Review Paper

Androgen Treatment of Depressive Symptoms in Older Men: A Systematic Review of Feasibility and Effectiveness

Nathalie T Shamlian MD, FRCPC¹, Martin G Cole MD, FRCPC²

Objective: To determine the feasibility and effectiveness of androgen treatment of depressive symptoms in older men. Method: We searched MEDLINE, PsycINFO, Eric, HealthStar, and the Cochrane Database of Systematic Reviews for potentially relevant articles. The bibliographies of relevant articles were searched for additional references and experts were consulted. Seventeen reports (8 open and 12 randomized trials; 3 articles reported studies of 2 designs) met the following inclusion criteria: original investigation; published in English or French, use of acknowledged criteria or scale for depression, and open or randomized trial of androgen treatment. Four criteria assessed study validity: randomization, double blinding, comparability of treatment and control groups at baseline, and description of dropouts. We abstracted, tabulated, and compared information from each article. Results: Most studies had methodological limitations. With regard to feasibility, there were few reported withdrawals due to adverse events. With regard to effectiveness, 6 of 8 open trials had positive results, and 5 of 12 randomized trials had positive results, although 1 was equivocal. However, testosterone was combined with antidepressant medication in 3 of these positive randomized trials. Only 2 open trials enrolled subjects whose mean age was 60 years or over. One had positive results, and the other had negative results. Conclusion: Androgen therapy may be feasible in the short term, but there is little evidence that it is an effective treatment for depressive symptoms in older men. (Can J Psychiatry 2006;51:295–299) Click here for author affiliations.  Click here for informationon funding and support.  Clinical Implications Little evidence supports the use of androgen therapy for depressive symptoms in older men. Limitations The literature search was restricted to articles published in English or French. There was heterogeneity in study designs, populations, androgens used, dosages, duration of therapy, and outcome measures. Key Words: androgen therapy, depression, aged Résumé : Le traitement à l’androgène des symptômes dépressifs chez les hommes âgés : une étude systématique de la faisabilité et de l’efficacité

For 3 reasons, depressive disorders in the elderly are becoming one of the most frequent clinical situations challenging health care professionals. First, people aged 65 years and over represent the fastest-growing population in Canada. Their numbers rose by almost 60% between 1981 and 1998, reaching a total of 3.7 million (for Health Canada 2003 statistics, see www.hc-sc.gc.ca/seniors-aines). Second, the prevalence rate of major depression in the elderly ranges from 1% to 4% in the community (1); the rate rises to 10% to 30% in elderly medical inpatients (2). Third, major depression in many elderly individuals is a chronic disorder characterized by recovery and recurrence. A recent study confirmed that elderly subjects suffer from more rapid recurrence than younger patients, with a median time to recurrence of 90 weeks in the group aged 65 to 79 years (3).

Moreover, older patients seem to suffer much more frequently from minor or subthreshold depression than from major depression. Among those aged 55 years and over in a community sample, the prevalence rate was 12.9% (1). Among older men, subthreshold depression is associated with as much disability as MDD and with an increased risk of death, compared with control subjects not suffering from depression (4,5). Beekman and others followed a group of 277 elderly subjects with depression for 6 years (average age at the beginning of the study, 71.8 years) (1). The most frequent diagnosis was subthreshold depression (75%, compared with 8% having MDD); 66% of these patients had a chronic-intermittent or chronic course of disease over the 6 years.

The prevalence rate of low levels of total testosterone in men aged over 60 years is reported to be approximately 20% (6,7). Low levels of testosterone are associated with depressive symptoms that include lack of energy, an irritable or sad mood, and decreased enjoyment of life, as determined by the St-Louis Testosterone Deficiency Questionnaire (8). Patients can also report a decreased sense of well-being, fatigue, decreased libido, and erectile dysfunction (7).

Shores and others examined the link between diminished testosterone levels and the incidence of depressive illness in a historical cohort study (9). They concluded that a total testosterone level of 200 ng/dL or less increased the risk of depression approximately fourfold in older men (that is, those aged 45 years and over). Seidman and others found that most of the elderly men with dysthymic disorder in their study had total testosterone levels in the hypogonadal range (mean 295 mg/dL)—levels significantly lower than the mean levels obtained in their groups with and without depression (10).

