Background
Fetal environment has a substantial influence on an individual’s health throughout their life course. Animal models of hypertensive disease of pregnancy have demonstrated adverse health outcomes among offspring exposed to hypertensive disease of pregnancy in utero. Although there are numerous descriptions of the neonatal, infant, and pediatric outcomes of human offspring affected by hypertensive disease of pregnancy, there are few data in US populations on later life outcomes, including mortality.
Objective
To assess risk for early mortality among offspring of pregnancies complicated by hypertensive disease of pregnancy.
Study Design
This is a retrospective cohort study of offspring born to women with singleton or twin pregnancies between 1947 and 1967 with birth certificate information in the Utah Population Database. We identified offspring from delivery diagnoses of gestational hypertension, preeclampsia, or eclampsia. Offspring from these pregnancies (exposed) were matched to offspring of pregnancies without hypertensive disease of pregnancy (unexposed) by maternal age at delivery, birth year, sex, and multiple gestation. We also identified unexposed siblings of exposed offspring for a separate sibling analysis. Mortality follow-up of all offspring continued through 2016, at which time they would have been 49–69 years old. Adjusted hazard ratios for cause-specific mortality comparing exposed with unexposed offspring were estimated using Cox proportional hazard models.
Results
We compared mortality risks for 4050 exposed offspring and 6989 matched unexposed offspring from the general population and 7496 unexposed siblings. Mortality risks due to metabolic, respiratory, digestive, nervous, and external causes of death did not differ between exposed and unexposed groups. Mortality risks from cardiovascular disease were greater in exposed offspring compared with unexposed offspring (adjusted hazard ratio, 1.57; 95% confidence interval, 1.16–2.12). In sex-specific models among the general population, cardiovascular disease mortality was significantly associated with exposure among male patients (adjusted hazard ratio, 1.92; 95% confidence interval, 1.27–2.88) but not among female patients (adjusted hazard ratio, 0.97; 95% confidence interval, 0.81–1.94). An interaction between hypertensive disease of pregnancy exposure and birth order on cardiovascular disease mortality was significant ( P =.047), suggesting that the effect of hypertensive disease of pregnancy on cardiovascular disease mortality increased with higher birth order. Among siblings, the association between hypertensive disease of pregnancy exposure and cardiovascular disease mortality was not significant (adjusted hazard ratio, 1.39; 95% confidence interval, 0.99–1.95), and this was also true for sex-specific analyses of males (adjusted hazard ratio, 1.26; 95% confidence interval, 0.81–1.94) and females (adjusted hazard ratio, 1.71; 95% confidence interval, 0.96–3.04). As in the general population, there was a significant interaction between hypertensive disease of pregnancy exposure and birth order on cardiovascular disease mortality ( P =.011).
Conclusion
In a US population, overall mortality risks are greater for offspring of pregnancies complicated by hypertensive disease of pregnancy compared with unexposed offspring. Among siblings, there was not a significant association between hypertensive disease of pregnancy exposure and cardiovascular disease mortality.
A growing body of evidence suggests that the fetal environment has a substantial influence on an individual’s health throughout their life course. The theory of fetal programming indicates that alterations in the prenatal nutritional, hormonal, or environmental milieu can alter fetal gene expression, leading to permanent effects on a range of physiological functions. This hypothesis was proposed by Barker, who demonstrated an association between low birthweight and cardiovascular disease later in life. Animal models of hypertensive disease of pregnancy (HDP) have demonstrated increases in fetal hypoxic–ischemic injury and subsequent abnormalities in postnatal growth, apoptosis rates, and neuronal migration in affected fetuses. These findings suggest that HDP, either alone or in combination with growth restriction, may induce changes leading to the susceptibility of offspring to chronic diseases later in life.
Why was this study conducted?
To assess mortality risk among offspring of pregnancies complicated by hypertensive disease of pregnancy.
Key findings
In a US population, male offspring of pregnancies complicated by hypertensive disease of pregnancy experience excess mortality risks, particularly from cardiovascular disease, when compared with offspring of pregnancies not complicated by hypertensive disease of pregnancy.
What does this add to what is known?
This study confirms that adults born from pregnancies complicated by hypertensive disease of pregnancy are not only at risk for cardiovascular disease, but also, their risk of mortality from it is significantly greater. These individuals could be targeted by earlier screening for cardiovascular disease and may benefit from interventions to improve long-term health outcomes.
HDP complicates 5%–10% of pregnancies in the United States. Multiple studies have assessed the association between maternal HDP and adverse neonatal, infant, and adult outcomes. However, these studies have limited numbers of participants, and there are few data on the US population on later life–course outcomes. In addition, the vast majority of these studies, including the ones that were able to follow offspring into adulthood, have focused on morbidity outcomes such as cardiovascular disease (CVD). Data regarding mortality outcomes among affected offspring are lacking. To address this knowledge gap, we examined the risk of all-cause and cause-specific mortality in offspring born from pregnancies complicated by HDP in relation both to population-based matched controls and to their unaffected siblings.
