Novel Donor-Recipient Allocation Model Predicts Early Graft Loss in Liver Transplant Recipients

Source: DGNews  |  Presented at AASLD

By Kari Oakes

WASHINGTON, DC -- November 8, 2013 -- Researchers have developed a Donor-Recipient Allocation Model (DReAM) that accurately predicts early graft loss based on liver donor-recipient matching.

Findings were presented here on November 3 at the 64th Annual Meeting of the American Association for the Study of Liver Diseases (AASLD).

The first 3 months after liver transplantation are risky. Almost one-third of all graft failures happen during that time, and accurate predictors of early graft loss are lacking, according to Mario Angelico, MD, University and Policlinico Tor Vergata, Rome, Italy, and colleagues.

To develop the model, researchers scrutinised data from the Italian Liver Match database, which included 1,480 adult patients undergoing first liver transplant from deceased heart-beating donors from June 2007 to May 2009, and data from OPN/UNOS, which comprised 9,740 adults undergoing first liver transplant during the same time period.

Data was analysed for factors leading to early graft loss or patient death within 90 days following transplantation. Final analysis included data from 1,430 patients and 7, 781 patients in both groups, respectively.

Of the patients in the Liver Match cohort, 173 (11.7%) experienced early graft loss or death within 3 months, as did 706 patients (7.2%) in the OPN/UNOS cohort.

Significant predictors of early graft loss included donor age, donation after cardiac death, cold ischaemia time, donor body mass index (BMI), donor height, recipient creatinine, bilirubin, disease aetiology, prior upper abdominal surgery, and portal thrombosis.

The data was used to generate an early graft loss-donor risk index (EGL-DRI) and an early graft loss-recipient risk index (EGL-RRI). When combined, these 2 indices produced the DReAM model, which was then validated in the OPTN/UNOS series.

The DReAM model showed a c-statistics (0.66) higher than the Donor-Model for End-Stage Liver Disease (D-MELD) score (0.58; P = 0.0003), higher than the Survival Outcomes Following Liver Transplantation (SOFT) score (0.59; P = .0718), and higher than the Balance of Risk (BAR) score (0.57; P = .0067).

Updated DReAM modelling allowed predictive accuracy to reach a c-statistic as high as 0.76.

DReAM allows for more accurate prediction of early graft loss in liver transplant recipients, promoting better resource allocation and patient outcomes, the authors concluded.

[Presentation title: Prediction of Early Graft Loss After Liver Transplantation Based on Donor Recipient Matching. The Liver Match Study. Abstract 1133]

Source