Combined effects of recipient age and model for end-stage liver disease score on liver transplantation outcomes.

BACKGROUND

The proportion of older patients awaiting liver transplantation (LT) is rising. Although increased age and LT-MELD are known to increase the risk of graft loss, no studies have explored whether there is a synergistic effect between LT-age and LT-MELD.

METHODS

All US adult, non-Status 1 recipients of primary deceased donor LT from 2/05 to 1/10 without MELD exceptions were included (n=15,677). Recipients were categorized by LT-age [18-59 yr (n=11,966), 60-64 yr (n=2181), 65-69 yr (n=1177), and ≥70 yr (n=343)] and LT-MELD [low (<20, n=5290), mid (20-27, n=5112), and high (≥28, n=5265)]. Adjusted Cox models evaluated the independent and combined effects of LT-age and LT-MELD on graft loss (death or re-LT).

RESULTS

LT-age ≥70 yr (HR=1.65, 95% CI 1.08-1.82) and LT-MELD ≥28 (HR=1.46, 95% CI 1.02-1.47) were independently associated with increased risk of graft loss (P<0.001). In a model allowing for the interaction between LT-age and LT-MELD, the risk of graft loss for recipients ≥70 years with MELD ≥28 was higher than predicted by the additive model (HR=2.38, 95% CI 1.73-3.27, P<0.001) resulting in 1-year graft survival of 56%. However, the increased risk of graft loss in recipients ≥70 years was attenuated at lower LT-MELD <28. Furthermore, the interaction term was not significant for any other LT-age and LT-MELD combination.

CONCLUSION

Our analyses suggest that recipients should not be excluded solely based on age; however, LT for recipients ≥70 years at high LT-MELD scores should be undertaken cautiously.