Physiologic Treatment of Severe Hypertension in Pregnancy and Postpartum

Abstract

We aimed to evaluate physiologic treatment of severe hypertension. This was a retrospective cohort study of pregnant and postpartum patients with severe blood pressure (systolic blood pressure ≥160 mmHg or diastolic blood pressure ≥110 mmHg) treated with intravenous labetalol or hydralazine at a single tertiary care center from 2013 to 2018. Patients were classified as having physiologic treatment if they had hyperdynamic physiology (pulse pressure≥65 mmHg) and received labetalol or had vasoconstrictive physiology (diastolic BP≥100 mmHg) and received hydralazine. The primary outcome was number of antihypertensive doses to achieve non-severe blood pressure. Of 1,120 patients included in the analysis, 653 had physiologic treatment and 467 non-physiologic treatment with 16 (1.4%) excluded for inability to classify physiology. Physiologic treatment was associated with fewer antihypertensive doses (1.4±0.9 doses vs 1.6±1.4 doses; aβ −0.28, 95% CI −0.42 to −0.14) and lower odds of medication conversion (2.5% vs 4.7%; aOR 0.48, 95% CI 0.24–0.93), but no difference in time to non-severe BP (31 minutes, IQR 16–66 minutes vs 34 minutes, IQR 15–76 minutes; aHR 1.0, 95% CI 0.9–1.2). Physiologic treatment of severe hypertension warrants further evaluation.

Précis:

Treatment of acute severe hypertension based on hyperdynamic versus vasoconstrictive physiology was associated with fewer antihypertensive doses and less frequent conversion to alternate medication to achieve non-severe hypertension.

INTRODUCTION

Hypertension in pregnancy is a major cause of maternal and neonatal morbidity and mortality.[1–3] Specifically, severe hypertension, defined as systolic blood pressure (SBP) ≥160 mmHg or diastolic blood pressure (DBP) ≥110 mmHg[2–3], can lead to maternal stroke, myocardial infarction, renal injury, abruption, and fetal death. The American College of Obstetricians and Gynecologists (ACOG) recommends treatment of severe hypertension within 30–60 minutes with intravenous (IV) labetalol, IV hydralazine, or oral immediate release nifedipine.[3]

A Cochrane review did not find significant differences regarding efficacy or safety between these antihypertensive medications [4], but it is possible they have differing effectiveness based on underlying physiology and medication mechanism of action.[5–7] Labetalol acts by inhibiting adrenergic stimulation of endothelial cell function,[8] theoretically making it more useful for a hyperdynamic state, whereas hydralazine acts through vasodilation,[9] potentially making it more effective for a vasoconstricted state. Our objective was to evaluate the association between treatment of severe acute hypertension based on hyperdynamic versus vasoconstrictive physiology and BP control.

METHODS

This was a retrospective cohort study of pregnant and postpartum patients (within six weeks) with severe BP treated with IV labetalol or hydralazine (based on baseline physiology) at a single tertiary care center (2013–2018). Under Institutional Review Board approval (IRB 300007245), patient data from Cerner-based electronic medical records were extracted. Those with a fetal death prior to severe BP, whose physiology could not be classified, or who never had a non-severe BP were excluded. We only considered the first episode of severe BP in the first pregnancy with data available for each patient.

BP physiology was characterized by pulse pressure (PP=SBP-DBP) and DBP with physiologic treatment defined based on BP physiology and antihypertensive medication received. Patients with PP ≥65 mmHg were considered to have hyperdynamic physiology and classified as having physiologic treatment if they received labetalol. Patients with DBP ≥100 mmHg were considered to have vasoconstrictive physiology and were classified as having physiologic treatment if they received hydralazine. Patients with both high PP and DBP were classified as having vasoconstrictive physiology in the primary analysis, and the opposite assumption was made in a sensitivity analysis.

The primary outcome was number of antihypertensive doses received before first non-severe BP. Secondary outcomes included conversion to alternative medication, time to first non-severe BP, and complications including maternal hypotension (defined as BP <90/60 within one hour of IV labetalol or hydralazine administration), placental abruption, non-reassuring fetal status requiring delivery, and maternal intensive care unit admission.

Multivariable linear, logistic and Cox proportional hazard regression estimated the association between physiologic treatment and outcomes with adjustment for covariates and testing for effect modification by chronic hypertension.

RESULTS

Of 1,120 included patients, 653 (58.3%) received physiologic treatment and 467 (41.7%) non-physiologic treatment (Appendix 1, available online at http://links.lww.com/xxx). Notably, only 16 (1.4%) were excluded for inability to classify BP physiology. Compared with individuals who received non-physiologic treatment, those who received physiologic treatment were older, had higher body mass index, and were more likely to have pre-gestational diabetes (Table 1). There were no significant differences in race and ethnicity, preeclampsia, magnesium prophylaxis or postpartum status (Table 1).

