Abstract
Objectives:
It is important to understand relationships of gestational weight gain with adverse pregnancy outcomes in women with chronic hypertension, given their high baseline risk of adverse outcomes. We assessed associations of gestational weight gain with adverse pregnancy outcomes in women with chronic hypertension by pre-pregnancy body mass index categories.
Study Design:
We identified 14,369 women with chronic hypertension using electronic health records from 3 integrated health care delivery systems (2005-2014). Gestational weight gain-for-gestational age charts were used to calculate gestational weight gain z-scores, which account for gestational age. Modified Poisson regression models using generalized estimating equations were used to calculate relative risks and 95% confidence intervals, adjusted for sociodemographic and medical characteristics.
Main Outcome Measurements:
Preeclampsia, preterm delivery, cesarean delivery, neonatal intensive care unit admission, birthweight (extracted from the electronic health record)
Results:
In women with normal weight or overweight, low gestational weight gain (z-score<−1) was associated with 27-28% greater risk of preterm delivery and 48-82% greater risk of small-for-gestational age birthweight, while high gestational weight gain (z-score>1) was associated with 40-90% greater risk of preeclampsia and 59-113% greater risk of LGA. In women with obesity, low GWG was associated with 27-54% lower risk of several adverse pregnancy outcomes, including preeclampsia and cesarean delivery.
Conclusions:
In women with chronic hypertension and normal weight or overweight, moderate gestational weight gain may confer the lowest risk of adverse outcomes. In women with chronic hypertension and obesity, low gestational weight gain may be necessary for the lowest risk of adverse pregnancy outcomes.
Keywords: birthweight, cesarean, gestational weight gain, neonatal intensive care unit, preeclampsia, preterm birth
INTRODUCTION
Gestational weight gain (GWG) is a potentially modifiable risk factor for pregnancy complications. The Institute of Medicine (IOM) recommends different ranges for GWG depending on pre-pregnancy body mass index (BMI) category, recognizing the joint influence of BMI and GWG on pregnancy outcomes [1]. In the general population of pregnant women, GWG below and above the IOM guidelines is associated with several adverse pregnancy outcomes, with highest risks of adverse outcomes in women with obesity and high GWG [2, 3].
Pregnant women with chronic hypertension are at high risk for several complications, such as preeclampsia, preterm delivery, cesarean delivery, neonatal intensive care unit (NICU) admission, and low birthweight [4, 5]. Because of this high baseline risk of adverse outcomes, relationships of GWG with pregnancy outcomes in women with chronic hypertension may be different than those in healthy pregnant women. It is important to understand how GWG impacts maternal and infant outcomes in women with chronic hypertension to inform potential GWG guidelines specific to this high-risk population.
Previous studies have reported inconsistent associations of GWG outside the IOM guidelines with pregnancy outcomes in women with chronic hypertension; however, these previous studies have not fully accounted for gestational age (GA) in these relationships [6-8]. GA at delivery must be carefully considered when studying relationships between GWG and pregnancy outcomes because women with a longer gestation have greater total GWG at delivery. Approaches that do not fully account for the correlation between GWG and GA [9, 10] may artificially inflate associations of low GWG with adverse outcomes, suggesting stronger relationships than truly exist. To account for the correlation between GWG and GA at delivery, standardized GWG-for-GA z-score charts have been developed [11, 12]. These z-scores allow for comparison of GWG between women with adverse pregnancy outcomes and women without adverse pregnancy outcomes at the same point in pregnancy. This is in contrast to previous methods comparing GWG between women with adverse pregnancy outcomes and women without adverse pregnancy outcomes at delivery (which tends to be at a later GA for women without adverse pregnancy outcomes) [13]. In this study, we used GWG-for-GA z-scores to assess the association of GWG with several pregnancy complications and neonatal outcomes by pre-pregnancy BMI category in women with chronic hypertension.
METHODS
Study Setting and Population
We used data from a retrospective cohort study of women with hypertension during pregnancy within three integrated health care delivery systems in the United States: Kaiser Permanente Northern California, Kaiser Permanente Southern California, and Kaiser Permanente Washington. This study has been previously described [14]. Briefly, women were included if they were enrolled in one of the three health plans from 16 weeks gestation through delivery and delivered a singleton live birth or stillbirth from January 1, 2007 through December 31, 2014. Deliveries between January 1, 2005 and December 31, 2006 at Kaiser Permanente Northern California were also included. Time periods were based on availability of blood pressure data in the electronic health record (EHR) at each site.
