User login
Which nutritional therapies are safe and effective for depression?
ST. JOHN’S WORT is effective for short-term relief of mild to moderate depression (strength of recommendation [SOR]: A; 1 systematic review). Its safety profile is superior to older antidepressants; data comparing it with newer antidepressants (such as selective serotonin reuptake inhibitors) are limited (SOR: A, 1 systematic review).
A small but statistically significant clinical benefit has been demonstrated for saffron, lavender, borage, dan zhi xiao yao (SOR: B, 1 systematic review and 3 randomized controlled trials), folate (SOR: A, 1 systematic review), and S-adenosylmethionine (SAMe) (SOR: A, 1 meta-analysis and 1 systematic review). Most trials of these preparations were short and small, limiting the ability to detect adverse effects.
Tryptophan (SOR: A, 1 systematic review) and 5-hydroxytryptophan (5-HTP) (SOR: A, 1 systematic review) have demonstrated superiority over placebo in alleviating symptoms of depression, but concerns exist about their safety.
N-3 long-chain polyunsaturated fatty acids (n-3 PUFAs) and omega-3 fatty acids don’t appear effective in treating major depressive disorder (SOR: A, 1 systematic review.)
Evidence summary
TABLE W1 summarizes study results and recommendations for nutritional therapies for depression.1-16
St. John’s wort works as well as standard antidepressants
A recent Cochrane review suggested that St. John’s wort is more effective than placebo in patients with mild to moderate depression and as effective as standard antidepressants.1
Other supplements also have benefits
A systematic review of 4 small randomized controlled trials (RCTs) suggested that saffron (30 mg) is superior to placebo in treating short-term depression (6 weeks). Treatment and outcomes were equivalent to fluoxetine and imipramine.2 A later RCT yielded results consistent with the systematic review.3
Combined lavender tincture (60 drops per day) and imipramine were more effective than imipramine alone in 1 small RCT.4
Borage, a traditional Persian medicine, was superior to placebo in reducing depressive symptoms in 1 small RCT.2
Dan zhi xiao yao, a traditional Chinese medicine, was as effective as the tricyclic anti-depressant maprotiline in 1 small RCT.2
Three RCTs suggested that folate may be used to supplement conventional treatments for depression, but it isn’t clear whether this would help patients with normal folate levels.5
A meta-analysis of 13 controlled clinical trials and a later systematic review of 11 articles including 2 RCTs concluded that SAMe is more effective than placebo and as efficacious as tricyclic antidepressants in treating major depression in adults. However, further trials are needed to answer questions about absorption, mechanism of action, and bioavailability.6,7
Tryptophan’s benefit comes with risk
In a Cochrane review of 2 RCTs, tryptophan and 5-HTP were superior to placebo in alleviating symptoms of depression. However, some published case reports have linked tryptophan use to potentially fatal eosinophilia-myalgia syndrome.8
No clear evidence for inositol or n-3 PUFAs
A Cochrane review of 4 small double-blind RCTs investigating inositol as a nutritional supplement in depression treatment failed to find clear evidence of therapeutic benefit.9
Three RCTs demonstrated significantly higher red blood cell membrane levels of n-3 PUFAs in nondepressed patients compared with depressed patients.10 However, a systematic review of 12 RCTs failed to demonstrate any benefit of n-3 PUFA supplementation over placebo in treating depressed mood.11 The authors concluded that larger trials are needed to demonstrate efficacy because of marked heterogeneity among the RCTs.
Safety issues. A recent Cochrane review found fewer adverse effects for St. John’s wort than tricyclic antidepressants.1 The most common adverse reactions were sensitivity to light, skin symptoms, gastrointestinal symptoms, and agitation. Data comparing St. John’s wort with newer antidepressants are lacking.
St. John’s wort does have pharmacokinetic interactions and should not be taken concurrently with other antidepressants, immunosuppressants, anti-HIV drugs, cou-marin-type anticoagulants, or certain antineoplastic agents.17
Reviews of meta-analyses, case reports, population studies, RCTs, and other literature have reported virtually no adverse effects for PUFAs; 18 trials investigating saffron, lavender, borage, dan zhi xiao yao, folate, SAMe, and inositol also reported no safety concerns. However, the size and duration of these studies limit their ability to detect significant problems.2,5,6,9 As previously noted, concerns exist regarding an association between tryptophan and eosinophilia-myalgia syndrome.8
Recommendations
The World Federation of Societies of Biological Psychiatry doesn’t recommend St. John’s wort for moderate to severe depression, but suggests it can be considered for treating mild to moderate depressive episodes provided the prescriber considers potential pharmacokinetic interactions with other medications and understands possible variations in purity and potency of extracts.19 The Federation also states that St. John’s wort is an alternative for patients reluctant to take traditional antidepressants.
TABLE W1
What the studies say about nutritional therapies for depression
| Supplement | Study type | Number of subjects | Comparison group | Outcome measure | Results | Conclusion | SOR |
|---|---|---|---|---|---|---|---|
| Borage (Echium amoenum) | 1 small RCT | 352,16 | Placebo | HAM-D | Improved HAM-D scores significantly at week 4 (borage 18. 3±3. 9 vs placebo 21. 9±3. 9; t=2. 51; P=. 02); no significant difference at Week 62,16 | Superior to placebo in reducing symptoms of depression | B |
| Dan zhi xiao yao | 1 small RCT | 632 | Maprotiline | HAM-D, SDS, SAS, scale for traditional Chinese medicine syndrome and symptom differentiation | 87% depression reduction (dan zhi xiao yao) vs 84% depression reduction (maprotiline) | As effective as maprotiline in treating depression | B |
| Folate | Cochrane review of 3 RCTs | 2475 | Studies 1 and 2: folate vs folate + other treatment (Study 1: low folate levels; Study 2: normal folate levels) Study 3: folate vs trazodone (normal folate levels) | HAM-D | Superior to placebo (NNT=5, defined as 50% reduction in HAM-D); comparable to trazodone (RR=0. 97; 95% CI, 0. 14-2. 01)7 | May have role as supplement to other treatments for depression Efficacy unclear in patients with normal folate levels | A |
| Inositol | Cochrane review of 4 RCTs | 1419 | Studies 1-3: placebo plus conventional antidepressants Study 4: placebo only | HAM-D, MADRS | Pooled estimate of effect of all 3 studies (SMD= -0. 08; 95% CI, -0. 45 to 0. 30) | No clear evidence of therapeutic benefit | A |
| Lavender (Lavandula angustifolia) | 1 small RCT | 454 | Imipramine | HAM-D | Imipramine plus lavender showed significant effect compared with imipramine alone (f=26. 87; Df=3. 01; P<. 0001) | Synergistic effect suggested when used with imipramine | B |
| n-3 long-chain polyunsaturated fatty acids | Systematic review including 3 RCTs; 10 meta-analysis of 12 RCTs11 | 10210 103211 | Various comparison groups included | Serum SFAs, MUFAs, PUFAs; RBC membrane levels n-3 PUFAs2 HAM-D, BDI3 | Systematic review:10 Study 1: n=30; ES=3. 61 Study 2: n=24; ES=1. 2 Study 3: n=48; ES=2. 43 Meta-analysis:11 Pooled ES=0. 13; 95% CI, 0. 01-0. 25 | Significantly higher RBC membrane levels of n-3 PUFAs in nondepressed vs depressed patients10 No significant effect for supplementation11 Larger trials with adequate power needed2,3 | A |
| S-adenosyl-methionine (SAMe) | Meta-analysis of 13 RCTs,6 systematic review including 2 RCTs7 | 3996 787 | Placebo and conventional antidepressants | HAM-D | NNT=2. 5 for HAM-D decrease of >25%; 6 NNT=6. 25 for HAM-D decrease of >50%6 | May have role in treatment of major depression Further trials are needed to address unanswered questions about absorption, mechanism of action, and bioavailability7 | A |
| Saffron (Crocus sativus) | Systematic review of 4 small RCTs, 1 later RCT | 3012 4013 4014 4015 403 | Imipramine12 Placebo13,15 Fluoxetine5,14 | HAM-D | Systematic review: Study 1: imipramine and saffron equally efficacious (f=2. 91; P=. 09)12 Study 2: Improved HAM-D scores: -12. 20±4. 67 (saffron) vs -5. 10±4. 71 (placebo) (P<. 0001)13 Study 3: Improved HAM-D scores: saffron petal -12. 00±4. 10; fluoxetine -13. 50±4. 91; difference between 2 treatments not significant (P=. 27)14 Study 4: Improved HAM-D scores: -14. 01±5. 53 (saffron petal) vs -5. 05±4. 63 (placebo) (P<. 0001)15 Study 5:5 NNT=10 | Efficacy of extract and petal suggested to treat mild to moderate depression Large-scale trials are warranted | B |
| St. John’s wort (Hypericum perforatum L. ) | Cochrane review of 29 RCTs | 54891 | SSRIs, tri/tetracyclic antidepressants, placebo | Responder rate ratio | St. John’s wort vs placebo: 9 larger trials: RR=1. 28; 95% CI, 1. 10-1. 491 9 smaller trials: RR=1. 87; 95% CI, 1. 22-2. 871 St. John’s wort vs SSRIs: 12 trials: RR=1. 00; 95% CI, 0. 90-1. 111 St. John’s wort vs tricyclics: 5 trials: RR=1. 02; 95% CI, 0. 90-1. 151 | Effective for treating mild to moderate depression | A |
| Tryptophan and 5-hydroxy-tryptophan (5-HTP) | Cochrane review of 2 RCTs | 648 | Placebo | HAM-D | NNT=2. 78 vs placebo (OR=4. 1; 95% CI, 1. 28-13. 15 | Superior to placebo Insufficient evidence regarding safety | A |
| BDI, Beck Depression Inventory; CI, confidence interval; DF, degrees of freedom; ES, effect size; F, F statistic; HAM-D, Hamilton Depression Rating Scale; MADRS, Mont-gomery-Asberg Depression Rating Scale; MUFAs, monounsaturated fatty acids; n-3 PUFAs, n-3 long-chain polyunsaturated fatty acids; NNT, number needed to treat; OR, odds ratio; PUFAs, polyunsaturated fatty acids; RBC, red blood cell; RCT, randomized controlled trial; RR, relative risk; SAS, self-rating anxiety scale; SDS, self-rating depression scale; SFAs, saturated fatty acids; SMD, standard weighted mean difference; SOR, strength of recommendation; SSRI, selective serotonin reuptake inhibitor. | |||||||
1. Linde K, Bemer MM, Kriston L. St. John’s wort for major depression. Cochrane Database Syst Rev. 2008;(4):CD000448.-
2. Sarris J. Herbal medicines in the treatment of psychiatric disorders: a systematic review. Phytother Res. 2007;21:703-716.
3. Akhondzadeh Basti A, Moshiri E, Noorbala AA, et al. Comparison of petal of Crocus sativus L. and fluoxetine in the treatment of depressed outpatients. Prog Neuropsychopharmacol Biol Psychiatry. 2006;31:439-442.
4. Akhondzadeh S, Kashani L, Fotouhi A, et al. Comparison of Lavandula angustifolia Mill. tincture and imipramine in the treatment of mild to moderate depression. Prog Neuropsychopharmacol Biol Psychiatry. 2003;27:123-127.
