Postpartum management of hypertension and effect on readmission rates

Open AccessPublished:October 29, 2021DOI:https://doi.org/10.1016/j.ajogmf.2021.100517

      BACKGROUND

      Postpartum hypertension is a source of significant morbidity and mortality in the United States. While advances have been made in the peripartum management of hypertension, there is little data to guide ongoing management postpartum.

      OBJECTIVE

      To determine whether an association exists between (1) hospital readmission and (2) hypertension in the 12 hours before discharge and the prescription of antihypertensive medications at the time of discharge. The secondary objective included evaluating the median time to readmission for hypertensive complications.

      STUDY DESIGN

      This was a retrospective cohort study of all women with peripartum hypertension at a single tertiary care center over a 3-year period (2017–2019). Peripartum hypertension was defined as any systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg on 2 occasions, 4 hours apart, in the electronic medical record during the patients’ admission for delivery. As potential risk factors for readmission, we also identified if the patients were discharged with a prescription for antihypertensive medication and assessed the blood pressure measurements during the 12 hours before discharge. The primary outcome of interest was postpartum readmission because of hypertensive complications. Readmission was defined as emergency room evaluation or hospital readmission because of hypertensive complications. Analysis was stratified into 4 comparison groups on the basis of the blood pressure and antihypertensive medications on discharge. The rate of postpartum readmissions was calculated. The risks of readmission were estimated using logistic regression and were adjusted for appropriate confounding variables.

      RESULTS

      Of 14,577 women who gave birth during the study period, 3480 (24%) met the definition of peripartum hypertension. Of those, 176 (5.1%) were readmitted within a median of 3 days from discharge. Sixty percent of patients readmitted had an International Classification of Diseases, Tenth Revision code for peripartum hypertension assigned by providers during their admission. Women with systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg before discharge were at a higher risk of readmission irrespective of being discharged with antihypertensive medication. Compared with those who were discharged normotensive, women who had hypertension in the 12 hours before discharge and were discharged with an antihypertensive prescription were at a significantly increased risk of readmission, adjusted odds ratio, 2.90; 95% confidence interval, 1.11–7.57.

      CONCLUSION

      Untreated hypertension within 12 hours before discharge was associated with a 32% higher risk of readmission in those who were not prescribed antihypertensive medications at discharge and a 3-fold increased risk of readmission in patients discharged on antihypertensive medication. These findings highlight the importance of treatment to normalize the blood pressure for at least 12 hours before discharge.

      Keywords

       Why was this study conducted?

      Recent studies have focused on improving the identification of postpartum hypertensive complications, with no data on how to manage patients with such complications. This study was meant to evaluate if the management of blood pressure at the time of discharge in those with hypertension during their hospitalization for delivery impacted their risk of readmission.

       Key findings

      Among the patients discharged with a prescription for antihypertensive medications, those discharged in a hypertensive state were almost 3 times more likely to be readmitted than their normotensive counterparts.

       The median time to readmission was 72 hours from hospital dischargeWhat does this add to what is known?

      Our data suggest that treatment for blood pressures ≤140/90 for at least 12 hours before discharge, significantly reduces the readmission risk and provides hypothesis-generating data for future interventional trials.

