Hypertension during pregnancy is an important cause of maternal morbidity and mortality.1 Avoidance of prolonged maternal exposure to severe hypertension (systolic blood pressure [BP] 160 mm Hg or greater or diastolic BP 110 mm Hg or greater) is paramount to the prevention of serious adverse outcomes such as intracranial hemorrhage and death.2 The American College of Obstetricians and Gynecologists (ACOG) recently updated national guidelines to improve identification and care of women with hypertension in pregnancy and published three treatment algorithms for hypertensive emergency.3,4 Each starts with a first-line antihypertensive agent: intravenous (IV) labetalol, IV hydralazine, or oral nifedipine.3 Subspecialty consultation is advised if BPs remain severely elevated after approximately 1 hour of treatment, suggesting that most patients should achieve BP control within 60 minutes of initial treatment. Blood pressure control has previously been defined in comparative antihypertensive trials as systolic BP 150 mm Hg or less and diastolic BP 100 mm Hg or less.5–7
Obstetric hypertensive emergency is frequently treated in an emergency department or obstetric triage unit. At Women & Infants Hospital, a tertiary care center affiliated with Brown University, women are treated in a specialty emergency department run by obstetrician–gynecologists, which also houses obstetric triage. We aimed to assess treatment outcomes associated with a quality improvement intervention designed to align clinical practice in this women's emergency department with updated ACOG guidelines and hypothesized that more women would achieve goal BP within an hour of therapy after using this more aggressive treatment approach.
MATERIALS AND METHODS
After institutional review board and quality council approval (Women & Infants Hospital [#15-0064] on August 11, 2015), we implemented a quality improvement intervention in the emergency department of Women & Infants Hospital and conducted a cohort study of women treated for obstetric hypertensive emergency before and after the intervention to assess its effects. As described previously, Women & Infants Hospital is a tertiary care center affiliated with Brown University. It is the largest obstetric hospital in Rhode Island, with more than 8,000 deliveries per year and is a referral center for patients from Rhode Island, eastern Connecticut, and southeastern Massachusetts. The emergency department, which also houses obstetric triage, is run by obstetrician–gynecologists and additionally staffed by obstetrics and gynecology resident housestaff (eight residents per training year), certified nurse–midwives, and nurse practitioners.
The quality improvement initiative involved several steps (Fig. 1). The obstetrics and gynecology physician practice guideline and nursing policy for gestational hypertension and preeclampsia were revised. The nursing policy specifically outlines acceptable BP monitoring procedures, including patient positioning and appropriate cuff size and application. The policy also mandates that any severe-range BP reading obtained automatically be confirmed with a manual measurement. Both documents were reviewed by physician and nursing leadership and voted on for acceptance by the medical staff before publication as a clinical resource on the hospital's secure network. In conjunction with nursing educators, management aids (laminated posters and small reference cards designed for staff to wear with their hospital identification badge) outlining the obstetric hypertensive emergency treatment algorithms were developed for use in real-time clinical care (Fig. 2). The management aids were intended to help staff recognize and communicate the presence of a hypertensive emergency to facilitate more timely IV access and treatment. Posters were placed in common areas and adjacent to every clinical workstation in individual treatment rooms within the emergency department. The reference cards were distributed to all emergency department staff during brief educational sessions (approximately 10–15 minutes) conducted during staff meetings or clinical work hours. During these sessions, the principal investigator (R.J.F.) explained the importance of treating acute-onset severe hypertension in pregnant and postpartum women as an emergency, reviewed the BP parameters to initiate treatment as well as goal BP ranges after treatment, and discussed the new treatment algorithms in detail. Finally, the computerized physician order entry set used for evaluation and treatment of gestational hypertension and preeclampsia was updated. This included changing the doses of antihypertensive agents to reflect those contained in the ACOG treatment algorithms (20, 40, or 80 mg IV labetalol; 5 or 10 mg IV hydralazine, and 10 or 20 mg oral nifedipine) and also added instructions for how often to provide consecutive doses of each of the three drugs, how often to monitor the patient's BP during and after treatment, and goals for BP control (maintain between 140–150 and 90–100 mm Hg). Each consecutive antihypertensive dose required a clinician order and could not be administered automatically by nursing staff. The laboratory section was also updated to include assessment of serum alanine aminotransferase in addition to aspartate aminotransferase (other laboratory tests included in the order set were complete blood count with platelets, serum creatinine, and urine protein-to-creatinine ratio). All steps of the quality improvement initiative were completed by December 2015.
