Echocardiography versus right-sided heart catheterization among lung transplantation candidates

Background. Right-heart-catheterization and transthoracic echocardiography are routine tests to measure pulmonary artery systolic pressure among lung transplantation candidates. Echocardiography may be as accurate as right-heart-catheterization, without the inherent risks of an invasive test. Methods. We examined the correlation between pulmonary pressures estimated by echocardiography versus right-heart-catheterization among lung transplantation candidates and their correlation to measurements during lung transplantation. Our cohort included all lung transplantation candidates during 1997 through 2004 who initially underwent pulmonary pressure evaluation by right-heart-catheterization and echo cardiography, as well as measurements during lung transplantation. Results. Of the 106 candidates, evaluation by transthoracic echocardiography was possible in 79 (74.5%). Median pulmonary systolic pressures by right-heart-catheterization was 44.0 [33.2-50.0] mm Hg and by echocardiography 40.0 [32.5-51.5] mm Hg (r = 0.80, p < 0.0001). In 14 (17.7%) patients the difference between the 2 methods was > 20 min Hg. The median time interval between echocardiography and right-heart-catheterization was 65 [40-155] days. The median value of pulmonary systolic pressure measured during lung transplantation in 44 (70.1%) of 62 patients was 39.5 [31.0-50.0] mm Hg. The time interval right-heart-catheterization-to-lung transplantation was 143 [87-339] days and echocardiography-to-lung transplantation 229 [130-367] days. The correlation between measurements during lung transplantation and initial measurements by right-heart-catheterization and echocardiography were r = 0.50 and r = 0.31, respectively, with corresponding p values of p = 0.001 and p = 0.07. Conclusions. For lung transplantation candidates and a suitable transthoracic echocardiography estimate of pulmonary systolic pressure, the need for right-heart catheterization, with its inherent risks for complications, may be foregone. The weak correlation between the initial and intraoperative measurements, probably stemming from the significant time interval, suggests that serial measurements may be needed.