      Introduction

      In the United States, hypertensive disorders account for approximately 7% of maternal mortality, with approximately 70% of these deaths occurring postpartum.
      • Petersen EE
      • Davis NL
      • Goodman D
      • et al.
      Vital signs: pregnancy-related deaths, United States, 2011-2015, and strategies for prevention, 13 states, 2013-2017.
      • Say L
      • Chou D
      • Gemmill A
      • et al.
      Global causes of maternal death: a WHO systematic analysis.
      • Davis NL
      • Smoots AN
      • Goodman DA.
      Pregnancy-related deaths: data from 14 U.S. Maternal mortality review committees, 2008–2017.
      In addition, hypertension (HTN) is associated with an increased risk of postpartum readmission at a substantial societal and medical cost.
      • Davis NL
      • Smoots AN
      • Goodman DA.
      Pregnancy-related deaths: data from 14 U.S. Maternal mortality review committees, 2008–2017.
      Current guidelines, including those from the American College of Obstetricians and Gynecologists and the National Institute for Health and Care Excellence recommend the postnatal treatment of hypertension with antihypertensive medication in those with blood pressure (BP) ≥150/100 mm Hg.
      Hypertension in pregnancy
      Report of the American College of Obstetricians and Gynecologists’ Task Force on Hypertension in Pregnancy.
      ,
      National Institute for Health and Care Excellence
      Hypertension in pregnancy: diagnosis and management.
      However, a systematic review by Cairns et al
      • Cairns AE
      • Pealing L
      • Duffy JMN
      • et al.
      Postpartum management of hypertensive disorders of pregnancy: a systematic review.
      highlighted a paucity of data relating to optimal BP thresholds for initiating antihypertensive treatment in the postpartum period.
      Given the lack of data regarding the management of hypertension in the postpartum period and the significant impact that this period has on maternal morbidity and mortality, we designed this study to evaluate the impact of medical management and persistent HTN on the risk of readmission in the early postpartum period. Because most types of severe morbidity and mortality are rare, we used readmission as a surrogate that still represented a significant medical complication but occurred with a frequency common enough to be studied. We hypothesized that the adequate treatment of postpartum hypertension would reduce the risk of postpartum readmission for hypertensive complications. Therefore, the objective of this study was to determine if the presence of hypertension within 12 hours before discharge and prescription for antihypertensive medication at the time of discharge from the hospital would affect the risk of hypertensive complications and requiring readmission in the postpartum period.

      Materials and Methods

      We performed a retrospective cohort study of all the included pregnant women with peripartum hypertension (pHTN) during their admission for delivery at a single tertiary care center at the Methodist Women's Hospital in Omaha, Nebraska, over a 3-year period (2017–2019). pHTN was defined as systolic BP ≥140 mm Hg or diastolic BP ≥90 mm Hg on at least 2 occasions 4 hours apart in the electronic medical record at any point during admission for delivery. The inclusion criteria were any pregnancy during the study period in which HTN was identified on the basis of the criteria described.
      All the patients with a delivery in the previous 12 months that presented with a postpartum emergency room visit or hospital admission during the study period at any hospital in the Methodist healthcare system were identified. To identify which visits should be reviewed, these encounters were electronically screened to identify if the patient had HTN during the emergency room visit or admission using patient level data. Because of the abbreviated nature of some emergency room visits, any visit with a single elevated BP reading on the basis of previously defined cutoffs were reviewed. Each readmission underwent chart review by the study authors (T.L. and B.C.) to determine if the readmission was related to complications of hypertension or preeclampsia. Readmissions because of other complications, including injury, infection, etc were excluded from analysis. Admissions determined to be related to HTN or to HTN-related complications were included in the readmission cohort. Patients were also included in the readmission cohort if they were discharged from the emergency room visit but if the reviewers felt that the documentation was consistent with the hemolysis, elevated liver enzymes, and low platelet count syndrome, severe gestational HTN, or preeclampsia with severe features and should have resulted in readmission. Patients were evaluated but the meeting criteria for gestational HTN or preeclampsia without severe features were excluded. This study was approved by the institutional review board on December 15, 2020, approval number 1459.
      The patient demographic information was compared between those with a usual postpartum course and those who required readmission for hypertension using appropriate univariate tests. A P value of <.05 was considered statistically significant. The primary exposures of interest included documented hypertension within 12 hours of discharge and antihypertensive prescription at the time of discharge. On the basis of these 2 exposure variables, the data were stratified into 4 groups. The rates of readmission for each of the 4 groups were calculated and compared using the chi squared test. On the basis of the calculated rates of readmission, the risk of readmission associated with hypertension within 12 hours of discharge was estimated using logistic regression. The regression models were stratified by antihypertensive medications at discharge. To avoid overadjustment, the backward stepwise elimination method was used to select the appropriate confounding variables for inclusion in the regression modeling. The initial model included the gestational age at delivery, insurance status, type of pregnancy-related HTN, and the length of stay. Variables where the confidence interval crossed 1 were sequentially removed and the model was reanalyzed. The variable was completely eliminated if the odds ratio of the reanalyzed model was within a 10% difference of the previous model. The final regression model in the patients discharged without medications included the gestational age at delivery, type of hypertensive disease, and the length of stay. In the patients discharged with medications, the final regression model included the health insurance type and the type of hypertensive disease. These analyses were performed using Stata Statistical Software, version 12 (StataCorp LLC, College Station, TX).

