- Email: [email protected]
Apolipoprotein E e4 and its prevalence in early childhood death due to sudden infant death syndrome or to recognised causes
Early Human Development (2008) 84, 549–554
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
w w w. e l s e v i e r. c o m / l o c a t e / e a r l h u m d e v
Apolipoprotein E e4 and its prevalence in early childhood death due to sudden infant death syndrome or to recognised causes Julie-Clare Becher b , Jean W. Keeling a , Jeanne Bell a,⁎, Betty Wyatt a , Neil McIntosh b a b
Division of Pathology (Neuropathology), University of Edinburgh, Scotland, United Kingdom Section of Child Life & Health, University of Edinburgh, Scotland, United Kingdom
Received 4 January 2008; accepted 8 January 2008
KEYWORDS Sudden infant death syndrome; Apolipoprotein E; Alleles; Sudden unexpected death in infancy
Abstract Background: Specific genetic polymorphisms have been shown to be more common in unexplained infant death. The APOE genotype exhibits opposite effects at the extremes of age with protective effects of e4 on perinatal mortality but detrimental effects as age progresses. Objective: To determine whether the APOE e4 allele is associated with early childhood (1 week– 2 years) unexplained death (‘sudden infant death syndrome’, SIDS) or with recognised causes (non-SIDS) and to compare these cohorts with published perinatal and adult data. Methods: DNA was extracted from spleen tissue of children dying in South East Scotland between 1990 and 2002. APOE alleles (e2, e3, e4) were determined using PCR. Comparisons of allele frequencies between groups were made. Results: There were 167 SIDS cases and 117 non-SIDS cases. Allele distributions of SIDS cases were similar to healthy newborns. Allele distributions of non-SIDS cases were more similar to adults than to healthy newborns. The percentage of children with at least one e4 allele was significantly lower in non-SIDS compared to SIDS (p = 0.016). Non-SIDS cases had a higher frequency of e3 compared to SIDS cases (p = 0.01) and to healthy newborns (0.005). Conclusions: Children dying from identified causes have different APOE allele distributions from SIDS cases, but are similar to adults. Children dying from SIDS have an allele distribution comparable to healthy newborns. The prevalence of e4 in SIDS is not of an order to contribute significantly to the age-related decline in e4. © 2008 Published by Elsevier Ireland Ltd.
⁎ Corresponding author. Division of Pathology (Neuropathology), Alexander Donald Building, Western General Hospital, Crewe Road, Edinburgh, EH4 2XU, Scotland, United Kingdom. Tel.: +44 131 537 1975; fax: +44 131 537 1013. E-mail address: [email protected] (J. Bell). 0378-3782/$ - see front matter © 2008 Published by Elsevier Ireland Ltd. doi:10.1016/j.earlhumdev.2008.01.002
550
1. Introduction Despite the reduction in unexplained infant deaths following widely publicised recommended changes in care practices, sudden unexpected death in infancy (SUDI) remains a numerically important group of deaths in the first year of life. Whilst the likelihood of a second similar death within a family is low, these undoubtedly occur and studies have estimated this increase to be between 1.9 and 9.5 times greater than the background risk of SUDI within the population studied [1–4]. Although there is no doubt that a small proportion of SUDI might be explained by a single entity such as an inherited metabolic disorder [5,6] it is accepted that many unexplained infant deaths are likely to involve a combination of factors including individual predisposition, age-specific risks and an exogenous stressor [7]. Host genotype is likely to be one of these factors and various polymorphisms, including those of IL-10 and the serotonin transporter gene, have been identified as contributing to the risk of sudden infant death syndrome (SIDS) [8,9]. In addition, abnormalities in neuronal numbers and function have recently been reported in a series of SIDS infants [10]. We hypothesised that Apolipoprotein E (APOE) polymorphisms might be a factor in the causation of SIDS as APOE shows isoform-specific effects on neuronal repair and protection and as such may modulate maturational changes and response to injury within the central nervous system of the infant [11]. The APOE genotype is controlled by three alleles at a single locus producing six different genotypes. The proportions of these genotypes and their related alleles has been established in many geographically and racially different groups [12]. An excess of the e4 allele in the fetus appears protective against early fetal death and death in the perinatal period [13,14]. However, in adults, the APOE e4 genotype is linked to ischaemic stroke [15] and neurodegenerative disorders such as late onset Alzheimer's disease [16]. An age-related decline in the APOE e4 allele has been established previously [17] and suggests that individuals with this allele may be predisposed to mortality during childhood or early to mid-adulthood. In addition, because of the small increase in susceptibility to SUDI within families, we wished to explore the hypothesis that infants who die in the first year of life may have over-representation of the e4 allele. This study examines the prevalence of the different APOE alleles amongst a group of infants who died in South East Scotland from known or unknown causes and compares the allele distribution of these groups with those of a population of healthy live born Scottish infants [17], Scottish perinatal deaths [14] and healthy Scottish adults [18].
2. Methods Ethics approval was obtained from the Lothian Research Ethics Committee. Early childhood deaths (1 week–2 years) between 1990 and 2002 were identified from the records of the Pathology Department of the Royal Hospital for Sick Children, Edinburgh. Autopsy examinations of children dying either in hospital or in the community in South East Scotland are carried out in this department.
J.-C. Becher et al. All childhood deaths were investigated to a similar standard according to a protocol. This included full organ dissection, tissue and fluid sampling for microscopy, bacterial culture, rapid identification techniques and culture for viruses and biochemical investigations. The brain was fixed in buffered formalin before examination by a neuropathologist. The spinal cord was examined and toxicological investigations done in selected cases. Cases were initially selected on age alone, but perusal of causes of death following completion of all autopsy investigations was done to ensure that similar numbers of explained and unexplained deaths were selected. Many of the necropsies were carried out on instruction from a Procurator Fiscal (equivalent to a Coroner in England and Wales). The explained deaths included both sudden and nonsudden deaths from both natural and non-natural causes. Medical histories were reviewed at time of death and a summary made by the pathologist. In all deaths occurring at home, there was a death scene investigation by police and in suspicious cases, by a pathologist. Cases were categorised by autopsy findings into two groups, explained (non-SIDS) and unexplained with no or only minor pathological findings (SIDS). The small number of pathologists involved and the close interdisciplinary approach to these deaths, with joint paediatric and forensic pathological investigation of many deaths, means diagnostic shift over the study period would have been minimal. Any death in which initial categorisation was unclear underwent further investigation by the Procurator Fiscal and the police with additional interviews and pathological investigations, although five cases remained undetermined. From archival files, a block of formalin-fixed paraffinembedded tissue, usually spleen, was retrieved from the departmental archives for each case. Five sections were cut from tissue blocks and stored in 2 ml screw cap tubes until analysis. These sections, identified by number only, were processed by the Department of Neuropathology for the determination of APOE genotype without knowledge of autopsy findings. From the deaths which occurred during the period of study, a sample of spleen was fresh frozen at autopsy for genotype analysis. The APOE genotype was determined for each case using a polymerase chain reaction (PCR) method adapted from Hixson and Vernier [19] as described previously [14]. The genotype for each sample was determined from the migration pattern of bands in a metaphor gel (Fig. 1). APOE e2/2 is characterised by the presence of 91 and 81 bp fragments. APOE e4/4 is characterised by a unique 72 bp fragment while APOE e3/3 lacks the 81 and 72 bp fragments. Heterozygotes have a combination of different fragments. PCR was repeated for cases in which the band pattern was faint or equivocal until a definite result was obtained. All samples were coded for APOE genotype analysis. When all results were obtained the cases were assigned to their group of origin, SIDS or non-SIDS. The subjects investigated in this study consist of 284 childhood deaths and previously published data from 371 healthy newborns, 251 perinatal deaths (185 stillborn infants and 66 neonatal deaths) and 400 adults. Genotype comparisons within the population of childhood deaths were made between SIDS and non-SIDS cohorts, and with the populations of perinatal deaths and healthy adults. The cohort of
APOE genotype in early childhood death
551 Table 1
Cause of death in non-SIDS cases Number
Congenital anomaly Congenital heart disease Other Infection Non-accidental injury Accidental injury Perinatal causes Asphyxia-related Immaturity-related Other/specific causes Renal failure Total
36 7 29 10 24 7 3 1 117
Figure 1 Metaphor gel showing the migration pattern for ApoE genotypes. Band 1 shows the control ladder.
