inflammatory protein that is elevated in MM, prostate cancer, and other malignancies.

Researchers found that tasquinimod can reduce tumor growth and improve survival in mouse models of MM. And these effects are associated with reduced angiogenesis in the bone marrow.

Cindy Lin, PhD, of the Wistar Institute in Philadelphia, Pennsylvania, and her colleagues presented these findings as abstract 1364.* The research was supported by Active Biotech and Ipsen, the companies developing tasquinimod.

Dr Lin noted that tasquinimod has already been tested in clinical trials of prostate cancer and initially appeared to be very effective. However, recent results from a phase 3 trial suggested the drug does not confer a favorable risk-benefit ratio for this population.

So Active Biotech and Ipsen decided to discontinue all trials of tasquinimod in prostate cancer. But the preclinical results observed in MM suggest tasquinimod may hold promise for treating these patients.

Activity in MM

Dr Lin said previous preclinical experiments revealed that myeloid-derived suppressor cells are involved in regulating MM progression, and these cells produce S100A9.

Tasquinimod is a quinoline-3-carboxamide derivative that binds to S100A9 and inhibits interaction with its receptors. So Dr Lin and her colleagues decided to investigate the antitumor effect of the drug in mouse models of MM.

The researchers initially tested tasquinimod in a syngeneic MM model, randomizing mice to treatment or control. Mice in the treatment group received tasquinimod at 30 mg/kg/day in their drinking water for 28 days.

Tasquinimod significantly improved survival in this model (P<0.005). All control mice died within 30 days of tumor inoculation, but about 40% of tasquinimod-treated mice were still alive more than 80 days out.

The researchers then tested tasquinimod in xenograft models of human MM. The drug significantly reduced tumor size in both H929 (P=0.0042) and RPMI-8226 models (P=0.0003). Treatment significantly improved survival in the H929 (P=0.0008) and RPMI-8226 models as well (P=0.0243).

Dr Lin said she and her colleagues did not see any side effects of treatment in any of the mice.

Investigating the mechanism

To determine if the antitumor effect of tasquinimod is, in fact, mediated through inhibition of S100A9, the researchers administered the drug to S100A9-knockout mice with MM. Tasquinimod did not improve survival in these mice, which suggests its anti-MM effects are mediated through S100A9 inhibition.

“To try and investigate some of the mechanisms of how survival is improved in tasquinimod-treated, tumor-bearing mice, we looked at a variety of different things, including angiogenesis,” Dr Lin said.

“We used CD31 immunohistochemistry to look at angiogenesis, and, in the untreated mice, we didn’t see a lot of staining. But in the tumor-bearing mice, there was a lot more staining [P=0.0231]. And when we gave the mice tasquinimod, angiogenesis was significantly decreased [P<0.0001].”

The researchers also looked at different angiogenic factors. And they found that, compared to control-treated mice with MM, tumor-bearing mice that received tasquinimod had a significant decrease in serum levels of VEGF, FGF2, tissue factor, and endoglin.

The team is now assessing the effects of S100A9 and tasquinimod on megakaryocytes and platelets, 2 of the major cell populations that promote angiogenesis.

*Information in the abstract differs from that presented at the meeting.

One of the key indicators of a nation’s health is how well it can care for its young. Despite many advances in medical care and improvements in access to care, infant mortality remains a significant concern worldwide. According to the World Health Organization, the leading cause of death among children under age 5 is preterm birth complications. With an estimated 15 million babies born prematurely (prior to 37 weeks’ gestation) globally each year, it is vital for ob.gyns. to uncover ways to predict, diagnose early, and treat the causes of preterm birth.

While the challenges to infant health could be considered more of an issue in developing countries, here in the United States, the Centers for Disease Control and Prevention estimates that 1 in 9 babies is born prematurely. Preterm birth-related causes of death (i.e., breathing and feeding problems and disabilities) accounted for 35% of all infant deaths in 2010.

The World Health Organization (WHO) lists the United States as one of the top 10 countries with the greatest number of preterm births, despite the fact that we spend approximately 17.1% of our gross domestic product in total health care expenditures – the highest rate among our peer nations.

In the April 2014 edition of Master Class, we discussed one of the primary causes of preterm birth, bacterial infections, and specifically the need for ob.gyns. to rigorously screen patients for asymptomatic bacteriuria, which can lead to pyelonephritis. This month, we examine another biologic marker of preterm birth, cervical length.

Seminal studies of transvaginal sonography to measure cervical length during pregnancy and predict premature birth were published more than 2 decades ago. This work showed that a short cervix at 24 and 28 weeks’ gestation predicted preterm birth. Since then, clinical studies have demonstrated the utility of cervical length screening in women with prior preterm pregnancies. In the last decade, three large, randomized human trials have examined the usefulness of universal cervical length screening (Am. J. Obstet. Gynecol. 2012;207:101-6). However, the results of these trials have given practitioners a confusing picture of the predictability of this biologic marker.

Given the complexity of the “to screen or not to screen” issue, we have devoted this Master Class to a discussion on the role of cervical length screening and the prediction of preterm birth. Our guest author this month is Dr. Erika Werner, an assistant professor in ob.gyn (maternal-fetal medicine) in the department of obstetrics and gynecology at Brown University, in Providence, R.I., and an expert in the area of preterm birth.

Dr. Reece, who specializes in maternal-fetal medicine, is vice president for medical affairs at the University of Maryland, Baltimore, as well as the John Z. and Akiko K. Bowers Distinguished Professor and dean of the school of medicine. Dr. Reece said he had no relevant financial disclosures. He is the medical editor of this column. Contact him at [email protected].

One of the key indicators of a nation’s health is how well it can care for its young. Despite many advances in medical care and improvements in access to care, infant mortality remains a significant concern worldwide. According to the World Health Organization, the leading cause of death among children under age 5 is preterm birth complications. With an estimated 15 million babies born prematurely (prior to 37 weeks’ gestation) globally each year, it is vital for ob.gyns. to uncover ways to predict, diagnose early, and treat the causes of preterm birth.

While the challenges to infant health could be considered more of an issue in developing countries, here in the United States, the Centers for Disease Control and Prevention estimates that 1 in 9 babies is born prematurely. Preterm birth-related causes of death (i.e., breathing and feeding problems and disabilities) accounted for 35% of all infant deaths in 2010.

The World Health Organization (WHO) lists the United States as one of the top 10 countries with the greatest number of preterm births, despite the fact that we spend approximately 17.1% of our gross domestic product in total health care expenditures – the highest rate among our peer nations.

In the April 2014 edition of Master Class, we discussed one of the primary causes of preterm birth, bacterial infections, and specifically the need for ob.gyns. to rigorously screen patients for asymptomatic bacteriuria, which can lead to pyelonephritis. This month, we examine another biologic marker of preterm birth, cervical length.

Seminal studies of transvaginal sonography to measure cervical length during pregnancy and predict premature birth were published more than 2 decades ago. This work showed that a short cervix at 24 and 28 weeks’ gestation predicted preterm birth. Since then, clinical studies have demonstrated the utility of cervical length screening in women with prior preterm pregnancies. In the last decade, three large, randomized human trials have examined the usefulness of universal cervical length screening (Am. J. Obstet. Gynecol. 2012;207:101-6). However, the results of these trials have given practitioners a confusing picture of the predictability of this biologic marker.

Given the complexity of the “to screen or not to screen” issue, we have devoted this Master Class to a discussion on the role of cervical length screening and the prediction of preterm birth. Our guest author this month is Dr. Erika Werner, an assistant professor in ob.gyn (maternal-fetal medicine) in the department of obstetrics and gynecology at Brown University, in Providence, R.I., and an expert in the area of preterm birth.

Dr. Reece, who specializes in maternal-fetal medicine, is vice president for medical affairs at the University of Maryland, Baltimore, as well as the John Z. and Akiko K. Bowers Distinguished Professor and dean of the school of medicine. Dr. Reece said he had no relevant financial disclosures. He is the medical editor of this column. Contact him at [email protected].

One of the key indicators of a nation’s health is how well it can care for its young. Despite many advances in medical care and improvements in access to care, infant mortality remains a significant concern worldwide. According to the World Health Organization, the leading cause of death among children under age 5 is preterm birth complications. With an estimated 15 million babies born prematurely (prior to 37 weeks’ gestation) globally each year, it is vital for ob.gyns. to uncover ways to predict, diagnose early, and treat the causes of preterm birth.

While the challenges to infant health could be considered more of an issue in developing countries, here in the United States, the Centers for Disease Control and Prevention estimates that 1 in 9 babies is born prematurely. Preterm birth-related causes of death (i.e., breathing and feeding problems and disabilities) accounted for 35% of all infant deaths in 2010.

The World Health Organization (WHO) lists the United States as one of the top 10 countries with the greatest number of preterm births, despite the fact that we spend approximately 17.1% of our gross domestic product in total health care expenditures – the highest rate among our peer nations.

In the April 2014 edition of Master Class, we discussed one of the primary causes of preterm birth, bacterial infections, and specifically the need for ob.gyns. to rigorously screen patients for asymptomatic bacteriuria, which can lead to pyelonephritis. This month, we examine another biologic marker of preterm birth, cervical length.

Seminal studies of transvaginal sonography to measure cervical length during pregnancy and predict premature birth were published more than 2 decades ago. This work showed that a short cervix at 24 and 28 weeks’ gestation predicted preterm birth. Since then, clinical studies have demonstrated the utility of cervical length screening in women with prior preterm pregnancies. In the last decade, three large, randomized human trials have examined the usefulness of universal cervical length screening (Am. J. Obstet. Gynecol. 2012;207:101-6). However, the results of these trials have given practitioners a confusing picture of the predictability of this biologic marker.

Given the complexity of the “to screen or not to screen” issue, we have devoted this Master Class to a discussion on the role of cervical length screening and the prediction of preterm birth. Our guest author this month is Dr. Erika Werner, an assistant professor in ob.gyn (maternal-fetal medicine) in the department of obstetrics and gynecology at Brown University, in Providence, R.I., and an expert in the area of preterm birth.

Dr. Reece, who specializes in maternal-fetal medicine, is vice president for medical affairs at the University of Maryland, Baltimore, as well as the John Z. and Akiko K. Bowers Distinguished Professor and dean of the school of medicine. Dr. Reece said he had no relevant financial disclosures. He is the medical editor of this column. Contact him at [email protected].

Rates of preterm birth in the United States have been falling since 2006, but the rates of early preterm birth in singletons (those under 34 weeks’ gestation), specifically, have not trended downward as dramatically as have late preterm birth in singletons (34-36 weeks). According to 2015 data from the National Vital Statistics Reports, the rate of early preterm births is still 3.4% in all pregnancies and 2.7% among singletons.

While the number of neonates born before 37 weeks of gestation remains high – approximately 11% in 2013 – and signifies a continuing public health problem, the rate of early preterm birth is particularly concerning because early preterm birth is more significantly associated with neonatal mortality, long-term morbidity and extended neonatal intensive care unit stays, all leading to increased health care expenditures.

Finding predictors for preterm birth that are stronger than traditional clinical factors has long been a goal of ob.gyns. because the vast majority of all spontaneous preterm births occur to women without known risk factors (i.e., multiple gestations or prior preterm birth).

Cervical length in the midtrimester is now a well-verified predictor of preterm birth, for both low- and high-risk women. Furthermore, vaginal progesterone has been shown to be a safe and beneficial intervention for women with no known risk factors who are diagnosed with a shortened cervical length (< 2 cm), and cervical cerclage has been suggested to reduce the risk of preterm birth for women with a history of prior preterm birth who also have a shortened cervical length.

Some are now advocating universal cervical length screening for women with singleton gestations, but before universal screening is mandated, the downstream effect of such a change in practice must be considered.

Backdrop to screening

Cervical length measurement was first investigated more than 25 years ago as a possible predictor of preterm birth. In 1996, a prospective multicenter study of almost 3,000 women with singleton pregnancies showed that the risk of preterm delivery is inversely and directly related to the length of the cervix, as measured with vaginal ultrasonography (N. Engl. J. Med. 1996;334:567-72).

In fact, at 24 weeks’ gestation, every 1 mm of additional cervical length equates to a significant decrease in preterm birth risk (odds ratio, 0.91). Several other studies, in addition to the landmark 1996 study, have similarly demonstrated this inverse relationship between preterm birth risk and cervical length between 18 and 24 weeks’ gestation.

However, the use of cervical measurement did not achieve widespread use until more than a decade later, when researchers began to identify interventions that could prolong pregnancy if a short cervix was diagnosed in the second trimester.

For example, Dr. E.B. Fonseca’s study of almost 25,000 asymptomatic pregnant women, demonstrated that daily vaginal progesterone reduced the risk of spontaneous delivery before 34 weeks by approximately 44% in women identified with a cervical length of 1.5 cm or less (N. Engl. J. Med. 2007;357:462-9). The vast majority of the women in this study had singleton pregnancies.

Shortly thereafter, Dr. S.S. Hassan and her colleagues completed a similar trial in women with singleton gestations and transvaginal cervical lengths between 1.0 and 2.0 cm at 20-23 weeks’ gestation. In this trial, nightly progesterone gel (with 90 mg progesterone per application) was associated with a 45% reduction in preterm birth before 33 weeks and a 38% reduction in preterm birth before 35 weeks (Ultrasound. Obstet. Gynecol. 2011;38:18-31).

A meta-analysis led by Dr. Roberto Romero, which included the Fonseca and Hassan trials, looked specifically at 775 women with a midtrimester cervical length of 2.5 cm or less. Women with a singleton gestation who had no history of preterm birth had a 40% reduction in the rate of early preterm birth when they were treated with vaginal progesterone (Am. J. Obstet. Gynecol. 2012;206:124-e1-19).

The benefits of identifying a short cervix likely extend to women with a history of prior preterm birth. A patient-level meta-analysis published in 2011 demonstrated that cervical cerclage placement was associated with a significant reduction in preterm birth before 35 weeks’ gestation in women with singleton gestations, previous spontaneous preterm birth, and cervical length less than 2.5 cm before 24 weeks’ gestation (Obstet. Gynecol. 2011;117:663-71).

The possible benefits of diagnosing and intervening for a shortened cervix have tipped many experts and clinicians toward the practice of universal cervical length screening of all singleton pregnancies. Research has shown that we can accurately obtain a cervical-length measurement before 24 weeks, and that we have effective and safe interventions for cases of short cervix: cerclage in women with a history of preterm birth who are already receiving progesterone, and vaginal progesterone in women without such a history.

 

Screening certainties and doubts

In 2011, my colleagues and I compared the cost effectiveness of two approaches to preterm birth prevention in low-risk pregnancies: no screening versus a single transvaginal ultrasound cervical-length measurement in all asymptomatic, low-risk singleton pregnant individuals between 18 and 24 weeks’ gestation.

In our model, women identified as having a cervical length less than 1.5 cm would be offered vaginal progesterone. Based on published data, we assumed there would be a 92% adherence rate, and a 45% reduction in deliveries before 34 weeks with progesterone treatment.

We found that in low-risk pregnancies, universal transvaginal cervical-length ultrasound screening and progesterone intervention would be cost effective and in many cases cost saving. We estimated that screening would prevent 248 early preterm births – as well as 22 neonatal deaths or neonates with long-term neurologic deficits – per 100,000 deliveries.

Our sensitivity analyses showed that screening remained cost saving under a range of clinical scenarios, including varied preterm birth rates and predictive values of a shortened cervix. Screening was not cost saving, but remained cost effective, when the expense of a transvaginal ultrasound scan exceeds $187 or when vaginal progesterone is assumed to reduce the risk of early preterm delivery by less than 20% (Ultrasound Obstet. Gynecol. 2011;38;32-37).

Neither the American College of Obstetricians and Gynecologists nor the Society for Maternal-Fetal Medicine support mandated universal transvaginal ultrasound cervical length screening. Both organizations state, however, that the approach may be considered in women with singleton gestations without prior spontaneous preterm birth.

Interestingly, Thomas Jefferson University in Philadelphia, which uses a universal screening program for singleton gestations without prior preterm birth, has recently published data that complicate the growing trend toward universal cervical length screening.

The Philadelphia clinicians followed a strategy whereby women with a transvaginal cervical length of 2 cm or less were prescribed vaginal progesterone (90 mg vaginal progesterone gel, or 200 mg micronized progesterone gel capsules). Those with a cervical length between approximately 2 cm and 2.5 cm were asked to return for a follow-up cervical length measurement before 24 weeks’ gestation.

What they found in this cohort was surprising: a rate of short cervix that is significantly lower than what previous research has shown.

Among those screened, 0.8% of women had a cervical length of 2 cm or less on an initial transvaginal ultrasonogram. Previously, a prevalence of 1%-2% for an even shorter cervical length (less than 1.5 cm) was fairly consistent in the literature.

As Dr. Kelly M. Orzechowski and her colleagues point out, the low incidence of short cervix “raises questions regarding whether universal transvaginal ultrasonogram cervical length screening in low-risk asymptomatic women is beneficial” (Obstet. Gynecol. 2014;124:520-5).

In our 2011 cost-effectiveness analysis, we found that screening was no longer a cost-saving practice when the incidence of cervical length less than 1.5 cm falls below 0.8%. Screening remained cost effective, however.

Recently, we found that if the Philadelphia protocol is followed and the U.S. population has an incidence of shortened cervix similar to that described by Dr. Orzechowski and her colleagues, universal cervical length screening in low-risk singleton pregnancies is cost effective but not cost saving. Furthermore, we found several additional plausible situations in this unpublished analysis in which universal screening ceased to be cost effective.

Thus, before we move to a strategy of mandated universal screening, we need better population-based estimates of the incidence of short cervix in a truly low-risk population.

We also must consider the future costs of progesterone. It is possible that costs may increase significantly if vaginal progesterone wins approval from the Food and Drug Administration for this indication.

Finally, if universal cervical length screening is to become the standard of care, we need policies in place to prevent misuse of the screening technology that would inevitably drive up costs without improving outcomes. For example, we must ensure that one cervical length measurement does not transition into serial cervical length measurements over the course of pregnancy, since measurement after 24 weeks has limited clinical utility. Similarly, progesterone use for a cervical length less than or equal to 2.0 cm cannot progress to progesterone for anyone approaching 2.0 cm (i.e. 2.5 cm or even 3 cm) as there is no evidence to suggest a benefit for women with longer cervixes.

Over time, it would be beneficial to have additional data on how best to manage patients who have a cervical length of 2 cm-2.5 cm before 24 weeks’ gestation. Many of us ask these women to return for a follow-up measurement and some may prescribe progesterone. However, we lack evidence for either approach; while a cervical length measurement less than 2.5 cm is clearly associated with an increased risk of preterm birth, the benefit of treatment has been demonstrated only with a cervical length of 2 cm or less.

 

Today and the future

For women with a history of preterm birth, cervical length screening is now routine. For low-risk pregnant women – those without a history of previous spontaneous preterm delivery – various approaches are currently taken. Most physicians recommend assessing the cervical length transabdominally at the time of the 18-20-week ultrasound, and proceeding to transvaginal ultrasonography if the cervical length is less than 3 cm or 3.5 cm.

To reliably image the cervix with transabdominal ultrasound, it should be performed with a full bladder and with the understanding that the cervix appears longer (6 mm longer, on average) when the bladder is full (Aust. N. Z. J. Obstet. Gynaecol. 2014;54:250-55).

Transvaginal ultrasound has been widely recognized as a sensitive and reproducible method for detecting shortened cervical length. Overall, this tool has several advantages over the transabdominal approach. However, the lack of universal access to transvaginal ultrasound and to consistently reliable cervical length measurements have been valid concerns of those who oppose universal transvaginal ultrasound cervical length screening.

Such concerns likely will lessen over time as transvaginal ultrasound continues to become more pervasive. Several years ago, the Perinatal Quality Foundation set standards for measuring the cervix and launched the Cervical Length Education and Review (CLEAR) program. When sonographers and physicians obtain training and credentialing, there appears to be only a 5%-10% intraobserver variability in cervical length measurement. (The PQF’s initial focus in 2005 was the Nuchal Translucency Quality Review program.)

Increasingly, I believe, transvaginal ultrasound cervical length measurement will be utilized to identify women at high risk for early preterm birth so that low-risk women can receive progesterone and high-risk women (those with a history of preterm birth) can be considered as candidates for cerclage placement. In the process, the quality of clinical care as well as the quality of our research data will improve. Whether and when such screening will become universal, however, is still uncertain.

Dr. Werner reported that she has no financial disclosures relevant to this Master Class.

Rates of preterm birth in the United States have been falling since 2006, but the rates of early preterm birth in singletons (those under 34 weeks’ gestation), specifically, have not trended downward as dramatically as have late preterm birth in singletons (34-36 weeks). According to 2015 data from the National Vital Statistics Reports, the rate of early preterm births is still 3.4% in all pregnancies and 2.7% among singletons.

While the number of neonates born before 37 weeks of gestation remains high – approximately 11% in 2013 – and signifies a continuing public health problem, the rate of early preterm birth is particularly concerning because early preterm birth is more significantly associated with neonatal mortality, long-term morbidity and extended neonatal intensive care unit stays, all leading to increased health care expenditures.

Finding predictors for preterm birth that are stronger than traditional clinical factors has long been a goal of ob.gyns. because the vast majority of all spontaneous preterm births occur to women without known risk factors (i.e., multiple gestations or prior preterm birth).

Cervical length in the midtrimester is now a well-verified predictor of preterm birth, for both low- and high-risk women. Furthermore, vaginal progesterone has been shown to be a safe and beneficial intervention for women with no known risk factors who are diagnosed with a shortened cervical length (< 2 cm), and cervical cerclage has been suggested to reduce the risk of preterm birth for women with a history of prior preterm birth who also have a shortened cervical length.

Some are now advocating universal cervical length screening for women with singleton gestations, but before universal screening is mandated, the downstream effect of such a change in practice must be considered.

Backdrop to screening

Cervical length measurement was first investigated more than 25 years ago as a possible predictor of preterm birth. In 1996, a prospective multicenter study of almost 3,000 women with singleton pregnancies showed that the risk of preterm delivery is inversely and directly related to the length of the cervix, as measured with vaginal ultrasonography (N. Engl. J. Med. 1996;334:567-72).

In fact, at 24 weeks’ gestation, every 1 mm of additional cervical length equates to a significant decrease in preterm birth risk (odds ratio, 0.91). Several other studies, in addition to the landmark 1996 study, have similarly demonstrated this inverse relationship between preterm birth risk and cervical length between 18 and 24 weeks’ gestation.

However, the use of cervical measurement did not achieve widespread use until more than a decade later, when researchers began to identify interventions that could prolong pregnancy if a short cervix was diagnosed in the second trimester.

For example, Dr. E.B. Fonseca’s study of almost 25,000 asymptomatic pregnant women, demonstrated that daily vaginal progesterone reduced the risk of spontaneous delivery before 34 weeks by approximately 44% in women identified with a cervical length of 1.5 cm or less (N. Engl. J. Med. 2007;357:462-9). The vast majority of the women in this study had singleton pregnancies.

Shortly thereafter, Dr. S.S. Hassan and her colleagues completed a similar trial in women with singleton gestations and transvaginal cervical lengths between 1.0 and 2.0 cm at 20-23 weeks’ gestation. In this trial, nightly progesterone gel (with 90 mg progesterone per application) was associated with a 45% reduction in preterm birth before 33 weeks and a 38% reduction in preterm birth before 35 weeks (Ultrasound. Obstet. Gynecol. 2011;38:18-31).

A meta-analysis led by Dr. Roberto Romero, which included the Fonseca and Hassan trials, looked specifically at 775 women with a midtrimester cervical length of 2.5 cm or less. Women with a singleton gestation who had no history of preterm birth had a 40% reduction in the rate of early preterm birth when they were treated with vaginal progesterone (Am. J. Obstet. Gynecol. 2012;206:124-e1-19).

The benefits of identifying a short cervix likely extend to women with a history of prior preterm birth. A patient-level meta-analysis published in 2011 demonstrated that cervical cerclage placement was associated with a significant reduction in preterm birth before 35 weeks’ gestation in women with singleton gestations, previous spontaneous preterm birth, and cervical length less than 2.5 cm before 24 weeks’ gestation (Obstet. Gynecol. 2011;117:663-71).

The possible benefits of diagnosing and intervening for a shortened cervix have tipped many experts and clinicians toward the practice of universal cervical length screening of all singleton pregnancies. Research has shown that we can accurately obtain a cervical-length measurement before 24 weeks, and that we have effective and safe interventions for cases of short cervix: cerclage in women with a history of preterm birth who are already receiving progesterone, and vaginal progesterone in women without such a history.

 

Screening certainties and doubts

In 2011, my colleagues and I compared the cost effectiveness of two approaches to preterm birth prevention in low-risk pregnancies: no screening versus a single transvaginal ultrasound cervical-length measurement in all asymptomatic, low-risk singleton pregnant individuals between 18 and 24 weeks’ gestation.

