Postpartum readmissions for hypertensive disorders in pregnancy during the COVID-19 pandemic.

BACKGROUND: Hypertensive disorders in pregnancy are one of the most common causes of readmission in the postpartum period. Because of the COVID-19 pandemic, early hospital discharge was encouraged for patients who were medically stable, because hospitalization rates among SARS-CoV-2-infected patients steadily increased in 2020. The impact of an early discharge policy on postpartum readmission rates among patients with hypertensive disorders in pregnancy is unknown.
OBJECTIVE: This study aimed to compare the postpartum readmission rates of patients with hypertensive disorders in pregnancy before and after implementation of an early discharge policy owing to the COVID-19 pandemic. STUDY
DESIGN: This was a quality improvement, retrospective cohort study of postpartum patients with antenatal hypertensive disorders in pregnancy who delivered and were readmitted because of hypertensive disorders in pregnancy at the New York University Langone Health medical center from March 1, 2019 to February 29, 2020 (control cohort) and from April 1, 2020 to March 31, 2021 (COVID-19 cohort). During the pandemic, our institution introduced an early discharge policy for all postpartum patients to be discharged no later than 2 days postpartum during the delivery admission if deemed medically appropriate. The reduction in postpartum length of stay was accompanied by the continuation of patient education, home blood pressure monitoring, and outpatient follow-up. The primary outcome was the comparison of the readmission rates for patients with postpartum hypertensive disorders in pregnancy. Data were analyzed using Fisher's Exact tests, chi-square tests, and Wilcoxon rank-sum tests with significance defined as P<.05.
RESULTS: There was no statistical difference in the readmission rates for patients with postpartum hypertensive disorders in pregnancy before vs after implementation of an early discharge policy (1.08% for the control cohort vs 0.59% for the COVID-19 cohort). The demographics in each group were similar, as were the median times to readmission (5.0 days; interquartile range, 4.0-6.0 days vs 6.0 days; interquartile range, 5.0-6.0 days; P=.13) and the median readmission length of stay (3.0 days; interquartile range, 2.0-4.0 days vs 3.0 days; interquartile range, 2.0-4.0 days; P=.45). There was 1 intensive care unit readmission in the COVID-19 cohort and none in the control cohort (P=.35). There were no severe maternal morbidities or maternal deaths.
CONCLUSION: These findings suggest that policies calling for a reduced postpartum length of stay, which includes patients with hypertensive disorders in pregnancy, can be implemented without impacting the hospital readmission rate for patients with hypertensive disorders in pregnancy. Continuation of patient education and outpatient surveillance during the pandemic was instrumental for the outpatient postpartum management of the study cohort. Further investigation into best practices to support early discharges is warranted.

A. Why was this study conducted? This study aimed to determine if an early discharge policy was a reasonable protocol for postpartum patients with hypertensive disorders in pregnancies.

B. What are the key findings? With implementation of an early discharge policy, the readmission rate for patients with hypertensive disorders in pregnancy was not higher in comparison to the pre-pandemic readmission rate.

C. What does this study add to what is already known? This study adds to the existing data regarding safe postpartum length of stay and the effects of the pandemic on an obstetric patient population at high risk for readmission.

Alt-text: Unlabelled box

Introduction

Hypertensive disorders in pregnancy (HDP) are one of the most common causes for obstetric-related readmissions. Previous studies have published postpartum readmission rates due to HDP ranging from 0.4% to 9.3%.2, 3, 4, 5, 6 Severe maternal morbidities related to postpartum hypertension and preeclampsia include stroke, eclampsia, and death. , Between 2014-2017, HDP comprised 6.6% of pregnancy-related deaths. These serious risks are why prompt and thorough evaluations are warranted for patients who exhibit any signs or symptoms of HDP in pregnancy and postpartum. The 2013 American College of Obstetricians and Gynecologists (ACOG) Hypertension in Pregnancy Executive Summary suggests that postpartum patients with HDP should “be monitored in the hospital or that equivalent outpatient surveillance be performed for at least 72 hours postpartum and again 7-10 days after delivery or earlier in women with symptoms.”

Standard postpartum hospitalization length of stay (LOS) is 48 hours after a vaginal delivery and 72 hours after a cesarean delivery. For postpartum patients, early discharge can be considered for those deemed medically appropriate and if the patient desires. Due to the pandemic, standard labor and delivery protocols were modified to mitigate virus transmission and decrease hospitalizations amid rising COVID-19 infection and hospitalization rates. Our institution introduced an early discharge policy (EDP) for all postpartum patients, regardless of mode of delivery. Little is known about the impact of early discharge on readmission rates for postpartum patients with HDP. We aimed to compare the postpartum readmission rates due to HDP before and after implementation of an early discharge policy during the COVID-19 pandemic.