Materials and Methods
In this retrospective historical cohort study, we identified all offspring of women with singleton or twin pregnancies with birth certificate information in the Utah Population Database (UPDB). The overall design, methods, and development of the UPDB have been reported previously. The UPDB is a comprehensive research database of linked records, which includes individual genealogical records from the Genealogical Society of Utah, official statewide birth and death certificates, and hospital discharge and ambulatory surgery records from the Utah Department of Health. It includes data on more than 11 million individuals. The study was approved by the Resource for Genetic and Epidemiologic Research, which oversees UPDB access, and the University of Utah Institutional Review Board.
The population represented in UPDB is predominantly of northern and western European Caucasian ancestry, reflecting the European-based pioneer population settling in Utah in the mid-1800s. The subjects under consideration were born as early as 1947, the first year that birthweight was recorded on Utah birth certificates. Using the text documented by the delivering provider on these birth certificate records, we identified pregnancies that were complicated by gestational hypertension, preeclampsia, eclampsia, and suspected HDP. We included the latter when the records indicated the presence of hypertension and edema or pregnancy-induced hypertension, but no definitive diagnosis noted on the birth certificate. Of note, HDP diagnoses and classifications were based solely on what was reported by the delivering provider on the birth certificate. These birth certificate data identified offspring of pregnancies with and without HDP for inclusion in our analyses. For all analyses, the date of death was available in UPDB based on genealogies, Utah death certificates, and the Social Security Death Index. The Social Security Death Index is a national database and captures the death dates of persons in the Social Security Administration, including those who moved out of Utah. The causes of death were determined using Utah death certificates and the International Classification of Diseases (revisions spanning ICD-7 to ICD-10).
Exposed offspring were identified as children delivered from pregnancies affected by HDP. Unexposed offspring were identified as children delivered to mothers with no history of HDP in any pregnancy. Exposed offspring were matched to unexposed offspring from the general population in 1:2 ratio by birth year, sex, maternal age at birth, and multiple gestation. We did not match by parity to avoid conditioning on the future. Offspring were excluded if they had maternal age at birth younger than age 12 or after age 50 years, missing paternal age at birth or paternal age at birth younger than age 15 years, missing gestation age or gestation age less than 20 weeks or greater than 44 weeks, missing birthweight or birthweight less than 500 g or greater than 6000 g. We also excluded offspring if they died before their first birthday. We employed this restriction to focus our attention on postinfancy outcomes. This study only included offspring born between 1947 and 1967 to allow sufficient follow-up time to observe mortality events.
We performed 2 analyses. First, we compared mortality risks of exposed offspring with matching unexposed offspring from the general population (general population analysis). Exposed offspring with no matching unexposed offspring were excluded. Analyses were adjusted for birth order (the order of the child’s birth within the mother’s reproductive history), maternal parity, stratified by case ID to accommodate the matching design, and clustered by mother ID to account for natural clustering among siblings. We did not adjust for low birthweight and gestational age because exposed offspring are at greater risk of lower birthweights and shorter gestation age. If included in the models, we would have variables that would correlate with each other in the model. Second, we compared mortality risks of exposed offspring with their unexposed siblings to control for unmeasured genetic and environmental factors (sibling analysis). Exposed offspring with no unexposed siblings were excluded. Because the oldest individuals in our cohort were 69 years old at the end of our study period, we did not anticipate being able to capture all differences in mortality; rather, we intended to determine whether in utero HDP exposure is associated with early mortality.
Demographic characteristics of exposed and unexposed offspring were compared using t tests for continuous variables and χ 2 tests for categorical variables. Conditional Cox proportional hazard models were used for all-cause mortality analysis, and standard competing risk models were used for specific causes of death, including circulatory system disease, respiratory system disease, cancer, nervous system/mental disorders, endocrine/nutritional/metabolic diseases, digestive system disease, and external causes. Analyses were adjusted for sex, maternal age at birth, multiple gestation, and clustered by mother ID. Because HDP is associated with birth order (maternal parity), we tested for interaction between HDP, birth order, and mortality. R version 3.5.0 (R Foundation for Statistical Computing, Vienna, Austria) was used to perform all analyses. P values <.05 from 2-sided tests were considered statistically significant.
