The Hidden Burden of Liver-Related Mortality in Patients with Cirrhosis and Low MELD score
A poster presented at the 2018 meeting of the American Association for the Study of Liver Diseases.The MELD score has provided the basis for organ allocation in liver transplantation for nearly two decades and is reliable in predicting 90 day mortality in patients with higher scores. However, it is unclear which patients with low MELD-Na (<15) have a higher than expected liver-related mortality and thus would benefit from liver transplantation. We sought to quantify the burden of liver-related mortality in the overall low MELD population as most studies have utilized national transplant databases which only include patients referred for transplantation,. Methods: We utilized HealthLNK, a Chicago-wide database, which incorporates the de-identified electronic medical records from six health care institutions: five large academic medical centers (Northwestern Medicine, University of Chicago Hospitals and Clinic, Rush University Medical Center, University of Illinois at Chicago Medical Center, and Loyola University Medical Center) and one large academic safety net health care system (Cook County Health and Hospitals System). We included adult patients with icd-9 codes of cirrhosis whose MELD-Na scores were never higher than 15 and who died during the study period between January 1, 2006, and December 31, 2012. Records were linked with the state wide death registry and causes of death based on death certificate were attributed to liver related, non-liver related, and non-descript adjudicated by a transplant surgeon and two hepatologists. Results: A total of 20,122 patients with cirrhosis were captured during the above time period. Of these patients, 7,834 had a usable MELD-Na score with 3,717 under 15 for the entire time period. Of this low-MELD-Na subgroup there were 494 deaths (13.3% mortality). Causes of death in this group were 163 (32.9%) Liver related, 144 (29.0%) Non-Liver related, 85 (17.1%) Non-Descript and 104 (21%) missing. Within the Liver related group, 42 died of infection, 17 died of hepatocellular carcinoma, 15 died of portal hypertensive complications, 10 patients died of bleeding and the remaining 79 patients did not have enough detail to attribute to a sub-category. Conclusion: Despite persistently low MELD during a six year time period, 1/3 of patients deaths were attributable to liver disease. Future research should aim at predicting those with a high risk of liver-related mortality in the low MELD population, such that they might benefit from referral and early liver transplantation to reduce this risk.