5. Taylor MJ, Carney S, Geddes J, et al. Folate for depressive disorders. Cochrane Database Syst Rev. 2003;(2):CD003390.-
6. Bressa GM. S-adenosyl-l-methionine (SAMe) as antidepres-sant. Acta Neurol Scand Suppl. 1994;154:7-14.
7. Williams AL, Girard C, Jui D, et al. S-adenosylmethionine (SAMe) as treatment for depression. Clin Invest Med. 2005;28:132-139.
8. Shaw K, Turner J, Del Mar C. Tryptophan and 5-hydroxy-tryptophan for depression. Cochrane Database Syst Rev. 2002;(1):CD003198.-
9. Taylor MJ, Wilder H, Bhagwager Z, et al. Inositol for depressive disorders. Cochrane Database Syst Rev. 2004;(2):CD004049.-
10. Williams AL, Katz D, Ali A, et al. Do essential fatty acids have a role in the treatment of depression? J Affect Disord. 2006;93:117-123.
11. Appleton KM, Hayward RC, Gunnell D, et al. Effects of n-3 longchain polyunsaturated fatty acids on depressed mood. Am J Clin Nutr. 2006;84:1308-1316.
12. Akhondzadeh S, Fallah-Pour H, Afkham K, et al. Comparison of Crocus sativus L. and imipramine in the treatment of mild to moderate depression. BMC Complement Altern Med. 2004;4:12.-
13. Akhondzadeh S, Tahmacebi-Pour N, Noorbala AA, et al. Crocus sativus L. in the treatment of mild to moderate depression. Phytother Res. 2005;19:148-151.
14. Noorbala AA, Akhondzadeh S, Tahmacebi-Pour N, et al. Hydro-alcoholic extract of Crocus sativus L. versus fluoxetine in the treatment of mild to moderate depression. J Ethnopharmacol. 2005;97:281-284.
15. Moshiri E, Basti AA, Noorbala AA, et al. Crocus sativus L. (petal) in the treatment of mild-to-moderate depression. Phytomedicine. 2006;13:607-611.
16. Sayyah M, Sahhah M, Kamalinejad M. A preliminary randomized double blind clinical trial on the efficacy of aqueous extract of Echium amoenum in the treatment of mild to moderate major depression. Prog Neuropsychopharmacol Biol Psychiatry. 2006;30:166-169.
17. Schulz V Safety of St. John’s wort extract compared to synthetic antidepressants. Phytomedicine. 2006;13:199-204.
18. Lee S, Gura KM, Kim S, et al. Current clinical applications of ome-ga-6 and omega-3 fatty acids. Nutr Clin Pract. 2006;21:323-341.
19. Bauer M, Bschor T, Pfennig A, et al. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for biological treatment of unipolar depressive disorders in primary care. World J Biol Psychiatry. 2007;8:67-104.
ST. JOHN’S WORT is effective for short-term relief of mild to moderate depression (strength of recommendation [SOR]: A; 1 systematic review). Its safety profile is superior to older antidepressants; data comparing it with newer antidepressants (such as selective serotonin reuptake inhibitors) are limited (SOR: A, 1 systematic review).
A small but statistically significant clinical benefit has been demonstrated for saffron, lavender, borage, dan zhi xiao yao (SOR: B, 1 systematic review and 3 randomized controlled trials), folate (SOR: A, 1 systematic review), and S-adenosylmethionine (SAMe) (SOR: A, 1 meta-analysis and 1 systematic review). Most trials of these preparations were short and small, limiting the ability to detect adverse effects.
Tryptophan (SOR: A, 1 systematic review) and 5-hydroxytryptophan (5-HTP) (SOR: A, 1 systematic review) have demonstrated superiority over placebo in alleviating symptoms of depression, but concerns exist about their safety.
N-3 long-chain polyunsaturated fatty acids (n-3 PUFAs) and omega-3 fatty acids don’t appear effective in treating major depressive disorder (SOR: A, 1 systematic review.)
Evidence summary
TABLE W1 summarizes study results and recommendations for nutritional therapies for depression.1-16
St. John’s wort works as well as standard antidepressants
A recent Cochrane review suggested that St. John’s wort is more effective than placebo in patients with mild to moderate depression and as effective as standard antidepressants.1
Other supplements also have benefits
A systematic review of 4 small randomized controlled trials (RCTs) suggested that saffron (30 mg) is superior to placebo in treating short-term depression (6 weeks). Treatment and outcomes were equivalent to fluoxetine and imipramine.2 A later RCT yielded results consistent with the systematic review.3
Combined lavender tincture (60 drops per day) and imipramine were more effective than imipramine alone in 1 small RCT.4
Borage, a traditional Persian medicine, was superior to placebo in reducing depressive symptoms in 1 small RCT.2
Dan zhi xiao yao, a traditional Chinese medicine, was as effective as the tricyclic anti-depressant maprotiline in 1 small RCT.2
Three RCTs suggested that folate may be used to supplement conventional treatments for depression, but it isn’t clear whether this would help patients with normal folate levels.5
A meta-analysis of 13 controlled clinical trials and a later systematic review of 11 articles including 2 RCTs concluded that SAMe is more effective than placebo and as efficacious as tricyclic antidepressants in treating major depression in adults. However, further trials are needed to answer questions about absorption, mechanism of action, and bioavailability.6,7
Tryptophan’s benefit comes with risk
In a Cochrane review of 2 RCTs, tryptophan and 5-HTP were superior to placebo in alleviating symptoms of depression. However, some published case reports have linked tryptophan use to potentially fatal eosinophilia-myalgia syndrome.8
No clear evidence for inositol or n-3 PUFAs
A Cochrane review of 4 small double-blind RCTs investigating inositol as a nutritional supplement in depression treatment failed to find clear evidence of therapeutic benefit.9
Three RCTs demonstrated significantly higher red blood cell membrane levels of n-3 PUFAs in nondepressed patients compared with depressed patients.10 However, a systematic review of 12 RCTs failed to demonstrate any benefit of n-3 PUFA supplementation over placebo in treating depressed mood.11 The authors concluded that larger trials are needed to demonstrate efficacy because of marked heterogeneity among the RCTs.
Safety issues. A recent Cochrane review found fewer adverse effects for St. John’s wort than tricyclic antidepressants.1 The most common adverse reactions were sensitivity to light, skin symptoms, gastrointestinal symptoms, and agitation. Data comparing St. John’s wort with newer antidepressants are lacking.
St. John’s wort does have pharmacokinetic interactions and should not be taken concurrently with other antidepressants, immunosuppressants, anti-HIV drugs, cou-marin-type anticoagulants, or certain antineoplastic agents.17
Reviews of meta-analyses, case reports, population studies, RCTs, and other literature have reported virtually no adverse effects for PUFAs; 18 trials investigating saffron, lavender, borage, dan zhi xiao yao, folate, SAMe, and inositol also reported no safety concerns. However, the size and duration of these studies limit their ability to detect significant problems.2,5,6,9 As previously noted, concerns exist regarding an association between tryptophan and eosinophilia-myalgia syndrome.8
Recommendations
The World Federation of Societies of Biological Psychiatry doesn’t recommend St. John’s wort for moderate to severe depression, but suggests it can be considered for treating mild to moderate depressive episodes provided the prescriber considers potential pharmacokinetic interactions with other medications and understands possible variations in purity and potency of extracts.19 The Federation also states that St. John’s wort is an alternative for patients reluctant to take traditional antidepressants.
TABLE W1
What the studies say about nutritional therapies for depression
| Supplement | Study type | Number of subjects | Comparison group | Outcome measure | Results | Conclusion | SOR |
|---|---|---|---|---|---|---|---|
| Borage (Echium amoenum) | 1 small RCT | 352,16 | Placebo | HAM-D | Improved HAM-D scores significantly at week 4 (borage 18. 3±3. 9 vs placebo 21. 9±3. 9; t=2. 51; P=. 02); no significant difference at Week 62,16 | Superior to placebo in reducing symptoms of depression | B |
| Dan zhi xiao yao | 1 small RCT | 632 | Maprotiline | HAM-D, SDS, SAS, scale for traditional Chinese medicine syndrome and symptom differentiation | 87% depression reduction (dan zhi xiao yao) vs 84% depression reduction (maprotiline) | As effective as maprotiline in treating depression | B |
| Folate | Cochrane review of 3 RCTs | 2475 | Studies 1 and 2: folate vs folate + other treatment (Study 1: low folate levels; Study 2: normal folate levels) Study 3: folate vs trazodone (normal folate levels) | HAM-D | Superior to placebo (NNT=5, defined as 50% reduction in HAM-D); comparable to trazodone (RR=0. 97; 95% CI, 0. 14-2. 01)7 | May have role as supplement to other treatments for depression Efficacy unclear in patients with normal folate levels | A |
| Inositol | Cochrane review of 4 RCTs | 1419 | Studies 1-3: placebo plus conventional antidepressants Study 4: placebo only | HAM-D, MADRS | Pooled estimate of effect of all 3 studies (SMD= -0. 08; 95% CI, -0. 45 to 0. 30) | No clear evidence of therapeutic benefit | A |
| Lavender (Lavandula angustifolia) | 1 small RCT | 454 | Imipramine | HAM-D | Imipramine plus lavender showed significant effect compared with imipramine alone (f=26. 87; Df=3. 01; P<. 0001) | Synergistic effect suggested when used with imipramine | B |
| n-3 long-chain polyunsaturated fatty acids | Systematic review including 3 RCTs; 10 meta-analysis of 12 RCTs11 | 10210 103211 | Various comparison groups included | Serum SFAs, MUFAs, PUFAs; RBC membrane levels n-3 PUFAs2 HAM-D, BDI3 | Systematic review:10 Study 1: n=30; ES=3. 61 Study 2: n=24; ES=1. 2 Study 3: n=48; ES=2. 43 Meta-analysis:11 Pooled ES=0. 13; 95% CI, 0. 01-0. 25 | Significantly higher RBC membrane levels of n-3 PUFAs in nondepressed vs depressed patients10 No significant effect for supplementation11 Larger trials with adequate power needed2,3 | A |
| S-adenosyl-methionine (SAMe) | Meta-analysis of 13 RCTs,6 systematic review including 2 RCTs7 | 3996 787 | Placebo and conventional antidepressants | HAM-D | NNT=2. 5 for HAM-D decrease of >25%; 6 NNT=6. 25 for HAM-D decrease of >50%6 | May have role in treatment of major depression Further trials are needed to address unanswered questions about absorption, mechanism of action, and bioavailability7 | A |
| Saffron (Crocus sativus) | Systematic review of 4 small RCTs, 1 later RCT | 3012 4013 4014 4015 403 | Imipramine12 Placebo13,15 Fluoxetine5,14 | HAM-D | Systematic review: Study 1: imipramine and saffron equally efficacious (f=2. 91; P=. 09)12 Study 2: Improved HAM-D scores: -12. 20±4. 67 (saffron) vs -5. 10±4. 71 (placebo) (P<. 0001)13 Study 3: Improved HAM-D scores: saffron petal -12. 00±4. 10; fluoxetine -13. 50±4. 91; difference between 2 treatments not significant (P=. 27)14 Study 4: Improved HAM-D scores: -14. 01±5. 53 (saffron petal) vs -5. 05±4. 63 (placebo) (P<. 0001)15 Study 5:5 NNT=10 | Efficacy of extract and petal suggested to treat mild to moderate depression Large-scale trials are warranted | B |
| St. John’s wort (Hypericum perforatum L. ) | Cochrane review of 29 RCTs | 54891 | SSRIs, tri/tetracyclic antidepressants, placebo | Responder rate ratio | St. John’s wort vs placebo: 9 larger trials: RR=1. 28; 95% CI, 1. 10-1. 491 9 smaller trials: RR=1. 87; 95% CI, 1. 22-2. 871 St. John’s wort vs SSRIs: 12 trials: RR=1. 00; 95% CI, 0. 90-1. 111 St. John’s wort vs tricyclics: 5 trials: RR=1. 02; 95% CI, 0. 90-1. 151 | Effective for treating mild to moderate depression | A |
| Tryptophan and 5-hydroxy-tryptophan (5-HTP) | Cochrane review of 2 RCTs | 648 | Placebo | HAM-D | NNT=2. 78 vs placebo (OR=4. 1; 95% CI, 1. 28-13. 15 | Superior to placebo Insufficient evidence regarding safety | A |
| BDI, Beck Depression Inventory; CI, confidence interval; DF, degrees of freedom; ES, effect size; F, F statistic; HAM-D, Hamilton Depression Rating Scale; MADRS, Mont-gomery-Asberg Depression Rating Scale; MUFAs, monounsaturated fatty acids; n-3 PUFAs, n-3 long-chain polyunsaturated fatty acids; NNT, number needed to treat; OR, odds ratio; PUFAs, polyunsaturated fatty acids; RBC, red blood cell; RCT, randomized controlled trial; RR, relative risk; SAS, self-rating anxiety scale; SDS, self-rating depression scale; SFAs, saturated fatty acids; SMD, standard weighted mean difference; SOR, strength of recommendation; SSRI, selective serotonin reuptake inhibitor. | |||||||
ST. JOHN’S WORT is effective for short-term relief of mild to moderate depression (strength of recommendation [SOR]: A; 1 systematic review). Its safety profile is superior to older antidepressants; data comparing it with newer antidepressants (such as selective serotonin reuptake inhibitors) are limited (SOR: A, 1 systematic review).
