Child-Turcotte-Pugh versus MELD-XI identify distinct high-risk populations for heart transplantation following ventricular assist device placement

Research output: Contribution to journalArticlepeer-review

Abstract

Background: Patients with end-stage heart failure frequently have significant congestive hepatopathy requiring hepatology assessment prior to heart transplantation listing. An elevated Model for End-stage Liver Disease score with modification to exclude INR (MELD-XI) has been associated with increased mortality following heart transplantation (HT). This study's primary aim was to examine whether Child-Turcotte-Pugh (CTP) classification is associated with post-transplant mortality in patients bridged to transplant with left ventricular assist devices. Methods and Results: We conducted a retrospective analysis of 134 patients from our center. Age, CTP class, and MELD-XI at HT were included in the multivariate model for the primary outcome, which demonstrated a significant association between 1-year mortality and CTP class (CTP-A HR:.08, CI.01–.46, P <.01; CTP-B HR:.25, CI.05–1.2, P =.08; reference group CTP-C), and MELD-XI (HR: 1.15; CI: 1.03–1.28; P =.01), but no significant difference for age (HR:.97; CI:.93–1.01; P =.15). Only 13/33 patients with CTP improvement after assist device also had improvement in MELD-XI. Conclusions: Patients with relatively low MELD-XI scores with discordantly high CTP classification may be a distinct subset for whom MELD-XI underestimates the risk of mortality after heart transplantation compared to CTP.

Original languageEnglish
Article numbere14617
JournalClinical Transplantation
Volume36
Issue number5
DOIs
StatePublished - May 2022

Keywords

  • heart transplantation
  • heart-assist devices
  • liver diseases
  • severity of illness index

Fingerprint

Dive into the research topics of 'Child-Turcotte-Pugh versus MELD-XI identify distinct high-risk populations for heart transplantation following ventricular assist device placement'. Together they form a unique fingerprint.

Cite this