OPTN/SRTR 2020 Annual Data Report: Liver

Abstract

This year was marked by the COVID-19 pandemic, which altered transplant program activity and affected waitlist and transplant outcomes. Still, 8906 liver transplants were performed, an all-time high, across 142 centers in the United States, and pretransplant as well as graft and patient survival metrics, continued to improve. Living donation activity decreased after several years of growth. As of June 30, 2020, 98989 liver transplant recipients were alive with a functioning graft, and in the context of increasing liver transplant volume, the size of both the adult and pediatric liver transplant waitlists have decreased.On February 4, 2020, shortly before the pandemic began, a new liver distribution policy based on acuity circles was implemented, replacing donor service area-and region-based boundaries. A policy change to direct pediatric livers to pediatric recipients led to an increase in deceased donor transplant rates and a decrease in pretransplant mortality rate among children, although the absolute number of pediatric transplants did not increase in 2020. Among adults, alcohol-associated liver disease became the predominant indication for liver transplant in 2020. After implementation of the National Liver Review Board and lower waitlist priority for most exception cases in 2019, fewer liver transplants were being performed via exception points, and the transplant rate between those with and without hepatocellular carcinoma has equalized. Women continue to experience higher pretransplant mortality and lower rates of liver transplant than men.

Deep learning and the future of the Model for End‐Stage Liver Disease–sodium score

“…While Nagai et al's detailed analysis of their model's performance on different substrata of their cohort is a strength of their work, further study is required to account for the significant proportion of patients excluded from their cohort (including patients who received a transplantation prior to 90 days on the waiting list, and patients listed with exception points), who compete for the same organs. Given that most patients with high MELD‐Na scores at listing receive a transplantation well within 90 days, [ 10 ] these exclusions mean this model cannot be validated without bias on the population of patients who enter the waiting list at high risk. Statistical identification of populations on whom a model's predictions are unreliable remains an active area of research in machine learning, and one which must shape the design of future deep learning–based clinical risk scores.…”

Intraoperative Doppler Ultrasound for Detection of Early Postoperative Vascular Complications in Orthotopic Liver Transplants

“…Doppler ultrasonography (DUS) is the current modality of choice in monitoring postoperative vascular complications. It is a non-invasive and cost-efficient technique for determining adequate perfusion and outflow of the graft [ 5 , 6 ]. The use of DUS to detect postoperative vascular complications following LT is well-documented.…”

Development of Machine Learning Models Predicting Estimated Blood Loss during Liver Transplant Surgery

“…Although liver transplantation (LT) has emerged as the treatment of choice for patients with end-stage liver disease, patients who undergo LT experience a variety of complications, including infection, mortality, and surgical re-intervention [ 1 , 2 ]. The incidence of major hemorrhage and transfusion during LT over the past decade has decreased significantly [ 3 ], but major bleeding during surgery is commonly expected [ 4 , 5 ].…”