Abstract
Background:
To examine the association of delivery mode with severe maternal morbidity events during delivery hospitalization among patients with pulmonary hypertension.
Methods:
This retrospective cohort study used the Premier inpatient administrative database. Patients delivering ≥ 25 weeks gestation from January 1, 2016 to September 30, 2020 and pulmonary hypertension were included. The primary analysis compared intended vaginal delivery (i.e. trial of labor) to intended cesarean delivery (intention to treat analysis). A sensitivity analysis was conducted comparing vaginal delivery to cesarean delivery (as treated analysis). The primary outcome was non-transfusion severe maternal morbidity during the delivery hospitalization. Secondary outcomes included blood transfusion (4 or more units) and readmission to the delivery hospital within 90 days from discharge from delivery hospitalization.
Results:
The cohort consisted of 727 deliveries. In the primary analysis, there was no difference in non-transfusion morbidity between intended vaginal delivery and intended cesarean delivery groups (adjusted odds ratio [aOR] 0.75, 95% confidence interval [CI]: 0.49, 1.15). In secondary analyses, intended cesarean delivery was not associated with blood transfusion (aOR 0.71, 95% CI: 0.34, 1.50) nor readmission within 90 days (aOR 0.60, 95% CI: 0.32, 1.14). In the sensitivity analysis, cesarean delivery was associated with a three-fold higher risk of non-transfusion morbidity compared to vaginal delivery (adjusted OR 2.64, 95% CI: 1.54, 3.93), a three-fold higher risk of blood transfusion (adjusted OR 3.06, 95% CI: 1.17, 7.99) and two-fold higher risk of readmission within 90 days (adjusted OR 2.20, 95% CI: 1.09, 4.46) compared to vaginal delivery.
Conclusion:
Among pregnant patients with pulmonary hypertension, a trial of labor was not associated with a higher risk of morbidity compared to an intended cesarean delivery. One third of patients who required an intrapartum cesarean delivery had a morbidity event demonstrating the increased risk of adverse events in this group.
INTRODUCTION
Pulmonary hypertension confers a substantially high risk of maternal morbidity and mortality such that pregnancy is contraindicated, and termination of pregnancy is advised.1,2 Nevertheless, some women with pulmonary hypertension carry pregnancies beyond 25 weeks. Knowledge of the current trends in peripartum care and outcomes in obstetric patients with pulmonary hypertension is limited to that presented in small case series and registry reports with limited sample sizes.3,4 Women with cardiovascular disease are often encouraged to have vaginal deliveries as the hemodynamic changes, inflammatory stress, and risks of bleeding, infection and air embolism are often lower with vaginal deliveries compared with cesarean deliveries.1,5 However, guidelines suggest that women with pulmonary hypertension should have cesarean delivery, though evidence is mounting that trial of labor is safe and may carry a lower risk of mortality in women with pulmonary hypertension.3,4,6 Rates of morbidity by delivery mode among patients with pulmonary hypertension have yet to be examined in large datasets.
The aim of this study was to examine the association between mode of delivery and severe maternal morbidity events among patients with pulmonary hypertension who carry pregnancies beyond 25 weeks’ gestation. We hypothesized that an intended vaginal delivery (i.e. labor or a trial of labor) has lower rates of morbidity compared with intended cesarean delivery.
METHODS
Study Design and Population
This study was deemed exempt from review by the Institutional Review Board. The requirement for written informed consent was waived by the Institutional Review Board. We conducted a retrospective cohort study using a large administrative database that represents approximately 25% of all discharges from in-hospital births in the United States (Premier Healthcare Database, Premier Inc., Charlotte, North Carolina). The Premier Healthcare Database is a de-identified data set and contains patient demographic information, length of stay, readmissions, hospital characteristics, and International Classification of Disease (10th edition) diagnosis and procedure codes, Healthcare Common Procedure Coding System Codes and Current Procedural Terminology codes. Premier contains a detailed date-stamped record of all patient-specific billing information, including procedures, medications, and diagnostic and therapeutic services. The readmission data tables in Premier allow for linkage with hospital encounter data, facilitating identification of postpartum readmissions to the delivery hospital. Data for readmissions to the delivery hospital were available for the full 90 days after all deliveries. This study follows the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.7
The cohort for this study consisted of patients 12 to 55 years old, who had both an International Classification of Diseases, Tenth Edition (ICD-10) diagnosis code for delivery after 25 weeks’ gestation (Z3A.25-Z3A.42) and for pulmonary hypertension (I27.0, I27.20–24, I27.29, I27.81–83, I27.89, I27.9) (Figure 1). Planned vaginal delivery was defined as an ICD-10 code for vaginal delivery (O80 or Z37.x or procedure codes: 10E0XZZ, 10D07Z3, 10D07Z4, 10D07Z5, 10D07Z6, 10D07Z7, 10D07Z8) and cesarean delivery was defined by presence of an ICD-10 procedure codes (10D00Z0, 10D00Z1, 10D00Z2) from January 1, 2016, to September 30, 2020. Patients with ICD-10 codes consistent with a delivery (i.e. Z37.*, O80, 10E0XZZ, 0W8NXZZ) were also classified as vaginal deliveries if no ICD-10 codes for cesarean delivery were present. Operative deliveries (e.g., vacuum or forceps) were classified as vaginal deliveries unless a cesarean delivery code was also present.
