Background and Purpose- The FOCUS trial (Fluoxetine or Control Under Supervision) showed that fluoxetine did not improve modified Rankin Scale scores (mRS) but increased the risk of fractures. We aimed to describe the fractures, their impact on mRS and factors associated with fracture risk. Methods- A United Kingdom, multicenter, parallel-group, randomized, placebo-controlled trial. Patients ≥18 years with a clinical stroke and persisting deficit assessed 2 to 15 days after onset were eligible. Consenting patients were allocated fluoxetine 20 mg or matching placebo for 6 months. The primary outcome was the mRS at 6 months and secondary outcomes included fractures. Results- Sixty-five of 3127 (2.1%) patients had 67 fractures within 6 months of randomization; 43 assigned fluoxetine and 22 placebo. Fifty-nine (90.8%) had fallen and 26 (40%) had fractured their neck of femur. The effect of fluoxetine on mRS (common odds ratio =0.951) was not significantly altered by excluding fracture patients (common odds ratio =0.961). Cox proportional hazards modeling showed that only age >70 year (hazard ratio =1.97; 95% CI, 1.13-3.45; P=0.017), female sex (hazard ratio =2.13; 95% CI, 1.29-3.51; P=0.003), and fluoxetine (hazard ratio =2.00; 95% CI, 1.20-3.34; P=0.008) were independently associated with fractures. Conclusions- Most fractures resulted from falls. Although many fractures were serious, and likely to impair patients' function, the increased fracture risk did not explain the lack of observed effect of fluoxetine on mRS. Only increasing age, female sex, and fluoxetine were independent predictors of fractures. Clinical Trial Registration- URL: http://www.controlled-trials.com. Unique identifier: ISRCTN83290762.
Keywords: fluoxetine; fractures; odds ratio; randomi; risk; zed controlled trial.