Table 1:

Baseline Characteristics by Physiologic and Non-Physiologic Treatment of Severe Hypertension in Pregnancy and Postpartum

	Physiologic (n = 653)	Non-Physiologic (n = 467)	p-value
Demographic, medical and obstetric characteristics
Maternal age (years)	29.2 ± 6.3	28.2 ± 6.4	0.02
Race			0.20
Black	395 (60.5)	274 (58.7)
White	219 (33.5)	152 (32.6)
None of the above	39 (6.0)	41 (8.8)
Hispanic Ethnicity	31 (4.8)	30 (6.4)	0.22
Maternal body mass index (kg/m2)	35.8 ± 9.7	33.3 ± 9.5	<0.001
Nulliparous	287 (44.0)	211 (45.2)	0.68
Chronic hypertension	276 (42.3)	170 (36.4)	0.05
Oral antihypertensive medication use before severe blood pressure	206 (31.6)	140 (30.0)	0.58
Labetalol	153 (23.4)	110 (23.6)	0.96
Nifedipine	45 (6.9)	29 (6.2)	0.65
Amlodipine	45 (6.9)	25 (5.4)	0.29
Hydrochlorothiazide	45 (6.9)	21 (4.5)	0.09
Preeclampsia	605 (92.7)	424 (90.8)	0.26
IV magnesium prophylaxis	620 (95.0)	439 (94.0)	0.49
Pre-gestational diabetes	114 (17.5)	33 (7.1)	<0.001
Tobacco use	84 (12.9)	66 (14.1)	0.54
Characteristics at time of severe-range BP
Postpartum	128 (19.6)	98 (21.0)	0.57
Gestational age (weeks) if pregnant	32.2 ± 6.5	31.6 ± 7.9	0.21

Data presented as mean ± standard deviation or n (%)

Chi-square and student’s t-test used for comparisons

Physiologic treatment was associated with fewer antihypertensive doses to achieve non-severe BP (1.4±0.9 doses vs 1.6±1.4 doses; aβ −0.28, 95% CI −0.42 to −0.14, Table 2). Chronic hypertension did not modify this association (interaction p=0.86). Findings were consistent when individuals who met both BP criteria (n=152) were re-classified as having hyperdynamic physiology in a sensitivity analysis. Physiologic treatment compared with non-physiologic treatment was associated with lower odds of switching medications (2.5% vs 4.7%; aOR 0.48, 95% CI 0.24–0.93). The time to achieve a non-severe BP was similar between groups (physiologic: 31 minutes, IQR 16–66 minutes vs non-physiologic: 34, IQR 15–76 minutes; aHR 1.0, 95% CI 0.9–1.2). There were no differences in other maternal or fetal complications.

Table 2:

Association between Physiologic Treatment of Severe Hypertension and Maternal and Neonatal Outcomes

	Physiologic (n = 653)	Non-physiologic (n = 467)	β (95% CI)	aβ (95% CI)*
Number of antihypertensive doses†	1.4 ± 0.9	1.6 ± 1.4	−0.27 (−0.41 to −0.14)	−0.28 (−0.42 to −0.14)
Gestational age at delivery (weeks)	33.4 ± 4.4	33.2 ± 4.6	0.28 (−0.31 to 0.87)	0.28 (−0.33 to 0.88)
			OR (95% CI)	aOR (95% CI)*
Change to alternative medication†	16 (2.5)	22 (4.7)	0.51 (0.26–0.98)	0.48 (0.24–0.93)
Hypotension <1 hour after treatment	1 (0.2)	1 (0.2)	0.72 (0.05–11.45)	0.63 (0.04–11.28)
Placental abruption‡	14 (2.1)	7 (1.5)	1.44 (0.58–3.59)	1.62 (0.64–4.08)
Maternal ICU admission‡	17 (2.6)	12 (2.6)	1.01 (0.48–2.14)	0.94 (0.44–2.02)
Cesarean delivery for non-reassuring fetal status‡§	94 (14.4)	67 (14.4)	1.00 (0.72–1.41)	0.90 (0.63–1.27)

Data presented as mean ± standard deviation or n (%) unless otherwise noted

^*Adjusted for age, BMI, and pregestational diabetes. Referent=non-physiologic treatment.

^†Before achieving non-severe BP

^‡During pregnancy

^§Noted if delivery by cesarean was indicated for non-reassuring fetal status

DISCUSSION

Physiologic treatment was associated with fewer antihypertensive doses to achieve non-severe BP and a lower likelihood of conversion to alternative medication. The time to non-severe BP was similar between treatment approaches and there was no difference in complications.

Our findings extend current knowledge regarding use of IV antihypertensive medications for treatment of severe BP in pregnancy and postpartum. While the 2013 Cochrane review did not find any significant differences in the efficacy of labetalol or hydralazine for treatment of severe BP in an unselected population,[4] our findings suggest labetalol may be beneficial for hyperdynamic physiology and hydralazine for vasoconstrictive physiology. However, given the retrospective study design, we are limited by potential for unmeasured confounding and cannot prove causality. This analysis did not address oral nifedipine use, recurrent episodes of severe BP, and may not be generalizable to all populations. Although treatment based on physiology was associated with fewer anti-hypertensive doses, the differences are modest and may not be clinically significant. Antihypertensive treatment of severe BP based on physiology should be evaluated further in larger studies.