3.2. Comparison of early childhood deaths with healthy newborns (Tables 2A and 2B) perinatal deaths was contemporaneously similar occurring throughout Scotland [14] as were the healthy term newborns from South East Scotland [17]. The population of healthy adults were aged between 45 and 60 years and had been selected from primary care lists in North East Scotland in the mid-1970s [18]. The presence of Hardy–Weinberg equilibrium was sought in all populations to ensure they were free from outside influences, such as mutation, migration and random selection. Comparisons of allele distribution between groups were made using the Chi square test and where a p value of less than 0.05 was considered significant. The Mann–Whitney U test was used for comparison of the age of children at death.
The allele distribution of SIDS cases is similar to that of healthy newborn infants particularly with respect to the elevated percentage of cases in both groups who possess at least one e4 allele. Non-SIDS cases have a higher frequency of e3 (p = 0.005) and lower frequency of e4 compared to healthy newborns (p = 0.03). Table 2A APOE allele distributions in early childhood deaths compared to healthy newborns, perinatal deaths and healthy adults
3. Results There were 289 children dying between 1 week and 2 years. Of these 167 died of SIDS and 117 children died from known causes (non-SIDS). In five cases the category of death could not be confidently assigned and these five infants were not considered further. The median age at death of SIDS and nonSIDS cohorts was 16 wks (IQR 6.5, 32 wks) and 11.5 wks (IQR 5.3, 22 wks) respectively, (p = 0.048). 98 (59%) of SIDS cases and 70 (60%) of non-SIDS cases were male. 21 (13%) of SIDS cases and 4 (3%) of non-SIDS cases were b 37 weeks gestation at birth (p = 0.013). Table 1 shows details of the cause of death in non-SIDS cases. Tables 2A and 2B show allele and genotype distributions for each of the cohorts. The genotypic profile of the cohort of childhood deaths proved to be in Hardy–Weinberg equilibrium and therefore could be assumed to be free from outside evolutionary forces (p = 0.99). Previous work has demonstrated such equilibrium in the other groups we studied [14,17].
3.1. Comparison of SIDS with non-SIDS
All early childhood deaths (n = 284, alleles 568) SIDS (n = 167, alleles 334) Non-SIDS (n = 117, alleles 234) Healthy newborns [17] (n = 371, alleles 742) All perinatal deaths [14] (n = 251, alleles 502) Stillborn infants (n = 185, alleles 370) Neonatal deaths (n = 66, alleles 132) Adults [18] (n = 400, alleles 800) a
Non-SIDS cases have a higher frequency of e3 (p = 0.01) and lower frequency of e4 (p = 0.039) compared to SIDS cases (Tables 2A and 2B). The percentage of cases with at least one e4 allele is significantly higher in SIDS compared to non-SIDS (p = 0.016) (Table 3).
b c d e f
e2
e3
e4
n (%)
n (%)
n (%)
39 (7)
434 (76)
95 (17)
Total alleles 568
26 (8) e 243 (73) a
65 (19) b 334
13 (6) f 191 (82) c
30 (13) d 234
63 (8)
538 (72)
141 (19)
742
64 (13) 365 (72)
73 (15)
502
48 (13) 274 (74)
48 (13)
370
16 (12)
91 (69)
25 (19)
132
66 (8)
616 (77)
118 (15)
800
vs non-SIDS, p = 0.01. vs non-SIDS, p = 0.039. vs healthy newborns, p = 0.005. vs healthy newborns, p = 0.03. vs perinatal deaths, p = 0.02. vs perinatal deaths, p = 0.002.
552 Table 2B adults
J.-C. Becher et al. Frequency of APOE genotype in early childhood deaths compared to healthy newborns, perinatal deaths and healthy
Cohort
All early childhood deaths SIDS Non-SIDS Healthy newborns [17] All perinatal deaths [14] Stillborn infants Neonatal deaths Adults [18]
Apolipoprotein E genotype 2/2
3/3
4/4
2/3
2/4
3/4
n (%)
n (%)
n (%)
n (%)
n (%)
n (%)
0 0 0 2 5 3 2 2
166 (59%) 88 (52.6%) 78 (66.7%) 199 (53.6%) 138 (54.9%) 110 (60%) 28 (42.4%) 233 (58.2%)
7 (2.4%) 4 (2.4%) 3 (2.6%) 15 (4%) 3 (1.2%) 3 (1.6%) 0 4 (1%)
30 (10.5%) 18 (10.8%) 12 (10.3%) 44 (11.9%) 38 (15.1%) 27 (14.6%) 11 (16.7%) 51 (12.8%)
9 (3.1%) 8 (4.8%) 1 (0.9%) 15 (4%) 16 (6.4%) 15 (8.1%) 1 (1.5%) 11 (2.7%)
72 49 23 96 51 27 24 99
(0.5%) (2%) (1.6%) (3%) (0.5%)
3.3. Comparison of infant deaths with perinatal deaths (Tables 2A and 2B) Both SIDS and non-SIDS cohorts have a lower frequency of the e2 allele compared to perinatal deaths (p = 0.02 for SIDS and p = 0.002 for non-SIDS).
3.4. Comparisons of early childhood deaths with adults (Tables 2A and 2B) Although healthy newborns differ from adults with respect to e4, and the percentage of e4 is found to be elevated in SIDS to the same degree as in healthy newborns, there was no difference between SIDS cases and adults. The allele distribution of non-SIDS cases does not differ significantly from that of adults.
