RESUMEN
Globally, diarrhoeal diseases are the second leading cause of death among children under 5 years old. Few case-control studies on the aetiology of diarrhoea have been conducted in China. A case-control study on 922 children under 5 years old who presented with diarrhoea and individually matched controls was conducted in China between May 2011 and January 2013. Quantitative PCR was used to analyze stool samples for 10 diarrhoeal pathogens. Potential enteric pathogens were detected in 377 (81.8%) of 461 children with diarrhoea and 215 controls (46.6%, p <0.001). Rotavirus, norovirus GII, Shigella and adenovirus were qualitatively associated with diarrhoea. Using receiver operating characteristic curves, the optimal cutoff threshold for defining a symptomatic individual was 72, 5840, and 10(4) copies per reaction for rotavirus (odds ratio 259), norovirus GII (odds ratio 10.6) and Shigella (odds ratio 5.1). The attributable fractions were 0.18 for rotavirus, 0.08 for norovirus GII, 0.01 for Shigella and 0.04 for adenovirus. Coinfections between pathogens were common. Two pairs, rotavirus and adenovirus, and norovirus GII and Salmonella were positively associated. The co-occurrence of rotavirus and sapovirus, astrovirus, enterotoxigenic Escherichia coli or Campylobacter jejuni only occurred in children with disease. Coinfection was not correlated with clinical symptoms. Quantitative data are critical. Our results indicate that increased pathogen loads increase the OR between diarrhoea and rotavirus, norovirus GII and Shigella. Coinfections with rotavirus and norovirus GII are common and occur in a nonrandom distribution. Despite testing for ten diarrhoeal pathogens, over two-thirds of cases do not have a recognized attributable cause.