An integrated random forest– and least absolute shrinkage and selection operator (LASSO)-derived nomogram for predicting postoperative nosocomial infections in colorectal cancer patients
Authors' conclusion
Our nomogram enables precise stratification of colorectal cancer patients by their postoperative infection risk, highlighting perioperative factors-such as operative duration, surgical approach, and ASA grade-that warrant targeted management.
Study context
- Population
- Patients who underwent colorectal cancer surgery between 2020 and 2024
- Participants
- 1,760 patients
- Exposure
- Training vs Validation
- Trainingn = 1,146
- Validationn = 614
- Outcome
- Postoperative nosocomial infection (any hospital-acquired infection, including lower respiratory tract, surgical-site, multiple-site and other-site infections); nomogram performance (AUC, calibration, DCA)
- Follow-up
- Within 30 days after surgery
Methodological detailsprediction model development and temporal validation, random forest and LASSO feature selection, multivariable logistic regression nomogram
- Prediction model development and temporal validation
- Random forest and LASSO feature selection
- Multivariable logistic regression nomogram
Main finding
As reported in the abstract
Nomogram discrimination (AUC)
| Group | AUC |
|---|---|
| Training | 0.860 |
| Validation | 0.843 |
AUC
Training0.860
Validation0.843
Observational data: associations, not effects.
Further findings
Incidence of postoperative hospital-acquired infection
- Postoperative hospital-acquired infection, all patients (166/1,760)
- 9.43%
- Lower respiratory tract infection most common subtype, followed by surgical-site infection
Independent predictors (multivariable logistic regression)
- LASSO retained eight predictors; age, coronary heart disease, perioperative blood transfusion, colostomy, surgical approach, ASA grade, persistent fever for 3 consecutive days and postoperative complications independently associated with infection
Original publication
An integrated random forest– and LASSO-derived nomogram for predicting postoperative nosocomial infections in colorectal cancer patients. Perioper Med (Lond). Published 26 February 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →