Maternal morbidity in postpartum severe preeclampsia by obstetric delivery volume

Background

Pre-eclampsia is a leading cause of maternal morbidity and mortality in the United States. Emerging data suggests that postpartum pre-eclampsia may be associated with a higher incidence of maternal morbidity compared to hypertensive disorders of pregnancy (HDP) diagnosed antenatally. Understanding postpartum maternal risk across facilities with a spectrum of obstetric services is critical with the rising rates of pre-eclampsia in all healthcare settings.

Objective

We investigated the relationship between facility delivery volume and rates of nontransfusion severe maternal morbidity (SMM) among patients readmitted postpartum for pre-eclampsia with severe features.

Study Design

This is a retrospective cohort study using the Nationwide Readmissions Database (2015–2019) of postpartum patients readmitted for pre-eclampsia with severe features. Our primary outcome was nontransfusion SMM during readmission, defined per U.S. Centers for Disease Control and Prevention criteria. We also evaluated SMM, cardiac SMM, and individual morbidities. The exposure variable was the number of annual deliveries at the readmitting facility. Restricted cubic splines with 4 knots were used to assess the functional form of the relationship between obstetric delivery volume and nontransfusion SMM; a linear relationship was identified as optimal. Logistic regression was used to estimate adjusted odds ratios (aOR) which controlled for maternal age, nontransfusion SMM at delivery, expanded obstetric comorbidity index, and HDP during delivery.

Results

The cohort included 29,472 patients readmitted with postpartum pre-eclampsia with severe features. The primary payer was 55% private and 42% governmental. Median age was 31.4 years. Most patients did not have prior HDP (65%) or chronic hypertension (86%) diagnosis antenatally. The median interval from delivery hospitalization to readmission was 3.9 days (25th percentile–75th percentile: 2.2–6.5). Nontransfusion SMM occurred in 7% of patients readmitted to facilities with >2000 deliveries compared to 9% with 1 to 2000 deliveries, and 52% without any delivery hospitalizations. The most common SMM was pulmonary edema and heart failure, observed in 4% of readmissions. We observed that for every increase in 1000 deliveries, the odds of a nontransfusion SMM at readmission decreased by 3.5% (aOR: 0.965; 95% confidence interval: 0.94, 0.99).

Conclusion

Nontransfusion SMM for postpartum readmissions with pre-eclampsia with severe features was inversely associated with readmitting hospital delivery volume. This information may guide risk-reducing initiatives for identifying strategies to optimize postpartum care at facilities with lower or no delivery volume.