The predictive value of the MELD-XI score for short- and long-term mortality in elderly patients with non-ST elevation myocardial infarction.

Abstract

In this study, we investigated the utility of the Model for End-stage Liver Disease excluding INR (MELD-XI) score in predicting short- and long-term mortality in elderly patients with non-ST elevation myocardial infarction (NSTEMI) who underwent coronary angiography (CAG). In total, we analyzed 228 elderly NSTEMI patients above the age of 75. We used the modified 5-item frailty index and the Charlson Comorbidity Index (CCI) to assess the comorbidities. The MELD-XI score was calculated using the logarithmic relationship between the serum creatinine and total bilirubin. The median long-term follow-up was 530 [interquartile range (IQR) = 303-817] days and the short- and long-term mortality rates were 11.8% (n = 27) and 16.4% (n = 33), respectively. Patients who did not survive had a substantially higher MELD-XI score than those who did [10.1 (IQR = 7.8-15.1) vs. 4.5 (IQR = 1.9-6.9), p < 0.001, respectively]. Multivariable Cox regression analyses indicated that the MELD-XI score predicted both short- and long-term mortality independently. When the MELD-XI score, serum creatinine, and total bilirubin area under the curve (AUC) values were compared to predict long-term mortality, the MELD-XI score had the highest value (AUC: 0.833), followed by the serum creatinine (AUC: 0.741), and the total bilirubin (AUC: 0.723). The accuracy of the MELD-XI score was further tested with the GRACE risk score, which demonstrated noninferiority. This was the first investigation which indicated that elderly NSTEMI patients with a high MELD-XI score had poor prognosis in the short- and long-term period.