Blood Pressure in Young Adults Born at Very Low Birth WeightNovelty and Significance
Adults Born Preterm International Collaboration
Adults born preterm at very low birth weight (VLBW; <1500 g) have higher blood pressure than those born at term. It is not known whether all VLBW adults are at risk or whether higher blood pressure could be attributed to some of the specific conditions underlying or accompanying preterm birth. To identify possible risk or protective factors, we combined individual-level data from 9 cohorts that measured blood pressure in young adults born at VLBW or with a more stringent birth weight criterion. In the absence of major heterogeneity, we performed linear regression analysis in our pooled sample of 1571 adults born at VLBW and 777 controls. Adults born at VLBW had 3.4 mm Hg (95% confidence interval, 2.2–4.6) higher systolic and 2.1 mm Hg (95% confidence interval, 1.3–3.0) higher diastolic pressure, with adjustment for age, sex, and cohort. The difference in systolic pressure was present in men (1.8 mm Hg; 95% confidence interval, 0.1–3.5) but was stronger in women (4.7 mm Hg; 95% confidence interval, 3.2–6.3). Among the VLBW group, blood pressure was unrelated to gestational age, maternal smoking, multiple pregnancy, retinopathy of prematurity, or bronchopulmonary dysplasia. Blood pressure was higher than that of controls among VLBW adults unexposed to maternal preeclampsia. Among those exposed, it was even higher, especially if born appropriate for gestational age. In conclusion, although female sex and maternal preeclampsia are additional risk factors, the risk of higher blood pressure is not limited to any etiologic subgroup of VLBW adults, arguing for vigilance in early detection of high blood pressure in all these individuals.
Adults born very preterm (VP; <32 weeks of gestation) or at very low birth weight (VLBW; <1500 g) have higher blood pressure than their peers born at term.1–10 A recent meta-analysis showed that VLBW or VP adults have on average 3.3 mm Hg higher systolic pressure than controls.7 Another meta-analysis, including studies in adults born at any degree of prematurity, concluded that the mean difference between adults born preterm and controls was 4.2 mm Hg for systolic and 2.6 mm Hg for diastolic blood pressure.8 These differences were more pronounced among women (systolic/diastolic 4.9/2.9 mm Hg) but clearly present among men as well (2.0/1.3 mm Hg).8 These differences are considerable given that, at the population level, a 2 mm Hg reduction in diastolic pressure is estimated to result in a 7% to 14% reduction in mortality from ischemic heart disease and 9% to 19% from stroke with greatest reductions in the youngest age groups.11
Although these meta-analyses have been important in confirming the association between very preterm birth and adult blood pressure, they have, apart from sex, not been able to assess any other risk factors or protective factors for high blood pressure among adults born very preterm. This would be a crucial step in identifying underlying mechanisms, which then could serve as targets for prevention.
The higher blood pressure among adults born VLBW/VP could arise from dissimilar conditions that lead to preterm birth or from different postnatal conditions and complications of prematurity or from variation in adult lifestyle behaviors or characteristics such as body size. A substantial proportion of VP or VLBW infants have experienced impaired fetal growth, which is a frequent cause of medically indicated preterm delivery. It is often accompanied by maternal preeclampsia,12,13 a hypertensive condition that may also share genetic susceptibility with high blood pressure.14 Counterintuitively, previous single-center studies have found similarly increased blood pressures in adults born VP or VLBW regardless of whether they have been born appropriate for gestational age (AGA) or small for gestational age (SGA)1,3,10 or whether they have been exposed to maternal preeclampsia.3,10,15 Notably, most individual studies remain inconclusive because any subgroup analyses that are based on one cohort at a time have limited power.
Adults born very preterm are shorter and, at least in some studies,1 thinner with a lower lean body mass than those born at term. Despite the fact that body mass index (BMI) and height affect blood pressure, adjustment for adult body size has had little effect on the associations between very preterm birth and adult blood pressure.1,2,15–17 Dissimilar methods of size adjustments hamper the use of reported aggregated data in conventional meta-analysis and therefore have not been able to assess whether the associations differ according to the modifying effect of adult body size.