Women were considered to have chronic hypertension if they met any of the following criteria, based on the American College of Obstetricians and Gynecologists definition of chronic hypertension [15]: 1) blood pressure ≥140/90 mmHg on two separate days within a 30 day period between the start of pregnancy and 20 weeks gestation, 2) at least one antihypertensive medication prescription fill (Table S1) in the 120 days before pregnancy and at least one diagnosis code for hypertension (International Classification of Disease [ICD]-9 code: 401-405, 437.2, 642.00-642.34, 642.7X, 642.9X, 760.0) between one year prior to pregnancy and 20 weeks gestation, or 3) one blood pressure ≥140/90 mmHg between the start of the pregnancy and 20 weeks gestation accompanied by a hypertension diagnosis code and a antihypertensive medication prescription fill within 7 days of the high blood pressure value. Of 16,358 eligible pregnancies, stillbirths (n=190), pregnancies in underweight women (n=74), and pregnancies with missing pre-pregnancy BMI (n=1,179) were excluded for a final analytic sample of 14,915 pregnancies. For pregnancies with missing data for GWG z-score or covariates (11% missing GWG z-score, 3.8% missing education, 3.7% missing parity, 0.3% missing race/ethnicity), we used multiple imputation using the Markov Chain Monte Carlo method under the assumption of multivariate normality with 100 imputations to impute GWG z-scores, education, parity, and race/ethnicity. After imputation, we excluded pregnancies with missing data for outcomes [16] (N=4 missing preterm delivery, N=544 missing small-for-GA (SGA), and N=546 missing large-for-GA (LGA)), for a total of 14,369 pregnancies among 13,356 women.
This study was approved by Institutional Review Boards at all three study sites and the states of California and Washington.
Data Collection
This study used EHR data, supplemented by state birth certificate data. Estimated due date was obtained from the EHR. For this study, the start of pregnancy was calculated as estimated due date minus 280 days.
Exposure: Gestational weight gain
Pre-pregnancy weight was defined as clinically measured weight closest to the start of pregnancy in the 12 months prior to pregnancy. If weight in the 12 months prior to pregnancy was not available, measured weight before 90 days of pregnancy or self-reported pre-pregnancy weight from the birth certificate was used. Total GWG was calculated as the difference between last clinically measured prenatal weight within 7 days before delivery and pre-pregnancy weight. Since preeclampsia diagnosis typically occurs before delivery, it would not be appropriate to include GWG that occurred after preeclampsia diagnosis in analyses with preeclampsia as the outcome. GWG up to preeclampsia diagnosis was calculated as the difference between weight measured within the 14 days prior to preeclampsia diagnosis and pre-pregnancy weight. GWG-for-GA charts specific for each pre-pregnancy BMI category were used to calculate GWG z-score, which is a measure of the number of standard deviations a woman’s GWG is from the mean GWG at the GA of the outcome of interest [11, 12]. One GWG z-score corresponds to different magnitudes of GWG in different pre-pregnancy BMI categories (Table S2).
Outcomes
Preeclampsia/eclampsia was identified using diagnosis codes (ICD-9 codes: 642.4-642.7) in inpatient encounters starting at 20 weeks gestation. We reviewed 45 charts to examine the validity of this approach and found a positive predictive value of 93%. Preterm delivery was defined as GA at delivery <37 weeks. Cesarean delivery was identified using procedure codes (ICD-9 codes: 74.0-74.2, 74.4, 74.9). NICU admission during the first 28 days of life was identified using hospitalization and billing records. SGA and LGA were calculated according to national sex- and race-specific birthweight reference curves [17]. SGA was defined as birthweight-for-GA <10th percentile. LGA was defined as birthweight-for-GA >90th percentile.
Covariates
Covariate data were obtained from the EHR and birth certificates. Race/ethnicity was categorized as: non-Hispanic white, Hispanic, non-Hispanic black, non-Hispanic Asian, other. Maternal education was categorized as: did not complete college, completed college. Parity was categorized as: nulliparous, multiparous. Antihypertensive medication use during pregnancy was defined as having at least one antihypertensive medication fill (Table S1) between start of pregnancy and delivery.
Pre-pregnancy BMI category
Pre-pregnancy BMI was obtained from the EHR and categorized using standard BMI cut-points for adults: underweight (<18.5 kg/m2), normal weight (18.5-24.9 kg/m2), overweight (25-29.9 kg/m2), obese (≥30 kg/m2) [18].
Statistical Analyses
All analyses were stratified by pre-pregnancy BMI category. We excluded women with underweight due to the small number of women in this category (N=45). We calculated descriptive statistics using mean and standard deviation for continuous variables, and frequency and percent for categorical variables by pre-pregnancy BMI category.
