Predicting prolonged hospital stay following hip dysplasia surgery in the pediatric population

Abstract

Previous studies have shown that minimizing the length of hospital stay (LOS) following surgical procedures reduces costs and can improve the patients’ quality of life and satisfaction. However, this relationship has not been defined following operative treatment for developmental dysplasia of the hip (DDH). Therefore, the purpose of this study was to determine the most important nonmodifiable and modifiable factors that can predispose patients to require a prolonged LOS following hip dysplasia surgery. From 2012 to 2019, a national pediatric database was used to identify pediatric patients who underwent surgery for hip dysplasia. Demographic, clinical, and comorbidity variables were analyzed in a patient cohort who had a normal LOS versus one with an extended LOS using chi-square tests and analysis of variance. Statistically significant variables (P value <0.05) were inputted into an artificial neural network model to determine the level of importance. Out of 10,816 patients, 594 (5.5%) had a prolonged LOS following DDH surgery. The five most important variables to predict extended LOS following hip dysplasia surgery were increased operative time (importance = 0.223), decreased BMI (importance = 0.158), older age (importance = 0.101), increased preoperative international normalized ratio (importance = 0.096), and presence of cardiac comorbidities (importance = 0.077). Operative time, BMI, age, preoperative international normalized ratio, and cardiac comorbidities had the greatest effect on predicting prolonged LOS postoperatively. Evaluating factors that impact patients’ LOS can help optimize costs and patient outcomes.