Liver dysfunction as predictor of prognosis in patients with amyloidosis: utility of the Model for End-stage Liver disease (MELD) scoring system

Francesco Cappelli, Samuele Baldasseroni, Franco Bergesio, Valentina Spini, Alessia Fabbri, Paola Angelotti, Elisa Grifoni, Paola Attanà, Francesca Tarantini, Niccolò Marchionni, Alberto Moggi Pignone, Federico Perfetto

Amyloidosis prognosis is often related to the onset of heart failure and a worsening that is concomitant with kidney-liver dysfunction; thus the Model for End-stage Liver disease (MELD) may be an ideal instrument to summarize renal-liver function. Our aim has been to test the MELD score as a prognostic tool in amyloidosis. We evaluated 128 patients, 46 with TTR-related amyloidosis and 82 with AL amyloidosis. All patients had a complete clinical and echocardiography evaluation; overall biohumoral assessment included troponin I, NT-proBNP, creatinine, total bilirubin and INR ratio. The study population was dichotomized at the 12 cut-off level of MELD scores; those with MELD score >12 had a lower survival compared to controls in the study cohort (40.7 vs 66.3 %; p = 0.006). Either as a continuous and dichotomized variable, MELD shows its independent prognostic value at multivariable analysis (HR = 1.199, 95 % CI 1.082-1.329; HR = 2.707, 95 % CI 1.075-6.817, respectively). MELD shows a lower prognostic sensitivity/specificity ratio than troponin I and NT-proBNP in the whole study population and AL subgroup, while in TTR patients MELD has a higher sensitivity/specificity ratio compared to troponin and NT-proBNP (ROC analysis-AUC: 0.853 vs 0.726 vs 0.659). MELD is able to predict prognosis in amyloidosis. A MELD score >12 selects a subgroup of patients with a higher risk of death. The predictive accuracy seems to be more evident in TTR patients in whom currently no effective scoring systems have been validated.