Benefits with Fluoxetine Therapy after Ischemic Stroke

Honolulu—Data from the phase 3 FLAME (Fluoxetine in Motor Recovery of Patients with Acute Ischaemic Stroke) trial demonstrate that starting patients on a 20-mg dose of daily fluoxetine promptly after an ischemic stroke contributes to better recovery of motor function at 3 months. As a result of the improved function, François Chollet, MD, PhD, head of the neurovascular department at Toulouse University Hospital and chief of the Toulouse Institute for Neurosciences in France, said participants taking fluoxetine were also more likely than those given placebo to be independent in activities of daily living at 90 days. Dr. Chollet, a lead investigator for the FLAME trial, presented his team’s findings at the AAN meeting. Dr. Chollet began by discussing the French investigators’ rationale for evaluating fluoxetine—a selective serotonin reuptake inhibitor (SSRI) typically prescribed to treat depression and other mood disorders—in stroke patients. He noted that the plasticity of the brain often allows it to reorganize itself in response to damage, resulting in partial or even full recovery of some neurologic functions. Over the past few decades, researchers have examined whether drugs and other interventions can be used to shape the brain’s remodeling process after stroke, with varying degrees of success. Dr. Chollet said the decision to use an SSRI stemmed from a combination of animal studies demonstrating that the serotonergic system facilitates motor output and findings from a handful of in vivo and clinical studies suggesting that SSRIs have positive effects on motor function recovery following stroke. He pointed to a 2001 clinical trial that correlated single-dose fluoxetine 20 mg with improved hand function after stroke and overactivation of motor cortices on magnetic resonance imaging. “This was the sort of proof of concept that we were seeking,” Dr. Chollet explained. FLAME’s primary outcome measure was mean change in the Fugl-Meyer motor scale (FMMS) score from baseline to trial conclusion at 90 days. Secondary end points included mean change in National Institutes of Health Stroke Scale (NIHSS), modified Rankin Scale (mRS), and MontgomeryÅsberg Depression Rating Scale (MADRS) assessments from study outset to completion. Patients were eligible for the double-blind, placebo-controlled trial if they experienced a recent acute ischemic stroke resulting in hemiparesia or hemiplegia, producing moderate-to-severe motor deficit as indicated by a score ≤55 on the FMMS. Individuals scoring >20 on the NIHSS were excluded, along with those with residual disability from an earlier stroke or with premorbid disability or cognitive deficits serious enough to hinder assessments. Patients with clinical depression or treated with antidepressants, neuroleptics, or benzodiazepines in the month prior to randomization were also excluded. A total of 118 stroke patients from 9 stroke centers across France were enrolled within 5 to 10 days of stroke onset and randomized to fluoxetine 20 mg/day (n=59) or placebo (n=59) for 90 days. Dr. Chollet said all patients received standard care from the facility’s inpatient stroke team and physiotherapy. A physiotherapist administered motor function tests at baseline and again at 30 days and 90 days postrandomization. NIHSS, mRS, and MADRS scores were also assessed according to this timeline. After censoring patients who died or withdrew, the 90-day statistical analysis included data for 56 patients assigned to placebo and 57 given fluoxetine. Dr. Chollet said the analysis corrected for some differences in baseline characteristics between the groups. Patients in the fluoxetine group tended to be older than those in the placebo arm (mean age, 66.4 vs 62.9 years, respectively), and more patients randomized to fluoxetine had a prior stroke (16.9% vs 6.8%). Another adjustment accounted for differences in mean FMMS score at baseline, which was higher in the fluoxetine group than the placebo group (17.1±11.7 vs 13.4±8.8, respectively). FMMS scores at 3 months demonstrated significant improvement in overall motor function among patients treated with fluoxetine. “This is true for the [over]all score and for the upper limb and lower limb part,” Dr. Chollet said. The fluoxetine group had an adjusted mean gain of 34.0 points (95% confidence interval [CI], 29.7-38.4) in overall FMMS scores compared with 24.3 points (95% CI, 19.9-28.7) in the placebo arm (P=.003). FMMS scores reflected a mean gain of 22.9 points (95% CI, 18.6-27.1) in upper limb motor function with fluoxetine compared with 13.1 points (95% CI, 8.9-17.4) with placebo (P=.002); patients recovered slightly less lower limb motor function, with a mean increase of 12.8 points (95% CI, 11.1-14.5) in the fluoxetine group versus 9.5 points (95% CI, 7.8-11.2) in the placebo arm (P=.010). Minimal difference in total NIHSS scores was observed between the fluoxetine and placebo groups at 90 days (5.8 vs 6.9, respectively). “However, if we look at the motor items of the NIHSS, there is a statistically significant difference in favor of fluoxetine,” Dr. Chollet said. Patients treated with fluoxetine had a mean motor score on the NIHSS of 4.7 compared with 6.3 for patients given placebo (P=.012). Investigators found no significant difference in MADRS scores between the groups, although the adjusted mean change slightly favored fluoxetine over placebo ( –0.1 vs +3.2, respectively; P=.032). Researchers identified 17 patients in the placebo group as clinically depressed compared with 4 patients in the fluoxetine group (P=.002), and Dr. Chollet said this might suggest that fluoxetine protected against depression. Fluoxetine was also associated with improvement in mRS scores. “If we look at Rankin score 1 or 2, which means the patient is independent in daily living activities, the number of patients independent is higher—significantly higher—in the fluoxetine group than in the placebo group,” said Dr. Chollet, noting that 26.3% of patients given fluoxetine had an mRS score of 0 to 2 compared with only 8.9% of patients taking placebo (P=.015). Rates of serious adverse events were low, and Dr. Chollet said fluoxetine was well tolerated. Of the most common adverse events, nausea, diarrhea, and partial seizure were more prevalent among fluoxetine users; whereas a greater proportion of patients given placebo experienced hepatic enzyme disorders, psychiatric disorders, and insomnia. Both arms reported equal rates of hyponatremia and abdominal pain. Dr. Chollet cited FLAME’s small patient sample as a limitation of the study as well as the fact that patients were selectively recruited according to degree of motor deficiency rather than enrolled consecutively. He also cited the lack of follow-up beyond 3 months as a concern. Because fluoxetine has a hormonal target rather than a vascular one such as intravenous thrombolysis with artery deocclusion, which Dr. Chollet said is the only treatment currently validated for the acute phase of stroke, it might produce benefits in areas such as mood, motivation, and attention. “One can accept that if attention is better and motivation is better, patients participate better in their rehabilitation program,” Dr. Chollet said. He concluded that the drug is now in the public domain, and FLAME’s positive findings suggest its use in stroke is probably in the public health interest, but that more studies are needed.