106 - Examining Racial Disparities in MELD Scores Predictions for Post-TIPS Mortality and Readmission

Monday, March 31, 2025

12:10 PM – 1:10 PM CT

Location: Music City Center, Expo Poster Area

Poster Presenter(s)

Arsenios K. Chriskos

Medical Student
Aristotle University of Thessaloniki, Greece

Author/Co-author(s)

Lucas C. Struycken, MD (he/him/his)

Assistant Professor
University of Arizona, United States
DG

Daniel Goldberg, MD

Assistant Professor
University of Arizona, United States
GW

Gregory J. Woodhead, MD, PhD

Assistant Professor
University of Arizona, College of Medicine, United States
IV

Ilaria Vittoria De Martini, MD

Assistant Professor
The University of Arizona, United States
JH

Jack Hannallah, MD, MBA, MPH

Assistant Professor
University of Arizona - Banner University Medical, United States
Shamar Young, MD

Professor
University of Arizona, United States

Purpose: Transjugular intrahepatic portosystemic shunt (TIPS) placement is a well studied treatment for symptomatic portal hypertension. Model for End-Stage Liver Disease (MELD), MELD-NA and MELD 3.0 have all been utilized to provide pre-TIPS risk stratification. However, racial demographics remain an understudied area in terms of outcomes following TIPS.

Materials and Methods: This single-center retrospective study examines outcomes following TIPS procedures with a focus on racial demographics and MELD scores. The study analyzed all patients who underwent TIPS, using Spearman's rank correlation coefficient to evaluate differences between racial groups. Patients were categorized by race, Non-Hispanic (NH) or Hispanic, and MELD scores prior to the procedure, with a threshold of 20 applied to assess its impact on outcomes. In total 6 groups of patients were analyzed. NH and Hispanics with MELD, MELD Na and MELD 3.0 < 20 and NH and Hispanics with all MELD scores mentioned ≥20.

Results: The study included 200 TIPS procedures, with 65% (130/200) of patients being NH and 35% (70/200) Hispanic. The cohort comprised 60.5% male (121/200) and 39.5% female (79/200) aged 32 to 73 years. As expected, NH patients with a MELD score of ≥20 had a 22.6% increased risk of readmission or death within 1 month post-TIPS(p=0.017), while Hispanics had a 38.9% higher chance (p=0.002). No significant difference in outcomes was found when comparing NH and Hispanic patients with a MELD (p=0.772), MELD NA (p=0.596), and MELD 3.0 (p=0.875) < 20. Similarly, when comparing NH and Hispanic patients with a MELD (p=0.109) or MELD 3.0 (p=0.192) ≥20 no significant difference was found. However, Hispanic patients with a MELD NA ≥20 had a significantly greater likelihood of perishing or being readmitted within 30 days (p=0.037) as compared to NH patients.

Conclusion: These results may indicate that Hispanic patients with a MELD NA ≥20 may be at an increased risk for readmission or death at 30 days, as compared to non-Hispanic patients. Interestingly similar results were not seen in the MELD or MELD 3.0 analyses. This may suggest that MELD NA may provide more information on pre-TIPS risk assessment than MELD or MELD 3.0 in patients of Hispanic descent. As new MELD tests are developed and our awareness of outcome variations in different ethnicities increases, the need to personalize the risk stratification method to the patient will increase as well and further data evaluating these differences is needed.