Poststroke Infections are an Independent Risk Factor for Poor Functional Outcome after Three-months in Thrombolysed Stroke Patients

Abstract

Background Infections are common complications in patients with acute ischemic stroke; however, the pathophysiology of the stroke-induced immunodepression is still under debate. Although it has been shown that increased mortality and longer hospital stay are associated with the presence of poststroke infections, it remains unclear if early poststroke infections occurring in the first seven-days have an effect on the overall functional outcome. Aims Aim of our study was to identify the frequency of poststroke infections in thrombolysed stroke patients and to analyze their effect on the outcome after three-months. Methods From 1998 to 2011, all patients in our institution undergoing thrombolysis for acute ischemic stroke were included into a prospective database. Baseline variables, clinical, radiographic, and laboratory data were collected prospectively. Outcome measures included symptomatic intracerebral hemorrhage per European-Australasian Acute Stroke Study II criteria, mortality, and modified Rankin Scale at three-months. Logistic regression models were used to identify independent predictors for poor outcome where appropriate. Results One thousand sixteen patients were included; of them, 36·3% had an infection during the first week. Pneumonia (9·6%) and urinary tract infections (5·4%) were most frequent. Severity of stroke ( P < 0·0001), infarct size ( P < 0·0001), atrial fibrillation ( P = 0·005), and cardio embolic cause of stroke ( P < 0·0001) were associated with infections. Age (odds ratio 1·089, 95% confidence interval 1·064–1·115, P < 0·0001), severity of stroke (odds ratio 1·111, 95% confidence interval 1·073–1·149; P < 0·0001) history of diabetes (odds ratio 0·555. 95% confidence interval 0·357–0·864; P = 0·009), infarct size (odds ratio 4·256 95% confidence interval 2·697–6·745; P < 0·0001), infections (odds ratio 1·548, 95% confidence interval 1·008–2·376; P = 0·046), and symptomatic intracerebral hemorrhage were independent predictors for poor outcome after three-months. Conclusions In our cohort of thrombolysed stroke patients, poststroke infections were frequent in patients with severe cardio embolic stroke, a large infarct, and a longer hospital stay; those patients have a higher risk of infection and a poorer functional outcome after three-months. This risk increases after occurrence of symptomatic intracerebral hemorrhage. Prevention of infection with antibiotic therapy or other prophylactic treatment could potentially lead to a better functional outcome and further randomized studies on this aspect are needed.