Risk factors for noma disease: a 6-year, prospective, matched case-control study in Niger.

BACKGROUND:Noma is a poorly studied disease that leads to severe facial tissue destruction in children in developing countries, but the cause remains unknown. We aimed to identify the epidemiological and microbiological risk factors associated with noma disease. METHODS:We did a prospective, matched, case-control study in Niger between Aug 1, 2001, and Oct 31, 2006, in children younger than 12 years to assess risk factors for acute noma. All acute noma cases were included and four controls for each case were matched by age and home village. Epidemiological and clinical data were obtained at study inclusion. We undertook matched-paired analyses with conditional logistic regression models. FINDINGS:We included 82 cases and 327 controls. Independent risk factors associated with noma were: severe stunting (odds ratio [OR] 4·87, 95% CI 2·35-10·09) or wasting (2·45, 1·25-4·83); a high number of previous pregnancies in the mother (1·16, 1·04-1·31); the presence of respiratory disease, diarrhoea, or fever in the past 3 months (2·70, 1·35-5·40); and the absence of chickens at home (1·90, 0·93-3·88). After inclusion of microbiological data, a reduced proportion of Fusobacterium (4·63, 1·61-13·35), Capnocytophaga (3·69, 1·48-9·17), Neisseria (3·24, 1·10-9·55), and Spirochaeta in the mouth (7·77, 2·12-28·42), and an increased proportion of Prevotella (2·53, 1·07-5·98), were associated with noma. We identified no specific single bacterial or viral pathogen in cases. INTERPRETATION:Noma is associated with indicators of severe poverty and altered oral microbiota. The predominance of specific bacterial commensals is indicative of a modification of the oral microbiota associated with reduced bacterial diversity. FUNDING:Gertrude Hirzel Foundation.