Pregnant (20 weeks of gestation or greater) and postpartum (6 weeks postpartum or less) women with a hypertensive emergency (systolic BP 160 mm Hg or greater, diastolic BP 110 mm Hg or greater, or both) treatment initiated in the emergency department of Women & Infants Hospital were eligible for inclusion. Patients were assigned to preintervention and postintervention groups, which spanned from September 1, 2014, to September 30, 2015, and December 1, 2015, to November 30, 2016, respectively. Patients were identified for inclusion if they received IV labetalol, IV hydralazine, or oral nifedipine during their emergency department course. Patients who received antihypertensive medication but were not severely hypertensive at the time of the administration, or for another indication such as tocolysis, were excluded. A single author (R.J.F.) collected maternal demographic and treatment characteristics as well as pregnancy and neonatal outcome data by individual patient electronic chart review. Treatment characteristics recorded included all BP values from initial severe reading to goal BP and the type, date, and time of each antihypertensive agent administered. Blood pressure values were manually entered into the electronic medical record system by nursing staff. The timing of medication administration was recorded automatically when a patient's wrist band was scanned, unless a manual override was required. Double data entry was performed to minimize error; data were initially recorded onto a paper data collection form and subsequently transferred to an electronic database. Missing data fields and outliers were reported to the primary investigator before analysis for clarification (ie, data truly missing from medical record, inadvertently not recorded, or entered inaccurately). If a patient had multiple emergency department presentations for a hypertensive emergency, treatment characteristics were collected from only the first episode of care to maintain data independence.
The a priori primary outcome was achievement of goal BP (systolic BP 150 mm Hg or less and diastolic BP 100 mm Hg or less) within an hour of initial antihypertensive therapy (and maintained for at least 2 hours). Based on direct clinical observation in the unit, it was estimated that 40% of women would achieve the primary outcome in the preintervention group. To detect a ±15% difference in the postintervention group, with a power of eighty percent and type 1 error of 5%, a sample size of 346 women was required. Planned secondary outcomes included times from first severe-range BP to 1) first treatment and 2) achieving goal BP. Baseline characteristic and outcome data comparisons were performed using Fisher exact tests for categorical variables and Student t test or Wilcoxon rank-sum tests for continuous variables. Nonparametric tests were used for data that were not normally distributed. All tests were two-tailed, and a P value of <.05 was used to define statistical significance.
Three hundred forty-six women were included (Fig. 3). There were equal numbers of pregnant (n=124) and postpartum (n=49) women in the preintervention and postintervention groups. When baseline characteristics of women treated before and after the intervention were compared, there were no significant differences (Table 1); maternal age, race–ethnicity, parity, and pregnancy plurality were all similar. The presence of comorbidities such as pre-existing or gestational diabetes mellitus, chronic hypertension, and history of preeclampsia were also similar between groups. Of women with chronic hypertension, 54% were prescribed an oral antihypertensive agent (labetalol, nifedipine, or methyldopa) at the time of their emergency department presentation in the preintervention group compared with 69% in the postintervention group (P=.15).
The median gestational age (36 vs 35 weeks in the preintervention and postintervention groups, respectively) and number of days postpartum (6 vs 7 days in the preintervention and postintervention groups, respectively) at the time of first emergency department presentation for treatment of hypertensive emergency were also similar (Table 2). The most common chief complaint was hypertension, either at home or in an outpatient office, followed by neurologic (eg, headache) or labor symptoms. More women had both aspartate aminotransferase and alanine aminotransferase measured (vs aspartate aminotransferase alone) on initial laboratory assessment in the postintervention group (n=129 vs n=56 in the preintervention group). Pulmonary edema at the time of initial presentation was rare in both groups. Nonreassuring fetal heart tones during BP stabilization were uncommon in both groups (Table 3). Fewer pregnant women were documented to report a headache after treatment in the postintervention group: 33% vs 58% (P<.001; Table 3). There were otherwise no significant differences in maternal or neonatal outcomes between the preintervention and postintervention groups for both pregnant and postpartum women (Tables 3 and 4).
