Can blood pressure trajectories indicate who is at risk for developing hypertensive disorders of pregnancy?

OBJECTIVE:

The prediction and prevention of hypertensive disorders of pregnancy (HDP) is of paramount importance for the reduction of adverse perinatal outcomes. Prenatal care involves the routine, serial measurement of maternal blood pressures (BPs) in an outpatient setting. Previous literature has shown that the BPs of patients with uncomplicated pregnancies tend to fall in the second trimester before rising again in the third, whereas the BPs of those who develop HDP do not follow this trajectory. However, these studies were mostly conducted in selected samples with BPs collected per research protocol.^1,2

This study aimed to compare BP trajectories of pregnant women who did and did not develop HDP in an analysis of data from a large, diverse, prospective clinical cohort in which BP was measured during routine antenatal visits.

STUDY DESIGN:

This was a secondary analysis of pregnant women ≥18 years old and <18 weeks gestation who were enrolled from 3 NYU Langone Health-affiliated hospitals beginning in March 2016. Informed consent was obtained, and demographic and obstetrical data were collected from electronic medical records (EMRs) and questionnaires.

BP readings from gestational weeks 6 to 40 were retrospectively extracted from EMRs for outpatient antenatal visits of all participants who had delivered by December 2019. Participants with chronic hypertension (cHTN) or those with missing BP readings beyond 20 weeks’ gestation were excluded. HDP diagnoses (gestational hypertension, preeclampsia, eclampsia) were identified by International Classification of Diseases, 10th Revision codes and confirmed by manual chart review. Participants with HDP who were found on manual review to have elevated BPs <20 weeks’ gestation were removed from the analysis. Implausible BP readings (systolic <40 mm Hg, diastolic <30 mm Hg) were also removed. Trajectories of systolic, diastolic, and mean arterial pressure readings (averaged over 2-week increments) were compared between those who developed HDP and those who did not. A sensitivity analysis was conducted to ensure that the removal of cHTN participants did not significantly change our results. Analyses were conducted using SAS version 9.4 (SAS Institute, Cary, NC).

RESULTS:

Of 2025 participants, 1813 had complete, plausible antenatal BP data available. Among those, 175 (9.65%) developed HDP. Participants with HDP were significantly more likely to have a higher prepregnancy body mass index, carry twins, deliver by cesarean delivery, and deliver at an earlier gestational age (Table).

TABLE. Demographics and pregnancy characteristics of participants with and without hypertensive disorders of pregnancy.

Characteristics	Non-HDP participants n=1638	HDP participants n=175	P value
Age, mean (SD)	31.84 (5.39)	31.53 (6.35)	.53
Race or ethnicity, n (%)			.34
Non-Hispanic White	562 (34.35)	56 (32.00)
Hispanic	799 (48.84)	91 (52.00)
Non-Hispanic Black	81 (4.95)	14 (8.00)
Non-Hispanic Asian	151 (9.23)	10 (5.71)
Non-Hispanic other	13 (0.79)	1 (0.57)
Non-Hispanic multiple	30 (1.83)	3 (1.71)
Insurance, n (%)			.85
Private	802 (48.96)	87 (49.71)
Public	836 (51.04)	88 (50.29)
Household income, n (%)			.47
<$30,000	254 (15.96)	33 (19.08)
$30,000–$99,999	310 (19.48)	31 (17.92)
≥$100,000	619 (38.91)	72 (41.62)
Don’t know	408 (25.64)	37 (21.39)
Education, n (%)			.86
High school or less	538 (32.84)	54 (30.86)
Some college but no degree	178 (10.87)	22 (12.57)
Associate or bachelor’s degree	448 (27.35)	50 (28.57)
Postgraduate degree	474 (28.94)	49 (28.00)
Employment status, n (%)			.33
Employed	1056 (64.71)	119 (68.39)
Not employed	576 (35.29)	55 (31.61)
Marital status, n (%)			.35
Married or living with partner	1452 (88.70)	150 (85.71)
Divorced or separated	34 (2.08)	3 (1.71)
Single or widowed	151 (9.22)	22 (12.57)
Body mass index, mean (SD)	25.97 (5.55)	27.69 (6.56)	.001
Parity, n (%)			.26
Nulliparous	806 (49.24)	94 (53.71)
Parous	831 (50.76)	81 (46.29)
Gestation, n (%)			.02
Singleton	1596 (98.46)	167 (95.98)
Twins	25 (1.54)	7 (4.02)
Alcohol use, n (%)			.46
Never user	551 (33.64)	59 (33.71)
Past user	825 (50.37)	94 (53.71)
Current user	262 (16.00)	22 (12.57)
Smoking, n (%)			.93
Never user	1493 (91.15)	159 (90.86)
Past user	122 (7.45)	14 (8.00)
Current user	23 (1.40)	2 (1.14)
Delivery method, n (%)			.03
Vaginal	1092 (66.67)	102 (58.29)
Cesarean	546 (33.33)	73 (41.71)
GA at delivery, mean (SD)	39.19 (1.56)	38.17 (2.28)	<.001
Assisted reproduction, n (%)	105 (6.41)	15 (8.57)	.27

