Selective Serotonin Reuptake Inhibitor Discontinuation Syndrome
Abstract & Commentary
Synopsis: Abrupt discontinuation of SSRIs, mimicking both noncompliance and intentional medication discontinuation, is commonly associated with diverse somatic symptoms. Interruption of fluoxetine is associated with fewer adverse events.
Source: Rosenbaum JF, et al. Selective serotonin reuptake inhibitor discontinuation syndrome: A randomized clinical trial. Biol Psychiatry 1998;44:77-87.
Recent reports describe a series of symptoms following discontinuation of selective serotonin reuptake inhibitors (SSRIs). The most common symptoms include dizziness, headache, nausea, vomiting, diarrhea, insomnia, irritability, electric shock sensations, and depressed mood and anxiety. Noting that discontinuation-emergent symptoms may occur with noncompliance as well as with physician-initiated medication discontinuation, and hypothesizing that the symptom complex may be less common with longer half-life drugs, Rosenbaum and colleagues present the first prospective, placebo-controlled study of the phenomena.
The Subjects consisted of 242 patients with depression adequately controlled with standard doses of fluoxetine (Prozac), sertraline (Zoloft), or paroxetine (Paxil) for a minimum of four months. Treatment was interrupted for 5-8 days with double-blind placebo substitution. To make sure the study was blinded, placebo substitution was randomized to occur at one of two predetermined follow-up visits, and 17% of the sample were randomized not to receive placebo substitution at all.
Symptoms were assessed using standard assessments and a 43-item questionnaire designed to evaluate expected discontinuation-emergent symptoms. Assessments were administered by clinicians that were blinded to the timing of placebo substitution or continuous SSRI treatment. Consistent with Rosenbaum et al’s hypothesis, the mean number of discontinuation-emergent symptoms was significantly lower in fluoxetine-treated patients than in either the sertraline-treated or paroxetine-treated patients (both P < 0.001). The mean number of events in the sertraline-treated patients was significantly lower than in the paroxetine-treated patients (P = 0.020), but, on other measures, differences between the sertraline and paroxetine groups were not statistically significant. The most frequently reported symptoms varied somewhat by antidepressant medication. In the fluoxetine-treated group, common symptoms consisted of the following: worsened mood (22%), irritability (17%), agitation (16%), confusion (14%), headache (14%), fatigue (16%), and emotional lability (13%). In sertraline-treated patients, common symptoms consisted of the following: worsened mood (28%), irritability (38%), agitation (37%), dizziness (29%), confusion (23%), headache (31%), nervousness (31%), crying (26%), fatigue (23%), emotional lability (31%), trouble sleeping (22%), anger (28%), dreaming (25%), nausea (14%), amnesia (17%), sweating (17%), depersonalization (17%), muscle aches (14%), unsteady gait (15%), panic (15%), sore eyes (14%), shaking (11%), muscle tension (14%), and chills (11%). In the paroxetine-treated group, common symptoms consisted of the following: worsened mood (45%), irritability (35%), agitation (31%), dizziness (50%), confusion (42%), headache (34%), nervousness (34%), crying (40%), fatigue (32%), emotional lability (26%), trouble sleeping (39%), dreaming (37%), anger (29%), nausea (40%), amnesia (24%), sweating (24%), depersonalization (21%), muscle aches (23%), unsteady gait (23%), panic (21%), sore eyes (15%), diarrhea (24%), shaking (21%), muscle tension (11%), and chills (18%). Increased depression was confirmed by increases in the Hamilton Depression Rating Scale and the Montgomery Asberg Depression Rating Scales. The incidence of depression relapse during medication interruption was 2% in fluoxetine-treated patients, 14% in sertraline-treated patients, and 27% in paroxetine-treated patients.
Comment by Lauren B. Marangell, MD
This is a well-designed prospective study of a randomized, double-blind interruption of antidepressant medication, with a systematic method for documenting discontinuation-emergent signs and symptoms. As Rosenbaum et al point out, there appears to be a meaningful relationship between plasma half-life and the occurrence of discontinuation-emergent symptoms, which is expected because the rate of decrease in serum concentration in the absence of continuous dosing is more precipitous for shorter half-life drugs. The findings of the study are clinically relevant both for planned medication discontinuation and for evaluating patients who are receiving ongoing antidepressant treatment. When antidepressants are discontinued at the end of a treatment period, the current data underscore the importance of a planned medication taper for most antidepressants (perhaps all except fluoxetine). In addition, patients should be advised of potential discontinuation-emergent symptoms and reassured that if these symptoms develop, they will subside within 1-2 weeks. Perhaps more important is the issue of intermittent noncompliance, which is all too common in clinical practice. The current data underscore the need to consider noncompliance when a patient who is taking a SSRI presents with complaints ranging from sore eyes to dizziness. Finally, although the current study only addresses fluoxetine, sertraline, and paroxetine, similar symptoms have been noted with other short half-life serotonergic antidepressants, including venlafaxine (Effexor and Effexor XR), nefazodone (Serzone), and fluvoxemine (Luvox).