The primary outcome, the proportion of women achieving goal BP within 60 minutes of initial therapy, was not statistically different between groups with 41% achieving the outcome preintervention compared with 47% postintervention (P=.28; Table 5). Median time from first severe BP to first treatment was unchanged (30 minutes preintervention vs 29 minutes postintervention, P=.058); however, median time from first severe BP to achieving goal BP decreased significantly (122 minutes vs 95 minutes, P=.04). Intravenous labetalol and oral nifedipine were used more often than IV hydralazine as an initial antihypertensive agent both before and after the intervention (Table 5). In the preintervention group, just under half of women received oral nifedipine or IV labetalol; after the project, there was a statistically significant increase in the use of IV labetalol as the first antihypertensive agent, increasing from 47% to 76% (P<.001).
Several other secondary outcome measures were evaluated to assess markers of quality improvement intervention protocol adherence (Table 6). Median time between first and second BP measurements decreased (28 minutes preintervention vs 25 minutes postintervention, P=.02); however, confirmation of hypertensive emergency within 15 minutes was only achieved in approximately 20% of women in either group. The proportion of women treated within 15 minutes was 18% preintervention compared with 26% postintervention (P=.09). Consecutive doses of IV labetalol and oral nifedipine for persistent severe hypertension were administered according to the doses recommended by ACOG more often in the postintervention group.
The quality improvement initiative studied failed to achieve the primary outcome; there was no association between the initiative and an increased proportion of women achieving BP control within an hour of obstetric hypertensive emergency treatment. In fact, less than half of women achieved goal BP within an hour of therapy before or after the intervention. However, some changes in practice after the intervention were noted. First, median time to achieving goal BP after treatment was significantly reduced. Additionally, more women were treated with IV antihypertensive agents after the intervention, suggesting timelier IV access establishment. Antihypertensive treatments were also administered according to the new ACOG dosage recommendations more often in the postintervention group. These findings suggest that, for the women included in this study, staff used a more aggressive treatment approach after the initiative.
The benefits of implementing evidence-based guidelines and standardized protocols for treating hypertension in pregnancy have been demonstrated in prior work.8,9 von Dadelszen et al8 found an association between use of standardized management guidelines in maternity hospitals in British Columbia and decreased Provincial rates of adverse maternal and perinatal outcomes. Similarly, Clark et al9 found a decreased maternal death rate related to hypertensive disorders in the 6 years after the introduction of checklist-based disease-specific treatment protocols directed at preventing death from postcesarean pulmonary embolism, intracranial hemorrhage associated with hypertensive crisis, and postpartum hemorrhage. A contemporaneous study published by Shields et al10 in April 2017 showed a significant reduction in the incidence of eclampsia and severe maternal morbidity associated with improved hospital compliance with state and national treatment guidelines for severe hypertension in pregnancy. This publication resulted from a large collaboration of 23 California hospitals that received education and underwent monitoring for compliance with three components of state and national guidelines: 1) treatment of severe hypertension within 1 hour of verification, 2) use of magnesium sulfate for patients with critically elevated BP, and 3) early postpartum follow-up (within 1 or 2 weeks of discharge). The described findings are encouraging; the use of consistent, evidence-based practices can significantly improve the quality of patient care (reflected by improved outcomes).
