Development of a prognostic scoring system for predicting hospitalization duration in postpartum endometritis

Abstract

Background

We aim to identify risk factors contributing to extended rehospitalizations in patients diagnosed with postpartum endometritis requiring intravenous antibiotics.

Methods

This retrospective cohort study examined postpartum endometritis patients readmitted for treatment from 2014 to 2022, comparing short (≤ 48 h) and prolonged hospitalization (> 48 h). Data included patient demographics, medical history, presentation parameters, vaginal examination findings, sonographic data, laboratory results, and details of the current labor to create a scoring system predicting prolonged hospitalization risk.

Results

During the study, 270 women with postpartum endometritis were hospitalized. Among them, 61 (22.6%) had hospital stays ≤ 48 h, while 209 (77.4%) experienced hospitalization > 48 h. Upon readmission, compared to the group with short stays, patients in the > 48 h group exhibited significantly elevated heart rates (97.9 ± 18.3 vs. 89.7 ± 12.9 bpm; p < 0.002) and CRP levels (13.8 ± 9.8 mg/dL vs. 8.1 ± 7 mg/dL; p < 0.001), respectively. Ultrasound revealed higher rates of pelvic hematoma or abscess in the > 48 h group (35.4% vs. 13.1%, respectively; p = 0.02). Multivariable logistic regression identified independent associations between hospitalization > 48 h and rupture of membranes > 14.5 h adjusted odds ratio (aOR 1.29, 95% CI 0.16–0.6, p = 0.016), temperature > 37.25 °C at readmission (aOR 1.31, 95% CI 0.013–0.42, p < 0.001), and CRP > 6.5 mg/dL at readmission (aOR 1.27, 95% CI 0.09–0.4, p = 0.002). A predictive scoring system was developed, indicating risks for prolonged hospitalization from 0.5 to 0.8.

Conclusion

The scoring system developed to predict prolonged hospitalization in postpartum endometritis can provide clinicians with valuable insights for improved diagnosis and prognosis.