Objective
We previously reported associations between preeclampsia and the occurrence of metabolic and respiratory diseases in the offspring. In this article we examine whether the associations were due to preeclampsia or factors leading to preeclampsia.
Study Design
From 1978 through 2004, we identified 22,264 discordant sib-pairs in Denmark according to prenatal exposure to preeclampsia. Exposed children were compared to their unexposed siblings by using stratified Cox regression to estimate hazard ratios and 95% confidence intervals for metabolic and respiratory diseases.
Results
Exposed children had rather similar risks for metabolic disorders compared to their unexposed siblings. However, when the second child within each sib-pair was exposed, this child had an increased risk for respiratory diseases (hazard ratio, 1.55; 95% confidence interval, 1.20–2.01).
Conclusion
Factors leading to preeclampsia may shape susceptibility to metabolic or respiratory diseases in the offspring but a programming effect on respiratory diseases induced by preeclampsia cannot be ruled out.
Preeclampsia is a disease characterized by pregnancy-induced hypertension and proteinuria that affects 2-8% of all pregnancies. It modifies the intrauterine environment by influencing the pattern of hormonal signals and substrate transported to the fetus. The fetus has to adapt to this potential unfavorable intrauterine environment and this adaption may increase susceptibility to disease occurring later in life. We have previously reported that prenatal exposure to preeclampsia was associated with an increased risk of metabolic disorders (incidence rate ratio, 1.6; 95% confidence interval [CI], 1.5–1.7) and respiratory disorders (incidence rate ratio, 1.2; 95% CI, 1.1–1.2) among children born at term. These associations may be due to intrauterine programming induced by preeclampsia or risk factors shared by preeclampsia and the 2 outcomes but other explanations are also possible.
If intrauterine programming induced by preeclampsia explained our previous results, one would expect that children exposed to preeclampsia have an increased risk of metabolic and respiratory diseases compared to their unexposed full siblings. If the associations are due to confounding by genes or time-stable environmental factors shared by the family, one would expect that exposed and unexposed full siblings have similar risks since both siblings would to some extent be exposed to similar time-stable factors. Therefore, we compared the risk of being hospitalized for metabolic and respiratory diseases in children exposed to preeclampsia to this risk in their unexposed siblings.
Materials and Methods
We conducted a population-based sibling cohort study by using the Danish Civil Registration System, the Danish National Hospital Register, and Danish Medical Birth Registry. The Danish Civil Registration System was established in 1968 and holds data on vital status and family structure, which also permits long-term follow-up. We identified 538,716 sib-pairs (1,077,432 singleton-born children) from Jan. 1, 1978, through Dec. 31, 2004. We classified children as full siblings if they had the same mother and father in the records and as half-siblings if only the mother was the same.
Information on preeclampsia and hospitalizations in children were obtained from the Danish National Hospital Register that holds nationwide data on all admissions to any Danish hospital since 1977 and on all outpatient visits since 1995. The information on diagnoses was based on the Danish version of the International Classification of Diseases, Eighth Revision ( ICD-8 ) from 1977 through 1993, and the International Statistical Classification of Diseases, 10th Revision ( ICD-10 ) from 1994 onward. Preeclampsia was defined as being hospitalized due to relevant ICD-8 codes (637.03, 637.04, 637.09) and ICD-10 codes (O14.0, O14.1, O14.9). Eclampsia was defined according to ICD-8 (637.1) and ICD-10 (O15). Exposed children were born to pregnancies complicated by preeclampsia while unexposed children were born to pregnancies with no recorded diagnosis of preeclampsia. We identified metabolic disorders by relevant ICD-8 codes (240-279) and ICD-10 codes (E00-E90), and respiratory diseases by relevant ICD-8 codes (460-519) and ICD-10 codes (J00-J99). The 2 conditions were identified from the date of the first admission to hospital or first contact with the outpatient clinic. These diseases were selected because our previous study found associations between preeclampsia and the risk of these diseases in the offspring.
Information on gestational age, maternal age at birth, parity, and birthweight was obtained from the Danish Medical Birth Registry, which includes all births in Denmark since 1973. Gestational age recorded in the registry was, in the past, mainly based on the date of the last menstrual period, but has in recent years increasingly been replaced with ultrasound estimates. We coded children as small for gestational age (SGA) if the birthweight was in the lowest 10th percentile of the gestational week- and sex-specific distribution of birth weight for the study period. Interbirth interval was calculated as the time (in years) between 2 consecutive birth dates.
In Denmark, institutional review board approval is not needed for register studies, but the study was approved by the Danish Data Protection Agency.
We excluded concordant siblings according to prenatal exposure to preeclampsia (both siblings were exposed or both siblings were unexposed to preeclampsia) and children with missing values on important covariates. Furthermore, we only kept sib-pairs that consisted of firstborn and second-born children ( Figure ). The final study population consisted of 22,264 discordant sib-pairs (44,528 singleton-born children) according to prenatal exposure to preeclampsia (1 was exposed and 1 was unexposed within each sib-pair).
