H. George Nurnberg, MD

Among 25 to 30 million Americans in whom depression is diagnosed annually, 18 to 25 million are treated with antidepressants, of which 90% are SSRI or non-selective serotonin reuptake inhibitor (SRI) antidepressants, the most frequently prescribed medications for all outpatients aged 18 to 65 years.

 Treatment-emergent adverse effects (eg, sexual dysfunction [SD], weight gain, sleep disturbance) account for 75% of premature treatment terminations and place patients at significantly increased risk for relapse, recurrence, morbidity, and mortality.

 With treatment-associated SD reported to occur in 40% to 70% of patients using SRIs,

 an iatrogenic SD may develop in 10 to 15 million patients annually. If this issue is not addressed, as many as 87% of these patients will become noncompliant.

SRI-associated SD usually occurs early in treatment, is persistent, and rarely remits spontaneously. It is frequently overlooked or misattributed to major depressive disorder (MDD). Despite an up to 80% potential response rate to the first or second SRI prescribed, up to 68% withdrawal rates within the first 3 months of treatment indicate that patients do not continue their prescribed medication for a sufficient period.

SRI-associated SD is a major factor responsible for this.

 It is interesting, however, to note that although physicians underestimate the occurrence of iatrogenic SD, it significantly influences their selection or switching of antidepressant agents.

Only a small number of double-blind, randomized, placebo-controlled trials (RCTs) have studied management of SRI-associated SD, and most of these found the agents tested no better than placebo. The only agents that have shown broad-based efficacy and effectiveness are selective type 5 phosphodiesterase inhibitors (PDE5Is).

 Without evidence-based treatments of SRI-associated SD, physicians have relied on clinical wisdom, anecdotal reports, case reports, and open-label studies to manage SD, which has left patients exposed to excessive random drug interventions aimed at mitigating the adverse sexual effects of SRIs.

Finding effective treatments for SRI-associated SD would be a significant advance for improving MDD treatment outcomes. In addition, SRI-associated SD can serve as a model for understanding relationships between treatment-emergent adverse effects in other conditions in which SD may be both a symptom of the primary disorder and an adverse effect of the medication used to treat it. Advances in nonadrenergic-noncholinergic novel signal transduction and the development of PDE5Is offer new evidence-based treatments for SD of different etiologies, for improving depression management with antidepressant-associated SD, and for developing treatments in other conditions in which medication causes SD. This article reviews the comparative effectiveness of treatment options for SRI-associated SD.

Classification and prevalence of sexual dysfunction

 classification system for SD is based on a modified triphasic model and categorizes patients according to etiology, gender, and domain: interest, response, and orgasm. A separate category, sexual pain disorders, includes dyspareunia, vaginismus, and priapism. Premature ejaculation in men and premature orgasm in women are residual categories. 

 has a separate category for cases that are secondary to medical treatment or substance use; antidepressant-associated SD is classified as substance-induced.

Epidemiological surveys of SD report rates of approximately 40% in women and 30% in men. These rates may be somewhat overreported, depending on whether a "significant distress" criterion is required.

 The most common complaints in women are low sexual desire (32%), inability to achieve orgasm (26%), and sex not being pleasurable (23%). Premature ejaculation (31%), performance anxiety (18%), and low sexual desire (15%) are the most common in men. With increasing age, men and women report lower frequencies of sexual intercourse;increased rates of erectile dysfunction; and decreased lubrication, orgasm, and enjoyment of sexual activity.

Patients with MDD should be systematically evaluated for SD. De-creased libido is reported in about 70% of patients with depression but is not a reliable indicator because it is easily confused with anhedonia

; 49% of women and 26% of men with MDD reported no sexual activity in the preceding month.

 Patients with MDD report decreased genital sensitivity, decreased sexual interest, decreased sexual arousal, and orgasm difficulties. Overall, 48% of women and 60% of men reported delayed or absent orgasm.