Figure 1:
Consort diagram of cohort creation
Exposures, Outcomes, and Covariates
Our exposure variable of interest was intended delivery mode (Supplemental Table 1). Clinically, when counseling patients, it is important to draw a distinction between intended mode of delivery and actual mode of delivery. The decision to proceed with attempt for a vaginal delivery can end in a vaginal delivery or result in an intrapartum cesarean delivery. Thus, we used an intention to treat approach and considered the primary comparison to be intended vaginal delivery, which included both vaginal delivery and intrapartum cesarean delivery versus intended cesarean delivery, in which we included patients without a diagnosis code consistent with intrapartum cesarean delivery. Intrapartum cesarean delivery was defined as presence of an ICD-10 code for fetal distress, failed operative delivery, cord prolapse, fetal-maternal disproportion, obstructed labor, abnormal forces of labor, long labor, and failed induction (Supplemental Table 1). We also performed a sensitivity analysis in which the final mode of delivery was vaginal was compared to all cesarean delivery including intrapartum cesarean delivery (as treated analysis).
The primary outcome was defined a priori as a diagnosis of any non-transfusion severe maternal morbidity, as defined by the Centers for Disease Control and Prevention (CDC), during the delivery hospitalization.8 We adjusted the CDC definition to include acute events that were not included in the original definition, such as acute heart failure and aortic dissection codes and removed codes deemed to reflect chronic illness not acute morbidity events (e.g., Moya Moya disease) (Supplemental Table 2).8 Only morbidity events not present on admission were included as outcomes. Secondary maternal outcomes included blood transfusion of 4 or more units of packed red blood cells and readmission to delivery hospital within 90 days. Cardiovascular treatments provided at any point in the delivery hospitalization such as vasopressors, inotropes, pulmonary vasodilators and extracorporeal membrane oxygenator (ECMO) use were extracted and presented descriptively as they were used in small numbers of patients, so there was no ability to adjust the models for these outcomes. In-hospital mortality, intensive care stay, and length of stay are similarly presented descriptively.
Statistical Analysis
Multivariable logistic regression models were used to compare outcomes (primary: any non-transfusion severe maternal morbidity, secondary: blood transfusion of 4 or more units of packed red blood cells and readmission to delivery hospital within 90 days) by intended and actual mode of delivery. Covariates for the logistic regression models were selected a priori based on known maternal morbidity risk factors, prior literature, and subject-matter expertise of the authors and included: age, payor category and comorbidities present in the expanded obstetric comorbidity index with some adjustments (Supplemental Table 3).9
Demographic and clinical characteristics of the cohort are reported, stratified by delivery mode. Descriptive statistics were used to examine the study population with categorical variables reported as counts and frequencies, and continuous data reported as median with interquartile range. The results are reported as odds ratio with 95% confidence intervals. Statistical analyses were performed in the SAS System, Version 9.4 (SAS Institute, Cary, North Carolina). A two-sided alpha level of 0.05 was pre-specified as statistically significant.
RESULTS
The cohort consisted of 727 deliveries (vaginal delivery: n=238, intrapartum cesarean delivery: n=171, intended cesarean delivery: n=318). Baseline demographics and comorbidities are presented in Table 1. Overall, patients who had prior cesarean delivery, multiple gestation, gestational hypertension, preterm delivery and cardiomyopathy were more likely to be delivered via intended cesarean delivery. All patients with placenta accreta spectrum were delivered via intended cesarean delivery in this cohort.
