Liver stiffness measurement for non-invasive assessment of central venous pressure in patients with heart failure: A multicentre pilot study

Abstract

Aims: Central venous pressure (CVP) is an important variable in assessing heart failure (HF) patients. However, invasive CVP measurement using right heart catheterization is associated with potential complications, and accurate measurement requires careful attention to technique. In this multicentre pilot study, we aimed to evaluate whether non-invasively measured liver stiffness can be used to assess CVP in patients with HF.

Methods and results: Patients aged 18 to 85 years with a history of HF (or left ventricular ejection fraction [LVEF] ≤40%) who required central venous line placement with CVP measurement were enrolled in an open-label study. Patients underwent simultaneous invasive catheter CVP measurement and non-invasive liver stiffness measurement (LSM) using vibration-controlled transient elastography (VCTE). The primary endpoint was the agreement between CVP and LSM-derived CVP (-5.9 + 6.8 × ln[LSM]) in the supine position. Among the 38 patients who completed the study, 34 (89.5%) were male, the mean age was 63.8 years, and the median LVEF was 24.5%. CVP correlated with LSM-derived CVP (Spearman r = 0.53, p < 0.001). Bland-Altman analysis demonstrated a mean difference (measured CVP minus LSM-derived CVP) of 0.11 mmHg (95% confidence interval -1.31, 1.54). LSM identified clinically significant CVP elevation (>10 mmHg) with an area under the curve of 0.83. An LSM threshold >10 kPa yielded a sensitivity of 0.73 and a specificity of 0.78, correctly classifying CVP (>10 or ≤10 mmHg) in 76% of patients.

Conclusions: Liver stiffness measurement is a promising non-invasive surrogate for CVP in patients with HF. Future studies are warranted to validate these findings and to assess the role of LSM for the screening, diagnosis, monitoring, and prognostic stratification of patients with HF.

Keywords: Central venous pressure; Heart failure; Liver stiffness; Non‐invasive assessment; Right heart catheterization; Vibration‐controlled transient elastography.

Figure 1

Bedside liver stiffness measurement using vibration‐controlled transient elastography.

Figure 2

Agreement between central venous pressure (CVP) and liver stiffness measurement (LSM)‐derived CVP according to the Taniguchi et al.'s formula. (A) Correlation between invasively measured CVP in the supine position and LSM‐derived CVP (Taniguchi et al.'s formula): Spearman correlation coefficient: 0.53 (95% confidence interval [CI] 0.25, 0.73); intraclass correlation coefficient: 0.55 (95% CI 0.28, 0.74). (B) Bland–Altman plot: mean difference 0.11 mmHg (95% CI −1.31, 1.54); lower limit of agreement: −8.38 mmHg (95% CI −10.85, −5.92); upper limit of agreement: 8.61 mmHg (95% CI 6.14, 11.08).

Figure 3

Correlation between liver stiffness measurement (LSM) and invasively measured central venous pressure (CVP) in the supine position. Spearman's rank correlation coefficient: 0.53 (95% confidence interval 0.25, 0.73).

Figure 4

Discriminatory ability of liver stiffness measurement (LSM) for detecting clinically significant central venous pressure (CVP) elevation (CVP >10 mmHg). According to the Youden's index, the optimal LSM cut‐off value for detecting an elevated CVP was 13.3 kPa, with a sensitivity of 0.73 and a specificity of 0.93.