Testani et al presented retrospective correlation data on 618 paired cardiac catheterizations and echocardiograms with Doppler tricuspid regurgitation signals. The investigators reported limited agreement between Doppler estimated pulmonary artery (PA) systolic pressure and cardiac catheterization pressures.

Several points merit consideration. First, the investigators incorrectly reference Kircher et al’s report on noninvasive evaluation of the inferior vena cava to estimate right atrial pressure. In Testani et al, when the inferior vena cava was not well visualized (190 patients, or about 30% of the sample), the investigators assumed the right atrial pressure to be 14 mm Hg. However, this value was not mentioned by Kircher et al, who demonstrated that an inferior vena cava collapse of <50% or >50% determined whether right atrial pressure was ≤10 mm Hg (mean 6 ± 5 mm Hg) or ≥10 mm Hg, respectively and did not suggest adding an arbitrary value. In our experience, if any arbitrary value were added, it would be ≤5 mm Hg.

Second, the limitations of fluid-filled catheter tracings and the reproducibility of catheter-derived measurements should be emphasized when reporting the accuracy of measurement techniques. It is not clear that the investigators directly examined the tracings or relied on the procedure reports. Catheter-derived PA pressure recordings often show roller coaster tracings in response to respiration and defy measurement. The nonsimultaneous nature of the catheter-derived and echocardiographic measurements may provide another source of scatter. By correlating these hemodynamic “values” with the vague peak velocity of the tricuspid regurgitation jet, we believe that the study of Testani et al was preordained to demonstrate the failure of noninvasive measurement of peak systolic pulmonary pressure.

Finally, peak pressures during catheterization and echocardiography are known to be discordant when measuring velocities in aortic stenosis. The noninvasive evaluation of pulmonary hemodynamics can be successful if we avoid peak pressures and instead correlate the mean PA pressure by catheterization with the mean pressure by echocardiography. Mean PA pressure can be derived from the mean tricuspid regurgitation gradient, the peak pulmonary regurgitation gradient, the combination of the end-diastolic pulmonary regurgitation and peak tricuspid regurgitation gradients, and other methods.