The high frequency of depressive symptoms and low testosterone levels in older men has made testosterone an appealing treatment for depression in this population. As early as the 1940s, testosterone was used as a treatment for depression (11, 12). In recent years, there has been a growing literature on “andropause” and an increase in avertisements for testosterone medications. For example, the May 2003 issue of Doctors Review (see www.doctorsreview.com/index.html) contained advertisements for 3 different formulations of testosterone. Similarly, at the 2003 congress of Psychiatry in Hull, Quebec (« Sexe et Psyché », 37ième congrès annuel de l’AMPQ), 3 different specialties—urology, endocrinology, and psychiatry—presented on this topic. Thus the objective of our systematic review was to determine the evidence for the feasibility and effectiveness of androgen treatment of depressive symptomatology in older men. The review process, modified from that suggested by Oxman and others (13), involved systematic selection of articles, review of the quality of each study, abstraction of data, and qualitative synthesis of results.

Methods

Article Selection

One author used 3 different strategies to identify all potentially relevant articles. First, this author searched MEDLINE from 1966 to September 2004, using the following key terms: depression (exploded) or depressive mood; and successively, testosterone: deficiency, therapeutic use, hormonotherapy; testosterone therapy (exploded); testosterone, therapeutic use; and depression (exploded). We searched other databases in a similar way (PsycINFO from 1985 to September 2004, HealthStar from 1975 to September 2004, Eric from 1966 to September 2004, and the Cochrane Database of Systematic Reviews), but we found no additional articles. Second, potentially relevant articles were retrieved and the bibliographies screened for additional references. Finally, one author contacted specialists who had written or lectured on this subject (2 psychiatrists and an endocrinologist), as well as apposite pharmaceutical companies, to identify any relevant new, ongoing, or unpublished studies.

All potentially relevant articles were screened independently by both authors to meet the following 4 inclusion criteria: original investigation, published in English or French, use of acknowledged criteria for depression and dysthymia and (or) an accepted scale to measure depressive symptoms, and open or randomized trial of treatment with androgen therapy.

Assessment of Validity

Four elements concerning the quality of the studies (as described by Jadad and others, 14) were abstracted independently by both authors. These criteria were randomization, double blinding, comparability of treatment and control groups at baseline, and description of dropouts.

Data Synthesis

One author abstracted and tabulated the following information from each article: study design and duration; population from which the patients were selected; patient age, sex, and gonadal status; inclusion and exclusion criteria; number of patients enrolled and completing the study; description of treatment and control groups; frequency of secondary effects requiring withdrawal from treatment; assessment and diagnostic tools; and results. To determine feasibility, we tabulated for each study the frequency of withdrawal due to proven or potential secondary effects. To determine effectiveness, we compared measures before and after treatment in the case of open trials. In the case of randomized trials, we compared the measures between treatment groups or treatment and control groups. The studies used different depression rating scales, and sometimes multiple scales, but at least one scale (usually the HDRS) was the accepted measure of depressive symptoms. When 2 or more acceptable scales were used, we reported the results of the HDRS only; if there were conflicting results with different scales, we reported all the results.

Results

Article Selection

We retrieved 25 potentially relevant articles, of which 17 met the inclusion criteria (15–31). The interrater agreement for this phase of the study was 97%; consensus was obtained through discussion between the authors. The review includes 8 open trials and 12 RCTs because 3 articles (15,19,21) reported studies of 2 designs (see Note).

Assessment of Validity

Interrater agreement between the authors on the 4 criteria of methodological quality was 89%; consensus was achieved through discussion between the authors. Most of the randomized trials had one or more methodological limitations.

Data Synthesis

Studies and Populations. The studies included 856 patients. Patients were generally male , although female subjects were enrolled in 3 studies (17,25,26). Subjects’ ages ranged from 15 to 72 years. The mean age of the groups, when available, was approximately 40 years. In only 2 studies (20,22) were the mean ages greater than 60 years. Five studies (16,19,27,28, including studies 16A and 16B), enrolled subjects who were HIV-positive. The baseline testosterone status of subjects varied from study to study: some studies included eugonadal men, others included hypogonadal men or men with low-normal testosterone levels. Notably, testosterone values defined as normal varied from one study to the next because there is no clear consensus regarding what these values should be. Concerning depression, different syndromes were examined, including major depression, dysthymia, and subthreshold depressive symptoms. Four studies (15,23,24, including studies 15A and 15B) used mesterolone, 13 studies used testosterone preparations (16,18–22,27–31), and 3 used DHEA (17,25,26). Study duration was short, (3 to 14 weeks), except for 2 studies of 5 and 6 months’ duration, respectively (18,27).

