Risk prediction in infective endocarditis by modified MELD-XI score.

The suitability of the model for end-stage liver disease excluding international normalized ratio (MELD-XI) score to predict adverse outcomes in infective endocarditis (IE) patients remains uncertain. This study was performed to explore the prognostic value of the MELD-XI score and modified MELD-XI score for patients with IE. A total of 858 patients with IE were consecutively enrolled and classified into two groups: MELD-XI ≤ 10 (n = 588) and MELD-XI > 10 (n = 270). Multivariate analysis was performed to determine risk factors independent of MELD-XI score. Higher MELD-XI score was associated with higher in-hospital mortality (15.6 vs. 4.8%, p < 0.001) and major adverse clinical events (33.3 vs. 18.4%, p < 0.001). MELD-XI score was an independent predictor of in-hospital death (odds ratio [OR] = 1.06, 95% CI, 1.02-1.10, p = 0.005). Based on a multivariate analysis, NYHA class III or IV (3 points), C-reactive protein > 9.5 mg/L (4 points), and non-surgical treatment (6 points) were added to MELD-XI score. Modified MELD-XI score produced higher predictive power than previous (AUC 0.823 vs. 0.701, p < 0.001). The cumulative incidence of long-term mortality (median 29 months) was significantly higher in patients with modified MELD-XI score > 13 than those without (log-rank = 25.30, p < 0.001). Modified MELD-XI score was independently associated with long-term mortality (hazard ratio = 1.08, 95% CI, 1.04-1.12, p < 0.001). MELD-XI score could be used as a risk assessment tool in IE. Furthermore, modified MELD-XI score remained simple and more effective in predicting poor prognosis.