In our model, women identified as having a cervical length less than 1.5 cm would be offered vaginal progesterone. Based on published data, we assumed there would be a 92% adherence rate, and a 45% reduction in deliveries before 34 weeks with progesterone treatment.

We found that in low-risk pregnancies, universal transvaginal cervical-length ultrasound screening and progesterone intervention would be cost effective and in many cases cost saving. We estimated that screening would prevent 248 early preterm births – as well as 22 neonatal deaths or neonates with long-term neurologic deficits – per 100,000 deliveries.

Our sensitivity analyses showed that screening remained cost saving under a range of clinical scenarios, including varied preterm birth rates and predictive values of a shortened cervix. Screening was not cost saving, but remained cost effective, when the expense of a transvaginal ultrasound scan exceeds $187 or when vaginal progesterone is assumed to reduce the risk of early preterm delivery by less than 20% (Ultrasound Obstet. Gynecol. 2011;38;32-37).

Neither the American College of Obstetricians and Gynecologists nor the Society for Maternal-Fetal Medicine support mandated universal transvaginal ultrasound cervical length screening. Both organizations state, however, that the approach may be considered in women with singleton gestations without prior spontaneous preterm birth.

Interestingly, Thomas Jefferson University in Philadelphia, which uses a universal screening program for singleton gestations without prior preterm birth, has recently published data that complicate the growing trend toward universal cervical length screening.

The Philadelphia clinicians followed a strategy whereby women with a transvaginal cervical length of 2 cm or less were prescribed vaginal progesterone (90 mg vaginal progesterone gel, or 200 mg micronized progesterone gel capsules). Those with a cervical length between approximately 2 cm and 2.5 cm were asked to return for a follow-up cervical length measurement before 24 weeks’ gestation.

What they found in this cohort was surprising: a rate of short cervix that is significantly lower than what previous research has shown.

Among those screened, 0.8% of women had a cervical length of 2 cm or less on an initial transvaginal ultrasonogram. Previously, a prevalence of 1%-2% for an even shorter cervical length (less than 1.5 cm) was fairly consistent in the literature.

As Dr. Kelly M. Orzechowski and her colleagues point out, the low incidence of short cervix “raises questions regarding whether universal transvaginal ultrasonogram cervical length screening in low-risk asymptomatic women is beneficial” (Obstet. Gynecol. 2014;124:520-5).

In our 2011 cost-effectiveness analysis, we found that screening was no longer a cost-saving practice when the incidence of cervical length less than 1.5 cm falls below 0.8%. Screening remained cost effective, however.

Recently, we found that if the Philadelphia protocol is followed and the U.S. population has an incidence of shortened cervix similar to that described by Dr. Orzechowski and her colleagues, universal cervical length screening in low-risk singleton pregnancies is cost effective but not cost saving. Furthermore, we found several additional plausible situations in this unpublished analysis in which universal screening ceased to be cost effective.

Thus, before we move to a strategy of mandated universal screening, we need better population-based estimates of the incidence of short cervix in a truly low-risk population.

We also must consider the future costs of progesterone. It is possible that costs may increase significantly if vaginal progesterone wins approval from the Food and Drug Administration for this indication.

Finally, if universal cervical length screening is to become the standard of care, we need policies in place to prevent misuse of the screening technology that would inevitably drive up costs without improving outcomes. For example, we must ensure that one cervical length measurement does not transition into serial cervical length measurements over the course of pregnancy, since measurement after 24 weeks has limited clinical utility. Similarly, progesterone use for a cervical length less than or equal to 2.0 cm cannot progress to progesterone for anyone approaching 2.0 cm (i.e. 2.5 cm or even 3 cm) as there is no evidence to suggest a benefit for women with longer cervixes.

Over time, it would be beneficial to have additional data on how best to manage patients who have a cervical length of 2 cm-2.5 cm before 24 weeks’ gestation. Many of us ask these women to return for a follow-up measurement and some may prescribe progesterone. However, we lack evidence for either approach; while a cervical length measurement less than 2.5 cm is clearly associated with an increased risk of preterm birth, the benefit of treatment has been demonstrated only with a cervical length of 2 cm or less.

 

Today and the future

For women with a history of preterm birth, cervical length screening is now routine. For low-risk pregnant women – those without a history of previous spontaneous preterm delivery – various approaches are currently taken. Most physicians recommend assessing the cervical length transabdominally at the time of the 18-20-week ultrasound, and proceeding to transvaginal ultrasonography if the cervical length is less than 3 cm or 3.5 cm.

To reliably image the cervix with transabdominal ultrasound, it should be performed with a full bladder and with the understanding that the cervix appears longer (6 mm longer, on average) when the bladder is full (Aust. N. Z. J. Obstet. Gynaecol. 2014;54:250-55).

Transvaginal ultrasound has been widely recognized as a sensitive and reproducible method for detecting shortened cervical length. Overall, this tool has several advantages over the transabdominal approach. However, the lack of universal access to transvaginal ultrasound and to consistently reliable cervical length measurements have been valid concerns of those who oppose universal transvaginal ultrasound cervical length screening.

Such concerns likely will lessen over time as transvaginal ultrasound continues to become more pervasive. Several years ago, the Perinatal Quality Foundation set standards for measuring the cervix and launched the Cervical Length Education and Review (CLEAR) program. When sonographers and physicians obtain training and credentialing, there appears to be only a 5%-10% intraobserver variability in cervical length measurement. (The PQF’s initial focus in 2005 was the Nuchal Translucency Quality Review program.)

Increasingly, I believe, transvaginal ultrasound cervical length measurement will be utilized to identify women at high risk for early preterm birth so that low-risk women can receive progesterone and high-risk women (those with a history of preterm birth) can be considered as candidates for cerclage placement. In the process, the quality of clinical care as well as the quality of our research data will improve. Whether and when such screening will become universal, however, is still uncertain.

Dr. Werner reported that she has no financial disclosures relevant to this Master Class.

Rates of preterm birth in the United States have been falling since 2006, but the rates of early preterm birth in singletons (those under 34 weeks’ gestation), specifically, have not trended downward as dramatically as have late preterm birth in singletons (34-36 weeks). According to 2015 data from the National Vital Statistics Reports, the rate of early preterm births is still 3.4% in all pregnancies and 2.7% among singletons.

While the number of neonates born before 37 weeks of gestation remains high – approximately 11% in 2013 – and signifies a continuing public health problem, the rate of early preterm birth is particularly concerning because early preterm birth is more significantly associated with neonatal mortality, long-term morbidity and extended neonatal intensive care unit stays, all leading to increased health care expenditures.

Finding predictors for preterm birth that are stronger than traditional clinical factors has long been a goal of ob.gyns. because the vast majority of all spontaneous preterm births occur to women without known risk factors (i.e., multiple gestations or prior preterm birth).

Cervical length in the midtrimester is now a well-verified predictor of preterm birth, for both low- and high-risk women. Furthermore, vaginal progesterone has been shown to be a safe and beneficial intervention for women with no known risk factors who are diagnosed with a shortened cervical length (< 2 cm), and cervical cerclage has been suggested to reduce the risk of preterm birth for women with a history of prior preterm birth who also have a shortened cervical length.

Some are now advocating universal cervical length screening for women with singleton gestations, but before universal screening is mandated, the downstream effect of such a change in practice must be considered.

Backdrop to screening

Cervical length measurement was first investigated more than 25 years ago as a possible predictor of preterm birth. In 1996, a prospective multicenter study of almost 3,000 women with singleton pregnancies showed that the risk of preterm delivery is inversely and directly related to the length of the cervix, as measured with vaginal ultrasonography (N. Engl. J. Med. 1996;334:567-72).

In fact, at 24 weeks’ gestation, every 1 mm of additional cervical length equates to a significant decrease in preterm birth risk (odds ratio, 0.91). Several other studies, in addition to the landmark 1996 study, have similarly demonstrated this inverse relationship between preterm birth risk and cervical length between 18 and 24 weeks’ gestation.

However, the use of cervical measurement did not achieve widespread use until more than a decade later, when researchers began to identify interventions that could prolong pregnancy if a short cervix was diagnosed in the second trimester.

For example, Dr. E.B. Fonseca’s study of almost 25,000 asymptomatic pregnant women, demonstrated that daily vaginal progesterone reduced the risk of spontaneous delivery before 34 weeks by approximately 44% in women identified with a cervical length of 1.5 cm or less (N. Engl. J. Med. 2007;357:462-9). The vast majority of the women in this study had singleton pregnancies.

Shortly thereafter, Dr. S.S. Hassan and her colleagues completed a similar trial in women with singleton gestations and transvaginal cervical lengths between 1.0 and 2.0 cm at 20-23 weeks’ gestation. In this trial, nightly progesterone gel (with 90 mg progesterone per application) was associated with a 45% reduction in preterm birth before 33 weeks and a 38% reduction in preterm birth before 35 weeks (Ultrasound. Obstet. Gynecol. 2011;38:18-31).

A meta-analysis led by Dr. Roberto Romero, which included the Fonseca and Hassan trials, looked specifically at 775 women with a midtrimester cervical length of 2.5 cm or less. Women with a singleton gestation who had no history of preterm birth had a 40% reduction in the rate of early preterm birth when they were treated with vaginal progesterone (Am. J. Obstet. Gynecol. 2012;206:124-e1-19).

The benefits of identifying a short cervix likely extend to women with a history of prior preterm birth. A patient-level meta-analysis published in 2011 demonstrated that cervical cerclage placement was associated with a significant reduction in preterm birth before 35 weeks’ gestation in women with singleton gestations, previous spontaneous preterm birth, and cervical length less than 2.5 cm before 24 weeks’ gestation (Obstet. Gynecol. 2011;117:663-71).

The possible benefits of diagnosing and intervening for a shortened cervix have tipped many experts and clinicians toward the practice of universal cervical length screening of all singleton pregnancies. Research has shown that we can accurately obtain a cervical-length measurement before 24 weeks, and that we have effective and safe interventions for cases of short cervix: cerclage in women with a history of preterm birth who are already receiving progesterone, and vaginal progesterone in women without such a history.

 

Screening certainties and doubts

In 2011, my colleagues and I compared the cost effectiveness of two approaches to preterm birth prevention in low-risk pregnancies: no screening versus a single transvaginal ultrasound cervical-length measurement in all asymptomatic, low-risk singleton pregnant individuals between 18 and 24 weeks’ gestation.

In our model, women identified as having a cervical length less than 1.5 cm would be offered vaginal progesterone. Based on published data, we assumed there would be a 92% adherence rate, and a 45% reduction in deliveries before 34 weeks with progesterone treatment.

We found that in low-risk pregnancies, universal transvaginal cervical-length ultrasound screening and progesterone intervention would be cost effective and in many cases cost saving. We estimated that screening would prevent 248 early preterm births – as well as 22 neonatal deaths or neonates with long-term neurologic deficits – per 100,000 deliveries.

Our sensitivity analyses showed that screening remained cost saving under a range of clinical scenarios, including varied preterm birth rates and predictive values of a shortened cervix. Screening was not cost saving, but remained cost effective, when the expense of a transvaginal ultrasound scan exceeds $187 or when vaginal progesterone is assumed to reduce the risk of early preterm delivery by less than 20% (Ultrasound Obstet. Gynecol. 2011;38;32-37).

Neither the American College of Obstetricians and Gynecologists nor the Society for Maternal-Fetal Medicine support mandated universal transvaginal ultrasound cervical length screening. Both organizations state, however, that the approach may be considered in women with singleton gestations without prior spontaneous preterm birth.

Interestingly, Thomas Jefferson University in Philadelphia, which uses a universal screening program for singleton gestations without prior preterm birth, has recently published data that complicate the growing trend toward universal cervical length screening.

The Philadelphia clinicians followed a strategy whereby women with a transvaginal cervical length of 2 cm or less were prescribed vaginal progesterone (90 mg vaginal progesterone gel, or 200 mg micronized progesterone gel capsules). Those with a cervical length between approximately 2 cm and 2.5 cm were asked to return for a follow-up cervical length measurement before 24 weeks’ gestation.

What they found in this cohort was surprising: a rate of short cervix that is significantly lower than what previous research has shown.

Among those screened, 0.8% of women had a cervical length of 2 cm or less on an initial transvaginal ultrasonogram. Previously, a prevalence of 1%-2% for an even shorter cervical length (less than 1.5 cm) was fairly consistent in the literature.

As Dr. Kelly M. Orzechowski and her colleagues point out, the low incidence of short cervix “raises questions regarding whether universal transvaginal ultrasonogram cervical length screening in low-risk asymptomatic women is beneficial” (Obstet. Gynecol. 2014;124:520-5).

In our 2011 cost-effectiveness analysis, we found that screening was no longer a cost-saving practice when the incidence of cervical length less than 1.5 cm falls below 0.8%. Screening remained cost effective, however.

Recently, we found that if the Philadelphia protocol is followed and the U.S. population has an incidence of shortened cervix similar to that described by Dr. Orzechowski and her colleagues, universal cervical length screening in low-risk singleton pregnancies is cost effective but not cost saving. Furthermore, we found several additional plausible situations in this unpublished analysis in which universal screening ceased to be cost effective.

Thus, before we move to a strategy of mandated universal screening, we need better population-based estimates of the incidence of short cervix in a truly low-risk population.

We also must consider the future costs of progesterone. It is possible that costs may increase significantly if vaginal progesterone wins approval from the Food and Drug Administration for this indication.

Finally, if universal cervical length screening is to become the standard of care, we need policies in place to prevent misuse of the screening technology that would inevitably drive up costs without improving outcomes. For example, we must ensure that one cervical length measurement does not transition into serial cervical length measurements over the course of pregnancy, since measurement after 24 weeks has limited clinical utility. Similarly, progesterone use for a cervical length less than or equal to 2.0 cm cannot progress to progesterone for anyone approaching 2.0 cm (i.e. 2.5 cm or even 3 cm) as there is no evidence to suggest a benefit for women with longer cervixes.

Over time, it would be beneficial to have additional data on how best to manage patients who have a cervical length of 2 cm-2.5 cm before 24 weeks’ gestation. Many of us ask these women to return for a follow-up measurement and some may prescribe progesterone. However, we lack evidence for either approach; while a cervical length measurement less than 2.5 cm is clearly associated with an increased risk of preterm birth, the benefit of treatment has been demonstrated only with a cervical length of 2 cm or less.

 

Today and the future

For women with a history of preterm birth, cervical length screening is now routine. For low-risk pregnant women – those without a history of previous spontaneous preterm delivery – various approaches are currently taken. Most physicians recommend assessing the cervical length transabdominally at the time of the 18-20-week ultrasound, and proceeding to transvaginal ultrasonography if the cervical length is less than 3 cm or 3.5 cm.

To reliably image the cervix with transabdominal ultrasound, it should be performed with a full bladder and with the understanding that the cervix appears longer (6 mm longer, on average) when the bladder is full (Aust. N. Z. J. Obstet. Gynaecol. 2014;54:250-55).

Transvaginal ultrasound has been widely recognized as a sensitive and reproducible method for detecting shortened cervical length. Overall, this tool has several advantages over the transabdominal approach. However, the lack of universal access to transvaginal ultrasound and to consistently reliable cervical length measurements have been valid concerns of those who oppose universal transvaginal ultrasound cervical length screening.

Such concerns likely will lessen over time as transvaginal ultrasound continues to become more pervasive. Several years ago, the Perinatal Quality Foundation set standards for measuring the cervix and launched the Cervical Length Education and Review (CLEAR) program. When sonographers and physicians obtain training and credentialing, there appears to be only a 5%-10% intraobserver variability in cervical length measurement. (The PQF’s initial focus in 2005 was the Nuchal Translucency Quality Review program.)

Increasingly, I believe, transvaginal ultrasound cervical length measurement will be utilized to identify women at high risk for early preterm birth so that low-risk women can receive progesterone and high-risk women (those with a history of preterm birth) can be considered as candidates for cerclage placement. In the process, the quality of clinical care as well as the quality of our research data will improve. Whether and when such screening will become universal, however, is still uncertain.

Dr. Werner reported that she has no financial disclosures relevant to this Master Class.

Only recently has paternal postpartum depression (PPD) received much attention. Research has shown that maternal PPD is associated with negative outcomes in the child’s cognitive develop­ment and social and marital problems for the parents. Likewise, depressed fathers are less likely to play outside with their child and more likely to put the child to bed awake.¹

Recent studies reported that 10.4% of men experienced depression within 12 month of delivery.¹ Edmondson et al² estimate the prevalence of paternal PPD to be 8% between birth and 3 months, 26% from 3 to 6 months, and 9% from 6 to 12 months.

Risk factors
Risk factors for paternal PPD have not been studied extensively. Some studies have shown that immaturity, lack of social sup­port, first or unplanned pregnancies, marital relationship problems, and unemployment were the most common risk factors for depression in men postnatally.³ A history of depression and other psychiatric disorders also increases risk.⁴ Psychosocial factors, such as quality of the spousal relationship, parenting distress, and perceived parenting efficacy, contribute to paternal depression.

Similarly, depressed postpartum fathers experience higher levels of parenting dis­tress and a lower sense of parenting effi­cacy.⁵ Interestingly, negative life events were associated with increased risk for depression in mothers, but not fathers.³

Clinical presentation
Paternal PPD symptoms appear within 12 months after the birth of the child and last for at least 2 weeks. Signs and symptoms of depression in men might not resemble those seen in postpartum women. Men tend to show aggression, increased or easy irritabil­ity, and agitation, and might not seek help for emotional issues as readily as women do. Typical symptoms of depression often are present, such as sleep disturbance or changes in sleep patterns, difficulty concen­trating, memory problems, and feelings of worthlessness, hopelessness, inadequacy, and excess guilt with suicidal ideation.⁶

Making the diagnosis
Maternal PPD commonly is evaluated using the Edinburgh Postnatal Depression Scale- Partner (EDPS-P) or Postpartum Depression Screening Scale. However, studies are lack­ing to determine which diagnostic modal­ity is most accurate for diagnosing paternal PPD.

A paternal PPD screening tool could include the EDPS-P administered to moth­ers. Edmonson et al² determined an EDPS-P score of >10 was the optimal cut-off point for screening for paternal depression, with a sensitivity of 89.5% and a specificity of 78.2%, compared with a structured clini­cal interview. Fisher et al⁴ determined that the EDPS-P report was a reliable method for detecting paternal PPD compared with validated depression scales completed by fathers. Madsen et al⁵ determined the Gotland Male Depression Scale, which detects typical male depressive symptoms, also was effective in recognizing paternal PPD at 6 weeks postpartum.⁷

Treatment of paternal PPD
Specific treatment for paternal PPD has not been studied extensively. Psychotherapy targeted at interpersonal family relation­ships and parenting is indicated for mild depression, whereas a combination of psy­chotherapy and pharmacotherapy is recom­mended for moderate or severe depression.

Depending on specific patient factors, pharmacotherapy options include selective serotonin reuptake inhibitors (SSRIs), serotonin-norepinephrine reuptake inhibi­tors, tricyclic antidepressants, and atypical antipsychotics.⁸ SSRIs often are used because of their efficacy and relative lack of serious side effects, as demonstrated in numerous trials.² Recovery is more likely with combi­nation therapy than monotherapy.⁹ Fathers with psychosis or suicidal ideation should be referred for inpatient treatment.


Disclosures
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Only recently has paternal postpartum depression (PPD) received much attention. Research has shown that maternal PPD is associated with negative outcomes in the child’s cognitive develop­ment and social and marital problems for the parents. Likewise, depressed fathers are less likely to play outside with their child and more likely to put the child to bed awake.¹

Recent studies reported that 10.4% of men experienced depression within 12 month of delivery.¹ Edmondson et al² estimate the prevalence of paternal PPD to be 8% between birth and 3 months, 26% from 3 to 6 months, and 9% from 6 to 12 months.

Risk factors
Risk factors for paternal PPD have not been studied extensively. Some studies have shown that immaturity, lack of social sup­port, first or unplanned pregnancies, marital relationship problems, and unemployment were the most common risk factors for depression in men postnatally.³ A history of depression and other psychiatric disorders also increases risk.⁴ Psychosocial factors, such as quality of the spousal relationship, parenting distress, and perceived parenting efficacy, contribute to paternal depression.

Similarly, depressed postpartum fathers experience higher levels of parenting dis­tress and a lower sense of parenting effi­cacy.⁵ Interestingly, negative life events were associated with increased risk for depression in mothers, but not fathers.³

Clinical presentation
Paternal PPD symptoms appear within 12 months after the birth of the child and last for at least 2 weeks. Signs and symptoms of depression in men might not resemble those seen in postpartum women. Men tend to show aggression, increased or easy irritabil­ity, and agitation, and might not seek help for emotional issues as readily as women do. Typical symptoms of depression often are present, such as sleep disturbance or changes in sleep patterns, difficulty concen­trating, memory problems, and feelings of worthlessness, hopelessness, inadequacy, and excess guilt with suicidal ideation.⁶

Making the diagnosis
Maternal PPD commonly is evaluated using the Edinburgh Postnatal Depression Scale- Partner (EDPS-P) or Postpartum Depression Screening Scale. However, studies are lack­ing to determine which diagnostic modal­ity is most accurate for diagnosing paternal PPD.

A paternal PPD screening tool could include the EDPS-P administered to moth­ers. Edmonson et al² determined an EDPS-P score of >10 was the optimal cut-off point for screening for paternal depression, with a sensitivity of 89.5% and a specificity of 78.2%, compared with a structured clini­cal interview. Fisher et al⁴ determined that the EDPS-P report was a reliable method for detecting paternal PPD compared with validated depression scales completed by fathers. Madsen et al⁵ determined the Gotland Male Depression Scale, which detects typical male depressive symptoms, also was effective in recognizing paternal PPD at 6 weeks postpartum.⁷

Treatment of paternal PPD
Specific treatment for paternal PPD has not been studied extensively. Psychotherapy targeted at interpersonal family relation­ships and parenting is indicated for mild depression, whereas a combination of psy­chotherapy and pharmacotherapy is recom­mended for moderate or severe depression.

Depending on specific patient factors, pharmacotherapy options include selective serotonin reuptake inhibitors (SSRIs), serotonin-norepinephrine reuptake inhibi­tors, tricyclic antidepressants, and atypical antipsychotics.⁸ SSRIs often are used because of their efficacy and relative lack of serious side effects, as demonstrated in numerous trials.² Recovery is more likely with combi­nation therapy than monotherapy.⁹ Fathers with psychosis or suicidal ideation should be referred for inpatient treatment.


Disclosures
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Only recently has paternal postpartum depression (PPD) received much attention. Research has shown that maternal PPD is associated with negative outcomes in the child’s cognitive develop­ment and social and marital problems for the parents. Likewise, depressed fathers are less likely to play outside with their child and more likely to put the child to bed awake.¹

Recent studies reported that 10.4% of men experienced depression within 12 month of delivery.¹ Edmondson et al² estimate the prevalence of paternal PPD to be 8% between birth and 3 months, 26% from 3 to 6 months, and 9% from 6 to 12 months.

Risk factors
Risk factors for paternal PPD have not been studied extensively. Some studies have shown that immaturity, lack of social sup­port, first or unplanned pregnancies, marital relationship problems, and unemployment were the most common risk factors for depression in men postnatally.³ A history of depression and other psychiatric disorders also increases risk.⁴ Psychosocial factors, such as quality of the spousal relationship, parenting distress, and perceived parenting efficacy, contribute to paternal depression.

Similarly, depressed postpartum fathers experience higher levels of parenting dis­tress and a lower sense of parenting effi­cacy.⁵ Interestingly, negative life events were associated with increased risk for depression in mothers, but not fathers.³

Clinical presentation
Paternal PPD symptoms appear within 12 months after the birth of the child and last for at least 2 weeks. Signs and symptoms of depression in men might not resemble those seen in postpartum women. Men tend to show aggression, increased or easy irritabil­ity, and agitation, and might not seek help for emotional issues as readily as women do. Typical symptoms of depression often are present, such as sleep disturbance or changes in sleep patterns, difficulty concen­trating, memory problems, and feelings of worthlessness, hopelessness, inadequacy, and excess guilt with suicidal ideation.⁶

Making the diagnosis
Maternal PPD commonly is evaluated using the Edinburgh Postnatal Depression Scale- Partner (EDPS-P) or Postpartum Depression Screening Scale. However, studies are lack­ing to determine which diagnostic modal­ity is most accurate for diagnosing paternal PPD.

A paternal PPD screening tool could include the EDPS-P administered to moth­ers. Edmonson et al² determined an EDPS-P score of >10 was the optimal cut-off point for screening for paternal depression, with a sensitivity of 89.5% and a specificity of 78.2%, compared with a structured clini­cal interview. Fisher et al⁴ determined that the EDPS-P report was a reliable method for detecting paternal PPD compared with validated depression scales completed by fathers. Madsen et al⁵ determined the Gotland Male Depression Scale, which detects typical male depressive symptoms, also was effective in recognizing paternal PPD at 6 weeks postpartum.⁷

Treatment of paternal PPD
Specific treatment for paternal PPD has not been studied extensively. Psychotherapy targeted at interpersonal family relation­ships and parenting is indicated for mild depression, whereas a combination of psy­chotherapy and pharmacotherapy is recom­mended for moderate or severe depression.

