OriginalInvestigation | ObstetricsandGynecology ......Systolic blood pressure, after week 3, mean...

Original Investigation | Obstetrics and Gynecology

Racial Differences in Postpartum Blood Pressure Trajectories Among WomenAfter a Hypertensive Disorder of PregnancyAlisse Hauspurg, MD; Lara Lemon, PhD, PharmD; Camila Cabrera, MD; Amal Javaid, BS; Anna Binstock, MD; Beth Quinn, RN; Jacob Larkin, MD; Andrew R. Watson, MD;Richard H. Beigi, MD; Hyagriv Simhan, MD

Abstract

IMPORTANCE Maternal morbidity and mortality are increasing in the United States, most of whichoccur post partum, with significant racial disparities, particularly associated with hypertensivedisorders of pregnancy. Blood pressure trajectory after a hypertensive disorder of pregnancy has notbeen previously described.

OBJECTIVES To describe the blood pressure trajectory in the first 6 weeks post partum after ahypertensive disorder of pregnancy and to evaluate whether blood pressure trajectories differ by self-reported race.

DESIGN, SETTING, AND PARTICIPANTS This prospective cohort study included deliveries betweenJanuary 1, 2018, and December 31, 2019. Women with a clinical diagnosis of a hypertensive disorderof pregnancy were enrolled in a postpartum remote blood pressure monitoring program at the timeof delivery and were followed up for 6 weeks. Statistical analysis was performed from April 6 to17, 2020.

MAIN OUTCOMES AND MEASURES Mixed-effects regression models were used to display bloodpressure trajectories in the first 6 weeks post partum.

RESULTS A total of 1077 women were included (mean [SD] age, 30.2 [5.6] years; 804 of 1017 White[79.1%] and 213 of 1017 Black [20.9%]). Systolic and diastolic blood pressures were found to decreaserapidly in the first 3 weeks post partum, with subsequent stabilization (at 6 days post partum: mean[SD] peak systolic blood pressure, 146 [13] mm Hg; mean [SD] peak diastolic blood pressure, 95 [10]mm Hg; and at 3 weeks post partum: mean [SD] peak systolic blood pressure, 130 [12] mm Hg; mean[SD] peak diastolic blood pressure, 85 [9] mm Hg). A significant difference was seen in bloodpressure trajectory by race, with both systolic and diastolic blood pressure decreasing more slowlyamong Black women compared with White women (mean [SD] peak systolic blood pressure at 1week post partum: White women, 143 [14] mm Hg vs Black women, 146 [13] mm Hg; P = .01; mean[SD] peak diastolic blood pressure at 1 week post partum: White women, 92 [9] mm Hg vs Blackwomen, 94 [9] mm Hg; P = .02; and mean [SD] peak systolic blood pressure at 3 weeks post partum:White women, 129 [11] mm Hg vs Black women, 136 [15] mm Hg; P < .001; mean [SD] peak diastolicblood pressure at 3 weeks post partum: White women, 84 [8] mm Hg vs Black women, 91 [13] mmHg; P < .001). At the conclusion of the program, 126 of 185 Black women (68.1%) compared with 393of 764 White women (51.4%) met the criteria for stage 1 or stage 2 hypertension (P < .001).

CONCLUSIONS AND RELEVANCE This study found that, in the postpartum period, blood pressuredecreased rapidly in the first 3 weeks and subsequently stabilized. The study also found that,compared with White women, Black women had a less rapid decrease in blood pressure, resulting inhigher blood pressure by the end of a 6-week program. Given the number of women with persistent

(continued)

Key PointsQuestion Does postpartum blood

pressure trajectory after a hypertensive

disorder of pregnancy differ by race?

Findings In this prospective cohort

study that included 1077 women after a

hypertensive disorder of pregnancy,

blood pressure trajectories evaluated

using mixed-effects linear regression

models differed significantly by self-

reported race. At the conclusion of the

study, 68% of Black women and 51% of

White women met the criteria for stage

1 or stage 2 hypertension.

Meaning This study suggests that

postpartum blood pressure trajectories

indicate persistence of higher blood

pressures among Black women in this

cohort, which may have important

implications for postpartum morbidity

and mortality associated with

hypertensive and cardiovascular

conditions in this population.

+ Invited Commentary

+ Supplemental content

Author affiliations and article information arelisted at the end of this article.

Open Access. This is an open access article distributed under the terms of the CC-BY License.

JAMA Network Open. 2020;3(12):e2030815. doi:10.1001/jamanetworkopen.2020.30815 (Reprinted) December 22, 2020 1/12

Downloaded From: https://jamanetwork.com/ by a Non-Human Traffic (NHT) User on 08/01/2021

https://jama.jamanetwork.com/article.aspx?doi=10.1001/jamanetworkopen.2020.31122&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2020.30815


Abstract (continued)

hypertension at the conclusion of the program, these findings also appear to support the importanceof ongoing postpartum care beyond the first 6 weeks after delivery.

JAMA Network Open. 2020;3(12):e2030815. doi:10.1001/jamanetworkopen.2020.30815

Introduction

Recent evidence suggests that maternal morbidity and mortality are increasing in the United States,most of which occur in the 6 weeks after delivery, referred to as the “fourth trimester.”1 The cause ofthis increase is likely multifactorial, associated with advancing maternal age and medicalcomorbidities as well as the concomitant limitations of the health care system. Recent data from theCenters for Disease Control and Prevention2 demonstrate that Black women are 3 to 4 times morelikely to die in childbirth compared with women of other races/ethnicities, a finding that persistsacross socioeconomic strata and is thought to be associated with social determinants of health aswell as implicit and explicit biases within the health care system, resulting in inequitabletreatment.1,3,4

