Morbidity and Mortality Rates Following Cytoreductive Surgery Combined With Hyperthermic Intraperitoneal Chemotherapy Compared With Other High-Risk Surgical Oncology Procedures

IMPORTANCE Currently, rates of referral of patients with peritoneal metastasis in the United States who qualify for cytoreductive surgery combined with hyperthermic intraperitoneal chemotherapy (CRS/HIPEC) are low, in part because of the misperception of high morbidity and mortality rates. However, patients requiring major gastrointestinal surgical procedures with similar complication rates are routinely referred. OBJECTIVE To evaluate the relative safety of CRS/HIPEC. DESIGN, SETTING, AND PARTICIPANTS Retrospective cohort study of 34 114 patients who underwent CRS/HIPEC, right lobe hepatectomy, trisegmental hepatectomy, pancreaticoduodenectomy, and esophagectomy between January 1, 2005, and December 31, 2015, included in the American College of Surgeons National Surgical Quality Improvement Project (NSQIP) database. Data analysis was performed in 2018. MAIN OUTCOMES AND MEASURES Data from the NSQIP database were used to compare perioperative and 30-day postoperative morbidity and mortality rates of CRS/HIPEC (1822 patients) with other, well-accepted, high-risk surgical oncology procedures: right lobe hepatectomy (5109 patients), trisegmental hepatectomy (2449 patients), pancreaticoduodenectomy (Whipple) (16 793 patients), and esophagectomy (7941 patients). RESULTS For 34 114 patients, median (interquartile range [IQR]) age was 63 (55-71) years and 42% were female. Patients undergoing CRS/HIPEC tended to be younger, with a median age of 57 years, and esophagectomy had the highest median (IQR) American Society of Anesthesiologists classification (3 [3-3]). When compared with CRS/HIPEC, higher complication rates were reported in the following categories: (1) superficial incisional infection in Whipple and esophagectomy (5.4% [95% CI, 4.4%-6.4%] vs 9.7%[95% CI, 9.3%-10.1%] and 7.2%[95% CI, 6.6%-7.8%], respectively; P<.001); (2) deep incisional infection in Whipple (1.7%[95% CI, 1.1%-2.3%] vs 2.7%[95% CI, 2.5%2.9%]; P<.01); (3) organ space infection in right lobe hepatectomy (7.2%[95% CI, 6.0%-8.4%] vs 9.0% [95% CI, 8.2%-9.8%]; P =.02), trisegmental hepatectomy (12.4%[95% CI, 11.1%-13.7%]; P<.001), and Whipple (12.9% [95% CI, 12.4%-13.4%]; P<.001); and (4) return to the operating room for esophagectomy (6.8%[95% CI, 5.6%-8.0%] vs 14.4%[95% CI, 13.6%-15.2%]; P<.001). Median (IQR) length of hospital stay was lower in CRS/HIPEC (8 [5-11] days) than Whipple (10 [7-15] days) and esophagectomy (10 [8-16] days) (P<.001). Overall 30-day mortality was lower in CRS/HIPEC (1.1%; 95% CI, 0.6%-1.6%) compared with Whipple (2.5%; 95% CI, 2.3%-2.7%), right lobe hepatectomy (2.9%; 95% CI, 2.4%-3.4%), esophagectomy (3.0%; 95% CI, 2.6%-3.4%), and trisegmental hepatectomy (3.9%; 95% CI, 3.1%-4.7%) (P<.001). CONCLUSIONS AND RELEVANCE Comparative analysis revealed CRS/HIPEC to be safe, often safer across the spectrum of NSQIP safety metrics when compared with similar-risk oncologic procedures. Patient selection was important in achieving observed outcomes. High complication rates are a misperception from early CRS/HIPEC experience and should no longer deter referral of patients to experienced centers or impede clinical trial development in the United States.