The underlying strategy of the WHO for malaria control is early diagnosis and prompt treatment, which ideally requires a sensitive, quick, simple and cheap diagnostic technique. The standard method of definitive malaria diagnosis in South Africa is microscopic examination of Giemsa-stained thick blood smears. Various new and alternative diagnostic methods have been developed over the past number of years. The ParaSight™ (PS) Test dipstick and two fluorescent microscopic methods, namely Kawamoto’s acridine orange (AO) staining technique and quantitative buffy coat analysis (QBC®), were evaluated by comparing them with Giemsa-staining.
In a laboratory component of this study, cultured parasites were used to artificially infect blood. Serial parasite dilutions were prepared and sampled. In the field component 139 blood samples were collected at Manguzi Hospital, KwaZulu-Natal, from suspected malaria patients. All samples were examined independently with the four methods and by different microscopists (a professional, the author and a novice). In the field study, excluding retakes from the analysis, sensitivity of Giemsa-stained thick smear (GTS) = 89.1%, Giemsa-stained thin smear (GTnS) = 81.8%, AO-stained thick smear (KTS) = 94.5%, AO-stained thin smear (KTnS) = 81.8%, QBC = 87.3% and PS = 96.4%, while specificity was 100%, 100%, 97.7%, 97.7%, 100% and 90.9% respectively. The 100% sensitivity cut-off points (P/nO in the laboratory study were as follows: GTS = 84, GTnS = 1080, KTS = 140, KTnS = 84, QBC = 140 and PS = 30 P/n . The 100% sensitivity cut-off points of the unexperienced microscopist were as follows: GTS = 140, KTS = 85 and QBC = 389 V.
In the field study the correlation, based on the Kappa K statistic, between categorized test estimates of parasitaemia and Gold standard (G-std) parasitaemia, was as follows: for GTS K = 0.865, GTnS K = 0.627, KTS K = 0.761, KTnSK =0.576. In the laboratory study statistical differences (based on paired t-tests) between test estimates and G-std parasitaemia at individual dilutions were insignificant except at four dilutions in KTS and at one dilution in GTS. Associations between the G-std and GTS, GTnS, KTS and KTnS, based on Pearson’s correlation coefficients (r), were respectively 0.821, 0.743, 0.381 and 0.791 in the field study and respectively 0.947, 0.924, 0.856, 0.866 in the laboratory study. The correlation (r) between the G-std and indirect QBC counts was 0.755 in the field 0.735 in the laboratory study. Regression equations were derived.
Total handling time per sample for GTS, KTS, QBC and PS was 50.1, 11.1, 10.7 and 9.0 min respectively, where samples were processed individually; 9.6, 6.6, 6.3 and 2.2 min when processed in batches of 10; 5.9, 6.3, 6.4, and 1.6 min when processed in batches of 100. Cost of extra equipment per microscope (excluding microscope) was US$ 350 for Kawamoto’s method and R16 819for QBC. Cost per test was 10c for Giemsa, 10c for Kawamoto’ method, R5.80 for QBC and about R5.00 for PS.
On grounds of its consistent and superior sensitivity and its speed, the ParaSight™ Test is recommended as the diagnostic tool of the future to replace the more time-consuming and less sensitive standard microscopic diagnosis in malaria control programmes.
|Degree Type||Masters degree|
|Degree Description||MSc (Geography)|