Use of a proactive duplex ultrasound protocol for hemodialysis access

Abstract

Objective: Arteriovenous fistula (AVF) creation is the preferred approach for hemodialysis access; however, the maturation of AVFs is known to be poor. We established a proactive early duplex ultrasound (DUS) surveillance protocol for evaluating AVFs before attempted access. This study determined the effect of this protocol related to improving AVF maturation.

Methods: From 2008 to 2013, 153 patients received new upper extremity AVFs and an early DUS surveillance protocol at a single academic institution. The protocol involved an early DUS evaluation before hemodialysis cannulation of the AVF at 4 to 8 weeks after AVF creation. A positive DUS result was identified as a peak systolic velocity of >375 cm/s or a >50% stenosis on gray scale imaging, along with decreased velocity in the outflow vein. Patients with positive DUS findings underwent prophylactic endovascular or open intervention to assist with AVF maturation. Nature of secondary interventions, as well as AVF patency and maturation, were assessed. Overall clinical outcomes and fistula patency were investigated.

Results: During the study period, 183 upper extremity AVFs were created in 153 patients, including 82 radiocephalic, 63 brachiocephalic, and 38 brachiobasilic AVFs. A mortality rate of 43% (n = 66) was observed in a median follow-up period of 34.5 months (interquartile range, 19.6-46.9). A total of 164 early DUS were performed at a median of 6 weeks (interquartile range, 3.4-9.6 weeks) after the initial creation. Early DUS showed nine AVFs were occluded and were excluded from further analysis. Hemodynamically significant lesions were found in 62 AVFs (40%); however, only 17 (11%) were associated with an abnormal physical examination. Positive DUS finding prompted a secondary intervention in 81% of the patients. Among those with positive early DUS findings, AVF maturation was 70% in those undergoing a secondary intervention compared with 25% in those not undergoing a prophylactic intervention (P = .011). Primary-assisted patency for AVFs with early positive and negative DUS findings were 83% and 96% at 6 months, 64% and 89% at 1 year, and 52% and 82% at 2 years, respectively (P < .001).

Conclusions: Early DUS surveillance of AVFs before initial access is reasonable to identify problematic AVFs that may not be reliably detected on clinical examination. Although DUS criteria for AVFs have yet to be universally accepted, proactive early postoperative DUS interrogation assists in the early detection of dysfunctional AVFs and improvement of fistula maturation. Despite improved patency in those with positive DUS findings who undergo prophylactic secondary intervention, overall patency remains inferior to those without an abnormality detected on early DUS imaging.

Fig 1.

Patient survival during follow-up period shown with the standard error (SE).

Fig 2.

Arteriovenous fistula (AVF) outcomes according to early positive and duplex ultrasound (DUS) findings and interventions.

Fig 3.

Correlation of peak systolic velocity on duplex ultrasound (DUS) imaging and stenosis percentage on fistulogram.

Fig 4.

A, Primary assisted patency of total arteriovenous fistulas (AVFs). B, Primary assisted patency according to early positive and negative duplex ultrasound (DUS) findings. C, Primary assisted patency according to AVF configuration. Data are shown with the standard error (SE).