Affiliation:
1. Yong Loo Lin School of Medicine National University of Singapore Singapore Singapore
2. Department of Gastroenterology and Hepatology Singapore General Hospital Singapore Singapore
3. Biostatistics Unit, Yong Loo Lin School of Medicine National University of Singapore Singapore Singapore
4. SingHealth Duke‐NUS Transplant Centre Singapore Singapore
Abstract
AbstractBackground and AimThe model for end‐stage liver disease (MELD) was updated to MELDNa and recently to MELD3.0 to predict survival of cirrhotic patients. We validated the prognostic performance of MELD3.0 and compared with MELDNa and MELD amongst cirrhotic inpatients.MethodsDemographical, clinical, biochemical, and survival data of cirrhotic inpatients in Singapore General Hospital (SGH) from 01 January 2018 to 31 December 2018, were studied retrospectively. Patients were followed up from first admission in 2018 until death or until 01 April 2023. Area under the receiver operating characteristic curves (AUROC) were computed for the discriminative effects of MELD3.0, MELDNa, and MELD to predict 30‐, 90‐, and 365‐day mortalities. AUROC was compared with DeLong's test. The cutoff MELD3.0 score for patients at high risk of 30‐day mortality was determined using Youden's Index. Survival curves of patients with MELD3.0 score above and below the cutoff were estimated with Kaplan–Meier method and compared with log‐rank analysis.ResultsTotally 862 patients were included (median age 71.0 years [interquartile range, IQR: 64.0–79.0], 65.4% males, 75.8% Chinese). Proportion of patients with Child‐Turcotte‐Pugh classes A/B/C were 55.5%/35.5%/9.0%. Median MELD3.0/MELDNa/MELD scores were 12.2 (IQR: 8.7–18.3)/11.0 (IQR: 8.0–17.5)/10.3 (IQR: 7.8–15.0). Median time of follow‐up was 51.9 months (IQR: 8.5–59.6). The proportion of 30‐/90‐/365‐day mortalities was 5.7%/13.2%/26.9%. AUROC of MELD3.0/MELDNa/MELD in predicting 30‐, 90‐, and 365‐day mortalities, respectively, were 0.823/0.793/0.783, 0.754/0.724/0.707, 0.682/0.654/0.644 (P < 0.05). Optimal cutoff to predict 30‐day mortality was MELD3.0 > 19 (sensitivity = 67.4%, specificity = 82.4%). Patients with MELD3.0 > 19, compared with patients with MELD3.0 ≤ 19, had shorter median time to death (98.0 days [IQR: 28.8–398.0] vs 390.0 days [IQR: 134.3–927.5]), and higher proportion of 30‐day mortality (68.8% vs 43.0%) (P < 0.001).ConclusionMELD3.0 performs better than MELDNa and MELD in predicting mortality in cirrhotic inpatients. MELD3.0 > 19 predicts higher 30‐day mortality.