Research on diagnostic methods have strongyloidiasis divergent validity and incomplete by not reporting data on safety, efficiency and performance diagnosis. To assess validity, performance, efficiency and safety of four diagnostic conventional parasitological methods for detection of Strongyloides stercoralis infection in the period 1980-2013. Systematic review with meta-analysis, exhaustive and reproducible literature search of six databases. Quality of the articles was assessed and meta-analysis was performed under the random effects model, calculating sensitivity, specificity, likelihood ratios, predictive values, proportion of false results, accuracy, odds ratio and Youden index J and ROC curve using Meta-DiSc(es) and Epidat 3.1. 11 studies with 9,025 individuals were included. Sensitivity of the Baermann method was 72%, positive likelihood ratio (LR+) 228 and negative likelihood ratio (LR-) 0.32. The agar plate culture (APC) had a sensitivity of 89%, LR+ 341 and LR- 0.11. Stool sensitivity was 21%, LR+ 67 and LR- 0.67. Sensitivity of the formol-ether concentration was 48%, LR+ 110 and LR- 0.59. Areas under the ROC curve were 0.999 in Baermann and APC, 0.977 in the stool and 0.829 in formalin-ether concentration; specificity was 100% in all tests. The four conventional parasitological methods tested in this study to detect S. stercoralis can be helpful; however, agar plate culture and Baermann method are best suited.
Read full abstract