Depending on specific patient factors, pharmacotherapy options include selective serotonin reuptake inhibitors (SSRIs), serotonin-norepinephrine reuptake inhibi­tors, tricyclic antidepressants, and atypical antipsychotics.⁸ SSRIs often are used because of their efficacy and relative lack of serious side effects, as demonstrated in numerous trials.² Recovery is more likely with combi­nation therapy than monotherapy.⁹ Fathers with psychosis or suicidal ideation should be referred for inpatient treatment.


Disclosures
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Ah! the dream of opening private practice! Whether you’re a resident making less than minimum wage or a clinic employee seeing ever more patients, the allure is powerful. But, just because you’re whip-smart in matters of the mind, doesn’t mean you know how to run a business. To prevent your dream from succumbing to the siren’s allure, you’ll need to create a blueprint that gets you moving today, as well as prepare sys­tems that will endure over the years.

Establish a business model
Before signing a lease or scheduling patients, think through these fundamental questions, not just from a clinical perspec­tive but a business one as well:
   • What kind of care would you like to provide? If you want to practice psycho­therapy and medication, you’ll have fewer time slots to have to fill, but it may be more challenging to find patients who want and can afford psychotherapy from you as well.
   • Where do you want to practice? Time spent commuting rarely produces income, so how close do you want your office to be to where you live? Being able to walk to work is wonderful, but is where you live the best location for your patients?
For example, downtown areas in big cit­ies are good for providing a critical mass of patients, especially if you only want to manage patients’ medications. But if you want to see children and families, you should consider a location that is friendlier for them—usually more residential areas. Having a coffee shop nearby for waiting parents doesn’t hurt. If you work in a rural area, how easily can patients get to your office?
   • Which hours do you want to work? Many patients will want to see you at “prime time”—before or after their work day or during the weekend. This might, not coincidentally, be when you don’t want to work. Consider whether there is room for compromise: Can you work 1 or 2 early or late days? Can you do 1 weekend day once in a while? If you want to see children, can you regularly be available after school?
   • Will you accept insurance? Pros: The insurance companies will do the marketing for you; your practice will fill quickly; their checks don’t bounce; and, 98% of the time, the claims and payment process works just fine.
Cons: You will make less money per patient, in return for the higher volume of patients that are sent your way; the insur­ance companies won’t want to pay you more than they pay non-psychiatrists for psychotherapy; and the small amount of time that there are administrative problems can consume a disproportionate share of your sanity.

Run the numbers carefully
Next, think about the financial aspect. How much do you need to make, after you’ve paid business expenses and taxes, to be content? You might be tempted to work as many hours as possible, think­ing that every hour off is an hour that you could have billed. Shifting your viewpoint from “hours lost” to “hours free” is a nec­essary approach to reduce burnout.

Once you have figured out your finan­cial goal, do the math: multiply hours/ week × hourly rate × how many weeks/ year you’ll work to determine your annual income. Play around with the numbers to test your priorities, such as optimiz­ing daily hours vs vacation time vs charg­ing more or less.

Build your brand
This is your professional identity—the pic­ture of your practice that your colleagues and future patients will see and that will start to get those hours filled. How will you convey your strengths and personal­ity? The answer: Get out of the office.
   • Take clinicians who will refer patients to you out to lunch (and pick up the tab).
   • Give free talks to psychotherapists or primary care providers. Grand rounds, group practice meetings, or local clinical associations are potential venues. Give the organizer a menu of topic options that con­nect your clinical interests and theirs, and then create a dynamic presentation based on their feedback. Tip: Do not PowerPoint them to tears.
   • Start blogging. If you enjoy writing, use a blog to showcase your talent and expertise. It is free advertising and makes you seem like a trusted authority. However, don’t start a blog unless you can commit to posting regularly.

Proceed thoughtfully; seek advice
As you think through the matrix of issues presented above, each set of answers may lead to a deeper set of questions. Consultation with a colleague or mentor can save you valuable time. Although you don’t have to have all the answers before you open your practice, spending time thinking through these and other issues beforehand will optimize the chance that your dream becomes a reality.

 

Disclosure
Dr. Braslow is the founder of Luminello.com.

Ah! the dream of opening private practice! Whether you’re a resident making less than minimum wage or a clinic employee seeing ever more patients, the allure is powerful. But, just because you’re whip-smart in matters of the mind, doesn’t mean you know how to run a business. To prevent your dream from succumbing to the siren’s allure, you’ll need to create a blueprint that gets you moving today, as well as prepare sys­tems that will endure over the years.

Establish a business model
Before signing a lease or scheduling patients, think through these fundamental questions, not just from a clinical perspec­tive but a business one as well:
   • What kind of care would you like to provide? If you want to practice psycho­therapy and medication, you’ll have fewer time slots to have to fill, but it may be more challenging to find patients who want and can afford psychotherapy from you as well.
   • Where do you want to practice? Time spent commuting rarely produces income, so how close do you want your office to be to where you live? Being able to walk to work is wonderful, but is where you live the best location for your patients?
For example, downtown areas in big cit­ies are good for providing a critical mass of patients, especially if you only want to manage patients’ medications. But if you want to see children and families, you should consider a location that is friendlier for them—usually more residential areas. Having a coffee shop nearby for waiting parents doesn’t hurt. If you work in a rural area, how easily can patients get to your office?
   • Which hours do you want to work? Many patients will want to see you at “prime time”—before or after their work day or during the weekend. This might, not coincidentally, be when you don’t want to work. Consider whether there is room for compromise: Can you work 1 or 2 early or late days? Can you do 1 weekend day once in a while? If you want to see children, can you regularly be available after school?
   • Will you accept insurance? Pros: The insurance companies will do the marketing for you; your practice will fill quickly; their checks don’t bounce; and, 98% of the time, the claims and payment process works just fine.
Cons: You will make less money per patient, in return for the higher volume of patients that are sent your way; the insur­ance companies won’t want to pay you more than they pay non-psychiatrists for psychotherapy; and the small amount of time that there are administrative problems can consume a disproportionate share of your sanity.

Run the numbers carefully
Next, think about the financial aspect. How much do you need to make, after you’ve paid business expenses and taxes, to be content? You might be tempted to work as many hours as possible, think­ing that every hour off is an hour that you could have billed. Shifting your viewpoint from “hours lost” to “hours free” is a nec­essary approach to reduce burnout.

Once you have figured out your finan­cial goal, do the math: multiply hours/ week × hourly rate × how many weeks/ year you’ll work to determine your annual income. Play around with the numbers to test your priorities, such as optimiz­ing daily hours vs vacation time vs charg­ing more or less.

Build your brand
This is your professional identity—the pic­ture of your practice that your colleagues and future patients will see and that will start to get those hours filled. How will you convey your strengths and personal­ity? The answer: Get out of the office.
   • Take clinicians who will refer patients to you out to lunch (and pick up the tab).
   • Give free talks to psychotherapists or primary care providers. Grand rounds, group practice meetings, or local clinical associations are potential venues. Give the organizer a menu of topic options that con­nect your clinical interests and theirs, and then create a dynamic presentation based on their feedback. Tip: Do not PowerPoint them to tears.
   • Start blogging. If you enjoy writing, use a blog to showcase your talent and expertise. It is free advertising and makes you seem like a trusted authority. However, don’t start a blog unless you can commit to posting regularly.

Proceed thoughtfully; seek advice
As you think through the matrix of issues presented above, each set of answers may lead to a deeper set of questions. Consultation with a colleague or mentor can save you valuable time. Although you don’t have to have all the answers before you open your practice, spending time thinking through these and other issues beforehand will optimize the chance that your dream becomes a reality.

 

Disclosure
Dr. Braslow is the founder of Luminello.com.

Ah! the dream of opening private practice! Whether you’re a resident making less than minimum wage or a clinic employee seeing ever more patients, the allure is powerful. But, just because you’re whip-smart in matters of the mind, doesn’t mean you know how to run a business. To prevent your dream from succumbing to the siren’s allure, you’ll need to create a blueprint that gets you moving today, as well as prepare sys­tems that will endure over the years.

Establish a business model
Before signing a lease or scheduling patients, think through these fundamental questions, not just from a clinical perspec­tive but a business one as well:
   • What kind of care would you like to provide? If you want to practice psycho­therapy and medication, you’ll have fewer time slots to have to fill, but it may be more challenging to find patients who want and can afford psychotherapy from you as well.
   • Where do you want to practice? Time spent commuting rarely produces income, so how close do you want your office to be to where you live? Being able to walk to work is wonderful, but is where you live the best location for your patients?
For example, downtown areas in big cit­ies are good for providing a critical mass of patients, especially if you only want to manage patients’ medications. But if you want to see children and families, you should consider a location that is friendlier for them—usually more residential areas. Having a coffee shop nearby for waiting parents doesn’t hurt. If you work in a rural area, how easily can patients get to your office?
   • Which hours do you want to work? Many patients will want to see you at “prime time”—before or after their work day or during the weekend. This might, not coincidentally, be when you don’t want to work. Consider whether there is room for compromise: Can you work 1 or 2 early or late days? Can you do 1 weekend day once in a while? If you want to see children, can you regularly be available after school?
   • Will you accept insurance? Pros: The insurance companies will do the marketing for you; your practice will fill quickly; their checks don’t bounce; and, 98% of the time, the claims and payment process works just fine.
Cons: You will make less money per patient, in return for the higher volume of patients that are sent your way; the insur­ance companies won’t want to pay you more than they pay non-psychiatrists for psychotherapy; and the small amount of time that there are administrative problems can consume a disproportionate share of your sanity.

Run the numbers carefully
Next, think about the financial aspect. How much do you need to make, after you’ve paid business expenses and taxes, to be content? You might be tempted to work as many hours as possible, think­ing that every hour off is an hour that you could have billed. Shifting your viewpoint from “hours lost” to “hours free” is a nec­essary approach to reduce burnout.

Once you have figured out your finan­cial goal, do the math: multiply hours/ week × hourly rate × how many weeks/ year you’ll work to determine your annual income. Play around with the numbers to test your priorities, such as optimiz­ing daily hours vs vacation time vs charg­ing more or less.

Build your brand
This is your professional identity—the pic­ture of your practice that your colleagues and future patients will see and that will start to get those hours filled. How will you convey your strengths and personal­ity? The answer: Get out of the office.
   • Take clinicians who will refer patients to you out to lunch (and pick up the tab).
   • Give free talks to psychotherapists or primary care providers. Grand rounds, group practice meetings, or local clinical associations are potential venues. Give the organizer a menu of topic options that con­nect your clinical interests and theirs, and then create a dynamic presentation based on their feedback. Tip: Do not PowerPoint them to tears.
   • Start blogging. If you enjoy writing, use a blog to showcase your talent and expertise. It is free advertising and makes you seem like a trusted authority. However, don’t start a blog unless you can commit to posting regularly.

Proceed thoughtfully; seek advice
As you think through the matrix of issues presented above, each set of answers may lead to a deeper set of questions. Consultation with a colleague or mentor can save you valuable time. Although you don’t have to have all the answers before you open your practice, spending time thinking through these and other issues beforehand will optimize the chance that your dream becomes a reality.

 

Disclosure
Dr. Braslow is the founder of Luminello.com.

CASE Confused and weak
Mr. W, age 26, is brought to the emergency department (ED) by his parents for intermit­tent confusion, weakness, and increasing lethargy over the past 4 days. He is jaun­diced with mild abdominal pain, nausea, and vomiting.

Mr. W has a history of alcohol use disorder, drinking as much as 1 L of vodka a day. Six months ago, he was hospitalized for alcoholic hepatitis and severe hyponatremia.

In the ED, Mr. W is awake, alert, and ori­ented to person, place, and time. Vital signs are: pulse 89 beats per minute; blood pres­sure, 117/50 mm Hg; respirations, 15 breaths per minute; and temperature, 98.5ºF. Physical examination is notable for scleral icterus, jaun­dice, tender hepatomegaly, and asterixis.

Mr. W is not taking any medications. He reports that his most recent drink was the day before; however, his current alcohol intake is unknown.

Laboratory tests reveal altered hepatic func­tion, including elevated aspartate aminotrans­ferase (251 U/L), alanine aminotransferase (56 U/L), alkaline phosphatase (179 U/L), total bilirubin (15.4 mg/dL), and ammonia (143 U/L), impaired coagulation (international normalized ratio 2.39), and decreased albumin (2.7 g/dL). Other metabolic disturbances include: sodium, 104 mEq/L; chloride, <60 mEq/L; potassium, 2.2 mEq/L; and CO2, 44.5 mEq/L.

What is your differential diagnosis for Mr. W’s altered mental status?
   a) hepatic encephalopathy
   b) Wernicke’s encephalopathy
   c) hyponatremia
   d) drug intoxication
   e) head trauma

The authors’ observations
Hyponatremia is defined as a serum sodium concentration <136 mEq/L. Mr. W is considered to have severe hyponatremia because his serum sodium concentration is <125 mEq/L. Although commonly caused by an inability to suppress antidiuretic hormone, hyponatremia has several pos­sible causes (Figure 1).¹ Symptoms are nonspecific and are more visible when there is a large or rapid decrease in the serum sodium concentration. Most patients with a serum sodium concentration >125 mEq/L are asymptomatic. Mr. W, who had a serum sodium of 104 mEq/L, presented with several symptoms, includ­ing confusion, lethargy, nausea, vomiting, and weakness. Headache, muscle spasms, depressed reflexes, restlessness, and disori­entation also might be observed.²

Because of Mr. W’s impaired hepatic func­tion, elevated ammonia, and asterixis, hepatic encephalopathy could be contributing to his altered mental status. Suspect Wernicke’s encephalopathy in a patient with neurologic symptoms and a history of chronic alcohol abuse. In its classic form, Wernicke’s enceph­alopathy has acute onset, characterized by the triad of ataxia, global confusion, and ocu­lar abnormalities. However, this triad is not consistently or frequently encountered.³

Which tests would you order next?
   a) blood ethanol level
   b) urine drug screen
   c) serum osmolality
   d) CT of the head

 

EVALUATION Sober, yet sick
To rule out intoxication as the cause of Mr. W’s altered mental status, blood ethanol level and urine drug screens are obtained and found to be negative. CT of the head is negative for acute intracranial pathology.

Mr. W is admitted to the medical intensive care unit (MICU) for severe hyponatremia and altered mental status. Serum osmolality is 220 mOsm/kg (normal range 281 to 304 mOsm/kg). To further classify his hypo­tonic hyponatremia, volume status is assessed, and Mr. W is determined to be euvolemic. Thyroid-stimulating hormone and cortisol are within normal limits, eliminating hypothy­roidism and adrenal insufficiency as causes of his euvolemic hypotonic hyponatremia. Mr. W is treated for hyponatremia likely sec­ondary to syndrome of inappropriate antidi­uretic hormone (SIADH). SIADH is a diagnosis of exclusion that first requires ruling out hypo­thyroidism and glucocorticoid insufficiency (Figure 1).¹

The authors’ observations
Because hypokalemia is an independent pre­dictive factor for development of hyponatre­mia, it is necessary to evaluate the potassium level in all hyponatremic patients. Mr. W’s potassium level was 2.2 mEq/L on admis­sion. Serum sodium concentration is related to total exchangeable sodium, total body water, and total exchangeable potassium. Potassium depletion causes a shift of sodium into cells with a comparable exit of potas­sium from cells into extracellular fluid. The reverse process occurs during potassium repletion, leading to an increase in serum sodium concentration and making hypoka­lemia a risk factor for developing osmotic demyelination syndrome (ODS).⁴

Treating hyponatremia
Hyponatremia treatment depends on its severity, presence or absence of symptoms, and whether the hyponatremia is acute (<24 hours) or chronic (>48 hours).⁵

Because of Mr. W’s extremely low serum sodium concentration, predisposition to hyponatremia secondary to alcoholism, and history of severe hyponatremia, it is likely he is chronically hyponatremic.

In patients with chronic hyponatremia, neurological sequelae are associated with the need for a more rapid rate of correction of serum sodium. For most patients with chronic hyponatremia, a correction rate of ≤10 to 12 mEq/L during the first 24 hours and <18 mEq/L over 48 hours is recom­mended to avoid ODS.⁶

 

Evidence suggests, however, that this 1-day limit might be too high for some patients. Alcoholism, hypokalemia, mal­nutrition, and liver disease are present in a high percentage of patients who develop

ODS after correcting hyponatremia (Table 1).⁶ Therefore, for patients such as Mr. W who are at high risk of ODS, experts recommend a goal of 4 to 6 mEq/L/d with a correction rate of ≤8 mEq/L in any 24-hour period (Table 2).⁶

 

TREATMENT Sodium normalizes
Mr. W receives 1 L of normal saline in the ED before admission to the MICU. Once in the MICU, despite likely chronic hyponatremia, he receives hypertonic (3%) saline, followed by normal saline. Initially, he responds when the serum sodium concentration improves. Because of his likely SIADH, Mr. W is fluid-restricted for 4 days. Serum sodium returns to normal over 7 hospital days (Figure 2). To address the pro­found hypokalemia, Mr. S receives 30 mEq of potassium chloride in the ED, and potas­sium is repeated daily throughout his stay in the MICU.

Mr. W remains lethargic, with intermittent periods of confusion throughout the hospital stay. His altered mental status is attributed to hepatic encephalopathy secondary to alco­holic hepatitis. The Maddrey discriminant function is a calculation that stratifies patients with alcoholic hepatitis for risk of mortal­ity and the use of steroids. Because Mr. W shows a Maddrey discriminant function ≥32, he receives methylprednisolone, followed by pentoxifylline, and liver function tests trend down. He also receives lactulose throughout hospitalization.

By discharge on hospital day 9, Mr. W’s serum sodium is 138 mEq/L; serum potas­sium, 4.1 mEq/L. Total bilirubin and prothrom­bin remain elevated. Mr. W is discharged on lactulose, thiamine, folic acid, and a 1-month course of pentoxifylline, 400 mg, 3 times a day.

READMISSION Unsteady gait, nausea
Three days after discharge, Mr. W returns to the ED after experiencing a 20-second epi­sode of total body rigidity. He has an unsteady gait and worsening nausea and vomiting.

When Mr. W arrives in the ED, he confirms he is taking his discharge medications as pre­scribed. His parents report that he has con­sumed alcohol and Cannabis since discharge and has been taking his sibling’s prescription medications, including quetiapine.

In the ED, Mr. W is awake, alert, and ori­ented to person, place, and time. Vital signs are: pulse, 118 beats per minute; blood pres­sure, 128/73 mm Hg; respirations, 16 breaths per minute; and temperature, 98.5ºF. Physical examination, again, is notable for scleral icterus, jaundice, and asterixis. No focal neu­rologic deficits are noted.

Consistent with Mr. W’s previous admis­sion, laboratory values reveal altered hepatic function and impaired coagulation. The serum sodium level remains within normal limits at 136 mEq/L. However, again, meta­bolic disturbances include decreased chloride (97 mEq/L), potassium (2.9 mEq/L), and CO2 (18.2 mEq/L). CT on readmission is unchanged from the earlier hospitalization.

What is your differential diagnosis for Mr. W’s total body rigidity?
   a) seizure
   b) ODS
   c) drug intoxication
   d) neuroleptic malignant syndrome

EVALUATION Shaking and weakness
Once admitted to the hospital, Mr. W reports an episode of right upper-extremity “shak­ing,” followed by weakness. He remembers the entire event and denies tongue biting or incontinence. He is evaluated for pos­sible seizure, given his multiple risk factors, including drug and alcohol use, ingestion of quetiapine, and history of hyponatremia. Routine EEG is negative but prolactin level is elevated.

Mr. W’s mental status continues to wax and wane, prompting a neurology consult and MRI for further evaluation. MRI of the brain without contrast reveals restricted diffusion in the pons centrally, with extension bilaterally to the midbrain and thalami—findings con­sistent with central pontine myelinolysis. A neurology consultation reveals quadriparesis, paraparesis, dysarthria, and diplopia on exam­ination, all symptoms associated with central pontine myelinolysis.

The authors’ observations
ODS, including central and extrapontine myelinolysis, is a demyelinating condi­tion that occurs because of severe osmotic stress, most commonly secondary to the overly rapid correction of hyponatremia in patients with conditions leading to nutritional or electrolyte stress.⁷ Mr. W is considered at high risk of developing ODS because he fulfills the 5 criteria listed in Table 1.

Several psychiatric illnesses and neu­ropsychiatric medications could lead to hyponatremia. Many studies^8-10 have docu­mented hyponatremia and resulting ODS in patients with alcoholism, schizophrenia, anorexia, primary psychogenic polydipsia, and MDMA (3,4-methylenedioxymetham­phetamine) abuse. Hyponatremia is a side effect of several neuropsychiatric medica­tions, including serotonin reuptake inhibi­tors, lithium, tricyclic antidepressants, opioids, carbamazepine, oxcarbazepine, and antipsychotic polypharmacy. Other commonly used medications associated with hyponatremia include salt-losing diuretics, nonsteroidal anti-inflammatory drugs, and acetaminophen.⁷

Disease severity varies from asymptom­atic to coma or death. Symptoms, although some could reverse completely, typically are a combination of neuropsychiatric (ie, emotional lability, disinhibition, and other bizarre behaviors) and neurologic. Neurologic symptoms include confusion, impaired cognition, dysarthria, dyspha­gia, gait instability, weakness or paralysis, and generalized seizures. Severely affected patients could experience “locked-in syn­drome,” in which they are awake but unable to move or communicate. Also consistent with Mr. W’s case, ODS often presents ini­tially with delirium, seizures, or encepha­lopathy, followed by a lucid interval before symptoms develop.⁷

 

Diagnosis is based on the appearance of demyelinating brain lesions on CT or MRI. MRI is more sensitive than CT; however, even an MRI scan can appear normal for as long as 4 weeks after symptoms appear.⁷ Therefore, an initial negative radiologic study in a high-risk patient who develops neurologic symptoms does not exclude ODS. Earlier detection is possible with dif­fusion-weighted MRI, which is most sensi­tive and can detect lesions within 24 hours of developing symptoms.¹¹ The severity of the lesion does not correlate with severity of symptoms.

Studies reveal a considerable range in prognosis of patients with clinically symp­tomatic ODS. A study of 44 patients with central pontine myelinolysis, of which 42 had chronic alcoholism, reported that 34% had no significant functional deficits at follow-up, 34% had minor neurologic defi­cits, and 31% became dependent on per­sonal help. Outcome did not depend on the extent or severity of neurologic symptoms or the severity of concomitant systemic complications.¹²

Because of its poor prognosis, prevention of ODS is important. Because ODS com­monly is caused by overly rapid correction of hyponatremia, it is necessary to adhere to guidelines for treating chronic hypona­tremia (Table 2). If overcorrection occurs, therapeutic re-lowering of serum sodium can be considered, but has not been validated in controlled trials. Based mainly on case reports that suggest benefit from early re-lowering serum sodium in patients with ODS symptoms, experts recommend the following:  
   • administer desmopressin, 2 to 4 μg, every 8 hours parenterally  
   • replace water orally or as 5% dextrose in water intravenously (3 mL/kg/hr)  
   • check serum sodium hourly until serum is reduced to goal.⁶

Bottom Line

Hyponatremia is the most common electrolyte disorder encountered in practice. Osmotic demyelination syndrome often is preventable, with considerable morbidity and mortality. Psychiatrists should be aware of this condition because it could be an adverse effect of many psychiatric medications and there are some psychiatric illnesses in which hyponatremia is a potential risk. In hyponatremic patients with persistent nonspecific neurologic or neuropsychiatric symptoms and negative CT imaging, additional imaging, such as MRI, is warranted.

Related Resources

• Braun MM, Barstow CH, Pyzocha NJ. Diagnosis and management of sodium disorders: hyponatremia and hypernatremia. Am Fam Physician. 2015;91(5):299-307.
• Vaidya C, Ho W, Freda BJ. Management of hyponatremia: providing treatment and avoiding harm. Cleve Clin J Med. 2010;77(10):715-726.

Drug Brand Names
Carbamazepine • Tegretol
Oxcarbazepine • Trileptal
Desmopressin • Stimate, DDAVP
Lithium • Eskalith, Lithobid
Pentoxifylline • Trental, Pentoxil
Methylprednisolone • Medrol
Quetiapine • Seroquel

Disclosure
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

CASE Confused and weak
Mr. W, age 26, is brought to the emergency department (ED) by his parents for intermit­tent confusion, weakness, and increasing lethargy over the past 4 days. He is jaun­diced with mild abdominal pain, nausea, and vomiting.

