Clinical Inquiries

Evidence summary

Even though clinical research is lacking, acetaminophen is widely used to relieve low-back pain with no documented teratogenic effect (US Food and Drug Administration [FDA] category B). Nonsteroidal anti-inflammatory drugs are classified as FDA category D in the third trimester because of their documented association with oligohydramnios, premature closure of the ductus arteriosus, nephrotoxicity, and periventricular hemorrhage in the fetus.¹ Opioids are category C and a poor choice to treat low-back pain in pregnancy.²

Physical therapy and water aerobics relieve pain, reduce sick days
A 2007 Cochrane review of interventions for treating back pain in pregnancy analyzed 8 studies with a total of 1305 patients that examined the effects of adding physical therapy and acupuncture to usual care.³ In one RCT, 407 patients with and without pain received 5 30-minute individualized physical therapy exercise sessions, 2 45-minute group physical therapy classes, or standard care.^3,4 Low-back pain decreased with group physical therapy (P<.05; number needed to treat [NNT] = 3.2) and individual therapy (NNT = 2.1). Patients who received individual therapy had a 12% decrease in sick days.

A prospective trial of 258 patients, half of whom did water aerobics and half physical therapy, showed comparable results for the 2 interventions (NNT = 11.4 for decreased sick days; odds ratio = 0.38, 95% confidence interval [CI], 0.16-0.88).³

Acupuncture reduces pain and analgesic use
A prospective, randomized open study cited in the 2007 Cochrane review divided 72 patients at 24 to 37 weeks’ gestational age into a group that received acupuncture plus standard care and a standard-care-only control group.^3,5 Treatment sessions occurred one or 2 times per week until delivery or recovery. The acupuncture group reported decreased pain (60% vs 14% for controls; P<.01; NNT = 2.2) and improved function (43% vs 9% for controls; P<.001; NNT = 2.9). There was also a difference in analgesic use: 0% for the acupuncture group vs 14% for controls; P<.05; NNT = 7.1.

A 2009 RCT divided 159 patients at 25 to 38 weeks’ gestational age into 3 groups: auricular acupuncture at specific points for one week, sham auricular acupuncture at nonspecific points for one week, and controls. At the end of Week 1, 80% of the acupuncture group had a clinically significant reduction in pain compared with 56% in the sham acupuncture group and 36% in the control group (P = .001 acupuncture vs sham, NNT = 4.2; P<.0001 acupuncture vs controls, NNT = 2.3).⁶

Osteopathic manipulative therapy (OMT) decreases disability, but not pain
A 2010 RCT divided 144 third trimester patients into 3 groups that received usual obstetric care, sham ultrasound therapy plus usual obstetric care, or OMT.⁷ Pain remained similar among the 3 groups throughout the study. Using the 24-point Roland-Morris Disability Questionnaire, OMT decreased disability by 0.72 points (95% CI, 0.31-1.14; P<.001) compared with 0.35 points in the usual obstetric-care-only group (95% CI, –0.06 to 0.76; P = .09). Ultrasound had no effect.

Corticosteroid injection reduces pain in a small trial
A small RCT of injection with the corticosteroid triamcinolone at the sacrospinous ligament insertion in 36 women with low-back pain showed significant reduction in pain in 17 of 18 women in the triamcinolone group compared with 9 of 18 women in the control group (P<.01; NNT = 2).⁸

Evidence lacking on maternity support garments
A poor-quality systematic review of 10 studies (N = 1909) of maternity support garments found insufficient evidence because of the heterogeneity of the trials.⁹

Recommendations

The American College of Obstetricians and Gynecologists suggests the following measures to prevent and treat low-back pain in pregnancy:¹⁰

• wear low-heeled (not flat) shoes with good arch support
• get help when lifting heavy objects
• place one foot on a stool or box when standing for long periods
• place a board between the mattress and box spring if the bed is too soft
• squat down, bend knees, and keep back straight when lifting
• sit in chairs with good back support or use a small pillow to provide support
• sleep on side with pillows between knees for support
• apply heat, cold, or massage to the painful area.

Evidence summary

Even though clinical research is lacking, acetaminophen is widely used to relieve low-back pain with no documented teratogenic effect (US Food and Drug Administration [FDA] category B). Nonsteroidal anti-inflammatory drugs are classified as FDA category D in the third trimester because of their documented association with oligohydramnios, premature closure of the ductus arteriosus, nephrotoxicity, and periventricular hemorrhage in the fetus.¹ Opioids are category C and a poor choice to treat low-back pain in pregnancy.²

Physical therapy and water aerobics relieve pain, reduce sick days
A 2007 Cochrane review of interventions for treating back pain in pregnancy analyzed 8 studies with a total of 1305 patients that examined the effects of adding physical therapy and acupuncture to usual care.³ In one RCT, 407 patients with and without pain received 5 30-minute individualized physical therapy exercise sessions, 2 45-minute group physical therapy classes, or standard care.^3,4 Low-back pain decreased with group physical therapy (P<.05; number needed to treat [NNT] = 3.2) and individual therapy (NNT = 2.1). Patients who received individual therapy had a 12% decrease in sick days.

A prospective trial of 258 patients, half of whom did water aerobics and half physical therapy, showed comparable results for the 2 interventions (NNT = 11.4 for decreased sick days; odds ratio = 0.38, 95% confidence interval [CI], 0.16-0.88).³

Acupuncture reduces pain and analgesic use
A prospective, randomized open study cited in the 2007 Cochrane review divided 72 patients at 24 to 37 weeks’ gestational age into a group that received acupuncture plus standard care and a standard-care-only control group.^3,5 Treatment sessions occurred one or 2 times per week until delivery or recovery. The acupuncture group reported decreased pain (60% vs 14% for controls; P<.01; NNT = 2.2) and improved function (43% vs 9% for controls; P<.001; NNT = 2.9). There was also a difference in analgesic use: 0% for the acupuncture group vs 14% for controls; P<.05; NNT = 7.1.

A 2009 RCT divided 159 patients at 25 to 38 weeks’ gestational age into 3 groups: auricular acupuncture at specific points for one week, sham auricular acupuncture at nonspecific points for one week, and controls. At the end of Week 1, 80% of the acupuncture group had a clinically significant reduction in pain compared with 56% in the sham acupuncture group and 36% in the control group (P = .001 acupuncture vs sham, NNT = 4.2; P<.0001 acupuncture vs controls, NNT = 2.3).⁶

Osteopathic manipulative therapy (OMT) decreases disability, but not pain
A 2010 RCT divided 144 third trimester patients into 3 groups that received usual obstetric care, sham ultrasound therapy plus usual obstetric care, or OMT.⁷ Pain remained similar among the 3 groups throughout the study. Using the 24-point Roland-Morris Disability Questionnaire, OMT decreased disability by 0.72 points (95% CI, 0.31-1.14; P<.001) compared with 0.35 points in the usual obstetric-care-only group (95% CI, –0.06 to 0.76; P = .09). Ultrasound had no effect.

Corticosteroid injection reduces pain in a small trial
A small RCT of injection with the corticosteroid triamcinolone at the sacrospinous ligament insertion in 36 women with low-back pain showed significant reduction in pain in 17 of 18 women in the triamcinolone group compared with 9 of 18 women in the control group (P<.01; NNT = 2).⁸

Evidence lacking on maternity support garments
A poor-quality systematic review of 10 studies (N = 1909) of maternity support garments found insufficient evidence because of the heterogeneity of the trials.⁹

Recommendations

The American College of Obstetricians and Gynecologists suggests the following measures to prevent and treat low-back pain in pregnancy:¹⁰

• wear low-heeled (not flat) shoes with good arch support
• get help when lifting heavy objects
• place one foot on a stool or box when standing for long periods
• place a board between the mattress and box spring if the bed is too soft
• squat down, bend knees, and keep back straight when lifting
• sit in chairs with good back support or use a small pillow to provide support
• sleep on side with pillows between knees for support
• apply heat, cold, or massage to the painful area.

Evidence summary

Even though clinical research is lacking, acetaminophen is widely used to relieve low-back pain with no documented teratogenic effect (US Food and Drug Administration [FDA] category B). Nonsteroidal anti-inflammatory drugs are classified as FDA category D in the third trimester because of their documented association with oligohydramnios, premature closure of the ductus arteriosus, nephrotoxicity, and periventricular hemorrhage in the fetus.¹ Opioids are category C and a poor choice to treat low-back pain in pregnancy.²

Physical therapy and water aerobics relieve pain, reduce sick days
A 2007 Cochrane review of interventions for treating back pain in pregnancy analyzed 8 studies with a total of 1305 patients that examined the effects of adding physical therapy and acupuncture to usual care.³ In one RCT, 407 patients with and without pain received 5 30-minute individualized physical therapy exercise sessions, 2 45-minute group physical therapy classes, or standard care.^3,4 Low-back pain decreased with group physical therapy (P<.05; number needed to treat [NNT] = 3.2) and individual therapy (NNT = 2.1). Patients who received individual therapy had a 12% decrease in sick days.

A prospective trial of 258 patients, half of whom did water aerobics and half physical therapy, showed comparable results for the 2 interventions (NNT = 11.4 for decreased sick days; odds ratio = 0.38, 95% confidence interval [CI], 0.16-0.88).³

Acupuncture reduces pain and analgesic use
A prospective, randomized open study cited in the 2007 Cochrane review divided 72 patients at 24 to 37 weeks’ gestational age into a group that received acupuncture plus standard care and a standard-care-only control group.^3,5 Treatment sessions occurred one or 2 times per week until delivery or recovery. The acupuncture group reported decreased pain (60% vs 14% for controls; P<.01; NNT = 2.2) and improved function (43% vs 9% for controls; P<.001; NNT = 2.9). There was also a difference in analgesic use: 0% for the acupuncture group vs 14% for controls; P<.05; NNT = 7.1.

A 2009 RCT divided 159 patients at 25 to 38 weeks’ gestational age into 3 groups: auricular acupuncture at specific points for one week, sham auricular acupuncture at nonspecific points for one week, and controls. At the end of Week 1, 80% of the acupuncture group had a clinically significant reduction in pain compared with 56% in the sham acupuncture group and 36% in the control group (P = .001 acupuncture vs sham, NNT = 4.2; P<.0001 acupuncture vs controls, NNT = 2.3).⁶

Osteopathic manipulative therapy (OMT) decreases disability, but not pain
A 2010 RCT divided 144 third trimester patients into 3 groups that received usual obstetric care, sham ultrasound therapy plus usual obstetric care, or OMT.⁷ Pain remained similar among the 3 groups throughout the study. Using the 24-point Roland-Morris Disability Questionnaire, OMT decreased disability by 0.72 points (95% CI, 0.31-1.14; P<.001) compared with 0.35 points in the usual obstetric-care-only group (95% CI, –0.06 to 0.76; P = .09). Ultrasound had no effect.

