PCR for enteric pathogens in high-prevalence settings. What does a positive signal tell us?

Background: Molecular methods, in particular PCR, are increasingly used for the diagnosis of enteric pathogens in stool samples. In high-endemicity settings, however, asymptomatic carriage or residual DNA from previous infections will hamper the interpretation of positive test results. We assessed the quantitative dimension of this problem in schoolchildren in the rural highlands of Madagascar. Methods: Stool samples were collected from 410 apparently healthy Madagascan schoolchildren and analysed by multiplex real-time PCR for enteroinvasive bacteria (Salmonella spp., Shigella spp./enteroinvasive Escherichia coli (EIEC), Campylobacter jejuni, Yersinia spp.), enteric protozoa (Entamoebea histolytica, Giardia duodenalis, Cryptosporidium spp., Cyclospora spp.), and helminths (Ascaris lumbricoides, Ancylostoma spp., Necator americanus, Strongyloides stercoralis). Symptoms of gastrointestinal disease were assessed. Results: Among the 410 samples, we detected Giardia duodenalis in 195, Campylobacter jejuni in 91, Ascaris lumbricoides in 72, Cyclospora cayetanenesis in 68, Shigella spp./EIEC in 56, and Strongyloides stercoralis and Cryptosporum spp. in 1 case each. Salmonella spp., Yersinia spp. and hookworms were not observed. Relative risk assessment suggested few and incoherent associations with pathogen detections, indicating asymptomatic carriage or DNA residuals. Only 26.1% of the schoolchildren were tested negative for all analysed pathogens. Conclusions: The very high risk of detecting traces of asymptomatic carriage or residual DNA from previous infections limits the value of highly sensitive PCR for the causal attribution of detected enteric pathogens from stool samples to an infectious gastrointestinal disease in the high-endemicity setting. Evaluated standards for the interpretation of such results are needed both for the diagnostic routine and for epidemiological assessments.