Sex Differences in the Association Between Birth Weight and Total Cholesterol. A Meta-Analysis DEBBIE A. LAWLOR, PHD, CHRISTOPHER G. OWEN, PHD, ANNA A. DAVIES, MSC, PETER H. WHINCUP, PHD, SHAH EBRAHIM, MD, DEREK G. COOK, PHD, AND GEORGE DAVEY SMITH, DSC PURPOSE: Determine whether a sex difference exists in the association between birth weight and total cholesterol later in life. METHODS: Meta-analysis of within-study differences in regression coefficients of cholesterol on birth weight. RESULTS: A total of 34 regression coefficients from 30 studies were included in the analyses; these provided data on 33,650 males and 23,129 females. There was evidence that the inverse association between birth weight and total cholesterol was stronger in males compared to females. The pooled within- study difference in age-adjusted regression coefficients was -0.03 mmol/l (-0.06, -0.01), p Z 0.02 and the pooled within-study difference in age and body mass index adjusted regression coefficients was -0.04 mmol/l (-0.07, -0.02), p Z 0.002. There was no evidence of heterogeneity in these meta-analyses (both p values O 0.6). CONCLUSIONS: These results provide some evidence for a sex difference in the birth weight – total cholesterol association. This is consistent with studies of fetal growth which suggest that birth size reflects different biological processes for females and males. However, other very large studies are required to confirm this finding. Ann Epidemiol 2006;16:19–25. Ó 2006 Elsevier Inc. All rights reserved. KEY WORDS: Fetal Programming, Cholesterol, Birth Weight, Meta-Analysis, Sex Differences. INTRODUCTION A number of studies have examined the association of birth weight with total cholesterol in later life and a meta-analysis found a weak or modest inverse association (1). Interest- ingly, several of the primary studies, have found a strong inverse association in men but no or weak associations in women (2–7). A sex difference in the effect of birth weight on adult disease outcomes might be predicted from sex differences in fetal-growth patterns (8, 9). It has been suggested that faster growing male fetuses are more vulnerable to the effects of fetal undernutrition. Fetal demand for nutrients is influenced by growth trajectory resulting in a faster growing fetus being more vulnerable to impaired nutrition (8) (10) (11). There is some evidence that growth trajectories differ between males and females, with male fetuses growing on average at a faster rate than female fetuses. If fetal nutritional programming is the underlying mechanism for the association between birth weight and total cholesterol then one would expect the association to be stronger in males compared to females (8). A recent study of serial weekly ultrasound scans identified a more complex picture of sex differences in growth patterns, with a male delay in growth and development and a greater dependency on third trimester conditions compared to females (9). From these differences in growth patterns the authors predicted that birth weight would be more strongly inversely associated with adult diseases in later life among males compared to females (9). Individual studies or meta-analyses that do not assess the effect of sex on the associations between birth weight and CHD risk and risk factors may therefore have under- estimated the true effect in men (9). However, when we examined the possibility of a sex difference in the association between birth weight and blood-pressure using meta-analysis we found no evidence of a sex difference (12). Further examination of sex differences in birth weight-CHD risk factor associations is warranted and here we use meta- analysis to assess whether a sex difference exists in the association between birth weight and total cholesterol. From the Department of Social Medicine, University of Bristol, United Kingdom (D.A.L., A.A.D., S.E., G.D.S.); and Division of Community Health Sciences, St George’s, University of London, London, United Kingdom (C.G.O., P.H.W., D.G.C.). Address correspondence to D. A. Lawlor, Department of Social Medicine, University of Bristol, Canynge Hall, Whiteladies Road, Bristol BS8 2PR, UK. Tel: C44 (0)117 928 7267; fax: C44 (0)117 928 7365. E-mail: d.a.lawlor@bristol.ac.uk This work was supported in part by a Department of Health (UK) Career Scientist Award (D.A.L.) and a British Heart Foundation Senior Research Fellowship (C.G.O.). Received July 7, 2004; accepted April 26, 2005. Ó 2006 Elsevier Inc. All rights reserved. 1047-2797/06/$–see front matter 360 Park Avenue South, New York, NY 10010 doi:10.1016/j.annepidem.2005.04.006