[Analysis of perioperative efficacy and safety of cytoreductive surgery in the treatment of colorectal cancer peritoneal metastases].

Objective: To analyzed perioperative safety of cytoreductive surgery (CRS) for patients with colorectal cancer peritoneal metastasis (CRPM) and to construct a predictive model for serious advese events (SAE). Methods: A descriptive case-series study was conducted to retrospectively collect the clinicopathological data and treatment status (operation time, number of organ resection, number of peritoneal resection, and blood loss, etc.) of 100 patients with peritoneal metastases from colorectal cancer or appendix mucinous adenocarcinoma who underwent CRS at the Sixth Affiliated Hospital of Sun Yat-sen University from January 2019 to August 2021. There were 53 males and 47 females. The median age was 52.0 (39.0-61.8) years old. Fifty-two patients had synchronous peritoneal metastasis and 48 had metachronous peritoneal metastasis. Fifty-two patients received preoperative neoadjuvant therapy. Primary tumor was located in the left colon, the right colon and the rectum in 43, 28 and 14 cases, respectively. Fifteen patients had appendix mucinous adenocarcinoma. Measures of skewed distribution are expressed as M (range). Perioperative safety was analyzed, perioperative grade III or higher was defined as SAE. Risk factors associated with the occurrence of SAEs were analyzed using multivariate logistic regression. A nomogram was plotted by R software to predict SAE, the efficacy of which was evaluated using the area under the ROC curve (AUC) and correction curves. Results: The median peritoneal cancer index (PCI) score was 16 (1-39). Sixty-eight (68.0%) patients achieved complete tumor reduction (tumor reduction score: 0-1). Sixty-two patients were treated with intraperitoneal hyperthermic perfusion chemotherapy (HIPEC). Twenty-one (21.0%) patients developed 37 SAEs of grade III-IV, including 2 cases of ureteral injury, 6 cases of perioperative massive hemorrhage or anemia, 7 cases of digestive system, 15 cases of respiratory system, 4 cases of cardiovascular system, 1 case of skin incision dehiscence, and 2 cases of abdominal infection. No grade V SAE was found. Multivariate logistic regression analysis showed that CEA (OR: 8.980, 95%CI: 1.428-56.457, P=0.019), PCI score (OR: 7.924, 95%CI: 1.486-42.259, P=0.015), intraoperative albumin infusion (OR: 48.959, 95%CI: 2.115-1133.289, P=0.015) and total volume of infusion (OR: 24.729, 95%CI: 3.956-154.562, P=0.001) were independent risk factors for perioperative SAE in CRS (all P<0.05). Based on the result of multivariate regression models, a predictive nomogram was constructed. Internal verification showed that the AUC of the nomogram was 0.926 (95%CI: 0.872-0.980), indicating good prediction accuracy and consistency. Conclusions: CRS is a safe and effective method to treat CRPM. Strict screening of patients and perioperative fluid management are important guarantees for reducing the morbidity of SAE.