Relapse During SSRI Treatment for Depression


Medscape Psychiatry & Mental Health eJournal. 1996;1(5) 

In This Article


The present survey data provide preliminary evidence to suggest that while clinicians widely regard SSRIs as efficacious in the acute treatment of depression, many observe some degree of relapse during ongoing treatment. Respondents cited a modal frequency of relapse with SSRIs as being between 11% to 25% of cases over 1 year. This group of clinicians believed that depressive symptoms most often re-emerged at approximately 6 months after achieving an initial remission. SSRI dosage increases seem necessary at some point in the course of ongoing treatment in half of cases observed by respondents. Only 29% of the sample believed that long-term treatment with SSRIs and TCAs had a comparable course of remission, and 40% regarded relapse as being more likely with SSRIs.

These findings are consistent with data by Fava and colleagues[9] and Nirenberg and associates[7] in suggesting that loss of antidepressant efficacy with SSRIs is a distinct and common occurrence in the long-term treatment of depression, relative to lower relapse rates reported for long-term pharmacotherapy for depression with TCAs.[8]

Clinicians appear to view this attenuated antidepressant response as resolvable in almost all instances. Although a majority indicated that SSRI dosage increases may be the most often-tried and efficacious maneuver, a number of other pharmacologic strategies were reported to be effective; changes to other antidepressants were often cited as effective.

Investigators have speculated as to possible mechanisms that may underlie relapse during SSRI treatment. Cain[6] suggested that fluoxetine may have a therapeutic window, above which the patient may be clinically nonresponsive. This has been linked to the eventual build-up of fluoxetine's N-demethylated active metabolite, norfluoxetine, which may directly impede the antidepressant effect of fluoxetine.[16] Hence, antidepressant response may attenuate over time but could theoretically return following a drug washout or dosage decrease. However, such a mechanism would not readily account for regaining a lost SSRI response by raising the dose.

We previously suggested the possibility that eventual drug tolerance could result from serotonin saturation after prolonged drug administration, leading to down-regulation of CNS serotonin receptors.[10] In line with this possibility, strategies that facilitate the release of presynaptic serotonin (eg, pindolol augmentation with TCAs) may be worthy of further exploration.

A number of limitations must be noted regarding the current questionnaire data: First, the response rate of only 30% among those surveyed raises the possibility for significant nonresponse bias. Although major demographic variables did not separate those who did or did not participate in the survey, the possibility exists that those who responded may have held a particular interest in the pharmacology of affective disorders or relapse in depression, and hence may have drawn upon clinical experiences with less representative patient populations. All practitioners were drawn from a single, large, urban institution, which may not reflect treatment outcomes in other types of clinical settings. In addition, many of the respondents were senior psychiatrists in well-established, full-time clinical practices who saw new patients much less often than long-term, established patients. Thus the relatively low number of patients treated with an SSRI in the study year (16 +/- 8) may not reflect the clinical experience of practitioners who see and treat larger numbers of new patients presenting at or near the onset of their illness. Together, these factors could limit the generalizability of the findings. Patients' impressions of relapse might also differ from those of their physicians, and corroborative data from patients themselves would strengthen the findings. Additionally, all clinicians' observations were culled from their individual experiences working with depressed patients; the retrospective nature and nonsystematic recall of relevant cases leave these observations impressionistic.

Despite those constraints, these survey results point to a prominent awareness among many psychiatrists that relapse in the course of depressive illness is common. Given the current trend among many clinicians to favor SSRIs over TCAs for long-term use, due largely to lesser toxicity and better side-effect profiles, further investigation of the long-term efficacy of SSRIs is warranted.

Directions for future study to elucidate the phenomenon of attenuated response might include greater exploration of the serotonergic basis for depression. In this regard, neurobiologic markers such as tryptophan depletion[17,18] may prove useful for identifying patients whose depressive symptoms arise largely as a consequence of low CNS levels of serotonin. It would be interesting to observe whether patients whose depressive symptoms intensify following tryptophan depletion may be at greater risk for relapse after remission from using an SSRI, based on supersensitivity of serotonergic neurotransmission.

Further prospective studies might also examine whether associated clinical factors, such as those identified by respondents in the current survey, (eg, prior dysthymia or major depression, atypical depressive features, comorbid substance abuse) successfully predict future relapse with SSRIs. Depressive features such as melancholia, for example, have been reported to respond less well initially to SSRIs than to TCAs.[19] It remains to be determined whether characteristics like this, or other symptom-constellations in depression, are predictive of the longevity of a successful treatment response.

Online Resources


  • Treatment: A discussion of treatment options for depressive disorder.

  • Selective Serotonin Reuptatke Inhibitors: A review describing the significant differences between the currently available SSRIs and their role in the treatment of psychiatric disorders.

  • Psychotropic Drug Interactions: A reference document concerning psychotropic drug interactions. Under Specific Drug Interactions, there is a section concerning SSRIs.