Malaria among adult inpatients in two Tanzanian referral hospitals: a prospective study.
Academic Article
Overview
abstract
Most malaria research in sub-Saharan Africa has focused on children and pregnant women, but malaria among hospitalized adults in this region is poorly characterized. In this prospective study, we assessed the prevalence and clinical characteristics of malaria among the inpatient adults in two hospitals in Tanzania. We enrolled adults admitted with suspected malaria and performed routine thick blood smear (BS) and malaria rapid diagnostic tests (RDT). We also assessed malaria parasite clearance rates. We considered malaria status 'confirmed' or 'excluded' only in patients with two concordant tests. Malaria polymerase chain reaction (PCR) was performed in a subset of patients with discordant BS and RDT. After BS and RDT were performed on 579 adults with suspected malaria, malaria was excluded in 458/579 (79.1%) and confirmed in 16/579 (2.8%). One hundred and five out of 579 (18.1%) had discordant results. The prevalences of positive BS and positive RDT were 102/579 (17.6%) and 35/579 (6.0%), respectively, with only fair agreement (Kappa=0.354, p<0.0001). PCR results agreed with RDT in 35/35 (100%) of patients with a negative RDT but positive BS. PCR results also agreed with RDT in 9/13 (69.2%) of cases with a positive RDT but negative BS. Clinical correlates of malaria by multivariable analysis included subjective fever (OR 3.6 [1.0-12.3], p=0.04), headache (OR 3.1 [1.2-8.0], p=0.02) and vomiting (OR 2.7 [1.2-6.4], p=0.02). Malaria parasite clearance was significantly delayed in the HIV-infected group. Our study demonstrated only fair agreement between RDT and BS malaria tests among Tanzanian adult inpatients with suspected malaria. PCR generally agreed with RDT results. HIV was associated with delayed parasite clearance in adults with malaria. We recommend the routine use of RDTs for malaria diagnosis among adults admitted to hospitals in sub-Saharan Africa.
