In spite of the reduction in the incidence of sudden unexpected infant death (SUID) after the back to sleep campaign, the underlying mechanisms remain poorly understood. A retrospective analysis of causes of SUID during a 3-year period, based on a prospective investigation protocol. We included all infants admitted to our reference centre for SUID during the years 2005–7. Our protocol included review of the clinical history, examination of the death scene, biological samples analysis, skeleton radiograph and complete autopsy.
Microbiology: Blood: culture, bacterial and viral PCR, ARN16S, HIV-RNA level. Urinocult. Cerebrospinal fluid (CSF): culture and viral PCR. Nasal and rectal culture, influenzae A and B and respiratory syncytial virus identification. Lung biopsy: pathology and culture.
Metabolic: Plasmatic carnitine, aminoacidogram in urine and CSF.
Immune: Haemogram, haemoglobin electrophoresis, immunoglobulin, lymphocytic typing, C-reactive protein in blood.
Toxic: Blood and hair.
We obtained data on 24 infants (18 males, 6 females). The mean age was 5.6 months (range 0.5–17.5 months, median 4.5 months). In six cases, autopsy was refused by parents. SUID was considered attributable to congenital cardiopathy in two cases, aspiration in one, infections in 13 and gut disorders in three. Postmortem examination allowed us to identify causes in 19 out of 24. Infection is currently a major cause of SUID. Microbiological and autopsy examinations succeed in explaining SUID in many cases. However, it seems interesting to develop approaches that are less invasive and stressing for the family. Magnetic resonance imaging may be performed instead of autopsy, especially when parents refuse for religious reasons.