Predicting Early Graft Dysfunction and Mortality After Liver Transplant Using the De Ritis Ratio

Abstract

Predicting complications after liver transplantation (LT) remains challenging. We propose incorporating the De Ritis ratio (DRR), a widely known parameter of liver dysfunction, into current or future scoring models to predict early allograft dysfunction (EAD) and mortality after LT. A retrospective chart review was conducted on 132 adults receiving a deceased donor LT from April 2015 to March 2020 and their matching donors. Donor variables, postoperative liver function, and DRR were correlated with the occurrence of EAD, post-transplant complications expressed by the Clavien-Dindo score, and 30-day mortality as outcome variables. Early allograft dysfunction was observed in 26.5% of patients and 7.6% of patients who died within 30 days after transplant. Recipients were more likely to experience EAD when receiving grafts from donation after circulatory death (P=.04), donor risk index (DRI) >2 (P=.006), ischemic injury at time-zero biopsy (P=.02), longer secondary warm ischemia time (P < .05), or higher Clavien-Dindo scores (IIIb-V; P < .001). The DRI, total bilirubin, and DRR on postoperative day 5 yielded significant associations with the primary outcomes and were used to develop the Gala-Lopez score using a weighted scoring model. This accurately predicted EAD, high Clavien-Dindo, and 30-day mortality in 75%, 81%, and 64% of patients. Including recipient and donor variables in predictive models, and for the first time DRR, as a constituent, should be regarded to predict EAD, severe complications, and 30-day mortality post-LT. Further studies will be required to validate the present findings and their applicability when using normothermic regional and machine perfusion technologies.