Stenosis detection in native hemodialysis fistulas with MDCT angiography.

Abstract

The objective of our study was to assess the diagnostic value of 64-MDCT angiography in the evaluation of failing hemodialysis arteriovenous fistulas (AVFs) in comparison with conventional digital subtraction angiography (DSA). Thirty-six patients (22 men; mean age +/- SD, 65 +/- 15 years) with hemodialysis fistula dysfunction underwent MDCT angiography before DSA. Linear weighted kappa was used to calculate interobserver agreement for stenosis for both MDCT angiography and DSA on a 5-point scale. Accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for the detection of >/= 50% stenosis or occlusion on MDCT angiography was calculated using DSA as the standard of reference. Wilcoxon's signed rank test and Mann-Whitney U test were used to compare differences in image quality between MDCT angiography and DSA and between MDCT angiography with the patient's arm stretched overhead or alongside the body, respectively. Interobserver agreement for detecting stenosis was excellent for both DSA (kappa = 0.86; 95% CI, 0.81-0.91) and MDCT angiography (kappa = 0.82; 95% CI, 0.77-0.87). Accuracy, sensitivity, specificity, PPV, and NPV of MDCT angiography for detecting >/= 50% stenosis or occlusion was 92.0% (95% CI, 86.8-95.3%), 90.2% (77.8-96.3%), 92.8% (85.9-96.6%), 85.2% (72.3-92.9%), and 95.4% (89.0-98.3%), respectively. No significant difference in image quality was seen between MDCT angiography and DSA (p = 0.3008) or between MDCT angiography with the patient's arm stretched overhead or alongside the body (p = 0.2912). MDCT angiography is a reproducible and reliable imaging technique for detection of >/= 50% stenosis or occlusion in dysfunctional hemodialysis fistulas.