Individuals who are small at birth are at increased risk of ischaemic heart disease (IHD) in later life. One hypothesis to explain this association - the Barker hypothesis - is fetal adaptation to a suboptimum intrauterine environment. Gordon Smith and colleagues from the Queen Mother's Hospital, Glasgow, UK, retrospectively investigated whether pregnancy complications associated with low birthweight were related to risk of subsequent maternal IHD.
Around 130 000 women in Scotland who had a first singleton delivery between 1981 and 1985 were studied. The risk of subsequent maternal IHD hospital admission or death was double for women whose babies were in the smallest 20% of birthweight for gestational age. A doubling of risk of maternal IHD was also independently associated with preterm delivery and with pre-eclampsia (high blood pressure during pregnancy). Women with all three characteristics were 7 times more likely to be admitted to hospital or die from IHD. The investigators propose that genetic predisposition may explain the association between low birthweight and subsequent maternal IHD.
In an accompanying Commentary (p 1990) Neil Poulter from Imperial College School of Medicine, London, UK, cautions that the emphasis on first-born babies and the absence of data on smoking may confound the results of the study. He states: "Confounding by parental characteristics, genetic or environmental, will only be eliminated in studies of monozygotic twins that investigate associations between anthropometric features within the twin-pairs and the development of a cardiovascular risk factor (ie, hypertension or diabetes) or a cardiovascular event in later life."
Twins study suggests low birthweight is not independently associated with heart attack later in life
With the aim of reducing genetic and environmental confounding factors, Anna Hubinette and colleagues from the Karolinska Institute, Sweden, did a case-control study in Swedish twins born before 1958 of which only one twin had acute myocardial infarction (heart attack,[AMI]) later in life. 132 same-sex twins discordant for AMI were compared with 118 matched twin pairs without AMI. In comparisons between AMI cases and external matched control twins, cases had significantly lower birthweight (average 2556 g compared with 2699 g); average birthlength and head circumference was also reduced in twins with AMI compared with external controls. However, no difference in birthweight, birthlength, and head circumference was found within twin pairs where one twin had AMI. The investigators conclude that the lack of an association between birth characteristics and AMI within twin pairs suggests that previously reported associations may be influenced by genetic and early environmental factors, or by unmeasured maternal factors that operate independently of birthweight.
In his Commentary (p 1990) Neil Poulter identifies three factors that weaken the twins study: first, he argues that the study is small, with only 132 twin pairs discordant for AMI; second, he comments on the methodological weakness of using a single timepoint to assess disease or non-disease status in the twin pairs discordant for AMI; third, he comments on the lack of information about environmental risk factors for AMI such as exercise, alcohol intake, and smoking habits. He concludes: "More valid evidence would arise from an analysis of rates of genuine discordance for coronary heart disease (perhaps evaluated by a good indirect measure of coronary atherosclerosis) across strata of birthweight differences in a much larger sample of DNA-confirmed monozygotic twins."
Contact: Dr Gordon C S Smith, Department of Obstetrics and Gynaecology, The Queen Mother's Hospital, Yorkhill, Glasgow G3 8SJ, UK; T) +44 (0)141 201 0550, page 2358; M) +44 (0)7974 786121; F) +44 (0)141 357 3610; E) gcs4@cornell.edu
Professor Paul Lichtenstein, Department of Medical Epidemiology, Karolinska Institutet, Box 281, SE-171 77, Stockholm, Sweden; T) +46 8 728 74 24 or +46 176 234 302; F)+46 8 31 49 75; E) paul.lichtenstein@mep.ki.se or Professor Sven Cnattingius, Department of Medical Epidemiology, Karolinska Institutet, Box 281, SE-171 77 Stockholm, Sweden; T) +46 8 728 6190; F)+46 8 31 49 75; E) sven.cnattingius@mep.ki.se
Professor Neil R Poulter, Department of Clinical Pharmacology & Therapeutics, Cardiovascular Studies Unit, Imperial College School of Medicine, London W2 1PG, UK; T) +44 (0)20 7594 3446/3445; F) +44 (0)20 7594 3411; E) .poulter@ic.ac.uk
Journal
The Lancet