To overcome these limitations of aggregated-data meta-analysis and of individual studies, we performed an individual-participant meta-analysis in a combined data set of 9 cohorts including 1571 adults born at VLBW and 777 controls born at term or with a normal birth weight. Our main aim was to identify factors that either increase or decrease risk of high blood pressure among adults born with VLBW. We investigated whether their higher blood pressure, as compared with controls, is modified by fetal growth (reflected by birth-weight SD score), by length of gestation, by maternal smoking or preeclampsia, or by maternal gestational or essential hypertension, or by multiple pregnancy or neonatal conditions. These conditions include bronchopulmonary dysplasia (BPD) and retinopathy of prematurity, conditions that both reflect severity of neonatal morbidity. Our secondary aims were to increase the precision of previous estimates comparing blood pressure in VLBW women and men with that in controls and to assess whether differences in adult body size or socioeconomic status contribute to differences in blood pressure.
Inclusion of Studies
We made a systematic search in Medline for VP or VLBW and blood pressure and picked up 10 relevant research articles from the hundreds of titles and abstracts, and >100 full texts reviewed.18 In 2013, we reran this search, and we (P.H., E.K., and S.S.) initiated the current project by contacting all research groups that were, through personal contacts (cohort 9, Belfast) or based on the search (other cohorts), known to have followed up a cohort of adults born VP or at VLBW and that had publications on or were otherwise known to have blood pressure measured at a postpubertal age on >40 VLBW subjects by May 2013. All identified cohorts that also followed a control group were defined by a weight limit and none by a gestational age limit. We also included studies that had a more stringent inclusion criterion for birth weight (<1250, <1000, or <800 g) or for presence of BPD, in our primary analyses. In secondary analyses, we only included the pure VLBW cohorts (and not those with more stringent inclusion criteria). For convenience, we refer to our subjects of interest as VLBW throughout the article. The original studies followed up control subjects that were born at term or at a normal birth weight. Regarding our main question, identifying risk and protective factors that contribute to blood pressure within VLBW subjects, we also included the VLBW subjects of one study with detailed data, despite its lack of a control group: The Dutch Project on Preterm and Small for Gestational Age Infants (POPS). All 9 research groups that were contacted agreed to participate.
Inclusion of Subjects
Exclusion criteria of each of the 9 cohort studies, if any, are shown in Table S1 in the online-only Data Supplement. For the current analysis, we included subjects born preterm (36 weeks 6 days or less) at VLBW (birth weight ≤1500 g), see Table S2. The original comparison groups served as controls.
Blood pressure measurement techniques are described in Table S3. Our main outcomes were office systolic and diastolic blood pressure as continuous variables. As dichotomous outcomes, we also used hypertension, defined as systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg or known usage of antihypertensive medication. For prehypertension, the thresholds were 130 and 85 mm Hg.19
As a measure of socioeconomic status, we used maternal educational attainment, and as a proxy of fetal growth, we used birth-weight SD score (in relation to sex and length of gestation, based on current US criteria.20). We defined SGA as birth-weight SD score <−2.0. Other predictors included maternal preeclampsia, smoking during pregnancy, retinopathy of prematurity, and BPD (for details, see online-only Data Supplement).
We performed all analyses with IBM SPSS 22.0 software. We used multiple linear regression analysis for blood pressure as a continuous outcome and logistic regression for (pre)hypertension as a dichotomous outcome. We chose to use 0.05 as threshold for significance. All models were adjusted for cohort (dummy coded) and current age; they were also either adjusted for or stratified by sex. Age was missing for 31 subjects in the Cleveland cohort. For these subjects, we used the mean age of the respective cohort. Availability of data determined which cohorts could be included in each model. Any additional variable remained in further models if it made at least a 10% difference in the main predictor’s effect size. Interactions were assessed in separate models including a product term together with the main effect terms in the model. For interactions, we considered P<0.01 as statistically significant. Heterogeneity between the study cohorts was assessed as cohort×VLBW interactions. When presenting data for subgroups (such as VLBW subjects born SGA or AGA), we first present comparisons of both groups separately with controls. Then, we present comparisons between these subgroups, both with and without cohort 5 (POPS), for consistency.
