Version 2.0

Antidepressant-Induced Sexual Dysfunction Associated with Low Serum Free

Psychiatry On-Line 1999


Return to Psychiatry On-Line Contents

Alan Jay Cohen, M.D.
Private Practice and Assistant Clinical Professor of Psychiatry, UCSF



In the course of an evaluation for treatment of antidepressant-induced
sexual dysfunction (ASD) with a new agent, an unforeseen pattern emerged
in the pre-treatment laboratory assessment. Free serum testosterone
levels in both men and women study subjects were found to be below the
normal ranges in 75 percent of subjects in this small study. There were
no other consistent laboratory findings that could account for such a
high percentage correlation.
Further inquiries into the possible causes for decreased serum
testosterone and its association with ASD seems warranted.



Antidepressant-induced sexual dysfunction (ASD) is a well recognized
complication of treatment for mood and anxiety disorders, (Gitlin 1997).
Recent discoveries have helped to provide effective remedies for this
significant obstacle to patient compliance and successful treatment
outcome(Cohen 1997, Gitlin 1997, Bartlik 1995). However, no remedy is
100% effective. In addition, there is no fully satisfactory theory that
explains the physiologic mechanisms responsible for the varied aspects
of sexual dysfunction observed. In the course of an
evaluation of treatment for ASD in a community office-based research
setting, a striking pattern emerged in the laboratory screening
protocol. Free testosterone levels were found to be subnormal in 15 of
20 patients. No other consistent laboratory value nor physical
examination finding could account for this observation. Causes for
reduced free testosterone and its effect on sexual function are
discussed with implications for future research and treatment


Twenty subjects, ages 35 to 74 years, were evaluated for a double blind
placebo controlled trial of a dietary supplement combination for the
treatment of ASD. All of the subjects were using medication for the
treatment of mood disorder (DSM IV Criteria) included SSRI's, SNRI's,
bupropion, trazodone and mirtazipine. Screening physical exams and
laboratory studies included CBC, TSH, Prolactin, serum free
Testosterone, Serum Chemistries, and Urinalysis were done. The Arizona
Sexual Effects Change Scale (ASECS) was used as part of the clinical
assessment of ASD. In the course of the evaluation process, low serum
free testosterone was noted in 15 patients.



Twelve men and eight women were evaluated. Eight men had subnormal free
testosterone levels, two additional men had borderline low levels. Six
women had subnormal levels of free testosterone. The average age of male
subjects was 50.5 years. The male ASECS mean score was 20 with a mean
free Testosterone of 13.5 pg/ml. The laboratory range of free
Testosterone was 16 - 33 pg/ml. The average age of female subjects was
39.6 years; female ASECS score was 20, and the mean free Testosterone
level was 0.8 pg/ml. (normal range 0.8 - 3.0 pg/ml). (Laboratory ranges
were modified according to standardized norms for age; average free
testosterone levels decline slightly with increasing age.) Table #1
summarizes the data on all of the subjects in the study.
Prolactin levels were above normal in only two subjects (one male, one
female), both of whom were also found to have below normal levels of
free testosterone.

All of the other subjects had normal Prolactin levels. Thyroid
stimulating hormone was found to be normal in all subjects.

Table 1.

Sex Age Medication  ASECS score free T (pg./ml.)
M 35  venlafaxine 16 23.4
M 36 sertraline 21 5.2 *
M 43 paroxetine 18 13.5 *
M 45 venlafaxine 17 16.3 #
M 46 venlafaxine 20 13.2 *
M 46 paroxetine/mirtazepine 20 13.4 *
M 47 citalopram 19 29.0
M 47 fluoxetine 22 17.6 #
M 50 sertraline 17 6.2 *
M 53 nefazodone 25 11.1 *
M 54 bupropion 21 7.4*
M 74 venlafaxine 24 5.6 *
F 20 citalopram 29 1.7
F 31 venlafaxine 21 0.50*
F 37 paroxetine 23 0.70 *
F 41 paroxetine 19 1.5
F 44 sertraline 16 0.40 *
F 45 bupropion/trazodone 16 0.50 *
F 47 fluoxetine 20 0.50*
F 52 bupropion 16 0.40

(*denotes subnormal fT levels, # denotes borderline low free T levels)
ASECS score range is 5-30 , 5 is maximal sexual function, 30 is minimal score.


