RESUMO
BACKGROUND: Stroke prognosis is related to the multimorbidity profile. Moreover, performing an individual evaluation of most common cerebrovascular risk factors (CVRF) not always identifies patients with poor prognosis. Thus, we decided to evaluate multimorbidity profile, focusing on the Charlson Comorbidity Index (CCI) validated by Goldstein for ischaemic stroke (IS) patients, a score that measures a burden of comorbidities and its related mortality in the long-term survival of the EMMA Study (Study of Stroke Mortality and Morbidity). METHODS: Nine hundred fifty-nine individuals (median age 70 years) had validated data on the diagnosis of IS, main CVRF and clinical comorbidities pre index event such as atrial fibrillation (AF), stroke recurrence, diabetes, hypertension, heart failure and cancer. CCI modified by Goldstein was calculated, which includes 17 clinical conditions with scores ranging from 1 to 6 (0-31 points). Survival analyses were performed by Kaplan-Meier curves and Cox logistic regression models (cumulative hazard ratio [HR] with [95% CI]) for all-cause mortality at 180 days, and every 3 years up to 9-year follow-up. Mortality analyzes were performed by CCI categorized according to weight added to comorbidities (Reference group: zero, moderate: 1, severe: 2 and very severe: ≥3 points). We also tested the modification effect of AF and stroke recurrence including these conditions in the CCI. RESULTS: The overall survival rate was 47% (508 deaths/959). The worst survival (577, 95% CI 381-773 days) and the highest risk of death after stroke were observed in the very severe CCI group (HR 3.18; 95% CI 2.16-4.69) up to 9 years. The inclusion of previous AF and stroke in the CCI slightly increased the risk of death for very severe CCI (HR 3.27; 95% CI 2.07-5.18). CONCLUSIONS: A high burden of comorbidities represented an independent predictor of poor prognosis increasing the risk of dying by 2 to 3 times among IS up to 9 years in the EMMA study. The inclusion of other CVRF such as AF and stroke recurrence slightly modified all-cause mortality risk.