Feasibility. With regard to feasibility, 73% to 100% of subjects completed the studies in which they were enrolled. Only 5 studies reported withdrawals related to secondary effects (16,19,26,30,31). In most studies, the secondary effects were not specified.

Effectiveness. With regard to effectiveness, 6 of 8 open trials, involving 267 patients, had positive results with significant reduction in depressive symptoms (16–19,21,22). The differences in the results could not be attributed to differences in study characteristics.

Of 12 randomized trials, only 4 had positive results (24– 26,28); 7 had negative results (15,19,21,23,27,29,31), and the remaining trial was equivocal (30). Of the 3 best-quality studies, 2 had negative results (29,31), and 1 was equivocal (30). In 3 of the 5 positive studies (25,28,30) the testosterone preparation was combined with antidepressant medication. Testosterone status at baseline did not appear to be related to the results of either the open or randomized trials. We were unable to do a quantitative metaanalysis of the results of the randomized trials because different androgen preparations were used and only 5 of the 12 reports offered sufficient information to calculate effect sizes.

Discussion

We proposed to determine the feasibility and effectiveness of androgen treatment for depressive symptomatology in older men. Although we located 8 open trials and 12 randomized trials, only 2 open trials enrolled subjects whose mean ages were greater than 60 years. The results of one were positive, and the results of the other were negative. The results of the remaining 18 trials suggest that androgen therapy is feasible in younger men in the short term (that is, 3 to 14 weeks), but even in this younger population, there is little evidence that androgen therapy alone is an effective treatment for depressive symptoms.

This review highlights the variability of the different studies in this field of research. To start with, the treatments did not always use the same molecule. For example, mesterolone is a derivative of DHT, 1 of the 2 metabolites of testosterone after transformation in peripheral tissues. In the studies that used mesterolone, patients were not exposed to estradiol, which could explain in part the negative results obtained. Other studies used the adrenal androgen DHEA, which is converted into testosterone and estrogens.

Also, different types of testosterone preparations were used: gel in 2 studies and intramuscular preparations in 11 studies. Dosages (amount and frequency of doses) were quite different, ranging from DHEA 30 to 450 mg daily, mesterolone 50 to 450 mg daily, testosterone gel 5 to 10 g daily, and testosterone injection 100 mg once weekly to 400 mg biweekly. Interestingly, in 2 studies using transdermal testosterone (20,30), the authors reported that fewer than 50% of the hypogonadal patients had a doubling of their plasma-free testosterone level and that 3 of the 11 testosterone subjects had increases of 70 ng/dL or less in their total testosterone levels. This variability makes it difficult to conclude which specific treatment is useful.

The populations were heterogeneous: healthy eugonadal volunteers in some studies and HIV-positive patients with wasting or older patients with Parkinson’s disease in others. According to available data (mean age was not always given), the patients’ mean age was about 40 years in most studies. This limits the generalizability of positive results to older patients. Notably, advertising by pharmaceutical companies targets men suffering from andropause, even if evidence of efficacy in this age group, at least concerning mood symptoms, is limited because few older subjects have been studied.

In Grinspoon and others’ study, 35% of the testosterone effect on mood could be explained by increased weight (multivariate analysis), possibly indicating an indirect effect of testosterone (27). Possibly, depressive symptoms and low testosterone levels are markers of the same disease state in some patients, with no causal relation between them.

In another study, the results were negative for hypogonadal patients suffering from MDD (29). These patients were not allowed concomitant psychotropic medication during the blinded phase. In an open trial by the same authors, patients had a positive response to testosterone treatment used as an adjunct to treatment with selective serotonin reuptake inhibitors (18). These results suggest that testosterone treatment might be useful in hypogonadal patients with depressive symptoms, probably more as an adjunctive treatment than as a first-line treatment. In Rabkin and others’ study, results in subgroups of patients were examined to determine whether patients with MDD were more likely to respond to testosterone injections (400 mg biweekly) than were subjects with only subthreshold symptoms (31). The conclusion was negative.