Corticosteroid injection reduces pain in a small trial
A small RCT of injection with the corticosteroid triamcinolone at the sacrospinous ligament insertion in 36 women with low-back pain showed significant reduction in pain in 17 of 18 women in the triamcinolone group compared with 9 of 18 women in the control group (P<.01; NNT = 2).⁸

Evidence lacking on maternity support garments
A poor-quality systematic review of 10 studies (N = 1909) of maternity support garments found insufficient evidence because of the heterogeneity of the trials.⁹

Recommendations

The American College of Obstetricians and Gynecologists suggests the following measures to prevent and treat low-back pain in pregnancy:¹⁰

• wear low-heeled (not flat) shoes with good arch support
• get help when lifting heavy objects
• place one foot on a stool or box when standing for long periods
• place a board between the mattress and box spring if the bed is too soft
• squat down, bend knees, and keep back straight when lifting
• sit in chairs with good back support or use a small pillow to provide support
• sleep on side with pillows between knees for support
• apply heat, cold, or massage to the painful area.

Evidence summary

A meta-analysis of 8 randomized controlled trials (RCTs) with a total of 2000 patients found that weight loss through diet reduced BP in hypertensive patients.¹ Investigators recruited adult outpatients, 45 to 66 years of age, with primary hypertension (systolic BP, 128-178 mm Hg, diastolic BP, 72-107 mm Hg) and randomized them to dietary advice or usual care for 6 to 12 months.

Dietary advice resulted in greater weight loss over 6 to 12 months of follow-up (weighted mean difference [WMD], –4.0 kg; 95% confidence interval [CI], –4.8 to –3.2 kg) and greater BP reduction (WMD for systolic BP, –4.5 mm Hg; 95% CI, –7.2 to –1.8 mm Hg; WMD for diastolic BP, –3.2 mm Hg; 95% CI, –4.8 to –1.5 mm Hg).

Investigators didn’t report how long patients maintained the weight loss. Although 3 RCTs included encouragement to exercise, this meta-analysis didn’t evaluate benefits of combining these interventions.

The effects of exercise are less clear
A meta-analysis of 24 RCTs examined exercise and weight loss in adult outpatients with a mean age of 51.6 years; baseline body mass index (BMI), 25.9 kg/m²; resting systolic BP, 127 mm Hg; and resting diastolic BP, 77.7 mm Hg.² On average, participants walked for 38.3 minutes, 4.4 days per week, for 34.9 weeks at a relative intensity of 70.1% of predicted maximum heart rate (in 6 studies) or 56.3% oxygen consumption intensity (VO₂) (in 14 studies).

Walking significantly reduced body weight (WMD, –0.95 kg; P<.001) and BMI (WMD, –0.28 kg/m²; P = .015), leading to a significant reduction in diastolic BP (WMD, –1.54 mm Hg; P = .026) but not systolic BP (WMD, –1.06 mm Hg; P = .316). The authors didn’t report whether participants maintained the weight loss after the interventions.

In a meta-analysis of 8 RCTs and 18 observational studies, adult outpatients described as generally normotensive and overweight (mean age 49 years) wore pedometers to encourage weight loss with the goal of decreasing BP.³

Pedometer use for 3 to 104 weeks increased physical activity (for RCTs, a 2491-steps-per-day increase; 95% CI, 1098-3885 steps per day; for observational studies, a 2183-steps-per-day increase; 95% CI, 1571-2796 steps per day) and decreased BMI by 0.38 kg/m² (95% CI, 0.05-0.72 kg/m²). For an 80-kg, 170-cm tall person with a BMI of 27.7 kg/m², reducing BMI by 0.4 units translates to a 1.2-kg weight loss. This weight loss reduced systolic BP by 3.8 mm Hg (95% CI, 1.7-5.9 mm Hg), but not diastolic BP (–0.3 mm Hg; 95% CI, 0.02 to –0.46). Using a 10,000-steps-per-day goal (P = .001) and a step diary (P<.001) further increased walking.

Recommendations

The National Heart, Lung, and Blood Institute’s Joint National Committee says that healthy lifestyles are critical to preventing hypertension and reducing BP in people who are already hypertensive.⁴ Specifically, the Committee recommends weight reduction in overweight or obese individuals by increasing physical activity and using the Dietary Approaches to Stop Hypertension (DASH) eating plan. Combining 2 or more lifestyle modifications may enhance results.

The Committee also notes that a positive, empathetic relationship with a clinician is crucial in building trust and enhancing motivation to make lifestyle changes. It recommends setting mutual goals, ensuring adequate patient education, using frequent feedback, and involving all members of the health care team.

Evidence summary

A meta-analysis of 8 randomized controlled trials (RCTs) with a total of 2000 patients found that weight loss through diet reduced BP in hypertensive patients.¹ Investigators recruited adult outpatients, 45 to 66 years of age, with primary hypertension (systolic BP, 128-178 mm Hg, diastolic BP, 72-107 mm Hg) and randomized them to dietary advice or usual care for 6 to 12 months.

Dietary advice resulted in greater weight loss over 6 to 12 months of follow-up (weighted mean difference [WMD], –4.0 kg; 95% confidence interval [CI], –4.8 to –3.2 kg) and greater BP reduction (WMD for systolic BP, –4.5 mm Hg; 95% CI, –7.2 to –1.8 mm Hg; WMD for diastolic BP, –3.2 mm Hg; 95% CI, –4.8 to –1.5 mm Hg).

Investigators didn’t report how long patients maintained the weight loss. Although 3 RCTs included encouragement to exercise, this meta-analysis didn’t evaluate benefits of combining these interventions.

The effects of exercise are less clear
A meta-analysis of 24 RCTs examined exercise and weight loss in adult outpatients with a mean age of 51.6 years; baseline body mass index (BMI), 25.9 kg/m²; resting systolic BP, 127 mm Hg; and resting diastolic BP, 77.7 mm Hg.² On average, participants walked for 38.3 minutes, 4.4 days per week, for 34.9 weeks at a relative intensity of 70.1% of predicted maximum heart rate (in 6 studies) or 56.3% oxygen consumption intensity (VO₂) (in 14 studies).

Walking significantly reduced body weight (WMD, –0.95 kg; P<.001) and BMI (WMD, –0.28 kg/m²; P = .015), leading to a significant reduction in diastolic BP (WMD, –1.54 mm Hg; P = .026) but not systolic BP (WMD, –1.06 mm Hg; P = .316). The authors didn’t report whether participants maintained the weight loss after the interventions.

In a meta-analysis of 8 RCTs and 18 observational studies, adult outpatients described as generally normotensive and overweight (mean age 49 years) wore pedometers to encourage weight loss with the goal of decreasing BP.³

Pedometer use for 3 to 104 weeks increased physical activity (for RCTs, a 2491-steps-per-day increase; 95% CI, 1098-3885 steps per day; for observational studies, a 2183-steps-per-day increase; 95% CI, 1571-2796 steps per day) and decreased BMI by 0.38 kg/m² (95% CI, 0.05-0.72 kg/m²). For an 80-kg, 170-cm tall person with a BMI of 27.7 kg/m², reducing BMI by 0.4 units translates to a 1.2-kg weight loss. This weight loss reduced systolic BP by 3.8 mm Hg (95% CI, 1.7-5.9 mm Hg), but not diastolic BP (–0.3 mm Hg; 95% CI, 0.02 to –0.46). Using a 10,000-steps-per-day goal (P = .001) and a step diary (P<.001) further increased walking.

Recommendations

The National Heart, Lung, and Blood Institute’s Joint National Committee says that healthy lifestyles are critical to preventing hypertension and reducing BP in people who are already hypertensive.⁴ Specifically, the Committee recommends weight reduction in overweight or obese individuals by increasing physical activity and using the Dietary Approaches to Stop Hypertension (DASH) eating plan. Combining 2 or more lifestyle modifications may enhance results.

The Committee also notes that a positive, empathetic relationship with a clinician is crucial in building trust and enhancing motivation to make lifestyle changes. It recommends setting mutual goals, ensuring adequate patient education, using frequent feedback, and involving all members of the health care team.

Evidence summary

A meta-analysis of 8 randomized controlled trials (RCTs) with a total of 2000 patients found that weight loss through diet reduced BP in hypertensive patients.¹ Investigators recruited adult outpatients, 45 to 66 years of age, with primary hypertension (systolic BP, 128-178 mm Hg, diastolic BP, 72-107 mm Hg) and randomized them to dietary advice or usual care for 6 to 12 months.

Dietary advice resulted in greater weight loss over 6 to 12 months of follow-up (weighted mean difference [WMD], –4.0 kg; 95% confidence interval [CI], –4.8 to –3.2 kg) and greater BP reduction (WMD for systolic BP, –4.5 mm Hg; 95% CI, –7.2 to –1.8 mm Hg; WMD for diastolic BP, –3.2 mm Hg; 95% CI, –4.8 to –1.5 mm Hg).

Investigators didn’t report how long patients maintained the weight loss. Although 3 RCTs included encouragement to exercise, this meta-analysis didn’t evaluate benefits of combining these interventions.

The effects of exercise are less clear
A meta-analysis of 24 RCTs examined exercise and weight loss in adult outpatients with a mean age of 51.6 years; baseline body mass index (BMI), 25.9 kg/m²; resting systolic BP, 127 mm Hg; and resting diastolic BP, 77.7 mm Hg.² On average, participants walked for 38.3 minutes, 4.4 days per week, for 34.9 weeks at a relative intensity of 70.1% of predicted maximum heart rate (in 6 studies) or 56.3% oxygen consumption intensity (VO₂) (in 14 studies).

Walking significantly reduced body weight (WMD, –0.95 kg; P<.001) and BMI (WMD, –0.28 kg/m²; P = .015), leading to a significant reduction in diastolic BP (WMD, –1.54 mm Hg; P = .026) but not systolic BP (WMD, –1.06 mm Hg; P = .316). The authors didn’t report whether participants maintained the weight loss after the interventions.

In a meta-analysis of 8 RCTs and 18 observational studies, adult outpatients described as generally normotensive and overweight (mean age 49 years) wore pedometers to encourage weight loss with the goal of decreasing BP.³

Pedometer use for 3 to 104 weeks increased physical activity (for RCTs, a 2491-steps-per-day increase; 95% CI, 1098-3885 steps per day; for observational studies, a 2183-steps-per-day increase; 95% CI, 1571-2796 steps per day) and decreased BMI by 0.38 kg/m² (95% CI, 0.05-0.72 kg/m²). For an 80-kg, 170-cm tall person with a BMI of 27.7 kg/m², reducing BMI by 0.4 units translates to a 1.2-kg weight loss. This weight loss reduced systolic BP by 3.8 mm Hg (95% CI, 1.7-5.9 mm Hg), but not diastolic BP (–0.3 mm Hg; 95% CI, 0.02 to –0.46). Using a 10,000-steps-per-day goal (P = .001) and a step diary (P<.001) further increased walking.

Recommendations

The National Heart, Lung, and Blood Institute’s Joint National Committee says that healthy lifestyles are critical to preventing hypertension and reducing BP in people who are already hypertensive.⁴ Specifically, the Committee recommends weight reduction in overweight or obese individuals by increasing physical activity and using the Dietary Approaches to Stop Hypertension (DASH) eating plan. Combining 2 or more lifestyle modifications may enhance results.