Mr. W has a history of alcohol use disorder, drinking as much as 1 L of vodka a day. Six months ago, he was hospitalized for alcoholic hepatitis and severe hyponatremia.

In the ED, Mr. W is awake, alert, and ori­ented to person, place, and time. Vital signs are: pulse 89 beats per minute; blood pres­sure, 117/50 mm Hg; respirations, 15 breaths per minute; and temperature, 98.5ºF. Physical examination is notable for scleral icterus, jaun­dice, tender hepatomegaly, and asterixis.

Mr. W is not taking any medications. He reports that his most recent drink was the day before; however, his current alcohol intake is unknown.

Laboratory tests reveal altered hepatic func­tion, including elevated aspartate aminotrans­ferase (251 U/L), alanine aminotransferase (56 U/L), alkaline phosphatase (179 U/L), total bilirubin (15.4 mg/dL), and ammonia (143 U/L), impaired coagulation (international normalized ratio 2.39), and decreased albumin (2.7 g/dL). Other metabolic disturbances include: sodium, 104 mEq/L; chloride, <60 mEq/L; potassium, 2.2 mEq/L; and CO2, 44.5 mEq/L.

What is your differential diagnosis for Mr. W’s altered mental status?
   a) hepatic encephalopathy
   b) Wernicke’s encephalopathy
   c) hyponatremia
   d) drug intoxication
   e) head trauma

The authors’ observations
Hyponatremia is defined as a serum sodium concentration <136 mEq/L. Mr. W is considered to have severe hyponatremia because his serum sodium concentration is <125 mEq/L. Although commonly caused by an inability to suppress antidiuretic hormone, hyponatremia has several pos­sible causes (Figure 1).¹ Symptoms are nonspecific and are more visible when there is a large or rapid decrease in the serum sodium concentration. Most patients with a serum sodium concentration >125 mEq/L are asymptomatic. Mr. W, who had a serum sodium of 104 mEq/L, presented with several symptoms, includ­ing confusion, lethargy, nausea, vomiting, and weakness. Headache, muscle spasms, depressed reflexes, restlessness, and disori­entation also might be observed.²

Because of Mr. W’s impaired hepatic func­tion, elevated ammonia, and asterixis, hepatic encephalopathy could be contributing to his altered mental status. Suspect Wernicke’s encephalopathy in a patient with neurologic symptoms and a history of chronic alcohol abuse. In its classic form, Wernicke’s enceph­alopathy has acute onset, characterized by the triad of ataxia, global confusion, and ocu­lar abnormalities. However, this triad is not consistently or frequently encountered.³

Which tests would you order next?
   a) blood ethanol level
   b) urine drug screen
   c) serum osmolality
   d) CT of the head

 

EVALUATION Sober, yet sick
To rule out intoxication as the cause of Mr. W’s altered mental status, blood ethanol level and urine drug screens are obtained and found to be negative. CT of the head is negative for acute intracranial pathology.

Mr. W is admitted to the medical intensive care unit (MICU) for severe hyponatremia and altered mental status. Serum osmolality is 220 mOsm/kg (normal range 281 to 304 mOsm/kg). To further classify his hypo­tonic hyponatremia, volume status is assessed, and Mr. W is determined to be euvolemic. Thyroid-stimulating hormone and cortisol are within normal limits, eliminating hypothy­roidism and adrenal insufficiency as causes of his euvolemic hypotonic hyponatremia. Mr. W is treated for hyponatremia likely sec­ondary to syndrome of inappropriate antidi­uretic hormone (SIADH). SIADH is a diagnosis of exclusion that first requires ruling out hypo­thyroidism and glucocorticoid insufficiency (Figure 1).¹

The authors’ observations
Because hypokalemia is an independent pre­dictive factor for development of hyponatre­mia, it is necessary to evaluate the potassium level in all hyponatremic patients. Mr. W’s potassium level was 2.2 mEq/L on admis­sion. Serum sodium concentration is related to total exchangeable sodium, total body water, and total exchangeable potassium. Potassium depletion causes a shift of sodium into cells with a comparable exit of potas­sium from cells into extracellular fluid. The reverse process occurs during potassium repletion, leading to an increase in serum sodium concentration and making hypoka­lemia a risk factor for developing osmotic demyelination syndrome (ODS).⁴

Treating hyponatremia
Hyponatremia treatment depends on its severity, presence or absence of symptoms, and whether the hyponatremia is acute (<24 hours) or chronic (>48 hours).⁵

Because of Mr. W’s extremely low serum sodium concentration, predisposition to hyponatremia secondary to alcoholism, and history of severe hyponatremia, it is likely he is chronically hyponatremic.

In patients with chronic hyponatremia, neurological sequelae are associated with the need for a more rapid rate of correction of serum sodium. For most patients with chronic hyponatremia, a correction rate of ≤10 to 12 mEq/L during the first 24 hours and <18 mEq/L over 48 hours is recom­mended to avoid ODS.⁶

 

Evidence suggests, however, that this 1-day limit might be too high for some patients. Alcoholism, hypokalemia, mal­nutrition, and liver disease are present in a high percentage of patients who develop

ODS after correcting hyponatremia (Table 1).⁶ Therefore, for patients such as Mr. W who are at high risk of ODS, experts recommend a goal of 4 to 6 mEq/L/d with a correction rate of ≤8 mEq/L in any 24-hour period (Table 2).⁶

 

TREATMENT Sodium normalizes
Mr. W receives 1 L of normal saline in the ED before admission to the MICU. Once in the MICU, despite likely chronic hyponatremia, he receives hypertonic (3%) saline, followed by normal saline. Initially, he responds when the serum sodium concentration improves. Because of his likely SIADH, Mr. W is fluid-restricted for 4 days. Serum sodium returns to normal over 7 hospital days (Figure 2). To address the pro­found hypokalemia, Mr. S receives 30 mEq of potassium chloride in the ED, and potas­sium is repeated daily throughout his stay in the MICU.

Mr. W remains lethargic, with intermittent periods of confusion throughout the hospital stay. His altered mental status is attributed to hepatic encephalopathy secondary to alco­holic hepatitis. The Maddrey discriminant function is a calculation that stratifies patients with alcoholic hepatitis for risk of mortal­ity and the use of steroids. Because Mr. W shows a Maddrey discriminant function ≥32, he receives methylprednisolone, followed by pentoxifylline, and liver function tests trend down. He also receives lactulose throughout hospitalization.

By discharge on hospital day 9, Mr. W’s serum sodium is 138 mEq/L; serum potas­sium, 4.1 mEq/L. Total bilirubin and prothrom­bin remain elevated. Mr. W is discharged on lactulose, thiamine, folic acid, and a 1-month course of pentoxifylline, 400 mg, 3 times a day.

READMISSION Unsteady gait, nausea
Three days after discharge, Mr. W returns to the ED after experiencing a 20-second epi­sode of total body rigidity. He has an unsteady gait and worsening nausea and vomiting.

When Mr. W arrives in the ED, he confirms he is taking his discharge medications as pre­scribed. His parents report that he has con­sumed alcohol and Cannabis since discharge and has been taking his sibling’s prescription medications, including quetiapine.

In the ED, Mr. W is awake, alert, and ori­ented to person, place, and time. Vital signs are: pulse, 118 beats per minute; blood pres­sure, 128/73 mm Hg; respirations, 16 breaths per minute; and temperature, 98.5ºF. Physical examination, again, is notable for scleral icterus, jaundice, and asterixis. No focal neu­rologic deficits are noted.

Consistent with Mr. W’s previous admis­sion, laboratory values reveal altered hepatic function and impaired coagulation. The serum sodium level remains within normal limits at 136 mEq/L. However, again, meta­bolic disturbances include decreased chloride (97 mEq/L), potassium (2.9 mEq/L), and CO2 (18.2 mEq/L). CT on readmission is unchanged from the earlier hospitalization.

What is your differential diagnosis for Mr. W’s total body rigidity?
   a) seizure
   b) ODS
   c) drug intoxication
   d) neuroleptic malignant syndrome

EVALUATION Shaking and weakness
Once admitted to the hospital, Mr. W reports an episode of right upper-extremity “shak­ing,” followed by weakness. He remembers the entire event and denies tongue biting or incontinence. He is evaluated for pos­sible seizure, given his multiple risk factors, including drug and alcohol use, ingestion of quetiapine, and history of hyponatremia. Routine EEG is negative but prolactin level is elevated.

Mr. W’s mental status continues to wax and wane, prompting a neurology consult and MRI for further evaluation. MRI of the brain without contrast reveals restricted diffusion in the pons centrally, with extension bilaterally to the midbrain and thalami—findings con­sistent with central pontine myelinolysis. A neurology consultation reveals quadriparesis, paraparesis, dysarthria, and diplopia on exam­ination, all symptoms associated with central pontine myelinolysis.

The authors’ observations
ODS, including central and extrapontine myelinolysis, is a demyelinating condi­tion that occurs because of severe osmotic stress, most commonly secondary to the overly rapid correction of hyponatremia in patients with conditions leading to nutritional or electrolyte stress.⁷ Mr. W is considered at high risk of developing ODS because he fulfills the 5 criteria listed in Table 1.

Several psychiatric illnesses and neu­ropsychiatric medications could lead to hyponatremia. Many studies^8-10 have docu­mented hyponatremia and resulting ODS in patients with alcoholism, schizophrenia, anorexia, primary psychogenic polydipsia, and MDMA (3,4-methylenedioxymetham­phetamine) abuse. Hyponatremia is a side effect of several neuropsychiatric medica­tions, including serotonin reuptake inhibi­tors, lithium, tricyclic antidepressants, opioids, carbamazepine, oxcarbazepine, and antipsychotic polypharmacy. Other commonly used medications associated with hyponatremia include salt-losing diuretics, nonsteroidal anti-inflammatory drugs, and acetaminophen.⁷

Disease severity varies from asymptom­atic to coma or death. Symptoms, although some could reverse completely, typically are a combination of neuropsychiatric (ie, emotional lability, disinhibition, and other bizarre behaviors) and neurologic. Neurologic symptoms include confusion, impaired cognition, dysarthria, dyspha­gia, gait instability, weakness or paralysis, and generalized seizures. Severely affected patients could experience “locked-in syn­drome,” in which they are awake but unable to move or communicate. Also consistent with Mr. W’s case, ODS often presents ini­tially with delirium, seizures, or encepha­lopathy, followed by a lucid interval before symptoms develop.⁷

 

Diagnosis is based on the appearance of demyelinating brain lesions on CT or MRI. MRI is more sensitive than CT; however, even an MRI scan can appear normal for as long as 4 weeks after symptoms appear.⁷ Therefore, an initial negative radiologic study in a high-risk patient who develops neurologic symptoms does not exclude ODS. Earlier detection is possible with dif­fusion-weighted MRI, which is most sensi­tive and can detect lesions within 24 hours of developing symptoms.¹¹ The severity of the lesion does not correlate with severity of symptoms.

Studies reveal a considerable range in prognosis of patients with clinically symp­tomatic ODS. A study of 44 patients with central pontine myelinolysis, of which 42 had chronic alcoholism, reported that 34% had no significant functional deficits at follow-up, 34% had minor neurologic defi­cits, and 31% became dependent on per­sonal help. Outcome did not depend on the extent or severity of neurologic symptoms or the severity of concomitant systemic complications.¹²

Because of its poor prognosis, prevention of ODS is important. Because ODS com­monly is caused by overly rapid correction of hyponatremia, it is necessary to adhere to guidelines for treating chronic hypona­tremia (Table 2). If overcorrection occurs, therapeutic re-lowering of serum sodium can be considered, but has not been validated in controlled trials. Based mainly on case reports that suggest benefit from early re-lowering serum sodium in patients with ODS symptoms, experts recommend the following:  
   • administer desmopressin, 2 to 4 μg, every 8 hours parenterally  
   • replace water orally or as 5% dextrose in water intravenously (3 mL/kg/hr)  
   • check serum sodium hourly until serum is reduced to goal.⁶

Bottom Line

Hyponatremia is the most common electrolyte disorder encountered in practice. Osmotic demyelination syndrome often is preventable, with considerable morbidity and mortality. Psychiatrists should be aware of this condition because it could be an adverse effect of many psychiatric medications and there are some psychiatric illnesses in which hyponatremia is a potential risk. In hyponatremic patients with persistent nonspecific neurologic or neuropsychiatric symptoms and negative CT imaging, additional imaging, such as MRI, is warranted.

Related Resources

• Braun MM, Barstow CH, Pyzocha NJ. Diagnosis and management of sodium disorders: hyponatremia and hypernatremia. Am Fam Physician. 2015;91(5):299-307.
• Vaidya C, Ho W, Freda BJ. Management of hyponatremia: providing treatment and avoiding harm. Cleve Clin J Med. 2010;77(10):715-726.

Drug Brand Names
Carbamazepine • Tegretol
Oxcarbazepine • Trileptal
Desmopressin • Stimate, DDAVP
Lithium • Eskalith, Lithobid
Pentoxifylline • Trental, Pentoxil
Methylprednisolone • Medrol
Quetiapine • Seroquel

Disclosure
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

CASE Confused and weak
Mr. W, age 26, is brought to the emergency department (ED) by his parents for intermit­tent confusion, weakness, and increasing lethargy over the past 4 days. He is jaun­diced with mild abdominal pain, nausea, and vomiting.

Mr. W has a history of alcohol use disorder, drinking as much as 1 L of vodka a day. Six months ago, he was hospitalized for alcoholic hepatitis and severe hyponatremia.

In the ED, Mr. W is awake, alert, and ori­ented to person, place, and time. Vital signs are: pulse 89 beats per minute; blood pres­sure, 117/50 mm Hg; respirations, 15 breaths per minute; and temperature, 98.5ºF. Physical examination is notable for scleral icterus, jaun­dice, tender hepatomegaly, and asterixis.

Mr. W is not taking any medications. He reports that his most recent drink was the day before; however, his current alcohol intake is unknown.

Laboratory tests reveal altered hepatic func­tion, including elevated aspartate aminotrans­ferase (251 U/L), alanine aminotransferase (56 U/L), alkaline phosphatase (179 U/L), total bilirubin (15.4 mg/dL), and ammonia (143 U/L), impaired coagulation (international normalized ratio 2.39), and decreased albumin (2.7 g/dL). Other metabolic disturbances include: sodium, 104 mEq/L; chloride, <60 mEq/L; potassium, 2.2 mEq/L; and CO2, 44.5 mEq/L.

What is your differential diagnosis for Mr. W’s altered mental status?
   a) hepatic encephalopathy
   b) Wernicke’s encephalopathy
   c) hyponatremia
   d) drug intoxication
   e) head trauma

The authors’ observations
Hyponatremia is defined as a serum sodium concentration <136 mEq/L. Mr. W is considered to have severe hyponatremia because his serum sodium concentration is <125 mEq/L. Although commonly caused by an inability to suppress antidiuretic hormone, hyponatremia has several pos­sible causes (Figure 1).¹ Symptoms are nonspecific and are more visible when there is a large or rapid decrease in the serum sodium concentration. Most patients with a serum sodium concentration >125 mEq/L are asymptomatic. Mr. W, who had a serum sodium of 104 mEq/L, presented with several symptoms, includ­ing confusion, lethargy, nausea, vomiting, and weakness. Headache, muscle spasms, depressed reflexes, restlessness, and disori­entation also might be observed.²

Because of Mr. W’s impaired hepatic func­tion, elevated ammonia, and asterixis, hepatic encephalopathy could be contributing to his altered mental status. Suspect Wernicke’s encephalopathy in a patient with neurologic symptoms and a history of chronic alcohol abuse. In its classic form, Wernicke’s enceph­alopathy has acute onset, characterized by the triad of ataxia, global confusion, and ocu­lar abnormalities. However, this triad is not consistently or frequently encountered.³

Which tests would you order next?
   a) blood ethanol level
   b) urine drug screen
   c) serum osmolality
   d) CT of the head

 

EVALUATION Sober, yet sick
To rule out intoxication as the cause of Mr. W’s altered mental status, blood ethanol level and urine drug screens are obtained and found to be negative. CT of the head is negative for acute intracranial pathology.

Mr. W is admitted to the medical intensive care unit (MICU) for severe hyponatremia and altered mental status. Serum osmolality is 220 mOsm/kg (normal range 281 to 304 mOsm/kg). To further classify his hypo­tonic hyponatremia, volume status is assessed, and Mr. W is determined to be euvolemic. Thyroid-stimulating hormone and cortisol are within normal limits, eliminating hypothy­roidism and adrenal insufficiency as causes of his euvolemic hypotonic hyponatremia. Mr. W is treated for hyponatremia likely sec­ondary to syndrome of inappropriate antidi­uretic hormone (SIADH). SIADH is a diagnosis of exclusion that first requires ruling out hypo­thyroidism and glucocorticoid insufficiency (Figure 1).¹

The authors’ observations
Because hypokalemia is an independent pre­dictive factor for development of hyponatre­mia, it is necessary to evaluate the potassium level in all hyponatremic patients. Mr. W’s potassium level was 2.2 mEq/L on admis­sion. Serum sodium concentration is related to total exchangeable sodium, total body water, and total exchangeable potassium. Potassium depletion causes a shift of sodium into cells with a comparable exit of potas­sium from cells into extracellular fluid. The reverse process occurs during potassium repletion, leading to an increase in serum sodium concentration and making hypoka­lemia a risk factor for developing osmotic demyelination syndrome (ODS).⁴

Treating hyponatremia
Hyponatremia treatment depends on its severity, presence or absence of symptoms, and whether the hyponatremia is acute (<24 hours) or chronic (>48 hours).⁵

Because of Mr. W’s extremely low serum sodium concentration, predisposition to hyponatremia secondary to alcoholism, and history of severe hyponatremia, it is likely he is chronically hyponatremic.

In patients with chronic hyponatremia, neurological sequelae are associated with the need for a more rapid rate of correction of serum sodium. For most patients with chronic hyponatremia, a correction rate of ≤10 to 12 mEq/L during the first 24 hours and <18 mEq/L over 48 hours is recom­mended to avoid ODS.⁶

 

Evidence suggests, however, that this 1-day limit might be too high for some patients. Alcoholism, hypokalemia, mal­nutrition, and liver disease are present in a high percentage of patients who develop

ODS after correcting hyponatremia (Table 1).⁶ Therefore, for patients such as Mr. W who are at high risk of ODS, experts recommend a goal of 4 to 6 mEq/L/d with a correction rate of ≤8 mEq/L in any 24-hour period (Table 2).⁶

 

TREATMENT Sodium normalizes
Mr. W receives 1 L of normal saline in the ED before admission to the MICU. Once in the MICU, despite likely chronic hyponatremia, he receives hypertonic (3%) saline, followed by normal saline. Initially, he responds when the serum sodium concentration improves. Because of his likely SIADH, Mr. W is fluid-restricted for 4 days. Serum sodium returns to normal over 7 hospital days (Figure 2). To address the pro­found hypokalemia, Mr. S receives 30 mEq of potassium chloride in the ED, and potas­sium is repeated daily throughout his stay in the MICU.

Mr. W remains lethargic, with intermittent periods of confusion throughout the hospital stay. His altered mental status is attributed to hepatic encephalopathy secondary to alco­holic hepatitis. The Maddrey discriminant function is a calculation that stratifies patients with alcoholic hepatitis for risk of mortal­ity and the use of steroids. Because Mr. W shows a Maddrey discriminant function ≥32, he receives methylprednisolone, followed by pentoxifylline, and liver function tests trend down. He also receives lactulose throughout hospitalization.

By discharge on hospital day 9, Mr. W’s serum sodium is 138 mEq/L; serum potas­sium, 4.1 mEq/L. Total bilirubin and prothrom­bin remain elevated. Mr. W is discharged on lactulose, thiamine, folic acid, and a 1-month course of pentoxifylline, 400 mg, 3 times a day.

READMISSION Unsteady gait, nausea
Three days after discharge, Mr. W returns to the ED after experiencing a 20-second epi­sode of total body rigidity. He has an unsteady gait and worsening nausea and vomiting.

When Mr. W arrives in the ED, he confirms he is taking his discharge medications as pre­scribed. His parents report that he has con­sumed alcohol and Cannabis since discharge and has been taking his sibling’s prescription medications, including quetiapine.

In the ED, Mr. W is awake, alert, and ori­ented to person, place, and time. Vital signs are: pulse, 118 beats per minute; blood pres­sure, 128/73 mm Hg; respirations, 16 breaths per minute; and temperature, 98.5ºF. Physical examination, again, is notable for scleral icterus, jaundice, and asterixis. No focal neu­rologic deficits are noted.

Consistent with Mr. W’s previous admis­sion, laboratory values reveal altered hepatic function and impaired coagulation. The serum sodium level remains within normal limits at 136 mEq/L. However, again, meta­bolic disturbances include decreased chloride (97 mEq/L), potassium (2.9 mEq/L), and CO2 (18.2 mEq/L). CT on readmission is unchanged from the earlier hospitalization.

What is your differential diagnosis for Mr. W’s total body rigidity?
   a) seizure
   b) ODS
   c) drug intoxication
   d) neuroleptic malignant syndrome

EVALUATION Shaking and weakness
Once admitted to the hospital, Mr. W reports an episode of right upper-extremity “shak­ing,” followed by weakness. He remembers the entire event and denies tongue biting or incontinence. He is evaluated for pos­sible seizure, given his multiple risk factors, including drug and alcohol use, ingestion of quetiapine, and history of hyponatremia. Routine EEG is negative but prolactin level is elevated.

Mr. W’s mental status continues to wax and wane, prompting a neurology consult and MRI for further evaluation. MRI of the brain without contrast reveals restricted diffusion in the pons centrally, with extension bilaterally to the midbrain and thalami—findings con­sistent with central pontine myelinolysis. A neurology consultation reveals quadriparesis, paraparesis, dysarthria, and diplopia on exam­ination, all symptoms associated with central pontine myelinolysis.

The authors’ observations
ODS, including central and extrapontine myelinolysis, is a demyelinating condi­tion that occurs because of severe osmotic stress, most commonly secondary to the overly rapid correction of hyponatremia in patients with conditions leading to nutritional or electrolyte stress.⁷ Mr. W is considered at high risk of developing ODS because he fulfills the 5 criteria listed in Table 1.

Several psychiatric illnesses and neu­ropsychiatric medications could lead to hyponatremia. Many studies^8-10 have docu­mented hyponatremia and resulting ODS in patients with alcoholism, schizophrenia, anorexia, primary psychogenic polydipsia, and MDMA (3,4-methylenedioxymetham­phetamine) abuse. Hyponatremia is a side effect of several neuropsychiatric medica­tions, including serotonin reuptake inhibi­tors, lithium, tricyclic antidepressants, opioids, carbamazepine, oxcarbazepine, and antipsychotic polypharmacy. Other commonly used medications associated with hyponatremia include salt-losing diuretics, nonsteroidal anti-inflammatory drugs, and acetaminophen.⁷

Disease severity varies from asymptom­atic to coma or death. Symptoms, although some could reverse completely, typically are a combination of neuropsychiatric (ie, emotional lability, disinhibition, and other bizarre behaviors) and neurologic. Neurologic symptoms include confusion, impaired cognition, dysarthria, dyspha­gia, gait instability, weakness or paralysis, and generalized seizures. Severely affected patients could experience “locked-in syn­drome,” in which they are awake but unable to move or communicate. Also consistent with Mr. W’s case, ODS often presents ini­tially with delirium, seizures, or encepha­lopathy, followed by a lucid interval before symptoms develop.⁷

 

Diagnosis is based on the appearance of demyelinating brain lesions on CT or MRI. MRI is more sensitive than CT; however, even an MRI scan can appear normal for as long as 4 weeks after symptoms appear.⁷ Therefore, an initial negative radiologic study in a high-risk patient who develops neurologic symptoms does not exclude ODS. Earlier detection is possible with dif­fusion-weighted MRI, which is most sensi­tive and can detect lesions within 24 hours of developing symptoms.¹¹ The severity of the lesion does not correlate with severity of symptoms.

Studies reveal a considerable range in prognosis of patients with clinically symp­tomatic ODS. A study of 44 patients with central pontine myelinolysis, of which 42 had chronic alcoholism, reported that 34% had no significant functional deficits at follow-up, 34% had minor neurologic defi­cits, and 31% became dependent on per­sonal help. Outcome did not depend on the extent or severity of neurologic symptoms or the severity of concomitant systemic complications.¹²

Because of its poor prognosis, prevention of ODS is important. Because ODS com­monly is caused by overly rapid correction of hyponatremia, it is necessary to adhere to guidelines for treating chronic hypona­tremia (Table 2). If overcorrection occurs, therapeutic re-lowering of serum sodium can be considered, but has not been validated in controlled trials. Based mainly on case reports that suggest benefit from early re-lowering serum sodium in patients with ODS symptoms, experts recommend the following:  
   • administer desmopressin, 2 to 4 μg, every 8 hours parenterally  
   • replace water orally or as 5% dextrose in water intravenously (3 mL/kg/hr)  
   • check serum sodium hourly until serum is reduced to goal.⁶

Bottom Line

Hyponatremia is the most common electrolyte disorder encountered in practice. Osmotic demyelination syndrome often is preventable, with considerable morbidity and mortality. Psychiatrists should be aware of this condition because it could be an adverse effect of many psychiatric medications and there are some psychiatric illnesses in which hyponatremia is a potential risk. In hyponatremic patients with persistent nonspecific neurologic or neuropsychiatric symptoms and negative CT imaging, additional imaging, such as MRI, is warranted.