GWG z-scores were categorized into three categories to allow for non-linear associations and to aid in interpretation: <−1 (low), −1 to 1 (moderate), and >1 (high). Modified Poisson regression models [19] were used to estimate relative risks (RR) and 95% confidence intervals (CI) for GWG z-score category with preeclampsia, preterm delivery, cesarean delivery, NICU admission, SGA, and LGA. Moderate GWG was the reference category. Separate regression models were run for each outcome. Regression parameters were estimated using generalized estimating equations to account for some women having multiple pregnancies during the study period. Models were adjusted for maternal age at delivery (years), race/ethnicity, maternal education, Medicaid during pregnancy, nulliparity, tobacco use during pregnancy, and antihypertensive medication use during pregnancy. Models with SGA and LGA as the outcome were additionally adjusted for neonatal sex. To account for potential edema prior to preeclampsia diagnosis, we conducted a sensitivity analysis excluding GWG in the 28 days prior to preeclampsia diagnosis in analyses with preeclampsia as the outcome. Regression models were run using each imputed dataset. Results were combined using Rubin’s rules [20].
To explore possible non-linear associations across the range of GWG z-scores, risks of each outcome were plotted across the range of GWG z-score values with GWG z-score modeled as a restricted cubic spline with 5 knots determined by Harrell’s default percentiles [21], adjusted for the mean values of all covariates. Complete case data (N=12,324) was used for plotting after comparison of multiply imputed and complete case results in primary analyses (results were similar) because of challenges with plotting multiply imputed data. GWG z-scores below the 1st percentile (<−3) are not shown.
SAS 9.4 (Cary, NC) and Stata 16.0 (College Station, TX) were used for statistical analyses.
RESULTS
Most women had obesity entering pregnancy (68%, Table 1). Women with obesity had lower GWG than women with normal weight or overweight. The average GWG z-score in all pre-pregnancy BMI categories was <0, indicating that on average, women in our population gained less than the reference population in which the GWG z-score curves were developed [11, 12].
Table 1.
Characteristics of pregnancies in women with chronic hypertension by pre-pregnancy BMI category (N=14,369)
| Characteristics | Normal weight (N=2225) |
Overweight (N=3382) |
Obese (N=8762) |
|---|---|---|---|
| Sociodemographic characteristics | |||
| Maternal age (years) mean (SD) | 32.2 (5.9) | 32.8 (5.8) | 32.5 (5.5) |
| Race/ethnicity, n (%) | |||
| Non-Hispanic white | 892 (40) | 1104 (33) | 2808 (32) |
| Hispanic | 448 (20) | 961 (29) | 3301 (38) |
| Non-Hispanic Asian | 641 (29) | 805 (24) | 1004 (11) |
| Non-Hispanic black | 223 (10) | 476 (14) | 1556 (18) |
| Other | 10 (0.5) | 21 (1) | 75 (1) |
| Missing | 11 | 15 | 18 |
| At least college education, n (%) | 1211 (57) | 1531 (47) | 3100 (37) |
| Missing | 90 | 110 | 293 |
| Medicaid, n (%) | 99 (4) | 179 (5) | 751 (9) |
| Pregnancy characteristics | |||
| Total gestational weight gain (kg), mean (SD) | 13.4 (5.5) | 12.2 (6.4) | 9.1 (7.4) |
| Missing | 121 | 124 | 374 |
| Gestational weight gain z-score, mean (SD) | −0.5 (1.2) | −0.5 (1.0) | −0.2 (0.9) |
| Missing | 342 | 336 | 872 |
| Gestational weight gain z-score category, n (%) | |||
| <−1 | 586 (31) | 806 (26) | 1174 (15) |
| −1 to 1 | 1181 (63) | 2082 (68) | 6115 (78) |
| >1 | 116 (6) | 158 (5) | 601 (8) |
| Missing | 342 | 336 | 872 |
| Institute of Medicine gestational weight gain category, n (%) | |||
| Below guidelines | 789 (38) | 631 (19) | 2390 (28) |
| Within guidelines | 664 (32) | 869 (27) | 1985 (24) |
| Above guidelines | 651 (31) | 1758 (54) | 4013 (48) |
| Missing | 121 | 124 | 374 |
| Nulliparous, n (%) | 1005 (47) | 1321 (40) | 3249 (38) |
| Missing | 106 | 116 | 271 |
| Tobacco use during pregnancy, n (%) | 110 (5) | 142 (4) | 525 (6) |
| Diabetes in pregnancy, n (%) | |||
| Pre-existing diabetes | 130 (6) | 451 (13) | 2041 (23) |
| Gestational diabetes | 286 (13) | 577 (17) | 1605 (18) |
| Antihypertensive medication use during pregnancy, n (%) | 948 (43) | 1579 (47) | 4333 (49) |
| Gestational age at delivery (weeks), mean (SD) | 38.2 (2.5) | 38.1 (2.5) | 38.1 (2.5) |
| Female infant, n (%) | 1047 (47) | 1652 (49) | 4533 (52) |
| Missing | 2 | 0 | 1 |
| Adverse pregnancy outcomes | |||
| Preeclampsia, n (%) | 421 (19) | 763 (23) | 2026 (23) |
| Preterm delivery, n (%) | 416 (19) | 675 (20) | 1670 (19) |
| Cesarean delivery, n (%) | 508 (23) | 943 (28) | 3218 (37) |
| NICU admission, n (%) | 369 (17) | 601 (18) | 1668 (19) |
| Small-for-gestational age, n (%) | 363 (16) | 454 (13) | 833 (10) |
| Large-for-gestational age, n (%) | 88 (4) | 266 (14) | 1199 (14) |
SD, standard deviation
Women with normal weight
In women with normal weight, risk of LGA and preeclampsia increased with greater GWG z-score, and risk of NICU admission and SGA consistently decreased with greater GWG z-score (Figure 1 and 2). Within the low and moderate GWG categories, risk of preterm delivery consistently decreased and risk of cesarean delivery remained steady. Within the high GWG category, risk for both preterm delivery and cesarean delivery increased with greater GWG z-score. Low GWG was associated with 27% greater risk of preterm delivery, 82% greater risk of SGA, and 58% lower risk of LGA than moderate GWG. High GWG was associated with 40% greater risk of preeclampsia, 44% greater risk of cesarean delivery, and 61% lower risk of SGA than moderate GWG (Table 2).
