Aim Sudden unexpected death in infancy (SUDI) represents the commonest presentation of post-neonatal infant death in the UK. This audit reviews current practice in the investigation of SUDI deaths, with particular regard to the practice of microbiological sampling in emergency departments (ED) compared with samples obtained at the time of autopsy for establishing the cause of death, as suggested by current guidelines.

Methods Coronial autopsies performed for the indication of SUDI over a 4-year period at a single specialist centre were reviewed with particular regard to the findings of microbiological investigations performed in ED compared with those performed at the time of autopsy.

Results Of 229 SUDI postmortems performed during the period, there were 136 cases in which both bacteriological samples taken in ED and at autopsy were available, including 109 with blood cultures taken at both time points. 66 cases had sterile blood cultures in ED of which 37 (56%) showed positive microbiological growth from autopsy samples including nine (14%) cases with group II pathogens. Group II pathogens were identified from ED samples in six (6%) of the total cases; all but two cases of Staphylococcus aureus were not detected at autopsy.

Conclusion Blood cultures obtained at autopsy are associated with a significantly higher rate of positive microbial cultures compared with blood samples taken in life. Most represent easily identified postmortem translocation or overgrowth rather than infection as the cause of death. No cases with a final infective cause of death would have been missed if ED sampling had not been performed.