Pregnancy-induced hypertension and infant mortality: roles of birthweight centiles and gestational age XK Chen, a SW Wen, a G Smith, b Q Yang, c M Walker c a OMNI Research Group, Clinical Epidemiology Program, Ottawa Health Research Institute, University of Ottawa, Ottawa, Ontario, Canada b Queen’s Perinatal Research Unit, Department of Obstetrics and Gynecology, Kingston General Hospital, Queen’s University, Kingston, Ontario, Canada c OMNI Research Group, Department of Obstetrics & Gynecology, University of Ottawa, Ottawa, Ontario, Canada Correspondence: Dr SW Wen, OMNI Research Group, Clinical Epidemiology Program, Ottawa Health Research Institute, University of Ottawa, 501 Smyth Road, Box 241, Ottawa, Ontario, Canada K1H 8L6. Email swwen@ohri.ca Accepted 6 October 2006. Objective To assess the effect of pregnancy-induced hypertension (PIH) on infant mortality in different birthweight centiles (small for gestational age [SGA], appropriate for gestational age [AGA], and large for gestational age [LGA]) and gestational ages (early preterm, late preterm, and full term). Design Retrospective cohort study. Setting Linked birth and infant death data set of USA between 1995 and 2000. Population A total of 17 464 560 eligible liveborn singleton births delivered after 20th gestational week. Methods Multivariate logistic regression models were applied to evaluate the association between PIH and infant mortality, with adjustment of potential confounders stratified by birthweight centiles and gestational age. Main outcome measure Infant death (0–364 days) and its three components: early neonatal death (0–6 days), late neonatal death (7–27 days), and postneonatal death (28–364 days). Results PIH was associated with decreased risks of infant mortality, early neonatal mortality, and late neonatal mortality in both preterm and term SGA births, and PIH was associated with lower postneonatal mortality in preterm SGA births. PIH was associated with decreased risks of infant mortality, early neonatal mortality, late neonatal mortality and postneonatal mortality in preterm AGA births. Decreased risk of infant mortality and early neonatal mortality was associated with PIH in early preterm LGA births. Conclusions The association between PIH and infant mortality varies depending on different birthweight centiles, gestational age, and age at death. PIH is associated with a decreased risk of infant mortality in SGA births, preterm AGA births, and early preterm LGA births. Keywords Birthweight centile, gestational age, infant mortality, pregnancy-induced hypertension, preterm birth, small-for-gestational-age birth. Please cite this paper as: Chen X, Wen S, Smith G, Yang Q, Walker M. Pregnancy-induced hypertension and infant mortality: roles of birthweight centiles and gestational age. BJOG 2007;114:24–31. Introduction Pregnancy-induced hypertension (PIH) is a generic classi- fication of hypertension disorders occurring during preg- nancy that includes gestational hypertension (hypertension without proteinuria), pre-eclampsia (hypertension with pro- teinuria), and eclampsia (pre-eclampsia with seizures). 1 PIH remains a leading cause of maternal and infant morbidity and mortality worldwide, 2 complicating up to 10% of preg- nancies. 3 Its aetiology and pathophysiology are largely unknown. 4 The impact of PIH on fetal growth is complex. 2 Previous studies suggested that on one hand, PIH was associated with a significantly increased risk of preterm births, 5–7 low birth- weight, 7 and small-for-gestational-age (SGA) births 2,5–8 and on the other hand, PIH was associated with an increased rate of high birthweight 7 and large-for-gestational-age (LGA) births. 2,7,8 Preterm PIH was found to be associated with lighter, shorter, and leaner newborns, whereas term PIH increased rates of both larger and smaller newborns. 8 Most infants born to mothers with PIH were appropriate for ges- tational age (AGA) or even LGA. 2,8 Based on the complicated 24 ª 2007 The Authors Journal compilation ª RCOG 2007 BJOG An International Journal of Obstetrics and Gynaecology DOI: 10.1111/j.1471-0528.2006.01177.x www.blackwellpublishing.com/bjog Epidemiology