ABSTRACT
Background and Aims: In Sub-Saharan Africa management of adult patients with febrile illness consists very often of empirical antibacterial and ant malarial treatment. This study examines the frequency, species identification and antibiotic susceptibility of bacterial isolates from blood and determines the frequency of malaria and the proportion of verified malaria cases among presumptively treated patients at a Malawian hospital. Study Design: This is a cross-sectional survey. Place and Duration of Study: Patients were enrolled at the Medical Department of Kamuzu Central Hospital, a referral hospital in Lilongwe, Malawi, between October 2010 and March 2011. Methodology: Patients ≥18 years with an axillary temperature ≥37.5°C were included. Blood cultures, malaria rapid diagnostic tests (RDTs), thick blood smears and HIV testing were performed. Results: 180 patients (58.3% female, median age: 31 years) were enrolled. Out of 157 patients, 89 (56.7%) tested HIV positive. Bacteremia was found in 19 (10.6%) patients including 6 (31.6%) Streptococcus pneumoniae, 6 (31.6%) Escherichia coli and 5 (26.3%) Salmonella enterica (4 Salmonella enterica serotype Typhimurium and 1 Salmonella enterica serotype Typhi). S. typhimurium and E. coli isolates showed frequent resistance to chloramphenicol, ampicillin and cotrimoxazole. Ceftriaxone was given to 110 (61.1%) patients. Malaria was confirmed by positive smear and/or positive RDT(s) in 57 (31.7%) cases. Presumptive antimalarial treatment was administered to 120 (66.7%) patients, however only 54 (45%) of these tested malaria positive. Conclusion: Empirical treatment of bloodstream infections should be based on antibiotic susceptibility of common local pathogens. Clinically suspected malaria should be confirmed by using malaria diagnostic testing before treatment. The use of malaria RDTs has to be carefully supervised and adherence to test results is advisable.