Unique Clinically Relevant Prognostic Indicators After TIPS Placement in Cirrhosis Patients with Pre-Existing Kidney...

Background: Transjugular intrahepatic portosystemic shunt (TIPSS) outcomes in patients with moderate-to-severe pre-existing kidney disease (PKD, stages G3a-G4) remain poorly characterized. This study aimed to identify potential predictors of mortality specifically in patients with an eGFR 15-59 mL/min/1.73 m². Methods: We retrospectively analyzed 68 cirrhosis patients with PKD (eGFR < 60 mL/min/1.73 m²) undergoing a TIPSS between April 2021 and April 2024. Clinical outcomes, renal function changes, and 12-month survival were assessed. Statistical analyses included paired t-tests with false discovery rate adjustment and Kaplan-Meier survival analysis to identify potential predictors of mortality. Results: The cohort (mean age 61.0 ± 8.3 years, 83.8% male, 79.4% with PKD G3a-G3b) showed modest improvement in renal function (creatinine 1.93 to 1.75 mg/dL, p = 0.031), though this biochemical change did not predict survival. Overall mortality was 36.8% (95% CI: 25.4-49.5%) at mean follow-up of 6.7 months. Traditional severity scores (MELD, Child-Turcotte-Pugh) showed no significant association with survival (p > 0.05 for all comparisons). In exploratory analyses, mortality was significantly higher in patients with the following: (1) uncontrolled diabetes before a TIPSS (55.2% vs. 25.9%; RR 2.35, 95% CI: 1.08-5.15, p = 0.032); (2) post-TIPSS infection (70.0% vs. 31.0%; HR 5.44, 95% CI: 1.54-19.23, p = 0.009); and (3) post-procedural cardiac events (85.7% vs. 31.1%; p = 0.005). These associations persisted after false-discovery rate adjustment but require prospective validation given the modest sample size and wide confidence intervals. Conclusions: In this exploratory single-center study of patients with moderate PKD undergoing a TIPSS, we observed associations between mortality and pre-TIPSS poorly controlled diabetes, infections, and cardiac events. These hypothesis-generating findings suggest potential areas for future research. Prospective multi-center studies are needed to validate these associations and determine whether interventions targeting these factors improve outcomes.