Homocysteine levels, paraoxonase 1 (PON1) activity, and cardiovascular risk.

lyzed our study population by dividing the births within each gestational age category into smaller and larger birth weights by their z score of birth weight by gestational age. Adjusting for gestational week within each category as well as year of birth, larger infants did have the best survival for any given gestational age. For 22 to 27, 28 to 32, 33 to 36, and 37 to 42 weeks, the relative risks of mortality were 0.39 (95% confidence interval [CI], 0.30-0.49), 0.45 (95% CI, 0.40-0.50), 0.40 (95% CI, 0.36-0.46), and 0.53 (95% CI, 0.50-0.55), respectively. We agree that predisposing genetic factors contribute to prematurity as well as subsequent reproductive potential; this is an important area for future investigation. A twin-pair study would be ideal for examining genetic contributions while controlling for potential environmental confounders. Furthermore, comparisons between twin sibships would be useful for examining contributors to birth weight, given that twin siblings have identical gestational age at birth. We limited our study population to singleton births given the strong association and potential confounding effect between multiple gestation and preterm birth. Furthermore, compelling twin-pair studies require known zygosity (identical vs fraternal twinning) based on ultrasound confirmation, genetic confirmation, or self-reported zygosity based on resemblance. Further analyses to disentangle the relative contributions of birth weight and length of gestation to long-term health outcomes, including twin-pair studies, are essential to improving the understanding of the fetal origins of adult disease.