Correspondence of aortic valve area determination from transesophageal echocardiography, transthoracic echocardiography, and cardiac catheterization.

The correspondence of aortic valve area measurements from transesophageal echocardiography, transthoracic echocardiography, and cardiac catheterization was determined in 100 patients with severe aortic stenosis (aortic valve area < or = 0.75 cm2), moderate aortic stenosis (aortic valve area > 0.75 to < or = 1.2 cm2), mild aortic stenosis (aortic valve area > 1.2 to < or = 2.0 cm2), and nonstenotic aortic valves (aortic valve area > 2.0 cm2). Because high correlation does not require high agreement, data were assessed by analysis of agreement. Aortic valve areas determined by transthoracic echocardiography (1.43 +/- 0.76 cm2), transesophageal echocardiography (1.40 +/- 0.79 cm2), and cardiac catheterization (1.47 +/- 0.82 cm2) were similar. Correlations between methods were excellent (r values = 0.96). "Limits of agreement" between the three comparisons did not differ significantly. Similar levels of agreement when comparing these independent methods for determining the aortic valve area indicate that direct planimetry by transesophageal echocardiography, the continuity equation with transthoracic echocardiography, and the Gorlin formula are equally accurate and may be used interchangeably. Clinically important discrepancies between methods are uncommon and are readily settled by adding a third method.