reducing the propensity to develop neointimal hyperpla- sia. 1 Ligation necessarily requires additional dissection and mobilization of the segment to be ligated. We found that signs of venous hypertension were remarkably absent in our patients. This finding may reflect the venous flow coming out of the hand into the fistula, potentially promoting maturation, with patent valves preventing retrograde flow. The very few cases of retro- grade venous flow at later times were due to stenoses that developed in the proximal cephalic vein, close to the elbow. Regarding the potential for additional dilation, we reported that both venous and arterial diameters reach a plateau (Figure 4 in the original paper); this may serve as a natural limitation of flow in the artery. The maximal blood flow we have observed in any patient to date was 1400 mL/min at 18 months of follow-up. Neither venous nor arterial aneurysmal dilation occurred in our experi- ence. We believe that steal syndrome is extremely unlikely to occur using the RADAR technique, because ligation of the distal artery prevents the retrograde flow necessary for steal, and ligation is the first-line interven- tion recommended for hand ischemia. We did not encounter a single case of hand ischemia in our RADAR patients; however, we did not perform RADAR in cases of a diseased ulnar artery, because these patients have a high risk of ischemia with any arteriovenous fistula. Regarding fistula ligation, because the radial article pediclednot the isolated radial arterydis dissected for the RADAR procedure, the secondary dissection of the radial artery for fistula closure is technically easy, because it remains free of scar tissue. In addition, because the artery typically elongates a little during maturation of the arteriovenous fistula, it is likely that radial artery recon- struction will not require an interposition graft. Alterna- tively, in patients without ischemic symptoms, radial artery reconstruction is probably not necessary. However, our experience with ligation is necessarily limited. Regarding our traditional technique, our rate of primary patency at 1 year was 48%, lower but still similar to 55% reported in the meta-analysis by Al Jaishi et al. 2 We believe that our patency rate reflects our close surveillance to aggressively detect and treat stenoses. We are currently planning a randomized trial between conventional and RADAR techniques that will give contemporary rates. Regarding vein diameter as a factor for failure, the vein diameter was not a significant risk factor for juxta- anastomotic stenosis (P ¼ .07; Table III in the original paper). We believe that this data suggests that the RADAR technique has improved outcomes independent of venous diameter. We do not have experience with the end-to-end anastomotic configuration. If a patient can have this geometry, then RADAR can likely be performed, with the added benefit of minimal venous manipulation. Thank you for your comments. We agree that careful follow-up and aggressive appropriate intervention are important components of maintaining access. Nirvana Sadaghianloo, MD Serge Declemy, MD Department of Vascular Surgery University Hospital of Nice Nice, France Medical School University of NiceeSophia Antipolis Nice, France Alan Dardik, MD, PhD Department of Surgery Yale University School of Medicine New Haven, Conn Vascular Biology and Therapeutics Program Yale University School of Medicine New Haven, Conn REFERENCES 1. Sadaghianloo N, Dardik A, Jean-Baptiste E, Rajhi K, Haudebourg P, Declemy S, et al. Salvage of early-failing radiocephalic fistulae with techniques that minimize venous dissection. Ann Vasc Surg 2015;29:1475-9. 2. Al-Jaishi AA, Oliver MJ, Thomas SM, Lok CE, Zhang JC, Garg AX, et al. Patency rates of the arteriovenous fistula for hemodialysis: a systematic review and meta-analysis. Am J Kidney Dis 2014;63:464-78. http://dx.doi.org/10.1016/j.jvs.2016.08.102 Regarding “Carotid endarterectomy significantly improves postoperative laryngeal sensitivity” We enjoyed the paper by Hammer et al 1 regarding the effect of carotid endarterectomy (CEA) on laryngeal sensitivity and voice. Although voice alteration consists a relative common complication after CEA, it has not been extensively examined in the literature to date. The authors noted that a significant number of their patients, about 30%, developed a supraglottic hematoma on the operated side of the neck, this rather indicating a damage caused in the operating field than a damage caused during intubation. Herewith, we would like to underline our results in a similar group of patients. 2 In a comparable proportion of patients who underwent CEA (about 23%), a small but significant temporary alteration of voice quality was noted after measuring certain voice parameters. Interestingly, these patients were those patients who required an high level surgical dissection at the base of the skull. These patients had undergone a 930 Letters to the Editor Journal of Vascular Surgery March 2017