Hypertension complicates 10% to 20% of pregnancies in the United States and is significantlyassociated with maternal morbidity and mortality in the postpartum period.1,2,5 Black women are atincreased risk of hypertensive disorders of pregnancy, and hypertension and cardiovascular diseasesare more frequently associated with morbidity and mortality among Black women compared withWhite women.1,3 These differences are particularly notable in the postpartum period. Despite thisfinding, prior studies and clinical management guidelines focus overwhelmingly on antepartum andintrapartum management, with relatively little emphasis placed on postpartum management.6 Onereason for the lack of management guidelines is our limited understanding of the postpartum bloodpressure trajectory after delivery and hospital discharge. At present in the United States, women aretypically discharged from the hospital on postpartum days 2 to 4 and the American College ofObstetricians and Gynecologists (ACOG) recommends a single blood pressure check between 3 and10 days post partum for women with a hypertensive disorder of pregnancy.6,7 Women withpersistent hypertension or the need for titration of antihypertensive medications are typically seenmore frequently in the postpartum period for medication management; however, as there are noclear guidelines on optimal blood pressure management in this period, this varies by clinician andinstitution.8 Subsequently, women are typically seen at 4 to 6 weeks post partum for acomprehensive postpartum visit and referred to their primary care physician if there are additionalneeds for antihypertensive medication management.6 This strategy is limited by poor adherence tofollow-up, with prior studies showing visit attendance rates of 45% to 60% in this period.7,9

The ACOG and others have suggested the use of telemedicine or remote health careinterventions to facilitate care in the fourth trimester.7 A previous study has demonstrated feasibility,high engagement, retention, and patient satisfaction with a postpartum hypertension remotemonitoring program as well as improved adherence with postpartum visits.10 Given poor adherenceto recommended follow-up in the postpartum period and resultant lack of longitudinally collectedserial blood pressure measurements, the trajectory of a decrease in blood pressure in the postpartumperiod has not been previously well studied, to our knowledge. Prior work has focused on smallhistoric cohorts or a predominantly White population.11-13 Therefore, the aims of this prospectivecohort study were 2-fold. The first aim was to describe the trajectory of postpartum blood pressureafter a hypertensive disorder of pregnancy using remote blood pressure monitoring in a large,contemporaneous cohort. The second aim, given the disparities in maternal morbidity and mortalityassociated with hypertensive disorders in the postpartum period, was to investigate differences inpostpartum blood pressure trajectory by race.

JAMA Network Open | Obstetrics and Gynecology Racial Differences in Blood Pressure After a Hypertensive Disorder of Pregnancy




Methods

This is an ongoing quality improvement project that included women admitted to the postpartumunit of a single tertiary care hospital (University of Pittsburgh Medical Center [UPMC] Magee-Womens Hospital) between January 1, 2018, and December 31, 2019. Eligible women had 1 of thefollowing hypertension-related diagnoses: gestational hypertension, preeclampsia, eclampsia, ornew-onset postpartum hypertension. Women with prepregnancy chronic hypertension wereexcluded. To be included in the program, women must have been English-speaking and have accessto a text messaging–enabled mobile device. Diagnoses were made by the clinical care team accordingto ACOG criteria.6 Regardless of diagnosis, hypertension in pregnancy was defined as blood pressureof 140 mm Hg or more systolic or 90 mm Hg or more diastolic. Maternal, obstetric, andsociodemographic data were obtained from the electronic medical record and subsequently theClinical Data Warehouse at UPMC. The program was approved by the UPMC Quality ImprovementReview Committee and the University of Pittsburgh institutional review board, which waived arequirement for informed consent because the data were deidentified. The reporting followed theStrengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelinefor cohort studies.

A description of the remote monitoring program has been previously described.10 In brief, wecreated a remote patient monitoring platform that used Vivify Health as its core vendor. Themonitoring platform was integrated with electronic health records and the Clinical Data Warehousefor both ordering and results. Patients were enrolled in the program by their primary obstetricclinician while admitted as inpatients on the postpartum unit or after a postpartum readmission.After identification and verification of eligibility, the clinician placed an order in the electronic medicalrecord with the patient’s telephone number, which automatically generated a text message to enrollthe patient. We used the A&D UA-651 (A&D Medical) automatic upper arm blood pressure monitor.The patient was trained on the use of the blood pressure device by a program-specific nurse educatorprior to discharge from the hospital. During program enrollment, bedside nursing staff recordedblood pressure on both the home blood pressure monitoring device and the hospital device toconfirm accuracy. The program was managed through a nursing-staffed UPMC call center withdocumentation of telephone calls and blood pressures directly into the electronic medical record.

We designed nursing call center–driven blood pressure management and treatment algorithmsthat were developed by local expert stakeholders, consistent with national guidelines on goals forpostpartum hypertension management.14 After discharge from the hospital, women were promptedto check their blood pressure 5 days per week for the first week of the program and between 3 and5 times per week for the remainder of the program through 6 weeks post partum. Blood pressureswere reported using text messaging. The initial choice of antihypertensive agent was dictated by theclinical care team while the patient was an inpatient. After hospital discharge, titration of medicationor, in the case of medication initiation, selection of the antihypertensive agent was based on clinicaljudgment from the call center physician. There are currently no standardized management guidelinesfor specific antihypertensive agents or parameters for medication titration in the postpartumperiod.8 Women with symptoms including chest pain, severe headache, blurry vision or visiondisturbances, shortness of breath, or blood pressures of 180 mm Hg or more systolic or 110 mm Hgor more diastolic were referred to the emergency department for further evaluation. Discontinuationor downward titration of antihypertensive therapy is based on patients having 3 consecutive bloodpressures less than 120/70 mm Hg.

