Abstract
Introduction:
This study aims to develop a predictive model for identifying patients at increased risk of prolonged hospital stay after robotic-assisted radical prostatectomy, either via multi-port or single-port techniques, to improve postoperative care and assist in selecting ideal candidates for same-day discharge.
Methods:
The study included patients who underwent robotic-assisted radical prostatectomy at the University of Illinois at Chicago between January 2013 and December 2022. The dataset comprises preoperative variables such as age, race, body mass index, comorbidities, prostate-specific antigen levels, Gleason score, surgical approach, and length of hospital stay. A logistic regression model was employed to predict the likelihood of a hospital stay exceeding 24 hours. The model’s performance was assessed using the area under the Receiver Operating Characteristic curve and fivefold cross-validation.
Results:
The predictive model identified significant factors associated with prolonged hospital stays, including body mass index, prostate-specific antigen levels, Gleason score, surgical approach, and specific comorbidities. The transperitoneal approach was strongly associated with lower chances of same-day discharge, demonstrating an odds ratio of 4.23 (p < 0.00001). The model achieved an accuracy of 73.7% as evaluated by cross-validation.
Discussion:
The nomogram effectively predicts the likelihood of prolonged hospital stay following robotic-assisted radical prostatectomy, enabling surgeons to make informed decisions about patient counseling, surgical planning, and postoperative management. This model may assist in identifying ideal candidates for same-day discharge while improving hospital resource utilization. Future studies should validate this model externally and explore the impact of additional perioperative and social factors.
Get full access to this article
View all access options for this article.