Background: Fever in acute stroke occurs in up to 61% of patients, increases with the severity of stroke, and is a strong predictor of an unfavourable outcome. 
Objective: To determine the occurrence of fever in patients with acute stroke, the practice of antibiotic administration and its association with in-hospital mortality.
Methods: All patients with acute stroke (both ischemic and haemorrhagic) admitted to Neurology services were recruited in the study. The period of study was January 2017 and December 2017. The occurrence of fever, antibiotics (number, type and duration) and in-hospital mortality were recorded. 
Results: There were 177 patients (128 (72.3%) ischemic) in the study period. The mean age was 53.9 (SD: 14.0) years. 123 (69.5%) of the patients were males. Fever was present in 113/174 (64.9%). Fever maximally occurred on second day of ictus. Antibiotics were administered in 104/124(83.9%) of all patients. The median duration of antibiotics given were 9 days (IQR: 5.6; 17.3). The median number of antibiotics administered was 3 (IQR: 2; 5). The median duration of hospital stay was 11 (IQR: 7.0; 25.3) days. The in-hospital mortality was 29/155 (18.7%). There was no association of fever and antibiotic administration with in-hospital mortality (p value=0.413, OR 1.45 (0.59; 3.56))
Conclusion: The study depicts the frequent use of antibiotics in stroke patients with fever. There was however, no association with of fever and antibiotic administration with in-hospital mortality. This challenges the use of antibiotics in febrile stroke patients, as the outcome may be primarily determined by stroke severity.

FEVER AND ANTIBIOTIC USE IN PATIENTS WITH ACUTE STROKE: A COHORT STUDY