Probability of Transplant (POT) Index: A Novel Method to Predict the Likelihood of an Adult Liver Transplant.

J. Moffett,¹ M. Kueht,¹ H. Petrowsky,² I. Riaz,³ C. Nalty,⁴ S. Khaderi,¹ C. O'Mahony,¹ J. Goss,¹ A. Rana.¹

¹Division of Abdominal Transplantation, Baylor College of Medicine, Houston, TX
²Department of Visceral and Transplant Surgery, University Hospital of Zurich, Zurich, Switzerland
³Department of Internal Medicine, University of Arizona, Tucson, AZ
⁴Department of Epidemiology and Biostatistics, Texas A&M Health Sciences Center, College Station, TX.

Meeting: 2016 American Transplant Congress

Abstract number: B253

Keywords: Multivariate analysis, Prediction models, Waiting lists

Session Information

Session Name: Poster Session B: Liver: MELD, Allocation and Donor Issues (DCD/ECD)

Session Type: Poster Session

Date: Sunday, June 12, 2016

Session Time: 6:00pm-7:00pm

Presentation Time: 6:00pm-7:00pm

Location: Halls C&D

Background: Access to liver transplantation is limited by a multitude of factors, which are quite variable among transplant candidates, and is underscored by the shortage of deceased donor allografts. The primary objective of this study was to identify factors predicting the likelihood a listed candidate would undergo liver transplantation (OLT). The secondary objective was to develop and validate a model to predict the likelihood of OLT with hopes of identifying candidates that may benefit from extended criteria donors.

Methods: To identify independent factors predicting the likelihood of OLT, we performed a multivariate Cox regression analysis of 62,244 candidates from the United Network for Organ Sharing database. We excluded candidates with exception points. We used the results of our multivariate analysis to construct a predictive index which was then validated in a separate cohort of patients.

Results: We identified 9 significant predictors of OLT for listed candidates: lab MELD score at listing, blood type, region of listing, retransplantation, hemodialysis, low albumin levels, life support, spontaneous bacterial peritonitis (SBP), and hospital admission. The most significant predictors of liver transplantation were blood group AB (hazard ratio [HR], 2.04; confidence interval [CI], 1.94-2.16) and listing in region 3 (HR, 2.02; CI, 1.95-2.10). When these predictors were incorporated into a probability of transplant (POT) index, the C statistic was 0.74. When applied to a contemporaneous cohort of patients undergoing OLT over a 2 year period, the C statistic was 0.81.

Conclusions: The POT index uses readily available information to predict the likelihood a candidate will undergo OLT. Although further validation is necessary, in this study, the POT index performed consistently well in predicting the likelihood of OLT.

CITATION INFORMATION: Moffett J, Kueht M, Petrowsky H, Riaz I, Nalty C, Khaderi S, O'Mahony C, Goss J, Rana A. Probability of Transplant (POT) Index: A Novel Method to Predict the Likelihood of an Adult Liver Transplant. Am J Transplant. 2016;16 (suppl 3).

To cite this abstract in AMA style:

Moffett J, Kueht M, Petrowsky H, Riaz I, Nalty C, Khaderi S, O'Mahony C, Goss J, Rana A. Probability of Transplant (POT) Index: A Novel Method to Predict the Likelihood of an Adult Liver Transplant. [abstract]. Am J Transplant. 2016; 16 (suppl 3). https://atcmeetingabstracts.com/abstract/probability-of-transplant-pot-index-a-novel-method-to-predict-the-likelihood-of-an-adult-liver-transplant/. Accessed May 31, 2025.

« Back to 2016 American Transplant Congress