4. Discussion Despite widely publicised recommendations in infant care practices such as such as the ‘Back to Sleep’ campaign, advice about appropriate sleep clothing and covering and the importance of a smoke-free environment [20,21] SUDI remains an important group of deaths. Accurate and thorough autopsy examination reveals a cause of death in a proportion of these but a majority remain unexplained. This group (SIDS) accounts for around 20% of all post-neonatal deaths in the United Kingdom [22]. This paper has established the distribution of APOE alleles and genotypes in a population of Scottish children dying in the first 2 years of life from known (non-SIDS) and unknown causes (SIDS), and has shown clear differences between the two groups with over-representation of e4 in cases dying from SIDS. In this respect, SIDS resemble healthy newborns whereas children dying from established causes have an allele distribution similar to adults. Such disparity is at odds with the spectrum of characteristics which are common to both SIDS and explained deaths, such as prematurity, male gender, low maternal age and socioeconomic deprivation [23]. The small discrepancy in the age at death of each cohort is unlikely to explain this difference. It is generally accepted that SIDS deaths occur earlier in the first year of life than other SUDI [23]. Within this study, the SIDS cases
(25.3%) (29.3%) (19.7%) (25.9%) (20.3%) (14.6%) (36.4%) (24.8%)
Total
284 167 117 371 251 185 66 400
were older at death than non-SIDS cases. The cohort of nonSIDS however, comprised not only of explained deaths in children dying suddenly and unexpectedly (SUDI), but also of deaths in chronic conditions such as prematurity and congenital anomaly. The small numbers in each diagnostic group within the cohort of infants dying from recognised causes (Table 1) precluded further investigation of the role of APOE in specified causes of death such as infection, although there is evidence suggesting that APOE may have an immunomodulatory role both in the central nervous system and systemically [40]. The childhood and perinatal deaths studied in this paper occurred in the decade between 1990 and 2002. Although the adult population is historical, the genotype distribution of this group is entirely consistent with other well-described Caucasian populations [12]. All groups studied were shown to be in Hardy–Weinberg equilibrium and therefore the differences seen in genotype between groups cannot be explained by evolutionary forces such as genetic drift, selection or mutation. The APOE genotype distribution is known to differ between ethnic populations but we believe that the proportion of non-Caucasians within the Scottish population has changed negligibly over the twenty year study period and as such fluctuations in racial demographics is unlikely to explain any differences observed between populations. The presence of Hardy–Weinberg equilibrium supports this. The methodology of genotyping differed between deceased and living cohorts, as DNA was extracted from paraffin blocks rather than blood spots respectively.
Table 3 Comparison of the APOE e4 allele within subgroups of childhood deaths (other populations tabulated below)
All childhood deaths SIDS⁎ Non-SIDS Healthy newborns All perinatal deaths Adults ⁎p = 0.016 vs non-SIDS.
e4 carriers
Non-e4 carriers
Total
n (%)
n (%)
n
88 61 27 126 70 114
196 (69) 106 (64) 90 (77) 245 (66) 181 (72) 286 (71)
284 167 117 371 251 400
(31) (37) (23) (34) (28) (29)
APOE genotype in early childhood death Although genotyping may be affected by DNA breakage related to formalin fixation, we believe that the gel banding of Apolipoprotein E is highly characteristic and as such, error in interpretation of results is unlikely. It remains possible that the differences observed occurred by chance although the numbers in each population are not small. It is also possible that infants who were deemed to be healthy at birth died later in childhood, and thus have been counted twice. If this was true, we estimate from Scottish infant mortality rates, that at most this would apply to only one of this group and would be unlikely to significantly bias our results. Although a small minority of cases previously diagnosed as SIDS appear to be caused directly by genetic mutations, for example in the genes encoding medium chain acyl-CoA dehydrogenase [24] or cardiac ion channels [25], it is thought that in other cases a genetic polymorphism may predispose to death when an infant is in a critical situation. It is unlikely however that one single mutation or polymorphism is the predisposing factor in all SIDS cases. The widely debated triple risk hypothesis of SIDS causation encompasses individual vulnerability, a critical developmental period, and an exogenous stressor such as an upper respiratory infection [7]. Suggested genetic predisposing factors have included mutations in embryological genes important in the developing autonomic nervous system [26] as well as various polymorphisms of IL-10 [27], the 5HT transporter gene [9] and glucokinase [28]. Others have suggested that prenatal damage, particularly to the developing brainstem, may make an infant vulnerable to adverse events postnatally [29]. In adults, APOE e4 is associated with excess deposition of β-amyloid in the brain following head injury [30] and with an earlier onset of Alzheimer's disease [31]. At a cellular level APOE e4 is associated with less vigorous neuritic outgrowth and repair than the other APOE isoforms [32]. A variety of subtle forms of neuropathological damage has been described in SIDS infants [33] ranging from increased dendritic spines [34] to brainstem gliosis [35–37]. However the APOE genotype does not appear to be associated with brain injury among perinatal deaths [14] and in this study, the APOE genotype of SIDS cases was no different than in healthy newborn infants. The e4 allele is under-represented in perinatal deaths compared to healthy newborns suggesting an antagonistic pleiotropic effect of this allele at the extremes of age, affording protection during early development but having deleterious effects as age progresses. This theory is supported by beneficial effects of e4 on cognitive development in children [38,39]. This study showed that the e4 allele is less common in children dying from explained causes compared to healthy newborns. The role of the APOE genotype in adult brain injury and cognitive deterioration is the focus of much research but early protective effects are less well understood and may involve modulation of the immune response or mechanisms of CNS repair [40]. Although some of the explained deaths occurred following infection or brain injury, the causes of death in this group of non-SIDS are heterogeneous and a single unifying hypothesis is unlikely. There is substantial overlap in causation between early childhood and perinatal deaths, and this may explain the similarity of e4 proportion in these groups. It is notable that an unusually high frequency of e2 is confined to deaths occurring around or before birth. Paradoxically APOE e2 is considered to be neuroprotective later in life [40].
553 Previous work has shown that healthy Scottish newborns have an excess of e4 compared to healthy Scottish latemiddle aged adults suggesting an age-related decline due to early mortality or morbidity of individuals with the e4 allele [17]. Although the rate of post-neonatal death at 1.8/ 1000LBs is low [22] and the numbers unlikely to contribute substantially to this decline, we found no significant excess of e4 overall amongst children dying in early childhood although the subset of SIDS cases did show a small rise in e4. It may be inferred that the reduction in e4 seen in this selected population of healthy late-middle aged adults is more likely to be attributed to early adult mortality or morbidity. The e4 allele has known associations with atherosclerosis and hypercholesterolaemia [12], both of which are risk factors for ischaemic heart disease, a leading cause of premature morbidity and mortality in the Scottish population [41]. In conclusion the findings of this investigation indicate that APOE e4 is less common in infants dying from recognised causes in early childhood than in either healthy newborns or SIDS, suggesting that e4 may be protective against a subgroup of infant mortality. In contrast we found no association of APOE e4 with SIDS, a group who have a similar genotype distribution to healthy newborns. Longitudinal cohort studies of APOE genotype are required to fully explain the decline in e4 throughout life and the antagonistic effects of e4 at the extremes of age. As our understanding of the genome increases it has become clear that not only may genes be switched on and off but they may also be site-specific within a cell. Within such a dynamic cellular environment, proteomic and metabolomic research will be necessary to further our understanding of SIDS and poorly understood gene-environment interactions will require future exploration.
Acknowledgements The authors would like to thank the Scottish Cot Death Trust for the funding of the study. Ian Croy and Iain Anthony gave considerable help with the APOE methodology and Dr Rob Elton advised on the statistical analysis.
Competing interest statement Jeanne Bell, Jean Keeling and Neil McIntosh are consulted regularly by both defence and prosecution teams with regard to legal proceedings in infant deaths. There are no financial competing interests for any of the authors.
References [1] Peterson DR, Sabotta EE, Daling JR. Infant mortality among subsequent siblings of infants who died of sudden infant death syndrome. J Pediatr 1986;108:911–4. [2] Smith GC, Wood AM, Pell JP, Dobbie R. Sudden infant death syndrome and complications in other pregnancies. Lancet 2005;366:2107–11. [3] Beal SM, Blundell HK. Recurrence incidence of sudden infant death syndrome. Arch Dis Child 1988;63:924–30. [4] Oyen N, Skjaerven R, Irgens LM. Population-based recurrence risk of sudden infant death syndrome compared with other infant and fetal deaths. Am J Epidemiol 1996;144:300–5.
