Birth Weight and Hypertension:Do Not Confound withGenetic and Shared Environmental Factors

Genetic and Shared Environmental Factors Do Not Confound the Association Between Birth Weight and Hypertension
A Study Among Swedish Twins
Niklas Bergvall, MSc; Anastasia Iliadou, PhD; Stefan Johansson, MD; Ulf de Faire, MD, PhD; Michael S. Kramer, MD; Yudi Pawitan, PhD; Nancy L. Pedersen, PhD; Paul Lichtenstein, PhD; Sven Cnattingius, MD, PhD
From the Department of Medical Epidemiology and Biostatistics (N.B., A.I., S.J., Y.P., N.L.P., P.L., S.C.), Division of Cardiovascular Epidemiology, Institute of Environmental Medicine and Department of Cardiology, Karolinska University Hospital (U.d.F.), Karolinska Institutet, Stockholm, Sweden, and Departments of Pediatrics and of Epidemiology and Biostatistics, McGill University Faculty of Medicine, Montreal, Canada (M.S.K.).
Correspondence to Niklas Bergvall, MSc, Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, PO Box 281, SE–171 77 Stockholm, Sweden. E-mail niklas.bergvall@ki.se

Received November 7, 2006; accepted March 30, 2007.

Background— Studies have found associations between low birth weight and increased risks of cardiovascular diseases in adulthood. However, these associations could be due to confounding by genetic or socioeconomic factors.

Methods and Results— We performed a study on Swedish like-sexed twins with known zygosity who were born from 1926 to 1958. First, to obtain an overall effect of birth weight on risk of hypertension, we performed cohort analyses on all twins (n=16 265). Second, to address genetic and shared environmental confounding, we performed a nested co-twin control analysis within 594 dizygotic and 250 monozygotic twin pairs discordant for hypertension. Birth characteristics, including birth weight, were obtained from original birth records. Information from adulthood was collected from a postal questionnaire in 1973 (body mass index, height, smoking, and alcohol use) and from a telephone interview conducted from 1998 to 2002 (hypertension and socioeconomic status). Hypertension was defined as reporting both high blood pressure and treatment with antihypertensive medication. In the cohort analysis, the adjusted odds ratio for hypertension in relation to a 500-g decrease in birth weight was 1.42 (95% confidence interval, 1.25 to 1.61). In the co-twin control analyses, the corresponding odds ratios were 1.34 (95% confidence interval, 1.07 to 1.69) for dizygotic and 1.74 (95% confidence interval, 1.13 to 2.70) for monozygotic twins.

Conclusions— In the largest twin study on the fetal origins of hypertension, we found that decreased birth weight is associated with increased risk of hypertension independently of genetic factors, shared familial environment, and risk factors for hypertension in adulthood, including body mass index.