A small but statistically significant clinical benefit has been demonstrated for saffron, lavender, borage, dan zhi xiao yao (SOR: B, 1 systematic review and 3 randomized controlled trials), folate (SOR: A, 1 systematic review), and S-adenosylmethionine (SAMe) (SOR: A, 1 meta-analysis and 1 systematic review). Most trials of these preparations were short and small, limiting the ability to detect adverse effects.
Tryptophan (SOR: A, 1 systematic review) and 5-hydroxytryptophan (5-HTP) (SOR: A, 1 systematic review) have demonstrated superiority over placebo in alleviating symptoms of depression, but concerns exist about their safety.
N-3 long-chain polyunsaturated fatty acids (n-3 PUFAs) and omega-3 fatty acids don’t appear effective in treating major depressive disorder (SOR: A, 1 systematic review.)
Evidence summary
TABLE W1 summarizes study results and recommendations for nutritional therapies for depression.1-16
St. John’s wort works as well as standard antidepressants
A recent Cochrane review suggested that St. John’s wort is more effective than placebo in patients with mild to moderate depression and as effective as standard antidepressants.1
Other supplements also have benefits
A systematic review of 4 small randomized controlled trials (RCTs) suggested that saffron (30 mg) is superior to placebo in treating short-term depression (6 weeks). Treatment and outcomes were equivalent to fluoxetine and imipramine.2 A later RCT yielded results consistent with the systematic review.3
Combined lavender tincture (60 drops per day) and imipramine were more effective than imipramine alone in 1 small RCT.4
Borage, a traditional Persian medicine, was superior to placebo in reducing depressive symptoms in 1 small RCT.2
Dan zhi xiao yao, a traditional Chinese medicine, was as effective as the tricyclic anti-depressant maprotiline in 1 small RCT.2
Three RCTs suggested that folate may be used to supplement conventional treatments for depression, but it isn’t clear whether this would help patients with normal folate levels.5
A meta-analysis of 13 controlled clinical trials and a later systematic review of 11 articles including 2 RCTs concluded that SAMe is more effective than placebo and as efficacious as tricyclic antidepressants in treating major depression in adults. However, further trials are needed to answer questions about absorption, mechanism of action, and bioavailability.6,7
Tryptophan’s benefit comes with risk
In a Cochrane review of 2 RCTs, tryptophan and 5-HTP were superior to placebo in alleviating symptoms of depression. However, some published case reports have linked tryptophan use to potentially fatal eosinophilia-myalgia syndrome.8
No clear evidence for inositol or n-3 PUFAs
A Cochrane review of 4 small double-blind RCTs investigating inositol as a nutritional supplement in depression treatment failed to find clear evidence of therapeutic benefit.9
Three RCTs demonstrated significantly higher red blood cell membrane levels of n-3 PUFAs in nondepressed patients compared with depressed patients.10 However, a systematic review of 12 RCTs failed to demonstrate any benefit of n-3 PUFA supplementation over placebo in treating depressed mood.11 The authors concluded that larger trials are needed to demonstrate efficacy because of marked heterogeneity among the RCTs.
Safety issues. A recent Cochrane review found fewer adverse effects for St. John’s wort than tricyclic antidepressants.1 The most common adverse reactions were sensitivity to light, skin symptoms, gastrointestinal symptoms, and agitation. Data comparing St. John’s wort with newer antidepressants are lacking.
St. John’s wort does have pharmacokinetic interactions and should not be taken concurrently with other antidepressants, immunosuppressants, anti-HIV drugs, cou-marin-type anticoagulants, or certain antineoplastic agents.17
Reviews of meta-analyses, case reports, population studies, RCTs, and other literature have reported virtually no adverse effects for PUFAs; 18 trials investigating saffron, lavender, borage, dan zhi xiao yao, folate, SAMe, and inositol also reported no safety concerns. However, the size and duration of these studies limit their ability to detect significant problems.2,5,6,9 As previously noted, concerns exist regarding an association between tryptophan and eosinophilia-myalgia syndrome.8
Recommendations
The World Federation of Societies of Biological Psychiatry doesn’t recommend St. John’s wort for moderate to severe depression, but suggests it can be considered for treating mild to moderate depressive episodes provided the prescriber considers potential pharmacokinetic interactions with other medications and understands possible variations in purity and potency of extracts.19 The Federation also states that St. John’s wort is an alternative for patients reluctant to take traditional antidepressants.
TABLE W1
What the studies say about nutritional therapies for depression
| Supplement | Study type | Number of subjects | Comparison group | Outcome measure | Results | Conclusion | SOR |
|---|---|---|---|---|---|---|---|
| Borage (Echium amoenum) | 1 small RCT | 352,16 | Placebo | HAM-D | Improved HAM-D scores significantly at week 4 (borage 18. 3±3. 9 vs placebo 21. 9±3. 9; t=2. 51; P=. 02); no significant difference at Week 62,16 | Superior to placebo in reducing symptoms of depression | B |
| Dan zhi xiao yao | 1 small RCT | 632 | Maprotiline | HAM-D, SDS, SAS, scale for traditional Chinese medicine syndrome and symptom differentiation | 87% depression reduction (dan zhi xiao yao) vs 84% depression reduction (maprotiline) | As effective as maprotiline in treating depression | B |
| Folate | Cochrane review of 3 RCTs | 2475 | Studies 1 and 2: folate vs folate + other treatment (Study 1: low folate levels; Study 2: normal folate levels) Study 3: folate vs trazodone (normal folate levels) | HAM-D | Superior to placebo (NNT=5, defined as 50% reduction in HAM-D); comparable to trazodone (RR=0. 97; 95% CI, 0. 14-2. 01)7 | May have role as supplement to other treatments for depression Efficacy unclear in patients with normal folate levels | A |
| Inositol | Cochrane review of 4 RCTs | 1419 | Studies 1-3: placebo plus conventional antidepressants Study 4: placebo only | HAM-D, MADRS | Pooled estimate of effect of all 3 studies (SMD= -0. 08; 95% CI, -0. 45 to 0. 30) | No clear evidence of therapeutic benefit | A |
| Lavender (Lavandula angustifolia) | 1 small RCT | 454 | Imipramine | HAM-D | Imipramine plus lavender showed significant effect compared with imipramine alone (f=26. 87; Df=3. 01; P<. 0001) | Synergistic effect suggested when used with imipramine | B |
| n-3 long-chain polyunsaturated fatty acids | Systematic review including 3 RCTs; 10 meta-analysis of 12 RCTs11 | 10210 103211 | Various comparison groups included | Serum SFAs, MUFAs, PUFAs; RBC membrane levels n-3 PUFAs2 HAM-D, BDI3 | Systematic review:10 Study 1: n=30; ES=3. 61 Study 2: n=24; ES=1. 2 Study 3: n=48; ES=2. 43 Meta-analysis:11 Pooled ES=0. 13; 95% CI, 0. 01-0. 25 | Significantly higher RBC membrane levels of n-3 PUFAs in nondepressed vs depressed patients10 No significant effect for supplementation11 Larger trials with adequate power needed2,3 | A |
| S-adenosyl-methionine (SAMe) | Meta-analysis of 13 RCTs,6 systematic review including 2 RCTs7 | 3996 787 | Placebo and conventional antidepressants | HAM-D | NNT=2. 5 for HAM-D decrease of >25%; 6 NNT=6. 25 for HAM-D decrease of >50%6 | May have role in treatment of major depression Further trials are needed to address unanswered questions about absorption, mechanism of action, and bioavailability7 | A |
| Saffron (Crocus sativus) | Systematic review of 4 small RCTs, 1 later RCT | 3012 4013 4014 4015 403 | Imipramine12 Placebo13,15 Fluoxetine5,14 | HAM-D | Systematic review: Study 1: imipramine and saffron equally efficacious (f=2. 91; P=. 09)12 Study 2: Improved HAM-D scores: -12. 20±4. 67 (saffron) vs -5. 10±4. 71 (placebo) (P<. 0001)13 Study 3: Improved HAM-D scores: saffron petal -12. 00±4. 10; fluoxetine -13. 50±4. 91; difference between 2 treatments not significant (P=. 27)14 Study 4: Improved HAM-D scores: -14. 01±5. 53 (saffron petal) vs -5. 05±4. 63 (placebo) (P<. 0001)15 Study 5:5 NNT=10 | Efficacy of extract and petal suggested to treat mild to moderate depression Large-scale trials are warranted | B |
| St. John’s wort (Hypericum perforatum L. ) | Cochrane review of 29 RCTs | 54891 | SSRIs, tri/tetracyclic antidepressants, placebo | Responder rate ratio | St. John’s wort vs placebo: 9 larger trials: RR=1. 28; 95% CI, 1. 10-1. 491 9 smaller trials: RR=1. 87; 95% CI, 1. 22-2. 871 St. John’s wort vs SSRIs: 12 trials: RR=1. 00; 95% CI, 0. 90-1. 111 St. John’s wort vs tricyclics: 5 trials: RR=1. 02; 95% CI, 0. 90-1. 151 | Effective for treating mild to moderate depression | A |
| Tryptophan and 5-hydroxy-tryptophan (5-HTP) | Cochrane review of 2 RCTs | 648 | Placebo | HAM-D | NNT=2. 78 vs placebo (OR=4. 1; 95% CI, 1. 28-13. 15 | Superior to placebo Insufficient evidence regarding safety | A |
| BDI, Beck Depression Inventory; CI, confidence interval; DF, degrees of freedom; ES, effect size; F, F statistic; HAM-D, Hamilton Depression Rating Scale; MADRS, Mont-gomery-Asberg Depression Rating Scale; MUFAs, monounsaturated fatty acids; n-3 PUFAs, n-3 long-chain polyunsaturated fatty acids; NNT, number needed to treat; OR, odds ratio; PUFAs, polyunsaturated fatty acids; RBC, red blood cell; RCT, randomized controlled trial; RR, relative risk; SAS, self-rating anxiety scale; SDS, self-rating depression scale; SFAs, saturated fatty acids; SMD, standard weighted mean difference; SOR, strength of recommendation; SSRI, selective serotonin reuptake inhibitor. | |||||||
1. Linde K, Bemer MM, Kriston L. St. John’s wort for major depression. Cochrane Database Syst Rev. 2008;(4):CD000448.-
2. Sarris J. Herbal medicines in the treatment of psychiatric disorders: a systematic review. Phytother Res. 2007;21:703-716.
