Abstract
The evidence that coxsackievirus B plays a causal role in the etiology of insulin-dependent diabetes mellitus (lDDM) is reviewed. This hypothesis is biologically plausible; one variant of another picornavirus, encephalomyocarditis virus, causes diabetes in genetically susceptible mice,but prior infection with another, serologically indistinguishable variant prevents this. The seasonal distributions of infection due to coxsackievirus Band of IDDM are similar. Case reports document coxsackievirus B infection coincident with the onset of IDDM. More than one-third of patients with recent-onset IDDM have antibody to coxsackievirus B but have not necessarily had recent infection. Case-control studies show no consistent association, with odds ratios ranging from 0.7 to 20. In two prospective studies, IDDM did not follow coxsackievirus B epidemics, but the sample sizeswere small. Although epidemiologicdata are too inconsistent to allow one to conclude that coxsackievirus B is a frequent cause of IDDM, methodologic difficulties—illustrated in the animal model—suggest that it would be premature to discard this hypothesis.