Weight Gain Before the Third Trimester and Risk of Hypertensive Disorders of Pregnancy: A Prospective Cohort Study

Background

Hypertensive disorders of pregnancy (HDP), including gestational hypertension, preeclampsia, and eclampsia, are high-risk factors for future cardiovascular events in women [1,2]. Studies have shown that women with HDP and their offspring have significantly higher risks of cardiovascular events in the future [3–5]; therefore, early diagnosis and prevention of HDP may improve their long-term health. Obesity has been reported to be a modifiable risk factor for HDP in several studies [6–8], but the association between excessive gestational weight gain and the risk of HDP remains uncertain or insufficient [9–11]; greater weight gain is related to a longer pregnancy duration and edema during pregnancy, mainly in the third trimester. Considering HDP as a disorder typically identified by new-onset hypertension in the second half of pregnancy, mainly in the third trimester, the relationship between weight gain before the third trimester and the incidence of HDP are uncertain, and there have been few studies on this topic in the Chinese population.

Conventional measures of gestational weight gain (GWG) can introduce bias into epidemiologic studies of GWG and adverse perinatal outcomes, such as premature birth, neonatal death, stillbirth, and preeclampsia, because both are correlated with gestational duration [10,12,13]. The gestational age-specific z-score of weight gain is a new measure to evaluate weight gain, which is independent of gestational duration. Compared with GWG, it correlates better with preterm birth [12]. Whether the gestational age-specific z-score of weight gain is suitable to assess the risk of HDP remains undetermined.

The main aim of the present study was to assess weight gain in 4 gestational periods 16 weeks (14+0, 17+6), 20 weeks (18+0, 21+6), 24 weeks (22+0, 25+6), and 28 weeks (26+0, 29+6) – and the risk of HDP in 4 early-pregnancy BMI categories – underweight (<18.5 kg/m2), normal-weight (18.5~23.9 kg/m2), overweight (24.0~27.9 kg/m2), and obesity (≥28.0 kg/m2)]. We also aimed to assess these relationships using the gestational age-specific z-score for weight gain.

Material and Methods

STUDY DESIGN:

This was a prospective cohort study of singleton-pregnant women in Tianjin, China, from 2016. This study was approved by the Medical Ethics and Human Clinical Trial Committee of the local hospital, and all participants agreed to participate in the study and signed a written informed consent form. Participants were recruited from 19 community hospitals if they met the following criteria:

Inclusion criteria: (I) Singleton pregnancy; (II) Gestational age at enrolment <13 completed weeks; and (III) No presence of chronic hypertension. Exclusion criteria: (I) Incomplete or abnormal enrolment information or follow-up information; (II) Pregnancy termination for various reasons before 24 weeks of pregnancy; and (III) Systolic blood pressure ≥140 mmHg (1 mmHg=0.133 kPa) and/or diastolic blood pressure ≥90 mmHg before 20 weeks of pregnancy.

The participants completed a pre-designed questionnaire to collect baseline characteristics, including age, ethnicity, education level, family history, pregnancy, and delivery times. Height and weight at enrolment were simultaneously recorded. Gestational age was calculated according to the last menstrual period and checked by B-ultrasonic examination in early pregnancy. Pregnant women were followed up once every 4 weeks up to 28~30 weeks of gestation, and the diagnoses were recorded by the medical record system of the hospitals for delivery.

OBSERVATION OUTCOMES:

The early pregnancy BMI was calculated based on maternal height and weight recorded at the first prenatal visit. Gestational weight gain was defined as the difference between the weight at the corresponding gestational age and the weight at the first prenatal visit. Patients with weight gain above or below 4 SD of each group at each gestational age were flagged and excluded according to their weight trajectory and clinical plausibility. If there were multiple weight measurements during some periods, the last one was analysed. The weight gain z-score was calculated using BMI-specific z-score charts for Chinese women [14]. The z-scores were analysed as continuous variables and categorical variables, referring to the method used by Bodnar et al. [11]; the weight gain z-score was modelled as a 3-level categorical variable (<−1 SD, −1 to +1 SD, >+1 SD), where <−1 SD was the reference category, which was assumed to indicate low HDP risk.

The diagnostic criteria for HDP were based on the Guidelines for the Diagnosis and Treatment of Hypertensive Disorders in Pregnancy (2015) [15]. Gestational hypertension was defined as a systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg after 20 weeks of pregnancy, which returned to normal within 12 weeks after delivery, and a negative urine protein test. Preeclampsia was defined as a systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg after 20 weeks of pregnancy accompanied by any of the following: urine protein ≥0.3 g/24 h, urine protein/creatinine ratio ≥0.3, random urine protein ≥(+) (examination method when urinary protein quantification could not be performed), or no proteinuria but heart, lung, liver, kidney or other important organ involvement or abnormal changes in the blood system, digestive system, nervous system, or the fetus. Eclampsia was defined as seizures that occurred during preeclampsia that could not be explained by other causes.

STATISTICAL ANALYSES:

Quartiles and percentage counts were used to describe the baseline characteristics of pregnant women. Non-parametric tests and chi-square tests were used to analyze the differences between groups. Relative risks (RRs) with 95% confidence intervals (CIs) were obtained using general linear models to analyse the relationship between weight gain or weight gain z-scores at the corresponding gestational week and the risk of HDP. Moreover, analyses were adjusted for the age at enrolment, early-pregnancy BMI, primipara, and family history of hypertension. Statistical analyses were performed using SPSS version 23.0 (IBM Corp, Armonk, NY, USA) and STATA version 14.0 (Stata Corp, College Station, TX, USA).