The Committee also notes that a positive, empathetic relationship with a clinician is crucial in building trust and enhancing motivation to make lifestyle changes. It recommends setting mutual goals, ensuring adequate patient education, using frequent feedback, and involving all members of the health care team.

Evidence summary

A population-based cohort study centered in Olmsted County, Minn (N=50,080 births between 1966 and 1995) established a baseline rate of VTE among pregnant patients (105 total events; 0.2% incidence), and found an increased relative risk of VTE among pregnant and postpartum patients (RR=4.29; 95% confidence interval [CI], 3.49-5.22; P<.001) compared with nonpregnant patients. The incidence of VTE was 199.7 per 100,000 woman-years. The postpartum annual incidence of VTE was 5 times higher than antepartum (511.2 vs 95.8 per 100,000).¹

Thrombophilia testing typically isn’t useful
There is no evidence of improved outcomes from screening pregnant women with prior VTEs for some of the more common hypercoagulable conditions, including factor V Leiden, prothrombin G20210A mutation, protein C and S deficiency, and antiphospholipid syndrome. A recent Clinical Inquiry addressed this question for general medical patients with idiopathic deep venous thrombosis and found no quality evidence to support a thrombophilia work-up in most patients.² A subsequent review, which addressed pregnant patients specifically, made the same recommendation, that is, no quality evidence supports a thrombophilia work-up in patients at risk for VTE.³

How effective is prophylactic anticoagulation?
A meta-analysis in the American College of Chest Physicians (ACCP) Evidence-Based Clinical Practice Guidelines reviewing data from 1953 orthopedic and medical patients who were mostly postoperative (and not including pregnant women) found that prophylactic anticoagulation with LMWH for patients at risk for VTE produced a relative risk for recurrent VTE of 0.36 (95% CI, 0.20-0.67).⁴

In a more recent prospective cohort study, prophylactic LMWH was given to 177 of 286 (62% treated) patients according to risk-based scoring for recurrent VTE.⁵ The treatment protocol called for anticoagulation antepartum, postpartum, or both, depending on risk score (the higher the risk, the longer the period of thromboprophylaxis). Patients with previous pregnancy-associated VTE received both antepartum and postpartum anticoagulation. The study found recurrent VTE rates of 0.35% (95% CI, 0-1.03) antepartum and 0.7% (95% CI, 0-1.67) postpartum among treated patients.⁵

Data from an earlier report summarized the expected VTE rate in patients not exposed to anticoagulation prophylaxis. This prospective cohort study evaluated 125 pregnant women with a history of prior VTE who had anticoagulation withheld and determined the rate of recurrent antepartum and postpartum VTE. Three women had an antepartum VTE (2.4%; 95% CI, 0.2-6.9). Three additional women developed postpartum VTE, for a total of 6 VTEs (4.8%, no CI reported).⁶

LMWH is beneficial, but dosing can be tricky
Patients with a history of pregnancy-associated VTE—whether or not they have known thrombophilia—do benefit from routine ante- and postpartum thromboprophylaxis, per expert opinion in practice guidelines.^4,7,8 LMWH is the preferred agent because of its safety during pregnancy and ease of dosing.

Precise dosing is nonetheless difficult to determine because clinical studies in pregnant patients are lacking and renal clearance of LMWH increases during pregnancy. Most authors recommend doses between the prophylactic and therapeutic ranges.⁴ Subcutaneous enoxaparin, for example, can be given at 40 mg every 24 hours (more aggressive, thus higher-risk, dosing is as much as 1 mg/kg every 12 hours); dalteparin can be administered at 5000 units every 24 hours up to as much as 100 units/kg every 12 hours.⁹

Recommendations

The American College of Obstetricians and Gynecologists (ACOG) 2011 updated Practice Bulletin recommends thrombophilia testing for pregnant patients previously diagnosed with a pregnancy-associated VTE, although they acknowledge the lack of quality evidence to support this recommendation. ACOG also recommends ante- and postpartum thromboprophylaxis for such patients.⁹

The ACCP expert review recommends that all pregnant women diagnosed with VTE during a previous pregnancy wear graduated elastic compression stockings throughout pregnancy and for at least 6 weeks postpartum.⁷

The ACCP also recommends LMWH for all pregnant patients with a prior VTE. Additionally, the ACCP says that a thrombophilia work-up, while not routinely recommended, might be appropriate—contingent on additional risk assessment.⁷

Evidence summary

A population-based cohort study centered in Olmsted County, Minn (N=50,080 births between 1966 and 1995) established a baseline rate of VTE among pregnant patients (105 total events; 0.2% incidence), and found an increased relative risk of VTE among pregnant and postpartum patients (RR=4.29; 95% confidence interval [CI], 3.49-5.22; P<.001) compared with nonpregnant patients. The incidence of VTE was 199.7 per 100,000 woman-years. The postpartum annual incidence of VTE was 5 times higher than antepartum (511.2 vs 95.8 per 100,000).¹

Thrombophilia testing typically isn’t useful
There is no evidence of improved outcomes from screening pregnant women with prior VTEs for some of the more common hypercoagulable conditions, including factor V Leiden, prothrombin G20210A mutation, protein C and S deficiency, and antiphospholipid syndrome. A recent Clinical Inquiry addressed this question for general medical patients with idiopathic deep venous thrombosis and found no quality evidence to support a thrombophilia work-up in most patients.² A subsequent review, which addressed pregnant patients specifically, made the same recommendation, that is, no quality evidence supports a thrombophilia work-up in patients at risk for VTE.³

How effective is prophylactic anticoagulation?
A meta-analysis in the American College of Chest Physicians (ACCP) Evidence-Based Clinical Practice Guidelines reviewing data from 1953 orthopedic and medical patients who were mostly postoperative (and not including pregnant women) found that prophylactic anticoagulation with LMWH for patients at risk for VTE produced a relative risk for recurrent VTE of 0.36 (95% CI, 0.20-0.67).⁴

In a more recent prospective cohort study, prophylactic LMWH was given to 177 of 286 (62% treated) patients according to risk-based scoring for recurrent VTE.⁵ The treatment protocol called for anticoagulation antepartum, postpartum, or both, depending on risk score (the higher the risk, the longer the period of thromboprophylaxis). Patients with previous pregnancy-associated VTE received both antepartum and postpartum anticoagulation. The study found recurrent VTE rates of 0.35% (95% CI, 0-1.03) antepartum and 0.7% (95% CI, 0-1.67) postpartum among treated patients.⁵

Data from an earlier report summarized the expected VTE rate in patients not exposed to anticoagulation prophylaxis. This prospective cohort study evaluated 125 pregnant women with a history of prior VTE who had anticoagulation withheld and determined the rate of recurrent antepartum and postpartum VTE. Three women had an antepartum VTE (2.4%; 95% CI, 0.2-6.9). Three additional women developed postpartum VTE, for a total of 6 VTEs (4.8%, no CI reported).⁶

LMWH is beneficial, but dosing can be tricky
Patients with a history of pregnancy-associated VTE—whether or not they have known thrombophilia—do benefit from routine ante- and postpartum thromboprophylaxis, per expert opinion in practice guidelines.^4,7,8 LMWH is the preferred agent because of its safety during pregnancy and ease of dosing.

Precise dosing is nonetheless difficult to determine because clinical studies in pregnant patients are lacking and renal clearance of LMWH increases during pregnancy. Most authors recommend doses between the prophylactic and therapeutic ranges.⁴ Subcutaneous enoxaparin, for example, can be given at 40 mg every 24 hours (more aggressive, thus higher-risk, dosing is as much as 1 mg/kg every 12 hours); dalteparin can be administered at 5000 units every 24 hours up to as much as 100 units/kg every 12 hours.⁹

Recommendations

The American College of Obstetricians and Gynecologists (ACOG) 2011 updated Practice Bulletin recommends thrombophilia testing for pregnant patients previously diagnosed with a pregnancy-associated VTE, although they acknowledge the lack of quality evidence to support this recommendation. ACOG also recommends ante- and postpartum thromboprophylaxis for such patients.⁹

The ACCP expert review recommends that all pregnant women diagnosed with VTE during a previous pregnancy wear graduated elastic compression stockings throughout pregnancy and for at least 6 weeks postpartum.⁷

The ACCP also recommends LMWH for all pregnant patients with a prior VTE. Additionally, the ACCP says that a thrombophilia work-up, while not routinely recommended, might be appropriate—contingent on additional risk assessment.⁷

Evidence summary

A population-based cohort study centered in Olmsted County, Minn (N=50,080 births between 1966 and 1995) established a baseline rate of VTE among pregnant patients (105 total events; 0.2% incidence), and found an increased relative risk of VTE among pregnant and postpartum patients (RR=4.29; 95% confidence interval [CI], 3.49-5.22; P<.001) compared with nonpregnant patients. The incidence of VTE was 199.7 per 100,000 woman-years. The postpartum annual incidence of VTE was 5 times higher than antepartum (511.2 vs 95.8 per 100,000).¹

Thrombophilia testing typically isn’t useful
There is no evidence of improved outcomes from screening pregnant women with prior VTEs for some of the more common hypercoagulable conditions, including factor V Leiden, prothrombin G20210A mutation, protein C and S deficiency, and antiphospholipid syndrome. A recent Clinical Inquiry addressed this question for general medical patients with idiopathic deep venous thrombosis and found no quality evidence to support a thrombophilia work-up in most patients.² A subsequent review, which addressed pregnant patients specifically, made the same recommendation, that is, no quality evidence supports a thrombophilia work-up in patients at risk for VTE.³

How effective is prophylactic anticoagulation?
A meta-analysis in the American College of Chest Physicians (ACCP) Evidence-Based Clinical Practice Guidelines reviewing data from 1953 orthopedic and medical patients who were mostly postoperative (and not including pregnant women) found that prophylactic anticoagulation with LMWH for patients at risk for VTE produced a relative risk for recurrent VTE of 0.36 (95% CI, 0.20-0.67).⁴

In a more recent prospective cohort study, prophylactic LMWH was given to 177 of 286 (62% treated) patients according to risk-based scoring for recurrent VTE.⁵ The treatment protocol called for anticoagulation antepartum, postpartum, or both, depending on risk score (the higher the risk, the longer the period of thromboprophylaxis). Patients with previous pregnancy-associated VTE received both antepartum and postpartum anticoagulation. The study found recurrent VTE rates of 0.35% (95% CI, 0-1.03) antepartum and 0.7% (95% CI, 0-1.67) postpartum among treated patients.⁵

Data from an earlier report summarized the expected VTE rate in patients not exposed to anticoagulation prophylaxis. This prospective cohort study evaluated 125 pregnant women with a history of prior VTE who had anticoagulation withheld and determined the rate of recurrent antepartum and postpartum VTE. Three women had an antepartum VTE (2.4%; 95% CI, 0.2-6.9). Three additional women developed postpartum VTE, for a total of 6 VTEs (4.8%, no CI reported).⁶

LMWH is beneficial, but dosing can be tricky
Patients with a history of pregnancy-associated VTE—whether or not they have known thrombophilia—do benefit from routine ante- and postpartum thromboprophylaxis, per expert opinion in practice guidelines.^4,7,8 LMWH is the preferred agent because of its safety during pregnancy and ease of dosing.