Related Resources

• Braun MM, Barstow CH, Pyzocha NJ. Diagnosis and management of sodium disorders: hyponatremia and hypernatremia. Am Fam Physician. 2015;91(5):299-307.
• Vaidya C, Ho W, Freda BJ. Management of hyponatremia: providing treatment and avoiding harm. Cleve Clin J Med. 2010;77(10):715-726.

Drug Brand Names
Carbamazepine • Tegretol
Oxcarbazepine • Trileptal
Desmopressin • Stimate, DDAVP
Lithium • Eskalith, Lithobid
Pentoxifylline • Trental, Pentoxil
Methylprednisolone • Medrol
Quetiapine • Seroquel

Disclosure
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Liraglutide (rDNA origin) injection, approved by the FDA in 2010 for managing type 2 diabetes mellitus (T2DM), has a new formulation and indication for treating obesity in adults as an adjunct to a reduced-calorie diet and increased physical activity (Table 1).¹

Liraglutide, recommended dosage 3 mg/d (under the brand name Saxenda), is approved for adults with a body mass index (BMI) ≥30, or those with a BMI of ≥27 and a weight-related condition such as hypertension, T2DM, or high cholesterol.¹ (A 1.8-mg formulation, under the brand name Victoza, is FDA-approved for managing T2DM, but is not indicated for weight management.)

How it works
Liraglutide is a glucagon-like peptide-1 (GLP-1) receptor agonist. GLP-1, which regulates appetite and calorie intake, is found in several regions of the brain that are involved in regulating appetite. Patients taking liraglutide lose weight because of decreased calorie intake, not increased energy expenditure.

Liraglutide is endogenously metabolized without a specific organ as a major route of elimination.¹

Dosage and administration
Liraglutide is administered using a prefilled, multi-dose pen that can be injected in the abdomen, thigh, or upper arm. Recommended dosage is 3 mg/d, administered any time of day. Initiate dosage at 0.6 mg/d the first week, then titrate by 0.6 mg a week—to reduce the likelihood of adverse gastrointestinal symptoms—until 3 mg/d is reached.

Discontinue liraglutide if a patient has not lost at least 4% of body weight after 16 weeks of treatment, because it is unlikely the patient will achieve and sustain weight loss.

Efficacy
Liraglutide was studied in 3 clinical trials of obese and overweight participants who had a weight-related condition. Patients who had a history of major depressive disorder or suicide attempt were excluded from the studies. All participants in Studies 1 and 2 received instruction about following a reduced-calorie diet and increasing physical activity. In Study 3, patients were randomized to treatment after losing >5% of their body weight through reduced calorie intake and exercise; those who did not meet the required weight loss were excluded from the study. In these 56-week clinical studies:
   • of 3,731 participants without diabetes or a weight-related comorbidity, such as high blood pressure or high cholesterol, 62% of patients (n = 2,313) who took liraglutide lost ≥5% of their body weight from baseline, compared with 34% of participants who received placebo
   • of 635 participants with T2DM, 49% of patients (n = 311) treated with liraglutide lost ≥5% of their body weight compared with 16% placebo patients
   • of 422 participants with a weight-related comorbidity, 42% of patients (n = 177) lost ≥5% of their body weight compared with 21.7% of placebo patients.

Improvements in some cardiovascular disease risk factors were observed. Long-term follow up was not studied.

Contraindictations
Liraglutide is contraindicated in patients who have a personal or family history of medullary thyroid carcinoma or multiple endocrine neoplasia syndrome type 2. In a 104-week study, malignant thyroid C-cell carcinomas were detected in rats and mice given liraglutide, 1 and 3 mg/kg/d; however, it was not detected in groups given 0.03 and 0.2 mg/kg/d. It isn’t known whether liraglutide can cause thyroid C-cell tumors in humans.

Patients should not take liraglutide if they have hypersensitivity to liraglutide or any product components, are using insulin, are taking any other GLP-1 receptor agonist, or are pregnant.

Adverse effects
The most common reported adverse effects are nausea (39.3%), hypoglycemia in patients with T2DM (23%), diarrhea (20.9%), constipation (19.4%), and vomiting (15.7%) (Table 2). In clinical trials, 9.8% of patients discontinued treatment because of adverse effects, compared with 4.3% of those receiving placebo.

Liraglutide has low potential for pharmacokinetic drug-drug interactions related to cytochrome P450 and plasma protein binding. For a full list of drug-drug interactions, see the full prescribing information.¹

Liraglutide (rDNA origin) injection, approved by the FDA in 2010 for managing type 2 diabetes mellitus (T2DM), has a new formulation and indication for treating obesity in adults as an adjunct to a reduced-calorie diet and increased physical activity (Table 1).¹

Liraglutide, recommended dosage 3 mg/d (under the brand name Saxenda), is approved for adults with a body mass index (BMI) ≥30, or those with a BMI of ≥27 and a weight-related condition such as hypertension, T2DM, or high cholesterol.¹ (A 1.8-mg formulation, under the brand name Victoza, is FDA-approved for managing T2DM, but is not indicated for weight management.)

How it works
Liraglutide is a glucagon-like peptide-1 (GLP-1) receptor agonist. GLP-1, which regulates appetite and calorie intake, is found in several regions of the brain that are involved in regulating appetite. Patients taking liraglutide lose weight because of decreased calorie intake, not increased energy expenditure.

Liraglutide is endogenously metabolized without a specific organ as a major route of elimination.¹

Dosage and administration
Liraglutide is administered using a prefilled, multi-dose pen that can be injected in the abdomen, thigh, or upper arm. Recommended dosage is 3 mg/d, administered any time of day. Initiate dosage at 0.6 mg/d the first week, then titrate by 0.6 mg a week—to reduce the likelihood of adverse gastrointestinal symptoms—until 3 mg/d is reached.

Discontinue liraglutide if a patient has not lost at least 4% of body weight after 16 weeks of treatment, because it is unlikely the patient will achieve and sustain weight loss.

Efficacy
Liraglutide was studied in 3 clinical trials of obese and overweight participants who had a weight-related condition. Patients who had a history of major depressive disorder or suicide attempt were excluded from the studies. All participants in Studies 1 and 2 received instruction about following a reduced-calorie diet and increasing physical activity. In Study 3, patients were randomized to treatment after losing >5% of their body weight through reduced calorie intake and exercise; those who did not meet the required weight loss were excluded from the study. In these 56-week clinical studies:
   • of 3,731 participants without diabetes or a weight-related comorbidity, such as high blood pressure or high cholesterol, 62% of patients (n = 2,313) who took liraglutide lost ≥5% of their body weight from baseline, compared with 34% of participants who received placebo
   • of 635 participants with T2DM, 49% of patients (n = 311) treated with liraglutide lost ≥5% of their body weight compared with 16% placebo patients
   • of 422 participants with a weight-related comorbidity, 42% of patients (n = 177) lost ≥5% of their body weight compared with 21.7% of placebo patients.

Improvements in some cardiovascular disease risk factors were observed. Long-term follow up was not studied.

Contraindictations
Liraglutide is contraindicated in patients who have a personal or family history of medullary thyroid carcinoma or multiple endocrine neoplasia syndrome type 2. In a 104-week study, malignant thyroid C-cell carcinomas were detected in rats and mice given liraglutide, 1 and 3 mg/kg/d; however, it was not detected in groups given 0.03 and 0.2 mg/kg/d. It isn’t known whether liraglutide can cause thyroid C-cell tumors in humans.

Patients should not take liraglutide if they have hypersensitivity to liraglutide or any product components, are using insulin, are taking any other GLP-1 receptor agonist, or are pregnant.

Adverse effects
The most common reported adverse effects are nausea (39.3%), hypoglycemia in patients with T2DM (23%), diarrhea (20.9%), constipation (19.4%), and vomiting (15.7%) (Table 2). In clinical trials, 9.8% of patients discontinued treatment because of adverse effects, compared with 4.3% of those receiving placebo.

Liraglutide has low potential for pharmacokinetic drug-drug interactions related to cytochrome P450 and plasma protein binding. For a full list of drug-drug interactions, see the full prescribing information.¹

Liraglutide (rDNA origin) injection, approved by the FDA in 2010 for managing type 2 diabetes mellitus (T2DM), has a new formulation and indication for treating obesity in adults as an adjunct to a reduced-calorie diet and increased physical activity (Table 1).¹

Liraglutide, recommended dosage 3 mg/d (under the brand name Saxenda), is approved for adults with a body mass index (BMI) ≥30, or those with a BMI of ≥27 and a weight-related condition such as hypertension, T2DM, or high cholesterol.¹ (A 1.8-mg formulation, under the brand name Victoza, is FDA-approved for managing T2DM, but is not indicated for weight management.)

How it works
Liraglutide is a glucagon-like peptide-1 (GLP-1) receptor agonist. GLP-1, which regulates appetite and calorie intake, is found in several regions of the brain that are involved in regulating appetite. Patients taking liraglutide lose weight because of decreased calorie intake, not increased energy expenditure.

Liraglutide is endogenously metabolized without a specific organ as a major route of elimination.¹

Dosage and administration
Liraglutide is administered using a prefilled, multi-dose pen that can be injected in the abdomen, thigh, or upper arm. Recommended dosage is 3 mg/d, administered any time of day. Initiate dosage at 0.6 mg/d the first week, then titrate by 0.6 mg a week—to reduce the likelihood of adverse gastrointestinal symptoms—until 3 mg/d is reached.

Discontinue liraglutide if a patient has not lost at least 4% of body weight after 16 weeks of treatment, because it is unlikely the patient will achieve and sustain weight loss.

Efficacy
Liraglutide was studied in 3 clinical trials of obese and overweight participants who had a weight-related condition. Patients who had a history of major depressive disorder or suicide attempt were excluded from the studies. All participants in Studies 1 and 2 received instruction about following a reduced-calorie diet and increasing physical activity. In Study 3, patients were randomized to treatment after losing >5% of their body weight through reduced calorie intake and exercise; those who did not meet the required weight loss were excluded from the study. In these 56-week clinical studies:
   • of 3,731 participants without diabetes or a weight-related comorbidity, such as high blood pressure or high cholesterol, 62% of patients (n = 2,313) who took liraglutide lost ≥5% of their body weight from baseline, compared with 34% of participants who received placebo
   • of 635 participants with T2DM, 49% of patients (n = 311) treated with liraglutide lost ≥5% of their body weight compared with 16% placebo patients
   • of 422 participants with a weight-related comorbidity, 42% of patients (n = 177) lost ≥5% of their body weight compared with 21.7% of placebo patients.

Improvements in some cardiovascular disease risk factors were observed. Long-term follow up was not studied.

Contraindictations
Liraglutide is contraindicated in patients who have a personal or family history of medullary thyroid carcinoma or multiple endocrine neoplasia syndrome type 2. In a 104-week study, malignant thyroid C-cell carcinomas were detected in rats and mice given liraglutide, 1 and 3 mg/kg/d; however, it was not detected in groups given 0.03 and 0.2 mg/kg/d. It isn’t known whether liraglutide can cause thyroid C-cell tumors in humans.

Patients should not take liraglutide if they have hypersensitivity to liraglutide or any product components, are using insulin, are taking any other GLP-1 receptor agonist, or are pregnant.

Adverse effects
The most common reported adverse effects are nausea (39.3%), hypoglycemia in patients with T2DM (23%), diarrhea (20.9%), constipation (19.4%), and vomiting (15.7%) (Table 2). In clinical trials, 9.8% of patients discontinued treatment because of adverse effects, compared with 4.3% of those receiving placebo.

Liraglutide has low potential for pharmacokinetic drug-drug interactions related to cytochrome P450 and plasma protein binding. For a full list of drug-drug interactions, see the full prescribing information.¹

The less time that passes between the onset of psychosis and initiation of appropriate treatment, the greater the patient’s odds of recovery.¹ However, relapse prevention is a major clinical challenge because >80% of patients will relapse within 5 years, and, on average, 40% to 50% of patients with a first-episode schizophrenia will relapse within 2 years depending on the definition used and patient characteristics.² Although there are several explanations and contributing factors to relapses, non­adherence—partial or complete discontinuation of antipsychotics—is a primary risk factor, contributing to a 5-fold increase in relapse risk.³

As such, optimal antipsychotic selection, dosing, and monitoring play an important role in managing this illness. Patients with first-episode psychosis (FEP) are unusual in some ways, compared with patients with multiple episodes of psychosis and represent a different stage of schizophrenia.

In this 2-part series, we will discuss pharmacotherapy for FEP. This article focuses on antipsychotic selection, dosage, and duration of treat­ment among these patients. The second article, in the July 2015 issue, reviews the rationale and evidence for non-standard, first-line thera­pies, including long-acting injectable antipsychotics and clozapine.

Defining FEP
FEP refers to a patient who has presented, been evaluated, and received treatment for the first psychotic episode associated with a schizophre­nia spectrum diagnosis.⁴ FEP is part of a trajectory marked by tran­ sitional periods. The patient transitions from being “healthy” to a prodromal state characterized by: (1) nonpsychotic behav­ioral disturbances such as depression or obsessive-compulsive disorder, (2) attenu­ated psychotic symptoms not requiring treatment, then converting to (3) psychotic symptoms prompting initial presentation for antipsychotic pharmacotherapy, lead­ing to (4) a formal diagnosis of schizo­phreniform disorder and, subsequently, schizophrenia, requiring treatment to sta­bilize symptoms.

There are 2 critical periods along this continuum: prodromal stage and the dura­tion of untreated psychosis (DUP). The prodromal period is a retrospectively iden­tified time where the patient shows initial nonpsychotic disturbances (eg, cognitive and behavioral symptoms) before exhibit­ing clinical diagnostic criteria for a schizo­phrenia spectrum disorder. Approximately one-third of patients exhibiting these symptoms convert to psychosis within 1 year, and early treatment engagement at this stage has been shown to improve out­comes.⁵ The DUP is the time from when a patient has noticeable psychotic symptoms to initiation of drug treatment. The DUP is a consistent predictor of clinical out­come in schizophrenia, including negative symptoms, quality of life, and functional capacity.¹

Antipsychotic selection
Treatment goals for FEP patients include:
   • minimizing the DUP
   • rapidly stabilizing psychosis
   • achieving full symptomatic remission
   • preventing relapse.

Several treatment guidelines for manag­ing schizophrenia offer variable recommen­dations for initial antipsychotic treatment in patients with first-episode schizophre­nia (Table 1).^6-15 Most recom­mend second-generation antipsychotics (SGAs) over first-generation antipsy­chotics (FGAs)^6,8,9,13,15 with specific recom­mendations on minimizing neurologic and metabolic adverse effects—to which FEP patients are susceptible—by avoid­ing high-potency and neurotoxic FGAs (eg, haloperidol and fluphenazine),⁷ clo­zapine,^11,14 olanzapine,¹¹ or ziprasidone.¹⁴ Two guidelines—the National Institute for Health and Care Excellence and the Scottish Intercollegiate Guidelines Network—do not state a preference for antipsychotic selection.^10,12

The rationale for these recommendations is based on efficacy data, tolerability dif­ferences, FDA-approved indications, and recent FDA approvals with sparse post-marketing data. Of note, there are a lack of robust data for newer antipsychotics (eg, aripiprazole, paliperidone, iloperidone, asenapine, and lurasidone) in effectively and safely treating FEP; however, given the results of other antipsychotics studies, it is likely the efficacy and tolerability of these drugs can be extrapolated from experience with multi-episode patients.

Study design and demographics. Research studies of FEP share some simi­larities in study design; however, there is enough variability to make it difficult to compare studies and generalize find­ings (Table 2).¹⁶ The variability of DUP is a limitation when comparing studies because it is a significant predic­tor of clinical outcome. Patients who abuse substances—and often are more challeng­ing to treat¹⁷—typically are excluded from these trials, which could explain the high response rate documented in studies of first-episode schizophrenia.

 

In addition, some FEP patients included in clinical trials might not be truly antipsy­chotic naïve; an estimated 25% to 75% of patients in these studies are antipsychotic naïve. This is an important consideration when comparing data on adverse effects that occur early in treatment. Additionally, acknowledging the advantages and disad­vantages of how to handle missing data is critical because of the high dropout rate observed in these studies.¹⁸

Efficacy. There is a high response rate to antipsychotic therapy—ranging from 46% to 96%, depending on the study—in patients with first-episode schizophrenia.³ The response mainly is seen in reduction of positive symptoms because typically negative and cognitive symptoms do not respond to antipsychotics. One study reported only 29% of patients achieved both positive and negative symptom remis­sion.¹⁹ It is likely that secondary negative symptoms caused by social withdrawal, reduced speech, and avoidance improve when positive symptoms subside, but pri­mary negative symptoms endure.In general, there is a lack of evidence suggesting that 1 antipsychotic class or agent is more effective than another. Studies mainly assess effectiveness using the primary outcome measure of all-cause discontinuation, such as the Clinical Antipsychotic Trials of Intervention Effectiveness study.²⁰ This outcome mea­sure is a mixture of patient preference, tol­erability, and efficacy that provides a more generalizable gauge on how well the treat­ment works in the clinic rather than tightly regulated settings such as clinical trials. A recent meta-analysis supports no differ­ences in efficacy among antipsychotics in early-episode psychosis.²¹

 

Tolerability. Because there are no significant differences among antipsychotic classes or agents in terms of efficacy in first-episode schizophrenia, drug selection is guided mainly by (1) the adverse effect profile and (2) what should be avoided depending on patient-specific variables. Evidence sug­gests first-episode patients are more sen­sitive to adverse effects of antipsychotics, particularly neurologic side effects (see this article at CurrentPsychiatry.com for a table comparing adverse effects of antipsychot­ics in first-episode psychosis).^18,22-29 Overall adverse effect profiles remain similar across FEP or multi-episode patients, but tend to be more exaggerated in drug-naïve patients with FEP.

Regarding FGA side effects, McEvoy et al¹⁸ demonstrated the neuroleptic threshold occurs at 50% lower haloperi­dol dosages in patients with first-episode schizophrenia (2.1 mg/d) compared with multi-episode schizophrenia (4.3 mg/d). Other trials suggest SGAs are associated with a lower risk of extrapyramidal side effects (EPS) or use of adjunctive therapies such as anticholinergic drugs or benzo­diazepines.^23-27 An exception to this state­ment is that higher risperidone dosages (≥4 to 6 mg/d) have been found to have higher rates of EPS and use of adjunctive medica­tions to treat these symptoms in FEP.²⁶ This is important because studies report higher discontinuation rates with more severe adverse effects of antipsychotics.

Cardiometabolic effects are of particular concern in first-episode patients because most weight gain happens in the first 3 to 4 months of treatment and remains throughout the first year.^18,24,29,30 Studies have shown that olanzapine, quetiap­ine, and risperidone are associated with more clinically significant weight gain compared with haloperidol and ziprasi­done.^23-25 Olanzapine-associated weight gain has been reported to be twice that of quetiapine and risperidone.¹⁸ Regardless, the EUFEST trial did not find a difference in clinically significant weight gain after 12 months among the antipsychotics studied, including haloperidol and ziprasidone.²⁵

Weight gain associated with these anti­psychotics is accompanied by changes in fasting triglycerides, glucose, total choles­terol,²³ and high-density lipoprotein cho­lesterol as well as an increase in body mass index (BMI) categorization²⁹ (eg, shift from normal to overweight).^18,25 Patients with lower baseline BMI and in racial minor­ity groups might experience more rapid weight gain regardless of antipsychotic selection.^29,30

Hyperprolactinemia could be under-recognized and could contribute to early treatment discontinuation.³¹ Evidence in patients with first-episode schizophrenia suggests similar outcomes as those seen in multi-episode patients, in whom ris­peridone is associated with higher pro­lactin elevations and clinically significant hyperprolactinemia (eg, galactorrhea and gynecomastia) compared with olanzapine, quetiapine, and low-dose haloperidol.^18,23,24 However, there is a lack of studies that assess whether long-term therapy with strong D2 receptor antagonists increases the risk of bone demineralization or path­ological fractures when started before patients’ bones reach maximum density in their mid-20s.³¹

Antipsychotic dosing
Given the high rate of treatment response in FEP and patients’ higher sensitivity to antipsychotic adverse effects, particularly EPS, guidelines recommend antipsychotic dosages lower than those used for multi-episode schizophrenia,¹¹ especially FGAs. Based on trial data, commonly used dos­ages include:
   • haloperidol, ≤5 mg/d^23-25,29
   • olanzapine, 10 mg/d^18,23,25,29
   • risperidone, ≤4 to 6 mg/d.^18,24,29,32

In general, haloperidol and risperidone, 2 to 3 mg/d, were well tolerated and effec­tive in trials. Higher quetiapine dosages of 500 to 600 mg/d could be required.^11,18,25

According to a survey on prescribing practices of antipsychotic selection and dosing in first-episode schizophrenia,4 clinical prescribing practices tend to use unnecessarily high initial antipsychotic dosing compared with trial data. There also is variability in the usual target anti­psychotic dosage ranging from 50% lower dosages to normal dosages in chronic schizophrenia to above FDA-approved maximum dosages for olanzapine (which may be necessary to counteract tobacco-induced cytochrome P450 1A2 enzyme induction).

In addition, these clinicians reported prescribing aripiprazole, an antipsychotic with weaker evidence (eg, case reports, case series, open-label studies) support­ing its efficacy and tolerability in FEP. These prescribing practices could reflect attempts to reduce the DUP and achieve symptom remission, so long as tolerability is not a concern.

Essentially, prescribed dosages should be based on symptom improvement and tolerability. This ideal dosage will vary as illustrated by Kapur et al,³³ who reported that FEP patients (N = 20) given haloper­idol, 1 mg or 2.5 mg/d, had D2 receptor occupancy rates of 38% to 87%, which was significantly dose-related (1 mg/d mean = 59%, 2.5 mg/d mean = 75%). Clinical response and EPS significantly increased as D2 receptor occupancy exceeded 65% and 78%, respectively.

Antipsychotic response
When should you expect to see symp­tom improvement in patients with first-episode schizophrenia?
Emsley et al³⁴ reported a 77.6% response rate among first-episode patients (N = 522) treated with low dosages of risperidone (mean modal dosage [MMD] = 3.3 mg/d) and haloperidol (MMD = 2.9 mg/d). They found variable response times that were evenly dispersed over a 10-week period. Nearly one-quarter (22.5%) did not respond until after week 4 and 11.2% did not respond until after week 8. In a study of FEP patients (N = 112) treated with olanzapine (MMD = 11.8 mg/d) or risperi­done (MMD = 3.9 mg/d), Gallego et al³⁵ reported a cumulative response of 39.6% at week 8 and 65.1% at week 16.

 

Although there is evidence that, among multi-episode patients, early nonresponse to antipsychotic therapy could predict subsequent nonresponse,³⁶ the evidence is mixed for first-episode schizophrenia. Studies by Emsley et al³⁴ and Gallego et al³⁵ did not find that early nonresponse at weeks 1 or 2 predicted subsequent nonre­sponse at week 4 or later. However, other studies support the idea that early nonre­sponse predicts subsequent nonresponse and early antipsychotic response predicts future response in first-episode patients, with good specificity and sensitivity.^37,38

Overall, treatment response in first-episode schizophrenia is variable. An adequate antipsychotic trial may be lon­ger, 8 to 16 weeks, compared with 4 to 8 weeks in multi-episode patients. Because research suggests that failure to respond to treatment may lead to medication non­adherence,³⁹ it is reasonable to consider switching antipsychotics when a patient experiences minimal or no response to antipsychotic therapy at week 2; however, this should be a patient-specific decision.

How long should you continue therapy after symptom remission?
There is a lack of consensus on the dura­tion of therapy for a patient treated for first-episode schizophrenia because a small percentage (10% to 20%) do not relapse after the first psychotic episode.³ In general, treatment guidelines and expert consensus statements recommend at least 1 to 2 years of treatment before considering a discon­tinuation trial.^7,10-11 Discuss the benefits and risks of maintenance treatment with your patient and obtain informed consent. With patients with minimal insight, obtaining proper consent is not possible and the phy­sician must exercise judgment unilaterally, if necessary, after educating the family.

After at least 12 months of treatment, antipsychotic therapy could continue indefinitely, depending on patient-specific factors. There are no predictors for identi­fying patients who do not require mainte­nance therapy beyond the first psychotic episode. The absence of negative and cog­nitive deficits could provide clues that a patient might be a candidate for antipsy­chotic tapering.