3. Akhondzadeh Basti A, Moshiri E, Noorbala AA, et al. Comparison of petal of Crocus sativus L. and fluoxetine in the treatment of depressed outpatients. Prog Neuropsychopharmacol Biol Psychiatry. 2006;31:439-442.
4. Akhondzadeh S, Kashani L, Fotouhi A, et al. Comparison of Lavandula angustifolia Mill. tincture and imipramine in the treatment of mild to moderate depression. Prog Neuropsychopharmacol Biol Psychiatry. 2003;27:123-127.
5. Taylor MJ, Carney S, Geddes J, et al. Folate for depressive disorders. Cochrane Database Syst Rev. 2003;(2):CD003390.-
6. Bressa GM. S-adenosyl-l-methionine (SAMe) as antidepres-sant. Acta Neurol Scand Suppl. 1994;154:7-14.
7. Williams AL, Girard C, Jui D, et al. S-adenosylmethionine (SAMe) as treatment for depression. Clin Invest Med. 2005;28:132-139.
8. Shaw K, Turner J, Del Mar C. Tryptophan and 5-hydroxy-tryptophan for depression. Cochrane Database Syst Rev. 2002;(1):CD003198.-
9. Taylor MJ, Wilder H, Bhagwager Z, et al. Inositol for depressive disorders. Cochrane Database Syst Rev. 2004;(2):CD004049.-
10. Williams AL, Katz D, Ali A, et al. Do essential fatty acids have a role in the treatment of depression? J Affect Disord. 2006;93:117-123.
11. Appleton KM, Hayward RC, Gunnell D, et al. Effects of n-3 longchain polyunsaturated fatty acids on depressed mood. Am J Clin Nutr. 2006;84:1308-1316.
12. Akhondzadeh S, Fallah-Pour H, Afkham K, et al. Comparison of Crocus sativus L. and imipramine in the treatment of mild to moderate depression. BMC Complement Altern Med. 2004;4:12.-
13. Akhondzadeh S, Tahmacebi-Pour N, Noorbala AA, et al. Crocus sativus L. in the treatment of mild to moderate depression. Phytother Res. 2005;19:148-151.
14. Noorbala AA, Akhondzadeh S, Tahmacebi-Pour N, et al. Hydro-alcoholic extract of Crocus sativus L. versus fluoxetine in the treatment of mild to moderate depression. J Ethnopharmacol. 2005;97:281-284.
15. Moshiri E, Basti AA, Noorbala AA, et al. Crocus sativus L. (petal) in the treatment of mild-to-moderate depression. Phytomedicine. 2006;13:607-611.
16. Sayyah M, Sahhah M, Kamalinejad M. A preliminary randomized double blind clinical trial on the efficacy of aqueous extract of Echium amoenum in the treatment of mild to moderate major depression. Prog Neuropsychopharmacol Biol Psychiatry. 2006;30:166-169.
17. Schulz V Safety of St. John’s wort extract compared to synthetic antidepressants. Phytomedicine. 2006;13:199-204.
18. Lee S, Gura KM, Kim S, et al. Current clinical applications of ome-ga-6 and omega-3 fatty acids. Nutr Clin Pract. 2006;21:323-341.
19. Bauer M, Bschor T, Pfennig A, et al. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for biological treatment of unipolar depressive disorders in primary care. World J Biol Psychiatry. 2007;8:67-104.
1. Linde K, Bemer MM, Kriston L. St. John’s wort for major depression. Cochrane Database Syst Rev. 2008;(4):CD000448.-
2. Sarris J. Herbal medicines in the treatment of psychiatric disorders: a systematic review. Phytother Res. 2007;21:703-716.
3. Akhondzadeh Basti A, Moshiri E, Noorbala AA, et al. Comparison of petal of Crocus sativus L. and fluoxetine in the treatment of depressed outpatients. Prog Neuropsychopharmacol Biol Psychiatry. 2006;31:439-442.
4. Akhondzadeh S, Kashani L, Fotouhi A, et al. Comparison of Lavandula angustifolia Mill. tincture and imipramine in the treatment of mild to moderate depression. Prog Neuropsychopharmacol Biol Psychiatry. 2003;27:123-127.
5. Taylor MJ, Carney S, Geddes J, et al. Folate for depressive disorders. Cochrane Database Syst Rev. 2003;(2):CD003390.-
6. Bressa GM. S-adenosyl-l-methionine (SAMe) as antidepres-sant. Acta Neurol Scand Suppl. 1994;154:7-14.
7. Williams AL, Girard C, Jui D, et al. S-adenosylmethionine (SAMe) as treatment for depression. Clin Invest Med. 2005;28:132-139.
8. Shaw K, Turner J, Del Mar C. Tryptophan and 5-hydroxy-tryptophan for depression. Cochrane Database Syst Rev. 2002;(1):CD003198.-
9. Taylor MJ, Wilder H, Bhagwager Z, et al. Inositol for depressive disorders. Cochrane Database Syst Rev. 2004;(2):CD004049.-
10. Williams AL, Katz D, Ali A, et al. Do essential fatty acids have a role in the treatment of depression? J Affect Disord. 2006;93:117-123.
11. Appleton KM, Hayward RC, Gunnell D, et al. Effects of n-3 longchain polyunsaturated fatty acids on depressed mood. Am J Clin Nutr. 2006;84:1308-1316.
12. Akhondzadeh S, Fallah-Pour H, Afkham K, et al. Comparison of Crocus sativus L. and imipramine in the treatment of mild to moderate depression. BMC Complement Altern Med. 2004;4:12.-
13. Akhondzadeh S, Tahmacebi-Pour N, Noorbala AA, et al. Crocus sativus L. in the treatment of mild to moderate depression. Phytother Res. 2005;19:148-151.
14. Noorbala AA, Akhondzadeh S, Tahmacebi-Pour N, et al. Hydro-alcoholic extract of Crocus sativus L. versus fluoxetine in the treatment of mild to moderate depression. J Ethnopharmacol. 2005;97:281-284.
15. Moshiri E, Basti AA, Noorbala AA, et al. Crocus sativus L. (petal) in the treatment of mild-to-moderate depression. Phytomedicine. 2006;13:607-611.
16. Sayyah M, Sahhah M, Kamalinejad M. A preliminary randomized double blind clinical trial on the efficacy of aqueous extract of Echium amoenum in the treatment of mild to moderate major depression. Prog Neuropsychopharmacol Biol Psychiatry. 2006;30:166-169.
17. Schulz V Safety of St. John’s wort extract compared to synthetic antidepressants. Phytomedicine. 2006;13:199-204.
18. Lee S, Gura KM, Kim S, et al. Current clinical applications of ome-ga-6 and omega-3 fatty acids. Nutr Clin Pract. 2006;21:323-341.
19. Bauer M, Bschor T, Pfennig A, et al. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for biological treatment of unipolar depressive disorders in primary care. World J Biol Psychiatry. 2007;8:67-104.
Evidence-based answers from the Family Physicians Inquiries Network
Can counseling prevent or treat postpartum depression?
No, in most cases, counseling does not prevent postpartum depression (PPD), though it can treat the disorder. Overall, psychosocial interventions don’t offer a significantly greater benefit than standard care in preventing PPD—although studies do suggest a preventive benefit when the intervention is administered postnatally, in the home, and targeted toward individual at-risk women (strength of recommendation [SOR]: A, meta-analysis of 15 randomized, controlled trials [RCTs] and 1 subsequent RCT).
Psychotherapy and counseling—including interpersonal therapy, individual and group cognitive behavioral therapy (CBT), psychodynamic therapy, and nondirective counseling—are effective in treating PPD (SOR: A, systematic review of 15 RCTs and 1 later RCT). Not enough evidence exists to compare the benefits of antidepressant medication with CBT (SOR: B, 2 low-quality RCTs).
Do some research before you refer
Patrick O. Smith, PhD
University of Mississippi Medical Center, Department of Family Medicine, Jackson
Postpartum depression negatively impacts maternal satisfaction and is a major women’s health issue. Recognizing that psychosocial interventions are considered first-line, evidence-based treatments is important, but, beyond that, knowing how to locate a licensed professional who delivers these treatments may be critical to your patient.
One way to identify such a clinician is to use a Web-based search tool such as www.findapsychologist.org, provided by the National Register of Health Service Providers in Psychology (www.nationalregister.org). Once identified, contact the clinician and ask how s/he does what s/he does. If the answer is evidence-based treatments, you may have a strong candidate for treating a woman with PPD. Just remember: A referral is as important as the care you, yourself, provide.
Evidence summary
Prevention: No overall benefit, but some approaches may help
A Cochrane meta-analysis of pooled data from 15 RCTs (7697 women) found that psychological interventions didn’t prevent PPD based on comparison of initial depression scores with scores at the conclusion of the studies (relative risk [RR]=0.81; 95% confidence interval [CI], 0.65-1.02).1 Although some studies suggested short-term benefit (N=4091; RR=0.65; 95% CI, 0.43-1.00), benefits diminished over time and weren’t noted when the definition of depression was limited to an Edinburgh Postpartum Depression Scale (EPDS) score below 12 (out of a maximum of 30). Some differences were found when the data were stratified.
Certain interventions were found to prevent depressive symptoms (defined differently in the various studies). They were: home visits provided by healthcare professionals (2 RCTs, N=1663; RR=0.68; 95% CI, 0.55-0.84), interventions targeting at-risk women (7 RCTs, N=1162; RR=0.67; 95% CI, 0.51-0.89), and interventions begun postnatally (10 RCTs, N=6379; RR=0.76; 95% CI, 0.58-0.98). Notably, the level of training of providers of psychological interventions included in the meta-analysis was highly variable.