Table 1:
Baseline characteristics of the pulmonary hypertension cohort
| Vaginal delivery (n=238) | Intrapartum cesarean delivery (n=171) | Intended cesarean delivery (n=318) | Intended vaginal delivery (Vaginal delivery + Intrapartum cesarean delivery) (n=409) | All cesarean delivery (Intrapartum + Intended cesarean delivery) (n=489) | |
|---|---|---|---|---|---|
| Demographic Characteristics | |||||
| Age, years (median, [Q1, Q3]) | 30 [25, 34] | 30 [26, 35] | 32 [28, 36] | 30 [25, 34] | 32 [27, 36] |
| Primary Payor | |||||
| Managed care | 50 (21.0%) | 44 (25.7%) | 96 (30.2%) | 94 (23.0%) | 140 (28.6%) |
| Medicaid | 145 (60.9%) | 93 (54.4%) | 171 (53.8%) | 238 (58.2%) | 264 (54.0%) |
| Medicare | 7 (2.9%) | 6 (3.5%) | 18 (5.7%) | 13 (3.2%) | 24 (4.9%) |
| Other | 36 (15.1%) | 28 (16.4%) | 33 (10.4%) | 64 (15.6%) | 61 (12.5%) |
| Race | |||||
| Asian | 10 (4.2%) | 9 (5.3%) | 8 (2.5%) | 19 (4.6%) | 17 (3.5%) |
| Black | 69 (29.0%) | 64 (37.4%) | 100 (31.4%) | 133 (32.5%) | 164 (33.5%) |
| Other | 35 (14.7%) | 25 (14.6%) | 45 (14.2%) | 60 (14.7%) | 70 (14.3%) |
| Unknown | 9 (3.8%) | 9 (5.3%) | 13 (4.1%) | 18 (4.4%) | 22 (4.5%) |
| White | 115 (48.3%) | 64 (37.4%) | 152 (47.8%) | 179 (43.8%) | 216 (44.2%) |
| Ethnicity | |||||
| Hispanic | 30 (12.6%) | 21 (12.3%) | 46 (14.5%) | 51 (12.5%) | 67 (13.7%) |
| Non-Hispanic | 151 (63.4%) | 109 (63.7%) | 211 (66.4%) | 260 (63.6%) | 320 (65.4%) |
| Unknown | 57 (23.9%) | 41 (24.0%) | 61 (19.2%) | 98 (24.0%) | 102 (20.9%) |
| Hospital Characteristics | |||||
| Urban hospital | 216 (90.8%) | 162 (94.7%) | 289 (90.9%) | 378 (92.4%) | 451 (92.2%) |
| Teaching hospital | 176 (73.9%) | 118 (69.0%) | 225 (70.8%) | 294 (71.9%) | 343 (70.1%) |
| Bed size | |||||
| 000–099 | 7 (2.9%) | 2 (1.2%) | 3 (0.9%) | 9 (2.2%) | 5 (1.0%) |
| 100–199 | 13 (5.5%) | 10 (5.8%) | 18 (5.7%) | 23 (5.6%) | 28 (5.7%) |
| 200–299 | 23 (9.7%) | 20 (11.7%) | 30 (9.4%) | 43 (10.5%) | 50 (10.2%) |
| 300–399 | 32 (13.4%) | 29 (17.0%) | 45 (14.2%) | 61 (14.9%) | 74 (15.1%) |
| 400–499 | 32 (13.4%) | 23 (13.5%) | 39 (12.3%) | 55 (13.4%) | 62 (12.7%) |
| 500+ | 131 (55.0%) | 87 (50.9%) | 183 (57.5%) | 218 (53.3%) | 270 (55.2%) |
| Region of country | |||||
| Midwest | 60 (25.2%) | 41 (24.0%) | 62 (19.5%) | 101 (24.7%) | 103 (21.1%) |
| Northeast | 43 (18.1%) | 36 (21.1%) | 50 (15.7%) | 79 (19.3%) | 86 (17.6%) |
| South | 97 (40.8%) | 74 (43.3%) | 155 (48.7%) | 171 (41.8%) | 229 (46.8%) |
| West | 38 (16.0%) | 20 (11.7%) | 51 (16.0%) | 58 (14.2%) | 71 (14.5%) |
| Operative vaginal delivery | 29 (12.2%) | NA | NA | 29 (7.1%) | N/A |
| Comorbidities | |||||
| Gestational Diabetes | 32 (13.4%) | 21 (12.3%) | 42 (13.2%) | 53 (13.0%) | 63 (12.9%) |
| HIV/AIDS | 0 (0.0%) | 2 (1.2%) | 3 (0.9%) | 2 (0.5%) | 5 (1.0%) |
