The following is a summary of “Predictive accuracy of the boosted regression model in estimating the risk of venous thromboembolism following minimally invasive radical surgery in pharmacological prophylaxis-naïve men with prostate cancer,” published in the February 2024 issue of Surgery by Leong et al.

Venous thromboembolism (VTE) stands as a concerning yet preventable complication following urological surgeries, presenting clinicians with the complex task of balancing thromboprophylaxis benefits against associated risks amidst global practice variations. Against this backdrop, the study endeavors to explore the feasibility of adopting a risk-stratified approach for thromboembolism prophylaxis following radical prostatectomy. Leveraging a prospective database encompassing 522 men who underwent minimally invasive prostatectomy between February 2010 and October 2021, researchers employed a deterministic data linkage method to cross-link VTE events through electronic systems. Utilizing Community Health Index (CHI) numbers, patients were identified via electronic health records, with demographic and clinical characteristics including age, comorbidities, Gleason staging, and readmission details meticulously accrued.

The primary outcome of interest was the occurrence of VTE within 90 days post-surgery, with an ancillary aim to develop a risk-stratified scoring system. Employing statistical analyses via R-Statistical Software, the study group estimated the risk of VTE within 90 days utilizing a gradient-boosting decision trees (BRT) model. Their findings revealed that 1.1% (6/522) of patients developed deep vein thrombosis or pulmonary embolism within the specified timeframe. Further statistical analysis unveiled significant differences in body mass index (p = 0.016), duration of hospital stay (p < 0.001), and number of readmissions (p = 0.036) between patients who developed VTE and those who did not. Notably, BRT analysis identified 8 variables of relative importance in predicting VTE, with the constructed receiver operating curves (ROC) demonstrating robust discrimination power, reflected by an AUC of 0.97 (95% CI: 0.945, 0.999).

While their study offers valuable insights into post-minimally invasive prostatectomy VTE incidence and predictive modeling, it is essential to acknowledge the limitations inherent to a single-center study. Nevertheless, their findings underscore the low VTE incidence in men not receiving prophylaxis with low molecular weight heparin and suggest the potential utility of the proposed risk-scoring system in identifying higher-risk patients for targeted thromboprophylaxis interventions, thus facilitating personalized patient care.

 Source: wjso.biomedcentral.com/articles/10.1186/s12957-023-03170-y