Color-coded duplex sonography for noninvasive diagnosis and grading of renal artery stenosis.

The accuracy of color-coded duplex sonography (CCDS) in screening hypertensive patients for renal artery stenosis (RAS) was assessed using a semi-quantitative waveform analysis. Our special aims were to separate between moderate and high grade stenoses and to evaluate the accuracy of the method in imaging both the whole course of the renal arteries and accessory renal arteries. Included in the prospective, angiographically controlled study were 135 consecutive patients with 268 renal arteries, of which 195 arteries (73%) could be visualized both proximally and distally by CCDS. Only three of 15 accessory renal arteries could be identified by CCDS. In 42 RAS > or = 50% sensitivity of CCDS was 93%, specificity 92%, and overall accuracy 92%. The sensitivity in identifying RAS > or = 75% was 92%, and none of the high grade stenoses were missed. Because of difficulties in visualizing the middle portion of the renal artery, we carefully examined this part of the artery in 116 additional patients. Whereas the proximal and the distal parts could be visualized in 77% of the renal arteries, signals from the middle third could be derived only in 60% on the right, and in 39% on the left side. Provided that the renal arteries were visualized both proximally and distally, a hemodynamically effective RAS could be excluded with high probability. Moreover, exact grading of high-grade stenoses was possible in all cases but one. An advantage of CCDS over conventional duplex sonography appears to be the time-saving examination. Since a low prevalence of RAS impairs the positive predictive value of CCDS, the examination should be reserved for patients with a strong clinical suspicion of renovascular hypertension.