Results
General population analysis
We identified 5929 exposed offspring meeting our inclusion criteria within the UPDB. After we excluded those with missing data (1321), those who had died within the first year of life (366), and those who did not have at least one matching unexposed offspring (192), we were left with 4050 exposed offspring. We matched these to 6989 unexposed offspring ( Figure 1 ). Exposed offspring were born from mothers with the following categories of hypertensive disease: 3544 preeclampsia, 281 gestational hypertension, 204 eclampsia, and 21 with suspected HDP. Descriptive statistics are shown in Table 1 . Compared with matching unexposed offspring from the general population, exposed offspring had significantly lower birthweights, gestation age, preterm, birth order, and number of siblings.
| Exposed (N = 4050) | Unexposed (N = 6989) | P value | |
|---|---|---|---|
| Sex | .944 | ||
| Female | 1973 (48.7%) | 3411 (48.8%) | |
| Male | 2077 (51.3%) | 3578 (51.2%) | |
| Birthweight, g | 3078.1 ± 640.0 | 3276.0 ± 534.9 | <.001 a |
| Small for gestational age | .774 | ||
| No | 3671 (90.6%) | 6322 (90.5%) | |
| Yes | 379 (9.4%) | 667 (9.5%) | |
| Gestational age, wk | 39.0 ± 2.0 | 39.5 ± 1.6 | <.001 a |
| Preterm | <.001 a | ||
| No | 3586 (88.5%) | 6563 (93.9%) | |
| Yes | 464 (11.5%) | 426 (6.1%) | |
| Birth year | 1957.6 ± 5.5 | 1957.6 ± 5.5 | .868 |
| Birth order (continuous) | 2.8 ± 2.1 | 3.1 ± 2.1 | <.001 a |
| Birth order (categorical) | <.001 a | ||
| 1 | 1560 (38.7%) | 1694 (24.4%) | |
| 2 | 683 (17.0%) | 1567 (22.6%) | |
| 3 | 570 (14.2%) | 1269 (18.3%) | |
| 4 | 497 (12.3%) | 978 (14.1%) | |
| 5 | 317 (7.9%) | 613 (8.8%) | |
| 6 | 163 (4.0%) | 367 (5.3%) | |
| 7 | 108 (2.7%) | 189 (2.7%) | |
| 8 | 46 (1.1%) | 124 (1.8%) | |
| 9 | 34 (0.8%) | 61 (0.9%) | |
| 10 | 26 (0.6%) | 30 (0.4%) | |
| 11 | 9 (0.2%) | 12 (0.2%) | |
| 12 | 5 (0.1%) | 17 (0.2%) | |
| 13 | 5 (0.1%) | 12 (0.2%) | |
| 14 | 2 (0.0%) | 5 (0.1%) | |
| 15 | 2 (0.0%) | 2 (0.0%) | |
| 16 | 0 (0.0%) | 2 (0.0%) | |
| Number of siblings | 4.5 ± 2.2 | 4.9 ± 2.2 | <.001 a |
| Multiplicity | .600 | ||
| No | 3734 (92.9%) | 6457 (93.2%) | |
| Yes | 284 (7.1%) | 470 (6.8%) | |
| Mothers’ age at birth | 26.7 ± 7.1 | 26.9 ± 7.1 | .306 |
| Mothers’ HDP status | <.001 a | ||
| None | 0 (0.0%) | 6989 (100.0%) | |
| Eclampsia | 204 (5.0%) | 0 (0.0%) | |
| Gestational hypertension | 281 (6.9%) | 0 (0.0%) | |
| Pre-eclampsia | 3544 (87.5%) | 0 (0.0%) | |
| Questionable HDP | 21 (0.5%) | 0 (0.0%) |
Exposed offspring had significantly greater all-cause mortality risk than unexposed offspring (10.6% vs 8.7%; adjusted hazard ratio [aHR], 1.19; 95% confidence interval [CI], 1.07–1.32, P =.002). Table 2 describes the counts and mortality risks by most common causes of death among exposed and matched unexposed offspring in the general population. The most common causes of death in this population were external causes (trauma, suicide), followed by neoplasms and CVD.
| Cause of deaths | Exposed, N (%) | Unexposed, N (%) | aHR* (95% CI) | P value |
|---|---|---|---|---|
| Total deaths | 430 (10.6) | 609 (8.7) | 1.19 (1.07–1.32) a | .002 a |
| Neoplasms | 70 (1.7) | 94 (1.3) | 1.37 (1.04–1.81) a | .026 a |
| Metabolic/nutrition | 18 (0.4) | 28 (0.4) | 1.14 (0.62–2.09) | .678 |
| Nervous | 11 (0.3) | 27 (0.4) | 0.62 (0.33–1.16) | .133 |
| Cardiovascular | 63 (1.6) | 68 (1.0) | 1.57 (1.16–2.12) a | .004 a |
| Respiratory | 21 (0.5) | 29 (0.4) | 1.37 (0.82 – 2.27) | .228 |
| Digestive | 18 (0.4) | 25 (0.4) | 1.32 (0.74 – 2.35) | .342 |
| External causes | 120 (3.0) | 198 (2.8) | 0.97 (0.80 – 1.18) | .755 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
