Preoperative Inflammatory Markers Predict Postoperative Complications After Cytoreductive Surgery and Hyperthermic Intraperitoneal Chemotherapy for Peritoneal Colorectal Carcinomatosis

A retrospective study of a single-center prospective database of patients who underwent curative CRS-HIPEC for peritoneal carcinomatosis (PC) of colo-rectal cancer (CRC), between September 2012 and July 2023 at Reims Hospital-University was conducted. Inclusion criteria were: patients older than 18 years with a performance status ≤ 2, who underwent complete (no macroscopic residual disease) CRS-HIPEC was performed. Exclusion criteria were the presence of an infectious disease prior to surgery, CRS without HIPEC, incomplete CRS and expressed refusal to participate in the study.

During the month prior to surgery, patients underwent a thorough diagnostic workup, including chest and abdominopelvic computed tomography (CT) scans, complete haematological and biochemical tests, and cytobacteriological analysis of urine. Blood tests included leukocytes, neutrophils, lymphocytes, platelet count, albumin and CRP. Within one week before surgery, patients were examined by both the surgeon and the anaesthetist.

Variables with a normal distribution are presented as mean-SD, while those with a non-normal distribution are presented as median and range. Categorical variables are presented as counts and percentages. Chi-Square or Fisher's exact test was used to compare qualitative variables. Linear regression analysis was used to assess any association between preoperative inflammatory biomarkers and patient comorbidities and disease characteristics. Preoperative levels of inflammatory biomarkers were compared between patients with no postoperative complication (POC) and patients with POC and major POC using the non-parametric Mann-Whitney U test. Optimal cut-off values for significant inflammatory biomarkers were determined using the Youden index. Receiver operating curves (ROC) were calculated to determine the area under the curve (AUC). Univariable and multivariable analyses including cut-off values of significant biomarkers were performed in a binary logistic regression analysis and expressed as odds ratios (OR). All tests were two-tailed and results were considered significant if the p-value was < 0.05. Statistical analyses were performed using R software (version 4.0.5).