Biomarkers predict cardiac risks after transjugular intrahepatic portosystemic shunt (TIPS) placement: a single-center prospective study

Sunday, March 30, 2025

3:18 PM – 3:27 PM CT

Location: Music City Center, 202 A

Presenting Author(s)

MH

Matthew Henry, MD, MS (he/him/his)

Resident
Medical College of Wisconsin, United States

Author/Co-author(s)

HZ

Helena Zaldivar Alcantara, MD

Faculty
Medical College Of Wisconsin, United States
PS

Philip T. Skummer, MD, MPH

Resident
Medical College of Wisconsin, United States
AS

Achuthan Sourianarayanane, MD

Faculty
Medical College of Wisconsin, United States
JK

Joohyun Kim, MD

Faculty
Medical College of Wisconsin, United States
Sarah B. White, MD, FSIR, MS

Faculty
Froedtert & Medical College of Wisconsin, United States
Eric J. Hohenwalter, MD, FSIR

Professor, Chief of Vascular and Interventional Radiology
Medical College of Wisconsin, United States
KR

Kaila C. Redifer Tremblay, MD

Faculty
Medical College of Wisconsin, United States

Purpose: To evaluate the response of four different biomarker values to transjugular intrahepatic portosystemic shunt (TIPS) placement and determine their correlation with post-procedure outcomes.

Materials and Methods: Patients undergoing TIPS between 2020-2023 were included in this study. Endothelin-1 (ET-1), tumor necrosis factor alpha (TNF-α), B-type natriuretic peptide (BNP), and Endothelial Nitric Oxide Synthase (eNOS) were measured in patients at their baseline volume status before undergoing TIPS and at various time points post-TIPS, in addition to standard of care practices. Patients were assigned to the cardiac group if right atrial pressure (RAP) post-TIPS was ≥15 mmHg, there was a ≥10 mmHg change in RAP, and/or peak systolic right ventricle pressure ≥ 46 mmHg. The control group included patients who did not meet the threshold of right heart pressure measurement changes. Logistic regression was used to model the relationship between biomarker levels and disease status. Receiver operator characteristic (ROC) curves were generated with area under the curve to assess model performance. The optimal cutoff value for sensitivity and specificity was determined using Youden’s index.

Results:

A total of 23 TIPS were performed with 11 (48%) patients in the cardiac group and 12 (52%) patients in the control group. The two cohorts were comparable at baseline besides cirrhosis etiology: 11 (91.7%) patients in the control group had alcohol induced cirrhosis whereas 7 (63.6%) patients in the cardiac group had nonalcoholic steatohepatitis (p< 0.05). Five cardiopulmonary adverse events occurred in the cardiac arm: cardiac failure, diastolic dysfunction, arrhythmias, and abnormal stress test. One adverse event, mild pulmonary artery hypertension, occurred in the control group, a significant difference compared to the cardiac group (p< 0.05).

eNOS and TNF-α were significantly elevated in the cardiac group across all timepoints (p< 0.05). TNF-α demonstrated 100% sensitivity and 75% specificity for predicting elevated RAP post-TIPS, and eNOS had an 80% sensitivity and 75% specificity. While BNP and ET-1 levels were comparable between the two groups at baseline, they were significantly elevated in the cardiac group at 2 and 4-6 weeks compared to baseline (p< 0.05). Baseline ET-1 had a 100% sensitivity and 71% specificity for predicting cardiac adverse events within 1-year of TIPS, and TNF-a had a 100% sensitivity and 57% specificity.

Conclusion: Biomarkers play a significant role in risk stratification of patients, potentially aiding in the prediction of adverse cardiac outcomes post-TIPS.