Medical Complications Following Cerebral Infarction: Frequency, Timing, and Predictors in a Population-Based Study.

Abstract

P142 The purpose of this study was to determine the frequency, timing of occurrence, and predictors of selected complications and medical disorders following cerebral infarction (CI) in a population-based study. The Rochester Epidemiology Project medical records linkage system was used to identify all first cerebral infarctions (CI) among Rochester, Minnesota residents during 1985–1989. The comprehensive medical records were reviewed; past history and demographic data, post-CI complications, and cause of death were recorded. CI subtype, severity, and site were determined. The frequency, timing and predictors of complications after CI were determined using the Kaplan-Meier product limit method and Cox proportional hazards regression analysis. A total of 444 first cerebral infarctions (181 men, 263 women; mean age 75.3 years) were identified among residents of Rochester, Minnesota, during 1985–1989. During a follow-up of up to 10 years, some of the complications and other medical disorders included chest infection in 194 (44%), a fall requiring medical attention in 122 (27%), myocardial infarction in 84 (19%), seizure in 38 (9%), and PE in 14 (3%). The risk of chest infection after CI was 11% at 30 days, 22% at 1 year, and 40% at 5 years. The risk of seizure was 2% at 30 days, 4% at one year, and 9% at 5 years. Regression analysis revealed that stroke severity, prior history of COPD, older age, swallowing dysfunction requiring a PEG, and stroke recurrence were significant independent predictors of chest infection after first CI. Use of aspirin after CI was inversely related to occurrence of chest infection. Complications are common after cerebral infarction. Some of the complications are more likely to occur early after the CI. Increased awareness of these issues may lead to enhanced complication prevention techniques, and surveillance for the complications. The study provides unique population-based data regarding predictors of the important causes of morbidity and mortality following cerebral infarction, and may be compared to other non-population based data.