554 [5] Emery JL, Howat AJ, Variend S, Vawter GF. Investigation of inborn errors of metabolism in unexpected infant deaths. Lancet 1988;2:29–31. [6] Green A, Preece MA, Hardy D. More on the metabolic autopsy. Clin Chem 2002;48:964–5. [7] Guntheroth WG, Spiers PS. The triple risk hypotheses in sudden infant death syndrome. Pediatrics 2002;110:e64. [8] Korachi M, Pravica V, Barson AJ, Hutchinson IV, Drucker B. Interleukin 10 genotype as a risk factor for sudden infant death syndrome: determination of IL-10 genotype from waxembedded postmortem samples. FEMS Immunol Med Microbiol 2004;42:125–9. [9] Weese-Mayer DE, Zhou L, Berry-Kravis EM, Maher BS, Silvestri JM, Marazita ML. Association of the serotonin transporter gene with sudden infant death syndrome: a haplotype analysis. Am J Med Genet A 2003;122:238–45. [10] Paterson DS, Trachtenberg FL, Thompson EG, Belliveau RA, Beggs AH, Darnall R, et al. Multiple serotonergic brainstem abnormalities in sudden infant death syndrome. JAMA 2006; 296:2124–32. [11] Weisgraber KH, Mahley RW. Human apolipoprotein E: the Alzheimer's disease connection. FASEB J 1996;10:1485–94. [12] Davignon J, Gregg RE, Sing CF. Apolipoprotein E polymorphism and atherosclerosis. Arteriosclerosis 1988;8:1–21. [13] Zetterberg H, Palmer M, Ricksten A, Poirier J, Palmqvist L, Rymo L, et al. Influence of the apolipoprotein E epsilon4 allele on human embryonic development. Neurosci Lett 2002;324:189–92. [14] Becher JC, Keeling JW, McIntosh N, Wyatt B, Bell J. The distribution of apolipoprotein E alleles in Scottish perinatal deaths. J Med Genet 2006;43:414–8. [15] McCarron MO, Delong D, Alberts MJ. APOE genotype as a risk factor for ischemic cerebrovascular disease: a meta-analysis. Neurology 1999;53:1308–11. [16] Farrer LA, Cupples LA, Haines JL, Hyman B, Kukull WA, Mayeux R, et al. Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium. JAMA 1997;278:1349–56. [17] Becher JC, Bell JE, McIntosh N, Keeling JW. Distribution of Apolipoprotein E alleles in a Scottish healthy newborn population. Biol Neonate 2005;88:164–7. [18] Cumming AM, Robertson FW. Polymorphism at the apoprotein-E locus in relation to risk of coronary disease. Clin Genet 1984;25:310–3. [19] Hixson JE, Vernier DT. Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with HhaI. J Lipid Res 1990;31:545–8. [20] Fleming PJ, Blair PS, Bacon C, Bensley D, Smith I, Taylor E, et al. Environment of infants during sleep and risk of the sudden infant death syndrome: results of 1993–5 case-control study for confidential inquiry into stillbirths and deaths in infancy. Confidential enquiry into stillbirths and deaths regional coordinators and researchers. BMJ 1996;313:191–5. [21] Blair PS, Fleming PJ, Bensley D, Smith I, Bacon C, Taylor E, et al. Smoking and the sudden infant death syndrome: results from 1993– 5 case-control study for confidential inquiry into stillbirths and deaths in infancy. Confidential enquiry into stillbirths and deaths regional coordinators and researchers. BMJ 1996;313:195–8. [22] Scottish perinatal and infant mortality and morbidity report. ISD Scotland; 2004.
J.-C. Becher et al. [23] Leach CE, Blair PS, Fleming PJ, Smith IJ, Platt MW, Berry PJ, et al. Epidemiology of SIDS and explained sudden infant deaths. CESDI SUDI Research Group. Pediatrics 1999:104:e43. [24] Opdal SH, Rognum TO. The sudden infant death syndrome gene: does it exist? Pediatrics 2004;114:e506–12. [25] Schwartz PJ, Stramba-Badiale M, Segantini A, Austoni P, Bosi G, Giorgetti R, et al. Prolongation of the QT interval and the sudden infant death syndrome. N Engl J Med 1998;338:1709–14. [26] Weese-Mayer DE, Berry-Kravis EM, Zhou L, Maher BS, Curran ME, Silvestri JM, et al. Sudden infant death syndrome: casecontrol frequency differences at genes pertinent to early autonomic nervous system embryologic development. Pediatr Res 2004;56:391–5. [27] Opdal SH, Opstad A, Vege A, Rognum TO. IL-10 gene polymorphisms are associated with infectious cause of sudden infant death. Hum Immunol 2003;64:1183–9. [28] Forsyth L, Hume R, Howatson A, Busuttil A, Burchell A. Identification of novel polymorphisms in the glucokinase and glucose-6-phosphatase genes in infants who died suddenly and unexpectedly. J Mol Med 2005;83:610–8. [29] Kinney HC. Abnormalities of the brainstem serotonergic system in the sudden infant death syndrome: a review. Pediatr Dev Pathol 2005;8:507–24. [30] Nicoll JA, Roberts GW, Graham DI. Apolipoprotein E epsilon 4 allele is associated with deposition of amyloid beta-protein following head injury. Nat Med 1995;1:135–7. [31] Strittmatter WJ, Saunders AM, Schmechel D, Pericak-Vance M, Enghild J, Salvesen GS, et al. Apolipoprotein E: high-avidity binding to beta-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc Natl Acad Sci U S A 1993;90:1977–81. [32] Nathan BP, Chang KC, Bellosta S, Brisch E, Ge N, Mahley RW, et al. The inhibitory effect of apolipoprotein E4 on neurite outgrowth is associated with microtubule depolymerization. J Biol Chem 1995;270:19791–9. [33] Valdes-Dapena M. The sudden infant death syndrome: pathologic findings. Clin Perinatol 1992;19:701–16. [34] Obonai T, Yasuhara M, Nakamura T, Takashima S. Catecholamine neurons alteration in the brainstem of sudden infant death syndrome victims. Pediatrics 1998;101:285–8. [35] Naeye RL. Causes of the excessive rates of perinatal mortality and prematurity in pregnancies complicated by maternal urinary-tract infections. N Engl J Med 1979;300:819–23. [36] Kinney HC, Filiano JJ, Harper RM. The neuropathology of the sudden infant death syndrome. A review. J Neuropathol Exp Neurol 1992;51:115–26. [37] Ozawa Y, Okado N. Alteration of serotonergic receptors in the brain stems of human patients with respiratory disorders. Neuropediatrics 2002;33:142–9. [38] Oria RB, Patrick PD, Zhang H, Lorntz B, Castro Costa CM, Brito GA, et al. APOE4 protects the cognitive development in children with heavy diarrhea burdens in Northeast Brazil. Pediatr Res 2005;57:310–6. [39] Wright RO, Hu H, Silverman EK, Tsaih SW, Schwartz J, Bellinger D, et al. Apolipoprotein E genotype predicts 24-month bayley scales infant development score. Pediatr Res 2003;54:819–25. [40] Mahley RW, Rall Jr SC. Apolipoprotein E: far more than a lipid transport protein. Annu Rev Genomics Hum Genet 2000;1:507–37. [41] http://www.heartstats.org British Heart Foundation Statistics website. Last accessed 25th May 2007.