Availability of Original Data
Reanalyzing the data was consistent with ethics committee approvals and participants’ consents acquired for the original studies. The data used are the property of the participating institutions, and request for data should be directed to the contact persons of each institution (see Table S2). Such requests may be subject to Ethics Review and individual consent.
Characteristics of the 9 cohorts are shown in Tables S1 and S2 and blood pressures in each cohort in Figure S1. In the total sample, adjusted for sex, age, and cohort, VLBW subjects had 3.4 mm Hg (95% confidence interval (CI), 2.2–4.6) higher systolic and 2.1 mm Hg (95% CI, 1.3–3.0) higher diastolic pressure than controls (Table 1). The means of blood pressures within cohorts were typically higher for the VLBW subjects (Figure S1). As several of the P values for the cohort×VLBW×sex interaction analyses were <0.01 regarding systolic pressure as the dependent variable, we fitted separate models for men and women. The VLBW effects on systolic pressure in each of the cohorts were similar, with VLBW×cohort interaction P values ≥0.03 for men and ≥0.12 for women. For between-cohort differences on diastolic pressure among men, the only significant interaction P value was 0.004 (lower diastolic pressure in VLBW men in cohort 6, Vancouver, see means in Figure S1); among women, the lowest such P value was 0.12. Despite this, we retained these 17 men from cohort 6 in the main analyses because we considered their (negative) influence on the VLBW effect estimate would be negligible.
Comparison of Blood Pressure in Young Adults With VLBW and Term-Born Controls
Both VLBW women and VLBW men had higher systolic and higher diastolic blood pressures than did their peers born at term (Table 1). The difference was larger among women. Further shown in Table 1 is the division of the VLBW group to those born with extremely low bith weight (ELBW, birth weight ≤1000 g) and those with birth weight 1001–1500 g). Both groups had higher blood pressure than controls.
Adjustment for Maternal Education
Within cohorts with adequate education data (all except 7 and 9), the difference in systolic pressure between VLBW and control groups was smaller than in the total sample (Table 2). Adjustment for low maternal education did not alter the VLBW–control differences. Therefore, in further VLBW–control comparisons, we omitted maternal education data and included all cohorts.
Adjustment for Current Height and BMI
As compared with controls, the VLBW subjects were shorter, and VLBW women had lower BMI (Table S2). Table 3 shows how the difference in systolic blood pressure between VLBW and controls among men was 4.2 mm Hg (95% CI, 2.4–6.0; increased by 1.9 mm Hg) when current height and BMI were introduced to the models. The other differences increased a little less.
The effect of VLBW on hypertension was different among men than among women, with a P value for interaction of 0.05. Hypertension was present in 13% of VLBW men and 11% of control men. For women, these percentages were 7.1% and 3.3%. Odds ratios (ORs) for having hypertension were 1.1 (95% CI, 0.7–1.7) within men and 2.4 (95% CI, 1.3–4.6) within women. The pattern was the same for prehypertension.19
Small for Gestational Age
Of the 1571 subjects born at VLBW, 223 (14.2%) were born SGA. We first compared the VLBW SGA and VLBW AGA subjects separately with controls (Table 4). Both SGA VLBW and AGA VLBW subjects had higher systolic and diastolic blood pressures than controls, and these differences were similar. Accordingly, any differences between SGA and AGA VLBW subjects were not statistically significant. In sex-specific analyses, VLBW AGA men had on average higher systolic and diastolic blood pressure than control men. Mean blood pressures in VLBW SGA men were, however, similar to those in control men and lower than those in VLBW AGA men (P values for interaction sex×SGA among the VLBW group were 0.41 for systolic and 0.15 for diastolic pressure, Table S4).