 SD severity appears to be correlated with depression severity.

Product inserts for antidepressants significantly underestimate the risk of SD (0% to 14%) because they are derived from spontaneous reports instead of systematic assessments that would reveal rates as high as 70%.

 A causal connection is clear because SD develops in healthy volunteers over a 2- to 4-week period when taking SRIs; this remits on drug discontinuation.

Most patients with SRI-associated SD present with similar rates of multiple complaints, including orgasm (delayed or anorgasmia), decreased satisfaction, arousal disturbance (lubrication/erection), and decreased interest.

 presents the distribution of men and women with SRI-associated SD in our trials. To determine which is primary, secondary, or a halo effect requires a systematic and comprehensive evaluation that must take into account:	

Preexisting SD not associated with depression or anxiety (with or without treatment), relationship problems, risk factors (eg, obesity, hypertension, alcohol/drug abuse), and other external factors or lifestyle conditions.

SD associated with antidepressant treatment in the context of treatment response (remission), dose, other concurrent Axis I to Axis III conditions, and psychosocial (relationship, work, stress, financial) factors.

Over time, some patients with depression-associated SD will improve with treatment, while some will experience an exacerbation when taking an antidepressant. In other patients, antidepressant-associated SD will develop de novo.

 Women appear to experience a higher base rate of depression-associated SD than men; men experience antidepressant-associated SD more frequently but with less severity and improve less than women with effective antidepressant treatment.

Although SSRIs replaced tricyclic antidepressants as first-line agents because of their improved side-effect profile, more efficient dosing, and overall safety and efficacy, the initial enthusiasm was dampened as more attention was paid to their adverse effects on sexual function. Although the effects on sexual function suggest a link to serotonergic mechanisms, this is not the full explanation because different serotonin receptors have different influences on sexual function. For example, 5HT-2 and 5HT-3 receptor stimulation is inhibitory to sexual function compared with 5HT-1 stimulation, which is facilitory.

 The final expression of antidepressant-associated SD is a cumulative effect, derived from factors such as receptor selectivity, receptor binding, potency, and other properties of the specific antidepressant (

 This variability was demonstrated in a study of more than 6000 patients using standardized instruments to evaluate SD in a primary care setting. The study found the following rates of antidepressant- associated SD with various agents: bupropion IR (22%), bupropion SR (25%), nefazodone (28%), SSRIs (39% to 43%), venlafaxine (42%), and mirtazapine (41%), which were lower when patients with comorbidities and concurrent risk factors were factored out.

 However, while differences between SRIs and other antidepressants suggest differential effects, their clin-ical significance needs to be better understood.

Explanations of the causes and mechanisms of MDD and SRI treatment-emergent SD were essentially post hoc to assumptions that elevated central serotonin levels were causal. With the development of PDE5Is and the awareness of the role of nitric oxide and other novel signaling transmitters, the involvement of dopamine, adrenergic, cholinergic, and many other neu- rohormonal, neurotransmitter, and other peripheral agents (eg, 

-methyl- d-aspartate, neuropeptides, hormones, g-aminobutyric acid) in the complex sequencing of events that constitute biological and behavioral sexual function became better understood. An explanatory model, that could be empirically tested, was needed that recognized the extensive cross-talk and direct/indirect dynamic interactions occurring among various central, autonomic, and peripheral neurotransmitter and neuroendocrine systems.

Management of antidepressant-associated sexual dysfunction

In almost 5 decades of antidepressant use, treatment recommendations for antidepressant-associated SD have increased. They fall into 4 general categories: (1) use of antidotes by agonists, partial agonists, or antagonist agents; (2) avoidance by use of nonserotonergic antidepressants with neutral or proactive pharmacokinetics for sexual function; (3) augmentation with or switching to primarily nonserotonergic antidepressants after the emergence of treatment-associated SD; and (4) adaptation or tolerance by waiting, dose adjustment, or drug holiday.