Materials and Methods

This was an Institutional Review Board exempt, quality improvement, retrospective cohort study of postpartum patients with antenatal HDP who delivered and were readmitted postpartum for a HDP indication within 30 days at NYU Langone Health-Tisch Hospital and NYU Langone Hospital-Brooklyn between March 2019 to March 2021. The study cohort was divided into the following groups: patients who delivered prior to implementation of the early discharge policy (3/1/2019-2/29/2020, control group) and patients who delivered after implementation of the early discharge policy (4/1/2020-3/31/2021, COVID group).

Our institution's early discharge policy authorized all postpartum patients, including those with HDP and regardless of mode of delivery, to be discharged no later than 2 days postpartum during the delivery admission if deemed medically appropriate. The early discharge policy did not apply to the readmission time period. Criteria to determine medical appropriateness for discharge during the delivery admission included, but was not limited to having nonincreasing vaginal bleeding, no signs of infection, nonsevere range blood pressures at least 24 hours prior to discharge, and absence of symptoms related to preeclampsia. There was no change in criteria for determining medical appropriateness for discharge during the COVID period. At the time of discharge from the delivery admission, our obstetrical department provided HDP patients with education on signs and symptoms of preeclampsia, blood pressure monitoring kit, and outpatient follow-up 1-6 weeks postpartum. The practice of providing patient education and tools for outpatient surveillance were in place prior to the pandemic and was continued throughout the pandemic in order remain in compliance with ACOG recommendations. We evaluated the clinical course for patients readmitted with HDP during their delivery admissions and hospital readmissions. Demographic and clinical information were extracted via chart review from the electronic medical record.

The inclusion criteria were defined as individuals with antenatal HDP who were readmitted postpartum to our hospital due to HDP. The following hypertensive disorders in pregnancy were included: gestational hypertension (GHTN), chronic hypertension (CHTN), preeclampsia without severe features (PECwoSF), preeclampsia with severe features (PECwSF), chronic hypertension with superimposed preeclampsia without severe features (CHTN SIPECwoSF), chronic hypertension with superimposed preeclampsia with severe features (CHTN SIPECwSF), HELLP syndrome, and eclampsia. Exclusion criteria included patients without antenatal HDP, postpartum patients who were readmitted for alternative diagnoses, and patients who delivered at a different institution.

The primary outcome was comparison of postpartum HDP readmission rates between groups. The secondary outcomes were comparison of the following during the delivery admission and readmission periods: HDP diagnosis, maximum blood pressures during each hospitalization period, use of immediate-acting antihypertension medication, use of long-acting antihypertension medication, number of inpatient hypertensive crisis (defined as SBP >/= 160 mmHg and/or DBP >/= 110 mmHg), administration of intravenous magnesium postpartum, number of patients discharged home on long-acting antihypertension medication, amount of oral antihypertension medications discharged home on, LOS, time to readmission, ICU admission, postpartum eclampsia, postpartum stroke, and maternal death.

Study data were collected and managed in a secure REDCap electronic database. Statistical data were analyzed using RStudio version 4.0.3. Categorical variables were assessed using chi-square test and Fisher's Exact test as appropriate. Continuous variables were assessed using Wilcoxon rank-sum test. A P value of <0.05 was considered statistically significant.

Results

There were a total of 3,698 delivered patients with antenatal HDP among 18,951 total deliveries during the study period. Overall, there were 31 (0.16%) postpartum patients with antenatal HDP whom were readmitted due to HDP of 18,951 total deliveries. Prior to implementation of the early discharge policy, 20 postpartum patients of 1,847 total delivered patients with HDP (1.08%) were readmitted due to HDP prior to the COVID period and 11 postpartum patients of 1,851 total patients with HDP (0.59%) who delivered during the COVID period were readmitted due to HDP (p=0.10).