Predicting the treatment course
Research investigating clinical predic­tors or biomarkers that forecast whether a patient will respond to treatment is pre­liminary. Many characteristics have been identified (Table 3^{1,3,4,23,25,40}) and include shorter DUP,¹ poorer premorbid function,³ antipsychotic discontinuation,³ a trusting patient-doctor relationship,⁴¹ and antipsychotic-related adverse effects,^23,25 which are predictive of response, nonre­sponse, relapse, adherence, and nonadher­ence, respectively.
 

Bottom Line
The goals of pharmacological treatment of first-episode schizophrenia are to minimize the duration of untreated psychosis and target full remission of positive symptoms using the lowest possible antipsychotic dosages. Pharmacotherapy should continued for 1 to 2 years, with longer duration considered if it is discussed with the patient and with vigilant monitoring for adverse effects and suboptimal medication nonadherence to prevent relapse.
 

Editor’s note: The second article in this series in the July 2015 issue reviews the rationale and evidence for non-standard, first-line therapies, including long-acting injectable antipsychotics and clozapine.

 

Related Resources
• Recovery After an Initial Schizophrenia Episode (RAISE) Project Early Treatment Program. National Institute of Mental Health. http://raiseetp.org.
• Martens L, Baker S. Promoting recovery from first epi­sode psychosis: a guide for families. Centre for Addiction and Mental Health. http://www.camh.ca/en/hospital/ Documents/www.camh.net/AboutCAMH/Guideto CAMH/MentalHealthPrograms/SchizophreniaProgram/ 3936PromotingRecoveryFirstEpisodePsychosisfinal.pdf.

Drug Brand Names
Aripiprazole • Abilify                Lurasidone • Latuda
Asenapine • Saphris                Olanzapine • Zyprexa
Clozapine • Clozaril                 Paliperidone • Invega
Fluphenazine • Prolixin            Quetiapine • Seroquel
Iloperidone • Fanapt               Risperidone • Risperdal
Haloperidol • Haldol                Ziprasidone • Geodon

 

Disclosures
Dr. Gardner reports no financial relationships with any companies whose products are mentioned in this article or with manufacturers of competing products.
Dr. Nasrallah is a consultant to Acadia, Alkermes, Lundbeck, Janssen, Merck, Otsuka, and Sunovion, and is a speaker for Alkermes, Lundbeck, Janssen, Otsuka, and Sunovion.

The less time that passes between the onset of psychosis and initiation of appropriate treatment, the greater the patient’s odds of recovery.¹ However, relapse prevention is a major clinical challenge because >80% of patients will relapse within 5 years, and, on average, 40% to 50% of patients with a first-episode schizophrenia will relapse within 2 years depending on the definition used and patient characteristics.² Although there are several explanations and contributing factors to relapses, non­adherence—partial or complete discontinuation of antipsychotics—is a primary risk factor, contributing to a 5-fold increase in relapse risk.³

As such, optimal antipsychotic selection, dosing, and monitoring play an important role in managing this illness. Patients with first-episode psychosis (FEP) are unusual in some ways, compared with patients with multiple episodes of psychosis and represent a different stage of schizophrenia.

In this 2-part series, we will discuss pharmacotherapy for FEP. This article focuses on antipsychotic selection, dosage, and duration of treat­ment among these patients. The second article, in the July 2015 issue, reviews the rationale and evidence for non-standard, first-line thera­pies, including long-acting injectable antipsychotics and clozapine.

Defining FEP
FEP refers to a patient who has presented, been evaluated, and received treatment for the first psychotic episode associated with a schizophre­nia spectrum diagnosis.⁴ FEP is part of a trajectory marked by tran­ sitional periods. The patient transitions from being “healthy” to a prodromal state characterized by: (1) nonpsychotic behav­ioral disturbances such as depression or obsessive-compulsive disorder, (2) attenu­ated psychotic symptoms not requiring treatment, then converting to (3) psychotic symptoms prompting initial presentation for antipsychotic pharmacotherapy, lead­ing to (4) a formal diagnosis of schizo­phreniform disorder and, subsequently, schizophrenia, requiring treatment to sta­bilize symptoms.

There are 2 critical periods along this continuum: prodromal stage and the dura­tion of untreated psychosis (DUP). The prodromal period is a retrospectively iden­tified time where the patient shows initial nonpsychotic disturbances (eg, cognitive and behavioral symptoms) before exhibit­ing clinical diagnostic criteria for a schizo­phrenia spectrum disorder. Approximately one-third of patients exhibiting these symptoms convert to psychosis within 1 year, and early treatment engagement at this stage has been shown to improve out­comes.⁵ The DUP is the time from when a patient has noticeable psychotic symptoms to initiation of drug treatment. The DUP is a consistent predictor of clinical out­come in schizophrenia, including negative symptoms, quality of life, and functional capacity.¹

Antipsychotic selection
Treatment goals for FEP patients include:
   • minimizing the DUP
   • rapidly stabilizing psychosis
   • achieving full symptomatic remission
   • preventing relapse.

Several treatment guidelines for manag­ing schizophrenia offer variable recommen­dations for initial antipsychotic treatment in patients with first-episode schizophre­nia (Table 1).^6-15 Most recom­mend second-generation antipsychotics (SGAs) over first-generation antipsy­chotics (FGAs)^6,8,9,13,15 with specific recom­mendations on minimizing neurologic and metabolic adverse effects—to which FEP patients are susceptible—by avoid­ing high-potency and neurotoxic FGAs (eg, haloperidol and fluphenazine),⁷ clo­zapine,^11,14 olanzapine,¹¹ or ziprasidone.¹⁴ Two guidelines—the National Institute for Health and Care Excellence and the Scottish Intercollegiate Guidelines Network—do not state a preference for antipsychotic selection.^10,12

The rationale for these recommendations is based on efficacy data, tolerability dif­ferences, FDA-approved indications, and recent FDA approvals with sparse post-marketing data. Of note, there are a lack of robust data for newer antipsychotics (eg, aripiprazole, paliperidone, iloperidone, asenapine, and lurasidone) in effectively and safely treating FEP; however, given the results of other antipsychotics studies, it is likely the efficacy and tolerability of these drugs can be extrapolated from experience with multi-episode patients.

Study design and demographics. Research studies of FEP share some simi­larities in study design; however, there is enough variability to make it difficult to compare studies and generalize find­ings (Table 2).¹⁶ The variability of DUP is a limitation when comparing studies because it is a significant predic­tor of clinical outcome. Patients who abuse substances—and often are more challeng­ing to treat¹⁷—typically are excluded from these trials, which could explain the high response rate documented in studies of first-episode schizophrenia.

 

In addition, some FEP patients included in clinical trials might not be truly antipsy­chotic naïve; an estimated 25% to 75% of patients in these studies are antipsychotic naïve. This is an important consideration when comparing data on adverse effects that occur early in treatment. Additionally, acknowledging the advantages and disad­vantages of how to handle missing data is critical because of the high dropout rate observed in these studies.¹⁸

Efficacy. There is a high response rate to antipsychotic therapy—ranging from 46% to 96%, depending on the study—in patients with first-episode schizophrenia.³ The response mainly is seen in reduction of positive symptoms because typically negative and cognitive symptoms do not respond to antipsychotics. One study reported only 29% of patients achieved both positive and negative symptom remis­sion.¹⁹ It is likely that secondary negative symptoms caused by social withdrawal, reduced speech, and avoidance improve when positive symptoms subside, but pri­mary negative symptoms endure.In general, there is a lack of evidence suggesting that 1 antipsychotic class or agent is more effective than another. Studies mainly assess effectiveness using the primary outcome measure of all-cause discontinuation, such as the Clinical Antipsychotic Trials of Intervention Effectiveness study.²⁰ This outcome mea­sure is a mixture of patient preference, tol­erability, and efficacy that provides a more generalizable gauge on how well the treat­ment works in the clinic rather than tightly regulated settings such as clinical trials. A recent meta-analysis supports no differ­ences in efficacy among antipsychotics in early-episode psychosis.²¹

 

Tolerability. Because there are no significant differences among antipsychotic classes or agents in terms of efficacy in first-episode schizophrenia, drug selection is guided mainly by (1) the adverse effect profile and (2) what should be avoided depending on patient-specific variables. Evidence sug­gests first-episode patients are more sen­sitive to adverse effects of antipsychotics, particularly neurologic side effects (see this article at CurrentPsychiatry.com for a table comparing adverse effects of antipsychot­ics in first-episode psychosis).^18,22-29 Overall adverse effect profiles remain similar across FEP or multi-episode patients, but tend to be more exaggerated in drug-naïve patients with FEP.

Regarding FGA side effects, McEvoy et al¹⁸ demonstrated the neuroleptic threshold occurs at 50% lower haloperi­dol dosages in patients with first-episode schizophrenia (2.1 mg/d) compared with multi-episode schizophrenia (4.3 mg/d). Other trials suggest SGAs are associated with a lower risk of extrapyramidal side effects (EPS) or use of adjunctive therapies such as anticholinergic drugs or benzo­diazepines.^23-27 An exception to this state­ment is that higher risperidone dosages (≥4 to 6 mg/d) have been found to have higher rates of EPS and use of adjunctive medica­tions to treat these symptoms in FEP.²⁶ This is important because studies report higher discontinuation rates with more severe adverse effects of antipsychotics.

Cardiometabolic effects are of particular concern in first-episode patients because most weight gain happens in the first 3 to 4 months of treatment and remains throughout the first year.^18,24,29,30 Studies have shown that olanzapine, quetiap­ine, and risperidone are associated with more clinically significant weight gain compared with haloperidol and ziprasi­done.^23-25 Olanzapine-associated weight gain has been reported to be twice that of quetiapine and risperidone.¹⁸ Regardless, the EUFEST trial did not find a difference in clinically significant weight gain after 12 months among the antipsychotics studied, including haloperidol and ziprasidone.²⁵

Weight gain associated with these anti­psychotics is accompanied by changes in fasting triglycerides, glucose, total choles­terol,²³ and high-density lipoprotein cho­lesterol as well as an increase in body mass index (BMI) categorization²⁹ (eg, shift from normal to overweight).^18,25 Patients with lower baseline BMI and in racial minor­ity groups might experience more rapid weight gain regardless of antipsychotic selection.^29,30

Hyperprolactinemia could be under-recognized and could contribute to early treatment discontinuation.³¹ Evidence in patients with first-episode schizophrenia suggests similar outcomes as those seen in multi-episode patients, in whom ris­peridone is associated with higher pro­lactin elevations and clinically significant hyperprolactinemia (eg, galactorrhea and gynecomastia) compared with olanzapine, quetiapine, and low-dose haloperidol.^18,23,24 However, there is a lack of studies that assess whether long-term therapy with strong D2 receptor antagonists increases the risk of bone demineralization or path­ological fractures when started before patients’ bones reach maximum density in their mid-20s.³¹

Antipsychotic dosing
Given the high rate of treatment response in FEP and patients’ higher sensitivity to antipsychotic adverse effects, particularly EPS, guidelines recommend antipsychotic dosages lower than those used for multi-episode schizophrenia,¹¹ especially FGAs. Based on trial data, commonly used dos­ages include:
   • haloperidol, ≤5 mg/d^23-25,29
   • olanzapine, 10 mg/d^18,23,25,29
   • risperidone, ≤4 to 6 mg/d.^18,24,29,32

In general, haloperidol and risperidone, 2 to 3 mg/d, were well tolerated and effec­tive in trials. Higher quetiapine dosages of 500 to 600 mg/d could be required.^11,18,25

According to a survey on prescribing practices of antipsychotic selection and dosing in first-episode schizophrenia,4 clinical prescribing practices tend to use unnecessarily high initial antipsychotic dosing compared with trial data. There also is variability in the usual target anti­psychotic dosage ranging from 50% lower dosages to normal dosages in chronic schizophrenia to above FDA-approved maximum dosages for olanzapine (which may be necessary to counteract tobacco-induced cytochrome P450 1A2 enzyme induction).

In addition, these clinicians reported prescribing aripiprazole, an antipsychotic with weaker evidence (eg, case reports, case series, open-label studies) support­ing its efficacy and tolerability in FEP. These prescribing practices could reflect attempts to reduce the DUP and achieve symptom remission, so long as tolerability is not a concern.

Essentially, prescribed dosages should be based on symptom improvement and tolerability. This ideal dosage will vary as illustrated by Kapur et al,³³ who reported that FEP patients (N = 20) given haloper­idol, 1 mg or 2.5 mg/d, had D2 receptor occupancy rates of 38% to 87%, which was significantly dose-related (1 mg/d mean = 59%, 2.5 mg/d mean = 75%). Clinical response and EPS significantly increased as D2 receptor occupancy exceeded 65% and 78%, respectively.

Antipsychotic response
When should you expect to see symp­tom improvement in patients with first-episode schizophrenia?
Emsley et al³⁴ reported a 77.6% response rate among first-episode patients (N = 522) treated with low dosages of risperidone (mean modal dosage [MMD] = 3.3 mg/d) and haloperidol (MMD = 2.9 mg/d). They found variable response times that were evenly dispersed over a 10-week period. Nearly one-quarter (22.5%) did not respond until after week 4 and 11.2% did not respond until after week 8. In a study of FEP patients (N = 112) treated with olanzapine (MMD = 11.8 mg/d) or risperi­done (MMD = 3.9 mg/d), Gallego et al³⁵ reported a cumulative response of 39.6% at week 8 and 65.1% at week 16.

 

Although there is evidence that, among multi-episode patients, early nonresponse to antipsychotic therapy could predict subsequent nonresponse,³⁶ the evidence is mixed for first-episode schizophrenia. Studies by Emsley et al³⁴ and Gallego et al³⁵ did not find that early nonresponse at weeks 1 or 2 predicted subsequent nonre­sponse at week 4 or later. However, other studies support the idea that early nonre­sponse predicts subsequent nonresponse and early antipsychotic response predicts future response in first-episode patients, with good specificity and sensitivity.^37,38

Overall, treatment response in first-episode schizophrenia is variable. An adequate antipsychotic trial may be lon­ger, 8 to 16 weeks, compared with 4 to 8 weeks in multi-episode patients. Because research suggests that failure to respond to treatment may lead to medication non­adherence,³⁹ it is reasonable to consider switching antipsychotics when a patient experiences minimal or no response to antipsychotic therapy at week 2; however, this should be a patient-specific decision.

How long should you continue therapy after symptom remission?
There is a lack of consensus on the dura­tion of therapy for a patient treated for first-episode schizophrenia because a small percentage (10% to 20%) do not relapse after the first psychotic episode.³ In general, treatment guidelines and expert consensus statements recommend at least 1 to 2 years of treatment before considering a discon­tinuation trial.^7,10-11 Discuss the benefits and risks of maintenance treatment with your patient and obtain informed consent. With patients with minimal insight, obtaining proper consent is not possible and the phy­sician must exercise judgment unilaterally, if necessary, after educating the family.

After at least 12 months of treatment, antipsychotic therapy could continue indefinitely, depending on patient-specific factors. There are no predictors for identi­fying patients who do not require mainte­nance therapy beyond the first psychotic episode. The absence of negative and cog­nitive deficits could provide clues that a patient might be a candidate for antipsy­chotic tapering.

Predicting the treatment course
Research investigating clinical predic­tors or biomarkers that forecast whether a patient will respond to treatment is pre­liminary. Many characteristics have been identified (Table 3^{1,3,4,23,25,40}) and include shorter DUP,¹ poorer premorbid function,³ antipsychotic discontinuation,³ a trusting patient-doctor relationship,⁴¹ and antipsychotic-related adverse effects,^23,25 which are predictive of response, nonre­sponse, relapse, adherence, and nonadher­ence, respectively.
 

Bottom Line
The goals of pharmacological treatment of first-episode schizophrenia are to minimize the duration of untreated psychosis and target full remission of positive symptoms using the lowest possible antipsychotic dosages. Pharmacotherapy should continued for 1 to 2 years, with longer duration considered if it is discussed with the patient and with vigilant monitoring for adverse effects and suboptimal medication nonadherence to prevent relapse.
 

Editor’s note: The second article in this series in the July 2015 issue reviews the rationale and evidence for non-standard, first-line therapies, including long-acting injectable antipsychotics and clozapine.

 

Related Resources
• Recovery After an Initial Schizophrenia Episode (RAISE) Project Early Treatment Program. National Institute of Mental Health. http://raiseetp.org.
• Martens L, Baker S. Promoting recovery from first epi­sode psychosis: a guide for families. Centre for Addiction and Mental Health. http://www.camh.ca/en/hospital/ Documents/www.camh.net/AboutCAMH/Guideto CAMH/MentalHealthPrograms/SchizophreniaProgram/ 3936PromotingRecoveryFirstEpisodePsychosisfinal.pdf.

Drug Brand Names
Aripiprazole • Abilify                Lurasidone • Latuda
Asenapine • Saphris                Olanzapine • Zyprexa
Clozapine • Clozaril                 Paliperidone • Invega
Fluphenazine • Prolixin            Quetiapine • Seroquel
Iloperidone • Fanapt               Risperidone • Risperdal
Haloperidol • Haldol                Ziprasidone • Geodon

 

Disclosures
Dr. Gardner reports no financial relationships with any companies whose products are mentioned in this article or with manufacturers of competing products.
Dr. Nasrallah is a consultant to Acadia, Alkermes, Lundbeck, Janssen, Merck, Otsuka, and Sunovion, and is a speaker for Alkermes, Lundbeck, Janssen, Otsuka, and Sunovion.

The less time that passes between the onset of psychosis and initiation of appropriate treatment, the greater the patient’s odds of recovery.¹ However, relapse prevention is a major clinical challenge because >80% of patients will relapse within 5 years, and, on average, 40% to 50% of patients with a first-episode schizophrenia will relapse within 2 years depending on the definition used and patient characteristics.² Although there are several explanations and contributing factors to relapses, non­adherence—partial or complete discontinuation of antipsychotics—is a primary risk factor, contributing to a 5-fold increase in relapse risk.³

As such, optimal antipsychotic selection, dosing, and monitoring play an important role in managing this illness. Patients with first-episode psychosis (FEP) are unusual in some ways, compared with patients with multiple episodes of psychosis and represent a different stage of schizophrenia.

In this 2-part series, we will discuss pharmacotherapy for FEP. This article focuses on antipsychotic selection, dosage, and duration of treat­ment among these patients. The second article, in the July 2015 issue, reviews the rationale and evidence for non-standard, first-line thera­pies, including long-acting injectable antipsychotics and clozapine.

Defining FEP
FEP refers to a patient who has presented, been evaluated, and received treatment for the first psychotic episode associated with a schizophre­nia spectrum diagnosis.⁴ FEP is part of a trajectory marked by tran­ sitional periods. The patient transitions from being “healthy” to a prodromal state characterized by: (1) nonpsychotic behav­ioral disturbances such as depression or obsessive-compulsive disorder, (2) attenu­ated psychotic symptoms not requiring treatment, then converting to (3) psychotic symptoms prompting initial presentation for antipsychotic pharmacotherapy, lead­ing to (4) a formal diagnosis of schizo­phreniform disorder and, subsequently, schizophrenia, requiring treatment to sta­bilize symptoms.

There are 2 critical periods along this continuum: prodromal stage and the dura­tion of untreated psychosis (DUP). The prodromal period is a retrospectively iden­tified time where the patient shows initial nonpsychotic disturbances (eg, cognitive and behavioral symptoms) before exhibit­ing clinical diagnostic criteria for a schizo­phrenia spectrum disorder. Approximately one-third of patients exhibiting these symptoms convert to psychosis within 1 year, and early treatment engagement at this stage has been shown to improve out­comes.⁵ The DUP is the time from when a patient has noticeable psychotic symptoms to initiation of drug treatment. The DUP is a consistent predictor of clinical out­come in schizophrenia, including negative symptoms, quality of life, and functional capacity.¹

Antipsychotic selection
Treatment goals for FEP patients include:
   • minimizing the DUP
   • rapidly stabilizing psychosis
   • achieving full symptomatic remission
   • preventing relapse.

Several treatment guidelines for manag­ing schizophrenia offer variable recommen­dations for initial antipsychotic treatment in patients with first-episode schizophre­nia (Table 1).^6-15 Most recom­mend second-generation antipsychotics (SGAs) over first-generation antipsy­chotics (FGAs)^6,8,9,13,15 with specific recom­mendations on minimizing neurologic and metabolic adverse effects—to which FEP patients are susceptible—by avoid­ing high-potency and neurotoxic FGAs (eg, haloperidol and fluphenazine),⁷ clo­zapine,^11,14 olanzapine,¹¹ or ziprasidone.¹⁴ Two guidelines—the National Institute for Health and Care Excellence and the Scottish Intercollegiate Guidelines Network—do not state a preference for antipsychotic selection.^10,12

The rationale for these recommendations is based on efficacy data, tolerability dif­ferences, FDA-approved indications, and recent FDA approvals with sparse post-marketing data. Of note, there are a lack of robust data for newer antipsychotics (eg, aripiprazole, paliperidone, iloperidone, asenapine, and lurasidone) in effectively and safely treating FEP; however, given the results of other antipsychotics studies, it is likely the efficacy and tolerability of these drugs can be extrapolated from experience with multi-episode patients.

Study design and demographics. Research studies of FEP share some simi­larities in study design; however, there is enough variability to make it difficult to compare studies and generalize find­ings (Table 2).¹⁶ The variability of DUP is a limitation when comparing studies because it is a significant predic­tor of clinical outcome. Patients who abuse substances—and often are more challeng­ing to treat¹⁷—typically are excluded from these trials, which could explain the high response rate documented in studies of first-episode schizophrenia.

 

In addition, some FEP patients included in clinical trials might not be truly antipsy­chotic naïve; an estimated 25% to 75% of patients in these studies are antipsychotic naïve. This is an important consideration when comparing data on adverse effects that occur early in treatment. Additionally, acknowledging the advantages and disad­vantages of how to handle missing data is critical because of the high dropout rate observed in these studies.¹⁸

Efficacy. There is a high response rate to antipsychotic therapy—ranging from 46% to 96%, depending on the study—in patients with first-episode schizophrenia.³ The response mainly is seen in reduction of positive symptoms because typically negative and cognitive symptoms do not respond to antipsychotics. One study reported only 29% of patients achieved both positive and negative symptom remis­sion.¹⁹ It is likely that secondary negative symptoms caused by social withdrawal, reduced speech, and avoidance improve when positive symptoms subside, but pri­mary negative symptoms endure.In general, there is a lack of evidence suggesting that 1 antipsychotic class or agent is more effective than another. Studies mainly assess effectiveness using the primary outcome measure of all-cause discontinuation, such as the Clinical Antipsychotic Trials of Intervention Effectiveness study.²⁰ This outcome mea­sure is a mixture of patient preference, tol­erability, and efficacy that provides a more generalizable gauge on how well the treat­ment works in the clinic rather than tightly regulated settings such as clinical trials. A recent meta-analysis supports no differ­ences in efficacy among antipsychotics in early-episode psychosis.²¹

 

Tolerability. Because there are no significant differences among antipsychotic classes or agents in terms of efficacy in first-episode schizophrenia, drug selection is guided mainly by (1) the adverse effect profile and (2) what should be avoided depending on patient-specific variables. Evidence sug­gests first-episode patients are more sen­sitive to adverse effects of antipsychotics, particularly neurologic side effects (see this article at CurrentPsychiatry.com for a table comparing adverse effects of antipsychot­ics in first-episode psychosis).^18,22-29 Overall adverse effect profiles remain similar across FEP or multi-episode patients, but tend to be more exaggerated in drug-naïve patients with FEP.

Regarding FGA side effects, McEvoy et al¹⁸ demonstrated the neuroleptic threshold occurs at 50% lower haloperi­dol dosages in patients with first-episode schizophrenia (2.1 mg/d) compared with multi-episode schizophrenia (4.3 mg/d). Other trials suggest SGAs are associated with a lower risk of extrapyramidal side effects (EPS) or use of adjunctive therapies such as anticholinergic drugs or benzo­diazepines.^23-27 An exception to this state­ment is that higher risperidone dosages (≥4 to 6 mg/d) have been found to have higher rates of EPS and use of adjunctive medica­tions to treat these symptoms in FEP.²⁶ This is important because studies report higher discontinuation rates with more severe adverse effects of antipsychotics.

