(O P3) Hospital Variation in Perioperative Outcomes of Cytoreductive Surgery for Cancer: Implications for Quality Measurement

Background: Interest in cytoreductive surgery for advanced intraperitoneal malignancies continues to rise. Many studies have reported significant morbidity from these operations, yet also with variation or heterogeneity in practice patterns. Whether utilization of the usual hospital quality measures of perioperative outcomes are applicable in this patient population is unknown.

Methods: Cytoreductive operations performed for cancer between January 1, 2013 and June 30, 2018 were identified in the American College of Surgeons National Surgical Quality Improvement Program registry (via CPT codes such as 49203, 49204, 49205). Risk-adjusted hospital-level variation in 30-day death, serious morbidity, reoperation, readmission, and a composite of death or serious morbidity (DSM) were evaluated using hierarchical models. National Cancer Institute (NCI)-designated Cancer Center (NCI-CC) status was also explored.

Results: A total of 6,125 operations were included across 588 hospitals, 56 of which were NCI-CCs. Overall rates of death, serious morbidity, reoperation, readmission, and DSM were 1.5%, 12.8%, 3.5%, 8.6%, and 13.3%, respectively. Hierarchical covariance parameters for death, serious morbidity, reoperation, readmission, and DSM were 0.24 (p=0.28), 0.04 (p=0.19), 0.21 (p=0.05), 0.002 (p=0.47), and 0.04 (p=0.17), respectively. When compared to other hospitals, NCI-CCs had better risk-adjusted 30-day mortality (p<0.001), but not for serious morbidity (p=0.13), reoperation (p=0.21), readmission (p=0.60), or DSM (p=0.14).

Conclusion: In these data, hospital-level variation was unable to be detected using the usual measures of perioperative outcomes. Given the morbidity associated with cytoreductive surgery, this demonstrates a clear opportunity to better improve the way quality is measured for patients undergoing cytoreductive operations for cancer.