During the delivery admission there was no difference in the postpartum LOS between the two groups, regardless of mode of delivery. The median postpartum LOS during the delivery admission in the control group was 2.0 days, IQR 2.0-3.0 days while the median postpartum LOS in the COVID group was 2.0 days, IQR 2.0-3.0 days (p=0.58). When analyzed by mode of delivery, the median delivery admission LOS for postpartum patients whom had vaginal deliveries in the control group was 2.0 days, IQR 2.0-2.0 days compared to 2.0 days, IQR 2.0-2.0 days in the COVID group (p=0.65). For those whom had cesarean deliveries, the median delivery admission LOS for postpartum patients in the control group was 3.0 days, IQR 2.8-3.0 days compared to 2.0 days, IQR 2.0-3.0 days in the COVID group (p=0.17). (Table 2)

Table 2

Comparison of Hypertension Management Between Cohorts

	Delivery Admission			Readmission
	Control Cohort(N=20)	COVID Cohort(N=11)	P-value	Control Cohort(N=20)	COVID Cohort(N=11)	P-value
Hypertensive Disorder in Pregnancy Diagnosisb			0.66			0.63
Chronic Hypertension	4 (20.0)	5 (45.5)		0 (0)	0 (0)
Gestational Hypertension	10 (50.0)	5 (45.5)		0 (0)	0 (0)
Preeclampsia without Severe Features	3 (15.0)	1 (9.1)		0 (0)	0 (0)
Preeclampsia with Severe Features	1 (5.0)	0 (0)		14 (70.0)	6 (54.5)
Chronic Hypertension with Superimposed Preeclampsiawithout Severe Features	0 (0)	0 (0)		0 (0)	0 (0)
Chronic Hypertension with Superimposed Preeclampsia with Severe Features	2 (10.0)	0 (0)		6 (30.0)	5 (45.5)
Eclampsia	0 (0)	0 (0)		0 (0)	0 (0)
HELLP Syndrome	0 (0)	0 (0)		0 (0)	0 (0)
Length of Stayc (days)Vaginal DeliveryCesarean Delivery	2.0 (2.0-3.0)2.0 (2.0-2.0)3.0 (2.8-3.0)	2.0 (2.0-3.0)2.0 (2.0-2.0)2.0 (2.0-3.0)	0.580.650.17	3.0 (2.0-4.0)3.0 (2.0-4.0)2.0 (2.0-4.0)	3.0 (2.0-4.0)2.5 (2.3-2.8)4.0 (2.0-4.0)	0.450.450.29
Maximum Systolic Blood Pressurec (mmHg)	156.8 ± 15.0	152.2 ± 9.7	0.50	181.3 ± 14.9	178.5 ± 12.2	0.63
Maximum Diastolic Blood Pressurec (mmHg)	94.8 ± 9.1	97.6 ± 10.5	0.68	99.6 ± 9.5	97. 6 ± 6.6	0.49
Number of Inpatient Hypertensive Crisisc	1.0 (1.0-1.0)	-	-	1.0 (1.0-2.0)	1.0 (0.5-2.0)	0.90
Use of Immediate-Acting Antihypertension Medicationb
Oral Nifedipine Immediate Release	0 (0)	0 (0)	-	0 (0)	0 (0)	-
Intravenous Labetalol	3 (15)	0 (0)	0.54	11 (64.7)	5 (45.5)	0.63
Intravenous Hydralazine	0 (0)	0 (0)	-	6 (35.3)	4 (36.4)	0.74
Time to Immediate-Acting Antihypertension Treatment of Hypertensive Crisisc (minutes)
First Crisis	9.0 (7.5-17.0)	-	-	11.0 (6.0-14.0)	7.0 (5.3-10.0)	0.14
Second Crisis	-	-	-	10.0 (7.5-11.5)	10.0 (5.0-13.0)	0.93
Third Crisis	-	-	-	8.0 (7.0-10.0)	-	-
Received Intravenous Magnesium Postpartumb	3 (15.0)	-	-	20 (100)	11 (100)	1.0
Oral Long-Acting Antihypertension Medications Used While Inpatientb
Nifedipine Extended Release	1 (5.0)	2 (18.2)	0.28	16 (80.0)	8 (72.7)	0.68
Labetalol	4 (20.0)	3 (27.3)	0.68	9 (45.0)	7 (63.6)	0.46
Furosemide	0 (0)	0 (0)	-	0 (0)	0 (0)	-
Enalapril	0 (0)	0 (0)	-	1 (5.0)	0 (0)	1.0
Hydralazine	0 (0)	0 (0)	-	0 (0)	1 (9.1)	0.35
Patients Discharged Home on Long-Acting Antihypertension Medicationb	5 (25.0)	5 (45.5)	0.42	20 (100)	10 (90.1)	0.76
Number of Oral Antihypertension Medications Discharged Onc	1.0 (1.0-1.0)	1.0 (1.0-1.0)	-	1.0 (1.0-1.3)	1.0 (1.0-1.0)	0.52
Intensive Care Unit Admissionb	0 (0)	0 (0)	-	0 (0)	1 (9.1)	0.35
Postpartum Eclampsiab	0 (0)	0 (0)	-	0 (0)	0 (0)	-
Postpartum Strokeb	0 (0)	0 (0)	-	0 (0)	0 (0)	-
Maternal Deathb	0 (0)	0 (0)	-	0 (0)	0 (0)	-

aData shown as mean ± standard deviation

n (%)

median (interquartile range)