| Preexisting Diabetes | 15 (6.3%) | 14 (8.2%) | 44 (13.8%) | 29 (7.1%) | 58 (11.9%) |
| Prior cesarean birth | 9 (3.8%) | 29 (17.0%) | 154 (48.4%) | 38 (9.3%) | 183 (37.4%) |
| Multiple gestation | 3 (1.3%) | 4 (2.3%) | 19 (6.0%) | 7 (1.7%) | 23 (4.7%) |
| Asthma | 70 (29.4%) | 44 (25.7%) | 78 (24.5%) | 114 (27.9%) | 122 (24.9%) |
| Bleeding disorder | 17 (7.1%) | 19 (11.1%) | 19 (6.0%) | 36 (8.8%) | 38 (7.8%) |
| BMI ≥ 40 kg/m2 | 10 (4.2%) | 15 (8.8%) | 28 (8.8%) | 25 (6.1%) | 43 (8.8%) |
| Chronic hypertension | 38 (16.0%) | 39 (22.8%) | 84 (26.4%) | 77 (18.8%) | 123 (25.2%) |
| Hypertensive crisis | 1 (0.4%) | 2 (1.2%) | 5 (1.6%) | 3 (0.7%) | 7 (1.4%) |
| Chronic renal disease | 10 (4.2%) | 10 (5.8%) | 18 (5.7%) | 20 (4.9%) | 28 (5.7%) |
| Connective tissue or autoimmunedisease | 8 (3.4%) | 5 (2.9%) | 15 (4.7%) | 13 (3.2%) | 20 (4.1%) |
| Substance use disorder | 52 (21.8%) | 36 (21.1%) | 56 (17.6%) | 88 (21.5%) | 92 (18.8%) |
| Advanced maternal age | 47 (19.7%) | 40 (23.4%) | 85 (26.7%) | 87 (21.3%) | 125 (25.6%) |
| Preexisting anemia | 66 (27.7%) | 59 (34.5%) | 105 (33.0%) | 125 (30.6%) | 164 (33.5%) |
| Bariatric surgery | 2 (0.8%) | 3 (1.8%) | 7 (2.2%) | 5 (1.2%) | 10 (2.0%) |
| Gastrointestinal disease | 35 (14.7%) | 26 (15.2%) | 62 (19.5%) | 61 (14.9%) | 88 (18.0%) |
| Mental health disorder | 47 (19.7%) | 33 (19.3%) | 63 (19.8%) | 80 (19.6%) | 96 (19.6%) |
| Neuromuscular disease | 23 (9.7%) | 21 (12.3%) | 43 (13.5%) | 44 (10.8%) | 64 (13.1%) |
| Placental abruption | 2 (0.8%) | 3 (1.8%) | 4 (1.3%) | 5 (1.2%) | 7 (1.4%) |
| Placenta accreta spectrum | 0 (0.0%) | 0 (0.0%) | 5 (1.6%) | 0 (0.0%) | 5 (1.0%) |
| Preeclampsia/gestational hypertension | 47 (19.7%) | 51 (29.8%) | 85 (26.7%) | 98 (24.0%) | 136 (27.8%) |
| Preeclampsia with severe features/HELLP | 38 (16.0%) | 33 (19.3%) | 62 (19.5%) | 71 (17.4%) | 95 (19.4%) |
| Preterm delivery | 77 (32.4%) | 76 (44.4%) | 170 (53.5%) | 153 (37.4%) | 246 (50.3%) |
| Thyrotoxicosis | 10 (4.2%) | 3 (1.8%) | 6 (1.9%) | 13 (3.2%) | 9 (1.8%) |
| Congenital heart disease and Eisenmenger syndrome | 29 (12.2%) | 11 (6.4%) | 37 (11.6%) | 40 (9.8%) | 48 (9.8%) |
| Valvular heart disease | 72 (30.3%) | 50 (29.2%) | 113 (35.5%) | 122 (29.8%) | 163 (33.3%) |
| Cardiomyopathy | 43 (18.1%) | 26 (15.2%) | 77 (24.2%) | 69 (16.9%) | 103 (21.1%) |
Morbidity events were common across all modes of delivery. At least one morbidity even was present in 14.7% of vaginal deliveries, 38.6% of intrapartum cesarean deliveries, and 24.5% of intended cesarean deliveries. The most common morbidity events were acute respiratory distress syndrome (9.8%), mechanical ventilation (6.3%), acute heart failure (5.1%), acute renal failure (5.0%), sepsis (3.7%), shock (3.4%), disseminated intravascular coagulation, (2.8%) and pulmonary edema (2.6%) (Table 2).