The study durations were noticeably short, usually 6 to 8 weeks. This might result in negative findings if it takes more time for significant changes in symptoms to occur. However, patients who responded positively seemed to do so rapidly. Seidman and others reported a significant decrease in HDRS scores as early as Week 2 (18); Bloch and others reported that 47% of patients receiving DHEA responded by Week 3, whereas 60% responded by Week 6 (26); and O’Connor reported significant reductions in symptoms by Week 2 (21).

There has been concern about the safety of androgen treatment. In the short term, it appears that patients tolerate the treatment well and that there are few adverse effects. However, serious consequences of long-term exposure to androgen therapy have been documented, including heart failure, peripheral edema, acne, prostate disorders, gynecomastia, hypercalcemia, sleep apnea, and laboratory test abnormalities, (elevated gamma-glutamyltransferase, total bilirubine, hematocrit, and hyperlipidemia; decreased high-density lipoprotein; and hypokalemia) (32,33).

This review has 3 limitations. First, we limited our literature search to French and English articles. Second, we did not assess the reliability of the literature search or data abstraction. Third, we were unable to do a quantitative metaanalysis.

To conclude, androgen therapy may be feasible in the short term, but there is little evidence that it is an effective treatment of depressive symptoms in older men.

Note

Tables showing details of the studies reviewed for this paper are available in the online version of the article at www.cpa-apc.org/Publications/cjpHome.asp.

Funding and Support

No funding or other support was received for this paper.

References

1. Beekman ATF, Geerlings SW, Deeg DJH, Smit JH, Schoevers RS, de Beurs E, and others. The natural history of late-life depression. Arch Gen Psychiatry 2002;59:605–11.

2. Fenton FR, Cole MG, Engelsmann F, Mansouri I. Depression in older medical inpatients. Int J Geriatr Psychiatry 1994;9:279–84.

3. Mueller TI, Kohn R, Leventhal N, Leon AC, Solomon D, Coryell W, and others. The course of depression in elderly patients. Am J Geriatr Psychiatry 2004;12(1):22–9.

4. Beekman AT, Deeg DJ, Braam AW, Smit JH, Van Tilburg W. Consequences of major and minor depression in later life: a study of disability, well-being and service utilization. Psychol Med 1997;27:1397–409.

5. Penninx BWJH, Geerlings SW, Deeg DJH, van Eijk JTM, van Tilburg W, Beekman ATF. Minor and major depression and the risk of death in older persons. Arch Gen Psychiatry 1999;56:889–95.

6. Margolese HC. The male menopause and mood: testosterone decline and depression in the aging male—is there a link? J Geriatr Psychiatry 2000;13:93–100.

7. Lee J. Management of androgen decline in aging men. Geriatr Ageing 2003;6:23–6.

8. Okun MS, McDonald WM, DeLong MR. Refractory nonmotor symptoms in male patients with Parkinson disease due to testosterone deficiency. Arch Neurol 2002;59:807–11.

9. Shores MM, Sloan KL, Matsumoto AM, Moceri VM, Felher B, Kivlahan DR. Increased incidence of diagnosed depressive illness in hypogonadal older men. Arch Gen Psychiatry 2004;61:162–7.

10. Seidman SN, Araujo AB, Roose SP, Devanand DP, Xie S, Cooper TB, and others. Low testosterone levels in elderly men with dysthymic disorder. Am J Psychiatry 2002;159:456–9.

11. Danziger L, Schroeder HT, Unger AA. Androgen therapy for involutional melancholia. Arch Neurol Psychiatry 1944;51:457–61.

12. Altschule MD , Tillotson KJ. The use of testosterone in the treatment of depression. N Engl J Med 1948;239:1036–8.

13. Oxman AD, Cook DJ, Guyatt GH. Users’ guides to the medical literature: VI. How to use an overview. JAMA 1994;272:1367–71.

14. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJM, Gavaghan DJ, and others. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17(1):1–12.

15. Itil T, Herrmann WM, Blasucci D, Freedman A. Male horomones in the treatment of depression: effects of mesterolone. Prog Neuropsychopharmacol 1978;2:457–67.

16. Rabkin JG, Rabkin R, Wagner G. Testosterone replacement therapy in HIV Illness. Gen Hosp Psychiatry 1995;17:37–42.

17. Wolkowitz OM, Reus VI, Roberts E, Manfredi F, Chan T, Raum WJ, and others. Dehydroepiandrosterone (DHEA) treatment of depression. Soc Biol Psych 1997;41:311–8.

