RESUMO
BACKGROUND: Stroke patients are at increased risk for acquiring infections in the hospital and risk of readmission. We aimed to examine whether an infection acquired during the initial stroke admission contributes to increased risk of readmission and infection during readmission. METHODS: We performed a retrospective cohort study incorporating all adult ischemic stroke patients from three New York City hospitals from 2006 to 2016. A validated computer algorithm defined infections based on electronically-available laboratory culture data. Multivariable logistic regression was used to evaluate the crude and adjusted association of infections present on admission (IPOA) and healthcare-associated infections (HAI) with 60-day readmissions, and infection during readmission. RESULTS: Among the 10,436 stroke patients, 17% had infections during initial admission of which 52% were IPOA and 48% were HAI. The risk of readmission was significantly higher for those with HAIs (OR = 1.40; 95% CI: 1.20-1.64) and IPOA (OR = 1.26; 95% CI: 1.09-1.47). The presence of infection during the 60-day readmission was also independently predicted by HAI (OR = 3.27; 95% CI: 2.60-4.12) and IPOA (OR = 2.54; 95% CI: 2.01-3.22). Patients with a Gram-negative infection were not at higher odds for readmission compared to patients with a Gram-positive infection (OR 1.07, 95%CI 0.81-1.42). CONCLUSION: Among stroke patients, HAI and IPOA were predictors of readmission within 60 days and infection during readmission.