Prediction of Postoperative Urinary Tract Infection Following Benign Gynecologic Surgery.

Area of Science:

• Urology
• Surgical Outcomes
• Predictive Modeling

Background:

• Urinary tract infections (UTIs) are a common complication following pelvic surgery.
• Accurate prediction of post-operative UTI risk is crucial for patient management and preventative strategies.
• Existing models may not fully capture the complexity of risk factors in this patient population.

Purpose of the Study:

• To develop and validate a predictive model for urinary tract infection (UTI) occurring within 8 weeks after pelvic surgery.
• To evaluate the performance of various machine learning algorithms in predicting post-operative UTIs.
• To identify key clinical and procedural variables associated with UTI risk.

Main Methods:

• Retrospective analysis of data from 1,657 women undergoing pelvic surgery across three tertiary care centers.
• Utilized machine learning algorithms including logistic regression, decision trees (DTs), naive Bayes (NB), random forest (RF), gradient boosting (GB), and multilayer perceptron (MP).
• Internal validation using two datasets and external validation using a third dataset to assess model generalizability.

Main Results:

• The overall incidence of UTI was 10.4% in the training set and 16.1% in the external validation set.
• Gradient boosting (GB), decision trees (DT), and random forest (RF) models demonstrated high predictive accuracy with an Area Under the Curve (AUC) > 0.97 in the training data.
• External validation confirmed the model's discriminatory ability, with AUCs of 0.88 for DT, 0.88 for RF, and 0.90 for GB.

Conclusions:

• A robust prediction model with high discriminatory power can accurately identify patients at risk for UTI within 8 weeks of pelvic surgery.
• Machine learning models, particularly GB, DT, and RF, show significant promise for predicting post-operative UTIs.
• Further prospective validation and randomized trials are recommended to assess the clinical utility of this model in preventing post-operative UTIs.