Predictors of early infection in cerebral ischemic stroke

Background: Infection is the most common complication of stroke. Aim: To determine the risk factors and predictors of post-stroke infection (PSI), which developed within 7 days from the onset of acute ischemic stroke. Subjects: The study included 60 ischemic stroke patients admitted in the Neurology Department of Zagazig University, Egypt, who were subdivided into: [Non Stroke Associated Infection group (nSAI); 30 patients having stroke without any criteria of infection within 7 days from the onset and Stroke Associated Infection group (SAI); 30 patients having stroke with respiratory tract infection (RTI) or urinary tract infection within 7 days], in addition to 30 healthy sex and age-matching subjects as control. Methods: All the patients had a detailed history taking, thorough clinical general and neurological examination, laboratory tests (Urine analysis & urine culture, blood sugar, lipid profile and serum tumor necrosis factor-alpha (TNF-α) and interleukin (IL)-10), a chest radiography to assess RTI and brain computed tomography (CT) to exclude the hemorrhagic stroke and to confirm the ischemic stroke. Results: SAI patients were found to be significantly older with higher baseline blood glucose level. Also the number of patients with tube feeding, lower conscious level, more stroke severity and more large size infarcts were significantly higher in SAI patients. There was a significant elevation in the IL-10, a significant decrease in the TNF-α and a significant decrease in the TNF-α/ IL-10 ratio, in the SAI group. The baseline serum level of IL-10 ≥ 14.5 pg/ ml and size of infarct area > 3.5 cm3 were found to be the independent predictors of PSI. Conclusion: Patients with older age, tube feeding, lower conscious level, worse baseline stroke severity, large cerebral infarcts in CT scan, and increased IL-10 serum level were more susceptible to infection. The baseline serum level of IL-10 ≥ 14.5 pg/ ml and the size of infarct area > 3.5 cm3 were the independent predictors of PSI.