Epidemiology of sudden unexpected death in infants ('cot death') in Northern Ireland.

Abstract

Sudden unexpected death in infants ('cot death' or 'crib death') is now recognized as a clinical entity and existing knowledge has been recently sum marized (Valdes-Dapena, 1967; Froggatt, Lynas, and Marshall, 1968; Bergman, Beckwith and Ray, 1970). The 'final common pathway' of death (as sumed to be identical in most cases) is, however, unknown. The consensus view is that throughout life these infants are essentially healthy; they die because while passing through a developmental stage of physiological vulnerability some critical combination of extrinsic and intrinsic factors occurs which proves lethal through either (a) sl respiratory mechanism?in whose production various processes have been incriminated (e.g., Bergman et al., 1970; Ray, Beckwith, Hebestreit, and Bergman, 1970; Shaw, 1970); (b) a lethal cardiac arrhythmia or conduction disturbance?for which an anatomic basis has been demonstrated (James, 1968); or (c) a hypersensitivity reaction to antigens at present un identified but often thought to be cow's milk proteins (e.g., Parish et al., 1960b; Gunther, 1966). Numerous other hypotheses have been postulated but they are either untenable or purely speculative. One aetiological concept?lethal cardiac arrhyth mia?is based on data (Froggatt et al., 1968) and special examination of hearts (James, 1968) from a two-year study in Northern Ireland. This paper describes the epidemiological aspects of this study and examines how the data accord with the main hypotheses of causation. Other aspects of the study are presented elsewhere (Froggatt et al., 1968; Froggatt, 1970a; Froggatt, Lynas, and Marshall, 1971). Throughout the text the description of a difference as 'significant' means that it or a greater difference was unlikely to occur by chance in more than 5% of repeated trials, x2 is everywhere calculated on absolute numbers and, where appropriate, is cor rected for continuity. The literature on this condition is now extensive. Primary sources are only referenced where essential: thorough reviews and extensive bibliographies are given by Valdes-Dapena (1967, 1970) and Froggatt et al. (1968, 1971). Definition and Criteria of Inclusion There is no standard designation of the entity (or entities) under study: 'sudden and unexpected death in infants' (Rabson, 1949) and 'the sudden infant death syndrome' (Gold, Adelson, and Godek, 1964), and their variants, are most frequently used. The former indicates selection on clinical criteria alone, e.g., 'the death of a child who was thought to be in good health or whose terminal illness appeared to be so mild that the possibility of a fatal outcome was not anticipated' (Adelson and Kinney, 1956); while the latter involves clinical and pathological criteria, i.e., 'the sudden death of any infant or young child which is unexpected by history and in which a thorough post-mortem examination fails to demonstrate an adequate cause of death' (Bergman et ah, 1970, p. 18). This again is strictly equivalent to 'sudden unexpected unexplained death in infants' (Fitzgibbons et al, 1969) which, in turn, stems from the earlier 'sudden apparently unexplained death during infancy' (Werne and Garrow, 1953). In practice, the above and variant terms are used as synonyms: in this article we use for brevity the acronym SUD. None of the above is a certifiable cause of death; accordingly, we must state precisely our criteria. We selected cases on necropsy findings, taking as the criterion the absence of demonstrable pathology (on the thorough procedure of Marshall (1970)) conventionally accepted as sufficient to explain death. Subjects with recognizable anomalies, e.g., polycystic disease and venous sinus thrombosis, were excluded even though these lesions may not have 'caused' the death (sudden death is not un common in such children and is usually attributed, perhaps erroneously, to the underlying condition), while in some few instances, e.g., asphyxia due to suspected choking, and infanticide, the history contributed to exclusion. We thus avoided bias inherent in selecting on a case history criterion and could include children with varied premortem clinical findings. Our cases were literally 'un explained at necropsy'; since, however, most were also 'sudden and unexpected' and in the narrow age range which characterizes SUD (see below), our 119