Back to 2025 Abstracts
RISK FACTORS FOR WITHDRAWAL OF CARE AFTER ELECTIVE SURGERY FOR GASTROINTESTINAL CANCERS: A NATIONAL ANALYSIS
Rebecca Tang
*, Sarah Rudasill, Christy Cauley, Grace Lee, Hiroko Kunitake, Rocco Ricciardi
Massachusetts General Hospital Department of Surgery, Boston, MA
Introduction: Risk factors for mortality after gastrointestinal (GI) oncologic surgery have been described and used to guide surgical decision-making and counseling. Although postoperative mortality captures failure to rescue from severe complications, examining patterns of withdrawal of care provides a more nuanced and patient-oriented understanding of complicated postoperative pathways. This study therefore seeks to identify risk factors for withdrawal of care after surgery for GI cancers.
Methods: Adults undergoing elective surgery for GI cancers (2021-2022) were identified from the American College of Surgeons National Surgical Quality Improvement Program. Withdrawal of care within 30 days of surgery was abstracted. Risk factors for postoperative withdrawal of care were then identified using univariate analysis and multivariable logistic regression.
Results: 46,376 patients were identified who underwent elective resection for a GI cancer. On univariate analysis, patients who withdrew care were more likely to be older (81% vs 57% over 65 years, p<0.001) and more comorbid (94% vs 80% with ASA class 3 or greater, p<0.001). After controlling for demographic and preoperative factors, significant risk factors for postoperative withdrawal of care included age 65 years or greater (AOR 1.87, p<0.001), poor functional status (AOR 2.77, p=0.001), and presence of metastatic disease (AOR 2.54, p=<0.001). In addition, postoperative respiratory failure (AOR 4.62, p<0.001), septic shock (AOR 6.84, p<0.001), and reoperation (AOR 1.66, p=0.030) were significant predictors of withdrawal of care. Case complexity and operative time were not predictive. Of the 721 patients who withdrew care postoperatively, 375 (52%) expired within 30 days, 187 (26%) were discharged home, and 102 (14%) were discharged to a facility.
Conclusion: Elderly patients with poor functional status are and increased risk of withdrawal of care leading to death following elective oncologic surgery. These patients may benefit from preoperative goals of care conversations and comprehensive interdisciplinary perioperative optimization. Besides age and frailty, the strongest predictors of withdrawal of care were postoperative complications such as respiratory failure, septic shock, and reoperation. This suggests that mortality after elective oncologic surgery is related to complex and nuanced decision making beyond the initial consent process. Although goals of care should be discussed preoperatively, these should be revisited when major postoperative complications occur. Findings from this study may be used to guide these longitudinal goals of care conversations and decision-making throughout the perioperative period.
Table 1. Patient Characteristics
Table 2. Risk Factors for Withdrawal of Care Following Elective Surgery for Gastrointestinal Cancers
Back to 2025 Abstracts
