Screening For Plasmodium Falciparum Malaria Using Mrd Test In A High Transmission Area Of Ghana: Kintampo District

ABSTRACT

Background

Malaria has a high burden in sub-Saharan Africa which is caused by Plasmodium falciparum. Delayed or inaccurate diagnosis of malaria could lead to fatal consequences in patients seeking medical care for malaria. Malaria Rapid Diagnostic Tests (mRDT) provide an alternative in the absence of malaria microscopy to provide prompt and accurate diagnosis and is used to complement clinical diagnosis of individuals who report at health facilities with unspecific symptoms. Low sensitivity and specificity of mRDTs may lead to under-diagnosis or over diagnosis of malaria. Hence there is the need to confirm the validity of the mRDT in specific malaria transmission zones. The aim of this study was to assess the sensitivity and specificity of mRDT in the Kintampo Municipality of Ghana.

Methods

A secondary data analysis was done using a cross sectional study design data from the Kintampo Malarial Drug Trial conducted during the period of June 2005 to May 2006 to assess the efficacy of Artemisinin combination therapy. Cleaned data were analyzed using Stata 11.0 (STATACORP, TX.). Demographic characteristics of study children that were categorical in nature were summarized as proportions, while quantitative variables such as hemoglobin concentration was summarized as means based on their distributions. Sensitivity and specificity analysis of the mRDT was explored using malaria microscopy as the gold standard and Receiver Operating Characteristic (ROC) curve was used to determine the ability of the RDT test to correctly classify patients with or without malaria parasitaemia. Univariate logistic regression

v

was used to explore crude relationships between socio-demographic and potential confounding variables with clinical malaria. Clinical malaria was defined as reported or measured fever with malaria parasite counts >0. All statistically significant variables identified from the univariate analysis were adjusted for in a multivariate logistic regression model.

Results:

A total of nine hundred and thirty seven (937) children between 0- 10 years who were screened in the primary study were analyzed in this study. About 99% (923/937) of the children enrolled had both malaria parasitaemia and mRDT done. Using microscopy as the gold standard, the overall sensitivity and specificity of the mRDT used were 93.2% (95% CI: 90.7-95.2) and 83.0% (95% CI: 78.9-86.5) with a positive predictive value of (PPV) 88.1% (95% CI: 85.1- 90.6) and negative predictive value (NPV) of 90.1% (95% CI: 86.5-92.9) respectively. The risk of clinical malaria was 1.45 times more likely among children analyzed in this study with abnormal liver function test compared to those with normal liver function test [aOR 1.45 (95% CI:1.05-2.01), p=0.03] after adjusting for age and sex in a multivariate logistic regression model.

Conclusion

This mRDT maintains its high sensitivity and specificity among patients in the forest-savanna transitional zone of Ghana. The Malaria Rapid Diagnostic Test may be provided to peripheral public health facilities that do not have trained laboratory technicians, electricity and laboratory to appropriately diagnose malaria.