18. Seidman SN, Rabkin JG. Testoserone replacement therapy for hypogonadal men with SSRI-refractory depression. J Affect Disord 1998;48:157–61.

19. Rabkin JG, Wagner GJ, Rabkin R. Testosterone therapy for human immumodeficiency virus-positive men with and without hypogonadism. J Clin Psychopharmacol 1999;19(1):19–27.

20. Okun MS, Walter BL, McDonald WM, Tenover JL, Green J, Juncos JL, and others. Beneficial effects of testosterone replacement for the nonmotor symptoms of Parkinson disease. Arch Neurol 2002;59:1750–3.

21. O’Connor DB, Archer J, Hair WM, Wu FCW. Exogenous testosterone, agression, and mood in eugonadal and hypogonadal men. Physiol Behav 2002;75:557–66.

22. Perry PJ, Yates WR, Williams RD, Andersen AE, MacIndoe JH, Lund BC, and others. Testosterone therapy in late-life major depression in males. J Clin Psychol 2002; 63:1096–101.

23. Itil TM, Michael ST, Shapiro DM, Itil KZ. The effects of mesterolone, a male sex hormone in depressed patients (a double-blind controlled study). Methods Find Exp Clin Pharmacol 1984;6:331–7.

24. Vogel W, Klaiber EL, Braverman DM. A comparison of the antidepressant effects of a synthetic androgen (mesterolone) and amitriptyline in depressed men. J Clin Psychiatry 1985;1:6–8.

25. Wolkowitz OM, Reus VI, Keebler A, Nelson N, Friedland M, Brizendine L, and others. Double-blind treatment of major depression with dehydroepiandrosterone. Am J Psychiatry 1999;156:646–9.

26. Bloch M, Schmidt PJ, Danaceau MA, Adams LF, Rubinow DR. Dehydroepiandrosterone treatment of midlife dysthymia. Soc Biol Psych 1999;45:1533–41.

27. Grinspoon S, Corcoran C, Takara S, Baaj A, Basgoz N, Klibanski A. Effects of hypogonadism and testestorone administration on depression indices in HIV-Infected Men. J Clin Endocrinol Metab 2000;85:60–5.

28. Rabkin JG, Wagner GJ, Rabkin R. A double-blind, placebo-controlled trial of testosterone therapy for HIV-positive men with hypogonadal symptoms. Arch Gen Psychiatry 2000;57:141–7.

29. Seidman SN, Spatz E, Rizzo C, Roose SP. Testosterone replacement therapy for hypogonadal men with major depressive disorder: a randomized, placebo-controlled clinical trial. J Clin Psychiatry 2001;62:406–12.

30. Pope HG Jr, Cohane GH, Kanayama G, Siegel AJ, Hudson JI. Testosterone gel supplementation for men with refractory depression: a randomized, placebo-controlled trial. Am J Psychiatry 2003;160:105–11.

31. Rabkin JG, Wagner GJ. Testosterone versus fluoxetine for depression and fatigue in HIV/AIDS. J Clin Psychopharmacol 2004;24:379–85.

32. Berkow R, Fletcher AJ. The Merck manual of diagnosis and therapy. 6th ed. Rahway (NJ): Merck Research Laboratories; 1992.

33. Wyngaarden JB, Smith LH, Bennett JC. Cecil textbook of medicine. 19th ed. Philadelphia (PA): WB Saunders; 1992.

Author(s)

Manuscript received February 2005, revised, and accepted December 2005.

1. Psychiatre et Responsable de la Clinique de la mémoire, Département de psychiatrie, Service de psychogériatrie, Hôpital du Sacré-Cœur de Montréal, et Université de Montréal, Montréal, QC.

2. Senior Psychiatrist, Department of Psychiatry, St Mary’s Hospital and McGill University, Montreal, QC

Address for correspondence: Dr N Shamlian, Département de psychiatrie, Service de psychogériatrie, Hôpital du Sacré-Cœur de Montréal, 5400, boul Gouin Ouest, Montréal, QC H4J 1C5

e-mail: clinique.memoire.hsc@ssss.gouv.qc.ca

1 | 2