Association between birth weight and adult blood pressure in twins: historical cohort study.

OBJECTIVES To evaluate the associations in twins between within pair differences in birth weight and subsequent blood pressures as adults thereby removing the impact of potential parental confounding variables. DESIGN Historical cohort study. SETTING St Thomas's UK adult twin register, June 1992 to September 1995. PARTICIPANTS 492 pairs of female twins (mean age 54 years). MAIN OUTCOME MEASURES Mean within pair differences in adult blood pressure in each of four strata of within pair differences in birth weight (0, 1-500 g, 501-1000 g, >1000 g). Differences in blood pressure were analysed before and after adjustment for potential confounders between adult twins, after exclusion of those twin pairs including at least one twin taking antihypertensive drugs, and by zygosity. RESULTS Reported mean birth weights of heavier and lighter twins were 2.51 (SD 0.61) v 2.12 (0.59) kg respectively. A graded inverse relation between strata of within pair differences in birth weight and differences in adult blood pressure was apparent, with an adjusted blood pressure range of 8.7/5.1 mm Hg across the four strata (test for trend: systolic, P=0.05; diastolic, P=0.09). After excluding those women taking antihypertensive drugs the significance of the association was similar (systolic, P=0.04; diastolic, P=0.10). When differences in blood pressure were stratified for zygosity similar but non-significant trends were apparent. CONCLUSION It would seem that birth weight is inversely associated with adult blood pressure and that this association is independent of parental confounding variables probably including, in view of the findings in monozygotic twins, genetic factors. The observed blood pressure differences are likely to result from retarded intrauterine growth due to placental dysfunction rather than inadequate maternal nutrition.