Outcome of Pneumonia in children admitted at Kenyatta National Hospital

Abstract

Background: Acute lower respiratory tract infections cause considerable morbidity and mortality in children under- 5 years of age especially in developing countries. This is unlike in developed countries where outcome of pneumonia has been shown to improve over time possibly due to improved treatment as well as preventive measures. Locally at Kenyatta National Hospital (KNH), mortality rate due to pneumonia has remained high in the last two decades. This poor outcome has been attributed to some factors such as severity of pneumonia which tend to be of bacterial origin, malnutrition and HIV infection which are more prevalent in developing countries. The extent to which these factors influence pneumonia outcome at KNH is not known. Objective: To describe short term mortality outcome of children admitted with severe and very severe pneumonia at KNH as well as its association with HIV, weight for height Z scores, bacteremia, hypoxemia and clinical severity. Additionally describe initial antibiotic prescriptions for children admitted with severe or very severe pneumonia at KNH. Outcome measure: Death within one week of admission. Design: Descriptive cross-sectional study Setting: The study was carried out in Paediatric Filter Clinic (PFC) and general wards of Kenyatta National Hospital a national referral and teaching hospital. Method: Children aged between two and 59 months who were attended at PFC and evaluated for severe or very severe pneumonia as per WHO guidelines were consecutively recruited. A physical examination, oxygen saturation, blood culture, and serology for HIV were performed on all study participants. Lumbar puncture was performed on those who had features suggestive of meningitis. Information on antibiotic prescription both at PFC and wards was recorded on the day of admission. The study participants were then monitored in the wards within the week of admission to document the outcome. This information was obtained from the patients' files. Data was entered into SPSS and analysed using the same. RESULTS: A total of 251 children were studied. Patients admitted with very severe pneumonia were 73 (29.1%). In total, 33 (13.1%) children died within the week of admission. Of the deaths, 63.6% occurred within 48 hrs of admission, whereas the average duration of hospital stay for the survivors was 5.2 days. The factors that were associated with short term mortality in multivariate analysis were: HIV infection (OR 12.7 CI 4.9-33.0), inability to feed (OR 4.5 CI 1.8-11.4), and severe dehydration (OR 6.7 CI 2.3-19.3). Inappropriate choice of initial antibiotics as well as inadequate doses was noted both at PFC and in the wards. For instance 67% of children admitted with severe pneumonia received inappropriate choice of initial antibiotics. In HIV infected children admitted with pneumonia, 38.3% missed cotrimoxazole in the initial antibiotic prescription for t . treatment or prophylaxis against Pneumocystis jiroveci pneumonia. CONCLUSION The short term mortality rate of children admitted with pneumonia was 13.2%. Mortality was higher among those with the very severe form (72.7%). Early deaths (within 48 hours of admission) occurred in 64% of the children hence aggressive and timely interventions are needed in order to reduce mortality due to pneumonia. The factors that were highly associated with short term mortality in multivariate analysis were: HIV infection (OR 12.7 CI 4.9-33.0), inability to feed (OR 4.5 CI 1.8-11.4), and severe dehydration (OR 6.7 CI 2.3-19.3). Adequate and effective supportive therapy in feeding and fluids administration might therefore reduce mortality due to pneumonia. Inappropriate choice as well as inadequate doses of initial antibiotics was noted in children admitted with pneumonia both at PFC and in the wards.