      Results

      A total of 14,577 deliveries occurred between January 2017 and December 2019, of which 3480 (24%) met the inclusion criteria (Figure 1). During this period, 176 women were readmitted because of postpartum hypertension (Figure 2) for an overall readmission rate of 5.1%. Compared with the patients with a routine postpartum course, the patients requiring readmission delivered earlier (37 vs 38 weeks gestational age; P=.01), were less likely to rely on public insurance (15% vs 23.8%; P=.008), and were discharged sooner (2.2 vs 2.5 days; P=.04). The patients who were readmitted were more likely to have preeclampsia, on the basis of International Classification of Diseases, Tenth Revision (ICD-10) coding (59.6% vs 46.6%; P=.006). The 2 groups were otherwise similar in terms of maternal age, parity, race, multiple gestation, chronic hypertension, severity of pregnancy-related HTN, and body mass index (Table 1). The patients in the readmission cohort were seen in the emergency department or were readmitted a median of 3 days from the day of discharge (range, 0–50 days).
      Figure 1
      Figure 1Flow diagram for peripartum hypertension cohort.
      Lovgren. Management of peripartum hypertension and readmission rates. Am J Obstet Gynecol MFM 2021.
      pHTN, peripartum hypertension.
      Table 1Cohort characteristics
      Lovgren. Management of peripartum hypertension and readmission rates. Am J Obstet Gynecol MFM 2021.
      CharacteristicReadmitRoutine postpartumP value
      N1763314
      Maternal age33 (30–36)32 (29–36).18
      Gestational age at delivery37 (36–39)38 (37–39).01
      Parity2 (1–3)2 (1–2).89
      Race
       Black or African American9 (5.4)178 (5.4).36
       Caucasian143 (85.6)2624 (79.2)
       Multiple4 (2.4)171 (5.2)
       American Indian1 (0.6)25 (0.8)
       Asian4 (2.4)46 (1.4)
       Native Hawaiian05 (0.2)
       Hispanic or Latinx4 (2.4)184 (5.6)
       Other1 (0.6)27 (0.8)
       Unavailable or declined1 (0.6)54 (1.6)
      Public insurance25 (15.0)789 (23.8).008
      Multiple pregnancy11 (6.6)205 (6.2).83
      CHTN14 (8.4)245 (7.4).63
      Hypertensive disorders of pregnancy
      Determined by International Classification of Diseases, Tenth Revision, coding
      105 (59.6)1545 (46.6).006
      Type of hypertensive disorder of pregnancy
      Determined by International Classification of Diseases, Tenth Revision, coding
       None71 (40.3)1769 (53.4).002
       Without severe features67 (38.1)899 (27.1)
       With severe features
      Including hemolysis, elevated liver enzymes, and low platelet count syndrome.
      23 (13.1)462 (13.9)
       Superimposed on CHTN14 (8.0)180 (5.4)
       Eclampsia1 (0.6)4 (0.1)
      BMI34.6 (30.1–39.1)34.5 (29.9–39.7).69
      Cesarean delivery72 (43.1)1593 (48.1).21
      LOS2.2 (1.9–3.1)2.5 (1.9–3.2).04
      Data are presented as number (percentage), median (interquartile range). Normality assessed with Shapiro-Wilks test.