Statistical AnalysisFor the first objective, mixed-effects regression models were used to display blood pressuretrajectories in the first 6 weeks post partum, with weeks post partum as the timescale for all analyses.We used repeated blood pressure measurements to fit mixed-effects linear regression models witheach blood pressure measurement as the outcome, participant identifier as random intercepts, and




https://www.equator-network.org/reporting-guidelines/strobe/

weeks post partum as a fixed effect expressed using restricted cubic splines with 4 knots positionedat 0.6, 1.9, 3.6, and 5.9 weeks. The optimal number of knots was chosen comparing both Akaikeinformation criterion and bayesian information criterion between models. Mixed-effects regressionmodels were further adjusted for predefined covariates known to be associated with blood pressure,including prepregnancy body mass index (BMI; calculated as weight in kilograms divided by heightin meters squared), tobacco use, type of hypertensive disorder, and use of antihypertensivemedication.

For the second objective, we repeated all models stratified by race (White individuals vs Blackindividuals). Stratified mixed-effects regression models were again adjusted for the same covariates.The number and position of splines were not different by race. In addition, in sensitivity analyses werepeated all analyses including gestational weight gain and BMI at the time of delivery. Differencesbetween races in the associations between blood pressure and weeks post partum were tested vialikelihood ratio test between models with and without parameters representing the interactionbetween race and prepregnancy BMI, BMI at delivery, and gestational weight gain based on resultsof prior studies.13,15 In sensitivity analyses, we repeated all regressions stratified by type ofhypertensive disorder, use of antihypertensive medication, and obesity, as each could modify theassociation.

Statistical analysis was performed from April 6 to 17, 2020. All analyses were performed usingStata IC, version 16 software package (StataCorp LP). All P values were from 2-sided tests, and resultswere deemed statistically significant at P < .05.

Results

A total of 1114 women (mean [SD] age, 30.1 [5.6] years; and mean [SD] BMI, 29.6 [7.9]) were enrolledin the program between January 1, 2018, and December 31, 2019; 37 women reported less than 2blood pressure measurements and were excluded, leaving 1077 in the analytic sample. For the firstobjective, we included all 1077 women, who contributed 17 146 blood pressure measurements acrossthe first 6 weeks post partum. Overall, women contributed a mean (SD) of 20.3 (7.1) blood pressurevalues during the program. The demographic and delivery characteristics of the overall cohort areshown in eTable 1 and eTable 2 in the Supplement. Overall, in our cohort, 447 women (41.5%) hadgestational hypertension, and 630 women (58.5%) had preeclampsia, and of these 1077 women, 315(29.2%) had preeclampsia with severe features.

Figure 1 displays fitted blood pressure values for all participants during the first 6 weeks postpartum. The mean (SD) peak systolic blood pressure for the cohort was 146 (13) mm Hg at a medianof 6 days post partum (interquartile range [IQR], 4-9 days). Mean (SD) peak diastolic blood pressureat a median of 6 days post partum (IQR, 4-11 days) was of 95 (10) mm Hg. We observed a rapiddecrease in blood pressure in the first 3 weeks post partum (mean [SD] peak systolic blood pressure,

Figure 1. Fitted Blood Pressure (BP) Trajectory in First 6 Weeks Post Partum

Weeks post partum

150

140

130

120

110

Syst

olic

BP,

mm

Hg

Systolic BP trajectoryA

0 1 2 3 4 5 6

Weeks post partum

95

90

85

80

75

Dias

tolic

BP,

mm

Hg

Diastolic BP trajectoryB

0 1 2 3 4 5 6

Shaded area indicates 95% CI.





130 [12] mm Hg; and mean [SD] peak diastolic blood pressure, 85 [9] mm Hg). Subsequently, bloodpressures stabilized for the remaining 3 weeks of the program. Overall among women with apostpartum visit blood pressure value available or who reported a blood pressure value after thethird week post partum (n = 1007 [93.5%]), 543 women (53.9%) met criteria for stage 1hypertension and 200 women (19.9%) met criteria for stage 2 hypertension at the conclusion ofthe program.

For our second objective, we excluded 60 women of non-White or non-Black race. Comparedwith White women (n = 804), Black women (n = 213) were younger (mean [SD] age, 28.1 [6.1] vs 30.6[5.4] years) and were more likely to have public health insurance (139 [65.3%] vs 209 [26.0%]), bemultiparous (107 [50.2%] vs 303 [37.7%]), and to have lower mean (SD) diastolic blood pressure atthe initiation of prenatal care (70 [9] vs 74 [8] mm Hg) (Table 1). Infants born to Black women hada lower mean (SD) birth weight than those born to White women (2849 [800] vs 2966 [743] g)(Table 2). There were no differences by race in gestational age at delivery. There were also nodifferences by race in the highest blood pressure in the 24 hours prior to hospital discharge. Blackwomen had higher peak systolic and diastolic blood pressures compared with White women (mean[SD] maximum systolic blood pressure, 150 [14] vs 145 [13] mm Hg; P < .001; and mean [SD]maximum diastolic blood pressure, 98 [12] vs 94 [9] mm Hg; P < .001). In addition, Black womenreached peak blood pressures at slightly later time points post partum compared with White women(median time of maximum systolic blood pressure, 8 days [IQR, 5-15 days] vs 7 days [IQR, 4-10 days];and median time of maximum diastolic blood pressure, 9 days [IQR, 5-18 days] vs 7 days [IQR, 5-12days]). As shown in Figure 2, both systolic and diastolic blood pressure decreased more slowly in thefirst 6 weeks post partum among Black women compared with White women (mean [SD] peaksystolic blood pressure at 1 week post partum: White women, 143 [14] mm Hg vs Black women, 146[13] mm Hg; P = .01; mean [SD] peak diastolic blood pressure at 1 week post partum: White women,92 [9] mm Hg vs Black women, 94 [9] mm Hg; P = .02; and mean [SD] peak systolic blood pressureat 3 weeks post partum: White women, 129 [11] mm Hg vs Black women, 136 [15] mm Hg; P < .001;mean [SD] peak diastolic blood pressure at 3 weeks post partum: White women, 84 [8] mm Hg vsBlack women, 91 [13] mm Hg; P < .001). In analyses adjusting for clinical covariates (includingprepregnancy BMI, tobacco use, type of hypertensive disorder, and use of antihypertensivemedication), our findings were unchanged. We tested for interaction between race andprepregnancy BMI and found that the interaction term was not statistically significant. In addition,we tested for interaction between race and type of hypertensive disorder, preterm hypertensivedisorder, and parity and found no evidence of an interaction. Multivariable-adjusted blood pressure

