Current issues of ACP Journal Club are published in Annals of Internal Medicine


Benefit-risk trade-off was found for stent vs balloon coronary angioplasty

ACP J Club. 1995 Jan-Feb;122:10. doi:10.7326/ACPJC-1995-122-1-010

Source Citation

Fischman DL, Leon MB, Baim DS, et al. A randomized comparison of coronary-stent placement and balloon angioplasty in the treatment of coronary artery disease. N Engl J Med. 1994 Aug 25;331:496-501.



To compare the effectiveness of balloon-expandable coronary artery stent implantation with that of standard balloon angioplasty in patients with symptomatic coronary disease.


Randomized controlled trial with 6-month follow-up.


20 centers in North America, Japan, and Italy.


410 patients (mean age, 60 y; 78% men) with symptomatic ischemic heart disease and new lesions ≤ 15 mm in length with ≥ 70% stenosis in a coronary vessel ≥ 3 mm in diameter. Exclusion criteria were myocardial infarction (MI) within the past 7 days; contraindication to aspirin, dipyridamole, or warfarin sodium; left ventricular ejection fraction ≤ 40%; coronary thrombus; multiple focal lesions or diffuse disease; serious disease in the left main coronary artery; ostial lesions; or severe vessel tortuosity.


207 patients were allocated to stent placement, and 203 patients were allocated to standard balloon angioplasty.

Main Outcome Measures

Restenosis, minimal luminal diameter, event-free survival, revascularization of the target lesion, bleeding and vascular complications, and mean hospital stay.

Main Results

The clinical success rates were 96% in the stent group and 90% in the angioplasty group (P = 0.01). At 6 months, patients in the stent group had a larger luminal diameter than those in the angioplasty group (1.75 vs. 1.56 mm; P = 0.007) and a lower rate of restenosis (31.6% vs. 42.1%; {95% CI for the 10.5% absolute risk reduction, 0.2% to 20.8%; P = 0.046; relative risk reduction, 24.9%; number needed to treat, 10; CI, 5 to 472}*). Event-free survival was 81% in the stent group compared with 76% in the angioplasty group (P = 0.16). 10% of patients in the stent group had revascularization of the target lesion compared with 15% in the angioplasty group (P = 0.06). Bleeding and vascular complications occurred in 7% of patients in the stent group compared with 4% of those in the angioplasty group (P = 0.14). The mean hospital stay was longer in the stent group than in the angioplasty group (5.8 vs. 2.8 days, P < 0.001).


Angiographic outcomes were better, bleeding and vascular complications were more frequent, and hospital stays were longer in patients with symptomatic coronary artery disease who received stent implantation than in those who received standard coronary angioplasty. Clinical outcomes were similar for both groups.

Source of funding: In part, Johnson and Johnson Interventional Systems.

For article reprint: Dr. S. Goldberg, Jefferson Medical College, Division of Cardiology, Suite 403, 1025 Walnut Street, Philadelphia, PA 19107. FAX215-923-2539.

*Numbers calculated from data in article.


Benefit-risk trade-off was found for stent vs balloon coronary angioplasty

Coronary artery restenosis after balloon angioplasty occurs in 30% to 50% of patients. More than 50 randomized trials have failed to show a successful pharmacologic intervention, and the first 2 trials with directional atherectomy (1, 2) produced equivocal results. In these 2 stent trials, the angiographic results were remarkably similar for both post-procedural and follow-up minimum luminal diameter and percent diameter stenosis. Each showed an absolute 10% reduction in angiographic restenosis with stenting, a significant advance in this field.

The clinical outcome results can be interpreted in various ways. A more negative analysis than that presented by the authors can be summarized by tabulating the results for stenting rather than angioplasty in 100 patients. In the Benestent study, 10 patients were spared repeat angioplasty at a cost of 1 extra acute coronary artery closure, 2 more bypass operations, 10 extra bleeding or vascular complications, and 540 additional acute hospital days. The clinical outcome data in the Stent Restenosis Study (STRESS) are more difficult to interpret because the patients having angioplasty had a higher-than-expected bypass surgery rate (8.4%) and a lower-than-expected repeat angioplasty rate (12.4%) for a restenosis rate of 42%. For 100 patients, 1 less angioplasty and 3.5 fewer bypass operations with stenting cost 2 extra coronary artery closures, 4 more bleeding and vascular complications, and 300 extra acute hospital days. Not counted is the additional cost of the stent ($1600), procedure times that are often longer and involve more contrast medium use and radiation exposure than does angioplasty, rigorous periprocedural management, extended bed rest, and the inconvenience of monitoring anticoagulation for 1 month.

It should be noted that these trials were designed to test whether a new device was superior to balloon angioplasty in relatively large arteries (≥ 3.0 mm) that were at low risk for acute closure (short lesions, not ostial, without thrombus). It is a difficult comparison to win given the excellent results obtained by experienced high-volume operators with current angioplasty equipment. Moreover, restenosis is usually not a life-threatening condition. Most patients can be redilated electively with 1 balloon catheter and discharged from the hospital the next day.

Some coronary artery stenoses do not initially respond well to balloon angioplasty or recur despite successful angioplasty. Anecdotal experience suggests that stenting some of these lesions may produce a successful outcome. Stenting is already the treatment of choice of many for nonanastomotic saphenous vein bypass graft stenoses because restenosis rates are lower than with angioplasty (25% vs. 70%). A different coronary stent used for the unstable situation of acute closure that complicates angioplasty successfully decreases the rates of acute MI and emergency bypass surgery.

It is increasingly clear that the best means of decreasing the risk for restenosis is to obtain the largest possible acute luminal diameter. More recent stent deployment strategies (larger balloons), often using intravascular ultrasound to ensure complete stent expansion, can often totally relieve the stenosis or produce a larger diameter than the reference segments, an accomplishment not consistently obtainable by other catheter interventions. Because stenting in these 2 studies reduced restenosis rates despite a residual stenosis of 20%, the studies suggest that a greater effect may be achieved. Moreover, intravascular ultrasound-guided deployment and aggressive antiplatelet drug therapy may obviate the need for anticoagulation with heparin and warfarin, which would dramatically decrease length of hospital stay and number of groin complications, altering the cost analysis. In the future, it is possible that stenting will be the treatment of choice for arteries > 3 mm in diameter. Currently, however, balloon angioplasty remains the best strategy for de novo coronary artery stenoses.

Eric R. Bates, MD
University of Michigan Medical Center Ann Arbor, Michigan