Precise dosing is nonetheless difficult to determine because clinical studies in pregnant patients are lacking and renal clearance of LMWH increases during pregnancy. Most authors recommend doses between the prophylactic and therapeutic ranges.⁴ Subcutaneous enoxaparin, for example, can be given at 40 mg every 24 hours (more aggressive, thus higher-risk, dosing is as much as 1 mg/kg every 12 hours); dalteparin can be administered at 5000 units every 24 hours up to as much as 100 units/kg every 12 hours.⁹

Recommendations

The American College of Obstetricians and Gynecologists (ACOG) 2011 updated Practice Bulletin recommends thrombophilia testing for pregnant patients previously diagnosed with a pregnancy-associated VTE, although they acknowledge the lack of quality evidence to support this recommendation. ACOG also recommends ante- and postpartum thromboprophylaxis for such patients.⁹

The ACCP expert review recommends that all pregnant women diagnosed with VTE during a previous pregnancy wear graduated elastic compression stockings throughout pregnancy and for at least 6 weeks postpartum.⁷

The ACCP also recommends LMWH for all pregnant patients with a prior VTE. Additionally, the ACCP says that a thrombophilia work-up, while not routinely recommended, might be appropriate—contingent on additional risk assessment.⁷

Evidence summary

SCD in athletes is a rare event; researchers estimate the annual incidence at 0.5 to 2 per 100,000 athletes per year.^1,2 The rarity of SCD, lack of registries recording cases, multiple causes, and varying demographics limit accurate estimation of its incidence.³ Experts also debate whether the ability of a single EKG to detect a potentially lethal arrhythmia outweighs the expense and potential harm of false-positive results. No randomized controlled trials have assessed the efficacy of preparticipation screening, with or without EKG, to reduce SCD. The major studies evaluating this issue are observational and retrospective.

An Italian study suggests lower SCD mortality with EKG screening
An Italian observational study of 33,735 athletes reported results of a 25-year program of mandated SCD screening (1979-2004).⁴ Researchers found that the incidence of sudden death was 89% lower at the end of the screening program (0.4/100,000 athletes/year), compared with the baseline (3.6/100,000 athletes/year).

The program used highly trained sports medicine physicians and supplemented the history and physical examination with universal EKG screening. Screening began at 12 to 14 years of age and was repeated regularly as long as the athlete was engaging in competition. The incidence of SCD in nonathletes (0.79/100,000 nonathletes/year) didn’t change over the 25 years of the study.

Limitations of the study include the lack of a concurrent control group of unscreened athletes and time-dependent bias, also known as immortal time bias. (Studies with time-dependent outcomes in which the test of interest, such as EKG screening, and the outcome analyzed, such as SCD, occur during the same period are susceptible to immortal time bias. Athletes who may have died suddenly during the prescreening period would never have made it to the first screening, so the group of athletes who made it alive to the first screening already represented a selected lower-risk population whose characteristics contributed to the lower mortality rates in the postscreening period.)

Other study limitations included baseline differences in male-to-female ratio (82% male) and age range (12-35 years). In addition, the study used a short measurement time (approximately 3 years) to establish a baseline SCD incidence, resulting in a higher incidence than the average in other studies.

An Israeli study suggests otherwise
An uncontrolled observational study done in Israel compared the number of SCDs before and after implementation of a mandatory nationwide screening program for all athletes.⁵ The screening protocol included a medical questionnaire, physical examination, resting EKG, and exercise stress test. Researchers estimated the SCD incidence by scrutinizing newspaper reports of sudden deaths in competitive athletes for 2 time periods: 12 years before (1985-1997) and 12 years after (1997-2009) the intervention.

They identified 24 presumed cardiac deaths, all in male athletes 12 to 44 years of age (mean 23.9 years). The incidence of SCD in the athletes before and after this protocol was 2.54 and 2.66/100,000 athletes/year, respectively (P not significant).

An advantage of this study is the longer period used to estimate SCD incidence (12 years compared with approximately 3 years). A disadvantage is that the researchers calculated the SCD incidence by relying only on media reports rather than on a death registry.

A US comparison study supports the Israeli findings
In the United States, researchers compared SCD rates in high school and college athletes in Minnesota with the rates reported in the Italian study discussed previously.¹ They collected mortality data from several sources, including a national death registry, over a similar time period (1985-2007), during which routine EKG screening of Minnesota athletes wasn’t mandated or recommended.

During the 23-year study period, SCD mortality rates in young athletes in Minnesota remained stable at 0.97/100,000 athletes/year (range 0.5-1.3). This rate didn’t differ significantly from the death rate reported at the end of the Italian study.⁴ The US authors concluded that their data didn’t support the hypothesis that routine EKG screening in young athletes results in lower SCD mortality.¹

Recommendations

All major groups acknowledge that EKG screening improves the sensitivity of the preparticipation physical exam. The European Society of Cardiology Study Group of Sport Cardiology and the International Olympic Committee advocate routine screening.^6,7

The American Heart Association (AHA) Council on Nutrition, Physical Activity, and Metabolism and the American College of Sports Medicine don’t recommend routine EKGs as part of the preparticipation evaluation of school-aged athletes.^8,9

The AHA recommends that a qualified examiner perform a full history and physical exam, which includes assessments of 12 key risk factors (TABLE), and advocates cardiovascular referral for patients who show positive findings.

TABLE
12 cardiovascular risk factors to watch for during preparticipation physicals for school-aged athletes

Evidence summary

SCD in athletes is a rare event; researchers estimate the annual incidence at 0.5 to 2 per 100,000 athletes per year.^1,2 The rarity of SCD, lack of registries recording cases, multiple causes, and varying demographics limit accurate estimation of its incidence.³ Experts also debate whether the ability of a single EKG to detect a potentially lethal arrhythmia outweighs the expense and potential harm of false-positive results. No randomized controlled trials have assessed the efficacy of preparticipation screening, with or without EKG, to reduce SCD. The major studies evaluating this issue are observational and retrospective.

An Italian study suggests lower SCD mortality with EKG screening
An Italian observational study of 33,735 athletes reported results of a 25-year program of mandated SCD screening (1979-2004).⁴ Researchers found that the incidence of sudden death was 89% lower at the end of the screening program (0.4/100,000 athletes/year), compared with the baseline (3.6/100,000 athletes/year).

The program used highly trained sports medicine physicians and supplemented the history and physical examination with universal EKG screening. Screening began at 12 to 14 years of age and was repeated regularly as long as the athlete was engaging in competition. The incidence of SCD in nonathletes (0.79/100,000 nonathletes/year) didn’t change over the 25 years of the study.

Limitations of the study include the lack of a concurrent control group of unscreened athletes and time-dependent bias, also known as immortal time bias. (Studies with time-dependent outcomes in which the test of interest, such as EKG screening, and the outcome analyzed, such as SCD, occur during the same period are susceptible to immortal time bias. Athletes who may have died suddenly during the prescreening period would never have made it to the first screening, so the group of athletes who made it alive to the first screening already represented a selected lower-risk population whose characteristics contributed to the lower mortality rates in the postscreening period.)

Other study limitations included baseline differences in male-to-female ratio (82% male) and age range (12-35 years). In addition, the study used a short measurement time (approximately 3 years) to establish a baseline SCD incidence, resulting in a higher incidence than the average in other studies.

An Israeli study suggests otherwise
An uncontrolled observational study done in Israel compared the number of SCDs before and after implementation of a mandatory nationwide screening program for all athletes.⁵ The screening protocol included a medical questionnaire, physical examination, resting EKG, and exercise stress test. Researchers estimated the SCD incidence by scrutinizing newspaper reports of sudden deaths in competitive athletes for 2 time periods: 12 years before (1985-1997) and 12 years after (1997-2009) the intervention.

They identified 24 presumed cardiac deaths, all in male athletes 12 to 44 years of age (mean 23.9 years). The incidence of SCD in the athletes before and after this protocol was 2.54 and 2.66/100,000 athletes/year, respectively (P not significant).

An advantage of this study is the longer period used to estimate SCD incidence (12 years compared with approximately 3 years). A disadvantage is that the researchers calculated the SCD incidence by relying only on media reports rather than on a death registry.

A US comparison study supports the Israeli findings
In the United States, researchers compared SCD rates in high school and college athletes in Minnesota with the rates reported in the Italian study discussed previously.¹ They collected mortality data from several sources, including a national death registry, over a similar time period (1985-2007), during which routine EKG screening of Minnesota athletes wasn’t mandated or recommended.

During the 23-year study period, SCD mortality rates in young athletes in Minnesota remained stable at 0.97/100,000 athletes/year (range 0.5-1.3). This rate didn’t differ significantly from the death rate reported at the end of the Italian study.⁴ The US authors concluded that their data didn’t support the hypothesis that routine EKG screening in young athletes results in lower SCD mortality.¹

Recommendations

All major groups acknowledge that EKG screening improves the sensitivity of the preparticipation physical exam. The European Society of Cardiology Study Group of Sport Cardiology and the International Olympic Committee advocate routine screening.^6,7

The American Heart Association (AHA) Council on Nutrition, Physical Activity, and Metabolism and the American College of Sports Medicine don’t recommend routine EKGs as part of the preparticipation evaluation of school-aged athletes.^8,9

The AHA recommends that a qualified examiner perform a full history and physical exam, which includes assessments of 12 key risk factors (TABLE), and advocates cardiovascular referral for patients who show positive findings.

TABLE
12 cardiovascular risk factors to watch for during preparticipation physicals for school-aged athletes

Evidence summary

SCD in athletes is a rare event; researchers estimate the annual incidence at 0.5 to 2 per 100,000 athletes per year.^1,2 The rarity of SCD, lack of registries recording cases, multiple causes, and varying demographics limit accurate estimation of its incidence.³ Experts also debate whether the ability of a single EKG to detect a potentially lethal arrhythmia outweighs the expense and potential harm of false-positive results. No randomized controlled trials have assessed the efficacy of preparticipation screening, with or without EKG, to reduce SCD. The major studies evaluating this issue are observational and retrospective.

An Italian study suggests lower SCD mortality with EKG screening
An Italian observational study of 33,735 athletes reported results of a 25-year program of mandated SCD screening (1979-2004).⁴ Researchers found that the incidence of sudden death was 89% lower at the end of the screening program (0.4/100,000 athletes/year), compared with the baseline (3.6/100,000 athletes/year).

The program used highly trained sports medicine physicians and supplemented the history and physical examination with universal EKG screening. Screening began at 12 to 14 years of age and was repeated regularly as long as the athlete was engaging in competition. The incidence of SCD in nonathletes (0.79/100,000 nonathletes/year) didn’t change over the 25 years of the study.