Most of the evidence for antidotes comes from single case reports or open-treatment case series. Initially promising antidotes failed to separate from placebo in RCTs. Reports for the use of mianserin, cyproheptadine, 

 amantadine, loratadine, and other agents have not demonstrated significant improvement for antidepressant-associated SD over placebo. The conclusions in these trials are also limited by their methodological shortcomings.

 For example, buspirone was reported to be effective for SSRI-induced SD as a post hoc secondary finding in a treatment-augmentation study of refractory depression.

 Another prospective buspirone RCT for treatment of antidepressant-associated SD showed no significant effect with higher doses than in the augmentation study.

Another approach and variant of antidote approaches is to select an antidepressant agent with a receptor profile that is less likely to be associated with the development of SD (eg, bupropion, nefazodone, mirtazapine, reboxetine). There are data showing that when prescribed as initial agents, these antidepressants may be associated with lower rates of antidepressant-associated SD compared with some SSRIs.

 However, despite its widespread promotion and perception of efficacy, bupropion has not been shown to be significantly better than placebo as a treatment for SSRI-associated SD in RCTs.

Switching antidepressants in a patient with MDD who has SRI-associated SD and is responding to the antidepressant adds the risk that the patient may be switched from an effective agent to an agent that may not be effective, as well as the risk for potential relapse or recurrence.

Waiting for adaptation may be the most frequently used approach. Placebo-controlled studies, or those following patients over time, have shown that spontaneous remission occured in only 6% to 20% of patients.

 Dose reduction to reestablish sexual function is not a practical strategy, since it may result in a subtherapeutic level with consequent relapse or recurrence of depression. There is no evidence to support the use of a drug holiday, although one uncontrolled study found favorable results after patients discontinued their SSRI for up to 3 days and then restarted medication at the usual dose.

Improvement in sexual functioning during a drug holiday cannot be expected with a longer-acting drug such as fluoxetine, presumably because of the metabolite's long half-life. In addition, intermittent dosing puts patients at risk for serotonergic discontinuation syndrome and promotes noncompliance with medication.

With the exception of PDE5Is (sildenafil, tadalafil, vardenafil), the data on current SRI-associated SD management approaches with other antidotes, augmentation, switching antidepressants, or waiting for tolerance to develop, are without evidence and rely on clinical wisdom.

 Psychiatrists will need to familiarize themselves further with these PDE5I agents, which can offer effective management for SRI-associated SD.	

The currently available PDE5Is are competitive and selective inhibitors of cyclic guanosine monophosphate [cGMP] specific phosphodiesterase-type five [PDE-5]. They have been shown to be effective in treating erectile dysfunction (ED) with a diverse etiological spectrum in men and are well-tolerated oral agents that enable a natural response to sexual stimulation by acting at the end organ to permit vascular engorgement and penile erection.

 PDE5Is distributed in smooth muscle cells of the corpus cavernosum, blood vessels, and circulating platelets enhance the signal transduction of nitric oxide--cGMP on smooth-muscle relaxation in the corpus cavernosum by inhibiting PDE-5 catabolism of cGMP.

 PDE5Is have advanced the understanding of physiological mechanisms of SD and offer a new approach for the treatment of medication-associated SD.

The efficacy of PDE5Is for the treatment of ED in men with comorbid depression was first reported in a retrospective analysis of data from 3414 men with diverse etiologies, including current or past history of depression from 11 double-blind, placebo-controlled clinical trials.

 Sildenafil 25 to 100 mg taken approximately 1 to 2 hours before sexual activity showed significant efficacy on the erectile function domain of the International Index of Erectile Function (IIEF). Mean improvement in baseline ED scores was significantly higher in those treated with sildenafil than in those treated with placebo (

 < .001).PDE5I efficacy for ED with comorbid mild to moderate or major depression was demonstrated in 3 RCTs: in men with depression in remission, in men with untreated mild or major depression,and in men with residual ED and treated MDD.