GA, gestational age; HDP, hypertension disorders of pregnancy; SD, standard deviation.

Rajeev. Blood pressure trajectories and hypertensive disorders of pregnancy. Am J Obstet Gynecol MFM 2022.

Participants averaged 10.6 BP readings throughout pregnancy. There was a range of 452 to 1571 BP readings per 2-week interval. Those who developed HDP had a higher BP at all time points (Figure, Panes A–C). Participants with an outpatient systolic BP >114 at any time <20 weeks’ gestation had a 2.37 times (95% confidence interval, 2.08–2.69) greater odds of developing HDP than those who did not after adjusting for prepregnancy obesity and twin gestation. The sensitivity, specificity, positive predictive value, and negative predictive value at this cutoff were 91.3% (90.9–91.7), 18.4% (16.7–20.2), 91.1% (90.6–91.5), and 18.8% (17.1–20.6), respectively.

FIGURE. Average blood pressures across pregnancy for participants with and without HDP.

BP, blood pressure; HDP, hypertensive disorders of pregnancy; MAP, mean arterial pressure.

Rajeev. Blood pressure trajectories and hypertensive disorders of pregnancy. Am J Obstet Gynecol MFM 2022.

The nadir for systolic BP occurred around gestational week 20 for HDP participants and week 24 for non-HDP participants. There was no statistically significant difference in the prenadir and postnadir slopes between the 2 groups except for the postnadir systolic BP, which increased more rapidly among women who developed HDP.

CONCLUSION:

Contrary to the results of previous studies, we observed similar BP trajectories among participants with and without HDP during the first and second trimesters, with both demonstrating a midgestation dip. On average, HDP participants without cHTN or elevated BP <20 weeks’ gestation had significantly higher BPs at all time points during pregnancy, and those with early gestational systolic BP >114 mm Hg had more than twice the odds of developing HDP. Therefore, clinicians may benefit from considering patients with prenatal systolic BP >114 mm Hg as high risk for HDP, irrespective of early pregnancy BP decline.

Although most studies show a U-shaped BP trajectory during uncomplicated pregnancies, more recent studies in 2022 and 2021 showed a stable, or even upward, trajectory.^3,4 However, participants from these studies were ethnically homogenous––Haitian and Chinese populations, respectively–and therefore their results are less generalizable to the US population. Our large, diverse participant population increases the generalizability of our results. Limitations of our study include not being able to account for antenatal aspirin use and reliance on BP taken during routine obstetrical visits. The latter is also a strength, however, because our findings are more relevant to the clinical setting.

We observed a midpregnancy decline in BP among participants who did and those who did not develop HDP, bringing into question the previous understanding of the physiology of placental abnormalities in relation to BP trajectories. Our findings have implications for clinicians who are trying to predict which patients may develop HDP and be at risk of postpartum cardiovascular complications.