Drug-eluting stents (DES) have been a major advance in interventional cardiology, but evidence for using these devices does not exist for all types of lesions or for all subsets of patients. One area where data have been lacking is the indication of diseased aortocoronary saphenous vein grafts (SVGs).A randomized trial (1) reported in this issue of the Journal narrows the evidence gap. In the SOS (Stenting Of Saphenous Vein Grafts) trial (1), the primary end point of binary restenosis was lower (9% vs. 51%, p Ͻ 0.0001) after the use of paclitaxel-eluting stents than after the use of similar bare-metal stents for lesions in SVGs of diameters of 2.5 to 4.0 mm. An important finding from this 80-patient study was that all-cause mortality was similar between the 2 groups at a median follow-up of 1.5 years.
See page 919Previous randomized study. Concerns about late mortality after using DES for SVG lesions emerged in an earlier randomized study. The initial publication of the RRISC (Reduction of Restenosis in Saphenous Vein Grafts with Cypher Sirolimus-Eluting Stents) trial (2) reported that the primary end point of late loss was lower after the use of sirolimus-eluting stents than after the use of bare-metal stents, but a follow-up study found unfavorable clinical outcomes (3). After a median follow-up of 32 months, 11 deaths occurred in the group receiving sirolimus-eluting stents (29%) but none occurred in the group receiving bare-metal stents (p Ͻ 0.001). Three deaths were sudden, and 1 was caused by stent thrombosis (3). Although the findings added to concerns about the long-term safety of DES, the imbalance was likely caused by the play of chance in a 75-patient study neither adequately powered nor prospectively designed to show a mortality difference. In any case, the reversal of conclusions from the study (2,3) supported the recurring theme that all medical knowledge, even that from a single trial, is provisional. Cohort studies. No evidence of harm has appeared in a recent review (4) of 8 retrospective cohort studies of DES for SVG lesions, although results were mixed. Two newer reports have continued the trend of presenting null or positive results. Gioia et al. (5) reported that 106 patients treated with DES for SVG lesions had similar rates of event-free survival (81% vs. 82%, p ϭ 0.9), target vessel revascularization (14% vs. 14%, p ϭ 1.0), or death (6% vs. 6%, p ϭ 1.0) at 2 years, compared with 119 patients treated with bare-metal stents. Brodie et al. (6) reported that 825 patients treated with DES for SVG lesions had lower rates of target vessel revascularization at 9 months than 361 patients with bare-metal stents (7.5% vs. 9.4%, p Ͻ 0.05). Persistent challenge of restenosis. Bare-metal stents, which have been widely used to treat SVG lesions, have produced only limited evidence of efficacy in randomized trials. The landmark SAVED (Saphenous Vein De Novo) trial generated mixed results (7). The primary end point of binary restenosis was not reduced by stenting compared with balloon angioplasty (37% vs....