Figure 1.
Adjusted risk of preeclampsia across the GWG z-score distribution by pre-pregnancy BMI category (green=normal weight; orange=overweight; blue=obese)
Figure 2.
Adjusted risk of adverse outcomes across the GWG z-score distribution by pre-pregnancy BMI category (green=preterm delivery; yellow=cesarean delivery; blue=small-for-gestational age; orange=large-for-gestational age; pink=NICU admission
Table 2.
Associations of gestational weight gain z-score and pregnancy outcomes stratified by pre-pregnancy BMI category
| Preeclampsia | Preterm Delivery |
Cesarean Delivery |
NICU Admission |
Small-for- gestational age |
Large-for- gestational age |
|
|---|---|---|---|---|---|---|
| Gestational weight gain z-score* |
RR (95% CI) |
RR (95% CI) |
RR (95% CI) |
RR (95% CI) |
RR† (95% CI) |
RR† (95% CI) |
| Normal weight women | ||||||
| <−1 | 0.83 (0.67, 1.03) | 1.27 (1.04, 1.55) | 1.04 (0.87, 1.25) | 1.10 (0.89, 1.36) | 1.82 (1.48, 2.24) | 0.42 (0.22, 0.77) |
| −1 to 1 | Referent | Referent | Referent | Referent | Referent | Referent |
| >1 | 1.40 (1.02, 1.94) | 1.06 (0.70, 1.59) | 1.44 (1.06, 1.96) | 0.84 (0.52, 1.34) | 0.39 (0.20, 0.75) | 1.59 (0.84, 3.03) |
| Overweight women | ||||||
| <−1 | 0.87 (0.75, 1.02) | 1.28 (1.09, 1.49) | 0.89 (0.78, 1.02) | 1.23 (1.04, 1.45) | 1.48 (1.23, 1.78) | 0.43 (0.30, 0.63) |
| −1 to 1 | Referent | Referent | Referent | Referent | Referent | Referent |
| >1 | 1.90 (1.53, 2.34) | 1.45 (1.07, 1.95) | 1.16 (0.91, 1.47) | 1.14 (0.82, 1.58) | 0.49 (0.27, 0.88) | 2.13 (1.46, 3.10) |
| Obese women | ||||||
| <−1 | 0.68 (0.60, 0.78) | 0.99 (0.88, 1.13) | 0.73 (0.67, 0.81) | 1.00 (0.88, 1.14) | 1.68 (1.43, 1.96) | 0.46 (0.37, 0.56) |
| −1 to 1 | Referent | Referent | Referent | Referent | Referent | Referent |
| >1 | 1.64 (1.47, 1.82) | 1.34 (1.15, 1.56) | 1.22 (1.12, 1.33) | 1.32 (1.13, 1.53) | 0.48 (0.33, 0.69) | 2.01 (1.74, 2.33) |
CI, confidence interval; NICU, neonatal intensive care unit; RR, relative risk
Gestational weight gain z-scores of −1 and 1 correspond to total gestational weight gain at 38 weeks (mean gestational age at delivery in our cohort) of 10.6 kg and 21.7 kg among normal weight women, 8.4 kg and 23.1 kg among overweight women, 5.2 kg to 21.0 kg among grade 1 obese women, 1.4 kg and 19.5 kg among grade 2 obese women, and −2.5 kg and 19.3 kg among grade 3 obese women. Gestational weight gain at delivery was used in analyses of all outcomes except preeclampsia; for preeclampsia analyses, gestational weight gain at diagnosis (for women with preeclampsia) or delivery (for women without preeclampsia) was used. All analyses adjusted for maternal age at delivery, race/ethnicity, maternal education, Medicaid, nulliparity, tobacco use during pregnancy, and antihypertensive medication use during pregnancy.