Table 1. Demographic Characteristics by Self-reported Race

Characteristic White (n = 804) Black (n = 213) P valueAge, mean (SD), y 30.6 (5.4) 28.1 (6.1) <.001

Prepregnancy BMI, mean (SD) 29.4 (7.8) 30.6 (8.5) .07

Delivery BMI, mean (SD) 34.5 (6.8) 35.4 (7.5) <.001

Gestational weight gain, mean (SD), kg 14.2 (10.4) 12.2 (12.1) <.001

Gestational age at prenatal care establishment, mean (SD), wk 11.1 (6.3) 12.5 (6.6) .007

First prenatal systolic BP, mean (SD), mm Hg 118 (11) 119 (10) .12

First prenatal diastolic BP, mean (SD), mm Hg 74 (8) 70 (9) <.001

Insurance status, No. (%)

Private insurance 579 (72.0) 71 (33.3)

<.001Public insurance 209 (26.0) 139 (65.3)

Other 16 (2.0) 3 (1.4)

Primiparous, No. (%) 501 (62.3) 106 (49.8) <.001

Current tobacco use, No. (%) 81 (10.1) 33 (15.5) .02

Pregestational diabetes, No. (%) 24 (3.0) 11 (5.2) .12

Gestational diabetes, No. (%) 82 (10.2) 22 (10.3) .35

Abbreviations: BMI, body mass index (calculated asweight in kilograms divided by height in meterssquared); BP, blood pressure.




trajectories remained higher in Black women compared with White women during the first 6 weekspost partum (likelihood ratio test χ2 = 36.9 for systolic blood pressure; and χ2 = 42.8 for diastolicblood pressure; P < .001 for all). At the conclusion of the program, mean blood pressure was higheramong Black women compared with White women (mean [SD] systolic blood pressure, 131 [14] vs122 [11] mm Hg; P < .001; and mean [SD] diastolic blood pressure, 84 [12] vs 79 [9] mm Hg; P < .001).Similarly, Black women were significantly more likely than White women to meet criteria for stage 1or stage 2 hypertension at the conclusion of the program (126 of 185 [68.1%] vs 393 of 764 [51.4%];P < .001) (Table 3).

Sensitivity analyses were stratified by antihypertensive medication use, type of hypertensivedisorder, and obesity. We observed similar findings in both the unadjusted and adjusted models(eFigures 1-4 in the Supplement).

Finally, to understand race-specific blood pressure trajectories in context, we examined theincidence of hospital readmission for hypertension in the postpartum period and observed the

Table 2. Delivery and Postpartum Characteristics by Self-reported Race

Characteristic White (n = 804) Black (n = 213) P valueType of hypertension, No. (%)

Gestational hypertension 350 (43.5) 81 (38.0) .15

Preeclampsia 454 (56.5) 132 (62.0)

Cesarean delivery, No. (%) 343 (42.7) 101 (47.4) .21

Birth weight, mean (SD), g 2966 (743) 2849 (800) .05

Gestational age at delivery, mean (SD), wk 37.3 (2.7) 37.4 (3.2) .71

Highest systolic BP in 24 h prior to hospital discharge, mean (SD),mm Hg

140 (12.6) 140 (11.6) .47

Highest diastolic BP in 24 h prior to hospital discharge, mean (SD),mm Hg

88 (7.1) 88 (7.4) .59

Discharged with prescription for antihypertensives, No. (%)a 183 (22.8) 63 (25.6)

β-Blocker 77 (9.6) 23 (10.8)

.04Calcium channel blocker 119 (14.8) 42 (19.7)

Other 2 (0.2) 0

Maximum systolic BP, mean (SD), mm Hg 145 (13) 150 (14) <.001

Maximum diastolic BP, mean (SD), mm Hg 94 (9) 98 (12) <.001

Time of maximum systolic BP, median (IQR), days post partum 7 (4-10) 8 (5-15) .02

Time of maximum diastolic BP, median (IQR), days post partum 7 (5-12) 9 (5-18) .002

Hospital readmission, No. (%) 76 (9.5) 36 (16.9) .02

Seen for postpartum appointment, No. (%) 686 (87.8) 144 (72.0) <.001

Number of blood pressures reported through program, median (IQR) 17 (11-22) 14 (7-19) <.001

Last postpartum week reported, median (IQR) 6 (5-6) 6 (3-6) <.001

BP recorded after week 3 postp artum, No. (%) 764 (95.0) 185 (86.9) <.001

Abbreviations: BP, blood pressure; IQR,interquartile range.a Women may be discharged with more than 1

antihypertensive agent.

Figure 2. Fitted Blood Pressure (BP) Trajectory in First 6 Weeks Post Partum by Race

Weeks post partum

P < .001 P < .001

150

140

130

120

110

Syst

olic

BP,

mm

Hg

Systolic BP trajectoryA

0 1 2 3 4 5 6

Weeks post partum

95

90

85

80

75

Dias

tolic

BP,

mm

Hg

Diastolic BP trajectoryB

0 1 2 3 4 5 6

Black women

White women

Shaded area indicates 95% CI.





incidence to be higher among Black women compared with White women during the first 6 weekspost partum (36 of 213 [16.9%] vs 76 of 804 [9.5%]; P = .02). There were no maternal deaths in thecohort during this period.

