EP900 - Defining High-Risk Patients for Complications Following Simultaneous Colorectal and Liver Resection

Purpose/Background: The optimal approach to patients with synchronous colorectal cancer and liver metastases remains debated. Simultaneous resection offers the advantage of a definitive resection but raises concern for increased perioperative morbidity. Prior national studies have described overall complication rates but not individualized risk estimates. We developed predictive models to quantify patient-specific risks of major complications with simultaneous versus isolated resections and identify those who may benefit from a staged strategy.

Methods/Interventions: The ACS-NSQIP procedure-targeted colectomy and hepatectomy files (2014-2023) were queried for adults undergoing (1) simultaneous colorectal and hepatic resection (SR), (2) isolated colectomy (ICR), or (3) isolated hepatectomy (IHR). The primary outcome was 30-day serious complication, defined as a proxy to Clavien-Dindo grade ≥ III adverse events (septic shock, pneumonia, unplanned intubation, pulmonary embolism, acute renal failure, cardiopulmonary arrest, myocardial infarction, failure to wean off mechanical ventilation, stroke, anastomotic or bile leak, post-hepatectomy liver failure, reoperation, or death within 30 days). Two multivariable logistic models were trained/tested (80:20 split): one for SR and one for isolated resections. Each patient in the dataset was then assigned two predicted risks of serious complication: one under the simultaneous model and one under the isolated model. Patients whose predicted risk of simultaneous surgery exceeded the risk of isolated surgery by >10% were classified as "high-risk for complications after simultaneous resection." Demographics, comorbidities, laboratory variables, and operative factors were compared between high-risk and non-high-risk groups.

Results/Outcomes: A total of 50,806 patients were included: 35,221 (69.3%) underwent ICR, 13,323 (26.2%) IHR, and 2,262 (4.5%) SR. SR patients were younger (median age 59 vs 69 for ICR, 60 for IHR) with fewer comorbidities. Major complications occurred more frequently among patients who underwent SR (15.5%) compared to ICR (11.8%) and IHR (11.8%) (p < 0.001). SR was independently associated with higher morbidity compared to ICR (Odds ratio (OR): 1.32, 95% CI 1.18-1.50), and IHR (OR: 1.35, 95% CI 1.16-1.57). Model performance was fair (SR model C-statistic=0.70; isolated resection model C-statistic=0.68). Eight clinical features were present more frequently among patients in the "high-risk" phenotype: Age >75, ASA >3, diabetes, smoking history, COPD, CHF, functional dependence, and albumin < 2.5 (Figure 1). Interestingly, high-risk hepatectomy (lobectomy or trisectionectomy) was not more common in the high-risk phenotype.

Conclusion/Discussion: Simultaneous colorectal and hepatic resection carries higher morbidity than isolated resection. Using national data, we developed a patient-level model to identify those at excessive risk for complications from simultaneous surgery. Clinical factors-particularly advanced age, hypoalbuminemia, smoking, and functional dependence-were stronger predictors of adverse outcomes than operative complexity. This model may guide individualized selection for staged versus simultaneous approaches to reduce postoperative morbidity.

Radar plot showing the relative prevalence (scaled 0-100%) of eight clinical factors associated with increased risk of major complications.