Birth-Weight SD Score and Length of Gestation as Continuous Variables
We also assessed the linear effects of birth-weight SD score and gestational age on blood pressure. Because the inclusion criterion of all studies was based on birth weight, VLBW subjects born at later gestational ages were, by design, included in the original follow-up studies only if they had sufficiently low, birth-weight SD scores. This produces an artificial negative correlation between gestational age and birth-weight SD score. Therefore, we at first analyzed the linear effects of length of gestation within each stratum of birth-weight SD scores and then the effects of birth-weight SD score within strata of gestational age (Tables 5 and 6; Tables S5 and S6). The cohort-, sex-, and age-adjusted blood pressure differences from controls were similar. Accordingly, among the VLBW subjects, there was no association between birth-weight SD score and systolic/diastolic pressures in any of the length of gestation strata and no association between length of gestation and systolic/diastolic pressures in any of the birth-weight SD score categories/strata, with the exception of those with a birth-weight SD score over +1: among them, those with a lower length of gestation had higher systolic blood pressure (3.0 mm Hg per week). These associations within the strata were similar among men and women (Table S5). P values for interaction: sex×birth-weight SD score in gestational age strata, >0.08; sex×gestational age in birth-weight SD score strata, >0.10.
Maternal Preeclampsia and Hypertension
Data on exposure to preeclampsia among the VLBW subjects, unavailable only in 2 cohorts, showed that 13.9% were exposed to that condition (Table S2). Subjects born at VLBW who were not exposed to maternal preeclampsia had 4.0/2.0 mm Hg higher systolic/diastolic pressure than all controls (Table S7). For VLBW subjects exposed to maternal preeclampsia, this difference was 5.8/4.2 mm Hg.
Of the VLBW 928 subjects, 154 were exposed to maternal preeclampsia. Of those exposed, 30 (19.5%) were born SGA as compared with 45 of the 774 (5.8%) not exposed. Mean systolic pressure was highest in AGA infants exposed to maternal preeclampsia and lowest in SGA infants not exposed (Table S7). In all but the SGA nonpreeclampsia group, blood pressure was higher than that in controls (Table S7). This pattern was driven by group differences within women.
When SGA and preeclampsia were entered in simultaneous regression models within the VLBW subjects, preeclampsia was associated with 2.3 mm Hg (95% CI, −0.1 to 4.5) higher systolic and 1.8 mm Hg (95% CI, 0.5–3.1) higher diastolic blood pressure, whereas SGA status was associated with 2.7 mm Hg (95% CI, −0.4 to 5.7) lower systolic and 0.3 mm Hg (95% CI, −1.9 to 2.4) lower diastolic pressure; there was no interaction between maternal preeclampsia and SGA (P values ≥0.57). When cohort 5 (POPS) was included, being born SGA was associated with 2.2 mm Hg (95% CI, 0.3–4.1) lower systolic pressure, and preeclampsia was associated with 1.5 mm Hg (95% CI, −0.3 to 3.4) higher systolic and 1.8 mm Hg (95% CI, 0.5–3.1) higher diastolic pressure.
We analyzed the potential role of gestational hypertension in cohorts 1, 8, and 9. The respective numbers of VLBW subjects whose mother had this condition were 9, 13, and 2. In a reanalysis, excluding these subjects made no difference to the VLBW effect on systolic and diastolic pressures in men or women.
Data on maternal current hypertension, as assessed at offspring’s young adulthood, was available for cohorts 1, 2, and 8. Among this subgroup of 462 men and 555 women, VLBW effects on systolic or diastolic pressures, without having maternal hypertension in the model, were slightly less than that in all cohorts pooled (please compare Tables 1 and S8). Adding maternal hypertension to the model left the VLBW effects unchanged. There was no effect of maternal hypertension within men, but within women, systolic/diastolic pressures were higher by 4.2 mm Hg (95% CI, 1.3–7.1) or 2.9 mm Hg (95% CI, 0.6–5.2) when adjusted for age, cohort, and VLBW status.