Discussion

In this study, we found that, in the postpartum period, blood pressure decreased rapidly in the first 3weeks and subsequently stabilized, although a substantial proportion of women continued to havehypertension at 6 weeks post partum. We also found that, compared with White women, Blackwomen had a less rapid decrease in blood pressure post partum, resulting in higher blood pressureby the end of the 6-week program. By 6 weeks post partum, 68.1% of Black women met criteria forstage 1 or stage 2 hypertension, compared with 51.4% of White women (P < .001). Our findings havepotential implications for addressing postpartum maternal morbidity and mortality, as well as longer-term cardiovascular health, in this population. In this study, there were no differences by race in earlypregnancy blood pressure or in the highest blood pressure prior to hospital discharge. However, themore adverse postpartum blood pressure trajectories seen in Black women are consistent acrossvarious sensitivity analyses and translate to a higher incidence of hypertension-related hospitalreadmission.

There are several possible explanations for our findings. In previous studies evaluating theassociation of race with cardiovascular recovery after peripartum cardiomyopathy, cardiac recoverytakes twice as long in Black women compared with White women.16,17 Recent studies havedemonstrated the causal overlap between peripartum cardiomyopathy and hypertensive disordersof pregnancy.18,19 It is well documented both within our data as well as in prior studies thathypertension is exacerbated at 3 to 7 days post partum.20 The cause of this exacerbation is unclear;however, others have speculated that it may be due to mobilization of fluid during this period.20 Ifthese fluid shifts result in volume overload or subclinical heart failure, perhaps our findings maymirror the differences seen by race in peripartum cardiomyopathy. This finding would suggest thatadditional postpartum diuresis might be particularly important in this population, as has beendemonstrated by recent work in a predominantly African American population.21 Our findings areconsistent with recent work from Lopes Perdigao et al,13 who demonstrated in a cohort of 84 womenthat women with higher BMI and of Black race had higher mean blood pressure values and were lesslikely to have resolution of their hypertension by 16 days post partum.

Within the context of small randomized clinical trials or research studies, text message–basedpostpartum hypertension monitoring programs have been shown to increase patient engagementand improve adherence with blood pressure follow-up.9,22 A previous study has demonstrated thefeasibility of, high engagement with, retention in, and patient satisfaction with our remotemonitoring program as well as improved adherence with postpartum visits.10 Given poor adherenceto recommended follow-up in the postpartum period, the trajectory of a decrease in blood pressurepost partum has not been well studied previously. This trajectory may be particularly important inconsideration of rising maternal morbidity and mortality in the postpartum period.1 Furthermore,prior studies have shown lower adherence to recommended follow-up among Black women andwomen of lower socioeconomic status.7,23

Table 3. Mean Blood Pressure and Blood Pressure Category by Race at Conclusion of Program Among WomenWith a Blood Pressure Measured After 3 Weeks Post Partum

Characteristic White (n = 764) Black (n = 185) P valueSystolic blood pressure, after week 3, mean (SD), mm Hg 120 (11) 127 (14) <.001

Diastolic blood pressure, after week 3, mean (SD), mm Hg 78 (9) 81 (11) <.001

Stage 1 hypertension (≥130-139/80-89 mm Hg), No. (%) 393 (51.4) 126 (68.1) <.001

Stage 2 hypertension (≥140/90 mm Hg), No. (%) 139 (18.2) 60 (32.4) <.001




The American Heart Association and ACOG have identified hypertensive disorders of pregnancyas risk factors for later-life cardiovascular disease; however, effective evidence-based interventionshave not yet been adequately studied or implemented, to our knowledge.24-27 Prior studies haveshown that postpartum care after a hypertensive disorder of pregnancy is often fragmented, withouta systematic transition from the obstetrician to an internist or cardiologist.25 Given the fragmentedcare and poor attendance rates, our findings raise the question of whether universal home bloodpressure monitoring and management with antihypertensive therapy for women with persistenthypertension beyond the first 6 weeks post partum may be warranted. This is particularly importantgiven the proportion of women in our cohort with ongoing hypertension at 6 weeks post partumand our finding that Black women with a hypertensive disorder ended the program with systolicblood pressures 9 mm Hg higher systolic and diastolic blood pressures 5 mm Hg higher than Whitewomen. More important, there were no differences by race in prenatal blood pressure and ourfindings suggest that a hypertensive disorder of pregnancy may have an adverse association withblood pressure. We cannot assess from our findings whether this persistent postpartumhypertension represents new-onset chronic hypertension or the ongoing resolution of thehypertensive disorder of pregnancy. For women with public health insurance, in many states withinthe United States, Medicaid coverage lasts only through 60 days post partum. Most maternal deathsoccur in the postpartum period, with approximately one-third occurring after 6 weeks post partum.2

Strengths and LimitationsOur study has several strengths. This was a large, prospective cohort from a single tertiary carecenter. Participants provided real-time blood pressure measurements using validated, calibratedblood pressure monitors. We used standardized management protocols to reduce the impact ofclinician variability in management approaches. Our inclusion of prenatal and in-hospital postpartumblood pressure values also strengthens our findings.

Our study also has several limitations. Participants were enrolled in the program by a clinicalcare professional. We cannot exclude the possibility that women enrolled in the program differ fromthose who were not enrolled in the program or that implicit bias or structural racism were associatedwith participant enrollment in the program. The program requires that women have access to a textmessaging–enabled telephone. Prior studies have shown that approximately 92% to 96% ofreproductive-aged women have access to a smartphone.28 However, it is possible that a largerproportion of high-risk women may not have access to a textmessaging–enabled phone; thus, theapplicability of our findings to a broader population within the United States may be limited. Thesingle-site nature of the study limits the generalizability of our findings. In addition, all blood pressurevalues are self-reported through the program; thus, we cannot exclude the possibility thatparticipants may not report accurate measurements, as no external validation was performed. Aswith all analytic approaches, mixed-effects regression modeling has several limitations. However,owing to the structure of our data and the ability to account for some heterogeneity over time, thiswas the most appropriate approach. Black women reported statistically fewer blood pressure valuesthrough the program and reported their final blood pressure value at an earlier postpartum time thanWhite women. Although these differences are statistically significant, the clinical significance is likelyless important given the overall number of blood pressure values reported and the median follow-uptime across both groups with more than 85% follow-up beyond 3 weeks. However, we cannotexclude the possibility that our findings may be associated with bias in attendance vs nonattendanceat the postpartum visit and adherence to the remote monitoring program.