Additionally adjusted for infant sex.
Women with overweight
In women with overweight, risk of preeclampsia, cesarean delivery, and LGA increased and risk of SGA decreased with greater GWG z-score across the entire range of GWG z-scores (Figure 1 and 2). Associations of GWG z-score with risk of preterm delivery and NICU admission were U-shaped. Low GWG was associated with 28% greater risk of preterm delivery, 23% greater risk of NICU admission, 48% greater risk of SGA, and 57% lower risk of LGA than moderate GWG. High GWG was associated with 90% greater risk of preeclampsia, 45% greater risk of preterm delivery, and 113% greater risk of LGA than moderate GWG (Table 2).
Women with obesity
In women with obesity, risk of cesarean delivery and LGA increased and risk of SGA decreased with greater GWG z-score across the entire range of GWG z-scores (Figure 1 and 2). Risks of preterm delivery and risk of cesarean delivery were elevated at GWG z-scores of >1. Low GWG was associated with 32% lower risk of preeclampsia, 27% lower risk of cesarean delivery, 68% greater risk of SGA, and 54% lower risk of LGA than moderate GWG. High GWG was associated with 64% greater risk of preeclampsia, 34% greater risk of preterm delivery, 22% greater risk of cesarean delivery, 32% greater risk of NICU admission, 52% lower risk of SGA, and 101% greater risk of LGA than moderate GWG (Table 2).
Results from Sensitivity Analysis
Associations of high GWG with greater preeclampsia risk were slightly attenuated when excluding weight gain in the 28 days prior to preeclampsia diagnosis (Table S3).
DISCUSSION
Main Findings
Our results suggest that the relationship between GWG and preterm delivery, cesarean delivery, and NICU admission may depend on pre-pregnancy BMI among women with chronic hypertension. In women with chronic hypertension and normal weight and overweight, low GWG and high GWG were both associated with greater risk of several adverse pregnancy outcomes. In women with normal weight or overweight, low GWG was associated with greater risk of preterm delivery and SGA, while high GWG was associated with greater risk of preeclampsia and LGA. In women with chronic hypertension and obesity, low GWG was associated with lower risk of several adverse pregnancy outcomes, including preeclampsia and cesarean delivery. Our findings suggest that moderate GWG for pregnant women with chronic hypertension and normal weight or overweight may be optimal, while low GWG may be beneficial in women with chronic hypertension and obesity for optimal pregnancy outcomes.
We are not aware of previous studies using GWG z-scores in women with chronic hypertension. Three previous studies have reported associations of GWG according to the IOM guidelines with pregnancy outcomes in women with chronic hypertension [6-8]. The largest of these studies (n=101,259), conducted using US birth certificate data, reported associations of GWG above IOM guidelines in term pregnancies with greater risk of eclampsia, cesarean delivery, NICU admission, and LGA and lower risk of SGA, with no effect modification by pre-pregnancy BMI.7 Two smaller studies (n<1,000) reported associations of GWG above the IOM guidelines with greater risk of preeclampsia and greater risk of preterm delivery [6-8].
Our finding that pre-pregnancy BMI is an important factor in the associations of greater GWG z-score with preterm delivery, cesarean delivery, and NICU admission is different from previous studies in women with chronic hypertension [6, 7]. Our study is the first to use GWG z-scores to fully account for GA in these associations, which may explain observed differences in results. Associations of GWG with preterm delivery are expected to be most strongly affected by residual confounding by GA at delivery, since preterm birth is defined by GA at delivery.
Two previous studies of GWG z-scores and adverse pregnancy outcomes in the general population of pregnant women, an individual participant data meta-analysis of 39 cohorts [3] and a cohort study of 760,043 births in Pennsylvania [22], reported associations of greater GWG z-score with lower risk of SGA and greater risk of LGA, which is consistent with our observed results [3, 22]. Both studies also observed associations of both low and high GWG z-scores with greater risk of preterm birth overall and separately in women with normal weight or overweight (women with obesity were not included in the study) [22]. In our study, we observed similar associations of low and high GWG z-scores with greater risk of preterm birth in women with overweight, while only low GWG z-scores were associated with greater risk of preterm birth in women with normal weight. We are not aware of previous studies that have examined associations of GWG z-scores with cesarean delivery or NICU admission.