Of the entire delivered patient population during the study period, there was no significant difference in the racial/ethnic and socioeconomic status compared to the study population. Study cohort demographics of the control and COVID groups are shown in Table 1 . Patients in the control group had higher rates of vaginal deliveries (60.0% vs. 40.0%) compared to the COVID group where patients had higher rates of cesarean deliveries (81.8% vs. 18.2%) (p=0.08). (Table 1)

Table 1

Study Demographics

	Control Cohort(N=20)	COVID Cohort(N=11)	p-value
Agea (years)	35.9 ± 5.6	32.6 ± 6.1	0.09
Advanced Maternal Ageb	11 (55.0)	3 (27.3)	0.26
Race/Ethnicityb			0.64
African-American	4 (20.0)	3 (27.3)
White	14 (70.0)	6 (54.5)
Hispanic	2 (10.0)	1 (9.1)
Asian	0 (0)	0 (0)
Other	0 (0)	1 (9.1)
Nulliparousb	14 (70.0)	9 (81.8)	0.68
Body Mass Indexc (kg/m2)	32.6 ± 7.7	33.2 ± 4.4	0.56
Obesityb (BMI >30 kg/m2)	12 (60.0)	8 (72.7)	0.70
Singleton Gestationb	19 (95.0)	10 (90.9)	1.0
Gestational Age at Deliveryc (weeks)	38.4 (37.2-39.4)	37.7 (37.5-38.6)	0.56
Preterm Deliveryb (≤36 weeks gestation)	3 (15.0)	0 (0)	0.54
Chronic Hypertensionb	6 (30.0)	5 (45.5)	0.45
On Hypertension Medicationb	5 (83.3)	4 (80)	0.68
Gestational Diabetes Mellitusb	1 (5.3)	1 (9.1)	0.61
Preexisting Diabetes Mellitusb	0 (0)	0 (0)	-
Current Smokerb	2 (10.0)	0 (0)	0.53
PreEclampsia in Previous Pregnancyb	2 (10.5)	1 (9.1)	1.0
Mode of Deliveryb			0.08
Vaginal Delivery	12 (60.0)	2 (18.2)
Cesarean Delivery	8 (40.0)	9 (81.8)
Delivery Indicationb			0.55
Medically Indicated due to HDPDiagnosis	12 (60.0)	5 (45.5)
Medically Indicated due to Other Diagnoses	3 (15.0)	3 (27.3)
Elective Delivery	3 (15.0)	3 (27.3)
Spontaneous Labor	2 (10.0)	0 (0)

Data shown as mean ± standard deviation

n (%)

median (interquartile range)

The majority of patients in each group were delivered due to a HDP diagnosis (60.0% vs. 45.5%, 0=0.55). (Table 1) Sixteen (80%) patients in the control group underwent induction of labor compared to 7 (63.6%) in the COVID group (p=0.41). Of those who underwent induction of labor, there was a higher rate of patients who had a cesarean delivery due to failed induction of labor, arrest of dilation or fetal intolerance of labor among the COVID group (71.4% (n=5) vs. 37.5% (n=6), p=0.25). All patients in the COVID group tested negative for COVID upon admission based on PCR testing.

The HDP diagnosis in each group during the delivery admission and readmission periods are described in Table 2 . During the delivery admission, there were only 3 patients with severe forms of preeclampsia (PECwSF or CHTN SIPECwSF) and no patients in the COVID group had a severe form of preeclampsia. In comparison, all patients were readmitted with severe forms of preeclampsia in both groups.

The median postpartum LOS during the delivery admission (2.0 days, IQR 2.0-3.0 days vs. 2.0 days, IQR 2.0-3.0 days, p=0.58) and during the readmission period (3.0 days, IQR 2.0-4.0 days vs. 3.0 days, IQR 2.0-4.0 days, p=0.45) were similar between groups. (Table 2) There was no difference in the median time to readmission after delivery (5.0 days, IQR 4.0-6.0 days vs 6.0 days, IQR 5.0-6.0 days, p=0.13) in the two groups.

