Objective: To compare the diagnostic value of colour Duplex scanning with arteriography for the detection of arterial disease of the aortoiliac arteries, femoropopliteal arteries and the origins of the tibial vessels.Design: Prospective, semi-blind study.Setting Vascular laboratory and radiology department, University Hospital.Methods: A total of 1658 arterial segments in 148 limbs were studied both by colour Duplex scanning and digital subtraction arteriography. Individual arterial segments were classified on the basis of peak systolic velocity ratios < 2.0, greater than or equal to 2.0 or an absent doppIer signal, as 0-49%, 50-99% diameter reduced, or occluded. The same arterial segments were similarly classified on the basis of arteriography and the two modalities were compared using a Kappa (k) analysis.Results: The overall agreement between arteriography and colour-coded Duplex ions k = 0.74 (95% CI, 0.70-0.78), this indicates substantial agreement. Kappa values (95% CI) from the aortoiliac, femoropopliteal and the origins of the infrapopliteal arteries were k = 0.59 (0.49-0.73; moderate agreement), k = 0.80 (0.76-0.84; substantial agreement) and k = 0.48 (0.35-0.61; moderate agreement) respectively.Conclusion: We conclude that there is substantial agreement between colour-coded Duplex and arteriography of the lower limbs, and that the ability of colour-coded duplex to plan and guide lower limb vascular interventions requires investigation.

Objective: To assess the effect of multisegment disease upon the accuracy of lower limb colour Duplex scanning.Design: Prospective, semi-blind study.Setting: Vascular Laboratory and Radiology Department, University Hospital.Methods: A total of 148 limbs (1106 individual arterial segments) were examined from the distal aorta to the origins of the tibial arteries by colour-coded Duplex and arteriography. Individual segments were graded as 0-49%, 50-99% diameter reduced ol occluded on the basis of peak systolic velocity ratios < 2.01 greater than or equal to 2.0 or an absent Doppler signal, and compared with similarly graded segments from blinded angiographic studies. The agreement (Kappa analysis) between Duplex and arteriography in segments adjacent to at least one proximal or distal greater than or equal to 50% diameter reducing lesion was then compared to the agreement between segments free of adjacent disease.Results: For isolated areas of disease, the kappa value (95% confidence interval) of agreement between Duplex ultrasonography and arteriography was 0.63 (0.53-0.73) and in the presence of neighbouring disease the value was 0.78 (0.73-0.83).Conclusion: We conclude therefore that providing appropriate criteria are used, Duplex assessment of lower limb arterial disease is not adversely affected by adjacent disease.