Limitations of the study include the lack of a concurrent control group of unscreened athletes and time-dependent bias, also known as immortal time bias. (Studies with time-dependent outcomes in which the test of interest, such as EKG screening, and the outcome analyzed, such as SCD, occur during the same period are susceptible to immortal time bias. Athletes who may have died suddenly during the prescreening period would never have made it to the first screening, so the group of athletes who made it alive to the first screening already represented a selected lower-risk population whose characteristics contributed to the lower mortality rates in the postscreening period.)

Other study limitations included baseline differences in male-to-female ratio (82% male) and age range (12-35 years). In addition, the study used a short measurement time (approximately 3 years) to establish a baseline SCD incidence, resulting in a higher incidence than the average in other studies.

An Israeli study suggests otherwise
An uncontrolled observational study done in Israel compared the number of SCDs before and after implementation of a mandatory nationwide screening program for all athletes.⁵ The screening protocol included a medical questionnaire, physical examination, resting EKG, and exercise stress test. Researchers estimated the SCD incidence by scrutinizing newspaper reports of sudden deaths in competitive athletes for 2 time periods: 12 years before (1985-1997) and 12 years after (1997-2009) the intervention.

They identified 24 presumed cardiac deaths, all in male athletes 12 to 44 years of age (mean 23.9 years). The incidence of SCD in the athletes before and after this protocol was 2.54 and 2.66/100,000 athletes/year, respectively (P not significant).

An advantage of this study is the longer period used to estimate SCD incidence (12 years compared with approximately 3 years). A disadvantage is that the researchers calculated the SCD incidence by relying only on media reports rather than on a death registry.

A US comparison study supports the Israeli findings
In the United States, researchers compared SCD rates in high school and college athletes in Minnesota with the rates reported in the Italian study discussed previously.¹ They collected mortality data from several sources, including a national death registry, over a similar time period (1985-2007), during which routine EKG screening of Minnesota athletes wasn’t mandated or recommended.

During the 23-year study period, SCD mortality rates in young athletes in Minnesota remained stable at 0.97/100,000 athletes/year (range 0.5-1.3). This rate didn’t differ significantly from the death rate reported at the end of the Italian study.⁴ The US authors concluded that their data didn’t support the hypothesis that routine EKG screening in young athletes results in lower SCD mortality.¹

Recommendations

All major groups acknowledge that EKG screening improves the sensitivity of the preparticipation physical exam. The European Society of Cardiology Study Group of Sport Cardiology and the International Olympic Committee advocate routine screening.^6,7

The American Heart Association (AHA) Council on Nutrition, Physical Activity, and Metabolism and the American College of Sports Medicine don’t recommend routine EKGs as part of the preparticipation evaluation of school-aged athletes.^8,9

The AHA recommends that a qualified examiner perform a full history and physical exam, which includes assessments of 12 key risk factors (TABLE), and advocates cardiovascular referral for patients who show positive findings.

TABLE
12 cardiovascular risk factors to watch for during preparticipation physicals for school-aged athletes

Evidence Summary

Two RCTs examined the effect of testosterone on postmenopausal women with HSDD. One trial randomized 272 women ages 40 to 70 years to a 300-mcg transdermal testosterone patch (TTP; 142 women) or placebo (130 women).¹ At 6 months, women using the TTP reported more sexually satisfying episodes (1.69 vs 0.59 episodes in 4 weeks; P=.0089) and a minimal increase in sexual desire scores (12.2 vs 4.56 on a 100-point sexual desire scale; P=.0007) compared with women using placebo.

A second trial randomized 814 postmenopausal women (mean age 54.2 years) to placebo (277 women), a 150-mcg TTP (267 women), or a 300-mcg TTP (270 women).² At 24 weeks, women taking 300 mcg (but not 150 mcg) of testosterone reported a greater number of satisfying sexual episodes than women taking placebo (2.1 vs 0.7; P<.0001). The 300-mcg TTP caused more unwanted hair growth than placebo (19.9% vs 10.5%; no P value given). The study didn’t continue long enough to assess cardiovascular risks.

Bupropion may improve sexual function in premenopausal women
Two RCTs found benefit from bupropion for premenopausal women with HSDD. In the first, investigators randomized 232 women 20 to 40 years of age to bupropion sustained release (SR) 150 mg daily or placebo. They assessed sexual function at 12 weeks with the Brief Index of Sexual Functioning for Women—a scale with scores ranging from -16 (poor functioning) to +75 (maximum functioning), with a mean value in normal women of 33.6.³ Women taking bupropion reported greater increases in scores than women taking placebo (15.8 to 33.9, vs 15.5 to 16.9; P=.001) and no serious adverse events.

A second RCT randomized 66 premenopausal women (mean age 36.1 years) to take either bupropion SR 150 mg daily, increased to 300 mg daily after one week, or placebo.⁴ Researchers measured sexual responsiveness (arousal, pleasure, and orgasm) using the Change in Sexual Functioning Questionnaire at baseline and on Days 28, 56, 84, and 112. Women taking bupropion had higher scores by Day 28 than women taking placebo and maintained the difference through Day 112 (P=.05). The authors indicated that the clinical significance of the change is unclear.

Sildenafil increases low sexual desire associated with antidepressants
A double-blind RCT enrolling 98 premenopausal women (mean age 36.7 years) with sexual dysfunction related to SSRIs found that sildenafil (50-100 mg) improved sexual functioning more than placebo using the 7-point Clinical Global Impression score (sildenafil: 1.9 points; 95% confidence interval [CI], 1.6-2.3; placebo: 1.1 points; 95% CI, 0.8-1.5; P=.001).⁵

The investigators didn’t specify whether the change was clinically significant. However, another RCT that studied 881 pre- and postmenopausal women with HSDD unassociated with SSRIs found no difference between sildenafil (10-100 mg) and placebo.⁶

Recommendations

The US Food and Drug Administration doesn’t recommend androgens for female sexual dysfunction.

The Endocrine Society says that bupropion may be used for HSDD (although it isn’t licensed for such use) and doesn’t recommend long-term use of testosterone because of inadequate safety studies.⁷

The North American Menopause Society recommends testosterone therapy for postmenopausal women with HSDD.⁸

Evidence Summary

Two RCTs examined the effect of testosterone on postmenopausal women with HSDD. One trial randomized 272 women ages 40 to 70 years to a 300-mcg transdermal testosterone patch (TTP; 142 women) or placebo (130 women).¹ At 6 months, women using the TTP reported more sexually satisfying episodes (1.69 vs 0.59 episodes in 4 weeks; P=.0089) and a minimal increase in sexual desire scores (12.2 vs 4.56 on a 100-point sexual desire scale; P=.0007) compared with women using placebo.

A second trial randomized 814 postmenopausal women (mean age 54.2 years) to placebo (277 women), a 150-mcg TTP (267 women), or a 300-mcg TTP (270 women).² At 24 weeks, women taking 300 mcg (but not 150 mcg) of testosterone reported a greater number of satisfying sexual episodes than women taking placebo (2.1 vs 0.7; P<.0001). The 300-mcg TTP caused more unwanted hair growth than placebo (19.9% vs 10.5%; no P value given). The study didn’t continue long enough to assess cardiovascular risks.

Bupropion may improve sexual function in premenopausal women
Two RCTs found benefit from bupropion for premenopausal women with HSDD. In the first, investigators randomized 232 women 20 to 40 years of age to bupropion sustained release (SR) 150 mg daily or placebo. They assessed sexual function at 12 weeks with the Brief Index of Sexual Functioning for Women—a scale with scores ranging from -16 (poor functioning) to +75 (maximum functioning), with a mean value in normal women of 33.6.³ Women taking bupropion reported greater increases in scores than women taking placebo (15.8 to 33.9, vs 15.5 to 16.9; P=.001) and no serious adverse events.

A second RCT randomized 66 premenopausal women (mean age 36.1 years) to take either bupropion SR 150 mg daily, increased to 300 mg daily after one week, or placebo.⁴ Researchers measured sexual responsiveness (arousal, pleasure, and orgasm) using the Change in Sexual Functioning Questionnaire at baseline and on Days 28, 56, 84, and 112. Women taking bupropion had higher scores by Day 28 than women taking placebo and maintained the difference through Day 112 (P=.05). The authors indicated that the clinical significance of the change is unclear.

Sildenafil increases low sexual desire associated with antidepressants
A double-blind RCT enrolling 98 premenopausal women (mean age 36.7 years) with sexual dysfunction related to SSRIs found that sildenafil (50-100 mg) improved sexual functioning more than placebo using the 7-point Clinical Global Impression score (sildenafil: 1.9 points; 95% confidence interval [CI], 1.6-2.3; placebo: 1.1 points; 95% CI, 0.8-1.5; P=.001).⁵

The investigators didn’t specify whether the change was clinically significant. However, another RCT that studied 881 pre- and postmenopausal women with HSDD unassociated with SSRIs found no difference between sildenafil (10-100 mg) and placebo.⁶

Recommendations

The US Food and Drug Administration doesn’t recommend androgens for female sexual dysfunction.

The Endocrine Society says that bupropion may be used for HSDD (although it isn’t licensed for such use) and doesn’t recommend long-term use of testosterone because of inadequate safety studies.⁷

The North American Menopause Society recommends testosterone therapy for postmenopausal women with HSDD.⁸

Evidence Summary

Two RCTs examined the effect of testosterone on postmenopausal women with HSDD. One trial randomized 272 women ages 40 to 70 years to a 300-mcg transdermal testosterone patch (TTP; 142 women) or placebo (130 women).¹ At 6 months, women using the TTP reported more sexually satisfying episodes (1.69 vs 0.59 episodes in 4 weeks; P=.0089) and a minimal increase in sexual desire scores (12.2 vs 4.56 on a 100-point sexual desire scale; P=.0007) compared with women using placebo.

A second trial randomized 814 postmenopausal women (mean age 54.2 years) to placebo (277 women), a 150-mcg TTP (267 women), or a 300-mcg TTP (270 women).² At 24 weeks, women taking 300 mcg (but not 150 mcg) of testosterone reported a greater number of satisfying sexual episodes than women taking placebo (2.1 vs 0.7; P<.0001). The 300-mcg TTP caused more unwanted hair growth than placebo (19.9% vs 10.5%; no P value given). The study didn’t continue long enough to assess cardiovascular risks.

Bupropion may improve sexual function in premenopausal women
Two RCTs found benefit from bupropion for premenopausal women with HSDD. In the first, investigators randomized 232 women 20 to 40 years of age to bupropion sustained release (SR) 150 mg daily or placebo. They assessed sexual function at 12 weeks with the Brief Index of Sexual Functioning for Women—a scale with scores ranging from -16 (poor functioning) to +75 (maximum functioning), with a mean value in normal women of 33.6.³ Women taking bupropion reported greater increases in scores than women taking placebo (15.8 to 33.9, vs 15.5 to 16.9; P=.001) and no serious adverse events.

