Liver Stiffness Linked to Heart Failure Severity, Outcomes

Patrice Wendling

January 29, 2018

OSAKA, JAPAN — Liver stiffness measured on elastography at discharge reflects heart failure (HF) severity and is a strong predictor of adverse events, including cardiac death and HF rehospitalization, a new study suggests.[1]

Interest in transient elastography is growing beyond its role in staging liver fibrosis to cardiology because of the relationship between hepatic dysfunction and HF. Recent work has shown that liver stiffness on elastography is associated with right-sided filling pressure and compares favorably with echocardiography in patients with HF who do not have chronic liver disease.

First author, Dr Tatsunori Taniguchi (Osaka University Graduate School of Medicine, Suita, Japan), explained to theheart.org | Medscape Cardiology that assessing the right heart with echocardiography is challenging, whereas elastography "is a very simple, easy method that can be finished in a matter of minutes without pain for the patients and has small inter- and intraoperator variability."

Their report was published online January 17 in JACC Cardiovascular Imaging.

For the study, 171 hospitalized patients with patients (mean age, 65 years; 68% male) underwent single liver elastography (FibroScan, Echosens) exam before discharge and were stratified into three liver stiffness (LS) tertiles: low (≤4.7 kPa, corresponding to an estimated right atrial pressure [RAP] of 4.6 mm Hg); middle (4.7 to ≤6.9 kPa, estimated RAP, 6.9 mm Hg), and high (>6.9 kPa, estimated RAP ≥7.0 mm Hg).

On the basis of the results, the researchers found the highest LS tertile was significantly associated with NYHA class 3 HF, jugular venous distention, moderate/severe tricuspid regurgitation, low hemoglobin and hematocrit levels, high levels of direct bilirubin and type IV collagen 7S domain, and a large inferior vena cava (IVC) diameter.

With a median follow-up of 203 days, Kaplan-Meier curves showed that patients with the highest LS had a significantly higher risk for death or HF rehospitalization than those in the other two groups (low vs high tertiles, P<0.001; middle vs high tertiles, P=0.002).

LS also significantly predicted cardiac events in fully adjusted Cox regression models (hazard ratio [HR], 1.12 per 1-kPa increase; 95% CI, 1.08–1.16).

"The higher the liver stiffness, the greater the risk of the patient to die. "Stone liver, heart in danger is, in a way, the message of this work," Drs Mathieu Pernot and Olivier Villemain (both Institut Langevin, Paris, France) write in a related editorial.[2]

Prognostic Power

The researchers also examined serum B-type natriuretic peptide (BNP), and unsurprisingly it also predicted cardiac events (HR, 2.11 per 1-log-unit increase; 95% CI, 1.5–3.01). Although BNP and LS were poorly correlated, higher LS remained associated with a higher prevalence of cardiac events even after adjustment for natural log-transformed (ln) BNP.

"LS shows a different aspect of heart failure compared to BNP, so we can assess the patient with both parameters to get good outcomes. So for me, this is the most surprising," coauthor Dr Tomohito Ohtani (Osaka University Graduate School of Medicine) said in an interview.

The common liver function tests (LFTs) of γ-glutamyl transpeptidase, alkaline phosphatase, and direct bilirubin were also significantly associated with cardiac event risk in univariate and adjusted analyses that included age, sex, estimated glomerular filtration rate, and ln BNP. When LS was added to the model, however, none of the LFTs remained associated with cardiac event rate, Taniguchi and colleagues report.

Further, a category-free net reclassification analysis showed a significant improvement in risk stratification for adverse events for LS in addition to clinical variables (0.691; 95% CI, 0.352–1.029).

Finally, an LS cutoff of 10.1 kPa showed a sensitivity of 73% and specificity of 90% for risk for cardiac events at 90 days. LS also showed better discriminatory power for detecting short-term risk than did IVC diameter (C-statistic, 0.823 vs 0.698).

"Surrogates of Surrogates"

Commenting for theheart.org | Medscape Cardiology, Dr Clyde Yancy (Northwestern University, Chicago, IL) said the study is helpful and will inform future research, but the issue that must be reconciled with these and other data is that "we continue to look at surrogates of surrogates."

"If the surrogate marker we really want to know about is the filling pressure of the heart, then whether we're looking at a biomarker or change in impedance or in this case the tension within the liver, we are really looking at a surrogate that only gives us information about another surrogate," he said. "And what we really need to understand is what's happening to the patient in totality."

In the future, LS might be a unique component of a multimodality screening process that includes point-of-care or home biomarkers, changes in weight and symptoms, and even remotely monitored pulmonary artery pressures, Yancy said.

"We will find something that has sufficient fidelity that allows us to streamline the decision-making process, but we're not there yet," he said.

The editorialists say the study represents "a big step forward in understanding the clinical usefulness of LS," but also that "there remains a long way to go for LS to become a reliable and reproducible tool for the cardiologist."

Pernot and Villemain point out that of 226 eligible patients, 37 were excluded because of organic hepatic disease and another 18 because of invalid LS measurements. Also, the authors themselves state that the link between central venous pressure and LS was "conceivable" but not certain.

The study was supported in part by a grant-in-aid from the Japanese Society for the Promotion of Science and a research grant from Osaka Heart Club. The authors, the editorialists, and Yancy have disclosed no relevant financial relationships.

Follow Patrice Wendling on Twitter: @pwendl. For more from theheart.org | Medscape Cardiology, follow us on Twitter and Facebook.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....

Recommendations