Interpretation
Findings from our study are important for informing clinical recommendations for GWG in women with chronic hypertension that account for risks and benefits by pre-pregnancy BMI category. Consistent with previous studies, we observed associations of high GWG with greater risk of preeclampsia and LGA in all pre-pregnancy BMI categories [2, 3]. These results suggest that reducing GWG in women with chronic hypertension may result in lower risk of preeclampsia and LGA. As women with chronic hypertension are at high risk for preeclampsia [4, 5], these results also suggest that women with high GWG in all pre-pregnancy BMI categories may require greater monitoring for preeclampsia.
Our findings suggest that the relationship between GWG and preterm delivery, cesarean delivery, and NICU admission may depend on pre-pregnancy BMI among women with chronic hypertension. In women with obesity prior to pregnancy, high GWG was associated with greater risk of preterm delivery, cesarean delivery, and NICU admission. About 2 in 3 women in our chronic hypertension cohort had obesity, which conveys a baseline high risk for these adverse outcomes [4, 23-28]. Reducing GWG in women with chronic hypertension and obesity may reduce the risk of preterm delivery, cesarean delivery, NICU admission, in addition to preeclampsia and LGA. While low GWG is also associated with greater risk of SGA, the corresponding lower risk of NICU admission suggests that this greater risk of SGA may not represent an adverse health outcome.
GWG targets, comparable to the IOM guidelines, have not yet been developed using GWG z-scores, making comparison to the IOM guidelines challenging. At 38 weeks gestation (the mean GA at delivery in our cohort) one GWG z-score corresponds to different magnitudes of GWG in different pre-pregnancy BMI categories, ranging from 6.2 kg in women who are normal weight prior to pregnancy to 13.1 kg in women with class III obesity prior to pregnancy (Table S2). It will be crucial for future research to identify optimal GWG z-score ranges by pre-pregnancy BMI category, weighing the varying severity of adverse pregnancy outcomes in women with chronic hypertension, to allow comparison with current IOM GWG guidelines, to increase clinical utility of GWG z-scores, and inform future GWG guidelines. The recommended lower and upper GWG z-score values at each GA can be graphically represented as GWG z-score charts for clinical use (similar to birthweight and childhood growth charts currently in use in pediatric practice).
Strengths and Limitations
Our large study with a diverse population of women with chronic hypertension used clinically measured weights from the EHR to calculate GWG, which likely reduced misclassification of GWG compared to previous studies that used birth certificate data [29]. Our use of GWG z-scores improves upon previous research in pregnant women with chronic hypertension, which may overestimate associations of low GWG with adverse pregnancy outcomes, by more accurately addressing confounding by GA.
It is possible that our measure of GWG up to the date of preeclampsia diagnosis is capturing GWG as a symptom of the disease. However, our sensitivity analyses showed associations of GWG with preeclampsia were modestly attenuated when excluding weight gained in the 28 days prior to preeclampsia diagnosis, which is consistent with results from a previous study [8]. This suggests that GWG due to edema prior to preeclampsia diagnosis does not fully explain observed associations of greater GWG with greater preeclampsia risk. We applied GWG-for-GA curves developed using a population of pregnant women with uncomplicated pregnancies to our cohort of women with chronic hypertension. To the extent that the GWG distribution differs between women with chronic hypertension and women with uncomplicated pregnancies, this may have introduced measurement error in GWG z-score in our study. Although we adjusted for important sociodemographic and medical characteristics in our analysis, as is the case in any observational study, we cannot rule out that residual confounding may have biased our findings.
Conclusions
In women with chronic hypertension, it is important to consider pre-pregnancy BMI in the relationship between GWG and adverse pregnancy outcomes. Interventions addressing high GWG may help reduce adverse pregnancy outcomes in all pre-pregnancy BMI categories, though GWG targets for optimal pregnancy outcomes may differ by pre-pregnancy BMI. For women with chronic hypertension and normal weight or overweight, moderate GWG may result in optimal pregnancy outcomes. For women with chronic hypertension and obesity, low GWG may be beneficial for optimal pregnancy outcomes. Clinical counseling and lifestyle interventions targeting GWG in pregnant women with chronic hypertension should take pre-pregnancy BMI into account.
Supplementary Material
HIGHLIGHTS.
Gestational weight gain is a potentially modifiable risk factor for adverse pregnancy outcomes.
In women with normal weight or overweight, moderate gestational weight gain may result in the lowest risk of adverse pregnancy outcomes.
In women with obesity, low gestational weight gain may be beneficial for the lowest risk of adverse pregnancy outcomes.
Acknowledgments
FUNDING
This study was funded by the Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD) R01HD082141. SEB was partly funded by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) T32DK11668401 and NICHD K99HD100585. NICHD and NIDDK were not involved in the conduct of this research or preparation of this manuscript.