A second RCT randomized 66 premenopausal women (mean age 36.1 years) to take either bupropion SR 150 mg daily, increased to 300 mg daily after one week, or placebo.⁴ Researchers measured sexual responsiveness (arousal, pleasure, and orgasm) using the Change in Sexual Functioning Questionnaire at baseline and on Days 28, 56, 84, and 112. Women taking bupropion had higher scores by Day 28 than women taking placebo and maintained the difference through Day 112 (P=.05). The authors indicated that the clinical significance of the change is unclear.

Sildenafil increases low sexual desire associated with antidepressants
A double-blind RCT enrolling 98 premenopausal women (mean age 36.7 years) with sexual dysfunction related to SSRIs found that sildenafil (50-100 mg) improved sexual functioning more than placebo using the 7-point Clinical Global Impression score (sildenafil: 1.9 points; 95% confidence interval [CI], 1.6-2.3; placebo: 1.1 points; 95% CI, 0.8-1.5; P=.001).⁵

The investigators didn’t specify whether the change was clinically significant. However, another RCT that studied 881 pre- and postmenopausal women with HSDD unassociated with SSRIs found no difference between sildenafil (10-100 mg) and placebo.⁶

Recommendations

The US Food and Drug Administration doesn’t recommend androgens for female sexual dysfunction.

The Endocrine Society says that bupropion may be used for HSDD (although it isn’t licensed for such use) and doesn’t recommend long-term use of testosterone because of inadequate safety studies.⁷

The North American Menopause Society recommends testosterone therapy for postmenopausal women with HSDD.⁸

Evidence summary

Multiple head-to-head RCTs of oral treatments for toenail onychomycosis demonstrate that terbinafine 250 mg per day for at least 12 weeks is superior to pulse itraconazole, weekly fluconazole, or pulse terbinafine (TABLE).^1-5 In these studies the number needed to treat (NNT) favoring daily terbinafine ranged from 2 to 12.

Recurrence is less common in patients who take terbinafine daily. In a prospective cohort study of 73 patients (21-81 years of age) followed for 5 years after clinical and mycological cure, onychomycosis recurred in 7 of 59 (12%) patients treated with daily terbinafine and 5 of 14 (36%) treated with pulse itraconazole (P=.046; NNT=4.2).⁶

TABLE
Oral treatments for onychomycosis: RCTs reveal how they compare

No interactions in patients with diabetes, the elderly
A prospective open study of 89 diabetic patients with longstanding toenail onychomycosis, treated with terbinafine 250 mg/d for 12 weeks (mean age 56 years, 42% with insulin-dependent diabetes mellitus), showed a clinical cure rate of 57% at 48 weeks. No hypoglycemic episodes were reported during the treatment phase, and no changes in liver enzymes or creatinine levels occurred.⁷

An open-label trial of 75 patients older than 65 years compared terbinafine alone (34 patients) with terbinafine and nail debridement (41 patients). Subjects took 250 mg terbinafine per day for 12 weeks; 73 (97.3%) took concomitant medications, including antihypertensives (64%), diabetic medications (25%), and lipid-lowering agents (47%).⁸ No clinically significant drug interactions or elevations in liver function tests occurred. Three patients (4%) withdrew from the study because of drug-related adverse effects (nausea, headache, or flank pain).

Recommendations

No major American medical organization has published guidelines addressing treatment of onychomycosis. The British Association of Dermatologists’ guidelines (2003) recommend terbinafine as first-line treatment for fungal toenail infections, with itraconazole as the next best alternative.9

Evidence summary

Multiple head-to-head RCTs of oral treatments for toenail onychomycosis demonstrate that terbinafine 250 mg per day for at least 12 weeks is superior to pulse itraconazole, weekly fluconazole, or pulse terbinafine (TABLE).^1-5 In these studies the number needed to treat (NNT) favoring daily terbinafine ranged from 2 to 12.

Recurrence is less common in patients who take terbinafine daily. In a prospective cohort study of 73 patients (21-81 years of age) followed for 5 years after clinical and mycological cure, onychomycosis recurred in 7 of 59 (12%) patients treated with daily terbinafine and 5 of 14 (36%) treated with pulse itraconazole (P=.046; NNT=4.2).⁶

TABLE
Oral treatments for onychomycosis: RCTs reveal how they compare

No interactions in patients with diabetes, the elderly
A prospective open study of 89 diabetic patients with longstanding toenail onychomycosis, treated with terbinafine 250 mg/d for 12 weeks (mean age 56 years, 42% with insulin-dependent diabetes mellitus), showed a clinical cure rate of 57% at 48 weeks. No hypoglycemic episodes were reported during the treatment phase, and no changes in liver enzymes or creatinine levels occurred.⁷

An open-label trial of 75 patients older than 65 years compared terbinafine alone (34 patients) with terbinafine and nail debridement (41 patients). Subjects took 250 mg terbinafine per day for 12 weeks; 73 (97.3%) took concomitant medications, including antihypertensives (64%), diabetic medications (25%), and lipid-lowering agents (47%).⁸ No clinically significant drug interactions or elevations in liver function tests occurred. Three patients (4%) withdrew from the study because of drug-related adverse effects (nausea, headache, or flank pain).

Recommendations

No major American medical organization has published guidelines addressing treatment of onychomycosis. The British Association of Dermatologists’ guidelines (2003) recommend terbinafine as first-line treatment for fungal toenail infections, with itraconazole as the next best alternative.9

Evidence summary

Multiple head-to-head RCTs of oral treatments for toenail onychomycosis demonstrate that terbinafine 250 mg per day for at least 12 weeks is superior to pulse itraconazole, weekly fluconazole, or pulse terbinafine (TABLE).^1-5 In these studies the number needed to treat (NNT) favoring daily terbinafine ranged from 2 to 12.

Recurrence is less common in patients who take terbinafine daily. In a prospective cohort study of 73 patients (21-81 years of age) followed for 5 years after clinical and mycological cure, onychomycosis recurred in 7 of 59 (12%) patients treated with daily terbinafine and 5 of 14 (36%) treated with pulse itraconazole (P=.046; NNT=4.2).⁶

TABLE
Oral treatments for onychomycosis: RCTs reveal how they compare

No interactions in patients with diabetes, the elderly
A prospective open study of 89 diabetic patients with longstanding toenail onychomycosis, treated with terbinafine 250 mg/d for 12 weeks (mean age 56 years, 42% with insulin-dependent diabetes mellitus), showed a clinical cure rate of 57% at 48 weeks. No hypoglycemic episodes were reported during the treatment phase, and no changes in liver enzymes or creatinine levels occurred.⁷

An open-label trial of 75 patients older than 65 years compared terbinafine alone (34 patients) with terbinafine and nail debridement (41 patients). Subjects took 250 mg terbinafine per day for 12 weeks; 73 (97.3%) took concomitant medications, including antihypertensives (64%), diabetic medications (25%), and lipid-lowering agents (47%).⁸ No clinically significant drug interactions or elevations in liver function tests occurred. Three patients (4%) withdrew from the study because of drug-related adverse effects (nausea, headache, or flank pain).

Recommendations

No major American medical organization has published guidelines addressing treatment of onychomycosis. The British Association of Dermatologists’ guidelines (2003) recommend terbinafine as first-line treatment for fungal toenail infections, with itraconazole as the next best alternative.9

Evidence summary

A 2012 Cochrane review and meta-analysis of 9 RCTs found that 1261 patients who used SMBG showed a small but statistically significant decrease in HbA1c at 6 months compared with 1063 controls. In 2 other RCTs, patients using SMBG showed a nonsignificant decrease in HbA1c compared with control subjects at 12 months (TABLE 1).¹

TABLE 1
Difference in HbA1c by duration of self-monitoring*

Another meta-analysis reported similar findings. The study grouped 9 RCTs based on the duration of SMBG and examined the change in HbA1c from baseline. In 5 of the trials, SMBG for 6 months was associated with a small decrease in HbA1c, but in the other 4, SMBG for longer than one year didn’t significantly change HbA1c levels.²

Baseline values make a difference
A meta-analysis of 9 RCTs demonstrated that SMBG was marginally superior to non-SMBG for reducing HbA1c when the baseline value was >8%. SMBG didn’t lower HbA1c in patients with a baseline HbA1c <8%. The greatest change in HbA1c occurred in patients with baseline values >10% (TABLE 2).³

In another meta-analysis, 12 of 15 RCTs found SMBG to be better than non-SMBG at reducing HbA1c when the baseline was >8%.⁴

TABLE 2
Patients benefit from self-monitoring when baseline HbA1c is >8%

Limitations of studies
Limitations of the studies (TABLES 1 AND 2) reviewed included methodological quality,^1,3 limited patient compliance reporting,³ heterogeneity,^1,2,4 and small sample size.^2,3

More frequent self-monitoring has no effect
A systematic review of 4 RCTs with a total of 637 patients compared frequent SMBG (4-7 times a week) with less frequent self-monitoring (1-2 times a week) for periods ranging from 3 to 12 months and found no difference in reduction of values (HbA1c reduction difference between the groups = –0.21; 95% confidence interval, –0.57 to 0.15).⁴

Recommendations

The American Diabetes Association advocates SMBG as a guide for patients who use oral or medical nutrition therapies for diabetes. Patients should receive initial instruction in SMBG and routine follow-up evaluation of their technique and ability to use data to adjust therapy.⁵

The American Association of Clinical Endocrinologists (AACE) advises that SMBG can be initiated at the same time as medical therapy, lifestyle modification, specific diabetes education, or dietary consultation. If HbA1c levels are above target, the AACE recommends more frequent SMBG: preprandially, 2 hours postprandially, occasionally between 2 am and 3 am, during illness, or anytime a low glucose level is suspected.⁶

Evidence summary

A 2012 Cochrane review and meta-analysis of 9 RCTs found that 1261 patients who used SMBG showed a small but statistically significant decrease in HbA1c at 6 months compared with 1063 controls. In 2 other RCTs, patients using SMBG showed a nonsignificant decrease in HbA1c compared with control subjects at 12 months (TABLE 1).¹

TABLE 1
Difference in HbA1c by duration of self-monitoring*

Another meta-analysis reported similar findings. The study grouped 9 RCTs based on the duration of SMBG and examined the change in HbA1c from baseline. In 5 of the trials, SMBG for 6 months was associated with a small decrease in HbA1c, but in the other 4, SMBG for longer than one year didn’t significantly change HbA1c levels.²

Baseline values make a difference
A meta-analysis of 9 RCTs demonstrated that SMBG was marginally superior to non-SMBG for reducing HbA1c when the baseline value was >8%. SMBG didn’t lower HbA1c in patients with a baseline HbA1c <8%. The greatest change in HbA1c occurred in patients with baseline values >10% (TABLE 2).³

In another meta-analysis, 12 of 15 RCTs found SMBG to be better than non-SMBG at reducing HbA1c when the baseline was >8%.⁴