      BMI, body mass index; CHTN, chronic hypertension; LOS, length of stay.
      a Determined by International Classification of Diseases, Tenth Revision, coding
      b Including hemolysis, elevated liver enzymes, and low platelet count syndrome.
      Figure 2.
      Figure 2
      Figure 2Flow diagram for postpartum readmissions with hypertensive complications.
      Lovgren. Management of peripartum hypertension and readmission rates. Am J Obstet Gynecol MFM 2021.
      HTN, hypertension.
      We analyzed the 4 groups on the basis of the BP 12 hours before discharge and whether or not they were discharged with a prescription for antihypertensive medication (Table 2) as follows: (1) Normotensive, discharged with antihypertensive prescription; (2) Normotensive discharged without antihypertensive prescription; (3) Hypertensive, discharged with antihypertensive prescription; (4) Hypertensive, discharged without antihypertensive prescription. The readmission rates increased from 2% to 4.1% to 5.7% to 5.9% (P=.02) in these groups, respectively.
      Table 2Rate of readmission by antihypertensive prescription and blood pressure in the 12 hours before discharge
      Lovgren. Management of peripartum hypertension and readmission rates. Am J Obstet Gynecol MFM 2021.
      Status at DischargeReadmit, n (%)Routine care, n (%)P value
      Normotension with aHP5 (2.0)249 (98).02
      Normotension without aHP59 (4.1)1395 (95.9)
      HTN with aHP32 (5.7)534 (94.3)
      HTN without aHP71 (5.9)1136 (94.1)
      aHP, antihypertensive prescription; HTN, hypertension.
      Further analysis of these variables found that the patients who were discharged with an antihypertensive prescription and were hypertensive in the 12 hours before discharge were almost 3 times more likely to be readmitted (adjusted odds ratio, 2.90; 95% confidence interval, [1.11–7.57]) (Table 3).
      Table 3Odds of readmission stratified by antihypertensive prescription at discharge and blood pressure 12 hours before discharge
      Lovgren. Management of peripartum hypertension and readmission rates. Am J Obstet Gynecol MFM 2021.
      Comparator GroupsN (%)OR (95% CI)aOR (95% CI)
      No aHP after discharge
      Adjusted for gestational age, type of hypertensive disorder of pregnancy, length of stay
       Normotensive before discharge59 (4.1)ReferenceReference
       Hypertensive before discharge79 (6.5)1.48 (1.04–2.11)1.41 (0.99–2.01)
      aHP after discharge
      Adjusted for health insurance and type of hypertensive disorder of pregnancy.
       Normotensive before discharge5 (2.0)ReferenceReference
       Hypertensive before discharge33 (5.8)2.98 (1.15–7.75)2.90 (1.11–7.57)
      aHP, antihypertensive prescription; aOR, adjusted odds ratio; CI, confidence interval; OR, odds ratio.
      a Adjusted for gestational age, type of hypertensive disorder of pregnancy, length of stay
      b Adjusted for health insurance and type of hypertensive disorder of pregnancy.