Table 2:
Mortality and Morbidity Events at delivery hospitalization by delivery mode
| Vaginal delivery (n=238) | Intrapartum cesarean delivery (n=171) | Not Labored cesarean delivery (n=318) | Intended vaginal delivery (vaginal delivery + Intrapartum cesarean delivery) (n=409) | Cesarean delivery (Intrapartum + Intended cesarean delivery) (n=489) | Intended: p-value Intended vaginal delivery vs. Intended cesarean delivery | As treated: p-value actual vaginal delivery vs. intrapartum + Intended cesarean delivery | |
|---|---|---|---|---|---|---|---|
| In-hospital Mortality | 0 (0.0%) | <10 | <10 | <10 | <10 | ||
| Non-Transfusion morbidity | 35 (14.7%) | 66 (38.6%) | 78 (24.5%) | 101 (24.7%) | 144 (29.4%) | 0.96 | <0.001 |
| Individual Morbidity Elements | |||||||
| Acute myocardial infarction | 0 (0.0%) | 0 (0.0%) | <10 | 0 (0.0%) | <10 | ||
| Dissection (not aneurysm) | 0 (0.0%) | 0 (0.0%) | <10 | 0 (0.0%) | <10 | ||
| Acute renal failure | <10 | 16 (9.4%) | 18 (5.7%) | 18 (4.4%) | 34 (7.0%) | ||
| Acute respiratory distress syndrome | 10 (4.2%) | 29 (17.0%) | 32 (10.1%) | 39 (9.5%) | 61 (12.5%) | ||
| Amniotic fluid embolism | 0 (0.0%) | <10 | <10 | <10 | <10 | ||
| Cardiac arrest | 0 (0.0%) | <10 | <10 | <10 | <10 | ||
| Ventricular fibrillation | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | ||
| Conversion of cardiac rhythm | <10 | <10 | <10 | <10 | <10 | ||
| Disseminated intravascular coagulation | <10 | <10 | <10 | 12 (2.9%) | 16 (3.3%) | ||
| Heart failure or arrest during surgery or procedure | 0 (0.0%) | <10 | <10 | <10 | <10 | ||
| Puerperal cerebrovascular disorders | <10 | <10 | <10 | <10 | <10 | ||
| Pulmonary edema | <10 | <10 | <10 | 13 (3.2%) | 14 (2.9%) | ||
| Acute heart failure | 11 (4.6%) | <10 | 17 (5.3%) | 20 (4.9%) | 26 (5.3%) | ||
| Severe anesthesia complications | <10 | <10 | <10 | <10 | <10 | ||
| Sepsis | <10 | 12 (7.0%) | <10 | 20 (4.9%) | 19 (3.9%) | ||
| Shock | <10 | 11 (6.4%) | <10 | 16 (3.9%) | 20 (4.1%) | ||
| Sickle cell disease with crisis | 0 (0.0%) | <10 | <10 | <10 | <10 | ||
| Air and thrombotic embolism | <10 | <10 | <10 | 11 (2.7%) | 15 (3.1%) | ||
| Temporary tracheostomy | 0 (0.0%) | 0 (0.0%) | <10 | 0 (0.0%) | <10 | ||
| Ventilation | <10 | 19 (11.1%) | 23 (7.2%) | 23 (5.6%) | 42 (8.6%) | ||
| Hysterectomy | <10 | <10 | <10 | <10 | <10 | ||
| Intensive care stay | 40 (16.8%) | 56 (32.7%) | 117 (36.8%) | 96 (23.5%) | 173 (35.4%) | ||
| Length of stay after delivery | 1 [1, 2] | 3 [2, 5] | 3 [2, 5] | 2 [1, 3] | 3 [2, 5] | ||
| Hospital Readmission (n,%) | |||||||
| 42-day Readmission | <10 | 25 (14.6%) | 25 (7.9%) | 34 (8.3%) | 50 (10.2%) | ||
| 90-day Readmission | 12 (5.0%) | 27 (15.8%) | 30 (9.4%) | 39 (9.5%) | 57 (11.7%) | 0.96 | 0.004 |
| Vasopressor use (any) * | <10 | 18 (10.5%) | 36 (11.3%) | 23 (5.6%) | 54 (11.0%) | ||
| Norepinephrine | <10 | 11 (6.4%) | 17 (5.3%) | 16 (3.9%) | 28 (5.7%) | ||
| Vasopressin | 0 (0.0%) | 15 (8.8%) | 26 (8.2%) | 15 (3.7%) | 41 (8.4%) | ||
| Both | 0 (0.0%) | <10 | <10 | <10 | 15 (3.1%) | ||
| Inotrope use (any) † | <10 | <10 | 23 (7.2%) | 11 (2.7%) | 32 (6.5%) | ||
| Epinephrine | 0 (0.0%) | <10 | <10 | <10 | <10 | ||
| Dobutamine | <10 | <10 | 10 (3.1%) | <10 | 14 (2.9%) | ||
| Dopamine | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | ||
| Milrinone | <10 | <10 | 11 (3.5%) | <10 | 14 (2.9%) | ||
| More than one inotrope | 0 (0.0%) | 0 (0.0%) | <10 | 0 (0.0%) | <10 | ||
| ECMO use | 0 (0.0%) | <10 | <10 | <10 | <10 | ||
| Any pulmonary vasodilator | 48 (20.2%) | 49 (28.7%) | 107 (33.6%) | 97 (23.7%) | 156 (31.9%) | ||
| Calcium Channel Blockers | 41 (17.2%) | 40 (23.4%) | 87 (27.4%) | 81 (19.8%) | 127 (26.0%) | ||
| Prostacyclin Analogues | <10 | <10 | 15 (4.7%) | 11 (2.7%) | 23 (4.7%) | ||
| Phosphodiesterase-5 Inhibitors | <10 | <10 | 18 (5.7%) | 10 (2.4%) | 23 (4.7%) | ||
| Obstetric morbidity | |||||||
| Blood transfusion <4 units red blood cells | 27 (11.3%) | 36 (21.1%) | 50 (15.7%) | 63 (15.4%) | 86 (17.6%) | ||