TABLE 2
Patients benefit from self-monitoring when baseline HbA1c is >8%

Limitations of studies
Limitations of the studies (TABLES 1 AND 2) reviewed included methodological quality,^1,3 limited patient compliance reporting,³ heterogeneity,^1,2,4 and small sample size.^2,3

More frequent self-monitoring has no effect
A systematic review of 4 RCTs with a total of 637 patients compared frequent SMBG (4-7 times a week) with less frequent self-monitoring (1-2 times a week) for periods ranging from 3 to 12 months and found no difference in reduction of values (HbA1c reduction difference between the groups = –0.21; 95% confidence interval, –0.57 to 0.15).⁴

Recommendations

The American Diabetes Association advocates SMBG as a guide for patients who use oral or medical nutrition therapies for diabetes. Patients should receive initial instruction in SMBG and routine follow-up evaluation of their technique and ability to use data to adjust therapy.⁵

The American Association of Clinical Endocrinologists (AACE) advises that SMBG can be initiated at the same time as medical therapy, lifestyle modification, specific diabetes education, or dietary consultation. If HbA1c levels are above target, the AACE recommends more frequent SMBG: preprandially, 2 hours postprandially, occasionally between 2 am and 3 am, during illness, or anytime a low glucose level is suspected.⁶

Evidence summary

A 2012 Cochrane review and meta-analysis of 9 RCTs found that 1261 patients who used SMBG showed a small but statistically significant decrease in HbA1c at 6 months compared with 1063 controls. In 2 other RCTs, patients using SMBG showed a nonsignificant decrease in HbA1c compared with control subjects at 12 months (TABLE 1).¹

TABLE 1
Difference in HbA1c by duration of self-monitoring*

Another meta-analysis reported similar findings. The study grouped 9 RCTs based on the duration of SMBG and examined the change in HbA1c from baseline. In 5 of the trials, SMBG for 6 months was associated with a small decrease in HbA1c, but in the other 4, SMBG for longer than one year didn’t significantly change HbA1c levels.²

Baseline values make a difference
A meta-analysis of 9 RCTs demonstrated that SMBG was marginally superior to non-SMBG for reducing HbA1c when the baseline value was >8%. SMBG didn’t lower HbA1c in patients with a baseline HbA1c <8%. The greatest change in HbA1c occurred in patients with baseline values >10% (TABLE 2).³

In another meta-analysis, 12 of 15 RCTs found SMBG to be better than non-SMBG at reducing HbA1c when the baseline was >8%.⁴

TABLE 2
Patients benefit from self-monitoring when baseline HbA1c is >8%

Limitations of studies
Limitations of the studies (TABLES 1 AND 2) reviewed included methodological quality,^1,3 limited patient compliance reporting,³ heterogeneity,^1,2,4 and small sample size.^2,3

More frequent self-monitoring has no effect
A systematic review of 4 RCTs with a total of 637 patients compared frequent SMBG (4-7 times a week) with less frequent self-monitoring (1-2 times a week) for periods ranging from 3 to 12 months and found no difference in reduction of values (HbA1c reduction difference between the groups = –0.21; 95% confidence interval, –0.57 to 0.15).⁴

Recommendations

The American Diabetes Association advocates SMBG as a guide for patients who use oral or medical nutrition therapies for diabetes. Patients should receive initial instruction in SMBG and routine follow-up evaluation of their technique and ability to use data to adjust therapy.⁵

The American Association of Clinical Endocrinologists (AACE) advises that SMBG can be initiated at the same time as medical therapy, lifestyle modification, specific diabetes education, or dietary consultation. If HbA1c levels are above target, the AACE recommends more frequent SMBG: preprandially, 2 hours postprandially, occasionally between 2 am and 3 am, during illness, or anytime a low glucose level is suspected.⁶

Evidence summary

A double-blind RCT of 191 women who had been treated for breast cancer (two-thirds were taking tamoxifen) and were having at least 14 hot flashes per week randomized the women to one of 4 groups: once-daily extended-release venlafaxine at 37.5, 75, or 150 mg, or placebo.¹ Higher doses of venlafaxine produced greater reductions in hot flash scores from baseline than did the lowest dose or placebo: 150 mg lowered scores by 61% (95% confidence interval [CI], 48%-75%); 75 mg also decreased scores by 61% (95% CI, 50%-68%); 37.5 mg reduced scores by 37% (95% CI, 26%-54%); and placebo lowered scores by 27% (95% CI, 11%-34%).

At the 75-and 150-mg doses, the number needed to treat to reduce the number of hot flashes by 50% was about 3. However, these doses caused dose-dependent adverse effects, including dry mouth, constipation, nausea, and decreased appetite, whereas the lowest dose produced an adverse-effect rate equal to placebo (no statistics supplied for these comparisons).

Venlafaxine significantly decreased the frequency of hot flashes
A pair of double-blind RCTs, reported together, evaluated hot flash frequency among 77 breast cancer survivors with at least 7 hot flashes a week who took either extended- release venlafaxine (37.5 or 75 mg daily) or placebo for 6 weeks.² Patients then crossed over to take either venlafaxine or placebo for another 6 weeks. Compared with placebo, venlafaxine 37.5 mg reduced the mean daily hot flash frequency by 22% (P<.001); venlafaxine 75 mg reduced it by 14% (P<.013).

In an open-label case series, 40 breast cancer patients with at least 20 hot flashes per week received extended-release venlafaxine 37.5 mg once daily for 8 weeks.³ Patients reported a 53% reduction in hot flash frequency compared with baseline (P<.001).

Gabapentin reduces hot flashes at 300 mg tid
A double-blind RCT of 420 women with breast cancer who had at least 14 hot flashes a week compared the effectiveness of gabapentin 100 mg tid, 300 mg tid, or placebo for 8 weeks.⁴ Investigators found no significant difference between the 100 mg tid dosing and placebo. However, gabapentin at 300 mg tid reduced hot flashes by 44% (an average of 2 fewer per day) compared with placebo (P<.007).

The study didn’t assess adverse effects, but noted that the withdrawal rates were similar between the groups receiving gabapentin and placebo (12%-17%). An authoritative online reference lists the 2 most common adverse effects of gabapentin as dizziness (11%-28%) and somnolence (5.5%-25%).⁵

Women prefer venlafaxine to gabapentin
A multicenter, open-label crossover trial involving 66 women with a history of breast cancer who had at least 14 hot flashes per week evaluated patient preference when treated with extended-release venlafaxine 75 mg daily and gabapentin 300 mg tid, each for 1 month.⁶ More patients preferred taking venlafaxine than gabapentin (68% vs 32%; P<.01), although both treatments reduced the numbers of hot flashes.

Recommendations

In a patient education statement, the American Cancer Society states that women with premature menopause caused by cancer treatment may do well with exercise and relaxation techniques alone. The statement mentions venlafaxine, fluoxetine, and paroxetine as adjunct therapy.⁷

Evidence summary

A double-blind RCT of 191 women who had been treated for breast cancer (two-thirds were taking tamoxifen) and were having at least 14 hot flashes per week randomized the women to one of 4 groups: once-daily extended-release venlafaxine at 37.5, 75, or 150 mg, or placebo.¹ Higher doses of venlafaxine produced greater reductions in hot flash scores from baseline than did the lowest dose or placebo: 150 mg lowered scores by 61% (95% confidence interval [CI], 48%-75%); 75 mg also decreased scores by 61% (95% CI, 50%-68%); 37.5 mg reduced scores by 37% (95% CI, 26%-54%); and placebo lowered scores by 27% (95% CI, 11%-34%).

At the 75-and 150-mg doses, the number needed to treat to reduce the number of hot flashes by 50% was about 3. However, these doses caused dose-dependent adverse effects, including dry mouth, constipation, nausea, and decreased appetite, whereas the lowest dose produced an adverse-effect rate equal to placebo (no statistics supplied for these comparisons).

Venlafaxine significantly decreased the frequency of hot flashes
A pair of double-blind RCTs, reported together, evaluated hot flash frequency among 77 breast cancer survivors with at least 7 hot flashes a week who took either extended- release venlafaxine (37.5 or 75 mg daily) or placebo for 6 weeks.² Patients then crossed over to take either venlafaxine or placebo for another 6 weeks. Compared with placebo, venlafaxine 37.5 mg reduced the mean daily hot flash frequency by 22% (P<.001); venlafaxine 75 mg reduced it by 14% (P<.013).

In an open-label case series, 40 breast cancer patients with at least 20 hot flashes per week received extended-release venlafaxine 37.5 mg once daily for 8 weeks.³ Patients reported a 53% reduction in hot flash frequency compared with baseline (P<.001).

Gabapentin reduces hot flashes at 300 mg tid
A double-blind RCT of 420 women with breast cancer who had at least 14 hot flashes a week compared the effectiveness of gabapentin 100 mg tid, 300 mg tid, or placebo for 8 weeks.⁴ Investigators found no significant difference between the 100 mg tid dosing and placebo. However, gabapentin at 300 mg tid reduced hot flashes by 44% (an average of 2 fewer per day) compared with placebo (P<.007).

The study didn’t assess adverse effects, but noted that the withdrawal rates were similar between the groups receiving gabapentin and placebo (12%-17%). An authoritative online reference lists the 2 most common adverse effects of gabapentin as dizziness (11%-28%) and somnolence (5.5%-25%).⁵

Women prefer venlafaxine to gabapentin
A multicenter, open-label crossover trial involving 66 women with a history of breast cancer who had at least 14 hot flashes per week evaluated patient preference when treated with extended-release venlafaxine 75 mg daily and gabapentin 300 mg tid, each for 1 month.⁶ More patients preferred taking venlafaxine than gabapentin (68% vs 32%; P<.01), although both treatments reduced the numbers of hot flashes.

Recommendations

In a patient education statement, the American Cancer Society states that women with premature menopause caused by cancer treatment may do well with exercise and relaxation techniques alone. The statement mentions venlafaxine, fluoxetine, and paroxetine as adjunct therapy.⁷

Evidence summary

A double-blind RCT of 191 women who had been treated for breast cancer (two-thirds were taking tamoxifen) and were having at least 14 hot flashes per week randomized the women to one of 4 groups: once-daily extended-release venlafaxine at 37.5, 75, or 150 mg, or placebo.¹ Higher doses of venlafaxine produced greater reductions in hot flash scores from baseline than did the lowest dose or placebo: 150 mg lowered scores by 61% (95% confidence interval [CI], 48%-75%); 75 mg also decreased scores by 61% (95% CI, 50%-68%); 37.5 mg reduced scores by 37% (95% CI, 26%-54%); and placebo lowered scores by 27% (95% CI, 11%-34%).

At the 75-and 150-mg doses, the number needed to treat to reduce the number of hot flashes by 50% was about 3. However, these doses caused dose-dependent adverse effects, including dry mouth, constipation, nausea, and decreased appetite, whereas the lowest dose produced an adverse-effect rate equal to placebo (no statistics supplied for these comparisons).