Retinopathy of Prematurity
Data on retinopathy of prematurity were available for cohorts 1, 3, 7, and 8. Retinopathy of at least stage 3 was seen from 3.2% to 14.7% of the VLBW/ELBW subjects. A history of at least stage-3 retinopathy was not associated with current blood pressure. Neither was retinopathy of any degree associated with systolic or diastolic pressures.
When VLBW subjects from multiple pregnancies were compared with VLBW singletons, their systolic and diastolic pressures were similar, with mean differences of −0.2 mm Hg (95% CI, −1.4 to 1.8) and −0.2 mm Hg (95% CI, −1.3 to 1.0).
Smoking During Pregnancy
In cohorts 1, 3, 5, 8, and 9, data on maternal smoking during pregnancy were available for 766 VLBW subjects and missing for 254. Among all these VLBW subjects, maternal smoking was not associated with blood pressure (mean differences −1.9 mm Hg [−3.8 to 0.1 mm Hg] for systolic and −0.6 mm Hg [−2.0 to 0.7 mm Hg] for diastolic pressure). Stratification by gestational age showed no effect of smoking during pregnancy on blood pressures in the 2 lowest gestational age categories (<28 or 28–31 completed weeks). For those 178 born at ≥32 completed weeks, it was associated with 4.5 mm Hg (95% CI, 0.8–8.3) lower systolic pressure. However, the P value for maternal smoking during pregnancy×gestational age category was 0.09.
The frequency of BPD differed between cohorts (Table S2). As compared with the controls, VLBW subjects either with or without BPD had similarly higher systolic pressure, difference with controls being 3.0 mm Hg (95% CI, 1.2–4.8) and 4.4 mm Hg (95% CI, 2.9–5.9), respectively. For diastolic pressure, findings were similar.
Our individual-participant meta-analysis has one main finding. Unlike existing reports, we were able to evaluate the long-term effects of clinically relevant subgroups of VLBW adults. With our large pooled data providing adequate power and precision, we show that the differences between VLBW adults and controls were to a great extent similar regardless of fetal growth disturbance, or with multiple birth, maternal smoking, or postnatal complications within the VLBW group. The only perinatal characteristic associated with blood pressure was maternal preeclampsia; however, even VLBW adults not exposed to maternal preeclampsia had higher blood pressure than controls. As our secondary finding, we show that the blood pressure differences between VLBW adults and controls became more pronounced when adjustments were made for the smaller body size of VLBW adults.
Two previous meta-analyses have been published on blood pressure in adults born preterm.7,8 Both were based on aggregate data from published studies and focused on assessing publication bias and methodological quality of the original studies and confounding by socioeconomic position, with no attempt to study prenatal and postnatal risk or protective factors, which were a key focus of the present study. One of these meta-analyses included 994 VLBW or VP subjects from 10 studies and reported a difference of 2.5 mm Hg (95% CI, 1.7–3.3) for systolic blood pressure with 787 controls; diastolic blood pressure or sex-stratified results were not reported.7 The other meta-analysis included 1871 subjects from 15 studies plus 14 192 male conscripts from a large register study, all born at any degree of prematurity and not necessarily preterm. Differences in systolic blood pressure with controls were 4.9 mm Hg (95% CI, 3.3–6.6) for women and 2.0 mm Hg (95% CI, 0.5–3.5) for men. Although that meta-analysis included both studies comparing VLBW or VP adults with controls and population-based cohort studies, the differences were similar to those found by us between VLBW adults and controls: 4.7 mm Hg (95% CI, 3.2–6.3) among women and 1.8 mm Hg (95% CI, 0.1–3.5) among men. Moreover, our study is consistent with both meta-analyses by showing that adjustment for socioeconomic position had a negligible effect on the associations.
