Buttonhole needling of arteriovenous fistulae (AVFs) was first described 30 years ago, but little evidence has been reported to show how it might differ from the standard rope-ladder technique. We carried out a randomized control trial comparing these two techniques. All suitable patients within the region were considered for recruitment. Patients were then randomized to continue with traditional rope-ladder needling or changed to buttonhole needling. A mean of 13.7 (median, 11) sharp needles were required to create a track. Nine of 22 patients in the buttonhole group reduced or stopped local anesthetic (LA) for needle insertion compared with one of 25 in the traditional group. The diameter of the buttonhole AVFs remained unchanged, whereas the control group increased in size by 30% +/- 7% (p < 0.01), equivalent to an absolute increase of 5 mm. There was no difference in bleeding times, but there was a preference for the buttonhole technique with 21 of the 22 patients and 15 of 23 nurses preferring buttonhole needling. Buttonhole needling is preferred by both patients and staff, reduces discomfort during needling and reduces AVF enlargement. There is a low level of complication.