Prediction model for prolonged hospitalization following cesarean delivery.

INTRODUCTION

A rise in the rate of cesarean delivery (CD) has been found to be associated with a higher length of hospital stay, making it a public health concern. We aimed to evaluate risk factors for prolonged hospitalization following CD.

METHODS

A retrospective cohort study, in a single tertiary medical center, was conducted (2011-2019). Cesarean deliveries were categorized into three groups according to the postpartum length of stay (a) up to 3 days (the routine post cesarean hospital stay in our center, reference group) (b) 4-9 days, and (c) 10 days or above (prolonged hospitalization). Risk factors were examined using univariate analysis as well as multivariate logistic regression. A specific risk prediction score was developed to predict the need for prolonged hospitalization and ROC curve was assessed utilizing the performance of our model.

RESULTS

Overall, 87,424 deliveries occurred during the study period. Of them, 19,732 (22.5%) were cesarean deliveries. Hospitalization period was distributed as follows: 10,971 (55.6%) women were hospitalized for up to 3 days, 7,576 (38.4%) stayed for 4-9 days and 1,185 (6%) had a prolonged hospitalization period (≥10 days). Using multivariate analysis, multiple pregnancy (OR = 1.29, 95%CI 1.05-1.58), preterm delivery < 37 weeks (OR = 8.32, 95%CI 6.7-10.2), Apgar score < 7 (OR = 1.41, 95%CI 1.11-1.78) and non-elective CD (OR = 1.44, 95%CI 1.15-1.8) were identified as independent risk factors for prolonged hospitalization. Antenatal thrombocytopenia (PLT < 100 K) was found to be a protective factor (OR = 0.51, 95%CI 0.28-0.92). Our score model included antenatal risk factors and was found to be predicting the outcome, with an AUC of 0.845 (95%CI 0.83-0.86, p-value < 0.001).

CONCLUSION

A prediction score model for prolonged hospitalization after CD may be beneficial for risk assessment and post-partum management.