Use of high-resolution endoluminal sonography to measure the radius and wall thickness of esophageal varices

Background: Measurement of variceal wall tension theoretically provides the most accurate method of predicting future variceal bleeding. Using high-resolution endoluminal sonography in 45 patients with known portal hypertension, we measured and correlated the two previously unmeasured variables involved in the calculation of variceal wall tension (radius and wall thickness) by the Laplace equation. Methods: A 20 MHz 6.2F ultrasound transducer was used to image esophageal varices during standard esophagoscopy. All images were captured on videotape and later reviewed by two blinded investigators. Outer and inner variceal wall circumferences were measured at a cross section of each varix. The radius of each varix and the variceal wall thickness were calculated. The radius of each varix was then correlated with its wall thickness. The interobserver and intraobserver variabilities were measured. Results: The mean variceal radius was .86 +/- .34 cm for the inner radius and 1.48 +/- .41 cm for the outer radius; mean variceal wall thickness was .099 +/- 0.037 cm. Intraobserver and interobserver correlation for the radius was r=.98 and r=.97, respectively. The intraobserver and interobserver correlations for the wall thickness were r=.92 and r=.91, respectively. Variceal radius did not correlate with the wall thickness of the varix. Conclusions: High-resolution endoluminal sonography provides a method far the accurate measurement of esophageal variceal radius and wall thickness. Variceal radius does not correlate with variceal wall thickness, implying that variceal wall tension cannot be accurately estimated by measurement of variceal size alone. Combining these data with measurements of variceal pressure should allow for the direct determination of wall tension and, subsequently, identification of patients at risk for variceal bleeding.