Amoebiasis is a human disease produced by Entamoeba histolytica which causes widespread mortality and morbidity worldwide through diarrheal disease and abscess establishment in parenchymal tissues such as liver, lung, and brain. The true prevalence of infection is unknown for most areas of the world due to the difficulty to characterise Entamoeba histolytica versus other non-pathogenic amoebas with identical morphology, as Entamoeba dispar, and Entamoeba moshkovskii. To overcome microscopy misidentification issues, we tested a nested multiplex polymerase chain reaction (PCR) and a real-time PCR on 194 stool samples collected from incoming dysentery patients in Cairo hospitals diagnosed with E. histolytica by microscopy. Nested PCR showed only 20 (10.3%) samples positive to E. histolytica and 17 (8.7%) to E. dispar. The real-time PCR detected only 19 and 11 samples positive to E. histolytica and E. dispar respectively, showing less sensitivity than the nested PCR. The data show that prevalence of E. histolytica in Cairo is lower when specific diagnosis methods are used instead of traditional microscopy, allowing to differentiate between morphologically identical human amoebas species.

Introduction: A test comparison of in-house and commercial real-time PCR (qPCR) kits for the detection of human parasites and microsporidia in stool samples was conducted without a gold standard. Three different commercial kits were included in the comparison, with a range of 3-15 different PCR targets, while 14 targets were covered by in-house testing, so not all 16 target pathogens were covered by all assays. Methods: Residual materials from nucleic acid extractions of stool samples with very high likelihood of being colonized or infected by at least one enteric parasite species or microsporidia were tested. In all, 500 DNA samples were analyzed, but due to limited sample volume, only 250 of the 500 samples were tested per assay. Each sample was assessed with the qPCR platforms being compared and cycle threshold (Ct) values were included in a descriptive comparison. Results: Depending on the assay applied, qPCR detected per 250 tested samples Giardia duodenalis (184-205), Blastocystis spp. (174-183), Trichuris trichiura (118-120), Ascaris lumbricoides (79-96), Necator americanus (78-106), Hymenolepis nana (40-42), Cryptosporidium spp. (27-36), Dientamoeba fragilis (26-28), Schistosoma spp. (13-23), Enterobius vermicularis (8-14), Entamoeba histolytica (7-16), Strongyloides stercoralis (6-38), Cyclospora spp. (6-13), Taenia spp. (1-4), microsporidia (1-5), and Ancylostoma spp. (1-2). Inter-assay agreement kappa was almost perfect (0.81-1) for Dientamoeba fragilis, Hymenolepis nana, Cryptosporidium spp., and Ascaris lumbricoides, substantial (0.61-0.8) for Necator americanus, Blastocystis spp., Ancylostoma spp., Giardia duodenalis, Schistosoma spp., Trichuris trichiura, and Enterobius vermicularis, moderate (0.41-0.6) for Entamoeba histolytica, fair (0.21-0.4) for microsporidia, slight (0-0.2) for Cyclospora spp. and Strongyloides stercoralis, and poor ( <0) for Taenia spp. Conclusions: Varying inter-assay agreement makes interpretation of microsporidia and parasite PCR in stool samples challenging. Infra-assay agreement had been controlled during the developing of the assays. Future studies, e.g., with optimized nucleic acid procedures and including microscopically characterized samples, are advisable.