Howard S. Kirshner, MD
Do selective serotonin reuptake inhibitors (SSRIs) improve recovery after stroke?
Included studies compared any SSRI (e.g., fluvoxamine, fluoxetine, sertraline, citalopram, or paroxetine), at any dose and for any duration or reason, with placebo or usual care in stroke survivors within 12 months of stroke or a reported mean time < 12 months since stroke (sudden-onset focal neurologic disturbance, assumed to be vascular in origin, and lasting > 24 h), and reported ≥ 1 outcome of interest. Study drugs with mixed pharmacologic effects that included SSRI actions were excluded. Primary outcomes were dependence and disability. Secondary outcomes included death, gastrointestinal effects, neurologic deficit, and depression.
Cochrane Depression Anxiety and Neurosis Group Trials Register (Nov 2011); Cochrane Central Register of Controlled Trials (2011, Issue 8); Cochrane Stroke Group Trials Register, MEDLINE, EMBASE/Excerpta Medica, CINAHL, AMED (Allied and Complementary Medicine), PsycINFO (all to August 2011); PsycBITE (Psychological Database for Brain Impairment Treatment Efficacy) (March 2012); clinical trials registers; pharmaceutical Web sites; reference lists; reviews; and citations of included studies were searched for randomized controlled trials (RCTs). 52 RCTs (n = 4059, mean age range 55 to 77 y) studying fluoxetine (28 RCTs), sertraline (7 RCTs), paroxetine (10 RCTs), citalopram (5 RCTs), escitalopram (1 RCT), and sertraline or fluoxetine (1 RCT) were included in the meta-analysis. Risk for bias was low for randomization (14 RCTs), concealment (9 RCTs), blinding (16 RCTs), and loss-to-follow-up (30 RCTs).
The main results are in the Table.
In patients recovering from stroke, selective serotonin reuptake inhibitors somewhat improved measures of dependence and disability.
Selective serotonin reuptake inhibitors (SSRIs) vs placebo or usual care (control) after stroke*
*NS = not significant; other abbreviations defined in Glossary. Weighted event rates, RRR, RRI, NNT, NNH, and CI calculated from control event rates and risk ratios in article using a random-effects model.
†Based on a modified Rankin scale of 3 to 5, score range 0 to 6, lower = better.
‡All results favor SSRIs. To correct for differences in the direction of scales (i.e., higher values = better outcomes for some scales but worse outcomes for other scales), the mean values of 1 set of trials was multiplied by −1.
Kirshner HS. Review: SSRIs somewhat improve dependence and disability after stroke. Ann Intern Med. ;158:JC12. doi: 10.7326/0003-4819-158-6-201303190-02012
Download citation file:
Published: Ann Intern Med. 2013;158(6):JC12.
Results provided by:
Copyright © 2018 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use