Prevalence and Prognostic Features of ECG Abnormalities in Acute Stroke

2017 
Abstract Background Africa has a growing burden of stroke with associated high morbidity and a 3-year fatality rate of 84%. Cardiac disease contributes to stroke occurrence and outcomes, but the precise relationship of abnormalities as noted on a cheap and widely available test, the electrocardiogram (ECG), and acute stroke outcomes have not been previously characterized in Africans. Objectives The study assessed the prevalence and prognoses of various ECG abnormalities among African acute stroke patients encountered in a multisite, cross-national epidemiologic study. Methods We included 890 patients from Nigeria and Ghana with acute stroke who had 12-lead ECG recording within first 24 h of admission and stroke classified based on brain computed tomography scan or magnetic resonance imaging. Stroke severity at baseline was assessed using the Stroke Levity Scale (SLS), whereas 1-month outcome was assessed using the modified Rankin Scale (mRS). Results Patients' mean age was 58.4 ± 13.4 years, 490 were men (55%) and 400 were women (45%), 65.5% had ischemic stroke, and 85.4% had at least 1 ECG abnormality. Women were significantly more likely to have atrial fibrillation, or left ventricular hypertrophy with or without strain pattern. Compared to ischemic stroke patients, hemorrhagic stroke patients were less likely to have atrial fibrillation (1.0% vs. 6.7%; p = 0.002), but more likely to have left ventricular hypertrophy (64.4% vs. 51.4%; p = 0.004). Odds of severe disability or death at 1 month were higher with severe stroke (AOR: 2.25; 95% confidence interval: 1.44 to 3.50), or atrial enlargement (AOR: 1.45; 95% confidence interval: 1.04 to 2.02). Conclusions About 4 in 5 acute stroke patients in this African cohort had evidence of a baseline ECG abnormality, but presence of any atrial enlargement was the only independent ECG predictor of death or disability.
    • Correction
    • Cite
    • Save
    • Machine Reading By IdeaReader
    37
    References
    0
    Citations
    NaN
    KQI
    []