Retrograde occlusive arteriography of hemodialysis access: failure to detect inflow lesions?

Once thought to be a minor player in hemodialysis (HD) access dysfunction relative to outflow stenosis, inflow stenosis has recently come to be viewed as a major cause of access failure. Indeed, recent literature has shown that up to 40% of all accesses referred for dysfunction have an inflow lesion. Imaging of the inflow segment has been traditionally performed by interventional nephrologists via retrograde occlusive arteriography (ROA). Recent advances in our understanding of ROA have cast the technique in a negative light, with the possibility of vascular complications and poor diagnostic yield coming to the fore. Using a prospectively collected, vascular access database, we identified 18 consecutive patients who received imaging of inflow lesions by ROA and direct arteriogram (DA). The mean percent luminal stenoses were found to be 59.89 ± 24 and 79.06 ± 17.8 (p = 0.009) for the ROA vs. DA groups, respectively. Using multiple regression analysis, DA was found to be associated with detecting higher degree of luminal stenosis (β = 19.17, 95% CI 6.28-32.05, p = 0.006). This small case series provides evidence on the theoretical concern that ROA does not adequately evaluate inflow lesions. We may conclude that by relying solely on ROA, interventional nephrologists may be failing to detect a subset of hemodynamically significant inflow lesions.