Document Type


Degree Name



Dept. of Public Health & Preventive Medicine


Oregon Health & Science University


Background: Patients are prioritized for liver transplantation (LT) by their anticipated 90-day waitlist mortality using the MELD score, however the MELD underestimates waitlist mortality when hyponatremia is present. A revised MELD that incorporates the added mortality due to hyponatremia, the MELD-Na, was shown to reduce waitlist mortality in hyponatremic patients in a modeling study. In UNOS Region 6, regional agreement has resulted in prioritization of cirrhotic patients with hyponatremia for LT using a MELD-Na exception since 2008. Research question: Does the use of a MELD -Na exception decrease 90-day mortality for hyponatremic patients on the liver transplant waitlist in Region 6. Specific Aims: (1) Identify and describe characteristics of the following groups from the UNOS transplant database: a. ‘MELD-Na’ group: Region 6 patients who received a MELD-Na exception b. ‘Normonatremic’ group: Region 6 normonatremic patients c. ‘Hyponatremic without MELD-Na’ group: Region 6 hyponatremic patients who did not receive a MELD-Na exception (2) Evaluate whether patients receiving MELD-Na exceptions in Region 6 have similar 90 day waitlist mortality as normonatremic patients in Region 6. (3) Compare waitlist mortality in patients with MELD-Na exceptions to hyponatremic patients who did not receive a MELD -Na exception. (4) Evaluate the influence of other predictors on waitlist mortality, using competing risk regression. Methods: In the UNOS registry, we selected all patients listed for liver transplant in Region 6 from January 2010 to June 2014 who received a MELD-Na prioritization exception based on regional agreement. We compared waitlist mortality of the groups using competing risk regression. Results: The sub hazard ratio for death was 52% lower in the MELD-Na group relative to the hyponatremic without exceptions group when taking into account the competing risk of LT and adjusted for MELD score, age, and encephalopathy (SHR=0.48, 95% CI 0.27 -0.87, p=0.016). The sub hazard ratio of death was 43% lower in the MELD-Na group compared to the normonatremic group, taking into account the competing risk of LT, and adjusted for MELD score, year on the waitlist, age, and etiology (SHR=0.57, 95% CI 0.30 - 1.07, p=0.081). Age was an influential predictor in normonatremic and hyponatremic patients without an exception.

Conclusion: There is a strong association of decreased mortality in hyponatremic patients who received a MELD -Na exception compared to hyponatremic patients without the exception. Therefore, the MELD-Na score had the intended effect of decreasing waitlist mortality in hyponatremic patients. I did not find a significant difference in mortality between MELD-Na exception patients and normonatremic patients, but the evidence for this lack of difference is not strong, and there may be a difference that this study could not detect. This raises the question of whether the MELD-Na exception conferred a waitlist survival benefit to hyponatremic patients. Before being adopted as the new determinant for LT priority there should be further research examining a possible survival benefit conferred by the MELD-Na to hyponatremic patients.




School of Medicine

Available for download on Monday, June 04, 2018