The novelty of the present study lies with the thorough investigation of how different reasons for prematurity and neonatal conditions predict blood pressure in young adults. This became possible with our detailed prenatal and neonatal data on individual participants. Although there was a relatively wide variation of length of gestation and fetal growth (birth-weight SD score) within the VLBW group, there was no convincing association between these variables and blood pressure. Systolic differences with controls were slightly less pronounced for the VLBW-SGA than the VLBW-AGA groups, but the difference between VLBW-SGA and VLBW-AGA groups did not attain statistical significance, and neither did associations of systolic or diastolic pressure with birth-weight SD score or length of gestation as continuous variables. The effects of gestational age at birth and fetal growth have been difficult to assess in previous studies, because most studies used a birth weight–based selection criterion. Consequently, subjects with more advanced gestational ages were included only if they were sufficiently growth restricted. The large number of subjects in our study allowed us to analyze the data in relatively narrow strata of length of gestation or birth-weight SD scores. Thus, we should have detected any meaningful association, if such existed.
Divided into subgroups above or below 1000 g, our VLBW adults showed similarly higher blood pressures (Table 1). Our findings among our VLBW subjects are at variance with the well-replicated linear associations between higher blood pressure and lower birth weight21,22 or shorter length of gestation23–25 among the general population. They suggest that the effects of VLBW birth per se override the effects of intrauterine growth restriction and degree of immaturity at birth on blood pressure. Our findings on postnatal adverse conditions support this idea. Although these conditions are dependent on the increasing severity of disease or degree of immaturity, we found no association of blood pressure with either retinopathy of prematurity or BPD. Both neonatal conditions serve as markers of specific vascular conditions: retinopathy is a sign of vascular injury in the eye associated with hyperoxia, whereas BPD may be associated with higher pulmonary vascular resistance and hypoxia. Although the lack of association with blood pressure argues against a major role of postnatal conditions, it should be interpreted with caution because our data do not capture the full picture of the postnatal clinical course. Postnatal clinical course itself has changed because of major developments in neonatal treatments during recent decades.
Maternal predisposition for hypertension could confound associations between pregnancy circumstances and later blood pressure. Our results suggest major confounding is unlikely, as the higher blood pressures among VLBW group survived adjustment for maternal hypertension. We, however, found that blood pressure was highest in VLBW adults whose mothers had preeclampsia. Previous studies in single VLBW cohorts1–3,5,15 have left the existence of such an association unclear although it is present in adults born at or closer to term.26 Interestingly, the high blood pressures among those exposed to maternal preeclampsia were particularly evident among adults born VLBW AGA rather than those born VLBW SGA. This finding suggests that common reasons underlying both preeclampsia and blood pressure could explain the association between the 2, rather than direct programming effects that typically would be associated with fetal growth restriction. Furthermore, although maternal preeclampsia is an additional predictor of high blood pressure, it is important to note that even VLBW subjects not exposed to maternal preeclampsia have higher blood pressure than controls.
VLBW adults are on average shorter and in several studies have slightly lower BMI than their term-born counterparts. However, it is likely that they still have unfavorable adiposity, similar to those born at term with SGA.27 We found that the difference in blood pressure between VLBW and control groups was strengthened when adjusted for height and BMI, which are associated with blood pressure.28 Although the appropriateness of adjustment variables on a potential causal pathway, such as height and BMI between VLBW birth and blood pressure, is debated within the epidemiological community,29 this adjustment is analogous to comparing VLBW individuals to term-born counterparts with a similar height and BMI.
