Bahar M. Beaver, MD
The issue of whether SSRIs are beneficial in overall post-stroke recovery is a yet unsettled question. The FLAME (Fluoxetine for Motor Recovery after Acute Ischemic Stroke) study in 2011 demonstrated that fluoxetine may help with improved motor recovery 3 months after stroke. However, the small sample size (118) was only one of many limitations of that trial. The authors of the TALOS study, published in the November issue of Stroke, set out to evaluate another SSRI (citalopram) and its effects on post-stroke recovery. In this double-blind, randomized, placebo-controlled study, 642 patients with ischemic stroke were randomly assigned to either citalopram 20mg (or 10mg if clinically indicated) or placebo once daily for 6 months. Study medication was started within 7 days after last known normal. Patients were followed for a total of 6 months using both in-person clinical evaluations and phone interviews. In the final analysis population, the citalopram group consisted of 268 patients, and the placebo group consisted of 284 patients. The mean age was 68 in both groups. The mean mRS score was 0.3 in each group. The mean NIHSS score was 5.3 in the citalopram group, and 4.8 in the placebo group.
The authors evaluated two co-primary outcomes: improvement in functional disability (mRS score) from one to six months, and composite vascular end point (TIA/stroke/MI) or vascular mortality during the first six months after stroke. In this publication, the authors did not report secondary outcome measures on cognition, prevention of depression, and well-being. When analyzing the co-primary outcomes, the study investigators did not find a difference between the two groups on any of the multiple outcome measures, including mRS at one and six months, any vascular event, and vascular mortality. There was no significant difference in neurologic, cardiovascular, hemorrhagic adverse events, or mortality when comparing the two treatment groups. However, there was a statistical difference in depression (p = 0.007) and other psychiatric symptoms (p = 0.04) between the two groups, with the patients in the citalopram group having a lower incidence of both.
Though the findings of this study are overall non-significant, there remains some hope in continuing to investigate the beneficial effects of SSRIs for stroke patients. One specific area of improvement with this study in particular may be a more detailed evaluation of functional outcomes. The mRS, though familiar and efficient, is not the most detailed assessment tool. Perhaps using a more extensive scale, such as the FIM (functional independence measure) instrument, in collaboration with our physical medicine and rehabilitation colleagues would provide more insight as to the true degree of recovery in post-stroke patients. Another weakness of this trial reported by the authors is the relatively mild stroke symptoms (mean NIHSS of 5) at the start of the study in the two groups. The authors highlighted that the low level of disability at the beginning of the trial period may have masked the overall potential for improvement in these patients. The beneficial effects of SSRIs in the nervous system have been widely reported, as have the potential detrimental effects. Owing to this discrepancy and the reality of post-stroke depression being so common in our patients, further studies are warranted in elucidating the true effect of SSRIs in post-stroke outcomes.