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
w w w. e l s e v i e r. c o m / l o c a t e / e a r l h u m d e v
Apolipoprotein E e4 and its prevalence in early childhood death due to sudden infant death syndrome or to recognised causes Julie-Clare Becher b , Jean W. Keeling a , Jeanne Bell a,⁎, Betty Wyatt a , Neil McIntosh b a b
Division of Pathology (Neuropathology), University of Edinburgh, Scotland, United Kingdom Section of Child Life & Health, University of Edinburgh, Scotland, United Kingdom
Received 4 January 2008; accepted 8 January 2008
KEYWORDS Sudden infant death syndrome; Apolipoprotein E; Alleles; Sudden unexpected death in infancy
Abstract Background: Specific genetic polymorphisms have been shown to be more common in unexplained infant death. The APOE genotype exhibits opposite effects at the extremes of age with protective effects of e4 on perinatal mortality but detrimental effects as age progresses. Objective: To determine whether the APOE e4 allele is associated with early childhood (1 week– 2 years) unexplained death (‘sudden infant death syndrome’, SIDS) or with recognised causes (non-SIDS) and to compare these cohorts with published perinatal and adult data. Methods: DNA was extracted from spleen tissue of children dying in South East Scotland between 1990 and 2002. APOE alleles (e2, e3, e4) were determined using PCR. Comparisons of allele frequencies between groups were made. Results: There were 167 SIDS cases and 117 non-SIDS cases. Allele distributions of SIDS cases were similar to healthy newborns. Allele distributions of non-SIDS cases were more similar to adults than to healthy newborns. The percentage of children with at least one e4 allele was significantly lower in non-SIDS compared to SIDS (p = 0.016). Non-SIDS cases had a higher frequency of e3 compared to SIDS cases (p = 0.01) and to healthy newborns (0.005). Conclusions: Children dying from identified causes have different APOE allele distributions from SIDS cases, but are similar to adults. Children dying from SIDS have an allele distribution comparable to healthy newborns. The prevalence of e4 in SIDS is not of an order to contribute significantly to the age-related decline in e4. © 2008 Published by Elsevier Ireland Ltd.
⁎ Corresponding author. Division of Pathology (Neuropathology), Alexander Donald Building, Western General Hospital, Crewe Road, Edinburgh, EH4 2XU, Scotland, United Kingdom. Tel.: +44 131 537 1975; fax: +44 131 537 1013. E-mail address: [email protected] (J. Bell). 0378-3782/$ - see front matter © 2008 Published by Elsevier Ireland Ltd. doi:10.1016/j.earlhumdev.2008.01.002
550
1. Introduction Despite the reduction in unexplained infant deaths following widely publicised recommended changes in care practices, sudden unexpected death in infancy (SUDI) remains a numerically important group of deaths in the first year of life. Whilst the likelihood of a second similar death within a family is low, these undoubtedly occur and studies have estimated this increase to be between 1.9 and 9.5 times greater than the background risk of SUDI within the population studied [1–4]. Although there is no doubt that a small proportion of SUDI might be explained by a single entity such as an inherited metabolic disorder [5,6] it is accepted that many unexplained infant deaths are likely to involve a combination of factors including individual predisposition, age-specific risks and an exogenous stressor [7]. Host genotype is likely to be one of these factors and various polymorphisms, including those of IL-10 and the serotonin transporter gene, have been identified as contributing to the risk of sudden infant death syndrome (SIDS) [8,9]. In addition, abnormalities in neuronal numbers and function have recently been reported in a series of SIDS infants [10]. We hypothesised that Apolipoprotein E (APOE) polymorphisms might be a factor in the causation of SIDS as APOE shows isoform-specific effects on neuronal repair and protection and as such may modulate maturational changes and response to injury within the central nervous system of the infant [11]. The APOE genotype is controlled by three alleles at a single locus producing six different genotypes. The proportions of these genotypes and their related alleles has been established in many geographically and racially different groups [12]. An excess of the e4 allele in the fetus appears protective against early fetal death and death in the perinatal period [13,14]. However, in adults, the APOE e4 genotype is linked to ischaemic stroke [15] and neurodegenerative disorders such as late onset Alzheimer's disease [16]. An age-related decline in the APOE e4 allele has been established previously [17] and suggests that individuals with this allele may be predisposed to mortality during childhood or early to mid-adulthood. In addition, because of the small increase in susceptibility to SUDI within families, we wished to explore the hypothesis that infants who die in the first year of life may have over-representation of the e4 allele. This study examines the prevalence of the different APOE alleles amongst a group of infants who died in South East Scotland from known or unknown causes and compares the allele distribution of these groups with those of a population of healthy live born Scottish infants [17], Scottish perinatal deaths [14] and healthy Scottish adults [18].
2. Methods Ethics approval was obtained from the Lothian Research Ethics Committee. Early childhood deaths (1 week–2 years) between 1990 and 2002 were identified from the records of the Pathology Department of the Royal Hospital for Sick Children, Edinburgh. Autopsy examinations of children dying either in hospital or in the community in South East Scotland are carried out in this department.
J.-C. Becher et al. All childhood deaths were investigated to a similar standard according to a protocol. This included full organ dissection, tissue and fluid sampling for microscopy, bacterial culture, rapid identification techniques and culture for viruses and biochemical investigations. The brain was fixed in buffered formalin before examination by a neuropathologist. The spinal cord was examined and toxicological investigations done in selected cases. Cases were initially selected on age alone, but perusal of causes of death following completion of all autopsy investigations was done to ensure that similar numbers of explained and unexplained deaths were selected. Many of the necropsies were carried out on instruction from a Procurator Fiscal (equivalent to a Coroner in England and Wales). The explained deaths included both sudden and nonsudden deaths from both natural and non-natural causes. Medical histories were reviewed at time of death and a summary made by the pathologist. In all deaths occurring at home, there was a death scene investigation by police and in suspicious cases, by a pathologist. Cases were categorised by autopsy findings into two groups, explained (non-SIDS) and unexplained with no or only minor pathological findings (SIDS). The small number of pathologists involved and the close interdisciplinary approach to these deaths, with joint paediatric and forensic pathological investigation of many deaths, means diagnostic shift over the study period would have been minimal. Any death in which initial categorisation was unclear underwent further investigation by the Procurator Fiscal and the police with additional interviews and pathological investigations, although five cases remained undetermined. From archival files, a block of formalin-fixed paraffinembedded tissue, usually spleen, was retrieved from the departmental archives for each case. Five sections were cut from tissue blocks and stored in 2 ml screw cap tubes until analysis. These sections, identified by number only, were processed by the Department of Neuropathology for the determination of APOE genotype without knowledge of autopsy findings. From the deaths which occurred during the period of study, a sample of spleen was fresh frozen at autopsy for genotype analysis. The APOE genotype was determined for each case using a polymerase chain reaction (PCR) method adapted from Hixson and Vernier [19] as described previously [14]. The genotype for each sample was determined from the migration pattern of bands in a metaphor gel (Fig. 1). APOE e2/2 is characterised by the presence of 91 and 81 bp fragments. APOE e4/4 is characterised by a unique 72 bp fragment while APOE e3/3 lacks the 81 and 72 bp fragments. Heterozygotes have a combination of different fragments. PCR was repeated for cases in which the band pattern was faint or equivocal until a definite result was obtained. All samples were coded for APOE genotype analysis. When all results were obtained the cases were assigned to their group of origin, SIDS or non-SIDS. The subjects investigated in this study consist of 284 childhood deaths and previously published data from 371 healthy newborns, 251 perinatal deaths (185 stillborn infants and 66 neonatal deaths) and 400 adults. Genotype comparisons within the population of childhood deaths were made between SIDS and non-SIDS cohorts, and with the populations of perinatal deaths and healthy adults. The cohort of
APOE genotype in early childhood death
551 Table 1
Cause of death in non-SIDS cases Number
Congenital anomaly Congenital heart disease Other Infection Non-accidental injury Accidental injury Perinatal causes Asphyxia-related Immaturity-related Other/specific causes Renal failure Total
36 7 29 10 24 7 3 1 117
Figure 1 Metaphor gel showing the migration pattern for ApoE genotypes. Band 1 shows the control ladder.