Although no observational study can prove causality, our study can give valuable insights on the mechanisms linking preterm birth at VLBW with adult blood pressure. A proportion of this relationship may be explained by common factors underlying maternal hypertension in pregnancy and offspring blood pressure. The lack of association with length of gestation or fetal growth argues against a major role of other conditions occurring before preterm birth and points to the postnatal period that, for those born at term, roughly corresponds to the third trimester of pregnancy. This period is sensitive for at least the development of the kidney. A suboptimal number of nephrons formed during this period may determine an individual’s course toward hypertension.30
The main strengths of the study are the large number of VLBW adults and controls and objectively recorded perinatal and neonatal data. We actively sought for and found unpublished data as well (cohorts 8 and 9, McMaster and Belfast). As to limitations, 2 of the studies we included used more stringent inclusion criteria, giving more weight to adults born at lowest birth weights in the pooled analysis. However, restricting the comparison with controls to those born at ELBW did not alter the difference in blood pressure. Although recruitment of controls was based on hospital catchment area for most cohorts, some used other approaches. Participation rates varied between 54% and 94% of those invited. These factors may increase heterogeneity. Moreover, a proportion of the variables were not available for all cohorts, reducing power for these comparisons. Adjustment for socioeconomic position is challenging in international studies because its classification and relevance to health differ from one society to another. For this same reason, we did not find adjustment for race/ethnicity meaningful. Prenatal and neonatal care has improved, and our data are unable to reveal the possible effects of modern treatments on blood pressure. However, our findings are directly relevant to VLBW survivors born in the 1970s or 1980s, now in their 30s and 40s. These survivors currently amount to ≈500 000 people in the United States alone.31
Thanks to modern neonatal care and increased survival, individuals born preterm at VLBW comprise ≈1 to 1.5% of the age classes ≤30 or 40 years. They have higher blood pressure than their peers born at term. This difference is seen in both sexes, although it is stronger among women. Although VLBW adults exposed to maternal preeclampsia have on average the highest blood pressure levels, higher blood pressure is also present among VLBW adults not exposed. Blood pressure among VLBW adults is not associated with fetal growth, gestational age, socioeconomic status; in other words, it is not limited to any subgroup of VLBW adults. This highlights the importance of vigilance in early detection of high blood pressure in all children and adults born preterm at VLBW.
Participating Centers and Investigators
In addition to individuals listed on the title page, the following investigators serve as contributors of the article. (For details, please see online-only Data Supplement.) Please note that one of the original coauthors, Professor Maureen Hack from CWRU, Cleveland, Ohio, sadly deceased on June 4, 2015. Cohort 1, Helsinki: Anna-Liisa Järvenpää, Sonja Strang-Karlsson, Riikka Pyhälä-Neuvonen, Katri Räikkönen, and Johan G Eriksson. Cohort 2, Cleveland: Maureen Hack and Mark Schluchter. Cohort 3, Melbourne: Catherine Callanan. Cohort 4, Trondheim: Marit S. Indredavik and Jon Skranes. Cohort 5, POPS: Karin van der Pal-de Bruin, Martijn Finken, and Yvonne Schönbeck. Cohort 6, Vancouver: Michael Whitfield and Anne Synnes. Cohort 7, Indomethacin, and cohort 8, McMaster: no additional contributors listed. Cohort 9, Belfast: Aisling Gough, Steven Caskey, Michael D Shields, and Henry L Halliday.
Sources of Funding
Helsinki Study of Very Low Birth Weight Adults was supported by Academy of Finland, Emil Aaltonen Foundation, Finnish Medical Society Duodecim, Finska Läkaresällskapet, Finnish Foundation for Cardiovascular Research, the Finnish Foundation for Pediatric Research, Finnish Special Governmental Subsidy for Health Sciences, Jalmari and Rauha Ahokas Foundation, Juho Vainio Foundation, Novo Nordisk Foundation, Päivikki and Sakari Sohlberg Foundation, Signe and Ane Gyllenberg Foundation, Maud Kuistila Memorial Foundation, Sigrid Jusélius Foundation, and Yrjö Jahnsson Foundation. CWRU study was supported by the National Institute of Child Health and Human Development grant R01 HD 34177. Melbourne study was supported by Royal Women’s Hospital Research Foundation and VicHealth Trondheim study was supported by St. Olav’s University Hospital, and Norwegian University of Science and Technology, Trondheim, Norway. Part of that study population was recruited from a multicenter study sponsored by the US National Institute of Child Health and Human Development, NIH (NICHD contract No. 1-HD-4-2803 and No. 1-HD-1-3127). Aisling Gough from the Belfast study was awarded a Research Forum for the Child International PhD studentship from Queen’s University Belfast, and Steven Caskey from the same study received a fellowship grant from the Northern Ireland Chest, Heart and Stroke Association. Cohort recruitment was also supported by research award from The Friends of Jessica Trust. Vancouver study was supported by British Columbia Medical Services Foundation, the Michael Smith Foundation for Health Research, and the Child and Family Research Institute.