This report is the first known documentation of reduced free
testosterone levels associated with ASD. Prior reports have mentioned
SSRI-induced prolactin elevations but none have described effects on
testosterone levels(Amsterdam 1997).
Certainly, drugs can play a role in decreasing testosterone levels.
Ketoconazole, megestrol, cimetidine, and spironolactone have all been
reported to lower testosterone levels(De Coster 1985, Griffin and Wilson
1998). Methadone and other opiates can suppress testosterone by reducing
LH levels centrally(Griffin and Wilson 1998) Anticonvulsants have been
associated with reduced free testosterone although epilepsy itself is also known to exhibit this effect (Herzog, 1992). Carbamazepine may increase metabolic cleaarance of testosterone and reduce LH levels. The P-450 CYP3A3/4 system is involved in the metabolism of testosterone (Griffin and Wilson, 1998). It is possible that antidepressants may be inducing the CYP3A3/4 isoenzyme with resultant enhanced metabolic clearance of testosterone and reduction in free hormone levels.  Changes
in sex hormone binding globulin levels can influence the quantity of
circulating free testosterone (Griffin and Wilson
1998). Certain medical conditions; cirrhosis, renal failure, HIV infection etc. have been associated with lower levels of testosterone (Griffin and Wilson, 1998). Even being a sports enthusiast may adversely effect testosterone levels (if the fan is on the losing side) (Bernhardt 1998). 
Studies investigating testosterone levels and mood disorders have shown
conflicting results (Seidman1998, Levitt, 1998). Levels of testosterone in 12 depressed males were
compared to age-matched normal controls by Levitt and Joffe in 1988.
No significant differences were noted between the two
groups.  Clearly, more research is needed to elucidate what role, if any, testosterone plays in the evaluation and treatment of antidepressant-induced sexual dysfunction. Further studies should take into account diurnal variations in hormone level, total and free levels of hormone as well as pre-and post-antidepressant levels. This report isn hindered by the limitations of a small number of subjects, lack of a control group, and no information on the testosterone level of subjects prior to the onset fo antidepressant use. Further studies should also include measurement of total testosterone levels as well as concomitant SHBG levels.


This is the first description of an association between low testosterone levels and antidepressant-induced sexual dysfunction known to this author. Further research is needed to evaluate this relationship in greater detail. It does open avenues of exploration regarding treatment of ASD utilizing hormone replacement.


1.)    Amsterdam J. et al  (1997) Breast enlargement during chronic
antidepressant therapy J Affective Disorders Nov.;46(2):151-156.
2.)   Bartlik B et al (1995) Psychostimulants apparently reverse sexual
dysfunction secondary to selective serotonin reuptake inhibitors. J Sex
Marital Ther. 21, (4):264-271.
3.)    Bernhardt PC et al (1998) Testosterone changes during vicarious
experiences of winning and losing among fans at sporting events. Physiol.
Behav Aug.;65(1):59-62.
4.)  Cohen AJ and Bartlik B (1998) Ginkgo biloba for
sexual dysfunction J Sex Marital Ther Apr-Jun 24:2 139-143.
5.) De Coster R et al (1985) Effect of a single administration of
ketoconazole on total and physiologically free plasma testosterone and 17
beta-oestradiol levels in healthy male volunteers. Eur J Clin Pharmacol
  6.) Gitlin M. (1997) Sexual side effects of psychotropic medications, in
     Psychiatric Clinics of North America:Annual of Drug Therapy, pg.61-90.
  7.)  Griffin J and Wilson J (1998) Disorders of the testes and the male
  tract in :Williams Textbook of Endocrinology 9th ed. W.B. Saunders Co.
_8.) Herzog AG (1995) Hormonal changes in epilepsy Epilepsia
  9.) Levitt A and Joffe R (1988) Total and free testosterone in depressed
men Acta Psychiatr Scand Mar;77(3)346-348.
10.) Sternbach H (1998) Age-associated testosterone decline in men:clinical
issues for psychiatry Am J Psychiatry Oct, 155:10,1310-1318.
11.) Seidman S and Rabkin J (1998) Testosterone replacement therapy for
hypogonadal men with SSRI-refractory depression. J Affect. Disord

The author wishes to thank Laura Stachel, M.D. for her assistance in the preparation of this report

Return to Psychiatry On-Line Contents