3.2. Comparison of early childhood deaths with healthy newborns (Tables 2A and 2B) perinatal deaths was contemporaneously similar occurring throughout Scotland [14] as were the healthy term newborns from South East Scotland [17]. The population of healthy adults were aged between 45 and 60 years and had been selected from primary care lists in North East Scotland in the mid-1970s [18]. The presence of Hardy–Weinberg equilibrium was sought in all populations to ensure they were free from outside influences, such as mutation, migration and random selection. Comparisons of allele distribution between groups were made using the Chi square test and where a p value of less than 0.05 was considered significant. The Mann–Whitney U test was used for comparison of the age of children at death.
The allele distribution of SIDS cases is similar to that of healthy newborn infants particularly with respect to the elevated percentage of cases in both groups who possess at least one e4 allele. Non-SIDS cases have a higher frequency of e3 (p = 0.005) and lower frequency of e4 compared to healthy newborns (p = 0.03). Table 2A APOE allele distributions in early childhood deaths compared to healthy newborns, perinatal deaths and healthy adults
3. Results There were 289 children dying between 1 week and 2 years. Of these 167 died of SIDS and 117 children died from known causes (non-SIDS). In five cases the category of death could not be confidently assigned and these five infants were not considered further. The median age at death of SIDS and nonSIDS cohorts was 16 wks (IQR 6.5, 32 wks) and 11.5 wks (IQR 5.3, 22 wks) respectively, (p = 0.048). 98 (59%) of SIDS cases and 70 (60%) of non-SIDS cases were male. 21 (13%) of SIDS cases and 4 (3%) of non-SIDS cases were b 37 weeks gestation at birth (p = 0.013). Table 1 shows details of the cause of death in non-SIDS cases. Tables 2A and 2B show allele and genotype distributions for each of the cohorts. The genotypic profile of the cohort of childhood deaths proved to be in Hardy–Weinberg equilibrium and therefore could be assumed to be free from outside evolutionary forces (p = 0.99). Previous work has demonstrated such equilibrium in the other groups we studied [14,17].
3.1. Comparison of SIDS with non-SIDS
All early childhood deaths (n = 284, alleles 568) SIDS (n = 167, alleles 334) Non-SIDS (n = 117, alleles 234) Healthy newborns [17] (n = 371, alleles 742) All perinatal deaths [14] (n = 251, alleles 502) Stillborn infants (n = 185, alleles 370) Neonatal deaths (n = 66, alleles 132) Adults [18] (n = 400, alleles 800) a
Non-SIDS cases have a higher frequency of e3 (p = 0.01) and lower frequency of e4 (p = 0.039) compared to SIDS cases (Tables 2A and 2B). The percentage of cases with at least one e4 allele is significantly higher in SIDS compared to non-SIDS (p = 0.016) (Table 3).
b c d e f
e2
e3
e4
n (%)
n (%)
n (%)
39 (7)
434 (76)
95 (17)
Total alleles 568
26 (8) e 243 (73) a
65 (19) b 334
13 (6) f 191 (82) c
30 (13) d 234
63 (8)
538 (72)
141 (19)
742
64 (13) 365 (72)
73 (15)
502
48 (13) 274 (74)
48 (13)
370
16 (12)
91 (69)
25 (19)
132
66 (8)
616 (77)
118 (15)
800
vs non-SIDS, p = 0.01. vs non-SIDS, p = 0.039. vs healthy newborns, p = 0.005. vs healthy newborns, p = 0.03. vs perinatal deaths, p = 0.02. vs perinatal deaths, p = 0.002.
552 Table 2B adults
J.-C. Becher et al. Frequency of APOE genotype in early childhood deaths compared to healthy newborns, perinatal deaths and healthy
Cohort
All early childhood deaths SIDS Non-SIDS Healthy newborns [17] All perinatal deaths [14] Stillborn infants Neonatal deaths Adults [18]
Apolipoprotein E genotype 2/2
3/3
4/4
2/3
2/4
3/4
n (%)
n (%)
n (%)
n (%)
n (%)
n (%)
0 0 0 2 5 3 2 2
166 (59%) 88 (52.6%) 78 (66.7%) 199 (53.6%) 138 (54.9%) 110 (60%) 28 (42.4%) 233 (58.2%)
7 (2.4%) 4 (2.4%) 3 (2.6%) 15 (4%) 3 (1.2%) 3 (1.6%) 0 4 (1%)
30 (10.5%) 18 (10.8%) 12 (10.3%) 44 (11.9%) 38 (15.1%) 27 (14.6%) 11 (16.7%) 51 (12.8%)
9 (3.1%) 8 (4.8%) 1 (0.9%) 15 (4%) 16 (6.4%) 15 (8.1%) 1 (1.5%) 11 (2.7%)
72 49 23 96 51 27 24 99
(0.5%) (2%) (1.6%) (3%) (0.5%)
3.3. Comparison of infant deaths with perinatal deaths (Tables 2A and 2B) Both SIDS and non-SIDS cohorts have a lower frequency of the e2 allele compared to perinatal deaths (p = 0.02 for SIDS and p = 0.002 for non-SIDS).
3.4. Comparisons of early childhood deaths with adults (Tables 2A and 2B) Although healthy newborns differ from adults with respect to e4, and the percentage of e4 is found to be elevated in SIDS to the same degree as in healthy newborns, there was no difference between SIDS cases and adults. The allele distribution of non-SIDS cases does not differ significantly from that of adults.
