The optimal gestational age to deliver patients with chronic hypertension on antihypertensive therapy

Abstract

Objective

To identify the optimal gestational age of planned delivery in pregnancies complicated by chronic hypertension requiring antihypertensive medications that minimizes the risk of adverse perinatal events and maternal morbidity.

Methods

Retrospective cohort study of singleton pregnancies after 37 weeks of gestation complicated by chronic hypertension on antihypertensive medication, delivered at 7 hospitals within an academic health system in New York from 12/1/2015 to 9/3/2020. Two comparisons were made (1) planned deliveries at 37–37^6/7weeks versus expectant management, (2) planned deliveries at 38–38^6/7weeks versus expectant management. Patients with other maternal or fetal conditions were excluded. The primary outcome was a composite of adverse perinatal outcomes including stillbirth, neonatal death, assisted ventilation, cord pH < 7.0, 5-minute Apgar ≤5, diagnosis of respiratory disorder, and neonatal seizures. The secondary outcomes included preeclampsia, eclampsia, primary cesarean delivery, postpartum readmission, and infant stay greater than 5 days. Odds ratios were estimated with multiple logistic regression and adjusted for confounding effects.

Results

A total of 555 patients met inclusion criteria. Patients who underwent planned delivery at 37 weeks compared to expectant management did not appear to be at higher risk of adverse perinatal outcomes (14.9% vs 10.4%, aOR 1.49, 95% CI: 0.77–2.88). Similarly, we did not find a difference in the primary outcome in patients who underwent planned delivery at 38 weeks versus those expectantly managed (9.7% vs 10.1%, (aOR 0.84, 95% CI: 0.39–1.76). There were no differences in the rates of primary cesarean or preeclampsia at 37 and 38 weeks.

Conclusion

Our findings suggest that there is no difference in neonatal or maternal outcomes for chronic hypertensive patients on medication if delivery is planned or expectantly managed at 37 or 38 weeks’ gestation.

Keywords:

• Pregnancy
• chronic hypertension
• preeclampsia
• delivery timing
• antihypertensive medication

Introductions

Chronic hypertension in pregnancy is associated with significant maternal and neonatal morbidity and is present in up to 1.5% of pregnant individuals [1]. Pregnancies complicated by chronic hypertension are at increased risk of preeclampsia, stroke, pulmonary edema, renal insufficiency, as well as poor fetal outcomes, such as stillbirth and growth restriction [2–5].

Due to these associated poor outcomes, for which the risk increases with advancing gestational age, early term delivery is often recommended. The American College of Obstetrics and Gynecology (ACOG) recommends that patients with isolated uncomplicated chronic hypertension not on medication be delivered between 38^0/7–39^6/7weeks, and those on antihypertensive medication be delivered between 37^0/7–39^6/7weeks [6]. Elective delivery prior to 39 weeks’ gestation, however, may be associated with increased risk of neonatal morbidities, such as respiratory complications, and higher rate of cesarean delivery [7,8]. Nevertheless, routine induction at 38 and 39 weeks in chronic hypertensive patients appeared to have the lowest rate of stillbirth and serious neonatal morbidity and decreased the rate for superimposed preeclampsia [9–11]. While these studies support consideration of early term delivery in patients with chronic hypertension, outcomes based on severity of chronic hypertension, including need for antihypertensive medication, were not evaluated.

Patients who require antihypertensive medications are presumed to have more advanced disease processes, thus justifying earlier delivery recommendations for maternal benefit. However, there are no prior studies evaluating if earlier delivery in patients with chronic hypertension on antihypertensive therapy is associated with any effect on adverse perinatal outcomes. The aim of our study is to identify the optimal gestational age of planned delivery in pregnancies complicated by chronic hypertension requiring antihypertensive medications that minimizes the risk of adverse perinatal events and maternal morbidity.

Materials and methods

This was a retrospective cohort study of all singleton pregnancies after 37 weeks’ gestation, complicated by chronic hypertension on antihypertensive medications, delivered at 7 hospitals within a large health system in New York. The study period was from 12/1/2015–9/3/2020. Institutional Review Board approval was obtained prior to conducting the study.

