Comparison of real-time PCR and the Kato-Katz method for the diagnosis of soil-transmitted helminthiasis and assessment of cure in a randomized controlled trial

Background Diagnosis of soil-transmitted helminths (STHs) in developing countries is commonly based on microscopic detection of eggs in stool samples, using the Kato-Katz (KK) method, which has a poor sensitivity for detecting light intensity infections. We compared the performance of the KK method and real-time PCR in the framework of a randomized trial, which evaluated four novel treatments againstTrichuris trichiuraand concomitant STH infections. Results Two stool samples obtained from 320 participants were examined at baseline and follow-up with quadruplicate KK and PCR analyses of one of the two samples using "bead-beating" for DNA extraction. At follow-up, 80 samples were negative according to both PCR and KK and 173 were positive with both methods for any of the STHs. Relative to PCR, the calculated sensitivity of KK at follow-up was 83.6%, 43.0% and 53.8% forT. trichiura, for hookworm and forAscaris lumbricoides, respectively. The sensitivity of PCR compared with KK at this time point was 89.1% forT. trichiura, 72.7% for hookworm and 87.5% forA. lumbricoides.Cure rates (CRs) forT. trichiuraandA. lumbricoideswere slightly lower with the PCR method. For hookworm CRs with KK were mostly significantly lower, namely 36.7%, 91.1%, 72.2% and 77.8% for moxidectin, moxidectin in combination with tribendimidine, moxidectin in combination with albendazole and albendazole in combination with oxantel pamoate, respectively, whereas with PCR the CRs were 8.3%, 82.6%, 37.1% and 57.1%, respectively. Conclusions In conclusion, a single real-time PCR is as sensitive as quadruplicate KK forT. trichiuraandA. lumbricoidesdetection but more sensitive for hookworm,which has an influence on the estimated treatment efficacy. PCR method with DNA extraction using the "bead-beating protocol" should be further promoted in endemic areas and laboratories that can afford the needed equipment. The study is registered at ISRCTN (no. 20398469).