Neutrophil-to-lymphocyte-ratio-based perioperative prognosis prediction model on early mortality after cytoreductive surgery with hyperthermic intraperitoneal chemotherapy

• DOI: 10.1016/j.asjsur.2022.01.104
• Publication Date: 2022-03-05T12:33:57Z
• Authors: Hao-Chien Hung; Po-Jung Hsu; Ting-Chang Chang; Hung-Hsueh Chou; Kuan-Gen Huang; Chyong-Huey Lai; Chao-Wei Lee; Ming-Chin Yu; Jeng-Fu You; Jun-Te Hsu; Ting-Jung Wu
• Keywords: Neutrophils; Cytoreduction Surgical Procedures; Hyperthermic Intraperitoneal Chemotherapy; Hyperthermia, Induced; Prognosis; Combined Modality Therapy; 3. Good health; Survival Rate; 03 medical and health sciences; 0302 clinical medicine; Humans; Lymphocytes; Peritoneal Neoplasms; Retrospective Studies

Combined cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (CRS-HIPEC) to treat peritoneal surface malignancy (PSM) has gained a positive result compared with palliative chemotherapy alone in several cancer types. However, its postoperative care could be challenging. We aimed to develop a predicting model on early mortality.We retrospectively reviewed 132 PSM patients who had received CRS-HIPEC. The optimal cut-off value of the neutrophil-to-lymphocyte ratio (NLR) was determined as 4.4 by using the receiver operating characteristic curve analysis with an area under the curve (AUC) of 0.75. The impact of NLR on survival was elucidated by comparing the pre-operative low (NLR≤ 4.4, n = 101) and high (NLR> 4.4, n = 31) groups using the Kaplan-Meier method. The significant variables selected in multivariate analysis on early mortality were used in prediction model development.Multivariate analysis showed that incomplete CRS, major postoperative complications, higher pre-operative NLR, and dynamic NLR changes were significant predictors of early mortality. Our perioperative prediction of prognosis (triple P) model contained four independent risks, and the AUC after classification was 0.860 (95% confidence interval [CI]: 0.773-0.947). External validation confirmed positive discrimination ability (AUC: 0.808, 95% CI: 0.666-0.950).In conclusion, our triple P model provides great determination in outcomes prediction and it is easily obtained, reliable, and applicable in routine practice.