Prevalence of human immunodeficiency virus infection among children under five years with severe or very severe pneumonia in Kenyatta National Hospital
Introduction Pneumonia in the context of Human Immunodeficiency Virus (HIV) and Acquired Immune Deficiency Syndrome (AIDS) is a problem of high magnitude particularly among children in sub-Saharan Africa. Studies from Southern African states have demonstrated children infected with HIV experienced an increased morbidity and mortality from pneumonia compared to HIV uninfected children. Study Justification There is little local data that has been able to clearly demonstrate the magnitude of this problem in Kenya and particularly at Kenyatta National Hospital (KNH). This information would be useful in planning health services and developing protocols or guidelines in the management of these children. Study Objectives This study's objective was to determine the prevalence of HIV infection in children aged two to 59 months with severe or very severe pneumonia admitted to KNH and to describe the clinical and immunologic staging as well as the short-term in-hospital mortality in these children. Methodology Children aged two to 59 months were recruited from the KNH Paediatric Emergency Unit (PEV) 24 hours a day, seven days a week for the entire period of the study. This was conducted jointly with other members of the 'Childhood Pneumonia Study Group'. HJV testing was offered according to the Provider Initiated Testing and Counselling (PITC) guidelines. Clinical and immunologic staging was carried out according to the new World Health Organisation (WHO) guidelines. Outcomes namely mortality and discharge status were followed for up to five days of in-patient treatment. Results 342 children were enrolled of whom 158 (46%) had severe pneumonia and 184 (54%) had very severe pneumonia. The median age of the cohort was 8.6 months (interquartile range (lQR) 5.1-14.4) with a male to female ratio of 1:1.2. 38/342 (11.1 %) were found to be HIV antibody positive, 31 (82%) of whom were < 18 months of age. Of these 29/31 (94%) were confirmed positive on DNA PCR giving a total of 36/342 HIV infected patients and an HIV prevalence of 10.6%. HIV prevalence among those with severe pneumonia was 8.2% while the prevalence among those with very severe pneumonia was higher at 12.6%. 23/36 (64%) of the HIV infected patients were male while 26/36 (74%) of the HIV infected patients were infants two to 11 months of age. 22(61%) of the 36 HIV infected patients were in WHO clinical stage 3 and 14(39%) in clinical stage 4.29/34(85%) of the HIV infected patients were immunosuppressed with 22(63%) having severe depressed CD4 counts. 17/27(63%) infants and 5/7(71 %) children aged 12-35 months had severe immunosuppression. The in-hospital mortality among HIV infected patients was 13.9% (5/36) more than double that of HIV uninfected patients which was 5.3% (16/305). Univariate predictors of early in-hospital mortality included; female gender Risk Ratio (RR) 6.25, very severe pneumonia RR 9.51, HIV infection RR 2.91, WHO HIV clinical stage 4, previous admission RR 4.35; all with a P value of < 0.05. Adjusted analysis showed HIV infection to be independently predictive of mortality in this cohort (RR 3.82, 95%CI 1.08 to 13.57, P value=0.038). Conclusions Theprevalence of HIV infection among children under five years admitted to KNH with severe or very severe pneumonia was 10.6%,38.9% of these children were in clinical stage 4 and 63% had severe depressed CD4 counts. In-hospital mortality among HIV infected children was 13.9% compared to HIV uninfected children who had a mortality rate of 5.3%. Recommendation Effort is needed to step-up early screening and diagnosis of HIV infection particularly among children with severe forms of pneumonia. This will assist in instituting early and aggressive management of both typical and atypical forms of pneumonia.