Correspondence
Dipyridamole plus Aspirin and Hemodialysis Graft Patency
N Engl J Med 2009; 361:720-721August 13, 2009
- Article
To the Editor:
Will the study by Dixon et al. (May 21 issue)1 change outcomes in clinical practice with respect to the use of dipyridamole plus aspirin for the treatment of hemodialysis graft stenosis? We suspect not. First, the external validity of the trial is difficult to assess, since the authors do not explain why a graft was chosen as the first access for nearly half the patients. Second, although treatment with dipyridamole plus aspirin modestly improved the predictably low primary patency of grafts,2 it is unlikely that such therapy actually reduced the risk of stenosis. Monthly flow monitoring was used to detect possible stenoses, yet confirmatory angiography was not performed in 53 of 127 patients with thrombosis who received dipyridamole plus aspirin (42%) and in 57 of 139 patients with thrombosis who received placebo (41%). Although access thrombosis is more complex than the simple paradigm of stenosis leading to thrombosis,3 many of these 110 patients may have had a stenosis that had gone undetected by flow monitoring.4 Thus, the modest between-group difference observed in angiographically detected stenoses could be entirely attributed to study design rather than differences in neointimal hyperplasia. This conclusion is far more likely, since there is no published evidence that dipyridamole inhibits neointimal hyperplasia in humans.
4 ReferencesAndy R. Weale, M.D.
David G. Cooper, M.D.
Gloucestershire Royal Hospital, Gloucester, United Kingdom
andy@weale.org. uk 1
Dixon BS ,Beck GJ ,Vazquez MA , et al. Effect of dipyridamole plus aspirin on hemodialysis graft patency. N Engl J Med 2009;360:2191-2201
Full Text | Web of Science | Medline2
Keuter XH ,Kessels AG ,de Haan MH ,van der Sande FM ,Tordoir JH . Prospective evaluation of ischemia in brachial-basilic and forearm prosthetic arteriovenous fistulas for hemodialysis. Eur J Vasc Endovasc Surg 2008;35:619-624
CrossRef | Web of Science | Medline3
Roy-Chaudhury P ,Sukhatme VP ,Cheung AK . Hemodialysis vascular access dysfunction: a cellular and molecular viewpoint. J Am Soc Nephrol 2006;17:1112-1127
CrossRef | Web of Science | Medline4
Tonelli M ,James M ,Wiebe N ,Jindal K ,Hemmelgarn B . Ultrasound monitoring to detect access stenosis in hemodialysis patients: a systematic review. Am J Kidney Dis 2008;51:630-640
CrossRef | Web of Science | Medline
To the Editor:
Dixon et al. conclude that dipyridamole plus aspirin had a significant effect in reducing the risk of stenosis of newly created arteriovenous grafts. As in previous studies,1,2 there was no difference in cumulative graft patency. According to Figure 2 of the article, the loss of primary unassisted patency was more than 30% at 3 months, slightly lower among patients receiving dipyridamole plus aspirin than among those receiving placebo. A subsequent loss of patency was similar. Perioperative factors, such as surgical technique, might have explained the initial difference. Interestingly, a significant effect of the clinical center on primary unassisted graft patency is reported in the Supplementary Appendix that accompanies the online version of the article, available at NEJM.org. A lack of angiographic data for 43% of patients with thrombosis in both study groups further weakens a causative link between the use of dipyridamole plus aspirin and the risk of stenoses. No study suggests that all arteriovenous grafts without thrombosis or flow reduction are without angiographic stenosis. The possible influence of surgical experience on successful maturation of arteriovenous fistulas was noted previously.3 To isolate the effect of surgical technique on vascular access outcomes, future studies may need to randomize pharmacologic interventions among participating surgeons.
3 ReferencesDirk M. Hentschel, M.D.
Brigham and Women's Hospital, Boston, MA
dhentschel@partners.org 1
Kaufman JS ,O'Connor TZ ,Zhang JH , et al. Randomized controlled trial of clopidogrel plus aspirin to prevent hemodialysis access graft thrombosis. J Am Soc Nephrol 2003;14:2313-2321
CrossRef | Web of Science | Medline2
Crowther MA ,Clase CM ,Margetts PJ , et al. Low-intensity warfarin is ineffective for the prevention of PTFE graft failure in patients on hemodialysis: a randomized controlled trial. J Am Soc Nephrol 2002;13:2331-2337
CrossRef | Web of Science | Medline3
Hentschel DM ,Asif A . Effect of clopidogrel on arteriovenous fistulas for dialysis. JAMA 2008;300:1647-1648
CrossRef | Web of Science | Medline
Author/Editor Response
Weale and Cooper question the external validity of our study on the basis of the frequent use of a graft as the first access. Grafts were the predominant and often first arteriovenous dialysis access type in the United States during the early phase of the trial. Subsequently, practice patterns evolved and fistulae became more prominent. We examined whether the use of dipyridamole plus aspirin was similarly effective for grafts placed as the first access (hazard ratio, 0.80; 95% confidence interval [CI], 0.65 to 0.99) or as a second or later access (hazard ratio, 0.83; 95% CI, 0.59 to 1.18) and found no significant difference (P=0.87). These findings support the generalizability of our trial's results for grafts placed subsequent to some other form of hemodialysis vascular access.
Weale and Cooper further question whether the use of dipyridamole plus aspirin reduced the risk of stenosis, since approximately 40% of thromboses were not evaluated with diagnostic angiography. Patients whose grafts were not studied angiographically often had thrombosis within the first 1 to 2 months, when surgical and perioperative factors rather than stenosis were the predominant causes of thrombosis. Nonetheless, as shown in Table 2 of our article, the effect of dipyridamole plus aspirin on the primary outcome, even in this group, was still in the same direction, and the difference between this group and patients with thrombosis who underwent angiography showing stenosis of 50% or more was not significant. Moreover, contrary to the assertion by Weale and Cooper, studies in humans have indicated that the use of dipyridamole plus aspirin slows the progression of stenosis, as cited in the article.1,2
Hentschel interpreted Figure 2 of the article as showing no effect of dipyridamole plus aspirin beyond 3 months and suggested that surgical technique rather than the inhibition of stenosis accounted for the results. The randomized, double-blind study design with stratification according to clinical center balanced the effect of surgical expertise on graft outcome in the two study groups. Thus, surgical expertise would not have accounted for the observed difference between patients receiving dipyridamole plus aspirin and those receiving placebo in the primary outcome. There is no evidence that the effect of dipyridamole plus aspirin on primary unassisted patency varied across clinical sites (see Table 1 in the Supplementary Appendix that accompanied the online version of our article). Moreover, contrary to Hentschel's interpretation, the cumulative incidence curves for the two study groups diverged only after the first month and continued beyond 6 months. This pattern is consistent with an effect of dipyridamole plus aspirin in slowing the progression of stenosis rather than an effect of variations in surgical technique.
2 ReferencesBradley S. Dixon, M.D.
University of Iowa, Iowa City, IA
bradley-dixon@uiowa.edu Harold I. Feldman, M.D.
University of Pennsylvania, Philadelphia, PA1
Chesebro JH ,Fuster V ,Elveback LR , et al. Effect of dipyridamole and aspirin on late vein-graft patency after coronary bypass operations. N Engl J Med 1984;310:209-214
Full Text | Web of Science | Medline2
Hess H ,Mietaschk A ,Deichsel G . Drug-induced inhibition of platelet function delays progression of peripheral occlusive arterial disease: a prospective double-blind arteriographically controlled trial. Lancet 1985;1:415-419
CrossRef | Web of Science | Medline
