Cardiac volume overload after long-term follow-up in TIPS patients

Transjugular intrahepatic portosystemic shunt (TIPS) is an effective treatment to reduce portal pressure in cirrhotic patients with symptomatic portal hypertension. Insertion of TIPS can be complicated by acute cardiac decompensation in rare cases, yet less is known about long-term cardiac outcome in these patients. Thus we analysed cardiac function during long-term follow-up after TIPS.

Patients that received TIPS or were seen for TIPS follow-up at the University Hospital Heidelberg between 2000 and 2013 were included for retrospective analysis. Patients with Budd Chiari syndrome (BCS) were excluded, as were patients without any clinical follow-up after TIPS placement. Cardiac Laboratory (high-sensitivity troponin T [hs-TnT] and N-terminal of the prohormone brain natriuretic peptide [NT-proBNP]) as well as findings on echocardiography were included for analysis. Last echocardiography before TIPS and last one at least 1 year after TIPS were included for each patient if available. Values are reported as means and standard deviation, the Wilcoxon test for paired samples was used to compare results. Continuous variables from echocardiography were dichotomized according to clinically relevant cut-off values and results pre- and post- TIPS were compared using Fisher's exact test. Statistical analyses were performed using SPSS.

A total of 157 patients were included in the final study cohort after 22 were excluded due to BCS and 31 because no follow-up after TIPS was available. Mean age at placement of TIPS was 61 years ( ± 12) and 51 (32.9%) patients were female. 31 patients were classified as having Child A, 93 as Child B and 33 as Child C cirrhosis at time of TIPS insertion. With regards to NT-proBNP and hs-TNT we found no significant difference after the first year of follow-up.

An enlarged (> 40 mm) left atrial diameter (LAD) was found in 43 of 102 patients before and in 28 of 45 patients at greater one year after TIPS insertion (P= 0.032). A trend towards more pathological results (> 55 mm) post-TIPS (2 of 46) were also observed for the End-diastolic diameter (EDD) compared to pre-TIPS (0 of 103, P= 0.094). The ratio of increased (> 35 mmHg) systolic pulmonary artery pressure (sPA) rose from 5 of 65 patients pre-TIPS to 13 of 42 after TIPS (P= 0.003). No difference was observed for septal thickness.

Analysis of paired samples revealed an increase of mean LAD from 37.41 mm ( ± 6.17) pre-TIPS to 41.67 mm ( ± 6.55) at more than one year post-TIPS (P= 0.001, n= 24). EDD increased from 44.00 mm ( ± 6.27) to 48.49 mm ( ± 4.50) (P= 0.002, n= 23). E'/A' ratio increased from 1.25 ( ± 0.69) to 1.35 ( ± 0.58) (P= 0.046, n= 6). Septal thickness was 10.93 mm ( ± 1.59) before and 11.56 mm ( ± 1.37) after TIPS (P= 0.012, n= 27). No significant differences were found for the end-systolic diameter, the right ventricular diameter and sPA.

In conclusion, TIPS is accompanied by long-term effects on cardiac function in cirrhotic patients. Our findings indicate that cardiac volume stress and an increase in pulmonary artery pressure is present in these patients. Thorough cardiac evaluation before TIPS should be performed and patients should undergo regular cardiopulmonary follow-up.

Corresponding author: Wannhoff, Andreas

E-Mail: [e-mail]