A later RCT of a 6-session cognitive-behavioral, midwife-administered intervention in mothers of preterm infants showed no preventive benefit (N=176; RR=1.02; 95% CI, 0.87-1.20).2
Treatment: Counseling helps, especially in the near term
A recent systematic review of 5 RCTs (N=450) investigated the effectiveness of interpersonal psychotherapy, CBT (individual and group), nondirective counseling, and psychodynamic therapy in reducing PPD symptoms.3
Interpersonal therapy (12 weekly sessions) significantly reduced PPD symptoms as measured by the Hamilton Depression Rating Scale (HAM-D) compared with a wait-list control group (1 RCT, N=120, RR=2.11; 95% CI, 1.04-4.28).
Individual CBT and ideal standard care (weekly 20- to 60-minute supportive meetings) were equally effective in reducing depression scores immediately postintervention and 6 months thereafter as measured by the EPDS (1 RCT, N=37). Although a trend toward greater benefit for CBT was noted, the study was underpowered to identify a significant difference.
Nondirective counseling reduced the proportion of women with depression (N=55; RR=0.49; 95% CI, 0.26-0.95) and lowered EPDS scores (N=193; treatment effect=-2.1; 95% CI, -3.8 to -0.3; P=.02) compared with routine primary care. Individual CBT also reduced EPDS scores, when compared to routine primary care (N=55; treatment effect=-2.7; 95% CI, -4.5 to -0.9; P=.003).
Psychodynamic therapy reduced the proportion of women with major depression (N=55; RR=1.89; 95% CI, 1.33-2.33).
All of these interventions improved PPD immediately following treatment compared with routine primary care, but the benefits were not sustained at long-term follow-up (6 months). Study limitations included failure to control for multiple comparisons, pretreatment group differences, differential attrition among groups, and lack of sufficient power.
A later RCT (N=121) also found psychological interventions (group CBT and group and individual counseling) to be superior to routine primary care, with individual counseling yielding the greatest improvement in PPD symptoms (P<.05).4
Antidepressants vs CBT: Too little information
Two RCTs compared antidepressant medications to CBT.3 In the first (N=87), fluoxetine and placebo were each paired with 1 or 6 CBT sessions. After 12 weeks of treatment, fluoxetine was superior to placebo as measured by mean symptom score reduction on the HAM-D, EPDS, and clinical interview schedule; 6 CBT sessions were superior to a single session as measured by mean symptom score reduction on the Hamilton Depression Scale and clinical interview schedule.5 No significant interaction effect was found.
The authors reported “highly significant” improvements, but didn’t specify significance level or provide adequate information to calculate number needed to treat. Interpretation of the findings is limited by methodologic weaknesses, high withdrawal rate, and exclusion of breast-feeding women.3
A second, small RCT (N=35) compared 12 weeks of paroxetine with a combination of paroxetine and CBT.6 Significant improvements—defined as percentage of patients in each group demonstrating at least a 50% score reduction on the HAM-D (paroxetine, 87.5%; combination, 78.9%) and EPDS (paroxetine, 61.5%; combination, 58.3%)—occurred in both groups (P<.01), but no difference was found between the groups. The study didn’t include a placebo control group.
Recommendations
The National Collaborating Centre for Women’s and Children’s Health recommends against offering educational interventions to pregnant women because such interventions haven’t been found to reduce PPD.7
The Scottish Intercollegiate Guidelines Network recommends “postnatal visits, interpersonal therapy, and/or antenatal preparation” to prevent PPD. To treat PPD, they recommend psychosocial interventions, preferably those that include more than 1 family member.8
1. Dennis CL, Creedy D. Psychosocial and psychological interventions for preventing postpartum depression. Cochrane Database Syst Rev. 2007;(4):CD001134.
2. Hagan R, Evans SF, Pope S. Preventing postnatal depression in mothers of very preterm infants: a randomized controlled trial. Br J Obstet Gynaecol. 2004;11:641-647.
3. Howard L. Postnatal depression. BMJ Clin Evid. 2006;14:1919-1931.
4. Milgrom J, Negri LM, Gemmill AW, et al. A randomized controlled trial of psychological interventions for postnatal depression. Br J Clin Psychol. 2005;44:529-542.
5. Appleby L, Warner R, Whitton A, et al. A controlled study of fluoxetine and cognitive-behavioral counseling in the treatment of postnatal depression. BMJ. 1997;314:932-936.
6. Misri S, Reebye P, Corral M, et al. The use of paroxetine and cognitive-behavioral therapy in postpartum depression and anxiety: a randomized controlled trial. J Clin Psychiatry. 2004;65:1236-1241.
7. National Collaborating Centre for Women’s and Children’s Health. Antenatal Care: Routine Care for the Healthy Pregnant Woman. London: RCOG Press; 2003. Available at: www.rcog.org.uk/resources/Public/pdf/Antenatal_Care.pdf. Accessed November 11, 2007.
8. Scottish Intercollegiate Guidelines Network. Postnatal Depression and Puerperal Psychosis. A National Clinical Guideline. Edinburgh, Scotland: Scottish Intercollegiate Guidelines Network (SIGN); 2002. Available at: www.sign.ac.uk/guidelines/fulltext/60/index.html. Accessed November 11, 2007.
No, in most cases, counseling does not prevent postpartum depression (PPD), though it can treat the disorder. Overall, psychosocial interventions don’t offer a significantly greater benefit than standard care in preventing PPD—although studies do suggest a preventive benefit when the intervention is administered postnatally, in the home, and targeted toward individual at-risk women (strength of recommendation [SOR]: A, meta-analysis of 15 randomized, controlled trials [RCTs] and 1 subsequent RCT).
Psychotherapy and counseling—including interpersonal therapy, individual and group cognitive behavioral therapy (CBT), psychodynamic therapy, and nondirective counseling—are effective in treating PPD (SOR: A, systematic review of 15 RCTs and 1 later RCT). Not enough evidence exists to compare the benefits of antidepressant medication with CBT (SOR: B, 2 low-quality RCTs).
Do some research before you refer
Patrick O. Smith, PhD
University of Mississippi Medical Center, Department of Family Medicine, Jackson
Postpartum depression negatively impacts maternal satisfaction and is a major women’s health issue. Recognizing that psychosocial interventions are considered first-line, evidence-based treatments is important, but, beyond that, knowing how to locate a licensed professional who delivers these treatments may be critical to your patient.
One way to identify such a clinician is to use a Web-based search tool such as www.findapsychologist.org, provided by the National Register of Health Service Providers in Psychology (www.nationalregister.org). Once identified, contact the clinician and ask how s/he does what s/he does. If the answer is evidence-based treatments, you may have a strong candidate for treating a woman with PPD. Just remember: A referral is as important as the care you, yourself, provide.
Evidence summary
Prevention: No overall benefit, but some approaches may help
A Cochrane meta-analysis of pooled data from 15 RCTs (7697 women) found that psychological interventions didn’t prevent PPD based on comparison of initial depression scores with scores at the conclusion of the studies (relative risk [RR]=0.81; 95% confidence interval [CI], 0.65-1.02).1 Although some studies suggested short-term benefit (N=4091; RR=0.65; 95% CI, 0.43-1.00), benefits diminished over time and weren’t noted when the definition of depression was limited to an Edinburgh Postpartum Depression Scale (EPDS) score below 12 (out of a maximum of 30). Some differences were found when the data were stratified.
Certain interventions were found to prevent depressive symptoms (defined differently in the various studies). They were: home visits provided by healthcare professionals (2 RCTs, N=1663; RR=0.68; 95% CI, 0.55-0.84), interventions targeting at-risk women (7 RCTs, N=1162; RR=0.67; 95% CI, 0.51-0.89), and interventions begun postnatally (10 RCTs, N=6379; RR=0.76; 95% CI, 0.58-0.98). Notably, the level of training of providers of psychological interventions included in the meta-analysis was highly variable.
A later RCT of a 6-session cognitive-behavioral, midwife-administered intervention in mothers of preterm infants showed no preventive benefit (N=176; RR=1.02; 95% CI, 0.87-1.20).2
Treatment: Counseling helps, especially in the near term
A recent systematic review of 5 RCTs (N=450) investigated the effectiveness of interpersonal psychotherapy, CBT (individual and group), nondirective counseling, and psychodynamic therapy in reducing PPD symptoms.3
Interpersonal therapy (12 weekly sessions) significantly reduced PPD symptoms as measured by the Hamilton Depression Rating Scale (HAM-D) compared with a wait-list control group (1 RCT, N=120, RR=2.11; 95% CI, 1.04-4.28).
Individual CBT and ideal standard care (weekly 20- to 60-minute supportive meetings) were equally effective in reducing depression scores immediately postintervention and 6 months thereafter as measured by the EPDS (1 RCT, N=37). Although a trend toward greater benefit for CBT was noted, the study was underpowered to identify a significant difference.
Nondirective counseling reduced the proportion of women with depression (N=55; RR=0.49; 95% CI, 0.26-0.95) and lowered EPDS scores (N=193; treatment effect=-2.1; 95% CI, -3.8 to -0.3; P=.02) compared with routine primary care. Individual CBT also reduced EPDS scores, when compared to routine primary care (N=55; treatment effect=-2.7; 95% CI, -4.5 to -0.9; P=.003).
Psychodynamic therapy reduced the proportion of women with major depression (N=55; RR=1.89; 95% CI, 1.33-2.33).
All of these interventions improved PPD immediately following treatment compared with routine primary care, but the benefits were not sustained at long-term follow-up (6 months). Study limitations included failure to control for multiple comparisons, pretreatment group differences, differential attrition among groups, and lack of sufficient power.
A later RCT (N=121) also found psychological interventions (group CBT and group and individual counseling) to be superior to routine primary care, with individual counseling yielding the greatest improvement in PPD symptoms (P<.05).4
Antidepressants vs CBT: Too little information
Two RCTs compared antidepressant medications to CBT.3 In the first (N=87), fluoxetine and placebo were each paired with 1 or 6 CBT sessions. After 12 weeks of treatment, fluoxetine was superior to placebo as measured by mean symptom score reduction on the HAM-D, EPDS, and clinical interview schedule; 6 CBT sessions were superior to a single session as measured by mean symptom score reduction on the Hamilton Depression Scale and clinical interview schedule.5 No significant interaction effect was found.
The authors reported “highly significant” improvements, but didn’t specify significance level or provide adequate information to calculate number needed to treat. Interpretation of the findings is limited by methodologic weaknesses, high withdrawal rate, and exclusion of breast-feeding women.3
A second, small RCT (N=35) compared 12 weeks of paroxetine with a combination of paroxetine and CBT.6 Significant improvements—defined as percentage of patients in each group demonstrating at least a 50% score reduction on the HAM-D (paroxetine, 87.5%; combination, 78.9%) and EPDS (paroxetine, 61.5%; combination, 58.3%)—occurred in both groups (P<.01), but no difference was found between the groups. The study didn’t include a placebo control group.
