Martha Kerr

August 18, 2010

August 18, 2010 — A new study to validate the accuracy of the Donor Risk Index (DRI) in predicting the outcome of a liver transplant shows that the algorithm is, in fact, not all that accurate and has limited value in clinical practice.

Dr. Marwan Abouljoud

The DRI was derived from large clinical trials; it put together characteristics of liver donors that could be used to predict the outcome of a liver transplant. According to lead author Marwan Abouljoud, MD, director of the Henry Ford Hospital Transplant Institute in Detroit, Michigan, the DRI does not appear to be accurate, for the purposes of the transplant surgeon, in assessing outcomes of an individual donor. Dr. Abouljoud presented these findings this week at the XLVII European Renal Association-European Dialysis and Transplant Association (ERA-EDTA) Congress in Vancouver, British Columbia.

The goals of the study were to confirm the effect of the DRI on liver transplant recipients and to evaluate further the effect donor and recipient characteristics on transplant outcomes.

The study looked at the DRI scores for 100 liver transplants performed during 2008. The investigators examined donor and recipient demographics, Model for End-Stage Liver Disease (MELD) scores, laboratory data, and operative factors. Pathologists involved in the cases were also consulted.

Mean recipient age was 54.5 ± 10.3 years, mean donor age was 42.5 ± 15.0 years, 39% of both groups were female, and 28% of transplant recipients had a primary diagnosis of hepatitis C virus infection. Average MELD score was 20 ± 8 (range, 9 to 40). The average DRI score was 1.51 ± 0.31 (range, 1.00 to 2.60).

The presence of hepatitis C virus RNA in recipients posttransplantation was significantly associated with graft failure and mortality (hazard ratio [HR], 0.223; P < .001). Elevated bilirubin levels in the recipient at the time of discharge were correlated with graft failure (HR, 1.185; P = .027).

The only DRI subcategory that was associated with graft loss in the study was a donor age older than 70 years (HR, 9.57; P = .006). DRI score had a near-significant correlation with graft failure (HR, 3.1; P = .088).

"We found the risks to be small for individual characteristics influencing outcomes, except for hepatitis C," said Dr. Abouljoud. Hepatitis in recipients posttransplantation had the greatest effect on mortality and graft failure. The mortality rate overall was 22%, and was 41% for those with recurrent hepatitis. The graft failure rate was 20% overall and was 44% for those with hepatitis.

"Looking at data from large studies is like looking under a microscope, and when the information from many of these national studies is translated to day-to-day practice, the information often doesn't prove to be practical, especially when you that consider patients are dying on transplant waiting lists," said Dr. Abouljoud.

"Conclusions are based on nationally collated and statistically processed data, and certain donor features are associated with certain risks of failure. Our study, as expected, shows that, at the program level, it is difficult to translate this into action items, such as when to avoid a particular donor for a particular recipient — especially one with a high risk of mortality while waiting," Dr. Abouljoud told Medscape Medical News.

"Our expected outcomes do not yet take donor factors into consideration that will ultimately account for added risks," he said. "All in all, this illustrates the complexity inherent in transplantation, including the decisions on both donor and recipient selection."

"Our findings suggest that the focus should be on the combined factors that affect organ quality and function, not one individual characteristic," Dr. Abouljoud concluded.

The study was funded by the Henry Ford Transplant Institute. Dr. Abouljoud has disclosed no relevant financial relationships.

XLVII European Renal Association-European Dialysis and Transplant Association (ERA-EDTA) Congress: Abstract 6. August 15, 2010.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....