Management of HDP between groups were overall similar. (Table 2) The mean maximum systolic blood pressure was similar in both group during the delivery admission (156.8 ± 15.0 mmHg vs. 152,2 ± 9.7 mmHg, p=0.50) and during the readmission period in both groups (181.3 ± 14.9 mmHg vs. 178.5 ± 12.2 mmHg, p=0.63). (Table 2) During the delivery admission, only 3 patients had one hypertensive crisis in the control group and no inpatient hypertensive crisis occurred in the COVID group. For the control group, the median time for administration of immediate-acting antihypertension treatment of persistent severe range blood pressures was 9.0 minutes (IQR 7.5-17.0 minutes). During the readmission period, there were patients with a range of 1 to 3 hypertensive crises in both groups. There was no significant difference in the time to administration of immediate-acting hypertension treatment during hypertensive crisis in either group. (Table 2)

Various long-acting antihypertension medications were used during the study period and there was no difference in the long-acting antihypertension medications used. (Table 2) Both groups were discharged home on similar quantities of oral long-acting antihypertension medications during each admission. (Table 2)

There was only 1 patient admitted to the ICU during the readmission period in the COVID group vs. 0 in the control group (p=0.35). This patient was readmitted to the ICU postpartum due to refractory hypertensive emergency related to preeclampsia with severe features requiring nicardipine infusion for management. There were no patients with eclampsia, stroke or maternal deaths in the entire study cohort.

Discussion

Principal Findings

After the implementation of an early discharge policy during the delivery admission there was no statistical difference in readmission rates for postpartum hypertensive disorders in pregnancy among patients with antenatal HDP.

Labor and delivery policy modifications due to the pandemic provided a unique opportunity in obstetrics to evaluate the safety of novel approaches to peripartum care. Valid concerns related to an early discharge policy include increase in postpartum readmissions, complications, and worse outcomes because hospital readmissions are associated with a high rate of morbidity and theoretically avoidable costs, particularly for patients with HDP. , This is consistent with other studies evaluating labor and delivery policy modifications due to COVID-19 that did not demonstrate an increase in adverse maternal postpartum outcomes with shorter postpartum lengths of stay in the overall obstetric population.13, 14, 15 Although there was not an overall difference in the LOS between groups during the delivery admission, our institution was able to safely implement an early discharge policy that did not increase the rate of postpartum readmission due to HDP or increase postpartum adverse outcomes such as postpartum eclampsia, stroke or death among the study cohort. As our institution's baseline cesarean delivery rate was similar between groups, there was a higher rate of cesarean deliveries in the COVID group that could potentially influence the lack of difference in the LOS amongst groups. The higher rate of cesarean deliveries amongst the COVID group could potentially be explained by the higher rate of cesarean deliveries of the patients who underwent induction of labor, most due to a HDP diagnosis.

Clinical and Research Implications

The overall baseline maternal postpartum readmission rate for all indications is 1-2%. However, medical conditions, such as HDP, are associated with higher rates of postpartum readmission. , During our study period, the readmission rate due to HDP was about 0.16%, which is lower than readmission rates from previous studies ranging from 0.4-9.3%. , Although the LOS between groups during the delivery admission were similar, an outpatient HDP management protocol for postpartum patients with HDP was continued after implementation of the early discharge policy to adhere with ACOG recommendations. It is important for healthcare institutions to implement an HDP outpatient HDP management protocol in accordance to ACOG recommendations as this is an essential component for early discharge policies. Future investigative aims should be directed towards best practices to develop postpartum discharge criteria and standardized outpatient follow-up for patients with HDP.

Strengths and Limitations

A strength of the study was the specific focus on readmissions in patients with antenatal HDP as HDP is one of the leading causes of postpartum readmissions as well as maternal morbidity and mortality. Limitations of the study include the retrospective nature of the study and the small sample size from a single institution which limits the generalizability of the results. The true sample size is limited by those without antenatal HDP and delivered patients who were readmitted to a different institution as we were unable to obtain this information. Additionally, while the COVID-19 pandemic offered an opportunity to compare groups with different postpartum LOS policies, it cannot be ignored that the pandemic caused wide reaching changes in New York's healthcare infrastructure and there may be difficulty in quantifying differences in the study cohort.

Conclusions

These findings suggest policies reducing postpartum lengths of stay can be implemented without increasing readmissions for hypertensive disorders in pregnancy. The reduction of postpartum LOS was accompanied by the continuation of a home blood pressure monitoring and introduction of outpatient hypertensive disorders in pregnancy management guidelines. Further research into optimal outpatient monitoring protocols for patients with HDP is warranted.