IMPORTANCEQuality improvement programs for colorectal cancer surgery have been introduced with benchmarking based on quality indicators, such as mortality. Detailed (pre)operative characteristics may offer relevant information for proper case-mix correction. OBJECTIVE To investigate the added value of machine learning to predict quality indicators for colorectal cancer surgery and identify previously unrecognized predictors of 30-day mortality based on a large, nationwide colorectal cancer registry that collected extensive data on comorbidities. DESIGN, SETTING, AND PARTICIPANTS All patients who underwent resection for primary colorectal cancer registered in the Dutch ColoRectal Audit between January 1, 2011, and December 31, 2016, were included. Multiple machine learning models (multivariable logistic regression, elastic net regression, support vector machine, random forest, and gradient boosting) were made to predict quality indicators. Model performance was compared with conventionally used scores. Risk factors were identified by logistic regression analyses and Shapley additive explanations (ie, SHAP values). Statistical analysis was performed between March 1 and September 30, 2020. MAIN OUTCOMES AND MEASURES The primary outcome of this cohort study was 30-day mortality. Prediction models were trained on a training set by performing 5-fold cross-validation, and outcomes were measured by the area under the receiver operating characteristic curve on the test set. Machine learning was further used to identify risk factors, measured by odds ratios and SHAP values. RESULTSThis cohort study included 62 501 records, most patients were male (35 116 [56.2%]), were aged 61 to 80 years (41 560 [66.5%]), and had an American Society of Anesthesiology score of II (35 679 [57.1%]). A 30-day mortality rate of 2.7% (n = 1693) was found. The area under the curve of the best machine learning model for 30-day mortality (0.82; 95% CI, 0.79-0.85) was significantly higher than the American Society of Anesthesiology score (0.74; 95% CI, 0.71-0.77; P < .001), Charlson Comorbidity Index (0.66; 95% CI, 0.63-0.70; P < .001), and preoperative score to predict postoperative mortality (0.73; 95% CI, 0.70-0.77; P < .001). Hypertension, myocardial infarction, chronic obstructive pulmonary disease, and asthma were comorbidities with a high risk for increased mortality. Machine learning identified specific risk factors for a complicated course, intensive care unit admission, prolonged hospital stay, and readmission. Laparoscopic surgery was associated with a decreased risk for all adverse outcomes. CONCLUSIONS AND RELEVANCEThis study found that machine learning methods outperformed conventional scores to predict 30-day mortality after colorectal cancer surgery, identified specific (continued)

No abstract

Purpose Scarce data are available on differences among index colectomies for colon cancer regarding reoperation for anastomotic leakage (AL) and clinical consequences. Therefore, this nationwide observational study aimed to evaluate reoperations for AL after colon cancer surgery and short-term postoperative outcomes for the different index colectomies. Methods Patients who underwent resection with anastomosis for a first primary colon carcinoma between 2013 and 2019 and were registered in the Dutch ColoRectal Audit were included. Primary outcomes were mortality, ICU admission, and stoma creation. Results Among 39,565 patients, the overall AL rate was 4.8% and ranged between 4.0% (right hemicolectomy) and 15.4% (subtotal colectomy). AL was predominantly managed with reoperation, ranging from 81.2% after transversectomy to 92.4% after sigmoid resection (p < 0.001). Median time to reoperation differed significantly between index colectomies (range 4–8 days, p < 0.001), with longer and comparable intervals for non-surgical reinterventions (range 13–18 days, p = 0.747). After reoperation, the highest mortality rates were observed for index transversectomy (15.4%) and right hemicolectomy (14.4%) and lowest for index sigmoid resection (5.6%) and subtotal colectomy (5.9%) (p < 0.001). Reoperation with stoma construction was associated with a higher mortality risk than without stoma construction after index right hemicolectomy (17.7% vs. 8.5%, p = 0.001). ICU admission rate was 62.6% overall (range 56.7–69.2%), and stoma construction rate ranged between 65.5% (right hemicolectomy) and 93.0% (sigmoid resection). Conclusion Significant differences in AL rate, reoperation rate, time to reoperation, postoperative mortality after reoperation, and stoma construction for AL were found among the different index colectomies for colon cancer, with relevance for patient counseling and perioperative management.