EVALUATION OF CHILDHOOD MALARIA MANIFESTATIONS, DIAGNOSTIC AND MANAGEMENT PRACTICES IN WESTERN KENYA HIGHLANDS

Abstract

Malaria remains a major killer in many parts of the world and challenges in accurate malaria diagnosis regrettably encourage widespread presumptive malaria treatment. The study conducted between September 2009 and May 2011 compared the sensitivity of Immunochromatographic malaria rapid diagnostic test with microscopy as the gold standard, using plasma, whole blood and serum among affected children (aged <12 years). Malaria rapid diagnostic test training and user-attitudes among healthcare professionals in the highlands and lowlands of western Kenya and malaria management practices in highland-based health centres were evaluated and determination of predominant malaria features, malaria parasite densities and the correlation of the latter with haemoglobin levels were also done. The results confirmed that immunochromatographic (ICT) malaria rapid test had 97% specificity to Plasmodium falciparum, while the 3% that tested negative using ICT were microscopically identified as Plasmodium malariae (1.5%) and Plasmodium ovale (1.5%), respectively. No child in the sample population had Plasmodium vivax. There was a highly significant difference in the species occurrence of malaria parasites (Fisher‘s exact test; p<0.0001).The sensitivity of ICT when using plasma and whole blood was the same (86.4%) but was comparatively less when serum was used (84.8%). There was a highly significant difference between the training and user-attitudes of health professionals in the highlands and lowlands of western Kenya towards rapid diagnostic tests (Pearson Chi Square; Pr <0.001). Fever, malaria-associated anaemia and sweating were elicited among 92.4%, 85.6% and 81.1% of the affected children, respectively. Pearson‘s analysis showed a negative correlation between haemoglobin levels and malaria parasite density but the negative correlation was statistically insignificant (‗R‘= -0.1171; p =0.181). Out of 153,530 patients evaluated for possible malaria across five highland-based health centres during 2001-2010 periods, 58.3% were presumptively treated, 33.3% had no malaria parasites and 8.4% had malaria parasites. Paired sample T- test showed a highly significant difference between the presumptively treated and laboratory-confirmed malaria across three (n=9771; mean=1.132E3; 95% C.I: 918.517 - 1.346E3; p < 0.001) and across two (n=118,255; mean=8119.571; 95% C.I: 5836.924-10402.219; p <0.001) highland-based health centres in western Kenya during years 2007 and 2001-2007, respectively. Immunochromatographic rapid diagnostic test is significantly sensitive and specific to Plasmodium falciparum; hence should be used to support microscopy method of malaria diagnosis. Although whole blood, plasma or serum is recommended for malaria diagnosis using immunochromatographic test, serum is less reliable than plasma or whole blood. Urgent training to sensitize qualified health professionals, especially in the highlands of western Kenya, to effectively use malaria rapid diagnostic tests, is vital. Investigations to establish and manage the causes of the high anaemia prevalence among children (aged twelve and below) with malaria, are urgently required. Measures to improve and update malaria morbidity statistics of children, to reduce presumptive malaria treatment and increase reliance on laboratory diagnosis of malaria should be identified and implemented.