Etiologic features of diarrheagenic microbes in stool specimens from patients with acute diarrhea in Thailand

Abstract

Many microbial species have been recognized as enteropathogens for humans. Here, we predicted the causative agents of acute diarrhea using data from multiplex quantitative PCR (qPCR) assays targeting 19 enteropathogens. For this, a case-control study was conducted at eight hospitals in Thailand. Stool samples and clinical data were collected from 370 hospitalized patients with acute diarrhea and 370 non-diarrheal controls. Multiple enteropathogens were detected in 75.7% and 13.0% of diarrheal stool samples using multiplex qPCR and bacterial culture methods, respectively. Asymptomatic carriers of enteropathogens were found among 87.8% and 45.7% of individuals by qPCR and culture methods, respectively. These results suggested the complexity of identifying causative agents of diarrhea. An analysis using the quantification cut-off values for clinical relevance drastically reduced pathogen-positive stool samples in control subjects from 87.8% to 0.5%, whereas 48.9% of the diarrheal stool samples were positive for any of the 11 pathogens. Among others, rotavirus, norovirus GII, Shigella/EIEC, and Campylobacter were strongly associated with acute diarrhea (P-value < 0.001). Characteristic clinical symptoms, epidemic periods, and age-related susceptibility to infection were observed for some enteropathogens. Investigations based on qPCR approaches covering a broad array of enteropathogens might thus improve our understanding of diarrheal disease etiology and epidemiological trends.

Figure 1

Detection of enteropathogens in 370 diarrheal and 370 non-diarrheal control subjects. (A) Proportions of patients and control subjects who tested positive by qPCR assays. (B) Number of isolates by bacterial culture methods. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 2

Number of pathogens detected per subject using qPCR among 370 each of diarrheal (an average of 2.7 ± 0.1 (standard error) pathogens) and non-diarrheal control subjects (an average of 2.3 ± 0.1 (standard error) pathogens) (A) and the distribution based on the quantitative cycle (Cq) cut-off values mentioned in Table 1 (B).