Epidemiology of shigellosis among children exposed to cases of Shigella dysentery: a multivariate assessment.

We followed 1,756 young, rural Bangladeshi children less than five years of age for one month after identification of sentinel Shigella patients in their neighborhoods. Two hundred nineteen (12%) children developed Shigella diarrhea (shigellosis) and 227 (13%) developed culture-negative dysentery. Shigella flexneri (60%) and S. dysenteriae, type 1 (15%) were the most common isolates among shigellosis cases. Within individual neighborhoods, there was poor agreement (Kappa = 0.21) between Shigella species isolated from sentinel patients and from additional cases detected during surveillance. The risk of shigellosis increased substantially after infancy and peaked in the second year of life. Severe stunting, as assessed by height-for-age, was associated with an increased risk of shigellosis (adjusted odds ratio [ORa] = 1.67, 95% confidence interval [CI] = 1.09-2.57, P < 0.05), while breast-feeding was protectively associated (ORa = 0.40, 95% CI = 0.24-0.69, P < 0.001). Only 43% of the shigellosis cases reported bloody stools; frank dysentery occurred more frequently in S. dysenteriae 1 infections than in S. flexneri infections (ORa = 5.04, 95% CI = 1.76-14.48, P < 0.01), and was also associated with severe stunting (ORa = 2.16, 95% CI = 1.01-4.58, P < 0.05). Our findings show that the high risk of shigellosis in residentially exposed Bangladeshi children results from multiple Shigella strains circulating concurrently within the same neighborhood; demonstrate that the risk is notably modified by host age, nutritional status, and dietary patterns; and illustrate that the classic picture of dysentery is relatively infrequent and is correlated with the infecting species and with host nutritional status.