Cardiometabolic effects are of particular concern in first-episode patients because most weight gain happens in the first 3 to 4 months of treatment and remains throughout the first year.^18,24,29,30 Studies have shown that olanzapine, quetiap­ine, and risperidone are associated with more clinically significant weight gain compared with haloperidol and ziprasi­done.^23-25 Olanzapine-associated weight gain has been reported to be twice that of quetiapine and risperidone.¹⁸ Regardless, the EUFEST trial did not find a difference in clinically significant weight gain after 12 months among the antipsychotics studied, including haloperidol and ziprasidone.²⁵

Weight gain associated with these anti­psychotics is accompanied by changes in fasting triglycerides, glucose, total choles­terol,²³ and high-density lipoprotein cho­lesterol as well as an increase in body mass index (BMI) categorization²⁹ (eg, shift from normal to overweight).^18,25 Patients with lower baseline BMI and in racial minor­ity groups might experience more rapid weight gain regardless of antipsychotic selection.^29,30

Hyperprolactinemia could be under-recognized and could contribute to early treatment discontinuation.³¹ Evidence in patients with first-episode schizophrenia suggests similar outcomes as those seen in multi-episode patients, in whom ris­peridone is associated with higher pro­lactin elevations and clinically significant hyperprolactinemia (eg, galactorrhea and gynecomastia) compared with olanzapine, quetiapine, and low-dose haloperidol.^18,23,24 However, there is a lack of studies that assess whether long-term therapy with strong D2 receptor antagonists increases the risk of bone demineralization or path­ological fractures when started before patients’ bones reach maximum density in their mid-20s.³¹

Antipsychotic dosing
Given the high rate of treatment response in FEP and patients’ higher sensitivity to antipsychotic adverse effects, particularly EPS, guidelines recommend antipsychotic dosages lower than those used for multi-episode schizophrenia,¹¹ especially FGAs. Based on trial data, commonly used dos­ages include:
   • haloperidol, ≤5 mg/d^23-25,29
   • olanzapine, 10 mg/d^18,23,25,29
   • risperidone, ≤4 to 6 mg/d.^18,24,29,32

In general, haloperidol and risperidone, 2 to 3 mg/d, were well tolerated and effec­tive in trials. Higher quetiapine dosages of 500 to 600 mg/d could be required.^11,18,25

According to a survey on prescribing practices of antipsychotic selection and dosing in first-episode schizophrenia,4 clinical prescribing practices tend to use unnecessarily high initial antipsychotic dosing compared with trial data. There also is variability in the usual target anti­psychotic dosage ranging from 50% lower dosages to normal dosages in chronic schizophrenia to above FDA-approved maximum dosages for olanzapine (which may be necessary to counteract tobacco-induced cytochrome P450 1A2 enzyme induction).

In addition, these clinicians reported prescribing aripiprazole, an antipsychotic with weaker evidence (eg, case reports, case series, open-label studies) support­ing its efficacy and tolerability in FEP. These prescribing practices could reflect attempts to reduce the DUP and achieve symptom remission, so long as tolerability is not a concern.

Essentially, prescribed dosages should be based on symptom improvement and tolerability. This ideal dosage will vary as illustrated by Kapur et al,³³ who reported that FEP patients (N = 20) given haloper­idol, 1 mg or 2.5 mg/d, had D2 receptor occupancy rates of 38% to 87%, which was significantly dose-related (1 mg/d mean = 59%, 2.5 mg/d mean = 75%). Clinical response and EPS significantly increased as D2 receptor occupancy exceeded 65% and 78%, respectively.

Antipsychotic response
When should you expect to see symp­tom improvement in patients with first-episode schizophrenia?
Emsley et al³⁴ reported a 77.6% response rate among first-episode patients (N = 522) treated with low dosages of risperidone (mean modal dosage [MMD] = 3.3 mg/d) and haloperidol (MMD = 2.9 mg/d). They found variable response times that were evenly dispersed over a 10-week period. Nearly one-quarter (22.5%) did not respond until after week 4 and 11.2% did not respond until after week 8. In a study of FEP patients (N = 112) treated with olanzapine (MMD = 11.8 mg/d) or risperi­done (MMD = 3.9 mg/d), Gallego et al³⁵ reported a cumulative response of 39.6% at week 8 and 65.1% at week 16.

 

Although there is evidence that, among multi-episode patients, early nonresponse to antipsychotic therapy could predict subsequent nonresponse,³⁶ the evidence is mixed for first-episode schizophrenia. Studies by Emsley et al³⁴ and Gallego et al³⁵ did not find that early nonresponse at weeks 1 or 2 predicted subsequent nonre­sponse at week 4 or later. However, other studies support the idea that early nonre­sponse predicts subsequent nonresponse and early antipsychotic response predicts future response in first-episode patients, with good specificity and sensitivity.^37,38

Overall, treatment response in first-episode schizophrenia is variable. An adequate antipsychotic trial may be lon­ger, 8 to 16 weeks, compared with 4 to 8 weeks in multi-episode patients. Because research suggests that failure to respond to treatment may lead to medication non­adherence,³⁹ it is reasonable to consider switching antipsychotics when a patient experiences minimal or no response to antipsychotic therapy at week 2; however, this should be a patient-specific decision.

How long should you continue therapy after symptom remission?
There is a lack of consensus on the dura­tion of therapy for a patient treated for first-episode schizophrenia because a small percentage (10% to 20%) do not relapse after the first psychotic episode.³ In general, treatment guidelines and expert consensus statements recommend at least 1 to 2 years of treatment before considering a discon­tinuation trial.^7,10-11 Discuss the benefits and risks of maintenance treatment with your patient and obtain informed consent. With patients with minimal insight, obtaining proper consent is not possible and the phy­sician must exercise judgment unilaterally, if necessary, after educating the family.

After at least 12 months of treatment, antipsychotic therapy could continue indefinitely, depending on patient-specific factors. There are no predictors for identi­fying patients who do not require mainte­nance therapy beyond the first psychotic episode. The absence of negative and cog­nitive deficits could provide clues that a patient might be a candidate for antipsy­chotic tapering.

Predicting the treatment course
Research investigating clinical predic­tors or biomarkers that forecast whether a patient will respond to treatment is pre­liminary. Many characteristics have been identified (Table 3^{1,3,4,23,25,40}) and include shorter DUP,¹ poorer premorbid function,³ antipsychotic discontinuation,³ a trusting patient-doctor relationship,⁴¹ and antipsychotic-related adverse effects,^23,25 which are predictive of response, nonre­sponse, relapse, adherence, and nonadher­ence, respectively.
 

Bottom Line
The goals of pharmacological treatment of first-episode schizophrenia are to minimize the duration of untreated psychosis and target full remission of positive symptoms using the lowest possible antipsychotic dosages. Pharmacotherapy should continued for 1 to 2 years, with longer duration considered if it is discussed with the patient and with vigilant monitoring for adverse effects and suboptimal medication nonadherence to prevent relapse.
 

Editor’s note: The second article in this series in the July 2015 issue reviews the rationale and evidence for non-standard, first-line therapies, including long-acting injectable antipsychotics and clozapine.

 

Related Resources
• Recovery After an Initial Schizophrenia Episode (RAISE) Project Early Treatment Program. National Institute of Mental Health. http://raiseetp.org.
• Martens L, Baker S. Promoting recovery from first epi­sode psychosis: a guide for families. Centre for Addiction and Mental Health. http://www.camh.ca/en/hospital/ Documents/www.camh.net/AboutCAMH/Guideto CAMH/MentalHealthPrograms/SchizophreniaProgram/ 3936PromotingRecoveryFirstEpisodePsychosisfinal.pdf.

Drug Brand Names
Aripiprazole • Abilify                Lurasidone • Latuda
Asenapine • Saphris                Olanzapine • Zyprexa
Clozapine • Clozaril                 Paliperidone • Invega
Fluphenazine • Prolixin            Quetiapine • Seroquel
Iloperidone • Fanapt               Risperidone • Risperdal
Haloperidol • Haldol                Ziprasidone • Geodon

 

Disclosures
Dr. Gardner reports no financial relationships with any companies whose products are mentioned in this article or with manufacturers of competing products.
Dr. Nasrallah is a consultant to Acadia, Alkermes, Lundbeck, Janssen, Merck, Otsuka, and Sunovion, and is a speaker for Alkermes, Lundbeck, Janssen, Otsuka, and Sunovion.

Recent studies suggest that most older adults main­tain sexual interest well into late life; many, however, experience sexual dysfunction. This article provides psychiatric practitioners with current information regard­ing sexuality and aging, as well as psychiatric and systemic medical comorbidities and sexual side effects of medi­cations. Practice guidelines for assessing and managing sexual dysfunction have been developed for use in many medical specialties, and such guidance would be welcome in psychiatric practice.

This article addresses the myth of “geriatric asexuality” and its potential impact on clinical practice, the effects of age-related physiological changes on sexual activity, the importance of sexuality in the lives of older adults, and sensitive questions clinicians can pose about geriatric sexu­ality. We also will discuss:  
   • the importance of including a sexual assessment in the comprehensive psychiatric evaluation  
   • recognizing sexual dysfunction  
   • providing appropriate management within a multi-disciplinary, collaborative approach.

Sexuality after 65
Regardless of age, sexual activity can provide a sense of com­fort and elicit a positive emotional and physical response.¹ Hillman² defined human sexuality as any combination of sex­ual behavior, emotional intimacy, and sense of sexual identity.

Sexuality in the aging population gen­erally is an understudied area, obscured by the myth of “geriatric asexuality” and subject to numerous psychosocial vari­ables.¹ Previous research, focused on a bio­logical perspective of sexuality, has largely overlooked psychological and social influ­ences.³ It has been assumed that, with age, physical and hormonal changes or chronic illness ordinarily reduce or eliminate sex­ual desire and sexual behavior.³ However, the majority of older adults (defined as age ≥65) report a moderate-to-high level of sexual interest well into late life.^1,3

Sexual function remains a subject often neglected in psychiatry. Sexual dysfunc­tions, as described in the DSM-5,⁴ do not include age-related changes in sexual func­tion. In addition to physiological changes, sexual difficulties can result from relation­ship strain, systemic medical or psychi­atric disorders, and sexual side effects of medications.

CASE REPORT
Mr. C, age 71 and married, is being treated for a major depressive episode that followed a course of shingles and persistent posther­petic neuralgia. Medications are: escitalo­pram, 20 mg/d; pregabalin, 150 mg/d; and ramipril, 5 mg/d. Mr. C is physically active and involved in social activities; he has no substance use history. He attends clinic visits with his wife.

Mr. C reports that despite significant improvement of his depressive and pain symptoms, he now experiences sexual dif­ficulties, which he seems hesitant to discuss in detail. According to his wife, Mr. C appears to lack sexual desire and has difficulty initi­ating and maintaining an erection. She asks Mr. C’s psychiatrist whether she should stop her estrogen treatment, intended to enhance her sexual function, given that the couple is no longer engaging in sexual intercourse.

Mr. C admits to missing physical inti­macy; however, he states, “If I have to make a choice between having sex with my wife and getting this depression out of my head, I’m going to pick getting rid of the depres­sion.” Mrs. C says she is becoming dissatisfied with their marriage and the limited time she and her husband now spend together. Mr. C’s psychiatrist suggests that Mr. C and his wife undergo couples counseling.

Physiological changes with aging
In both women and men, the reproductive system undergoes age-related physiologi­cal changes.

Women. In women, the phase of decline in ovarian function and resulting decline in sex steroid production (estradiol and pro­gesterone) is referred to as the climacteric, with menopause being determined retro­spectively by the cessation of a menstrual period for 1 year.⁵

Menopausal symptoms typically occur between age 40 and 58; the average age of menopause is 51.^6,7 Both estradiol and pro­gesterone levels decline with menopause, and anovulation and ovarian failure ensue. A more gradual decline of female testoster­one levels also occurs with aging, starting in the fourth decade of life.⁸

Clinical manifestations of menopause include vasomotor symptoms (ie, “hot flushes”), sleep disturbances, anxiety and depressive symptoms, decreased bone min­eral density, and increased risk of cardio­vascular disease.^6,7 Loss of estrogen as well as continued loss of testosterone can result in dyspareunia because of atrophy and decreased vulvar and vaginal lubrication, with sexual excitement achieved less quickly, and a decreased intensity of orgasm.⁷

Men. Research has shown that testosterone levels are highest in men in the second and third decades, with a subsequent gradual decline.⁹ Older men with a low testosterone level are described as experiencing “late-onset hypogonadism,” also known by the popularized term “andropause.”¹⁰ This is attributed to decreased activity at the tes­ticular and hypothalamic levels.¹⁰

Nonetheless, only a small fraction of older men with confirmed androgen defi­ciency are clinically symptomatic.^11,12 Low testosterone is associated with decreased libido; it can hinder morning erections, contribute to erectile dysfunction, and result in erections that require physical stimulation.¹³

 

Notably, erectile dysfunction involves several other etiologic factors: psychiatric (eg, relationship difficulties, depression), neurogenic (eg, spinal cord injury), endo­crine (eg, hyperprolactinemia), arteriogenic (eg, hypertension, type 2 diabetes mellitus), and drug-induced (eg, antidepressants, antihypertensives).¹⁴ A low testosterone level also has been associated with potential cognitive changes, decreased bone mineral density, metabolic syndrome (eg, increased risk of type 2 diabetes mellitus), and cardio­vascular mortality.¹⁰

Effects on sexual activity. How much age-related physiological changes impact sexual practices in the geriatric popula­tion is uncertain. A study following 3,302 women through menopause over 6 years found some decline in sexual activity; how­ever, this decline was not associated with increased sexual pain, decreased desire, or lack of arousal.¹⁵ Men continue to find ways to remain sexually active despite physiolog­ical changes (eg, erectile difficulties), but the etiology of sexual dysfunction in later life remains multi-modal, involving physical, psychological, and relational factors.^16,17

Sexual practices in older adults
Researchers for the National Social Life, Health, and Aging Project (NSHAP) have examined sexual activities, behaviors, and problems in >3,000 older community-dwelling men and women across the United States, using information collected from in-home interviews.18 This study found that sexual activity, defined as any mutually voluntary sexual contact with another person, declines with age; how­ever, even in the oldest age group (age 75 to 85), 39% of men and 17% of women reported being sexually active in the last 12 months. Among these persons, 54% reported sexual activity at least 2 times per month; 23% reported having sex at least once a week; and 32% reported engaging in oral sex. Partner availability predicted sexual activity.

Respondents with self-reported poor physical health were more likely to experi­ence sexual problems (eg, difficulty with erection or lubrication, dyspareunia, and lack of pleasure). The most commonly reported reason for sexual inactivity in those with a spouse or other intimate partner was the male partner’s poor physical health.¹⁸

A longitudinal study, part of the Women’s Healthy Ageing Project, examined changes in sexual function at late menopause com­pared with early menopause. Although the researchers also found an age-related decrease in sexual activity, 50% of their late-menopause respondents (mean age, 70; range, 64 to 77) still reported sexual activity in the previous month, with 35% of this subgroup reporting sexual activity at least once a week; 83% reported sexual thoughts or fantasies.¹⁹ Availability of a partner, absence of a history of depression, moder­ate (compared with no) alcohol consump­tion, and better cognitive function were significantly associated with a higher level of sexual activity.¹⁹

In the Successful Aging Evaluation study, conducted in San Diego County, California, community-dwelling older partnered adults age 50 to 99 (mean age, 75) were surveyed about their sexual health after a cogni­tive screen by telephone²⁰; rating scales for depression, anxiety, and physical function also were included. Results included 41% of men and 35% of women reporting sexual activity at least once a week, and only 21% of men and 24% of women reporting no sex­ual activity in the previous year. Depressive symptoms were most highly correlated with lack of sexual activity.²⁰

Overall, these studies reveal that posi­tive physical and mental health, access to a healthy partner, and a positive attitude toward sex are correlated with sexual activ­ity in later life, whereas barriers to sexual activity include lack of a healthy sexual partner, depression, and chronic systemic medical illnesses, such as coronary artery disease or type 2 diabetes mellitus.^13,17,21-24 Sexual activity and satisfaction have been positively associated with perceived general well-being and self-esteem.^25,26 Conversely, sexual difficulties secondary to disease can be a source of distress for couples.²⁷

Possibly overlooked? It is important to note that sexuality itself is a subjective area. Psychological intimacy is a universal phe­nomenon, and its physical expression may evolve as couples accommodate to age-related bodily changes. Means of achieving physical closeness, other than intercourse (eg, intimate touching, hand holding, kiss­ing, or even acts of caretaking), may not be adequately captured in studies that look specifically at sexual activity.

Taking a sexual history in a geriatric patient
Because sexuality can be an uncomfort­able topic for geriatric patients to discuss, sexual problems in this population often go unrecognized. It has been suggested that psychiatrists are more likely to inquire about sexual activity in middle-aged patients than geriatric patients with the same psychiatric presentation—perhaps illustrating a bias against taking a sexual history from a geriatric patient.²⁸ However, because many older patients can experi­ence depression or anxiety disorders in relation to normal sexual changes or sex­ual dysfunction within the context of their intimate relationships, it is essential to bring these issues to light.

Although a sexual history may not be the focus of a first clinical encounter, consider making such an assessment at a relatively early stage of patient care. The importance of such dialogue is 2-fold:
   • It demonstrates to the patient that talking about sexuality in a respectful and empathic manner is appropriate and can encourage patients to communicate more effectively about sexuality with clinicians and with sexual partners.
   • It helps elicit medical information needed to make an accurate diagnosis and provide adequate management.

 

How to begin. As a starting point to taking a sexual history, an open-ended invitation for the geriatric patient to share informa­tion may be best, such as “What would you like to tell me about your sexual life?” See further suggestions (Table 1) and examples of more detailed questions to ask once a dialogue has been initiated (Table 2). Additional factors that may contribute to sexual dysfunction are presented in Table 3.^1,27,29,30

CASE CONTINUED
In Mr. C’s case, an assessment of his sexual history, including risk factors for sexual dysfunction, is completed. Results from laboratory investigations, including a total testosterone level, are within normal limits.

Mr. C asks about using medications with fewer sexual side effects (he has been taking 3 medications that can contribute to sexual dysfunction). A gradual cross-taper of esci­talopram, 20 mg/d, to mirtazapine, 45 mg/d, is implemented, along with tapering prega­balin to 50 mg/d.

Mr. C’s psychiatric and pain symptom improvement is maintained. He notices a boost in his sexual desire but has minimal improvement in erectile dysfunction. He is encouraged to speak with his primary care physician about an antihypertensive agent with less impact on sexual function, as well as therapeutic agents for erectile dysfunc­tion; these, he declines.

At a subsequent visit, Mr. C reports feeling less apprehension about sexual performance. He is now willing to consider further medica­tion options with his primary care physician, and agrees to a recommendation for couples psychotherapy.

As illustrated in Mr. C’s case, the recom­mended sexual assessment and manage­ment strategies to consider at a minimum in psychiatric practice are listed in Table 4.

STI risk in geriatric patients
The risk of sexually transmitted infections (STIs), including human immunodeficiency virus (HIV), often is overlooked in sexually active older adults. Although STIs are more common among younger adults, there is recent evidence of increased incidence in the geriatric population31 (with the high­est risk of incident HIV and some STIs in older men who have sex with men³²). These increased rates can be explained, at least in part, by:
   • older men being less likely to use a condom during sexual activity
   • promotion of viral entry in older women through a drier, thinner vaginal wall
   • increased longevity of HIV-positive persons.³¹

Routine STI screening is not warranted in all older adults, but education and prevention strategies in sexually active seniors who are at greater risk of STIs are recommended. Particularly, clinicians should seek opportunities to discuss risk factors and safe sex practices (eg, using condoms, limiting number of sexual part­ners, practicing good hygiene, engaging in preventive care), and provide behavioral counseling where appropriate.^31,33

Additional considerations in geriatric sexuality
Because psychiatric and systemic medical conditions can hinder sexual function, it is essential to identify and manage these conditions. Several neuropsychiatric dis­orders, including mood and neurocogni­tive disorders, can not only cause sexual dysfunction, but also can raise ethical dilemmas for clinicians, such as reduced decisional capacity in cognitively impaired patients to consent to sexual activity.^1,34

In some patients, psychological, envi­ronmental, and pharmacological treatment options may help. A phosphodiesterase type 5 inhibitor for erectile dysfunction can be prescribed by the primary care phy­sician, a psychiatrist, or another specialist, depending on the physician’s expertise and comfort level.

Sequencing of sexual dysfunction. Notably, there is a common paradox in mood disorders. Decreased sexual interest or performance may represent an aspect of anhedonia associated with depres­sion, whereas sexual dysfunction could also result from medication use (particularly that of serotonergic antidepressants, such as selective serotonin reuptake inhibitors and serotonin-norepinephrine inhibitors), even as other depressive symptoms improve. Therefore, it is critical to analyze sequencing of sexual dysfunction—as part of the pre­senting mood symptoms or dysfunction induced by antidepressant treatment.

Geriatric sexuality in the digital age. Because older adults represent a rapidly growing segment of digital device users,³⁵ Internet use is likely to play a role in the future of sexuality and “digital intimacy,” in that older adults can engage in online sexual activities. The Internet also can be a tool to access medical education.

Related Resources
• Burghardt KJ, Gardner KN. Sildenafil for SSRI-induced sexual dysfunction. Current Psychiatry. 2013;12(4):29-32,A.
• Maciel M, Laganà L. Older women’s sexual desire prob­lems: biopsychosocial factors impacting them and barriers to their clinical assessment [published online January 5, 2014]. Biomed Res Int. 2014;2014:107217. doi: 10.1155/2014/107217.

Drug Brand Names
Bupropion • Wellbutrin, Zyban                   Mirtazapine • Remeron
Carbamazepine • Tegretol                         Oxcarbazepine • Trileptal
Clonidine • Catapres                                 Phenobarbital • Luminal
Donepezil • Aricept                                   Phenytoin • Dilantin
Escitalopram • Lexapro                             Pregabalin • Lyrica
Gabapentin • Neurontin                            Ramipril • Altace
Lamotrigine • Lamictal                              Rivastigmine • Exelon
Lithium • Eskalith, Lithobid                       Trazodone • Desyrel
Memantine • Namenda                             Valproic acid • Depakote

 

Disclosures
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Recent studies suggest that most older adults main­tain sexual interest well into late life; many, however, experience sexual dysfunction. This article provides psychiatric practitioners with current information regard­ing sexuality and aging, as well as psychiatric and systemic medical comorbidities and sexual side effects of medi­cations. Practice guidelines for assessing and managing sexual dysfunction have been developed for use in many medical specialties, and such guidance would be welcome in psychiatric practice.

This article addresses the myth of “geriatric asexuality” and its potential impact on clinical practice, the effects of age-related physiological changes on sexual activity, the importance of sexuality in the lives of older adults, and sensitive questions clinicians can pose about geriatric sexu­ality. We also will discuss:  
   • the importance of including a sexual assessment in the comprehensive psychiatric evaluation  
   • recognizing sexual dysfunction  
   • providing appropriate management within a multi-disciplinary, collaborative approach.

Sexuality after 65
Regardless of age, sexual activity can provide a sense of com­fort and elicit a positive emotional and physical response.¹ Hillman² defined human sexuality as any combination of sex­ual behavior, emotional intimacy, and sense of sexual identity.

Sexuality in the aging population gen­erally is an understudied area, obscured by the myth of “geriatric asexuality” and subject to numerous psychosocial vari­ables.¹ Previous research, focused on a bio­logical perspective of sexuality, has largely overlooked psychological and social influ­ences.³ It has been assumed that, with age, physical and hormonal changes or chronic illness ordinarily reduce or eliminate sex­ual desire and sexual behavior.³ However, the majority of older adults (defined as age ≥65) report a moderate-to-high level of sexual interest well into late life.^1,3

Sexual function remains a subject often neglected in psychiatry. Sexual dysfunc­tions, as described in the DSM-5,⁴ do not include age-related changes in sexual func­tion. In addition to physiological changes, sexual difficulties can result from relation­ship strain, systemic medical or psychi­atric disorders, and sexual side effects of medications.

CASE REPORT
Mr. C, age 71 and married, is being treated for a major depressive episode that followed a course of shingles and persistent posther­petic neuralgia. Medications are: escitalo­pram, 20 mg/d; pregabalin, 150 mg/d; and ramipril, 5 mg/d. Mr. C is physically active and involved in social activities; he has no substance use history. He attends clinic visits with his wife.

Mr. C reports that despite significant improvement of his depressive and pain symptoms, he now experiences sexual dif­ficulties, which he seems hesitant to discuss in detail. According to his wife, Mr. C appears to lack sexual desire and has difficulty initi­ating and maintaining an erection. She asks Mr. C’s psychiatrist whether she should stop her estrogen treatment, intended to enhance her sexual function, given that the couple is no longer engaging in sexual intercourse.

Mr. C admits to missing physical inti­macy; however, he states, “If I have to make a choice between having sex with my wife and getting this depression out of my head, I’m going to pick getting rid of the depres­sion.” Mrs. C says she is becoming dissatisfied with their marriage and the limited time she and her husband now spend together. Mr. C’s psychiatrist suggests that Mr. C and his wife undergo couples counseling.

Physiological changes with aging
In both women and men, the reproductive system undergoes age-related physiologi­cal changes.