Patients with a diagnosis of chronic hypertension on antihypertensive medications at the time of admission were identified using the system-wide electronic medical record (EMR) system. After patients were identified, the prenatal records were reviewed to confirm a diagnosis of chronic hypertension. Patients were managed by both generalists and maternal-fetal medicine providers, and antihypertensives were adjusted based on provider preference. Timing of delivery was based on provider discretion.

Inclusion criteria consisted of singleton gestations at or beyond 37^0/7weeks, with a diagnosis of chronic hypertension, treated with antihypertensive medications during the pregnancy. Exclusion criteria consisted of patients with preeclampsia prior to 37^0/7weeks, significant maternal comorbidities (pre-gestational diabetes or significant renal disease), fetal growth restriction, fetal anomalies, chronic hypertensive patients not on antihypertensive medication, and patients’ EMR’s lacking information regarding antihypertensive use. The sample included all eligible patients identified by querying the EMR. No formal power calculation was used to select the sample size.

Detailed information was collected including maternal demographics, medical and obstetrical history, insurance type, body mass index (BMI), tobacco use, aspirin use, antenatal corticosteroid administration for fetal lung maturity, history of preeclampsia in a prior pregnancy, creatinine level, presence of proteinuria, type and number of antihypertensive medications used, gestational age at delivery, maternal outcomes, and neonatal outcomes.

To establish the optimal timing of delivery, patients were grouped by the week of gestational age at which delivery was planned (37^0/7–37^6/7, 38^0/7–38^6/7, 39^0/7-and beyond). Planned delivery was defined as either induction or scheduled cesarean for either chronic hypertension or any other medical/obstetrical indication. Expectant management was defined as delivery outside of the gestational age window or those not scheduled within the specified week. Patients were grouped by planned delivery to determine if an intended intervention optimized perinatal outcome. Two comparisons were made. The first comparison was between planned deliveries from 37^0/7–37^6/7weeks versus expectant management (unplanned deliveries from 37^0/7–37^6/7weeks and all deliveries 38^0/7weeks or later). The second comparison was planned deliveries from 38^0/7–38^6/7weeks versus expectant management (unplanned deliveries from 38^0/7–38^6/7weeks and all deliveries 39^0/7weeks or later).

The primary outcome was a composite of adverse perinatal outcomes, which included stillbirth, neonatal death, assisted ventilation, cord pH < 7.0, 5-min Apgar ≤5, diagnosis of respiratory disorder (respiratory failure, respiratory distress, acute hypoxia respiratory distress syndrome of newborn), and neonatal seizures. The secondary group of outcomes included preeclampsia, eclampsia, primary cesarean delivery, postpartum readmission, and infant hospital stay greater than 5 days.

Statistical analysis

Continuous variables were described by mean (standard deviation). Categorical variables were described by frequencies (percentage). Hypothesis tests were conducted using logistic regression models for each outcome variable with effects specified for exposure group and adjustment factors. Adjustment factors were number of medications, race, obesity, and prior pre-eclampsia. These factors were chosen before the analysis due to their confounding effects on the relationship between the exposure and the outcomes [12–14]. Statistical significance was determined for each test at the significance level α = 0.05. Statistical analysis was generated using SAS software (SAS Institute Inc., Cary, NC, USA).

Results

Of the 1682 patients with chronic hypertension identified in the study period, 384 patients were excluded due to not receiving antihypertensive medication and 601 not reaching 37 weeks gestation. An additional 142 patients were excluded for maternal and pregnancy complications, leaving a total of 555 patients meeting inclusion criteria.

A total of 94 patients (16.9%) underwent a planned delivery between 37^0/7–37^6/7 weeks gestation compared to 461 (83.1%) patients who did not. Of the remaining patients still pregnant at 38 weeks, 113 patients (29%) underwent a planned delivery at 38^0/7–38^6/7weeks compared to 277 patients who underwent expectant management. Table 1 shows baseline characteristics. Compared to patients in the expectant management group, those who underwent planned delivery at 37 weeks were more likely to be of Black race or take nifedipine, and they were less likely to receive antenatal corticosteroids or take methyldopa. No clinical or sociodemographic differences were noted between patients who underwent 38 week planned delivery compared to expectant management.

Table 1. Baseline characteristics.

