Inaccuracy of Doppler echocardiographic estimates of pulmonary artery pressures in patients with pulmonary hypertension: implications for clinical practice.

BACKGROUND

Recent studies suggest that Doppler echocardiography (DE)-based estimates of pulmonary artery systolic pressure (PASP) may not be as accurate as previously believed. We sought to determine the accuracy of PASP measurements using DE compared with right-sided heart catheterization (RHC) in patients with pulmonary hypertension (PH).

METHODS

We compared DE estimates of PASP to invasively measure PASP during RHC in 160 consecutive patients with PH (part one). To account for possible changes in hemodynamics between DE and RHC, we then prospectively determined PASP in an additional 23 consecutive patients undergoing simultaneous RHC and DE (part two). Bland-Altman analyses were performed to evaluate the agreement between RHC and DE measurements of PASP. Accuracy was predefined as 95% limits of agreement within ± 10 mm Hg for PASP estimates.

RESULTS

In part one, there was moderate correlation between DE and RHC measurements of PASP (r = 0.68, P < .001). However, using Bland-Altman analysis, the bias for DE estimates of PASP was 2.2 mm Hg with 95% limits of agreement ranging from -34.2 to 38.6 mm Hg. DE estimates of PASP were determined to be inaccurate in 50.6% of patients. In part two, there was moderate correlation between DE and RHC measurements of PASP (r = 0.71, P < .01). However, despite simultaneous DE and RHC measurements, the bias for DE estimates of PASP was 8.0 mm Hg with 95% limits of agreement ranging from -28.4 to 44.4 mm Hg.

CONCLUSIONS

DE estimates of PASP are inaccurate in patients with PH and should not be relied on to make the diagnosis of PH or to follow the efficacy of therapy.