Women. In women, the phase of decline in ovarian function and resulting decline in sex steroid production (estradiol and pro­gesterone) is referred to as the climacteric, with menopause being determined retro­spectively by the cessation of a menstrual period for 1 year.⁵

Menopausal symptoms typically occur between age 40 and 58; the average age of menopause is 51.^6,7 Both estradiol and pro­gesterone levels decline with menopause, and anovulation and ovarian failure ensue. A more gradual decline of female testoster­one levels also occurs with aging, starting in the fourth decade of life.⁸

Clinical manifestations of menopause include vasomotor symptoms (ie, “hot flushes”), sleep disturbances, anxiety and depressive symptoms, decreased bone min­eral density, and increased risk of cardio­vascular disease.^6,7 Loss of estrogen as well as continued loss of testosterone can result in dyspareunia because of atrophy and decreased vulvar and vaginal lubrication, with sexual excitement achieved less quickly, and a decreased intensity of orgasm.⁷

Men. Research has shown that testosterone levels are highest in men in the second and third decades, with a subsequent gradual decline.⁹ Older men with a low testosterone level are described as experiencing “late-onset hypogonadism,” also known by the popularized term “andropause.”¹⁰ This is attributed to decreased activity at the tes­ticular and hypothalamic levels.¹⁰

Nonetheless, only a small fraction of older men with confirmed androgen defi­ciency are clinically symptomatic.^11,12 Low testosterone is associated with decreased libido; it can hinder morning erections, contribute to erectile dysfunction, and result in erections that require physical stimulation.¹³

 

Notably, erectile dysfunction involves several other etiologic factors: psychiatric (eg, relationship difficulties, depression), neurogenic (eg, spinal cord injury), endo­crine (eg, hyperprolactinemia), arteriogenic (eg, hypertension, type 2 diabetes mellitus), and drug-induced (eg, antidepressants, antihypertensives).¹⁴ A low testosterone level also has been associated with potential cognitive changes, decreased bone mineral density, metabolic syndrome (eg, increased risk of type 2 diabetes mellitus), and cardio­vascular mortality.¹⁰

Effects on sexual activity. How much age-related physiological changes impact sexual practices in the geriatric popula­tion is uncertain. A study following 3,302 women through menopause over 6 years found some decline in sexual activity; how­ever, this decline was not associated with increased sexual pain, decreased desire, or lack of arousal.¹⁵ Men continue to find ways to remain sexually active despite physiolog­ical changes (eg, erectile difficulties), but the etiology of sexual dysfunction in later life remains multi-modal, involving physical, psychological, and relational factors.^16,17

Sexual practices in older adults
Researchers for the National Social Life, Health, and Aging Project (NSHAP) have examined sexual activities, behaviors, and problems in >3,000 older community-dwelling men and women across the United States, using information collected from in-home interviews.18 This study found that sexual activity, defined as any mutually voluntary sexual contact with another person, declines with age; how­ever, even in the oldest age group (age 75 to 85), 39% of men and 17% of women reported being sexually active in the last 12 months. Among these persons, 54% reported sexual activity at least 2 times per month; 23% reported having sex at least once a week; and 32% reported engaging in oral sex. Partner availability predicted sexual activity.

Respondents with self-reported poor physical health were more likely to experi­ence sexual problems (eg, difficulty with erection or lubrication, dyspareunia, and lack of pleasure). The most commonly reported reason for sexual inactivity in those with a spouse or other intimate partner was the male partner’s poor physical health.¹⁸

A longitudinal study, part of the Women’s Healthy Ageing Project, examined changes in sexual function at late menopause com­pared with early menopause. Although the researchers also found an age-related decrease in sexual activity, 50% of their late-menopause respondents (mean age, 70; range, 64 to 77) still reported sexual activity in the previous month, with 35% of this subgroup reporting sexual activity at least once a week; 83% reported sexual thoughts or fantasies.¹⁹ Availability of a partner, absence of a history of depression, moder­ate (compared with no) alcohol consump­tion, and better cognitive function were significantly associated with a higher level of sexual activity.¹⁹

In the Successful Aging Evaluation study, conducted in San Diego County, California, community-dwelling older partnered adults age 50 to 99 (mean age, 75) were surveyed about their sexual health after a cogni­tive screen by telephone²⁰; rating scales for depression, anxiety, and physical function also were included. Results included 41% of men and 35% of women reporting sexual activity at least once a week, and only 21% of men and 24% of women reporting no sex­ual activity in the previous year. Depressive symptoms were most highly correlated with lack of sexual activity.²⁰

Overall, these studies reveal that posi­tive physical and mental health, access to a healthy partner, and a positive attitude toward sex are correlated with sexual activ­ity in later life, whereas barriers to sexual activity include lack of a healthy sexual partner, depression, and chronic systemic medical illnesses, such as coronary artery disease or type 2 diabetes mellitus.^13,17,21-24 Sexual activity and satisfaction have been positively associated with perceived general well-being and self-esteem.^25,26 Conversely, sexual difficulties secondary to disease can be a source of distress for couples.²⁷

Possibly overlooked? It is important to note that sexuality itself is a subjective area. Psychological intimacy is a universal phe­nomenon, and its physical expression may evolve as couples accommodate to age-related bodily changes. Means of achieving physical closeness, other than intercourse (eg, intimate touching, hand holding, kiss­ing, or even acts of caretaking), may not be adequately captured in studies that look specifically at sexual activity.

Taking a sexual history in a geriatric patient
Because sexuality can be an uncomfort­able topic for geriatric patients to discuss, sexual problems in this population often go unrecognized. It has been suggested that psychiatrists are more likely to inquire about sexual activity in middle-aged patients than geriatric patients with the same psychiatric presentation—perhaps illustrating a bias against taking a sexual history from a geriatric patient.²⁸ However, because many older patients can experi­ence depression or anxiety disorders in relation to normal sexual changes or sex­ual dysfunction within the context of their intimate relationships, it is essential to bring these issues to light.

Although a sexual history may not be the focus of a first clinical encounter, consider making such an assessment at a relatively early stage of patient care. The importance of such dialogue is 2-fold:
   • It demonstrates to the patient that talking about sexuality in a respectful and empathic manner is appropriate and can encourage patients to communicate more effectively about sexuality with clinicians and with sexual partners.
   • It helps elicit medical information needed to make an accurate diagnosis and provide adequate management.

 

How to begin. As a starting point to taking a sexual history, an open-ended invitation for the geriatric patient to share informa­tion may be best, such as “What would you like to tell me about your sexual life?” See further suggestions (Table 1) and examples of more detailed questions to ask once a dialogue has been initiated (Table 2). Additional factors that may contribute to sexual dysfunction are presented in Table 3.^1,27,29,30

CASE CONTINUED
In Mr. C’s case, an assessment of his sexual history, including risk factors for sexual dysfunction, is completed. Results from laboratory investigations, including a total testosterone level, are within normal limits.

Mr. C asks about using medications with fewer sexual side effects (he has been taking 3 medications that can contribute to sexual dysfunction). A gradual cross-taper of esci­talopram, 20 mg/d, to mirtazapine, 45 mg/d, is implemented, along with tapering prega­balin to 50 mg/d.

Mr. C’s psychiatric and pain symptom improvement is maintained. He notices a boost in his sexual desire but has minimal improvement in erectile dysfunction. He is encouraged to speak with his primary care physician about an antihypertensive agent with less impact on sexual function, as well as therapeutic agents for erectile dysfunc­tion; these, he declines.

At a subsequent visit, Mr. C reports feeling less apprehension about sexual performance. He is now willing to consider further medica­tion options with his primary care physician, and agrees to a recommendation for couples psychotherapy.

As illustrated in Mr. C’s case, the recom­mended sexual assessment and manage­ment strategies to consider at a minimum in psychiatric practice are listed in Table 4.

STI risk in geriatric patients
The risk of sexually transmitted infections (STIs), including human immunodeficiency virus (HIV), often is overlooked in sexually active older adults. Although STIs are more common among younger adults, there is recent evidence of increased incidence in the geriatric population31 (with the high­est risk of incident HIV and some STIs in older men who have sex with men³²). These increased rates can be explained, at least in part, by:
   • older men being less likely to use a condom during sexual activity
   • promotion of viral entry in older women through a drier, thinner vaginal wall
   • increased longevity of HIV-positive persons.³¹

Routine STI screening is not warranted in all older adults, but education and prevention strategies in sexually active seniors who are at greater risk of STIs are recommended. Particularly, clinicians should seek opportunities to discuss risk factors and safe sex practices (eg, using condoms, limiting number of sexual part­ners, practicing good hygiene, engaging in preventive care), and provide behavioral counseling where appropriate.^31,33

Additional considerations in geriatric sexuality
Because psychiatric and systemic medical conditions can hinder sexual function, it is essential to identify and manage these conditions. Several neuropsychiatric dis­orders, including mood and neurocogni­tive disorders, can not only cause sexual dysfunction, but also can raise ethical dilemmas for clinicians, such as reduced decisional capacity in cognitively impaired patients to consent to sexual activity.^1,34

In some patients, psychological, envi­ronmental, and pharmacological treatment options may help. A phosphodiesterase type 5 inhibitor for erectile dysfunction can be prescribed by the primary care phy­sician, a psychiatrist, or another specialist, depending on the physician’s expertise and comfort level.

Sequencing of sexual dysfunction. Notably, there is a common paradox in mood disorders. Decreased sexual interest or performance may represent an aspect of anhedonia associated with depres­sion, whereas sexual dysfunction could also result from medication use (particularly that of serotonergic antidepressants, such as selective serotonin reuptake inhibitors and serotonin-norepinephrine inhibitors), even as other depressive symptoms improve. Therefore, it is critical to analyze sequencing of sexual dysfunction—as part of the pre­senting mood symptoms or dysfunction induced by antidepressant treatment.

Geriatric sexuality in the digital age. Because older adults represent a rapidly growing segment of digital device users,³⁵ Internet use is likely to play a role in the future of sexuality and “digital intimacy,” in that older adults can engage in online sexual activities. The Internet also can be a tool to access medical education.

Related Resources
• Burghardt KJ, Gardner KN. Sildenafil for SSRI-induced sexual dysfunction. Current Psychiatry. 2013;12(4):29-32,A.
• Maciel M, Laganà L. Older women’s sexual desire prob­lems: biopsychosocial factors impacting them and barriers to their clinical assessment [published online January 5, 2014]. Biomed Res Int. 2014;2014:107217. doi: 10.1155/2014/107217.

Drug Brand Names
Bupropion • Wellbutrin, Zyban                   Mirtazapine • Remeron
Carbamazepine • Tegretol                         Oxcarbazepine • Trileptal
Clonidine • Catapres                                 Phenobarbital • Luminal
Donepezil • Aricept                                   Phenytoin • Dilantin
Escitalopram • Lexapro                             Pregabalin • Lyrica
Gabapentin • Neurontin                            Ramipril • Altace
Lamotrigine • Lamictal                              Rivastigmine • Exelon
Lithium • Eskalith, Lithobid                       Trazodone • Desyrel
Memantine • Namenda                             Valproic acid • Depakote

 

Disclosures
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Recent studies suggest that most older adults main­tain sexual interest well into late life; many, however, experience sexual dysfunction. This article provides psychiatric practitioners with current information regard­ing sexuality and aging, as well as psychiatric and systemic medical comorbidities and sexual side effects of medi­cations. Practice guidelines for assessing and managing sexual dysfunction have been developed for use in many medical specialties, and such guidance would be welcome in psychiatric practice.

This article addresses the myth of “geriatric asexuality” and its potential impact on clinical practice, the effects of age-related physiological changes on sexual activity, the importance of sexuality in the lives of older adults, and sensitive questions clinicians can pose about geriatric sexu­ality. We also will discuss:  
   • the importance of including a sexual assessment in the comprehensive psychiatric evaluation  
   • recognizing sexual dysfunction  
   • providing appropriate management within a multi-disciplinary, collaborative approach.

Sexuality after 65
Regardless of age, sexual activity can provide a sense of com­fort and elicit a positive emotional and physical response.¹ Hillman² defined human sexuality as any combination of sex­ual behavior, emotional intimacy, and sense of sexual identity.

Sexuality in the aging population gen­erally is an understudied area, obscured by the myth of “geriatric asexuality” and subject to numerous psychosocial vari­ables.¹ Previous research, focused on a bio­logical perspective of sexuality, has largely overlooked psychological and social influ­ences.³ It has been assumed that, with age, physical and hormonal changes or chronic illness ordinarily reduce or eliminate sex­ual desire and sexual behavior.³ However, the majority of older adults (defined as age ≥65) report a moderate-to-high level of sexual interest well into late life.^1,3

Sexual function remains a subject often neglected in psychiatry. Sexual dysfunc­tions, as described in the DSM-5,⁴ do not include age-related changes in sexual func­tion. In addition to physiological changes, sexual difficulties can result from relation­ship strain, systemic medical or psychi­atric disorders, and sexual side effects of medications.

CASE REPORT
Mr. C, age 71 and married, is being treated for a major depressive episode that followed a course of shingles and persistent posther­petic neuralgia. Medications are: escitalo­pram, 20 mg/d; pregabalin, 150 mg/d; and ramipril, 5 mg/d. Mr. C is physically active and involved in social activities; he has no substance use history. He attends clinic visits with his wife.

Mr. C reports that despite significant improvement of his depressive and pain symptoms, he now experiences sexual dif­ficulties, which he seems hesitant to discuss in detail. According to his wife, Mr. C appears to lack sexual desire and has difficulty initi­ating and maintaining an erection. She asks Mr. C’s psychiatrist whether she should stop her estrogen treatment, intended to enhance her sexual function, given that the couple is no longer engaging in sexual intercourse.

Mr. C admits to missing physical inti­macy; however, he states, “If I have to make a choice between having sex with my wife and getting this depression out of my head, I’m going to pick getting rid of the depres­sion.” Mrs. C says she is becoming dissatisfied with their marriage and the limited time she and her husband now spend together. Mr. C’s psychiatrist suggests that Mr. C and his wife undergo couples counseling.

Physiological changes with aging
In both women and men, the reproductive system undergoes age-related physiologi­cal changes.

Women. In women, the phase of decline in ovarian function and resulting decline in sex steroid production (estradiol and pro­gesterone) is referred to as the climacteric, with menopause being determined retro­spectively by the cessation of a menstrual period for 1 year.⁵

Menopausal symptoms typically occur between age 40 and 58; the average age of menopause is 51.^6,7 Both estradiol and pro­gesterone levels decline with menopause, and anovulation and ovarian failure ensue. A more gradual decline of female testoster­one levels also occurs with aging, starting in the fourth decade of life.⁸

Clinical manifestations of menopause include vasomotor symptoms (ie, “hot flushes”), sleep disturbances, anxiety and depressive symptoms, decreased bone min­eral density, and increased risk of cardio­vascular disease.^6,7 Loss of estrogen as well as continued loss of testosterone can result in dyspareunia because of atrophy and decreased vulvar and vaginal lubrication, with sexual excitement achieved less quickly, and a decreased intensity of orgasm.⁷

Men. Research has shown that testosterone levels are highest in men in the second and third decades, with a subsequent gradual decline.⁹ Older men with a low testosterone level are described as experiencing “late-onset hypogonadism,” also known by the popularized term “andropause.”¹⁰ This is attributed to decreased activity at the tes­ticular and hypothalamic levels.¹⁰

Nonetheless, only a small fraction of older men with confirmed androgen defi­ciency are clinically symptomatic.^11,12 Low testosterone is associated with decreased libido; it can hinder morning erections, contribute to erectile dysfunction, and result in erections that require physical stimulation.¹³

 

Notably, erectile dysfunction involves several other etiologic factors: psychiatric (eg, relationship difficulties, depression), neurogenic (eg, spinal cord injury), endo­crine (eg, hyperprolactinemia), arteriogenic (eg, hypertension, type 2 diabetes mellitus), and drug-induced (eg, antidepressants, antihypertensives).¹⁴ A low testosterone level also has been associated with potential cognitive changes, decreased bone mineral density, metabolic syndrome (eg, increased risk of type 2 diabetes mellitus), and cardio­vascular mortality.¹⁰

Effects on sexual activity. How much age-related physiological changes impact sexual practices in the geriatric popula­tion is uncertain. A study following 3,302 women through menopause over 6 years found some decline in sexual activity; how­ever, this decline was not associated with increased sexual pain, decreased desire, or lack of arousal.¹⁵ Men continue to find ways to remain sexually active despite physiolog­ical changes (eg, erectile difficulties), but the etiology of sexual dysfunction in later life remains multi-modal, involving physical, psychological, and relational factors.^16,17

Sexual practices in older adults
Researchers for the National Social Life, Health, and Aging Project (NSHAP) have examined sexual activities, behaviors, and problems in >3,000 older community-dwelling men and women across the United States, using information collected from in-home interviews.18 This study found that sexual activity, defined as any mutually voluntary sexual contact with another person, declines with age; how­ever, even in the oldest age group (age 75 to 85), 39% of men and 17% of women reported being sexually active in the last 12 months. Among these persons, 54% reported sexual activity at least 2 times per month; 23% reported having sex at least once a week; and 32% reported engaging in oral sex. Partner availability predicted sexual activity.

Respondents with self-reported poor physical health were more likely to experi­ence sexual problems (eg, difficulty with erection or lubrication, dyspareunia, and lack of pleasure). The most commonly reported reason for sexual inactivity in those with a spouse or other intimate partner was the male partner’s poor physical health.¹⁸

A longitudinal study, part of the Women’s Healthy Ageing Project, examined changes in sexual function at late menopause com­pared with early menopause. Although the researchers also found an age-related decrease in sexual activity, 50% of their late-menopause respondents (mean age, 70; range, 64 to 77) still reported sexual activity in the previous month, with 35% of this subgroup reporting sexual activity at least once a week; 83% reported sexual thoughts or fantasies.¹⁹ Availability of a partner, absence of a history of depression, moder­ate (compared with no) alcohol consump­tion, and better cognitive function were significantly associated with a higher level of sexual activity.¹⁹

In the Successful Aging Evaluation study, conducted in San Diego County, California, community-dwelling older partnered adults age 50 to 99 (mean age, 75) were surveyed about their sexual health after a cogni­tive screen by telephone²⁰; rating scales for depression, anxiety, and physical function also were included. Results included 41% of men and 35% of women reporting sexual activity at least once a week, and only 21% of men and 24% of women reporting no sex­ual activity in the previous year. Depressive symptoms were most highly correlated with lack of sexual activity.²⁰

Overall, these studies reveal that posi­tive physical and mental health, access to a healthy partner, and a positive attitude toward sex are correlated with sexual activ­ity in later life, whereas barriers to sexual activity include lack of a healthy sexual partner, depression, and chronic systemic medical illnesses, such as coronary artery disease or type 2 diabetes mellitus.^13,17,21-24 Sexual activity and satisfaction have been positively associated with perceived general well-being and self-esteem.^25,26 Conversely, sexual difficulties secondary to disease can be a source of distress for couples.²⁷

Possibly overlooked? It is important to note that sexuality itself is a subjective area. Psychological intimacy is a universal phe­nomenon, and its physical expression may evolve as couples accommodate to age-related bodily changes. Means of achieving physical closeness, other than intercourse (eg, intimate touching, hand holding, kiss­ing, or even acts of caretaking), may not be adequately captured in studies that look specifically at sexual activity.

Taking a sexual history in a geriatric patient
Because sexuality can be an uncomfort­able topic for geriatric patients to discuss, sexual problems in this population often go unrecognized. It has been suggested that psychiatrists are more likely to inquire about sexual activity in middle-aged patients than geriatric patients with the same psychiatric presentation—perhaps illustrating a bias against taking a sexual history from a geriatric patient.²⁸ However, because many older patients can experi­ence depression or anxiety disorders in relation to normal sexual changes or sex­ual dysfunction within the context of their intimate relationships, it is essential to bring these issues to light.

Although a sexual history may not be the focus of a first clinical encounter, consider making such an assessment at a relatively early stage of patient care. The importance of such dialogue is 2-fold:
   • It demonstrates to the patient that talking about sexuality in a respectful and empathic manner is appropriate and can encourage patients to communicate more effectively about sexuality with clinicians and with sexual partners.
   • It helps elicit medical information needed to make an accurate diagnosis and provide adequate management.

 

How to begin. As a starting point to taking a sexual history, an open-ended invitation for the geriatric patient to share informa­tion may be best, such as “What would you like to tell me about your sexual life?” See further suggestions (Table 1) and examples of more detailed questions to ask once a dialogue has been initiated (Table 2). Additional factors that may contribute to sexual dysfunction are presented in Table 3.^1,27,29,30

CASE CONTINUED
In Mr. C’s case, an assessment of his sexual history, including risk factors for sexual dysfunction, is completed. Results from laboratory investigations, including a total testosterone level, are within normal limits.

Mr. C asks about using medications with fewer sexual side effects (he has been taking 3 medications that can contribute to sexual dysfunction). A gradual cross-taper of esci­talopram, 20 mg/d, to mirtazapine, 45 mg/d, is implemented, along with tapering prega­balin to 50 mg/d.

Mr. C’s psychiatric and pain symptom improvement is maintained. He notices a boost in his sexual desire but has minimal improvement in erectile dysfunction. He is encouraged to speak with his primary care physician about an antihypertensive agent with less impact on sexual function, as well as therapeutic agents for erectile dysfunc­tion; these, he declines.

At a subsequent visit, Mr. C reports feeling less apprehension about sexual performance. He is now willing to consider further medica­tion options with his primary care physician, and agrees to a recommendation for couples psychotherapy.

As illustrated in Mr. C’s case, the recom­mended sexual assessment and manage­ment strategies to consider at a minimum in psychiatric practice are listed in Table 4.

STI risk in geriatric patients
The risk of sexually transmitted infections (STIs), including human immunodeficiency virus (HIV), often is overlooked in sexually active older adults. Although STIs are more common among younger adults, there is recent evidence of increased incidence in the geriatric population31 (with the high­est risk of incident HIV and some STIs in older men who have sex with men³²). These increased rates can be explained, at least in part, by:
   • older men being less likely to use a condom during sexual activity
   • promotion of viral entry in older women through a drier, thinner vaginal wall
   • increased longevity of HIV-positive persons.³¹

Routine STI screening is not warranted in all older adults, but education and prevention strategies in sexually active seniors who are at greater risk of STIs are recommended. Particularly, clinicians should seek opportunities to discuss risk factors and safe sex practices (eg, using condoms, limiting number of sexual part­ners, practicing good hygiene, engaging in preventive care), and provide behavioral counseling where appropriate.^31,33

Additional considerations in geriatric sexuality
Because psychiatric and systemic medical conditions can hinder sexual function, it is essential to identify and manage these conditions. Several neuropsychiatric dis­orders, including mood and neurocogni­tive disorders, can not only cause sexual dysfunction, but also can raise ethical dilemmas for clinicians, such as reduced decisional capacity in cognitively impaired patients to consent to sexual activity.^1,34

In some patients, psychological, envi­ronmental, and pharmacological treatment options may help. A phosphodiesterase type 5 inhibitor for erectile dysfunction can be prescribed by the primary care phy­sician, a psychiatrist, or another specialist, depending on the physician’s expertise and comfort level.

Sequencing of sexual dysfunction. Notably, there is a common paradox in mood disorders. Decreased sexual interest or performance may represent an aspect of anhedonia associated with depres­sion, whereas sexual dysfunction could also result from medication use (particularly that of serotonergic antidepressants, such as selective serotonin reuptake inhibitors and serotonin-norepinephrine inhibitors), even as other depressive symptoms improve. Therefore, it is critical to analyze sequencing of sexual dysfunction—as part of the pre­senting mood symptoms or dysfunction induced by antidepressant treatment.

Geriatric sexuality in the digital age. Because older adults represent a rapidly growing segment of digital device users,³⁵ Internet use is likely to play a role in the future of sexuality and “digital intimacy,” in that older adults can engage in online sexual activities. The Internet also can be a tool to access medical education.

Related Resources
• Burghardt KJ, Gardner KN. Sildenafil for SSRI-induced sexual dysfunction. Current Psychiatry. 2013;12(4):29-32,A.
• Maciel M, Laganà L. Older women’s sexual desire prob­lems: biopsychosocial factors impacting them and barriers to their clinical assessment [published online January 5, 2014]. Biomed Res Int. 2014;2014:107217. doi: 10.1155/2014/107217.

Drug Brand Names
Bupropion • Wellbutrin, Zyban                   Mirtazapine • Remeron
Carbamazepine • Tegretol                         Oxcarbazepine • Trileptal
Clonidine • Catapres                                 Phenobarbital • Luminal
Donepezil • Aricept                                   Phenytoin • Dilantin
Escitalopram • Lexapro                             Pregabalin • Lyrica
Gabapentin • Neurontin                            Ramipril • Altace
Lamotrigine • Lamictal                              Rivastigmine • Exelon
Lithium • Eskalith, Lithobid                       Trazodone • Desyrel
Memantine • Namenda                             Valproic acid • Depakote

 

Disclosures
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.