Footnotes
DISCLOSURE OF INTERESTS
LC is currently an employee of Genentech, A Member of the Roche Group. Dr. Chen is in a data infrastructure role that is not focused on any specific therapies. The current publication is based on work completed while she was employed by Kaiser Permanente Washington Health Research Institute. The remaining authors report no conflict of interest.
ETHICS APPROVAL
This study was approved by Institutional Review Boards at Kaiser Permanente Northern California, Kaiser Permanente Southern California, and Kaiser Permanente Washington and the states of California and Washington (approval date: August 24, 2015; IRBNet #1277850).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
REFERENCES
- [1].ACOG Committee opinion no. 548: weight gain during pregnancy, Obstetrics and gynecology 121(1) (2013) 210–2. [DOI] [PubMed] [Google Scholar]
- [2].Goldstein RF, Abell SK, Ranasinha S, Misso M, Boyle JA, Black MH, Li N, Hu G, Corrado F, Rode L, Kim YJ, Haugen M, Song WO, Kim MH, Bogaerts A, Devlieger R, Chung JH, Teede HJ, Association of Gestational Weight Gain With Maternal and Infant Outcomes: A Systematic Review and Meta-analysis, Jama 317(21) (2017) 2207–2225. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Santos S, Voerman E, Amiano P, Barros H, Beilin LJ, Bergstrom A, Charles MA, Chatzi L, Chevrier C, Chrousos GP, Corpeleijn E, Costa O, Costet N, Crozier S, Devereux G, Doyon M, Eggesbo M, Fantini MP, Farchi S, Forastiere F, Georgiu V, Godfrey KM, Gori D, Grote V, Hanke W, Hertz-Picciotto I, Heude B, Hivert MF, Hryhorczuk D, Huang RC, Inskip H, Karvonen AM, Kenny LC, Koletzko B, Kupers LK, Lagstrom H, Lehmann I, Magnus P, Majewska R, Makela J, Manios Y, McAuliffe FM, McDonald SW, Mehegan J, Melen E, Mommers M, Morgen CS, Moschonis G, Murray D, Ni Chaoimh C, Nohr EA, Nybo Andersen AM, Oken E, Oostvogels A, Pac A, Papadopoulou E, Pekkanen J, Pizzi C, Polanska K, Porta D, Richiardi L, Rifas-Shiman SL, Roeleveld N, Ronfani L, Santos AC, Standl M, Stigum H, Stoltenberg C, Thiering E, Thijs C, Torrent M, Tough SC, Trnovec T, Turner S, van Gelder M, van Rossem L, von Berg A, Vrijheid M, Vrijkotte T, West J, Wijga AH, Wright J, Zvinchuk O, Sorensen T, Lawlor DA, Gaillard R, Jaddoe V, Impact of maternal body mass index and gestational weight gain on pregnancy complications: an individual participant data meta-analysis of European, North American and Australian cohorts, BJOG : an international journal of obstetrics and gynaecology 126(8) (2019) 984–995. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Bramham K, Parnell B, Nelson-Piercy C, Seed PT, Poston L, Chappell LC, Chronic hypertension and pregnancy outcomes: systematic review and meta-analysis, Bmj 348 (2014) g2301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Panaitescu AM, Syngelaki A, Prodan N, Akolekar R, Nicolaides KH, Chronic hypertension and adverse pregnancy outcome: a cohort study, Ultrasound Obstet Gynecol 50(2) (2017) 228–235. [DOI] [PubMed] [Google Scholar]
- [6].Ornaghi S, Algeri P, Todyrenchuk L, Vertemati E, Vergani P, Impact of excessive pre-pregnancy body mass index and abnormal gestational weight gain on pregnancy outcomes in women with chronic hypertension, Pregnancy hypertension 12 (2018) 90–95. [DOI] [PubMed] [Google Scholar]
- [7].Yee LM, Caughey AB, Cheng YW, Association between gestational weight gain and perinatal outcomes in women with chronic hypertension, American journal of obstetrics and gynecology 217(3) (2017) 348.e1–348.e9. [DOI] [PubMed] [Google Scholar]
- [8].Siegel AM, Tita AT, Machemehl H, Biggio JR, Harper LM, Evaluation of Institute of Medicine Guidelines for Gestational Weight Gain in Women with Chronic Hypertension, AJP reports 7(3) (2017) e145–e150. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Hutcheon JA, Bodnar LM, Joseph KS, Abrams B, Simhan HN, Platt RW, The bias in current measures of gestational weight gain, Paediatr Perinat Epidemiol 26(2) (2012) 109–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Hutcheon JA, Bodnar LM, Good Practices for Observational Studies of Maternal Weight and Weight Gain in Pregnancy, Paediatr Perinat Epidemiol 32(2) (2018) 152–160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Hutcheon JA, Platt RW, Abrams B, Himes KP, Simhan HN, Bodnar LM, A weight-gain-for-gestational-age z score chart for the assessment of maternal weight gain in pregnancy, The American journal of clinical nutrition 97(5) (2013) 1062–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Hutcheon JA, Platt RW, Abrams B, Himes KP, Simhan HN, Bodnar LM, Pregnancy weight gain charts for obese and overweight women, Obesity (Silver Spring, Md.) 23(3) (2015) 532–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Johansson K, Hutcheon JA, Stephansson O, Cnattingius S, Pregnancy weight gain by gestational age and BMI in Sweden: a population-based cohort study, The American journal of clinical nutrition 103(5) (2016) 1278–84. [DOI] [PubMed] [Google Scholar]