↵* Centers and investigators participating in the APIC Collaboration are listed in the Appendix.
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.116.08167/-/DC1.
- Received July 19, 2016.
- Revision received July 19, 2016.
- Accepted August 1, 2016.
- © 2016 American Heart Association, Inc.
- Doyle LW,
- Faber B,
- Callanan C,
- Morley R
- Keijzer-Veen MG,
- Finken MJ,
- Nauta J,
- Dekker FW,
- Hille ET,
- Frölich M,
- Wit JM,
- van der Heijden AJ
- de Jong F,
- Monuteaux MC,
- van Elburg RM,
- Gillman MW,
- Belfort MB
- Parkinson JR,
- Hyde MJ,
- Gale C,
- Santhakumaran S,
- Modi N
- Rotteveel J,
- van Weissenbruch MM,
- Twisk JW,
- Delemarre-Van de Waal HA
- Hovi P
- Pickering TG,
- Hall JE,
- Appel LJ,
- Falkner BE,
- Graves J,
- Hill MN,
- Jones DW,
- Kurtz T,
- Sheps SG,
- Roccella EJ
- Olsen IE,
- Groveman SA,
- Lawson ML,
- Clark RH,
- Zemel BS
- Gamborg M,
- Byberg L,
- Rasmussen F,
- et al
- Lawlor DA,
- Ebrahim S,
- Davey Smith G
- Järvelin MR,
- Sovio U,
- King V,
- Lauren L,
- Xu B,
- McCarthy MI,
- Hartikainen AL,
- Laitinen J,
- Zitting P,
- Rantakallio P,
- Elliott P
- Johansson S,
- Iliadou A,
- Bergvall N,
- Tuvemo T,
- Norman M,
- Cnattingius S
- Davis EF,
- Lazdam M,
- Lewandowski AJ,
- Worton SA,
- Kelly B,
- Kenworthy Y,
- Adwani S,
- Wilkinson AR,
- McCormick K,
- Sargent I,
- Redman C,
- Leeson P
- 28.↵Emerging Risk Factors Collaboration. Adult height and the risk of cause-specific death and vascular morbidity in 1 million people: individual participant meta-analysis. Int J Epidemiol. 2012;41:1419–1433.
- Tu YK,
- West R,
- Ellison GT,
- Gilthorpe MS
- Lee KS,
- Kim BI,
- Khoshnood B,
- Hsieh HL,
- Chen TJ,
- Herschel M,
- Mittendorf R
Novelty and Significance
What Is New?
The increased blood pressure in adults who were born at very low birth weight (<1500 g) is stronger among women and those whose mothers had preeclampsia (high blood pressure) during pregnancy.
Importantly, however, the high blood pressure is present in all subgroups of these adults, regardless of the reason for preterm birth or complications they may have had after preterm birth.
What Is Relevant?
Approximately 1 to 1.5% of people are born preterm at very low birth weight.
Because of their higher blood pressure, they may be at an increased risk of coronary heart disease or stroke later in life. Treatment of high blood pressure, when necessary, could reduce these risks.
Preventive measures and timely treatment need to be targeted to all VLBW subjects and not limited to any subgroup.
Our findings highlight the importance of vigilance in early detection of high blood pressure in all groups of children and adults born preterm at VLBW.