Non-invasive assessment of central venous pressure in heart failure: a systematic prospective comparison of echocardiography and Swan-Ganz catheter

Abstract

Assessing hemodynamics, especially central venous pressure (CVP), is essential in heart failure (HF). Right heart catheterization (RHC) is the gold-standard, but non-invasive methods are also needed. However, the role of 2-dimensional echocardiography (2DE) remains uncertain, and 3-dimensional echocardiography (3DE) is not always available. This study investigated standardized and breathing-corrected assessment of inferior vena cava (IVC) volume using echocardiography (2DE and 3DE) versus CVP determined invasively using RHC. Sixty consecutive HF patients were included (82% male, age 54 ± 11 years, New York Heart Association class 2.23 ± 0.8, ejection fraction 46 ± 18.4%, brain natriuretic peptide 696.93 ± 773.53 pg/mL). All patients underwent Swan-Ganz RHC followed by 2DE and 3DE, and IVC volume assessment. On 2DE, mean IVC size was 18.3 ± 5.5 mm and 13.8 ± 6 mm in the largest deflection and shortest distention, respectively. Mean CVP from RHC was 9.3 ± 5.3 mmHg. Neither 2DE nor 3DE showed acceptable correlation with invasively measured CVP; IVC volume acquisition showed optimal correlation with RHC CVP (0.64; 95% confidence interval 0.46–0.77), with better correlation when mitral valve early diastole E wave and right ventricular end-diastolic diameter were added. Using a CVP cut-point of 10 mmHg, receiver operating characteristic curve showed true positivity (specificity) of 0.90 and sensitivity of 62% for predicting CVP. A validation study confirmed these findings and verified the high predictive value of IVC volume assessment. Neither 2DE nor 3DE alone can reliably mirror CVP, but IVC volume acquisition using echocardiography allows non-invasive and adequate approximation of CVP. Correlation with invasively measured pressure was strongest when CVP is > 10 mmHg.