Recommendations
The National Collaborating Centre for Women’s and Children’s Health recommends against offering educational interventions to pregnant women because such interventions haven’t been found to reduce PPD.7
The Scottish Intercollegiate Guidelines Network recommends “postnatal visits, interpersonal therapy, and/or antenatal preparation” to prevent PPD. To treat PPD, they recommend psychosocial interventions, preferably those that include more than 1 family member.8
No, in most cases, counseling does not prevent postpartum depression (PPD), though it can treat the disorder. Overall, psychosocial interventions don’t offer a significantly greater benefit than standard care in preventing PPD—although studies do suggest a preventive benefit when the intervention is administered postnatally, in the home, and targeted toward individual at-risk women (strength of recommendation [SOR]: A, meta-analysis of 15 randomized, controlled trials [RCTs] and 1 subsequent RCT).
Psychotherapy and counseling—including interpersonal therapy, individual and group cognitive behavioral therapy (CBT), psychodynamic therapy, and nondirective counseling—are effective in treating PPD (SOR: A, systematic review of 15 RCTs and 1 later RCT). Not enough evidence exists to compare the benefits of antidepressant medication with CBT (SOR: B, 2 low-quality RCTs).
Do some research before you refer
Patrick O. Smith, PhD
University of Mississippi Medical Center, Department of Family Medicine, Jackson
Postpartum depression negatively impacts maternal satisfaction and is a major women’s health issue. Recognizing that psychosocial interventions are considered first-line, evidence-based treatments is important, but, beyond that, knowing how to locate a licensed professional who delivers these treatments may be critical to your patient.
One way to identify such a clinician is to use a Web-based search tool such as www.findapsychologist.org, provided by the National Register of Health Service Providers in Psychology (www.nationalregister.org). Once identified, contact the clinician and ask how s/he does what s/he does. If the answer is evidence-based treatments, you may have a strong candidate for treating a woman with PPD. Just remember: A referral is as important as the care you, yourself, provide.
Evidence summary
Prevention: No overall benefit, but some approaches may help
A Cochrane meta-analysis of pooled data from 15 RCTs (7697 women) found that psychological interventions didn’t prevent PPD based on comparison of initial depression scores with scores at the conclusion of the studies (relative risk [RR]=0.81; 95% confidence interval [CI], 0.65-1.02).1 Although some studies suggested short-term benefit (N=4091; RR=0.65; 95% CI, 0.43-1.00), benefits diminished over time and weren’t noted when the definition of depression was limited to an Edinburgh Postpartum Depression Scale (EPDS) score below 12 (out of a maximum of 30). Some differences were found when the data were stratified.
Certain interventions were found to prevent depressive symptoms (defined differently in the various studies). They were: home visits provided by healthcare professionals (2 RCTs, N=1663; RR=0.68; 95% CI, 0.55-0.84), interventions targeting at-risk women (7 RCTs, N=1162; RR=0.67; 95% CI, 0.51-0.89), and interventions begun postnatally (10 RCTs, N=6379; RR=0.76; 95% CI, 0.58-0.98). Notably, the level of training of providers of psychological interventions included in the meta-analysis was highly variable.
A later RCT of a 6-session cognitive-behavioral, midwife-administered intervention in mothers of preterm infants showed no preventive benefit (N=176; RR=1.02; 95% CI, 0.87-1.20).2
Treatment: Counseling helps, especially in the near term
A recent systematic review of 5 RCTs (N=450) investigated the effectiveness of interpersonal psychotherapy, CBT (individual and group), nondirective counseling, and psychodynamic therapy in reducing PPD symptoms.3
Interpersonal therapy (12 weekly sessions) significantly reduced PPD symptoms as measured by the Hamilton Depression Rating Scale (HAM-D) compared with a wait-list control group (1 RCT, N=120, RR=2.11; 95% CI, 1.04-4.28).
Individual CBT and ideal standard care (weekly 20- to 60-minute supportive meetings) were equally effective in reducing depression scores immediately postintervention and 6 months thereafter as measured by the EPDS (1 RCT, N=37). Although a trend toward greater benefit for CBT was noted, the study was underpowered to identify a significant difference.
Nondirective counseling reduced the proportion of women with depression (N=55; RR=0.49; 95% CI, 0.26-0.95) and lowered EPDS scores (N=193; treatment effect=-2.1; 95% CI, -3.8 to -0.3; P=.02) compared with routine primary care. Individual CBT also reduced EPDS scores, when compared to routine primary care (N=55; treatment effect=-2.7; 95% CI, -4.5 to -0.9; P=.003).
Psychodynamic therapy reduced the proportion of women with major depression (N=55; RR=1.89; 95% CI, 1.33-2.33).
All of these interventions improved PPD immediately following treatment compared with routine primary care, but the benefits were not sustained at long-term follow-up (6 months). Study limitations included failure to control for multiple comparisons, pretreatment group differences, differential attrition among groups, and lack of sufficient power.
A later RCT (N=121) also found psychological interventions (group CBT and group and individual counseling) to be superior to routine primary care, with individual counseling yielding the greatest improvement in PPD symptoms (P<.05).4
Antidepressants vs CBT: Too little information
Two RCTs compared antidepressant medications to CBT.3 In the first (N=87), fluoxetine and placebo were each paired with 1 or 6 CBT sessions. After 12 weeks of treatment, fluoxetine was superior to placebo as measured by mean symptom score reduction on the HAM-D, EPDS, and clinical interview schedule; 6 CBT sessions were superior to a single session as measured by mean symptom score reduction on the Hamilton Depression Scale and clinical interview schedule.5 No significant interaction effect was found.
The authors reported “highly significant” improvements, but didn’t specify significance level or provide adequate information to calculate number needed to treat. Interpretation of the findings is limited by methodologic weaknesses, high withdrawal rate, and exclusion of breast-feeding women.3
A second, small RCT (N=35) compared 12 weeks of paroxetine with a combination of paroxetine and CBT.6 Significant improvements—defined as percentage of patients in each group demonstrating at least a 50% score reduction on the HAM-D (paroxetine, 87.5%; combination, 78.9%) and EPDS (paroxetine, 61.5%; combination, 58.3%)—occurred in both groups (P<.01), but no difference was found between the groups. The study didn’t include a placebo control group.
Recommendations
The National Collaborating Centre for Women’s and Children’s Health recommends against offering educational interventions to pregnant women because such interventions haven’t been found to reduce PPD.7
The Scottish Intercollegiate Guidelines Network recommends “postnatal visits, interpersonal therapy, and/or antenatal preparation” to prevent PPD. To treat PPD, they recommend psychosocial interventions, preferably those that include more than 1 family member.8
1. Dennis CL, Creedy D. Psychosocial and psychological interventions for preventing postpartum depression. Cochrane Database Syst Rev. 2007;(4):CD001134.
2. Hagan R, Evans SF, Pope S. Preventing postnatal depression in mothers of very preterm infants: a randomized controlled trial. Br J Obstet Gynaecol. 2004;11:641-647.
3. Howard L. Postnatal depression. BMJ Clin Evid. 2006;14:1919-1931.
4. Milgrom J, Negri LM, Gemmill AW, et al. A randomized controlled trial of psychological interventions for postnatal depression. Br J Clin Psychol. 2005;44:529-542.
5. Appleby L, Warner R, Whitton A, et al. A controlled study of fluoxetine and cognitive-behavioral counseling in the treatment of postnatal depression. BMJ. 1997;314:932-936.
6. Misri S, Reebye P, Corral M, et al. The use of paroxetine and cognitive-behavioral therapy in postpartum depression and anxiety: a randomized controlled trial. J Clin Psychiatry. 2004;65:1236-1241.
7. National Collaborating Centre for Women’s and Children’s Health. Antenatal Care: Routine Care for the Healthy Pregnant Woman. London: RCOG Press; 2003. Available at: www.rcog.org.uk/resources/Public/pdf/Antenatal_Care.pdf. Accessed November 11, 2007.
8. Scottish Intercollegiate Guidelines Network. Postnatal Depression and Puerperal Psychosis. A National Clinical Guideline. Edinburgh, Scotland: Scottish Intercollegiate Guidelines Network (SIGN); 2002. Available at: www.sign.ac.uk/guidelines/fulltext/60/index.html. Accessed November 11, 2007.
1. Dennis CL, Creedy D. Psychosocial and psychological interventions for preventing postpartum depression. Cochrane Database Syst Rev. 2007;(4):CD001134.
2. Hagan R, Evans SF, Pope S. Preventing postnatal depression in mothers of very preterm infants: a randomized controlled trial. Br J Obstet Gynaecol. 2004;11:641-647.
3. Howard L. Postnatal depression. BMJ Clin Evid. 2006;14:1919-1931.
4. Milgrom J, Negri LM, Gemmill AW, et al. A randomized controlled trial of psychological interventions for postnatal depression. Br J Clin Psychol. 2005;44:529-542.
5. Appleby L, Warner R, Whitton A, et al. A controlled study of fluoxetine and cognitive-behavioral counseling in the treatment of postnatal depression. BMJ. 1997;314:932-936.
6. Misri S, Reebye P, Corral M, et al. The use of paroxetine and cognitive-behavioral therapy in postpartum depression and anxiety: a randomized controlled trial. J Clin Psychiatry. 2004;65:1236-1241.
7. National Collaborating Centre for Women’s and Children’s Health. Antenatal Care: Routine Care for the Healthy Pregnant Woman. London: RCOG Press; 2003. Available at: www.rcog.org.uk/resources/Public/pdf/Antenatal_Care.pdf. Accessed November 11, 2007.
8. Scottish Intercollegiate Guidelines Network. Postnatal Depression and Puerperal Psychosis. A National Clinical Guideline. Edinburgh, Scotland: Scottish Intercollegiate Guidelines Network (SIGN); 2002. Available at: www.sign.ac.uk/guidelines/fulltext/60/index.html. Accessed November 11, 2007.
Evidence-based answers from the Family Physicians Inquiries Network
How does VTE risk for the patch and vaginal ring compare with oral contraceptives?
Evidence is conflicting with regard to the comparative frequency of venous thrombolic events (VTE) among women using the transdermal patch when compared to an oral contraceptive (OC), even though the patch produces a relatively high serum ethinyl estradiol (EE) level (strength of recommendation [SOR]: C, conflicting cohort case-control studies).
The vaginal ring has a risk of VTE comparable to that of an OC (SOR: B, 1 comparative study).
For now, base decisions on patient preference
Richard Williams, MD
University of Nevada School of Medicine, Reno
This review points out that we don’t have enough evidence to make a strong recommendation about oral or nonoral estrogen-containing contraceptives based on the risk of thromboembolic disease. All estrogen-containing contraceptives have similar side-effect profiles, regardless of the route of administration.
In my experience, the patch or ring appeals to women who have had difficulty with OCs and need a simpler dosing regimen to improve compliance. The choice between an oral estrogen-containing contraceptive and the patch or ring should be based on the patient’s preference, not the risk of thromboembolic disease, until we have evidence to suggest otherwise.