Results

A total of 6081 non-hypertensive singleton-pregnant women with a gestational age at enrolment of <13 completed weeks were enrolled. Among these women, 28 were eventually diagnosed with chronic hypertension or chronic hypertension complicated with preeclampsia; 166 women had terminated pregnancies for various reasons before 24 weeks; 355 women withdrew from follow-up or were lost to follow-up; 109 participants had incomplete enrolment or delivery information or illogical weight records; and 128 participants had ≤2 weight measurements in the 4 periods. Finally, 5295 pregnant women were included. The study flowchart is outlined in Figure 1.

The characteristics of the 5295 women who participated in the study are presented in Table 1. Among them, 512 (9.7%), 3231 (61.0%), 1156 (21.8%), and 396 (7.5%) women were underweight, normal-weight, overweight, and obese, respectively. There were 177 cases of preeclampsia, 3 cases of eclampsia, and 105 cases of gestational hypertension. The total incidence of HDP was 5.4%. The corresponding incidence rates of HDP in the underweight, normal-weight, overweight, and obese groups were 2.3%, 3.5%, 9.3%, and 13.1%, respectively. In this Chinese cohort, 96% were of Han ethnicity, which is the ethnic majority in China, and 69% were nulliparous. The mean maternal age at enrolment was 30 years. With the increase in early-pregnancy BMI, the chance of developing HDP gradually increased, and the mean weight gain during each gestational period gradually decreased. However, there was no significant difference in weight gain z-scores between the 4 BMI groups.

Tables 2 and 3 compare the characteristics of women with or without HDP overall and in each BMI group. Overall, compared to women who did not progress to HDP, women who eventually progressed to HDP had higher early-pregnancy BMIs, were more likely to be primipara, and had family histories of hypertension and caesarean section, but they had a lower gestational age at delivery and lower neonatal weight. There were no statistically significant differences in education level, gestational age at enrolment, or neonatal sex. In the underweight group, only weight gain and weight gain z-scores of women with HDP at approximately 28 weeks were higher than those of the control group, and there were no significant differences in the other parameters. In the normal-weight group, which accounted for 61% of the entire cohort, weight gain in women with HDP at approximately 24–28 weeks and weight gain z-scores at approximately 28 weeks were higher than those of the control group. In the overweight and obese groups, weight gain and weight gain z-scores of women with HDP at approximately 16 weeks were higher than those of normotensive women in the overweight group, while the opposite result was obtained in the obese group. There were no significant differences in weight gain and the weight gain z-scores in the other periods.

Crude and adjusted RRs for weight gain or weight gain z-scores before the third trimester and the risk of HDP according to early-pregnancy BMI category are shown in Table 4. In underweight women, weight gain at approximately 28 weeks was positively correlated with the risk of HDP (crude RR: 1.23, 95% CI: 1.02–1.47). After adjusting for confounding factors, such as age, early pregnancy BMI, primipara, and family history of hypertension, the adjusted RR was 1.21 (95% CI: 1.01–1.45). The association of the weight gain z-score at approximately 28 weeks with the risk of HDP was null. Considering the small sample and the low incidence of HDP in the underweight group, the test efficiency may have been insufficient. Weight gain and weight gain z-scores at approximately 28 weeks in the normal-weight group were associated with the risk of HDP (weight gain RR: 1.08, 95% CI: 1.02–1.15; weight gain z-score RR: 1.29, 95% CI: 1.05–1.58). Compared to the reference of low weight gain (<−1 SD), weight gain >+1 SD was associated with an approximately 2.0 times greater likelihood of HDP (RR: 2.08, 95% CI: 1.06–4.08). Weight gain and weight gain z-scores at the remaining time points were not related to the risk of HDP, and the relationships among the other periods were not significant. Among the overweight women, weight gain and weight gain z-scores at approximately 16 weeks were associated with an increased risk of HDP (weight gain RR: 1.12, 95% CI: 1.01–1.25; weight gain z-score RR: 1.32, 95% CI: 1.09–1.57). Compared to the reference of low weight gain (<−1 SD), weight gain >+1 SD at approximately 16 weeks significantly increased the risk of developing HDP (RR: 4.21, 95% CI: 1.36–13.01), while in obese women, there was a negative relationship between weight gain at approximately 16 weeks and the risk of HDP. Associations were null for the other periods.

Discussion

RESEARCH LIMITATIONS:

First, our study population was located in 19 communities in Tianjin, and the resident population is mostly of Han ethnicity. Therefore, this study can only represent the characteristics of the Han population in northern China. Second, considering the low incidence of HDP, the overall population of the study was not large, especially for the underweight and obese women, so the research data in these 2 groups were not sufficiently accurate. Third, some studies have shown that the relationship between weight gain and the risk of HDP subtypes was not the same. The present study did not analyze this relationship because of the number of people enrolled. Results are still being collated, and future large-sample cohort studies may provide a sufficient number of cases for analysis. Finally, using early-pregnancy BMI as the basis for grouping may have caused bias in enrolment. However, the pre-pregnancy weight mainly depended on the memory of the pregnant woman, and there may be recall errors. The grouping basis of the z-score chart in the Chinese population that our study represented is the early-pregnancy BMI, and the sensitivity analyses in the study indicated that the possibility of misclassification was small and acceptable [14].

Conclusions

Excessive weight gain before the third trimester, independent of age, early pregnancy BMI, primipara, and family history of hypertension, was associated with a greater risk of developing HDP among women with early-pregnancy normal weight, which may provide a chance to identify subsequent hypertensive disorders. The relationship between weight gain in early pregnancy and the risk of HDP remains unclear. Additional large-sample research with serial antenatal weight measure (including pre-pregnancy weight) are need to determine whether early-pregnancy weight gain is associated with HDP risk.