The aim of the study was to evaluate the diagnostic accuracy of Color Doppler US, CT Angiography (CTA), and GD-enhanced MR Angiography (MRA) compared with digital subtraction angiography (DSA) for the detection of renal artery stenosis in patients with clinically suspected renovascular hypertension. Fifty-eight patients with suspected renovascular hypertension were enrolled in the study. All patients underwent Color Doppler US, CTA and GD-enhanced MRA. DSA was the gold standard method for the number of renal arteries, existence and degree of stenosis, or evidence of fibromuscular dysplasia. DSA depicted 132 renal arteries, 16 stenoses, and 4 arteries with fibromuscular dysplasia. Color Doppler US failed to detect 1 main and 14 polar arteries. CTA depicted all main renal arteries and 7/16 polar arteries, but failed to detect stenosis in two accessory vessels. Likewise, MRA did not detect stenotic accessory renal arteries, depicted 9/16 polar renal arteries, but missed two main renal arteries. All methods depicted the four main renal arteries with fibromuscular dysplasia. The overall sensitivity, specificity, and positive and negative predictive accuracy were 75%, 89.6%, 60% and 94.6%, respectively, for color Doppler US; 94%, 93%, 71%, and 99%, respectively, for CTA; and 90%, 94.1%, 75%, and 98%, respectively, for GD-enhanced MRA. CTA and GD-enhanced MRA have comparable and satisfactory results with respect to the negative predictive accuracy of the suspected renal artery stenosis. The concept of an imaging algorithm including US as screening test when appropriate and CTA or MRA as the second step-procedure is suggested. Therefore, DSA may be reserved for cases with major discrepancies or therapeutic interventions.