      Discussion

       Principal findings

      In our cohort of women with pHTN, those who remained hypertensive and were discharged with a prescription for antihypertensive medication were significantly more likely to be readmitted to the hospital following discharge. These findings suggest that treatment to achieve normotension in the postpartum period may significantly reduce maternal morbidity.
      Last, the patients were readmitted in a median of 3 days, indicating that patient follow-up should occur within 24 to 48 hours after discharge if the goal is to identify illness and prevent readmission.

       Results

      There are no available data—observational or otherwise—to aid in determining the proper management of patients with hypertension in the postpartum period, despite the high prevalence of hypertensive disease and the burden on maternal morbidity and mortality. In a systematic review of antihypertensive therapy for hypertension during pregnancy and after childbirth, Firoz et al identified only 1 randomized trial of 38 subjects that assessed postpartum treatment.
      • Firoz T
      • Magee LA
      • MacDonell K
      • et al.
      Oral antihypertensive therapy for severe hypertension in pregnancy and postpartum: a systematic review.
      In addition, Sharma et al randomized 50 women to labetalol or extended release nifedipine in those with persistent postpartum hypertension (sustained BP ≥150/100 mm Hg), noting that both were effective in controlling BP; labetalol achieved control with fewer side effects with no report of how this impacted the clinical outcomes.
      • Sharma KJ
      • Greene N
      • Kilpatrick SJ.
      Oral labetalol compared to oral nifedipine for postpartum hypertension: a randomized controlled trial.
      Our hypothesis also expected antihypertensive medication to reduce to the rate of readmission, and although not statistically significant, the data trended toward a benefit in this group as well.

       Clinical implications

      When considering the management of hypertension, our study would indicate that permissive hypertension in the postpartum period is associated with an increased risk of admission, and a proactive approach to the management of hypertension may significantly impact the maternal morbidity and potentially, the mortality, in the postpartum period. Although the study is observational in nature, the data are compelling on the basis of the results. The paradigm of obstetrical patients being “cured” with delivery and permissive hypertension during the postpartum period is called into serious question.

       Research implications

      Our study provides significant hypothesis generation for future interventional trials, as this is an untapped area of research. Further trials should be utilized to confirm the findings of this study and evaluate the benefit of individual antihypertensive agents, patient compliance with antihypertensive medications, treatment side effects, and the cost-benefit analysis. Further trials should also be done to define the outpatient discharge follow-up for hypertensive patients and the appropriateness of readmission as a surrogate for more significant maternal morbidity and mortality.

       Strengths and limitations

      We acknowledge the limitations to this study, and they include the biases implicit in retrospective study design. The initial cohort (24%) is large compared with the rates of pHTN previously reported in the literature.
      • Greiner KS
      • Speranza RJ
      • Rincón M
      • Beeraka SS
      • Burwick RM.
      Association between insurance type and pregnancy outcomes in women diagnosed with hypertensive disorders of pregnancy.
      • Corrigan L
      • O'Farrell A
      • Moran P
      • Daly D
      Hypertension in pregnancy: prevalence, risk factors and outcomes for women birthing in Ireland.
      • Lo JO
      • Mission JF
      • Caughey AB.
      Hypertensive disease of pregnancy and maternal mortality.
      Because of the method used to identify the study cohort, there is a risk some patients who did not have pHTN met criteria for inclusion due to spurious BP measurements. Although it is novel, our data support our method as being clinically significant. In the cohort without an ICD-10 code for hypertension, there remained a readmission rate of 4%, indicating that although they were not clinically identified, they remained at a significant risk of complications of HTN and readmission related to hypertensive complications. Our choice of the methodology allowed the greatest possible inclusion in an area not previously researched, allowing future studies to further elucidate the preferred method.
      The sample size was also not large enough to assess the true outcomes of interest, and readmission was utilized as a surrogate. We chose readmission as a surrogate as this outcome represents a significant medical complication with implied personal and societal costs. The results were further limited by the small ‘n’ for some of the outcomes. Our data collection process was unable to account for those that may have been readmitted to another hospital, affecting the overall risk of readmission and potentially impacting the demographic characteristics of the readmission cohort. Finally, of those who were prescribed an antihypertensive medication, it is unknown if they were filled or were taken as directed.
      The major strengths of our study were the methods for selecting both the overall cohort and the readmission cohort. As we were evaluating the information that was previously unreported, we chose a method that would result in the most inclusive hypertension cohort and a method to result in the most accurate readmission group. This was accomplished by reviewing the readmission cohort at the chart level to ensure that the patients were properly included and excluded. The authors felt that this would allow our data to identify the areas to refine future research and serve as a comparator for future methodology.
      As noted previously, our methodology unexpectedly included a large portion of patients who were not clinically identified during their admission but were found to have a significant risk of readmission because of complications of HTN in the postpartum period. This group would not have been identified utilizing ICD-10 codes, and it represents a group that may benefit from further investigation.
      In contrast to the published guidelines recommending treatment for BP ≥150/100 mm Hg, we used the threshold for the treatment of BP ≥140/90 mm Hg, as this is the currently accepted threshold for the diagnosis of pregnancy-related hypertension.
      Hypertension in pregnancy
      Report of the American College of Obstetricians and Gynecologists’ Task Force on Hypertension in Pregnancy.
      ,
      National Institute for Health and Care Excellence
      Hypertension in pregnancy: diagnosis and management.
      ,
      Gestational hypertension and preeclampsia: ACOG Practice Bulletin, Number 222.
      This is further justified, given the American College of Cardiology and American Heart Association's recent recommendation changing the definition of hypertension to 130 mm Hg/80 mm Hg.
      • 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 Practice guidelines.