Our remote monitoring program concludes at 6 weeks post partum; thus, a further decrease inblood pressure beyond this period cannot be assessed. Furthermore, our cohort includes onlywomen with a hypertensive disorder of pregnancy. We cannot assess how these blood pressuretrajectories may differ from those of women without a diagnosed hypertensive disorder. Outside ofpregnancy, multiple randomized clinical trials have shown differing effects of antihypertensiveagents by race, such that the American College of Cardiology and American Heart Association




guidelines recommend calcium channel blockers as first-line antihypertensive agents in AfricanAmericans.29,30 These recommendations have not been assessed in a pregnant or postpartumpopulation. Within our cohort, β-blockers and calcium channel blockers were used equally amongBlack and White patients. Although our findings persisted among women not taking anyantihypertensive agent, we cannot exclude the possibility that the interaction of race and type ofantihypertensive agent used could have been associated with our results. Further study is needed toassess whether findings regarding race and effectiveness of specific antihypertensive agents areupheld in a pregnant and postpartum population. Finally, race is a heterogenous and multifacetedexposure that in our health care system encompasses an economic, social, and physiologicalconstruct that includes both biological and nonbiological aspects. It is possible that our findings aresecondary to other unmeasurable factors such as implicit and explicit bias within the structure of ourmedical institutions or the remote monitoring program itself. Unfortunately, because we used datafrom a program built into clinical care, we are unable to assess whether there are managementdifferences in response to elevated blood pressures by race. The structure of the program does notallow for accurate ascertainment of antihypertensive medication changes, as they are frequentlychanged through a verbal recommendation rather than through issuing a new prescription ordocumentation in the medical record. Blood pressure management through the program follows astandard protocol as described previously and information regarding patient race was not availableto the on-call physicians managing blood pressure through the program.10 Thus, we are unable toassess whether variable treatment based on race could have been associated with our findings, butthis remains of particular concern given prior work suggesting differential treatment based onrace31-33 and remains an area of active investigation for our team. We also continue to investigatewhether the differences we see in blood pressure trajectory are associated with the racial disparitiesin morbidity and mortality associated with hypertensive and cardiovascular conditions.

Conclusions

The findings of this cohort study suggest that, compared with White women, Black women have apostpartum blood pressure trajectory with persistence of higher blood pressure values that may beassociated with the increased risk of morbidity and mortality seen in this period. Given the difficultywith adherence to care in this period, our findings support the use of remote monitoring amongwomen at highest risk for hypertensive-related morbidity and suggest the importance of ongoingaccess to care beyond 6 weeks post partum.

ARTICLE INFORMATIONAccepted for Publication: October 30, 2020.

Published: December 22, 2020. doi:10.1001/jamanetworkopen.2020.30815

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Hauspurg Aet al. JAMA Network Open.

Corresponding Author: Alisse Hauspurg, MD, Department of Obstetrics, Gynecology and Reproductive Sciences,Magee-Womens Hospital, University of Pittsburgh School of Medicine, 300 Halket St, Pittsburgh, PA 15232 ([email protected]).

Author Affiliations: Magee-Womens Research Institute, University of Pittsburgh School of Medicine, Pittsburgh,Pennsylvania (Hauspurg, Larkin, Simhan); Department of Obstetrics, Gynecology and Reproductive Sciences,Magee-Womens Hospital, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania (Hauspurg,Lemon, Cabrera, Binstock, Quinn, Larkin, Beigi, Simhan); Department of Clinical Analytics, University of PittsburghMedical Center, Pittsburgh, Pennsylvania (Lemon); University of Pittsburgh School of Medicine, Pittsburgh,Pennsylvania (Javaid); Department of Surgery, University of Pittsburgh School of Medicine, Pittsburgh,Pennsylvania (Watson).





https://jamanetwork.com/journals/jamanetworkopen/pages/instructions-for-authors#SecOpenAccess/?utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2020.30815

mailto:[email protected]

mailto:[email protected]

Author Contributions: Dr Hauspurg had full access to all of the data in the study and takes responsibility for theintegrity of the data and the accuracy of the data analysis.

Concept and design: Hauspurg, Lemon, Quinn, Larkin, Beigi, Simhan.

Acquisition, analysis, or interpretation of data: Hauspurg, Lemon, Cabrera, Javaid, Binstock, Larkin, Watson.

Drafting of the manuscript: Hauspurg, Lemon, Cabrera, Larkin, Beigi, Simhan.

Critical revision of the manuscript for important intellectual content: Hauspurg, Javaid, Binstock, Quinn,Watson, Simhan.

Statistical analysis: Hauspurg, Lemon, Larkin, Simhan.

Administrative, technical, or material support: Hauspurg, Cabrera, Quinn, Watson, Beigi, Simhan.

Supervision: Larkin, Beigi, Simhan.

Conflict of Interest Disclosures: Dr Hauspurg reported receiving grants from the University of Pittsburgh duringthe conduct of the study. Dr Simhan reported being a co-founder of Naima Health LLC, which does work in thedomain of digital patient engagement, outside the submitted work. No other disclosures were reported.

Funding/Support: This work was supported by National Institutes of Health/Office of Research on Women’sHealth Building Interdisciplinary Research Careers in Women’s Health NIH K12HD043441 scholar funds to DrHauspurg.

Role of the Funder/Sponsor: The funding source had no role in the design and conduct of the study; collection,management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; anddecision to submit the manuscript for publication.