Evidence summary
Two nonoral estrogen-progestin contraceptives have been approved by the US Food and Drug Administration (FDA). OrthoEvra is a transdermal patch applied weekly for 3 consecutive weeks, followed by 1 patch-free week per cycle.1 The NuvaRing is a vaginal ring worn for 3 consecutive weeks in a 4-week cycle.2
The patch causes greater estrogen exposure than OCs or the ring
In November 2005, the FDA issued an update to the labeling of the OrthoEvra contraceptive patch, reporting increased systemic estrogen exposure, which may increase the risk of blood clots.3 The FDA warned that the transdermal patch exposes the user to 60% more estrogen than the typical birth control pill containing 35 μg EE.3 In January 2008, the FDA approved an additional update to include the results of a new study that found users of the patch to be at higher risk of developing VTE than OC users.3,4
One pharmacokinetic study found that exposure to EE differed among delivery systems. The area under the EE concentration-vs-time curve in the patch group was 1.6 times higher than in the OC group (P<.05) and 3.4 times higher than in the vaginal ring group (P<.05).2
So what’s the VTE risk? Two studies, contrasting conclusions
A nested case-control study—based on a Phar-Metrics longitudinal database of information from paid claims by managed care health plans—included 215,769 women between the ages of 15 and 44 years who had started using the patch or a norgestimate-EE combination OC since April 1, 2002, when OrthoEvra was first introduced on the US market.5 Investigators identified 68 diagnosed cases of VTE with no identifiable risk factors.
The overall incidence of VTE in this study was 52.8 per 100,000 women-years (95% confidence interval [CI], 35.8-74.9) among patch users and 41.8 per 100,000 women-years among OC users (95% CI, 29.4-57.6).5 The study concluded that the risk of nonfatal VTE for the patch isn’t higher than the risk for an OC containing 35 μg EE and norgestimate (odds ratio [OR]=0.9; 95% CI, 0.5-1.6; incidence rate ratio [IRR]=1.1; 95% CI, 0.7-1.8).
A recent update to the study evaluated an additional 17 months of data on new cases of women meeting the same criteria. The supplementary results proved consistent with earlier conclusions, indicating that the risk of nonfatal VTE for the patch is similar to the risk for the OC (OR=1.1; 95% CI, 0.6-2.1).6 Combined data from the original study and the update show that the OR for VTE is 1.0 (95% CI, 0.7-1.5) in users of the patch compared with users of the OC.6
Another nested case-control study—based on UnitedHealthcare insurance claims data and confirmatory chart reviews—showed contrasting results. The study included 340,377 women between the ages of 15 and 44 years who were new users of the patch or new and previous users of a norgestimate-EE combination OC from April 1, 2002 through December 31, 2004.3 Investigators verified 57 diagnoses of VTE, controlling for confounding factors. The incidence of VTE in this study was 40.8 per 100,000 women-years among patch users and 18.3 per 100,000 women-years among users of the norgestimate-35 μg EE OC. The study reported a more than 2-fold increased risk of VTE in patch users compared to OC users (OR=2.4; 95% CI, 1.1-5.5; IRR=2.2; 95% CI, 1.3-3.8).3,7
Do the differences between studies make a difference?
The 2 studies appear similar in design but have 2 major identifiable differences:
- The first study verified VTE diagnoses by claims for systemic anticoagulants, whereas the second study expanded its analysis by performing confirmatory chart reviews for VTE diagnoses.
- The first study included only new OC and patch users as of April 1, 2002, whereas the second study included new and experienced users of the OC as of April 1, 2002.
The significance of the differences in these studies is debatable; the results have yielded controversial, conflicting evidence.
Safety and tolerability are similar for the vaginal ring and OCs
A 1-year, open-label, randomized Phase III study of 1030 women compared the NuvaRing with a combination OC containing levonorgestrel and 30 μg EE. One case of deep venous thrombosis occurred in the NuvaRing group.
In reviewing the data, the authors concluded that the NuvaRing demonstrated comparable safety and tolerability to the OC.8 NuvaRing users experienced similar side effects compared with OC users.9
Recommendations
The World Health Organization Medical Eligibility Criteria for Contraceptive Use (WHOMEC) reports that long-term safety data for the estrogen-progestin contraceptive patch are not available.10 However, the limited studies that are available suggest a safety profile similar to that of combination OCs with comparable hormone formulations.
WHOMEC suggests that the guidelines for combination OCs also should apply to the patch and the ring. Women shouldn’t use these contraceptive methods if they have a history of VTE or current VTE or if they are undergoing major surgery that may include prolonged immobilization.10
1. Abrams LS, Skee D, Natarajan J, Wong FA. Pharmacokinetic overview of Ortho Evra/Evra. Fertil Steril. 2002;77(2 suppl):S3-S12.
2. van den Heuvel MW, van Bragt AJ, Alnabawy AK, Kaptein MC. Comparison of ethinylestradiol pharmacokinetics in three hormonal contraceptive formulations: the vaginal ring, the transdermal patch and an oral contraceptive. Contraception. 2005;72:168-174.
3. US Food and Drug Administration Center for Drug Evaluation and Research. Ortho Evra (norelgestromin/ethinyl estradiol) Information. Available at: www.fda.gov/cder/drug/infopage/orthoevra/default.htm. Accessed July 5, 2008.
4. Cole JA, Norman H, Doherty M, Walker AM. Venous thromboembolism, myocardial infarction, and stroke among transdermal contraceptive system users. Obstet Gynecol. 2007;109(2 Pt 1):339-346.
5. Jick SS, Kaye JA, Russmann S, Jick H. Risk of nonfatal venous thromboembolism in women using a contraceptive transdermal patch and oral contraceptives containing norgestimate and 35 mcg of ethinyl estradiol. Contraception. 2006;73:223-228.
6. Jick S, Kaye JA, Li L, Jick H. Further results on the risk of nonfatal venous thromboembolism in users of the contraceptive transdermal patch compared to users of oral contraceptives containing norgestimate and 35 mcg of ethinyl estradiol. Contraception. 2007;76:4-7.
7. Burkman RT. Transdermal contraceptive patch. In: Rose BD, ed. UpToDate [online database]. Waltham, Mass: UpToDate; 2008.
8. Oddsson K, Leifels-Fischer B, de Melo NR, et al. Efficacy and safety of a contraceptive vaginal ring (NuvaRing) compared with a combined oral contraceptive: a 1-year randomized trial. Contraception. 2005;71:176-182.
9. Gaffield ME, Curtis KM, Mohllajee AP, Peterson HB. Medical eligibility criteria for new contraceptive methods: combined hormonal patch, combined hormonal vaginal ring and the etonogestrel implant. Contraception. 2006;73:134-144.
10. Reproductive Health and Research World Health Organization. Medical Eligibility Criteria for Contraceptive Use. 3rd ed. Geneva, Switzerland: World Health Organization; 2004. Available at: www.who.int/reproductive-health/publications/mec/mec.pdf. Accessed September 12, 2008.
Evidence is conflicting with regard to the comparative frequency of venous thrombolic events (VTE) among women using the transdermal patch when compared to an oral contraceptive (OC), even though the patch produces a relatively high serum ethinyl estradiol (EE) level (strength of recommendation [SOR]: C, conflicting cohort case-control studies).
The vaginal ring has a risk of VTE comparable to that of an OC (SOR: B, 1 comparative study).
For now, base decisions on patient preference
Richard Williams, MD
University of Nevada School of Medicine, Reno
This review points out that we don’t have enough evidence to make a strong recommendation about oral or nonoral estrogen-containing contraceptives based on the risk of thromboembolic disease. All estrogen-containing contraceptives have similar side-effect profiles, regardless of the route of administration.
In my experience, the patch or ring appeals to women who have had difficulty with OCs and need a simpler dosing regimen to improve compliance. The choice between an oral estrogen-containing contraceptive and the patch or ring should be based on the patient’s preference, not the risk of thromboembolic disease, until we have evidence to suggest otherwise.
Evidence summary
Two nonoral estrogen-progestin contraceptives have been approved by the US Food and Drug Administration (FDA). OrthoEvra is a transdermal patch applied weekly for 3 consecutive weeks, followed by 1 patch-free week per cycle.1 The NuvaRing is a vaginal ring worn for 3 consecutive weeks in a 4-week cycle.2
The patch causes greater estrogen exposure than OCs or the ring
In November 2005, the FDA issued an update to the labeling of the OrthoEvra contraceptive patch, reporting increased systemic estrogen exposure, which may increase the risk of blood clots.3 The FDA warned that the transdermal patch exposes the user to 60% more estrogen than the typical birth control pill containing 35 μg EE.3 In January 2008, the FDA approved an additional update to include the results of a new study that found users of the patch to be at higher risk of developing VTE than OC users.3,4
One pharmacokinetic study found that exposure to EE differed among delivery systems. The area under the EE concentration-vs-time curve in the patch group was 1.6 times higher than in the OC group (P<.05) and 3.4 times higher than in the vaginal ring group (P<.05).2
So what’s the VTE risk? Two studies, contrasting conclusions
A nested case-control study—based on a Phar-Metrics longitudinal database of information from paid claims by managed care health plans—included 215,769 women between the ages of 15 and 44 years who had started using the patch or a norgestimate-EE combination OC since April 1, 2002, when OrthoEvra was first introduced on the US market.5 Investigators identified 68 diagnosed cases of VTE with no identifiable risk factors.
The overall incidence of VTE in this study was 52.8 per 100,000 women-years (95% confidence interval [CI], 35.8-74.9) among patch users and 41.8 per 100,000 women-years among OC users (95% CI, 29.4-57.6).5 The study concluded that the risk of nonfatal VTE for the patch isn’t higher than the risk for an OC containing 35 μg EE and norgestimate (odds ratio [OR]=0.9; 95% CI, 0.5-1.6; incidence rate ratio [IRR]=1.1; 95% CI, 0.7-1.8).
A recent update to the study evaluated an additional 17 months of data on new cases of women meeting the same criteria. The supplementary results proved consistent with earlier conclusions, indicating that the risk of nonfatal VTE for the patch is similar to the risk for the OC (OR=1.1; 95% CI, 0.6-2.1).6 Combined data from the original study and the update show that the OR for VTE is 1.0 (95% CI, 0.7-1.5) in users of the patch compared with users of the OC.6
Another nested case-control study—based on UnitedHealthcare insurance claims data and confirmatory chart reviews—showed contrasting results. The study included 340,377 women between the ages of 15 and 44 years who were new users of the patch or new and previous users of a norgestimate-EE combination OC from April 1, 2002 through December 31, 2004.3 Investigators verified 57 diagnoses of VTE, controlling for confounding factors. The incidence of VTE in this study was 40.8 per 100,000 women-years among patch users and 18.3 per 100,000 women-years among users of the norgestimate-35 μg EE OC. The study reported a more than 2-fold increased risk of VTE in patch users compared to OC users (OR=2.4; 95% CI, 1.1-5.5; IRR=2.2; 95% CI, 1.3-3.8).3,7
Do the differences between studies make a difference?
The 2 studies appear similar in design but have 2 major identifiable differences:
- The first study verified VTE diagnoses by claims for systemic anticoagulants, whereas the second study expanded its analysis by performing confirmatory chart reviews for VTE diagnoses.
- The first study included only new OC and patch users as of April 1, 2002, whereas the second study included new and experienced users of the OC as of April 1, 2002.
The significance of the differences in these studies is debatable; the results have yielded controversial, conflicting evidence.