| Blood transfusion 4 or more units red blood cells | <10 | 19 (11.1%) | 24 (7.5%) | 26 (6.4%) | 43 (8.8%) | 0.53 | 0.003 |
Norepinephrine or vasopressin or both
Epinephrine, dobutamine, dopamine and, or milrinone
Values <10 are suppressed
In the primary analysis (intention to treat), where intrapartum cesarean delivery was considered in the vaginal delivery group, after adjustment for age, payer category, and comorbidities, there was no difference in the risk of non-transfusion morbidity between the intended vaginal delivery and intended cesarean delivery groups (adjusted OR 0.75 95% CI: 0.49, 1.15). No patients with placenta previa had the morbidity composite outcome, so this comorbidity was not included in the model. The model results are summarized in Figure 2 and Supplemental Table 4.
Figure 2:
Summary of model results. Panel A: Intention to treat analysis (planned CD versus planned vaginal delivery) Panel B: As treated analysis (CD versus VD)
In the sensitivity analysis (as treated), where intrapartum cesarean delivery was included in the cesarean delivery group, after adjustment for age, payor category, and comorbidities, cesarean delivery was associated with at least a 50% increase in risk of non-transfusion morbidity compared to vaginal delivery based on the lower bound of the 95% CI (adjusted OR 2.46, 95% CI: 1.54, 3.93). The sensitivity analysis model results are summarized in Figure 2 and Supplemental Table 5.
Examining secondary outcomes, in the intention to treat analysis, cesarean delivery was not associated with blood transfusion of 4 or more units (adjusted OR 0.71, 95% CI: (0.34, 1.50) nor readmission within 90 days (adjusted OR 0.60, 95% CI: 0.32–1.14). In the sensitivity analysis (as treated), cesarean delivery was associated with a three-fold higher risk of blood transfusion of 4 or more units (adjusted OR 3.06, 95% CI: 1.17, 7.99) and two-fold higher risk of readmission within 90 days (adjusted OR 2.20, 95% CI: 1.09, 4.46) compared to vaginal delivery. Secondary outcomes are summarized in Table 2 and Figure 2. The cumulative incidence of readmissions, by mode of delivery is presented in Figure 3. The 90-day readmission rates were 5.0% for the vaginal delivery group, 15.8% for the intrapartum cesarean delivery group and 9.4% for the non-labored cesarean delivery group.
Figure 3:
Cumulative incidence of readmissions, by mode of delivery
More patients who had labored or intended cesarean delivery required norepinephrine, vasopressin or inotropic medications during the delivery hospitalization compared with patients who had vaginal deliveries (Table 2). Pulmonary vasodilators such as calcium channel blockers and prostacyclin analogues were more common among patients who had labored, or intended cesarean deliveries compared with patients who had vaginal deliveries (Table 2).
DISCUSSION
This large retrospective cohort study of pregnant patients with pulmonary hypertension demonstrated no differences in the risk of morbidity between patients who had a trial of labor and patients who delivered via intended cesarean delivery. When all cesarean delivery patients were compared to vaginal delivery patients, there was a 3-fold increase in risk of morbidity, and similar increases in rates of blood transfusion. A trial of labor is the preferred mode of delivery for patients with cardiac disease, yet cesarean deliveries are preferred by some clinicians for selected high-risk conditions such as pulmonary hypertension due to the concern for increased risk associated with unscheduled cesarean deliveries as well as the unpredictable length of labor.1,6 Our results suggest that either strategy - a trial of labor or planned cesarean delivery, may be reasonably considered for many patients with pulmonary hypertension. However, it should be noted that there is increased risk of morbidity among patients who go on to require intrapartum cesarean deliveries.
