Background: In epidemiologic research, patients with obesity have, on average, a lower risk for recurrent cardiovascular events after ischemic stroke compared with non-obese patients. Despite this “obesity paradox,” we hypothesized that clinical features associated with more severe metabolic disease would identify overweight and obese stroke patients at high risk for recurrent vascular events.
Methods: The IRIS trial examined the efficacy of pioglitazone compared with placebo, for prevention of stroke/ myocardial infarction (MI) among non-diabetic insulin resistant patients with a recent ischemic stroke or TIA. Patients were followed for a median of 4.8 years. Among 3,707 participants, we first examined risk of recurrent stroke/MI by obesity status at baseline (overweight or obese [BMI ≥25] vs <25) using Cox proportional hazards models with and without adjustment for sociodemographic factors, smoking and vascular disease history. Next, for 3,142 participants with BMI ≥ 25, a Cox model including 7 baseline features (systolic blood pressure, C-reactive protein, HOMA-IR, hemoglobin A1C, waist circumference, triglycerides and high density lipoprotein) was used to stratify patients into tertiles of “metabolic risk”. The hazard of stroke/MI was then calculated across risk tertiles(T).
Results: Overweight and obese participants had a lower incidence of stroke or MI compared with non-obese patients (10.1% vs.12.6%; hazard ratio [HR] 0.77, [0.59-0.99]) and the difference was attenuated after adjustment (HR 0.85 [0.66-1.11]). Among overweight and obese patients, a metabolic risk score identified patients at higher risk of stroke or MI (T3 12.2%, T2 9.2%, T1 9.0%; T3 vs. T1 HR 1.40 [1.08, 1.83]). This association remained significant after adjustments (HR 1.33 [1.02, 1.75]).
Conclusion: Although overweight and obese patients may on average have lower risk of recurrent stroke or MI after an ischemic stroke or TIA compared with non-obese individuals, those with advanced metabolic impairment also have a high absolute risk of recurrence.