Safety and tolerability are similar for the vaginal ring and OCs
A 1-year, open-label, randomized Phase III study of 1030 women compared the NuvaRing with a combination OC containing levonorgestrel and 30 μg EE. One case of deep venous thrombosis occurred in the NuvaRing group.
In reviewing the data, the authors concluded that the NuvaRing demonstrated comparable safety and tolerability to the OC.8 NuvaRing users experienced similar side effects compared with OC users.9
Recommendations
The World Health Organization Medical Eligibility Criteria for Contraceptive Use (WHOMEC) reports that long-term safety data for the estrogen-progestin contraceptive patch are not available.10 However, the limited studies that are available suggest a safety profile similar to that of combination OCs with comparable hormone formulations.
WHOMEC suggests that the guidelines for combination OCs also should apply to the patch and the ring. Women shouldn’t use these contraceptive methods if they have a history of VTE or current VTE or if they are undergoing major surgery that may include prolonged immobilization.10
Evidence is conflicting with regard to the comparative frequency of venous thrombolic events (VTE) among women using the transdermal patch when compared to an oral contraceptive (OC), even though the patch produces a relatively high serum ethinyl estradiol (EE) level (strength of recommendation [SOR]: C, conflicting cohort case-control studies).
The vaginal ring has a risk of VTE comparable to that of an OC (SOR: B, 1 comparative study).
For now, base decisions on patient preference
Richard Williams, MD
University of Nevada School of Medicine, Reno
This review points out that we don’t have enough evidence to make a strong recommendation about oral or nonoral estrogen-containing contraceptives based on the risk of thromboembolic disease. All estrogen-containing contraceptives have similar side-effect profiles, regardless of the route of administration.
In my experience, the patch or ring appeals to women who have had difficulty with OCs and need a simpler dosing regimen to improve compliance. The choice between an oral estrogen-containing contraceptive and the patch or ring should be based on the patient’s preference, not the risk of thromboembolic disease, until we have evidence to suggest otherwise.
Evidence summary
Two nonoral estrogen-progestin contraceptives have been approved by the US Food and Drug Administration (FDA). OrthoEvra is a transdermal patch applied weekly for 3 consecutive weeks, followed by 1 patch-free week per cycle.1 The NuvaRing is a vaginal ring worn for 3 consecutive weeks in a 4-week cycle.2
The patch causes greater estrogen exposure than OCs or the ring
In November 2005, the FDA issued an update to the labeling of the OrthoEvra contraceptive patch, reporting increased systemic estrogen exposure, which may increase the risk of blood clots.3 The FDA warned that the transdermal patch exposes the user to 60% more estrogen than the typical birth control pill containing 35 μg EE.3 In January 2008, the FDA approved an additional update to include the results of a new study that found users of the patch to be at higher risk of developing VTE than OC users.3,4
One pharmacokinetic study found that exposure to EE differed among delivery systems. The area under the EE concentration-vs-time curve in the patch group was 1.6 times higher than in the OC group (P<.05) and 3.4 times higher than in the vaginal ring group (P<.05).2
So what’s the VTE risk? Two studies, contrasting conclusions
A nested case-control study—based on a Phar-Metrics longitudinal database of information from paid claims by managed care health plans—included 215,769 women between the ages of 15 and 44 years who had started using the patch or a norgestimate-EE combination OC since April 1, 2002, when OrthoEvra was first introduced on the US market.5 Investigators identified 68 diagnosed cases of VTE with no identifiable risk factors.
The overall incidence of VTE in this study was 52.8 per 100,000 women-years (95% confidence interval [CI], 35.8-74.9) among patch users and 41.8 per 100,000 women-years among OC users (95% CI, 29.4-57.6).5 The study concluded that the risk of nonfatal VTE for the patch isn’t higher than the risk for an OC containing 35 μg EE and norgestimate (odds ratio [OR]=0.9; 95% CI, 0.5-1.6; incidence rate ratio [IRR]=1.1; 95% CI, 0.7-1.8).
A recent update to the study evaluated an additional 17 months of data on new cases of women meeting the same criteria. The supplementary results proved consistent with earlier conclusions, indicating that the risk of nonfatal VTE for the patch is similar to the risk for the OC (OR=1.1; 95% CI, 0.6-2.1).6 Combined data from the original study and the update show that the OR for VTE is 1.0 (95% CI, 0.7-1.5) in users of the patch compared with users of the OC.6
Another nested case-control study—based on UnitedHealthcare insurance claims data and confirmatory chart reviews—showed contrasting results. The study included 340,377 women between the ages of 15 and 44 years who were new users of the patch or new and previous users of a norgestimate-EE combination OC from April 1, 2002 through December 31, 2004.3 Investigators verified 57 diagnoses of VTE, controlling for confounding factors. The incidence of VTE in this study was 40.8 per 100,000 women-years among patch users and 18.3 per 100,000 women-years among users of the norgestimate-35 μg EE OC. The study reported a more than 2-fold increased risk of VTE in patch users compared to OC users (OR=2.4; 95% CI, 1.1-5.5; IRR=2.2; 95% CI, 1.3-3.8).3,7
Do the differences between studies make a difference?
The 2 studies appear similar in design but have 2 major identifiable differences:
- The first study verified VTE diagnoses by claims for systemic anticoagulants, whereas the second study expanded its analysis by performing confirmatory chart reviews for VTE diagnoses.
- The first study included only new OC and patch users as of April 1, 2002, whereas the second study included new and experienced users of the OC as of April 1, 2002.
The significance of the differences in these studies is debatable; the results have yielded controversial, conflicting evidence.
Safety and tolerability are similar for the vaginal ring and OCs
A 1-year, open-label, randomized Phase III study of 1030 women compared the NuvaRing with a combination OC containing levonorgestrel and 30 μg EE. One case of deep venous thrombosis occurred in the NuvaRing group.
In reviewing the data, the authors concluded that the NuvaRing demonstrated comparable safety and tolerability to the OC.8 NuvaRing users experienced similar side effects compared with OC users.9
Recommendations
The World Health Organization Medical Eligibility Criteria for Contraceptive Use (WHOMEC) reports that long-term safety data for the estrogen-progestin contraceptive patch are not available.10 However, the limited studies that are available suggest a safety profile similar to that of combination OCs with comparable hormone formulations.
WHOMEC suggests that the guidelines for combination OCs also should apply to the patch and the ring. Women shouldn’t use these contraceptive methods if they have a history of VTE or current VTE or if they are undergoing major surgery that may include prolonged immobilization.10
1. Abrams LS, Skee D, Natarajan J, Wong FA. Pharmacokinetic overview of Ortho Evra/Evra. Fertil Steril. 2002;77(2 suppl):S3-S12.
2. van den Heuvel MW, van Bragt AJ, Alnabawy AK, Kaptein MC. Comparison of ethinylestradiol pharmacokinetics in three hormonal contraceptive formulations: the vaginal ring, the transdermal patch and an oral contraceptive. Contraception. 2005;72:168-174.
3. US Food and Drug Administration Center for Drug Evaluation and Research. Ortho Evra (norelgestromin/ethinyl estradiol) Information. Available at: www.fda.gov/cder/drug/infopage/orthoevra/default.htm. Accessed July 5, 2008.
4. Cole JA, Norman H, Doherty M, Walker AM. Venous thromboembolism, myocardial infarction, and stroke among transdermal contraceptive system users. Obstet Gynecol. 2007;109(2 Pt 1):339-346.
5. Jick SS, Kaye JA, Russmann S, Jick H. Risk of nonfatal venous thromboembolism in women using a contraceptive transdermal patch and oral contraceptives containing norgestimate and 35 mcg of ethinyl estradiol. Contraception. 2006;73:223-228.
6. Jick S, Kaye JA, Li L, Jick H. Further results on the risk of nonfatal venous thromboembolism in users of the contraceptive transdermal patch compared to users of oral contraceptives containing norgestimate and 35 mcg of ethinyl estradiol. Contraception. 2007;76:4-7.
7. Burkman RT. Transdermal contraceptive patch. In: Rose BD, ed. UpToDate [online database]. Waltham, Mass: UpToDate; 2008.
8. Oddsson K, Leifels-Fischer B, de Melo NR, et al. Efficacy and safety of a contraceptive vaginal ring (NuvaRing) compared with a combined oral contraceptive: a 1-year randomized trial. Contraception. 2005;71:176-182.
9. Gaffield ME, Curtis KM, Mohllajee AP, Peterson HB. Medical eligibility criteria for new contraceptive methods: combined hormonal patch, combined hormonal vaginal ring and the etonogestrel implant. Contraception. 2006;73:134-144.
10. Reproductive Health and Research World Health Organization. Medical Eligibility Criteria for Contraceptive Use. 3rd ed. Geneva, Switzerland: World Health Organization; 2004. Available at: www.who.int/reproductive-health/publications/mec/mec.pdf. Accessed September 12, 2008.
1. Abrams LS, Skee D, Natarajan J, Wong FA. Pharmacokinetic overview of Ortho Evra/Evra. Fertil Steril. 2002;77(2 suppl):S3-S12.
2. van den Heuvel MW, van Bragt AJ, Alnabawy AK, Kaptein MC. Comparison of ethinylestradiol pharmacokinetics in three hormonal contraceptive formulations: the vaginal ring, the transdermal patch and an oral contraceptive. Contraception. 2005;72:168-174.
3. US Food and Drug Administration Center for Drug Evaluation and Research. Ortho Evra (norelgestromin/ethinyl estradiol) Information. Available at: www.fda.gov/cder/drug/infopage/orthoevra/default.htm. Accessed July 5, 2008.
4. Cole JA, Norman H, Doherty M, Walker AM. Venous thromboembolism, myocardial infarction, and stroke among transdermal contraceptive system users. Obstet Gynecol. 2007;109(2 Pt 1):339-346.
5. Jick SS, Kaye JA, Russmann S, Jick H. Risk of nonfatal venous thromboembolism in women using a contraceptive transdermal patch and oral contraceptives containing norgestimate and 35 mcg of ethinyl estradiol. Contraception. 2006;73:223-228.
6. Jick S, Kaye JA, Li L, Jick H. Further results on the risk of nonfatal venous thromboembolism in users of the contraceptive transdermal patch compared to users of oral contraceptives containing norgestimate and 35 mcg of ethinyl estradiol. Contraception. 2007;76:4-7.
7. Burkman RT. Transdermal contraceptive patch. In: Rose BD, ed. UpToDate [online database]. Waltham, Mass: UpToDate; 2008.
8. Oddsson K, Leifels-Fischer B, de Melo NR, et al. Efficacy and safety of a contraceptive vaginal ring (NuvaRing) compared with a combined oral contraceptive: a 1-year randomized trial. Contraception. 2005;71:176-182.
9. Gaffield ME, Curtis KM, Mohllajee AP, Peterson HB. Medical eligibility criteria for new contraceptive methods: combined hormonal patch, combined hormonal vaginal ring and the etonogestrel implant. Contraception. 2006;73:134-144.
10. Reproductive Health and Research World Health Organization. Medical Eligibility Criteria for Contraceptive Use. 3rd ed. Geneva, Switzerland: World Health Organization; 2004. Available at: www.who.int/reproductive-health/publications/mec/mec.pdf. Accessed September 12, 2008.
Evidence-based answers from the Family Physicians Inquiries Network