- [14].Chen L, Shortreed SM, Easterling T, Cheetham TC, Reynolds K, Avalos LA, Kamineni A, Holt V, Neugebauer R, Akosile M, Nance N, Bider-Canfield Z, Walker RL, Badon SE, Dublin S, Identifying hypertension in pregnancy using electronic medical records: The importance of blood pressure values, Pregnancy hypertension 19(2020) 112–118. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].ACOG Practice Bulletin No. 203: Chronic Hypertension in Pregnancy, Obstetrics and gynecology 133(1) (2019) e26–e50. [DOI] [PubMed] [Google Scholar]
- [16].von Hippel PT, 4. Regression with Missing Ys: An Improved Strategy for Analyzing Multiply Imputed Data, Sociological Methodology 37(1) (2007) 83–117. [Google Scholar]
- [17].Oken E, Kleinman KP, Rich-Edwards J, Gillman MW, A nearly continuous measure of birth weight for gestational age using a United States national reference, BMC pediatrics 3 (2003) 6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Centers for Disease Control and Prevention, About Adult BMI: How is BMI calculated and interpreted?, 2014. http://www.cdc.gov/healthyweight/assessing/bmi/adult_bmi/index.html#Interpreted. (Accessed October 3, 2014.
- [19].Zou G, A modified poisson regression approach to prospective studies with binary data, American journal of epidemiology 159(7) (2004) 702–6. [DOI] [PubMed] [Google Scholar]
- [20].Rubin D, Multiple Imputation for Nonresponse in Surveys, John Wiley, New York, 1987. [Google Scholar]
- [21].Harrell F Jr, Regression Modeling Strategies: With Applications to Linear Models, Logistic Regression, and Survival Analysis, Springer, New York, 2001. [Google Scholar]
- [22].Bodnar LM, Hutcheon JA, Parisi SM, Pugh SJ, Abrams B, Comparison of gestational weight gain z-scores and traditional weight gain measures in relation to perinatal outcomes, Paediatr Perinat Epidemiol 29(1) (2015) 11–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Bautista-Castano I, Henriquez-Sanchez P, Aleman-Perez N, Garcia-Salvador JJ, Gonzalez-Quesada A, Garcia-Hernandez JA, Serra-Majem L, Maternal obesity in early pregnancy and risk of adverse outcomes, PloS one 8(11) (2013) e80410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].El-Chaar D, Finkelstein SA, Tu X, Fell DB, Gaudet L, Sylvain J, Tawagi G, Wu Wen S, Walker M, The impact of increasing obesity class on obstetrical outcomes, Journal of obstetrics and gynaecology Canada : JOGC = Journal d'obstetrique et gynecologie du Canada : JOGC 35(3) (2013) 224–233. [DOI] [PubMed] [Google Scholar]
- [25].Athukorala C, Rumbold AR, Willson KJ, Crowther CA, The risk of adverse pregnancy outcomes in women who are overweight or obese, BMC pregnancy and childbirth 10 (2010) 56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Schummers L, Hutcheon JA, Bodnar LM, Lieberman E, Himes KP, Risk of adverse pregnancy outcomes by prepregnancy body mass index: a population-based study to inform prepregnancy weight loss counseling, Obstetrics and gynecology 125(1) (2015) 133–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Bicocca MJ, Mendez-Figueroa H, Chauhan SP, Sibai BM, Maternal obesity and the risk of early-onset and late-onset hypertensive disorders of pregnancy, Obstet Gynecol 136(1) (2020) 118–27. [DOI] [PubMed] [Google Scholar]
- [28].Robillard P, Dekker G, Boukerrou M, Boumahni B, Hulsey T, Scioscia M, Gestational weight gain and rate of late-onset preeclampsia: a retrospective analysis on 57 000 singleton pregnancies in Reunion Island, BMJ Open 10(7) (2020) e036549. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [29].Bodnar LM, Abrams B, Bertolet M, Gernand AD, Parisi SM, Himes KP, Lash TL, Validity of birth certificate-derived maternal weight data, Paediatr Perinat Epidemiol 28(3) (2014) 203–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.