Bedside assessment of cardiac hemodynamics: the impact of noninvasive testing and examiner experience

Knowledge of cardiac filling pressures is critical in the diagnosis and management of patients with dyspnea or heart failure. Echocardiography and B-natriuretic peptide (BNP) testing are commonly used to estimate these pressures, but their incremental value beyond physical examination remains unknown.

Right and left heart filling pressures were prospectively estimated as "normal" or "abnormal" by staff cardiologists and cardiovascular trainees based upon physical examination findings alone, or examination coupled with echocardiographic and BNP data in patients referred for cardiac catheterization. Net reclassification improvement was calculated to determine whether echocardiographic/BNP data had incremental value in the determination of right and left heart pressures.

Two hundred fifteen observations were made by 9 examiners in 116 consecutive patients. Right and left heart pressures were accurately predicted from examination alone in 71% and 60% of observations, respectively. Examination-based accuracy was greater for staff cardiologists compared with trainees for right heart (82 vs 67%, P=.03) and left heart pressures (71% vs 55%, P=.03). Exposure to echocardiographic and BNP data did not enhance accuracy beyond bedside examination alone, both for left heart pressures (net reclassification improvement=-0.004; 95% confidence interval, -0.12-0.12) and right heart pressures (net reclassification improvement=0.02, 95% confidence interval, -0.09-0.13).

Cardiac filling pressures can be estimated from physical examination with modest accuracy, which is enhanced with experience. While echocardiographic and BNP data predict cardiac filling pressures, they may not provide information of incremental value beyond examination alone. Rigorous teaching and practice of cardiac examination skills should continue to be emphasized during medical training.