Dear Sir,
Venneman, Bajanowski, Brinkmann et als’ article(1) on the reduction
of risk of SIDS associated with breast feeding largely confirms previous
findings. In 1965 Carpenter and Shaddick(2) compared 110 cases of sudden
unexplained deaths in infancy with 196 normal controls matched for age and
sex. They reported that at ages 0, 0.5, 1, 2, & 3 months 56.3%,
38.8%, 24.7%, 8.8%, and 5.6 % of cases were fully breast fed compared
with 70.4%, 58.1%, 41.8%, 26.5%, and 21.6% of controls at those ages.
These percentages, presented as a graph similar the Vennemann et al’s,
used all available records at each age.
Comparing these percentages with Venneman et al’s Fig.2, it may be
seen that initial breast feeding rates were similar for controls and much
higher for the cases. However, forty years ago the duration of breast
feeding in for both cases and controls was much shorter so that only 21.6%
of controls were fully breast fed at 3 months compared with 50% in the
later study.
Discriminant analysis, which approximates to multiple logistic
regression(3) showed that bottle feeding at two weeks, shared bed, soft
pillow, found with bedding partly or completely covering mouth and nose,
and respiratory symptoms were significant risk factors(2).
When identifying infants at increased risk of SIDS at or soon after
birth, intention to breast feed has been associated with a highly
significant reduction of risk(4). However, the CESDI SUDI study(5) and
the ECAS study(6) found that risk associated with bottle feeding from
birth was accounted for by factors detailing the circumstances of death,
in contrast to this new German study(1). However, when the type of
feeding in the last 7 days was included in the ECAS model the odds ratio
for complete breast feeding vs. bottle feeding was 0.50 with 95% CI
(0.33–0.76) after adjusting for 17 covariables(7). This suggested a
reduction of risk that was less but not significantly less than Vennemann
et al’s findings(1).
Estimated reduction in risk associated with partial breast feeding
differs between the studies. In the London and Cambridge study2, the
percentage of infants partially breast fed was greater for cases than for
controls at every age, suggesting that partial breast feeding was
associated with increased risk. In the ECAS study, the adjusted OR for
partial breast feeding in the last 7 days vs. bottle feeding is 0.85 and
is not significant(7). Vennemann et al’s Fig 1 shows the percentage of
cases and controls fully and partially breast fed by age. Superimposing
Fig 1 on Fig 2, it can be seen that the percentage of cases and controls
partially breast fed is similar up to 3 months of age. Thereafter the
percentage of controls partially breast fed progressively exceeds that of
the cases. This suggests that partial breast feeding in the first two
months may not be protective. The differences in estimates of the benefit
of partial breast feeding between the ECAS findings(7) and Vennemann et
al’s more recent findings(1) may be due to changes in the prevalence of
other risk factors to which infants are exposed and the nature of the
supplements infants were and are given.
Finally, the ECAS analysis(7)showed that bed sharing is associated
with prolongation of breast feeding. However, the reduction of risk
associated with breast feeding is cancelled out by the increased risk
associated with bed sharing.
R.G. Carpenter
Medical Statistics Unit,
London School of Hygiene & Tropical Medicine,
London, England.
References
1. Vennemann MM, Bajanowski T, Brinkmann B, Jorch G, Yücesan K,
Sauerland C, Mitchell EA, and the GeSID Study Group. Does breastfeeding
reduce the risk of Sudden Infant Death Syndrome? Pediatrics 2009; 123:
e406-e410.
2. Carpenter RG, Shaddick CW. Role of infection, suffocation, and
bottle feeding in Cot Death. Brit J Prev Soc Med. 1965; 19: 1–7.
3 Carpenter RG. Scoring to provide risk-related primary health care:
evaluation an up-dating during use. J Royal Statist Soc. Ser A 1983; 146:
1–32.
4. Carpenter RG, Gardner A, Mcweeny PM, Emery JL. A multistage
scoring system for identifying infants at risk of sudden death. Arch Dis
Childh. 1977; 52: 606–12.
5. Fleming P, Bacon C, Blair P, Berry PJ. Sudden unexpected deaths
in infancy. The CESDI SUDI studies 1993–1996. Stationery Office. 2000.
London.
6. Carpenter R, Irgens LM, Blair, P et al. Sudden unexplained infant
death in 20 regions in Europe: Case control study. Lancet 2004; 363:
185–91.
7 Carpenter RG. The hazards of bed sharing. Paediatrics & Child
Health (Canada) 2006. 11: Suppl A: 24A–28A.
Conflict of Interest:
None declared
eLetters: ![[Read eLetters]](/icons/misc/sectionDOWN.gif){kind=icon}
