Abstract

Three methods were used to investigate the role of infection in sudden unexpected infant death (SUD): (i) microbiological comparison of SUD victims and matched, live, community controls; (ii) postmortem classification of the contribution of infection to death; and (iii) case-control analysis of the relative risk associated with both infection and heavy wrapping. Limited sampling from the upper respiratory tract and gut in SUD victims and controls showed no significant excess of viral infection in the SUD victims (odds ratio = 1.98, 95% confidence interval (CI) 0.9 to 4.5). At postmortem examination, infection explained death in 3/95 babies and may have contributed to death in 37/95. Over 70 days of age, the combined presence of viral infection and wrapping in excess of 10 togs produced an odds ratio of SUD of 51.5 (95% CI 5.64 to 471.48) compared with wrapping of less than 6 togs. Viral infection was not a major risk factor as long as babies were lightly wrapped. In heavily wrapped babies the presence of a viral infection greatly increased the risk of SUD.