Validation of the McCluskey Index for Predicting Higher Blood Transfusion in Living Donor Liver Transplantation

Background Liver transplantation (LT) surgery is often associated with massive blood transfusion (MBT) due to the complex nature of the procedure and altered coagulation status. Predicting MBT is crucial for effective resource planning and patient management. Existing literature suggests that the McCluskey index has good predictive ability for MBT, although it has primarily been validated in cadaveric transplant programs. This study aimed to validate the index at our center, which mainly performs living donor-related LT (LDLT) using adult-related donors, and identify predictors of blood transfusion in our cohort. Methods We retrospectively analyzed data from 533 patients who underwent either cadaveric or LDLT between January 2019 and September 2022. Packed red blood cells (PRBC) were transfused to maintain a target hemoglobin level of 8-9 g/dL. In our study, transfusion of ≥6 PRBC (the 75th percentile) was defined as higher blood transfusion (HBT). A receiver operating characteristic curve was used to assess the McCluskey index's ability to predict HBT. Results Using the threshold of ≥6 PRBC, 32.6% of the study population required HBT. The McCluskey index, which showed an area under the curve (AUC) of 0.82 in the internally validated cohort, yielded an AUC of 0.66 in our cohort. Independent risk factors for HBT identified in our analysis included preoperative hemoglobin level (OR = 0.69, 95% CI: 0.61-0.78), preoperative Model for End-Stage Liver Disease score (OR = 1.71, 95% CI: 1.14-2.58), a history of ICU admission within 60 days prior to transplantation (OR = 1.63, 95% CI: 1.05-2.52), and deceased donor LT (OR = 3.52, 95% CI: 1.1-11.7). A scoring system incorporating these four independent risk factors predicted HBT with an AUC of 0.73, significantly higher than the McCluskey index (p = 0.02). Conclusions Most components of the McCluskey index did not appear to be independent risk factors for increased blood transfusion in our cohort, which was predominantly composed of LDLT cases. Nevertheless, as a composite score, the McCluskey index showed some predictive efficacy, though it was inferior to the prediction model based on the independent risk factors identified in our cohort.