Venlafaxine significantly decreased the frequency of hot flashes
A pair of double-blind RCTs, reported together, evaluated hot flash frequency among 77 breast cancer survivors with at least 7 hot flashes a week who took either extended- release venlafaxine (37.5 or 75 mg daily) or placebo for 6 weeks.² Patients then crossed over to take either venlafaxine or placebo for another 6 weeks. Compared with placebo, venlafaxine 37.5 mg reduced the mean daily hot flash frequency by 22% (P<.001); venlafaxine 75 mg reduced it by 14% (P<.013).

In an open-label case series, 40 breast cancer patients with at least 20 hot flashes per week received extended-release venlafaxine 37.5 mg once daily for 8 weeks.³ Patients reported a 53% reduction in hot flash frequency compared with baseline (P<.001).

Gabapentin reduces hot flashes at 300 mg tid
A double-blind RCT of 420 women with breast cancer who had at least 14 hot flashes a week compared the effectiveness of gabapentin 100 mg tid, 300 mg tid, or placebo for 8 weeks.⁴ Investigators found no significant difference between the 100 mg tid dosing and placebo. However, gabapentin at 300 mg tid reduced hot flashes by 44% (an average of 2 fewer per day) compared with placebo (P<.007).

The study didn’t assess adverse effects, but noted that the withdrawal rates were similar between the groups receiving gabapentin and placebo (12%-17%). An authoritative online reference lists the 2 most common adverse effects of gabapentin as dizziness (11%-28%) and somnolence (5.5%-25%).⁵

Women prefer venlafaxine to gabapentin
A multicenter, open-label crossover trial involving 66 women with a history of breast cancer who had at least 14 hot flashes per week evaluated patient preference when treated with extended-release venlafaxine 75 mg daily and gabapentin 300 mg tid, each for 1 month.⁶ More patients preferred taking venlafaxine than gabapentin (68% vs 32%; P<.01), although both treatments reduced the numbers of hot flashes.

Recommendations

In a patient education statement, the American Cancer Society states that women with premature menopause caused by cancer treatment may do well with exercise and relaxation techniques alone. The statement mentions venlafaxine, fluoxetine, and paroxetine as adjunct therapy.⁷

Evidence summary

Two double-blind RCTs that compared cocoa butter with placebo to prevent stretch marks in pregnant women found no difference. In the first, investigators enrolled 300 Afro-Caribbean women at 12 to 15 weeks’ gestation. Women used either 25% cocoa butter cream or a placebo cream comprised of emollients and vitamin E daily. Investigators monitored compliance and assessed stretch marks using a validated scale at 26 weeks, 36 weeks, and after delivery. Cocoa butter cream didn’t reduce stretch marks (44% vs 55% for placebo; P=.09). Three women (1 using cocoa butter, 2 using placebo) discontinued the cream because of mild self-limiting reactions.¹

In the second RCT, investigators randomized 210 nulliparous women (mainly with intermediate skin color) to use cocoa butter lotion with vitamin E or placebo. Women applied the lotion daily to their abdomen, breasts, and thighs, starting at 12 to 18 weeks’ gestation. Investigators assessed the severity of stretch marks either at delivery or postpartum using a validated scale. Cocoa butter lotion didn’t prevent stretch marks (45.1% vs 48.8% for placebo; P=.730).²

Save the olive oil for cooking
A nonblinded RCT that compared twice-daily olive oil massage with no treatment in 70 nulliparous women beginning at 18 to 20 weeks’ gestation found no reduction in stretch marks (45.7% vs 62.9% without olive oil massage; P=.115). The investigators didn’t report whether they performed a sample-size analysis to determine if the study was adequately powered to demonstrate no difference.³

Mixed, but mostly negative, results for multi-ingredient cream
A double-blind RCT found that Trofolastin cream containing Centella asiatica (also known as Gotu kola, a member of the parsley family), vitamin E, and collagen hydrolysates didn’t prevent pregnancy-related stretch marks among 80 women who applied the treatment beginning at 12 weeks’ gestation.

When investigators evaluated a subgroup of 18 women who had already developed stretch marks during puberty, they found that fewer of the women acquired additional stretch marks during pregnancy (11% vs 100% with placebo; P=.0001). The investigators didn’t calculate whether the sample was large enough to prove a significant difference.⁴

Fewer stretch marks with vitamin E in small flawed study
An older systematic review (1996) included a prospective RCT that randomized 50 women at 20 weeks’ gestation to massage their abdomen, thighs, and breasts with vitamin E ointment or perform no massage. The trial found fewer stretch marks with vitamin E ointment massage (odds ratio=0.26; 95% confidence interval, 0.08-0.84). The authors of the review described this RCT as poorly randomized and without blinding; investigators didn’t report whether the sample size was adequate to demonstrate a significant effect.⁵

Recommendations

The American Congress of Obstetricians and Gynecologists Web site states that although many creams, lotions, and oils on the market claim to prevent stretch marks, no proof exists that these treatments work. Using a heavy moisturizer may help keep skin soft, but it won’t help get rid of stretch marks.⁶

The American Academy of Dermatology Web site also says that a moisturizer can improve the appearance of stretch marks and reduce itchiness; sunless tanning products can hide the marks.⁷

Evidence summary

Two double-blind RCTs that compared cocoa butter with placebo to prevent stretch marks in pregnant women found no difference. In the first, investigators enrolled 300 Afro-Caribbean women at 12 to 15 weeks’ gestation. Women used either 25% cocoa butter cream or a placebo cream comprised of emollients and vitamin E daily. Investigators monitored compliance and assessed stretch marks using a validated scale at 26 weeks, 36 weeks, and after delivery. Cocoa butter cream didn’t reduce stretch marks (44% vs 55% for placebo; P=.09). Three women (1 using cocoa butter, 2 using placebo) discontinued the cream because of mild self-limiting reactions.¹

In the second RCT, investigators randomized 210 nulliparous women (mainly with intermediate skin color) to use cocoa butter lotion with vitamin E or placebo. Women applied the lotion daily to their abdomen, breasts, and thighs, starting at 12 to 18 weeks’ gestation. Investigators assessed the severity of stretch marks either at delivery or postpartum using a validated scale. Cocoa butter lotion didn’t prevent stretch marks (45.1% vs 48.8% for placebo; P=.730).²

Save the olive oil for cooking
A nonblinded RCT that compared twice-daily olive oil massage with no treatment in 70 nulliparous women beginning at 18 to 20 weeks’ gestation found no reduction in stretch marks (45.7% vs 62.9% without olive oil massage; P=.115). The investigators didn’t report whether they performed a sample-size analysis to determine if the study was adequately powered to demonstrate no difference.³

Mixed, but mostly negative, results for multi-ingredient cream
A double-blind RCT found that Trofolastin cream containing Centella asiatica (also known as Gotu kola, a member of the parsley family), vitamin E, and collagen hydrolysates didn’t prevent pregnancy-related stretch marks among 80 women who applied the treatment beginning at 12 weeks’ gestation.

When investigators evaluated a subgroup of 18 women who had already developed stretch marks during puberty, they found that fewer of the women acquired additional stretch marks during pregnancy (11% vs 100% with placebo; P=.0001). The investigators didn’t calculate whether the sample was large enough to prove a significant difference.⁴

Fewer stretch marks with vitamin E in small flawed study
An older systematic review (1996) included a prospective RCT that randomized 50 women at 20 weeks’ gestation to massage their abdomen, thighs, and breasts with vitamin E ointment or perform no massage. The trial found fewer stretch marks with vitamin E ointment massage (odds ratio=0.26; 95% confidence interval, 0.08-0.84). The authors of the review described this RCT as poorly randomized and without blinding; investigators didn’t report whether the sample size was adequate to demonstrate a significant effect.⁵

Recommendations

The American Congress of Obstetricians and Gynecologists Web site states that although many creams, lotions, and oils on the market claim to prevent stretch marks, no proof exists that these treatments work. Using a heavy moisturizer may help keep skin soft, but it won’t help get rid of stretch marks.⁶

The American Academy of Dermatology Web site also says that a moisturizer can improve the appearance of stretch marks and reduce itchiness; sunless tanning products can hide the marks.⁷

Evidence summary

Two double-blind RCTs that compared cocoa butter with placebo to prevent stretch marks in pregnant women found no difference. In the first, investigators enrolled 300 Afro-Caribbean women at 12 to 15 weeks’ gestation. Women used either 25% cocoa butter cream or a placebo cream comprised of emollients and vitamin E daily. Investigators monitored compliance and assessed stretch marks using a validated scale at 26 weeks, 36 weeks, and after delivery. Cocoa butter cream didn’t reduce stretch marks (44% vs 55% for placebo; P=.09). Three women (1 using cocoa butter, 2 using placebo) discontinued the cream because of mild self-limiting reactions.¹

In the second RCT, investigators randomized 210 nulliparous women (mainly with intermediate skin color) to use cocoa butter lotion with vitamin E or placebo. Women applied the lotion daily to their abdomen, breasts, and thighs, starting at 12 to 18 weeks’ gestation. Investigators assessed the severity of stretch marks either at delivery or postpartum using a validated scale. Cocoa butter lotion didn’t prevent stretch marks (45.1% vs 48.8% for placebo; P=.730).²

Save the olive oil for cooking
A nonblinded RCT that compared twice-daily olive oil massage with no treatment in 70 nulliparous women beginning at 18 to 20 weeks’ gestation found no reduction in stretch marks (45.7% vs 62.9% without olive oil massage; P=.115). The investigators didn’t report whether they performed a sample-size analysis to determine if the study was adequately powered to demonstrate no difference.³

Mixed, but mostly negative, results for multi-ingredient cream
A double-blind RCT found that Trofolastin cream containing Centella asiatica (also known as Gotu kola, a member of the parsley family), vitamin E, and collagen hydrolysates didn’t prevent pregnancy-related stretch marks among 80 women who applied the treatment beginning at 12 weeks’ gestation.

When investigators evaluated a subgroup of 18 women who had already developed stretch marks during puberty, they found that fewer of the women acquired additional stretch marks during pregnancy (11% vs 100% with placebo; P=.0001). The investigators didn’t calculate whether the sample was large enough to prove a significant difference.⁴

Fewer stretch marks with vitamin E in small flawed study
An older systematic review (1996) included a prospective RCT that randomized 50 women at 20 weeks’ gestation to massage their abdomen, thighs, and breasts with vitamin E ointment or perform no massage. The trial found fewer stretch marks with vitamin E ointment massage (odds ratio=0.26; 95% confidence interval, 0.08-0.84). The authors of the review described this RCT as poorly randomized and without blinding; investigators didn’t report whether the sample size was adequate to demonstrate a significant effect.⁵

Recommendations

The American Congress of Obstetricians and Gynecologists Web site states that although many creams, lotions, and oils on the market claim to prevent stretch marks, no proof exists that these treatments work. Using a heavy moisturizer may help keep skin soft, but it won’t help get rid of stretch marks.⁶

The American Academy of Dermatology Web site also says that a moisturizer can improve the appearance of stretch marks and reduce itchiness; sunless tanning products can hide the marks.⁷