Fish Oil and Hemodialysis Graft Patency

Abstract

MORE THAN 600 000 PATIENTS IN THE UNITED States have end-stage kidney disease, with the majority—nearly 400 000 patients— requiring long-term maintenance hemodialysis. Ready access to the bloodstream is essential for hemodialysis and may be provided by a surgically created arteriovenous fistula, arteriovenous graft, or central venous catheter. An arteriovenous fistula is preferred because of its lower risk of thrombosis and lower overall cost. However, delayed maturation and failure rates of 20% to 60% for new fistulas result in greater reliance on catheters, the least desirable type of hemodialysis access. Compared with a new fistula, an arteriovenous graft is easier to cannulate and can be used within a few weeks of surgical creation. Yet grafts are associated with a higher rate of thrombosis, with attendant increased morbidity and cost. Grafts were once the most common type of hemodialysis access in the United States. However, through the efforts of the Fistula First Initiative, the End-Stage Renal Disease Networks, and the medical community, the prevalence of grafts in the United States has recently declined to less than 25%. Given the challenges of fistula maturation, grafts remain an appealing option, if the problem of thrombosis can be solved. The underlying cause of thrombosis in grafts and fistulas is neointimal hyperplasia leading to stenosis, typically at the vein-graft or vein-artery anastomosis. Fish oil has measurable anti-inflammatory, antiproliferative, and antithrombotic effects as well as a salutary effect on endothelial function that could slow the development of neointimal hyperplasia and graft thrombosis. Fish oil has been shown to reduce distal occlusion in coronary vein grafts and is associated with lower overall cardiovascular disease mortality. In a small, randomized, placebo-controlled trial of 24 patients, supplementation with 4 g of fish oil daily (1 g/capsule, 4 capsules/d) decreased hemodialysis graft thrombosis from 75.6% to 14.9% at 1 year. In this issue of JAMA, Lok and colleagues report the results of a randomized placebo-controlled trial evaluating the effect of fish oil supplementation on patency of newly created hemodialysis grafts. Patients were randomized 7 days after graft surgery to receive either four 1-g capsules daily of fish oil (containing 400 mg of eicosapentaenoic acid and 200 mg of docosahexaenoic acid per capsule) or matching placebo capsules taken for 12 months. The primary outcome was the proportion of patients with loss of primary unassisted graft patency, defined as the first graft thrombosis or a surgical or radiological intervention on the graft. Because of slow recruitment, the study was terminated early before reaching its target goal of 232 patients. A total of 201 patients were randomized; 5 dropped out before receiving treatment, 2 in the fish oil group and 3 in the placebo group. A total 196 patients were ultimately treated, 99 with fish oil and 97 with placebo. The analysis was based on patients who completed the study rather than those randomized. At 12 months, loss of primary unassisted graft patency occurred in 48% of patients who received fish oil and 62% who received placebo, for a relative risk reduction of 22%, a difference that did not reach statistical significance. However, predefined secondary outcomes that measured the rate of occurrence of events, including the number of primary events per 1000 access-days (incidence rate ratio) and the loss of primary unassisted patency by the Kaplan-Meier method (hazard ratio), revealed reductions of 42% and 32%, respectively, both of which were statistically significant. Although other predefined secondary graft outcomes including the rate of thrombosis and rate of radiological or surgical interventions were also significantly reduced, cumulative graft patency was not prolonged. As is consistent with other trials of fish oil, Lok et al reported a statistically significant reduction in systolic and diastolic blood pressure and a reduction in the cardiovascular event rate among patients who received fish oil. However, the latter finding should be interpreted with caution, given the small sample size and overall small number of events. Nonetheless, given the high mortality from cardiovascular disease in the hemodialysis population, this latter observation should stimulate an appropriately powered trial evaluating the effect of fish oil for prolonging patient survival. In the study by Lok et al, there was no between-group difference in triglyceride levels, which has usually been a re-