Intraoperative packed red blood cell transfusion (iPRBT) and PCI-normalised iPRBT rates (iPRBT/PCI ratio) negatively affect short- and long-term outcomes of patients undergoing cytoreductive surgery and intraperitoneal chemotherapy - An analysis of 880 patients.

ABSTRACT

BACKGROUND: Most studies on the effects of intraoperative packed red blood cell transfusions (iPRBT) on patients undergoing cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC) have shown deleterious outcomes. It is unclear if this is a result of the transfusion itself or because iPRBTs serve as a surrogate of more advanced disease. METHODS: A retrospective analysis of 880 patients treated from 1996 to 2017. The effect of any exposure to iPRBT as well as the effect of peritoneal cancer index (PCI)-normalised iPRBT rates (ratio of iPRBT/PCI) on patients short- and long-term outcomes (recurrence-free (RFS) and overall survival (OS)) were assessed. Equally, the prognostic effect of postoperative PRBTs was analysed and adjusted for. RESULTS: Of the 880 patients included, only 26.4% had no iPRBT whereas 59.2% of patients had no postoperative PRBT. Patients with no iPRBTs had significantly lower PCIs, less high-grade complications, shorter ICU and hospital length of stay, as well as improved RFS and OS. Furthermore, high PCI-normalised iPRBTs resulted in worse perioperative and long-term outcomes, with a median OS of 41 months vs. 103 months (5-year survival rate 36.6% vs. 66.1%; p < 0.001) and median RFS of 13 months vs. 30 months (5-year RFS rate 18.3% vs. 37.6% p < 0.001) compared to those with a low iPRBT/PCI ratio. This independent effect was confirmed upon multivariable Cox regression analysis which corrected for important confounders including complexity of procedures and postoperative PRBTs (adjusted HR [aHR]2.04, 95%CI 1.36-3.04, p = 0.001 for OS; aHR 1.38, 95%CI 1.06-1.81, p = 0.017 for RFS). However, subgroup analysis (stratified by histopathologic disease entities) revealed that this independent prognostic effect was seen in high-grade mucinous appendiceal neoplasms, whereas PCI-normalised IPRBTs were not significantly prognostic in other histopathologic subgroups. CONCLUSION: iPRBTs significantly and independently impact perioperative and long-term outcomes of patients undergoing CRS/HIPEC. However, this effect mainly seems to occur in patients with high-grade mucinous neoplasms, whereas it may only be of borderline prognostic significance in other patient groups. The development of blood-sparing protocols may help improve outcomes of patients undergoing this complex oncologic procedure.