 Improvement from baseline in IIEF erectile function scores was significantly greater with a PDE5I than with placebo (

 < .0001). In addition, regardless of whether treatment was with PDE5I or placebo, the mean score on the Hamilton Rating Scale for Depression (HAM-D) improved to a greater extent in treatment responders than in nonresponders, suggesting that the relationship between ED and depression is complex and multidirectional.

The efficacy of sildenafil in the treatment of antidepressant-associated SD was first noted in several case reports and small open-label studies.

 A retrospective subanalysis of several double-blind, placebo-controlled trials showed significantly greater improvement in mean scores for erectile function, orgasm, and satisfaction domains of the IIEF (

 < .01), suggesting the effectiveness of sildenafil for antidepressant-associated SD.

The strongest evidence supporting PDE5I efficacy in the treatment of antidepressant-associated ED is provided by 3 prospective, double-blind RCTs in men with MDD in remission and SRI-associated SD. The men were enrolled on the basis of treatment with a selective or nonselective SRI for a minimum of 12 weeks (last 6 weeks stable), antidepressant-associated SD for at least 4 weeks, and no history of SD before the onset of depression or antidepressant treatment.

 Sildenafil was taken before sexual activity, initially at 50 mg, with the dose increased to 100 mg if needed. Sildenafil-assigned patients reported much/very much improved sexual function, including erectile function, ejaculation, orgasm, and satisfaction compared with those on placebo.

All patients maintained remission of depression, supporting the importance of maintaining medication adherence by managing treatment-emergent adverse effects to improve treatment outcomes. In addition, antidepressant- associated SD responders showed significant additional improvement in depression severity, even though in remission (HAM-D score of 7 or greater) at baseline. Although the FDA indication is specific to ED, as a whole, the empirical data for these studies are consistent with those showing that sildenafil provides significant improvements in SRI-associated SD, not limited to ED but including desire, delayed ejaculation/ orgasm, and satisfaction, which can directly or indirectly improve with treatment.

With continuation open-label treatment of SRI-associated SD in partial responders and nonresponders, 81% became full responders (75% of partial and nonresponders to double-blind sildenafil; and 85% of partial and nonresponders to blind placebo) (

). Significant improvement occurred in measures of erectile function and ejaculatory delay, with overall satisfaction in both groups with persisting remission of MDD in all patients who continued stable-dose SRI antidepressant treatment. Sildenafil was well tolerated, with no discontinuations attributed to adverse events in both open and blind phases.

 Similar studies with tadalafil and vardenafil treatment of depression-associated or antidepressant-associated ED showed similar results, with patients attaining successful intercourse during double-blind

The use of sildenafil and other PDE5Is for antidepressant-induced SD in women is off-label. With women exibiting a higher prevalence of MDD, greater severity of treatment-emergent SD, and greater use of SSRIs, it was expected that interest regarding the potential effectiveness of selective PDE5Is in women would occur. Mechanisms analogous to male sexual function apply in female sexual (dys)function. PDE5 and not-otherwise-specified isoforms are present in human genital tissue and can respond to the enhancement of the NO-cGMP axis.

Sildenafil treatment for female sexual disorder (FSD) is associated with clitoral erection, labial swelling, and vaginal lubrication upon sexual stimulation, but these reports are more discordant between subjective experience and physiological arousal. Initial reports of PDE5I-treatment of FSD described potential efficacy, but other reports in FSD of various etiologies were equivocal and raised important questions needing further study.

Open-label reports show improvements with SRI-associated SD treatment.

 9 of 10 women with MDD in remission, without preexisting SD and serotonergic antidepressant-induced anorgasmia, had effective reversal of delayed orgasm. 

Reports from a study of flexible-dose sildenafil for serotonergic antidepressant-associated SD in 100 women following the same protocol as in the study in men indicate that SRI-associated SD in women improved with sildenafil treatment while the patients continued their prescribed antidepressant.

