Prise en charge des morts inattendues du nourrisson par les centres de référence français : état des lieux des pratiques en 2013

In France, nearly 500 infants still die unexpectedly every year. In 2009, the French Institute for Public Health Surveillance published a survey showing great heterogeneity in the management of sudden unexpected infant death (SUID) cases. The aim of this study was to evaluate the actual diagnostic approach to SUID in the different reference centers in France and to determine the degree to which the 2007 recommendations of the French National Authority for Health (Haute Autorite de sante [HAS]) are applied. We conducted a multicenter cross-sectional epidemiological study by email sent to the 36 SIDS reference centers with questions on examinations usually performed in SIDS cases. We also submitted six SUID test cases for death classification to the different reference physicians. Twenty-nine of 36 centers (80.5%) responded. Among the recommended tests, only blood cultures, analysis of cerebrospinal fluid, and a proposal to autopsy are done in 100% of the centers. Other investigations are not carried out systematically: skeleton radiography (65.5%), cranial CT scan (58%), eye fundus (20.7%), metabolic analysis (65.5%), and blood toxicology (62%). The main reasons for non-completion of these tests were hospital practices, lack of resources, technical difficulties, cost of tests, and difficulty in interpreting results (50% reported not knowing the postmortem biological standards). None of the institutions apply the HAS recommendations entirely. The classification of causes-of-death test cases also varied between the centers, with a maximum of 62% concordance in their responses. In 2013, in France, there is still substantial heterogeneity in the diagnostic set-up of SUIDS, a non-exhaustive implementation of the recommendations of the French National Authority for Health, and an unsatisfactory SUIDS classification tool because of considerable discordance between physicians. These results explain the current difficulties in obtaining reliable epidemiological data, because many teams do not use all the investigations recommended to find the cause of death. Therefore, the establishment of a national registry would provide accurate and up-to-date epidemiological, environmental, medical, and biological data to identify the events causing death and propose appropriate means of prevention.