4. Discussion Despite widely publicised recommendations in infant care practices such as such as the ‘Back to Sleep’ campaign, advice about appropriate sleep clothing and covering and the importance of a smoke-free environment [20,21] SUDI remains an important group of deaths. Accurate and thorough autopsy examination reveals a cause of death in a proportion of these but a majority remain unexplained. This group (SIDS) accounts for around 20% of all post-neonatal deaths in the United Kingdom [22]. This paper has established the distribution of APOE alleles and genotypes in a population of Scottish children dying in the first 2 years of life from known (non-SIDS) and unknown causes (SIDS), and has shown clear differences between the two groups with over-representation of e4 in cases dying from SIDS. In this respect, SIDS resemble healthy newborns whereas children dying from established causes have an allele distribution similar to adults. Such disparity is at odds with the spectrum of characteristics which are common to both SIDS and explained deaths, such as prematurity, male gender, low maternal age and socioeconomic deprivation [23]. The small discrepancy in the age at death of each cohort is unlikely to explain this difference. It is generally accepted that SIDS deaths occur earlier in the first year of life than other SUDI [23]. Within this study, the SIDS cases
(25.3%) (29.3%) (19.7%) (25.9%) (20.3%) (14.6%) (36.4%) (24.8%)
Total
284 167 117 371 251 185 66 400
were older at death than non-SIDS cases. The cohort of nonSIDS however, comprised not only of explained deaths in children dying suddenly and unexpectedly (SUDI), but also of deaths in chronic conditions such as prematurity and congenital anomaly. The small numbers in each diagnostic group within the cohort of infants dying from recognised causes (Table 1) precluded further investigation of the role of APOE in specified causes of death such as infection, although there is evidence suggesting that APOE may have an immunomodulatory role both in the central nervous system and systemically [40]. The childhood and perinatal deaths studied in this paper occurred in the decade between 1990 and 2002. Although the adult population is historical, the genotype distribution of this group is entirely consistent with other well-described Caucasian populations [12]. All groups studied were shown to be in Hardy–Weinberg equilibrium and therefore the differences seen in genotype between groups cannot be explained by evolutionary forces such as genetic drift, selection or mutation. The APOE genotype distribution is known to differ between ethnic populations but we believe that the proportion of non-Caucasians within the Scottish population has changed negligibly over the twenty year study period and as such fluctuations in racial demographics is unlikely to explain any differences observed between populations. The presence of Hardy–Weinberg equilibrium supports this. The methodology of genotyping differed between deceased and living cohorts, as DNA was extracted from paraffin blocks rather than blood spots respectively.
Table 3 Comparison of the APOE e4 allele within subgroups of childhood deaths (other populations tabulated below)
All childhood deaths SIDS⁎ Non-SIDS Healthy newborns All perinatal deaths Adults ⁎p = 0.016 vs non-SIDS.
e4 carriers
Non-e4 carriers
Total
n (%)
n (%)
n
88 61 27 126 70 114
196 (69) 106 (64) 90 (77) 245 (66) 181 (72) 286 (71)
284 167 117 371 251 400
(31) (37) (23) (34) (28) (29)
APOE genotype in early childhood death Although genotyping may be affected by DNA breakage related to formalin fixation, we believe that the gel banding of Apolipoprotein E is highly characteristic and as such, error in interpretation of results is unlikely. It remains possible that the differences observed occurred by chance although the numbers in each population are not small. It is also possible that infants who were deemed to be healthy at birth died later in childhood, and thus have been counted twice. If this was true, we estimate from Scottish infant mortality rates, that at most this would apply to only one of this group and would be unlikely to significantly bias our results. Although a small minority of cases previously diagnosed as SIDS appear to be caused directly by genetic mutations, for example in the genes encoding medium chain acyl-CoA dehydrogenase [24] or cardiac ion channels [25], it is thought that in other cases a genetic polymorphism may predispose to death when an infant is in a critical situation. It is unlikely however that one single mutation or polymorphism is the predisposing factor in all SIDS cases. The widely debated triple risk hypothesis of SIDS causation encompasses individual vulnerability, a critical developmental period, and an exogenous stressor such as an upper respiratory infection [7]. Suggested genetic predisposing factors have included mutations in embryological genes important in the developing autonomic nervous system [26] as well as various polymorphisms of IL-10 [27], the 5HT transporter gene [9] and glucokinase [28]. Others have suggested that prenatal damage, particularly to the developing brainstem, may make an infant vulnerable to adverse events postnatally [29]. In adults, APOE e4 is associated with excess deposition of β-amyloid in the brain following head injury [30] and with an earlier onset of Alzheimer's disease [31]. At a cellular level APOE e4 is associated with less vigorous neuritic outgrowth and repair than the other APOE isoforms [32]. A variety of subtle forms of neuropathological damage has been described in SIDS infants [33] ranging from increased dendritic spines [34] to brainstem gliosis [35–37]. However the APOE genotype does not appear to be associated with brain injury among perinatal deaths [14] and in this study, the APOE genotype of SIDS cases was no different than in healthy newborn infants. The e4 allele is under-represented in perinatal deaths compared to healthy newborns suggesting an antagonistic pleiotropic effect of this allele at the extremes of age, affording protection during early development but having deleterious effects as age progresses. This theory is supported by beneficial effects of e4 on cognitive development in children [38,39]. This study showed that the e4 allele is less common in children dying from explained causes compared to healthy newborns. The role of the APOE genotype in adult brain injury and cognitive deterioration is the focus of much research but early protective effects are less well understood and may involve modulation of the immune response or mechanisms of CNS repair [40]. Although some of the explained deaths occurred following infection or brain injury, the causes of death in this group of non-SIDS are heterogeneous and a single unifying hypothesis is unlikely. There is substantial overlap in causation between early childhood and perinatal deaths, and this may explain the similarity of e4 proportion in these groups. It is notable that an unusually high frequency of e2 is confined to deaths occurring around or before birth. Paradoxically APOE e2 is considered to be neuroprotective later in life [40].
553 Previous work has shown that healthy Scottish newborns have an excess of e4 compared to healthy Scottish latemiddle aged adults suggesting an age-related decline due to early mortality or morbidity of individuals with the e4 allele [17]. Although the rate of post-neonatal death at 1.8/ 1000LBs is low [22] and the numbers unlikely to contribute substantially to this decline, we found no significant excess of e4 overall amongst children dying in early childhood although the subset of SIDS cases did show a small rise in e4. It may be inferred that the reduction in e4 seen in this selected population of healthy late-middle aged adults is more likely to be attributed to early adult mortality or morbidity. The e4 allele has known associations with atherosclerosis and hypercholesterolaemia [12], both of which are risk factors for ischaemic heart disease, a leading cause of premature morbidity and mortality in the Scottish population [41]. In conclusion the findings of this investigation indicate that APOE e4 is less common in infants dying from recognised causes in early childhood than in either healthy newborns or SIDS, suggesting that e4 may be protective against a subgroup of infant mortality. In contrast we found no association of APOE e4 with SIDS, a group who have a similar genotype distribution to healthy newborns. Longitudinal cohort studies of APOE genotype are required to fully explain the decline in e4 throughout life and the antagonistic effects of e4 at the extremes of age. As our understanding of the genome increases it has become clear that not only may genes be switched on and off but they may also be site-specific within a cell. Within such a dynamic cellular environment, proteomic and metabolomic research will be necessary to further our understanding of SIDS and poorly understood gene-environment interactions will require future exploration.
Acknowledgements The authors would like to thank the Scottish Cot Death Trust for the funding of the study. Ian Croy and Iain Anthony gave considerable help with the APOE methodology and Dr Rob Elton advised on the statistical analysis.
Competing interest statement Jeanne Bell, Jean Keeling and Neil McIntosh are consulted regularly by both defence and prosecution teams with regard to legal proceedings in infant deaths. There are no financial competing interests for any of the authors.
References [1] Peterson DR, Sabotta EE, Daling JR. Infant mortality among subsequent siblings of infants who died of sudden infant death syndrome. J Pediatr 1986;108:911–4. [2] Smith GC, Wood AM, Pell JP, Dobbie R. Sudden infant death syndrome and complications in other pregnancies. Lancet 2005;366:2107–11. [3] Beal SM, Blundell HK. Recurrence incidence of sudden infant death syndrome. Arch Dis Child 1988;63:924–30. [4] Oyen N, Skjaerven R, Irgens LM. Population-based recurrence risk of sudden infant death syndrome compared with other infant and fetal deaths. Am J Epidemiol 1996;144:300–5.