       Conclusions

      In summary, our data suggest that the proactive treatment of pHTN may significantly reduce readmission, and therefore, maternal morbidity. Future studies should evaluate the impact of short-term follow-up, efficacy of antihypertensive medications and validate readmission as a surrogate for more severe maternal morbidity and mortality.

      References

        • Petersen EE
        • Davis NL
        • Goodman D
        • et al.
        Vital signs: pregnancy-related deaths, United States, 2011-2015, and strategies for prevention, 13 states, 2013-2017.
        MMWR Morb Mortal Wkly Rep. 2019; 68: 423-429
        • Say L
        • Chou D
        • Gemmill A
        • et al.
        Global causes of maternal death: a WHO systematic analysis.
        Lancet Glob Health. 2014; 2: e323-e333
        • Davis NL
        • Smoots AN
        • Goodman DA.
        Pregnancy-related deaths: data from 14 U.S. Maternal mortality review committees, 2008–2017.
        Centers for Disease Control and Prevention, United States Department of Health and Human Services, Atlanta, GA2019
        • Hypertension in pregnancy
        Report of the American College of Obstetricians and Gynecologists’ Task Force on Hypertension in Pregnancy.
        Obstet Gynecol. 2013; 122: 1122-1131
        • National Institute for Health and Care Excellence
        Hypertension in pregnancy: diagnosis and management.
        2019 (Available at:) (Accessed January 22, 2021)
        • Cairns AE
        • Pealing L
        • Duffy JMN
        • et al.
        Postpartum management of hypertensive disorders of pregnancy: a systematic review.
        BMJ Open. 2017; 7e018696
        • Firoz T
        • Magee LA
        • MacDonell K
        • et al.
        Oral antihypertensive therapy for severe hypertension in pregnancy and postpartum: a systematic review.
        BJOG. 2014; 121: 1210-1218
        • Sharma KJ
        • Greene N
        • Kilpatrick SJ.
        Oral labetalol compared to oral nifedipine for postpartum hypertension: a randomized controlled trial.
        Hypertens Pregnancy. 2017; 36: 44-47
        • Greiner KS
        • Speranza RJ
        • Rincón M
        • Beeraka SS
        • Burwick RM.
        Association between insurance type and pregnancy outcomes in women diagnosed with hypertensive disorders of pregnancy.
        J Matern Fetal Neonatal Med. 2020; 33: 1427-1433
        • Corrigan L
        • O'Farrell A
        • Moran P
        • Daly D
        Hypertension in pregnancy: prevalence, risk factors and outcomes for women birthing in Ireland.
        Pregnancy Hypertens. 2021; 24: 1-6
        • Lo JO
        • Mission JF
        • Caughey AB.
        Hypertensive disease of pregnancy and maternal mortality.
        Curr Opin Obstet Gynecol. 2013; 25: 124-132
      1. Gestational hypertension and preeclampsia: ACOG Practice Bulletin, Number 222.
        Obstet Gynecol. 2020; 135: e237-e260
        • 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 Practice guidelines.
        Hypertension. 2018; 71: e13-e115