Angiography Versus Intravascular Ultrasound-Directed Stent Placement - AVID


Given conflicting data regarding the benefit of the routine use of intravascular ultrasound (IVUS) to guide stent implantation, the AVID trial tested the hypothesis that IVUS-guided bare-metal stent placement would be superior to stent placement guided by angiography alone.


IVUS-guided bare-metal stent placement would be superior to stent placement guided by angiography alone.

Study Design

Study Design:

Patients Enrolled: 800
Mean Follow Up: 12 months
Mean Patient Age: 62.5 years
Female: 30
Mean Ejection Fraction: 54%

Patient Populations:

  • Age >18 years
  • Elective PCI
  • RVD ≥2.5 mm by visual estimation


  • Coronary dissection not covered by a stent
  • TIMI flow grade <3 post-stent placement
  • Chronic total occlusion
  • Stent placement in sole remaining circulation or left main equivalent
  • Stent placement within an aneurysmal portion of a vessel such that complete stent-vessel wall contact could not be achieved
  • A bypass graft supplying a native vessel <2.0 mm by visual estimate
  • Cardiac transplantation
  • Performance of IVUS during the index procedure before stent placement

Primary Endpoints:

  • TLR at 12 months

Secondary Endpoints:

  • All-cause mortality at 12 months
  • MI at 12 months
  • Stent thrombosis at 12 months
  • CABG at 12 months
  • Major adverse cardiac events at 12 months
  • All the above endpoints, including the primary endpoint at 30 days and 6 months

Drug/Procedures Used:

No IVUS was performed before stent placement in either group. In the angiography-directed (AG) arm, procedural success was defined as <10% stent diameter stenosis compared with the distal reference vessel, by visual estimation. IVUS was performed after this in a blinded fashion. In the IVUS-directed (ID) arm, identical criteria as above were applied for angiographic success. IVUS was then performed in an unblinded fashion, and assessed for optimal stent placement, based on IVUS criteria, such as full apposition, smallest cross-sectional area (CSA) ≥90% of the distal reference vessel CSA, and absence of noncovered dissections.

Concomitant Medications:

Ticlopidine 250 mg was administered twice a day for 2-4 weeks, which could be replaced by clopidogrel 75 mg daily for 2 weeks after August 1998. Aspirin 325 mg/day was administered for an indefinite period. No patients received glycoprotein IIb/IIIa inhibitors.

Principal Findings:

A total of 800 patients were randomized, 406 to the AG arm, and 394 to the ID arm. Baseline characteristics were fairly similar between the two arms. About 19% had prior coronary artery bypass grafting (CABG) and 25% had prior percutaneous coronary intervention (PCI).

The vessel treated was the left anterior descending in 39%, and right coronary artery in 34%. More than one vessel was stented in 8% of the patients. The mean lesion length was 13.0 mm, with an distal reference vessel diameter (RVD) of 2.81 mm, although about 31% of the vessels had a RVD of <2.5 mm by angiographic core laboratory analysis.

In the ID arm, of those who did not meet IVUS criteria for optimal stent placement, only 37% of patients received further therapy for an underexpanded stent. Additional therapy in the ID arm was mainly necessary for residual stenosis >10% (29.2%) and nonapposition (10.7%). The ID arm had a higher stent usage (1.53 vs. 1.38, p = 0.01), as well as a larger maximal balloon size (3.71 vs. 3.53, p < 0.001). Final stent area expansion compared with the CSA of the distal reference vessel lumen was significantly better in the ID arm (90.4% vs. 84.6%, p < 0.001).

At 12 months, there was no difference in the incidence of target lesion revascularization (TLR) between the AD and ID arms (12.0% vs. 8.1%, p = 0.08), although a difference was noted in patients with a prestent RVD ≥2.5 mm (10.1% vs. 4.3%, p = 0.01), and prestent % stenosis >70% (14.2% vs. 3.1%, p = 0.002). There was similarly no difference in the incidence of death (1.9% vs. 3.3%, p = 0.39), myocardial infarction (MI) (5.1% vs. 6.8%, p = 0.32), or stent thrombosis (1.1% vs. 1.4%, p = 0.75).


The results of the AVID trial indicate that IVUS-directed bare-metal stenting is not associated with improved clinical outcomes in patients undergoing elective PCI, although certain patients such as those with an RVD of >2.5 mm, and present stenosis of >70% may benefit. Interestingly, although mandated in the protocol of this trial, only 37% of underexpanded stents received additional therapy in the IVUS-directed arm. This may have biased the results of the primary endpoint toward the null.

Earlier IVUS trials had failed to show a definite benefit with the use of IVUS for routine PCI. This trial seems to suggest that certain patients may derive benefit by reduction in TLR with IVUS guidance. However, this trial was conducted more than 10 years ago, and there have been several procedural and technological advances in PCI since then, including in the use of periprocedural therapy. It is thus unknown if these results are valid in a contemporary population. Moreover, it is unclear if a similar benefit would be noted with drug-eluting stents, which have a significantly lower incidence of TLR to begin with.


Russo RJ, Silva PD, Teirstein PS, et al. A randomized controlled trial of angiography versus intravascular ultrasound-directed bare-metal coronary stent placement (The AVID Trial). Circ Cardiovasc Intervent 2009;2:113-23.

Keywords: Myocardial Infarction, Angina, Stable, Metals, Thrombosis, Drug-Eluting Stents, Cross-Sectional Studies, Constriction, Pathologic, Coronary Artery Bypass, Stents, Percutaneous Coronary Intervention

< Back to Listings