	37 weeks			38 weeks
	Planned delivery 37–37.6 weeks n = 94	Expectant delivery beyond 37.0 weeks n = 461	p Value	Planned delivery 38–38.7 weeks n = 113	Expectant delivery beyond 38 weeks n = 277	p Value
Age	34.68 ± 5.12	34.74 ± 4.87	.92	35.16 ± 4.63	34.56 ± 4.81	0.26
Race			0.002			0.41
White	21 (22.3)	191 (41.4)		43 (38.0)	124 (44.8)
Black	37 (39.4)	122 (26.5)		33 (29.2)	66 (23.8)
Other/Unknown	36 (38.3)	148 (32.1)		37 (32.7)	87 (31.4)
Ethnicity			0.43			0.71
Hispanic	13 (13.8)	56 (12.2)		12 (10.6)	33 (11.9)
Not Hispanic	70 (74.5)	368 (79.8)		94 (83.2)	221 (79.8)
Unknown	11 (11.7)	37 (8.0)		7 (6.2)	23 (8.3)
Parity			0.60			0.61
0	33 (35.1)	156 (33.8)		34 (30.1)	99 (35.7)
1	38 (40.4)	169 (36.7)		47 (41.6)	94 (33.9)
>1	23 (24.5)	136 (29.5)		32 (28.3)	84 (30.3)
Insurance			0.23			0.33
Public Insurance	26 (27.7)	157 (34.1)		33 (29.2)	95 (34.3)
Private Insurance	68 (72.3)	304 (66.0)		80 (70.8)	182 (65.7)
BMI > 30	81 (86.2)	359 (77.9)	0.07	93 (82.3)	209 (75.5)	0.14
Tobacco Use	2 (2.1)	4 (0.9)	0.28	0 (0.00)	4 (1.4)	0.20
Proteinuria			0.61			0.36
Present	10 (10.6)	55 (11.9)		14 (12.4)	29 (10.5)
Not present	56 (59.6)	291 (63.1)		75 (66.4)	170 (61.4)
Unknown	28 (29.8)	115 (25.0)		24 (21.2)	78 (28.2)
Creatinine	0.60 ± 0.12	0.62 ± 0.15	0.21	0.63 ± 0.17	0.61 ± 0.13	0.32
Steroid	9 (9.6)	17 (3.7)	0.01	2 (1.8)	7 (2.5)	0.65
Aspirin	38 (40.4)	159 (34.5)	0.27	38 (33.6)	84 (30.3)	0.52
Prior Preeclampsia	19 (20.2)	72 (15.6)	0.27	13 (11.5)	47 (17.0)	0.17
GDMA	25 (26.6)	85 (18.4)	0.07	23 (20.4)	53 (19.1)	0.78
Maternal Comorbidities^a	8 (8.5)	40 (8.7)	0.96	12 (10.6)	21 (7.6)	0.33
Medication
Labetalol	71 (75.5)	352 (76.4)	0.86	83 (73.4)	211 (76.2)	0.57
Nifedipine	23 (24.5)	57 (12.4)	0.002	15 (13.3)	28 (10.1)	0.37
Amlodipine	2 (2.1)	18 (3.9)	0.40	7 (6.2)	10 (3.6)	0.27
Methyldopa	5 (5.3)	70 (15.2)	0.01	19 (16.8)	42 (15.2)	0.68
Hydralazine	0 (0)	5 (1.1)	0.31	1 (0.9)	1 (0.4)	0.51
Number of antihypertensive medications			0.80			0.12
1	87 (92.6)	423 (91.8)		102 (90.3)	262 (94.6)
>1	7 (7.5)	38 (8.2)		11 (9.7)	15 (5.4)

Data are mean (standard deviation) number (frequency).

^aMaternal comorbidities include autoimmune, cardiovascular, hematological, neurological disorders.

Maternal and neonatal outcomes of planned delivery between 37^0/7 to 37^6/7weeks compared to expectant management beyond 37 weeks are shown in Table 2, and outcomes of planned delivery between 38^0/7 to 38^6/7weeks compared to expectant management beyond 38 weeks are shown in Table 2. Of the patients who had a planned delivery between 37^0/7–37^6/7weeks, the primary neonatal adverse outcome occurred in 14 (14.9%) patients compared to 48 (10.4%) patients in the expectant management group. On logistical regression, these differences were not significant (aOR 1.49, 95% CI: 0.77–2.88). Among patients who underwent a planned delivery at 38^0/7–38^6/7weeks, the primary neonatal adverse outcome occurred in 11 (9.7%) compared to 28 (10.1%) of patients in the expectant group, which was also not found to be significant on logistical regression modeling (aOR 0.84, 95% CI: 0.39–1.78). The rates of individual neonatal outcomes are shown on Table 2.

