Definition of the sudden infant death syndrome | The BMJ

Intended for healthcare professionals


Definition of the sudden infant death syndrome

BMJ 1994; 308 doi: (Published 28 May 1994) Cite this as: BMJ 1994;308:1439
  1. C Rambaud,
  2. C Guilleminault,
  3. P Campbell
  1. Anatomic-Pathologique, Hospital Necker-Enfants Malades, 75743
  2. Paris Cedex 15, France Sleep Disorder Clinic, Palo Alto, CA 94304
  3. USA Victoria Institute of Forensic Pathology, South Melbourne, Victoria 3205, Australia.

    EDITOR, - The definition of the sudden infant death syndrome was called into question at the second international conference on the syndrome, in February 1992. As no consensus could be reached it was decided to keep the 1969 Beckwith definition: “the death of an infant or young child, which is unexpected by history and in whom a thorough necropsy examination fails to reveal an adequate cause of death.”

    The issue is what constitutes an “adequate cause of death.” To base the definition of the sudden infant death syndrome on a diagnosis of exclusion is questionable. There is no current consensus on the minimum postmortem investigations that are needed to ensure that every cause of death has been excluded. There is also no consensus among pathologists on what is an adequate cause of death. The problem is not related to the identification of lesions, as every careful paediatric pathologist reports similar findings,1 usually noted as “in-significant lesions.” This interpretation is the core of the problem and is not shared by French pathologists. They believe that the pathological findings can be considered to be an adequate cause of death in three quarters of the cases diagnosed as the sudden infant death syndrome in many countries.2,3

    The main difference between the French and other approaches to sudden infant death is the systematic protocol that involves many specialists at the time of death.4 Any infant who dies suddenly is brought to the emergency room and dealt with in the same way as an acutely ill child would be. Biological and microbiological samples are collected within 10 hours of the last sighting of the infant alive in 90% of cases. Paediatricians direct the sampling process and collect the paediatric and obstetric data, and microbiologists play a part well before the necropsy protocol starts. The extensive necropsy protocol is always performed by a paediatric pathologist with a major interest in the sudden infant death syndrome.4 All findings are reviewed at a multidisciplinary conference, and the decision concerning the cause of death is made there. This approach is very different from many investigations into sudden infant death that have been reported in the United States, for example.5

    The definition of the sudden infant death syndrome is crucial as many epidemiological data will be derived from infants labelled as having died of the syndrome by a bureaucratic, sometimes ill equipped system. These so called epidemiological data may misdirect researchers toward irrelevant protocols. It has been reassuring for parents to be given something that they consider to be a diagnosis (that is, the sudden infant death syndrome), and such a positive impact must be taken into consideration. Therefore, we suggest a modification of the definition of the syndrome. The sudden infant death syndrome should be diagnosed in every case of cot death fulfilling the classical clinical criteria for the syndrome (a 1 month to 1 year old infant, clinically well before the death or with only minor illness, is put to sleep and found dead), with a subsequent classification according to the main postmortem findings (for example, the sudden infant death syndrome with myocarditis). This systematic indication in every report of the main two or three findings, as a minimum requirement, may give more substance and medical usefulness to the term sudden infant death syndrome.


    View Abstract