554 [5] Emery JL, Howat AJ, Variend S, Vawter GF. Investigation of inborn errors of metabolism in unexpected infant deaths. Lancet 1988;2:29–31. [6] Green A, Preece MA, Hardy D. More on the metabolic autopsy. Clin Chem 2002;48:964–5. [7] Guntheroth WG, Spiers PS. The triple risk hypotheses in sudden infant death syndrome. Pediatrics 2002;110:e64. [8] Korachi M, Pravica V, Barson AJ, Hutchinson IV, Drucker B. Interleukin 10 genotype as a risk factor for sudden infant death syndrome: determination of IL-10 genotype from waxembedded postmortem samples. FEMS Immunol Med Microbiol 2004;42:125–9. [9] Weese-Mayer DE, Zhou L, Berry-Kravis EM, Maher BS, Silvestri JM, Marazita ML. Association of the serotonin transporter gene with sudden infant death syndrome: a haplotype analysis. Am J Med Genet A 2003;122:238–45. [10] Paterson DS, Trachtenberg FL, Thompson EG, Belliveau RA, Beggs AH, Darnall R, et al. Multiple serotonergic brainstem abnormalities in sudden infant death syndrome. JAMA 2006; 296:2124–32. [11] Weisgraber KH, Mahley RW. Human apolipoprotein E: the Alzheimer's disease connection. FASEB J 1996;10:1485–94. [12] Davignon J, Gregg RE, Sing CF. Apolipoprotein E polymorphism and atherosclerosis. Arteriosclerosis 1988;8:1–21. [13] Zetterberg H, Palmer M, Ricksten A, Poirier J, Palmqvist L, Rymo L, et al. Influence of the apolipoprotein E epsilon4 allele on human embryonic development. Neurosci Lett 2002;324:189–92. [14] Becher JC, Keeling JW, McIntosh N, Wyatt B, Bell J. The distribution of apolipoprotein E alleles in Scottish perinatal deaths. J Med Genet 2006;43:414–8. [15] McCarron MO, Delong D, Alberts MJ. APOE genotype as a risk factor for ischemic cerebrovascular disease: a meta-analysis. Neurology 1999;53:1308–11. [16] Farrer LA, Cupples LA, Haines JL, Hyman B, Kukull WA, Mayeux R, et al. Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium. JAMA 1997;278:1349–56. [17] Becher JC, Bell JE, McIntosh N, Keeling JW. Distribution of Apolipoprotein E alleles in a Scottish healthy newborn population. Biol Neonate 2005;88:164–7. [18] Cumming AM, Robertson FW. Polymorphism at the apoprotein-E locus in relation to risk of coronary disease. Clin Genet 1984;25:310–3. [19] Hixson JE, Vernier DT. Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with HhaI. J Lipid Res 1990;31:545–8. [20] Fleming PJ, Blair PS, Bacon C, Bensley D, Smith I, Taylor E, et al. Environment of infants during sleep and risk of the sudden infant death syndrome: results of 1993–5 case-control study for confidential inquiry into stillbirths and deaths in infancy. Confidential enquiry into stillbirths and deaths regional coordinators and researchers. BMJ 1996;313:191–5. [21] Blair PS, Fleming PJ, Bensley D, Smith I, Bacon C, Taylor E, et al. Smoking and the sudden infant death syndrome: results from 1993– 5 case-control study for confidential inquiry into stillbirths and deaths in infancy. Confidential enquiry into stillbirths and deaths regional coordinators and researchers. BMJ 1996;313:195–8. [22] Scottish perinatal and infant mortality and morbidity report. ISD Scotland; 2004.
J.-C. Becher et al. [23] Leach CE, Blair PS, Fleming PJ, Smith IJ, Platt MW, Berry PJ, et al. Epidemiology of SIDS and explained sudden infant deaths. CESDI SUDI Research Group. Pediatrics 1999:104:e43. [24] Opdal SH, Rognum TO. The sudden infant death syndrome gene: does it exist? Pediatrics 2004;114:e506–12. [25] Schwartz PJ, Stramba-Badiale M, Segantini A, Austoni P, Bosi G, Giorgetti R, et al. Prolongation of the QT interval and the sudden infant death syndrome. N Engl J Med 1998;338:1709–14. [26] Weese-Mayer DE, Berry-Kravis EM, Zhou L, Maher BS, Curran ME, Silvestri JM, et al. Sudden infant death syndrome: casecontrol frequency differences at genes pertinent to early autonomic nervous system embryologic development. Pediatr Res 2004;56:391–5. [27] Opdal SH, Opstad A, Vege A, Rognum TO. IL-10 gene polymorphisms are associated with infectious cause of sudden infant death. Hum Immunol 2003;64:1183–9. [28] Forsyth L, Hume R, Howatson A, Busuttil A, Burchell A. Identification of novel polymorphisms in the glucokinase and glucose-6-phosphatase genes in infants who died suddenly and unexpectedly. J Mol Med 2005;83:610–8. [29] Kinney HC. Abnormalities of the brainstem serotonergic system in the sudden infant death syndrome: a review. Pediatr Dev Pathol 2005;8:507–24. [30] Nicoll JA, Roberts GW, Graham DI. Apolipoprotein E epsilon 4 allele is associated with deposition of amyloid beta-protein following head injury. Nat Med 1995;1:135–7. [31] Strittmatter WJ, Saunders AM, Schmechel D, Pericak-Vance M, Enghild J, Salvesen GS, et al. Apolipoprotein E: high-avidity binding to beta-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc Natl Acad Sci U S A 1993;90:1977–81. [32] Nathan BP, Chang KC, Bellosta S, Brisch E, Ge N, Mahley RW, et al. The inhibitory effect of apolipoprotein E4 on neurite outgrowth is associated with microtubule depolymerization. J Biol Chem 1995;270:19791–9. [33] Valdes-Dapena M. The sudden infant death syndrome: pathologic findings. Clin Perinatol 1992;19:701–16. [34] Obonai T, Yasuhara M, Nakamura T, Takashima S. Catecholamine neurons alteration in the brainstem of sudden infant death syndrome victims. Pediatrics 1998;101:285–8. [35] Naeye RL. Causes of the excessive rates of perinatal mortality and prematurity in pregnancies complicated by maternal urinary-tract infections. N Engl J Med 1979;300:819–23. [36] Kinney HC, Filiano JJ, Harper RM. The neuropathology of the sudden infant death syndrome. A review. J Neuropathol Exp Neurol 1992;51:115–26. [37] Ozawa Y, Okado N. Alteration of serotonergic receptors in the brain stems of human patients with respiratory disorders. Neuropediatrics 2002;33:142–9. [38] Oria RB, Patrick PD, Zhang H, Lorntz B, Castro Costa CM, Brito GA, et al. APOE4 protects the cognitive development in children with heavy diarrhea burdens in Northeast Brazil. Pediatr Res 2005;57:310–6. [39] Wright RO, Hu H, Silverman EK, Tsaih SW, Schwartz J, Bellinger D, et al. Apolipoprotein E genotype predicts 24-month bayley scales infant development score. Pediatr Res 2003;54:819–25. [40] Mahley RW, Rall Jr SC. Apolipoprotein E: far more than a lipid transport protein. Annu Rev Genomics Hum Genet 2000;1:507–37. [41] http://www.heartstats.org British Heart Foundation Statistics website. Last accessed 25th May 2007.