Although the present study was not powered to detect a difference in adverse maternal or neonatal outcomes, by rigorously evaluating treatment characteristics and markers of protocol adherence, we were able to assess for both changes in clinical practice and identify areas for continued improvement. First, elimination of barriers to confirming and communicating the presence of obstetric hypertensive emergency within 15 minutes should be identified and addressed, because this would facilitate more timely treatment. Next, streamlining the administration of consecutive antihypertensive doses should be considered. In the current system, each consecutive dose requires a separate clinician order. The safety of reflexive nursing administration of consecutive doses for refractory hypertension within the established treatment algorithms should be evaluated. Finally, administration of a long-acting antihypertensive agent once a certain BP threshold is achieved should be considered. This was not specifically included in the protocol in the studied initiative, but could decrease the time required for patients to achieve sustained BP control.
The strengths of this study would first include the multidisciplinary collaboration of physician, nursing, and advanced practice providers on a quality improvement intervention that led to changed clinical practice. Next, the simplicity of the intervention would make it easy to replicate in another center and generalizable to other obstetric populations. Additionally, study data were abstracted directly from individual patient records and not limited to coding data. Study limitations include the retrospective nature of the data collection, which depends on completeness and assumes accuracy of the medical record. This study captured only patients who received treatment for severe hypertension in the emergency department and cannot comment on treatment of patients with obstetric hypertensive emergency on other hospital units. Given safety protocols in place, it is unlikely that a significant number of patients with hypertensive emergency went untreated in the emergency department during the study period. Finally, although the intervention should be generalizable to other obstetric populations, its use in emergency departments with different types of health care providers or patient acuity levels may yield different results.
Safe and effective care of patients experiencing an obstetric hypertensive emergency requires a shared understanding of not only treatment strategy, but also the urgency with which it should be applied amongst all health care providers. This study suggests that a simple, multidisciplinary treatment approach and clinical management aids can improve some aspects of obstetric hypertensive emergency treatment and highlights the need for ongoing quality improvement efforts.
1. GBD 2015 Maternal Mortality Collaborators. Global, regional, and national levels of maternal mortality, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet 2016;388:1775–812.
2. Clark SL, Belfort MA, Dildy GA, Herbst MA, Meyers JA, Hankins GD. Maternal death in the 21st century: causes, prevention, and relationship to cesarean delivery. Am J Obstet Gynecol 2008;199:36.e1–5.
3. Emergent therapy for acute-onset, severe hypertension during pregnancy and the postpartum period. Committee Opinion No. 692. American College of Obstetricians and Gynecologists. Obstet Gynecol 2017;129:e90–5.
4. Executive summary: hypertension in pregnancy. American College of Obstetricians and Gynecologists. Obstet Gynecol 2013;122:1122–31.
5. Shekhar S, Gupta N, Kirubakaran R, Pareek P. Oral nifedipine versus intravenous labetalol for severe hypertension during pregnancy: a systematic review and meta-analysis. BJOG 2016;123:40–7.
6. Rezaei Z, Sharbaf FR, Pourmojieb M, Youefzadeh-Fard Y, Motevalian M, Khazaeipour Z, et al. Comparison of the efficacy of nifedipine and hydralazine in hypertensive crisis in pregnancy. Acta Med Iran 2011;49:701–6.
7. Raheem IA, Saaid R, Omar SZ, Tan PC. Oral nifedipine versus intravenous labetalol for acute blood pressure control in hypertensive emergencies of pregnancy: a randomised trial. BJOG 2012;119:78–85.
8. von Dadelszen P, Sawchuck D, McMaster R, Douglas MJ, Lee SK, Saunders S, et al. The active implementation of pregnancy hypertension guidelines in British Columbia. Obstet Gynecol 2010;116:659–66.
9. Clark SL, Christmas JT, Frye DR, Meyers JA, Perlin JB. Maternal mortality in the United States: predictability and the impact of protocols on fatal postcesarean pulmonary embolism and hypertension-related intracranial hemorrhage. Am J Obstet Gynecol 2014;211:32.e1–9.
10. Shields LE, Wiesner S, Klein C, Pelletreau B, Hedriana HL. Early standardized treatment of critical blood pressure elevations is associated with a reduction in eclampsia and severe maternal morbidity. Am J Obstet Gynecol 2017;216:415.e1–5.