Aim: Although colour coded Doppler ultrasound (CCDU) is established as a non-invasive diagnostic tool for detecting renal artery stenoses (RAS), no uniform criterion for defining a hemodynamically relevant stenosis (i.e. angiographic stenosis of > or = 70% of the diameter) exists. We therefore investigated the predictive value of the renal-aortic flow-velocity ratio (RAR) > 3.5 m/s and a difference of < 0.05 between the left and right side for the resistance index according to Pourcelot (dRI) in the detection of a relevant RAS.
Patients and methods: We analysed 500 consecutive CCDU examinations of patients with hypertension retrospectively. An RAR > 3.5 and/or a lateral inequality of the RI < 0.05 were used as stenosis criterion.
Results: In 448 patients (90%) both renal arteries could be found, in 11 patients (2%) only the right artery, in 6 patients (1%) the left artery, and in 35 patients (7%) no renal artery was detectable. In 98 patients (19.6%), RAS was diagnosed, 69 (71%) of them underwent angiography. 38 patients presented an RAR > 3.5 plus dRI > 0.05. In 29 of these, angiography was performed. 96% of them presented with an RAS of > or = 70% and 4% showed an RAS of 40-69% (specificity 97%, sensitivity 76%). In 54 patients the RAR was > 3.5, but dRI < 0.05. 24% of the patients undergoing angiography (n = 37) presented with an RAS of > or = 70%, 68% with an RAS of 40-69%, and 8% with an RAS of < 40% (specificity 60%, sensitivity 100%). 44 hypertensive patients who underwent angiography after a CCDU examination not suggesting the presence of RAS were used as control group.
Conclusion: An experienced physician using a high quality colour-coded duplex-machine can reliably detect the renal arteries. The presence of RAS can be diagnosed with certainty by CCDU applying the criterion of RAR > 3.5, but the diagnosis of a one-sided haemodynamically relevant RAS can only be certain if the criterion of dRI > 0.05 is used in addition.