 The women with MDD in remission, without preexisting SD, and SRI-associated SD randomly assigned to sildenafil or placebo (50 or 100 mg) as needed before sexual activity reported significant improvement in SD; 67% of all SRI-associated SD treatment responders were treated with sildenafil and 33% with placebo (

The domains with the most significant improvement were orgasm delay and overall satisfaction. Female treatment responders had higher initial hormone levels of free testosterone and thyroxine. That suggests that PDE5I-treatment responders may have different hormone profiles than nonresponders, and hormonal status is an important marker associated with sildenafil treatment outcomes.

 This trial allowed female patients with SRI-associated SD to maintain treatment adherence and depression remission. The decision by industry to terminate trials pursuing an indication for PDE5I use in women may have been premature, and further studies are needed.

The evidence reviewed here suggests that the PDE5Is may effectively manage SRI-associated SD and support antidepressant treatment adherence in men and women. The message to physicians prescribing antidepressants or other medications with associated SD is to tell their patients not to stop the primary drugs if sexual adverse effects occur, but instead, encourage the patient to discuss these effects, and further assure them that these adverse effects can often be addressed without stopping medication. 

 summarizes recommended clinical guidelines for patients stabilized on a dose of prescribed antidepressant or other medication with sexual adverse effects.

Initial work has been promising, but further study is needed to determine whether PDE5I reversal of adverse medication effects on sexual function can be extended to women, and to other disease states or conditions in which treatment-associated adverse effects are likely to develop. The evidence-based literature for PDE5Is is compelling, particularly when compared with former expert-opinion-driven folk treatments. With the current emphasis on treatment effectiveness, active treatment of treatment-emergent adverse effects would seem to be essential and should be incorporated into efficacy studies. Keeping patients treatment-adherent is a direct and cost-effective way of improving disease management outcomes.

 National Health and Nutrition Examination Survey estimates of total prevalence (both diagnosed and undiagnosed) projected to year 2005. National Ambulatory Care Medical Survey: 1997 summary. 

 Greden JF. The burden of disease for treatment- resistant depression. 

 Nurnberg HG, Thompson PT, Hensley PL. Antidepressant medication change in a clinical treatment setting: a comparison of serotonin reuptake inhibitor effectiveness. 

Nurnberg HG, Hensley PL, Lauriello J. Sildenafil in the treatment of sexual dysfunction induced by selective serotonin reuptake inhibitors: an overview. 

 Depression Guideline Panel. Depression in primary care, vol 2--Treatment of major depression. Rockville, Md: US Department of Health and Human Services, 1993. Agency for Health Care Policy and Research Publication 93-0551.


 Kessler RC, Berglund P, Demler O, et al. The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R). 

Lin EH, von Korff M, Katon W, et al. The role of the primary care physician in patients' adherence to antidepressant therapy. 

 Katon W, von Korff M, Lin EH, et al. Adequacy and duration of antidepressant treatment in primary care. 

 Zimmerman M, Posternak M, Friedman M, et al. Which factors influence psychiatrists' selection of antidepressants? 

Rudkin L, Taylor MJ, Hawton K. Strategies for managing sexual dysfunction induced by antidepressant medication. 

Laumann EO, Paik A, Rosen RC. Sexual dysfunction in the United States, prevalence and predictors. 

Hawton K, Gath D, Day A. Sexual function in a community sample of middle-aged women with partners: effects of age, marital, socioeconomic, psychiatric, gynecological, and menopausal factors. 

Feldman HA, Goldstein I, Hatzichristou DG, et al. Impotence and its medical and psychosocial correlates: results of the Massachusetts Male Aging Study. 

 Hamilton MA. A rating scale for depression. 

H. George Nurnberg, MD

Articles

Options for Management of Serotonin Reuptake Inhibitor-Induced Sexual Dysfunction