Data from the Registry on Pregnancy and Cardiac Disease (ROPAC) registry demonstrated higher rates of maternal complications from a strategy of planned cesarean delivery, when compared with planned vaginal delivery, among patients delivering with maternal cardiac disease.10 Our results which focus solely on maternal patients with pulmonary hypertension and differ from these results in that when planned vaginal delivery is compared with planned cesarean delivery event rates appear broadly equivalent. To our knowledge, ROPAC data has yet to be applied the subpopulation of maternal cardiac patients with pulmonary hypertension to compare outcomes based on intended route of delivery.3 Thus difference in results may be driven by our specific focus on pulmonary hypertension. Alternatively, with respect to data sources, ROPAC as a prospectively gathered registry contains more clinically granular data than what is available in administrative datasets, including the datasets used for this study, and this may alter findings. Additionally, with respect to generalizability, ROPAC reflects the care and outcomes of delivered in facilities electing to participate in the registry, while the PHD is a collaboration of approximately 25% of inpatient admissions in the United States who provide data.
Strengths of our study include our large sample size; that our data were drawn from a large geographically diverse all-payer hospital administrative database, which is likely to be generalizable of United States practice; and that we were able to look to 90 day rehospitalization data. However, some limitations should be noted. Out-of-hospital death and readmissions to non-delivery hospitals were not available in this dataset, resulting in possible underreporting of these outcomes. The high 90-day readmission rate in this cohort highlights the intensive care and follow-up required to prevent adverse outcomes in these patients. Reliance on diagnosis and procedure codes for most of our variable definitions could result in uncertainty. Of note, we did use previously validated approaches such as using existing definitions of morbidity and comorbid conditions when possible, adapting for the unique needs of this paper. While we included a robust set of covariables, residual confounding may exist. Finally, this dataset does not allow for examination of severity of pulmonary hypertension which influences outcomes in these patients.4 Another limitation is that it is possible that variation in processes of care at individual delivery hospitals may be a confounder that we have not accounted for in this analysis. As we are limited by sample size and event rate among individual facilities, we cannot control for this analytically, but in a sensitivity analysis we added teaching status, urban versus rural status, bed count, and provider region to the models and both analyses intended mode of delivery and actual mode of delivery are very similar to findings from the main analyses without hospital characteristics. A limitation of this work is that we did not have access to granular data (vital signs, fetal heart tracings, progress notes, etc..) surrounding the decision to perform a planned cesarean delivery versus planned vaginal delivery. This limitation makes the analysis prone to possible residual confounding and precluded an additional propensity matched sensitivity analysis. Neonatal outcomes were not available in this dataset.
Our results should be confirmed in studies with observational designs with greater clinical detail about severity and type of pulmonary hypertension, right heart function and where delivery mode and outcomes can be verified. Given the high degree of morbidity associated with intrapartum cesarean delivery, efforts to predict which patients will require an intrapartum cesarean delivery may guide when a planned cesarean delivery may be beneficial. Whether the risk factors for needing cesarean delivery overlap with risk factors for morbidity or whether the cesarean delivery itself causes morbidity remains unknown. A combination of the two mechanisms could explain the higher rate of morbidity among patients who require intrapartum cesarean delivery.
Delivery planning for patients with pulmonary hypertension is best achieved with a multidisciplinary approach. These findings suggest that when evaluating potential delivery modality for individual patients with pulmonary hypertension, pregnancy heart teams may reasonably consider both planned vaginal or cesarean delivery as potential options. For example, severity of pulmonary hypertension and the need to have appropriate pregnancy heart team members available (e.g., cardiothoracic anesthesiology and ECMO teams) influence the decision to perform a planned cesarean delivery; our data suggest this is a similarly safe approach.
This large retrospective study examining risk of morbidity by mode of delivery in patients with pulmonary hypertension demonstrated comparable risk of morbidity between patients who had trials of labor and patients who had intended cesarean delivery. Thus, vaginal delivery is considered the preferred mode of delivery for patients with pulmonary hypertension, similar to many other cardiac diseases. Planned cesarean delivery can be reasonably offered in appropriately selected and counseled patients or when indicated by delivery resource needs.
Supplementary Material
KEY POINTS.
Question: Is there an association between delivery mode and severe maternal morbidity events among patients with pulmonary hypertension?