Table 2. Outcomes of planned delivery by gestational age for patients with chronic hypertension on antihypertensive medications.

Outcomes of Planned Delivery 37^0/7 to 37^6/7 weeks Gestation versus Expectant Management beyond 37 weeks
	Planned delivery 37.0–37.6 weeks n = 94	Expectant delivery beyond 37.0 weeks n = 461	aOR with 95% CI
Primary Outcomes	14 (14.8)	48 (10.4)	1.49 (0.77, 2.88)
Ph < 7.0	1 (0.2)	0 (0)
Respiratory Disorder	13 (13.8)	37 (8.0)
Mechanical Vent	2 (2.1)	3 (0.6)
Non-Invasive Resp Support	8 (8.5)	39 (8.5)
Apgar <5	2 (2.1)	4 (0.9)
Neonatal Death	0 (0)	0 (0)
Seizure	0 (0)	0 (0)
Primary C-Section	30 (31.9)	116 (25.1)	1.45 (0.87, 2.41)
Preeclampsia	30 (31.9)	146 (31.7)	0.92 (0.56, 1.52)
Eclampsia	0 (0)	0 (0)	n/a
Postpartum Readmit	11 (11.7)	34 (7.4)	1.53 (0.74, 3.19)
Neonatal Days > =5	8 (8.5)	32 (6.9)	1.25 (0.55, 2.86)
Outcome of Planned Delivery 38^0/7 to 38^6/7 weeks Gestation versus Expectant Management beyond 38 weeks
	Planned delivery 38–38.7 weeks n = 113	Expectant delivery beyond 38 weeks n = 277	aOR with 95% CI
Primary Outcomes	11 (9.7)	28 (10.1)	0.84 (0.39, 1.78)
Ph < 7.0	0 (0)	0 (0)
Respiratory Disorder	7 (6.2)	23 (8.3)
Mechanical Vent	1 (0.9)	1 (0.4)
Non-Invasive Resp Support	11 (9.7)	21 (7.6)
Apgar <5	1 (0.9)	0 (0)
Neonatal Death	0 (0)	0 (0)
Seizure	0 (0)	0 (0)
Primary C-Section	28 (24.8)	64 (23.1)	0.98 (0.58, 1.66)
Preeclampsia	28 (24.8)	75 (27.1)	0.83 (0.50, 1.40)
Eclampsia	0 (0)	0 (0)	n/a
Postpartum Readmit	5 (4.4)	22 (7.9)	0.54 (0.20, 1.46)
Neonatal Days > =5	4 (3.5)	19 (6.9)	0.48 (0.16, 1.46)

OR: Odds Ratio.

Patients who underwent planned delivery at 37^0/7–37^6/7weeks had similar primary cesarean rates compared to those expectantly managed (31.9% vs 31.6%). There was also no difference in primary cesarean rates among patients who underwent planned delivery at 38^0/7–38^6/7weeks compared to those expectantly managed (24.9% vs 23.1%).

There were no differences in preeclampsia rates among patients who had planned delivery at 37^0/7–37^6/7week versus expectant management (31.9% vs 31.7%) as well planned at delivery 38^0/7–38^6/7week versus expectant management (24.7% vs 27.1%). There were also no differences in other secondary outcomes including eclampsia, postpartum readmission, and infant hospital stay greater than 5 days between the two groups.

Discussion

For pregnant patients with chronic hypertension on antihypertensive medications, we did not find a difference in maternal or neonatal outcomes when comparing planned delivery at 37 weeks and 38 weeks versus expectant management. Thus, we were unable to identify an optimal week of gestational age which time such patients should be delivered.

Our findings suggest that planned delivery at 37 and 38 weeks’ gestations when compared to expectant management did not impact the rate of adverse perinatal or maternal outcomes. Our findings were similar to Harper et al. [9], a study evaluating chronic hypertensive patients managed both with and without antihypertensive therapy. The authors did not detect a difference in neonatal adverse outcomes or rates of preeclampsia in 37 to 38 weeks gestation. Our findings were also similar to Ram et al. [10], who found that routine induction at 38 and 39 weeks in patients with chronic hypertension was not associated with increased rates of primary cesarean. Instead, the authors found decreased rate of superimposed preeclampsia. Differences in patient population may be responsible for this contrasting finding.

