NORSTENT - Drug-eluting versus bare-metal stent for CAD (short)

Bonaa KH, et al. Drug-eluting or bare-metal stents for coronary artery disease. N Engl J Med 2016 [online]

Bottom-line: In patients with CAD who received PCI with coronary stent placement and received DAPT with ASA + clopidogrel x9 months, there were no difference in death or MI whether patients received 2nd-gen drug-eluting stents or bare-metal stents. 

Drug-eluting stents reduced the need for further revascularization by 3.3% (absolute) at 6 years.

Also worth mentioning: Regardless of stent type implanted, patients received 9 months of dual-antiplatelet therapy (DAPT; with clopidogrel). Despite this, patients with drug-eluting stents had lower risk of stent thrombosis than those with bare-metal stents. Combined with the evidence showing no statistically-significant difference in stent thrombosis risk between 3 vs 12 months of DAPT in patients with 2nd-generation drug-eluting stents, this provides evidence against using bare-metal stents in patients with increased risk of bleeding to permit shorter DAPT duration.

Context:

• Previous best-available evidence, which consisted of indirect/network meta-analysis of heterogeneous RCTs, suggested better long-term outcomes, including lower risk of death and stent thrombosis, with drug-eluting stents (especially "2nd generation" everolimus- or zotarolimus-eluting stents) compared to bare-metal stents.
• General principles:
  • Stent restenosis: Balloon angioplasty > bare-metal stent > drug-eluting stent
  • Stent thrombosis risk with stents: Bare-metal ~= 2nd-gen drug-eluting (everolimus, zotarolimus) < 1st-gen drug-eluting stents (sirolimus < paclitaxel) 
  • The overall balance of restenosis and in-stent thrombosis risk were thought to lead to lower death/MI with 2nd-gen drug-eluting stents versus bare-metal and 1st-gen drug-eluting stents

Patients (n=9013)

• Multicenter (all 8 Norwegian PCI cneters); every person who got PCI in Norway Sept 2008-Feb 2014 potentially eligible for this trial
• Inclusion:
  • Adults presenting with stable angina or ACS with lesion in native coronary arteries or coronary-artery grafts amenable to stent implantation
• Exclusion:
  • Previous coronary stent
  • Bifurcation lesion requiring 2+ stent technique
  • On warfarin
  • Life expectancy <5 y due to condition other than CAD
• 12,425 eligible -> 9013 randomized
• Average patient
  • 63 y/o
  • Male 75%
  • Smoker ~35%
  • Diabetes ~13%
  • PCI indication: stable angina 30%, UA 12%, NSTEMI 31%, STEMI 27%
  • Multivessel disease 40%
  • Procedure characteristics: 1-2 stents implanted, total length ~27 mm

Interventions

• I: Drug-eluting stent (DES)
  • ost common: Everolimus-eluting (Promus 67%, Xience 15%), zotarolimus-eluting (Endeavor Resolute 11%)
• C: Bare-metal stent (BMS)
  • Most common: Driver 43%, Integrity 22%, Liberte 18%)
• Co-interventions common to both groups:
  • ASA 75 mg/d indefinitely
  • Clopidogrel 75 mg/d x9 months
  • Other secondary prevention therapy per current guidelines

Results @ ~6 years

• No significant difference in primary outcome (death, non-fatal spontaneous MI)
  • DES 16.6% vs BMS 17.1% (p=0.66)
• Other non-significant secondary outcomes:
  • Death: 8.5% vs 8.4%
  • Spontaneous MI: 11.4% vs 12.5%
  • Stroke: 3.4% vs 3.0%
• Statistically-significantly different secondary outcomes:
  • Any revascularization (CABG or PCI): 16.5% vs 19.8% (p<0.001, number needed to treat 31 at 6 years)
  • Definite stent thrombosis: 0.8% vs 1.2% (p=0.05)

Issue with internal validity?

• No: Allocation-concealed, open-label RCT analyzing intention-to-treat population, 0% lost-to-follow-up
  • Low risk of allocation, performance, detection or attrition bias