Meeting Presentation: This study was presented at the 40th Annual Pregnancy Meeting of the Society forMaternal Fetal Medicine; February 6, 2020; Grapevine, Texas.

REFERENCES1. Hoyert DL, Miniño AM. Maternal mortality in the United States: changes in coding, publication and data release,2018. Natl Vital Stat Rep. 2020;69(2):1-18.

2. Centers for Disease Control and Prevention. Data on selected pregnancy complications in the United States.Published February 28, 2019. Accessed February 10, 2020. https://www.cdc.gov/reproductivehealth/maternalinfanthealth/pregnancy-complications-data.htm#

3. Lisonkova S, Joseph KS. Incidence of preeclampsia: risk factors and outcomes associated with early- versus late-onset disease. Am J Obstet Gynecol. 2013;209(6):544.e1-544.e12. doi:10.1016/j.ajog.2013.08.019

4. Johnson JD, Louis JM. Does race or ethnicity play a role in the origin, pathophysiology, and outcomes ofpreeclampsia? an expert review of the literature. Am J Obstet Gynecol. Published onlinne July 24, 2020. doi:10.1016/j.ajog.2020.07.038

5. Clapp MA, Little SE, Zheng J, Robinson JN. A multi-state analysis of postpartum readmissions in the UnitedStates. Am J Obstet Gynecol. 2016;215(1):113.e1-113.e10. doi:10.1016/j.ajog.2016.01.174

6. ACOG practice bulletin no. 202: gestational hypertension and preeclampsia. Obstet Gynecol. 2019;133(1):e1-e25. doi:10.1097/AOG.0000000000003018

7. ACOG committee opinion no. 736: optimizing postpartum care. Obstet Gynecol. 2018;131(5):e140-e150. doi:10.1097/AOG.0000000000002633

8. Cairns AE, Pealing L, Duffy JMN, et al. Postpartum management of hypertensive disorders of pregnancy:a systematic review. BMJ Open. 2017;7(11):e018696. doi:10.1136/bmjopen-2017-018696

9. Hirshberg A, Downes K, Srinivas S. Comparing standard office-based follow-up with text-based remotemonitoring in the management of postpartum hypertension: a randomised clinical trial. BMJ Qual Saf. 2018;27(11):871-877. doi:10.1136/bmjqs-2018-007837

10. Hauspurg A, Lemon LS, Quinn BA, et al. A postpartum remote hypertension monitoring protocol implementedat a hospital level. Obstet Gynecol. 2019;134(4):685-691. doi:10.1097/AOG.0000000000003479

11. Walters BNJ, Walters T. Hypertension in the puerperium. Lancet. 1987;2(8554):330. doi:10.1016/S0140-6736(87)90912-3

12. Walters BN, Thompson ME, Lee A, de Swiet M. Blood pressure in the puerperium. Clin Sci (Lond). 1986;71(5):589-594. doi:10.1042/cs0710589

13. Lopes Perdigao J, Hirshberg A, Koelper N, Srinivas SK, Sammel MD, Levine LD. Postpartum blood pressuretrends are impacted by race and BMI. Pregnancy Hypertens. 2020;20:14-18. doi:10.1016/j.preghy.2020.02.006




https://www.ncbi.nlm.nih.gov/pubmed/32510319

https://www.cdc.gov/reproductivehealth/maternalinfanthealth/pregnancy-complications-data.htm

https://www.cdc.gov/reproductivehealth/maternalinfanthealth/pregnancy-complications-data.htm

https://dx.doi.org/10.1016/j.ajog.2013.08.019




https://dx.doi.org/10.1097/AOG.0000000000003018



https://dx.doi.org/10.1136/bmjopen-2017-018696

https://dx.doi.org/10.1136/bmjqs-2018-007837


https://dx.doi.org/10.1016/S0140-6736(87)90912-3

https://dx.doi.org/10.1016/S0140-6736(87)90912-3

https://dx.doi.org/10.1042/cs0710589

https://dx.doi.org/10.1016/j.preghy.2020.02.006

14. American College of Obstetricians and Gynecologists; Task Force on Hypertension in Pregnancy. Hypertensionin pregnancy: report of the American College of Obstetricians and Gynecologists’ Task Force on Hypertension inPregnancy. Obstet Gynecol. 2013;122(5):1122-1131.

15. Colin Bell A, Adair LS, Popkin BM. Ethnic differences in the association between body mass index andhypertension. Am J Epidemiol. 2002;155(4):346-353. doi:10.1093/aje/155.4.346

16. Goland S, Modi K, Hatamizadeh P, Elkayam U. Differences in clinical profile of African-American women withperipartum cardiomyopathy in the United States. J Card Fail. 2013;19(4):214-218. doi:10.1016/j.cardfail.2013.03.004

17. Irizarry OC, Levine LD, Lewey J, et al. Comparison of clinical characteristics and outcomes of peripartumcardiomyopathy between African American and non–African American women. JAMA Cardiol. 2017;2(11):1256-1260. doi:10.1001/jamacardio.2017.3574

18. Gammill HS, Chettier R, Brewer A, et al. Cardiomyopathy and preeclampsia. Circulation. 2018;138(21):2359-2366. doi:10.1161/CIRCULATIONAHA.117.031527

19. Parikh P, Blauwet L. Peripartum cardiomyopathy and preeclampsia: overlapping diseases of pregnancy. CurrHypertens Rep. 2018;20(8):69. doi:10.1007/s11906-018-0868-9

20. Taylor RN, Roberts JM, Cunningham FG, Lindheimer MD. Chesley’s Hypertensive Disorders in Pregnancy. 4thed. Elsevier; 2015.

