FLAME (Stroke)

From Wiki Journal Club
Jump to navigation Jump to search
Chollet F, et al. "Fluoxetine for motor recovery after acute ischaemic stroke (FLAME): a randomised placebo-controlled trial". The Lancet Neurology. 2011. 10(2):123-130.
PubMedFull text

Clinical Question

Does the early administration of fluoxetine in addition to conventional physiotherapy improve motor recovery following ischemic strokes?

Bottom Line

Early administration of fluoxetine (initiated 5-10 days after stroke onset) in addition to conventional physiotherapy improves motor recovery in patients with moderate to severe motor deficits due to ischemic stroke.

Major Points

Several small case-series suggest that early SSRIs administration may promote motor recovery after ischemic stroke. This effect is proposed to be due to a non-vascular mechanism, possibly related to a serotonin-mediated suppression of post-stroke hyperexcitability in the unaffected hemisphere leading to improved neuroplasticity.[1]

Published in 2011, the Fluoxetine for Motor Recovery after Acute Ischaemic Stroke (FLAME) study followed 113 patients after acute stroke randomized to early fluoxetine or placebo. At 90 days, the group randomized to fluoxetine experienced higher Fugl-Meyer Motor Scale (FMMS) scores (34.0 vs. 24.3; P=0.003). The medication was well-tolerated.

While the FLAME trial is generally regarded as a well-designed, well-executed and sufficiently powered trial, its restrictive inclusion criteria, small sample size, and short follow-up limit the generalization of its findings in clinical practice. The results of the on-going Australian AFFINITY trial (anticipated to conclude in 2019, anticipated n=~1,600) may be better able to address the role of early fluoxetine for benefit to cognition, improving fatigue, mood and general functional ability in addition to motor recovery after ischemic stroke.


As of August 2015, no guidelines have been published that reflect the results of this trial.


  • Multi-centered, double-blinded, randomized control trial
  • N=113
    • Fluoxetine (n=59, 1 death, 1 withdrawal due to severe hypoxia)
    • Placebo (n=59, 1 death, 2 withdrawals due to pulmonary embolism and kidney tumor)
  • Setting: 9 stroke units in France
  • Enrollment: 2005-2009
  • Mean follow-up: 90 days after enrollment
  • Analysis: Intention-to-treat (presumably, not explicitly stated)
  • Primary outcome: Motor recovery by day 90


Inclusion Criteria

  • Acute ischemic stroke causing hemiparesis or hemiplegia 5-10 days
  • Age 18-85 years
  • Fugl-Meyer Motor Scale (FMMS) score ≤55 at time of randomization

Exclusion Criteria

  • Significant premorbid disability which can interfere with outcome assessment: Motor deficits from prior neurological insults, severe aphasia (productive or receptive), NIH Stroke Scale (NIHSS) >20
  • Patients with a clinically diagnosed depression or Montgomery Asberg Depression Rating Scale (MADRS) score >19
  • Patients on treatment with antidepressants, MAOI, neuroleptics or benzodiazepine within 1 month of randomization
  • Planned CEA
  • Renal or hepatic insufficiency, or another major illness which was anticipated to prevent follow up
  • Pregnancy

Baseline Characteristics

From the fluoxetine group.

  • Demographics: Age 66 years, 63% female
  • BMI: 26 kg/m2
  • CVA details:
    • Risk factors: DM 24%, HTN 66%, HLD 61%, smoker 51%, cardiac disease 58%, AF 10%, prior CVA 17%
    • Location: Carotid territory 86%, vertebrobasilar 10%, lacunar 3%
    • Severity:
      • FMMS score: Total 17, upper limb 5, lower limb 12
      • NIHHS score: Total 13, motor 9.9
      • Modified Rankin score:
        • 0-2: 0%
        • 3 (moderate disability): 3%
        • 4 (moderate-severe disability): 42%
        • 5 (severe disability): 54%
  • IV thrombolytics: 36%
  • MADRS score (depression rating): 6
  • Time from stroke to study treatment: 9 days


  • Randomization to a group:
    • Fluoxetine - Fluoxetine 20 mg po qday
    • Placebo
  • All participants underwent conventional physiotherapy and stroke care
  • If depression occurred during the 90 day follow up period:
    • Patients were encouraged to be placed on open label fluoxetine 20 mg po qday in addition to their study medication
    • If another antidepressant other than fluoxetine was used, the study medication was discontinued


Presented as fluoxetine vs. placebo. Adjusted scores control for age, history of CVA, and FMMS at inclusion. All outcomes are at day 90.

Primary Outcome

Motor recovery

Change in FMMS from baseline. The FMMS score is out of 100, higher numbers indicate greater motor recovery.

+36.4 vs. +21.9 (difference 14.5; 95% CI 7.3-21.6)
Adjusted: +34.0 vs. +24.3 (difference 9.8; 95% CI 3.4-16.1; P=0.003)

Secondary Outcomes

Upper limb recovery

Change FMMS subscore from baseline.

+24.2 vs. +11.8 (difference 12.4; 95% CI 5.9-18.9)
Adjusted: +22.9 vs. +13.1 (difference 9.7; 95% CI 3.6-15.9; P=0.002)
Lower limb recovery

Change FMMS subscore from baseline.

+12.2 vs. +10.1 (difference 2.1; 95% CI -0.4 to 4.6)
Adjusted: +12.8 vs. +9.5 (difference 3.3; 95% CI 0.8-5.7; P=0.010)
Adjusted NIHSS (n=112)
Total: 5.8 vs. 6.9 (P=0.151)
Motor: 4.7 vs. 6.3 (P=0.012)
Modified Rankin Scale 0-2 (n=112)
26% vs. 9% (P=0.015)
Adjusted: 34% vs. 11% (P=0.021)
Change in MADRS score (n=110)
0 vs. +3.1
Adjusted: -0.1 vs. +3.2 (P=0.032)

Additional Analyses

Study medication compliance
88 vs. 89 tablets taken in prior 90 days (P=0.722)
29% vs. 7% (P=0.002)

Subgroup Analysis

The primary outcome was still significant for comparison of TPA vs. no TPA.

Adverse Events

Similar between groups.


  • Small study
  • Unclear long-term effects of the intervention


French National Program for Clinical Research

Further Reading