Early clinical experience and comparison between percutaneous and surgical arteriovenous fistula

Percutaneous arteriovenous fistula (pAVF) has been recently developed as an alternative to surgical AVF (sAVF). We report our experience with pAVF in comparison with a contemporaneous sAVF group.

Charts of all 51 patients with pAVF performed at our institution were analyzed retrospectively, in addition to 51 randomly selected contemporaneous patients with sAVF (2018-2022) with available follow-up. Outcomes of interest were (i) procedural success rate, (ii) number of maturation procedures required, (iii) fistula maturation rates, and (iv) rates of tunneled dialysis catheter (TDC) removal. For patients on hemodialysis (HD), sAVF and pAVF were considered mature when the AVF was used for HD. For patients not on HD, pAVF were considered mature if flow rates of ≥500 mL/min were documented in superficial venous outflow; for sAVF, documentation of maturity based on clinical criteria was required.

Compared patients with sAVF, patients with pAVF were more likely to be male (78% vs 57%; P = .033) and less likely to have congestive heart failure (10% vs 43%; P < .001) and coronary artery disease (18% vs 43%; P = .009). Procedural success was achieved in 50 patients with pAVF (98%). Fistula angioplasties (60% vs 29%; P = .002) and ligation (24% vs 2%; P = .001) or embolization (22% vs 2%; P = .002) of competing outflow veins were more frequently performed on patients with pAVF. The surgical cohort had more planned transpositions (39% vs 6%; P < .001). When all maturation interventions were combined, pAVF required more maturation procedures, but this was not statistically significant (76% vs 53%; P = .692). When planned second-stage transpositions were excluded, pAVF had a statistically significant higher rate of maturation procedures (74% vs 24%; P < .001). Overall, 36 pAVF (72%) and 29 sAVF (57%) developed mature fistulas. This difference, however, was not statistically significant (P = .112). At the time of AVF creation, 26 patients with pAVF and 40 patients with sAVF were on HD, all through use of a TDC. Catheter removal was recorded in 15 patients with pAVF (58%) and 18 patients with sAVF (45%) (P = .314). The mean time until TDC removal in pAVF group was 146 ± 74 days, compared with 175 ± 99 in the sAVF group (P = .341).

Compared with sAVF, rates of maturation after pAVF seem to be similar, but this result may be related to the higher intensity of maturation procedures and patient selection. An analysis of appropriately matched patients will assist in elucidating the possible role of pAVF vis-a-vis sAVF.