21. Perdigao JL, Lewey J, Hirshberg A, et al LB 4: furosemide for accelerated recovery of blood pressurepostpartum: a randomized placebo controlled trial (FoR BP). Am J Obstet Gynecol. 2020;222(1):S759-S760. doi:10.1016/j.ajog.2019.11.1278

22. Hoppe KK, Williams M, Thomas N, et al. Telehealth with remote blood pressure monitoring for postpartumhypertension: a prospective single-cohort feasibility study. Pregnancy Hypertens. 2019;15:171-176. doi:10.1016/j.preghy.2018.12.007

23. Levine LD, Nkonde-Price C, Limaye M, Srinivas SK. Factors associated with postpartum follow-up andpersistent hypertension among women with severe preeclampsia. J Perinatol. 2016;36(12):1079-1082. doi:10.1038/jp.2016.137

24. Graves CR, Davis SF. Cardiovascular complications in pregnancy: it is time for action. Circulation. 2018;137(12):1213-1215. doi:10.1161/CIRCULATIONAHA.117.031592

25. Spaan J, Peeters L, Spaanderman M, Brown M. Cardiovascular risk management after a hypertensive disorderof pregnancy. Hypertension. 2012;60(6):1368-1373. doi:10.1161/HYPERTENSIONAHA.112.198812

26. Hauspurg A, Ying W, Hubel CA, Michos ED, Ouyang P. Adverse pregnancy outcomes and future maternalcardiovascular disease. Clin Cardiol. 2018;41(2):239-246. doi:10.1002/clc.22887

27. Haas DM, Parker CB, Marsh DJ, et al; NHLBI nuMoM2b Heart Health Study. Association of adverse pregnancyoutcomes with hypertension 2 to 7 years postpartum. J Am Heart Assoc. 2019;8(19):e013092. doi:10.1161/JAHA.119.013092

28. Pew Research Center. Mobile fact sheet. Published June 12, 2019. Accessed April 6, 2020. https://www.pewinternet.org/fact-sheet/mobile/

29. Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the prevention, detection, evaluation, and management of high blood pressure in adults:a report of the American College of Cardiology/American Heart Association Task Force on Clinical PracticeGuidelines. Hypertension. 2018;71(6):e13-e115. doi:10.1161/HYP.0000000000000065

30. Leenen FHH, Nwachuku CE, Black HR, et al; Antihypertensive and Lipid-Lowering Treatment to Prevent HeartAttack Trial Collaborative Research Group. Clinical events in high-risk hypertensive patients randomly assigned tocalcium channel blocker versus angiotensin-converting enzyme inhibitor in the antihypertensive and lipid-lowering treatment to prevent heart attack trial. Hypertension. 2006;48(3):374-384. doi:10.1161/01.HYP.0000231662.77359.de

31. Tedla YG, Schwartz SM, Silberman P, Greenland P, Passman RS. Racial disparity in the prescription ofanticoagulants and risk of stroke and bleeding in atrial fibrillation patients. J Stroke Cerebrovasc Dis. 2020;29(5):104718. doi:10.1016/j.jstrokecerebrovasdis.2020.104718

32. Berhie SH, Riley LE, Bryant AS. Disparities in 17-hydroxyprogesterone caproate offer and uptake inpregnancies at a risk of preterm birth. Am J Perinatol. 2019;36(10):1066-1071. doi:10.1055/s-0038-1676113

33. Yee LM, Liu LY, Sakowicz A, Bolden JR, Miller ES. Racial and ethnic disparities in use of 17-alphahydroxyprogesterone caproate for prevention of preterm birth. Am J Obstet Gynecol. 2016;214(3):374.e1-374.e6.doi:10.1016/j.ajog.2015.12.054




https://www.ncbi.nlm.nih.gov/pubmed/24150027

https://dx.doi.org/10.1093/aje/155.4.346

https://dx.doi.org/10.1016/j.cardfail.2013.03.004

https://dx.doi.org/10.1016/j.cardfail.2013.03.004

https://jama.jamanetwork.com/article.aspx?doi=10.1001/jamacardio.2017.3574&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2020.30815

https://dx.doi.org/10.1161/CIRCULATIONAHA.117.031527

https://dx.doi.org/10.1007/s11906-018-0868-9





https://dx.doi.org/10.1038/jp.2016.137

https://dx.doi.org/10.1038/jp.2016.137

https://dx.doi.org/10.1161/CIRCULATIONAHA.117.031592

https://dx.doi.org/10.1161/HYPERTENSIONAHA.112.198812

https://dx.doi.org/10.1002/clc.22887

https://dx.doi.org/10.1161/JAHA.119.013092

https://dx.doi.org/10.1161/JAHA.119.013092

https://www.pewinternet.org/fact-sheet/mobile/

https://www.pewinternet.org/fact-sheet/mobile/

https://dx.doi.org/10.1161/HYP.0000000000000065

https://dx.doi.org/10.1161/01.HYP.0000231662.77359.de

https://dx.doi.org/10.1161/01.HYP.0000231662.77359.de

https://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2020.104718

https://dx.doi.org/10.1055/s-0038-1676113


SUPPLEMENT.eTable 1. Demographics of Overall CohorteTable 2. Delivery and Postpartum CharacteristicseFigure 1. Fitted BP Trajectory (Solid Line) and 95% CI (Shaded Area) in First 6 Weeks Postpartum Among ObeseWomen (BMI �30 kg/m2) by RaceeFigure 2. Fitted BP Trajectory (Solid Line) and 95% CI (Shaded Area) in First 6 Weeks Postpartum Among Womenon No Antihypertensive Medications by RaceeFigure 3. Fitted BP Trajectory (Solid Line) and 95% CI (Shaded Area) in First 6 Weeks Postpartum Among Womenon Antihypertensive Medications at the Time of Discharge by RaceeFigure 4. Fitted BP Trajectory (Solid Line) and 95% CI (Shaded Area) in First 6 Weeks Postpartum Among WomenWith Preeclampsia by Race




OriginalInvestigation | ObstetricsandGynecology ......Systolic blood pressure, after week 3, mean...

Documents

Transcript of OriginalInvestigation | ObstetricsandGynecology ......Systolic blood pressure, after week 3, mean...