J. J. Kim1, E. Gifford1, V. T. Nguyen1, P. Chisum1, A. Zeng1, C. DeVirgilio1 1Harbor-UCLA Medical Center,Surgery,Torrance, CA, USA
Introduction: Mounting evidence suggests routine ultrasound vascular mapping prior to hemodialysis access surgery leads to higher rates of arteriovenous fistula (AVF) creation. Further, the diameter of vein by pre-operative ultrasound has been shown to correlate with subsequent fistula maturation. Studies define 2.5mm as a minimum vein diameter for autogenous fistula creation. However, whether ultrasonographic vein diameter accurately predicts intra-operative vein diameter is not clear.
Methods: Retrospective review of a prospectively collected database including all hemodialysis access procedures performed by a single surgeon between 2011 and 2014 was performed. Patients without pre-operative vascular mapping and those undergoing revision of AVF were excluded. Pre-operative ultrasound vascular mapping results as well as intra-operative measurements of artery and vein used for anastamosis were recorded. Comparison of ultrasound measurement to intra-operative measurement was performed of the same vein at the same site where the anastamosis was created. All vascular mappings were performed with rubber tourniquet in place and all intra-operative measurements were taken by a single surgeon after complete dissection of the vessel and dilatation of vein with heparinized saline. Two-tailed paired student's t-test was used to analyze vein diameters obtained by pre-operative ultrasound and intra-operative ruler. Chi-squared test was used to compare the number of veins meeting the 2.5mm minimum diameter between the two measurement modalities.
Results: One-hundred and sixty-one patients had pre-operative vascular mapping with subsequent intra-operative vessel measurements. The median age was 52 and most were male (60%, n=97,). One-hundred and three patients (64%) were already hemodialysis dependent at the time of surgery and 132 (82%) underwent their first access surgery. There were 79 brachiocephalic AVF, 37 brachiobasilic AVF, 31 radiocephalic AVF, 2 brachiobrachial AVF, and 11 arteriovenous graft placements. Mean vein diameter by vascular mapping was 3.2mm (SD 1.4) and mean vein diameter by intra-operative measurement was 4.4mm (SD 1.3). This difference was statistically significant (p<0.0001). Of the 161 veins, vascular mapping showed 103 (64%) to be adequate for AVF creation (diameter ≥ 2.5 mm) while intra-operative measurements showed 151 (94%) to be adequate for use (p<0.0001). In contrast, the mean diameter of artery was larger by pre-operative ultrasound (4.0mm vs. 3.4mm, p<0.0001).
Conclusion: Preoperative ultrasound vascular mapping with tourniquet may underestimate the size of vein found during surgery. Inadequate vein diameter measured by ultrasound should not preclude exploration for autogenous AVF creation.