Findings: In this retrospective cohort study that included 727 deliveries in people with pulmonary hypertension, there was no difference in non-transfusion morbidity between intended vaginal delivery and intended cesarean delivery groups (adjusted odds ratio 0.75, 95% confidence interval: 0.49, 1.15).
Meaning: Our results suggest that many patients with pulmonary hypertension may be reasonably offered a trial of labor, but given the increased risks associated with unplanned cesarean delivery, obstetric assessment of the patients probability of achieving vaginal delivery plays in important role in shared decision making.
Acknowledgements:
MLM receives grant support from the Foundation for Anesthesia Education and Research, Mentored Research Training Grant. JJF is supported by grant K12HD103083 from the National Institute of Child Health and Human Development (NICHD) of the U.S. National Institutes of Health. ME is supported by grant 1K01MH127309. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. We presented preliminary data analysis as an abstract (poster presentation) and as an oral presentation at the Society of Cardiovascular Anesthesiologists in May 2022, in Palm Springs, CA.
Glossary of Terms
- ICD-10
International Classification of Diseases, Tenth Edition
- CDC
Centers for Disease Control and Prevention
- ECMO
Extracorporeal membrane oxygenator
- ROPAC
Registry on Pregnancy and Cardiac Disease
Footnotes
Conflicts of interest: None
REFERENCES
- 1.Regitz-Zagrosek V, Roos-Hesselink JW, Bauersachs J, Blomstrom-Lundqvist C, Cifkova R, De Bonis M, Iung B, Johnson MR, Kintscher U, Kranke P, Lang IM, Morais J, Pieper PG, Presbitero P, Price S, Rosano GMC, Seeland U, Simoncini T, Swan L, Warnes CA, E. S. C. Scientific Document Group: 2018 ESC Guidelines for the management of cardiovascular diseases during pregnancy. Eur Heart J 2018; 39: 3165–3241 [DOI] [PubMed] [Google Scholar]
- 2.Thomas E, Yang J, Xu J, Lima FV, Stergiopoulos K: Pulmonary Hypertension and Pregnancy Outcomes: Insights From the National Inpatient Sample. J Am Heart Assoc 2017; 6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Sliwa K, van Hagen IM, Budts W, Swan L, Sinagra G, Caruana M, Blanco MV, Wagenaar LJ, Johnson MR, Webb G, Hall R, Roos-Hesselink JW, ROPAC Investigators: Pulmonary hypertension and pregnancy outcomes: data from the Registry Of Pregnancy and Cardiac Disease (ROPAC) of the European Society of Cardiology. Eur J Heart Fail 2016; 18: 1119–28 [DOI] [PubMed] [Google Scholar]
- 4.Meng ML, Landau R, Viktorsdottir O, Banayan J, Grant T, Bateman B, Smiley R, Reitman E: Pulmonary Hypertension in Pregnancy: A Report of 49 Cases at Four Tertiary North American Sites. Obstet Gynecol 2017; 129: 511–520 [DOI] [PubMed] [Google Scholar]
- 5.Easter SR, Rouse CE, Duarte V, Hynes JS, Singh MN, Landzberg MJ, Valente AM, Economy KE: Planned vaginal delivery and cardiovascular morbidity in pregnant women with heart disease. Am J Obstet Gynecol 2020; 222: 77 e1–77 e11 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hemnes AR, Kiely DG, Cockrill BA, Safdar Z, Wilson VJ, Al Hazmi M, Preston IR, MacLean MR, Lahm T: Statement on pregnancy in pulmonary hypertension from the Pulmonary Vascular Research Institute. Pulm Circ 2015; 5: 435–65 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, Initiative S: The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007; 370: 1453–7 [DOI] [PubMed] [Google Scholar]
- 8.Centers for Disease Control, National Center for Chronic Disease Prevention and Health Promotion, Division of Reproductive Health. Severe Maternal Morbidity in the United States. https://www.cdc.gov/reproductivehealth/maternalinfanthealth/severematernalmorbidity.html. 2021. Accessed.
- 9.Leonard SA, Kennedy CJ, Carmichael SL, Lyell DJ, Main EK: An Expanded Obstetric Comorbidity Scoring System for Predicting Severe Maternal Morbidity. Obstet Gynecol 2020; 136: 440–449 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ruys TP, Roos-Hesselink JW, Pijuan-Domenech A, Vasario E, Gaisin IR, Iung B, Freeman LJ, Gordon EP, Pieper PG, Hall R, Boersma E, Johnson MR, investigators R: Is a planned caesarean section in women with cardiac disease beneficial? Heart 2015; 101: 530–6 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.