Aorta-to-coronary saphenous vein grafts (SVGs) are the most widely used technique in patients who undergo coronary artery bypass graft (CABG) surgery, but data from the Randomized Multicenter Radial Artery Patency Study (RAPS; NCT00187356) demonstrate that radial artery grafts have better long-term angiographic patency. Stephen Fremes, MD, MSc, University of Toronto and Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada, presented findings from the 5-year analysis of RAPS, a multicenter, randomized clinical trial of 561 patients that compared both kinds of grafts.

One-year outcomes from RAPS were published in 2004 [Desai ND et al. N Engl J Med 2004]. The primary hypothesis was that radial artery grafts would be superior to SVG at 1 year and again at 5 years of follow-up. Patients who were undergoing nonemergency CABG with graftable triple-vessel disease and an estimated ejection fraction >35% were enrolled and received a left internal mammary artery bypass graft to the left anterior descending artery, one radial artery graft, and one SVG. The location of the radial artery graft (right coronary vs left circumflex) was selected at random, with the other artery receiving a SVG. In this type of randomization scheme, variance is minimized, since randomization is performed within rather than between patients, with each patient serving as his or her own control. The primary endpoint at 1 year was the proportion of total graft occlusion, and perfect graft patency (TIMI flow grade 3) was a secondary endpoint. Other secondary endpoints included proximal and distal anastomotic stenosis and stenosis in the body of the graft. Patients were excluded if they had renal insufficiency or the inability to utilize both potential conduits (ie, patients with varicose veins or vein stripping, nonpalpable ulnar arteries or positive Allen’s test on clinical exam, abnormal upper extremity Doppler ultrasonography, vasculitis, or Reynaud’s syndrome). The primary statistical analysis was performed on an intention-to-treat basis, with a p value of < 0.048 considered to indicate significant superiority, considering a single interim analysis.

Postoperative angiography was performed at 1 year in 440 of the 561 enrolled patients; complete graft occlusion was higher in SVGs than in radial artery grafts (13.6% vs 8.2%; p=0.009), a relative risk reduction of 40%. Diffuse narrowing of the graft was more frequent in radial artery grafts than SVGs (7.0% vs 0.9%; p=0.001). In patients with patent grafts, angiographic stenosis at the proximal anastomosis was higher with radial artery grafts than with SVGs (21.4% vs 11.1%, p < 0.001). Radial artery grafts had less stenosis in the graft body (5.7% vs 12.3%; p=0.003), with no significant difference at the distal anastomosis. Perfect graft patency (ie, TIMI grade flow 3) was similar for both grafts (87.7% for radial vs 85.7% for saphenous). Clinical endpoints could not be compared between graft strategies, considering that randomization was within rather than between patients; however, overall mortality was 1.4% at 1 year, and perioperative myocardial infarction was similar (~3%) between the radial and SVG regions.

Five-year angiographic follow-up was available in 269 patients. In this subgroup, the mean age was 60 years, 15% was female, one-third of procedures were for an urgent indication, and one-third was diabetic. In this 5-year follow-up analysis (mean interval from surgery 7.6 + 1.5 years), the authors swapped the original primary endpoint of proportion of total graft occlusion for functional graft occlusion (TIMI flow grade 0–2). Nevertheless, this subgroup still demonstrated an association with less total graft occlusion (TIMI grade flow 0) in the radial artery versus SVG group (8.9% vs 17.8%; OR, 0.50; 95% CI, 0.32 to 0.80; p=0.004). Functional graft occlusion was also lower for radial artery grafts compared with SVGs (12.0% vs 18.8%; OR, 0.64; 95% CI, 0.41 to 0.98; p=0.05). In grafts with TIMI 3 flow, proximal and distal anastomotic stenosis was similar for both grafts, but stenosis in the body of the graft was more common with SVGs (15.2% vs 6.7%; p=0.02). This translated into a reduction in complete occlusion or stenosis in the radial grafts (33.8% vs 21.9%; OR, 0.58; 95% CI, 0.40 to 0.86; p=0.004).

Overall, among patients who were undergoing elective CABG, the RAPS study demonstrated that radial arteries are associated with an approximate 9% sustained benefit from graft occlusion and less graft disease than saphenous veins at 5 years. This translates into a “number needed to treat” with radial bypass (in place of SVG) of ~12 patients to prevent 1 additional graft occlusion.