At the time the study was conducted, ACOG recommended maintaining blood pressure <160/110 mm Hg to decrease the overall rate of poor maternal outcomes and protect against sequela of hypertension, including stroke and pulmonary edema [15]. Pharmacological intervention in this study were made based on provider discretion. Antihypertensive medication was likely initiated and adjusted to maintain blood pressures less than 160/110 mm Hg. Regardless of blood pressure control, it can be presumed that chronic hypertensive patients who require treatment inherently have more significant disease, and can reasonably be considered a different patient population than those who have chronic hypertension but are not treated.

Our results were similar to Harper et al. who found increased rate of preeclampsia after 39 weeks, but no differences in preeclampsia rates among patients between 37–39 weeks [9]. Though dissimilarity in patient population exists, our findings illustrate that requiring therapeutic antihypertensive management may not increase the risk of developing preeclampsia. Initial findings did not show evidence of lower rates of preeclampsia if chronic hypertension was treated, though these studies were limited [16,17]. However, a recent retrospective cohort study did show that maintenance of blood pressures below 140/90 with antihypertensive medications decreased the rate of developing preeclampsia [18]. This was further confirmed by a randomized control trial which found a target blood pressure below 140/90 led to a lower rate of adverse maternal outcomes, including preeclampsia, without an increased risk of severe neonatal complications [19]. Based on these results, the Society of Maternal Fetal Medicine (SMFM) and ACOG recently updated recommendations to maintain blood pressure in chronic hypertensive patients <140/90 mm Hg [20]. Though our study did not examine blood pressure control, differences in preeclampsia rate may not have been detected among the cohorts because treatment with antihypertensive medications had impacted overall preeclampsia rates.

Overall, neonatal adverse outcomes were not different if patients had planned delivery at 37 or 38 weeks gestation. There were also lower rates of steroid use in the planned 37 week delivery group, but this did not appear to impact outcomes. Based on our limited sample size, we cannot exclude the possibility of a type II error. Future studies should attempt to include a larger number of subjects to determine if planned delivery at 37 weeks may be associated with adverse outcomes, and determine if differences in steroid use impacts outcomes.

This is the first study to evaluate whether the risk of adverse outcomes for pregnant patients with chronic hypertension on antihypertensive medication is impacted by timing of delivery. An additional strength of this study is the inclusion of data from multiple hospitals within a large health system, each serving a diverse patient population. In contrast to prior population cohort studies, this study utilized patient-level data obtained from a shared system-wide EMR system. This facilitated identification and exclusion of certain maternal and fetal complications, including fetal growth restriction, thus allowing for evaluation of chronic hypertension pregnancy outcomes in the absence of concurrent conditions [12].

One limitation of our study is that given its retrospective nature, it does not allow for causal inferences. A second limitation is the overall small sample size, which may not have allowed for adequate detection of differences between groups as only a limited number of patients with chronic hypertension on medication were delivered during our study period. A third limitation is that groups were based on gestational age by weeks, however, differences may have been better detected if groups were based on gestational days (e.g. 37^6/7 vs 38^0/7). Due to the number of patients in our study, this was not possible. Future studies should be conducted in a prospective manner, powered with a larger sample size.

Our study also did not have details available regarding blood pressure control of patients at the time of delivery. Future studies should stratify patients based on blood pressure control with stable hypertension on low-to-moderate dose antihypertensive medication compared to those who have difficult-to-control hypertension requiring up-titration of medication. This comparison, along with the recent change in recommendations, could provide more information if planned delivery at a specific gestational age would optimize outcomes.

In conclusion, there is no evidence of a difference in neonatal or maternal outcomes, including preeclampsia rate and primary cesarean, for patients with chronic hypertension on antihypertensive medications, when planning delivery at 37 or 38 weeks’ gestation compared to expectant management. While at this time it is not unreasonable to deliver chronic hypertensive patients on medications at or after 37 weeks, more prospective data is needed to further clarify the optimal gestational age of delivery for these patients. Our data suggests that adverse maternal or perinatal outcomes is not impacted if delivery is planned at 37 or 38 weeks’ gestation or expectantly managed.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.