From Wiki Journal Club
Jump to: navigation, search
Brilakis ES, et al. "Drug-eluting stents versus bare-metal stents in saphenous vein grafts: a double-blind, randomised trial". Lancet. 2018. 391(10134):1997-2007.
PubMedFull text

Clinical Question

In patients with a history of coronary artery bypass grafting undergoing percutaneous coronary intervention (PCI) of a saphenous vein graft, do bare-metal stents (BMS) result in a higher rate of target-vessel failure compared to drug-eluting stents (DES)?

Bottom Line

In patients with a history of coronary artery bypass grafting undergoing percutaneous coronary intervention (PCI) of a saphenous vein graft, bare-metal stents (BMS) result in similar rates of target-vessel failure (defined as cardiac death, target vessel myocardial infarction, and target vessel revascularization) as drug-eluting stents (DES). At 12 months, the target-vessel failure rate was 19% in the BMS group and 17% in the DES group. Adverse event rates were similar in each group.

Major Points

Multiple high quality randomized controlled trials (e.g., NORSTENT) have established drug-eluting stents (DES) as the standard of care for percutaneous coronary intervention (PCI) in native coronary arteries. When compared to bare-metal stents (BMS), use of DES has been shown to result in lower rates of in-stent restenosis (ISR) as well as possibly stent thrombosis, leading to fewer target-vessel failures over the long-term. Although the superiority of DES over BMS has been well-established for PCI involving native coronaries, the comparative efficacy of DES over BMS has not been well-studied in PCI involving saphenous vein grafts, an increasingly common procedure. Furthermore, vein grafts typically allow for placement of stents with larger diameter, which generally leads to lower ISR rates, which may obviate the benefit of DES over BMS. Thus, a randomized trial was needed to compare the performance of BMS versus DES in patients undergoing PCI of vein grafts.

The 2018 Drug-eluting stents vs bare-metal stents In saphenous Vein graft Angioplasty (DIVA) trial randomized 597 Veterans Administration (VA) patients undergoing PCI of an existing vein graft to revascularization using BMS or DES and assessed for a primary outcome of target-vessel failure (a composite of cardiac death, target-vessel MI, or target-vessel revascularization). The study was terminated early due to slow recruitment and funding issues. At 12 months, the rate of target-vessel failure was similar in both the BMS (19%) and DES (17%) groups. Target-vessel failure rates remained similar when including all available follow-up (median 2.7 years, 37% for BMS and 34% for DES). There were no significant differences in any of the component endpoints. The adverse effect/complication profiles were similar in both groups.

Overall, the DIVA trial suggests that BMS is equivalent to DES in the setting of PCI of vein graft lesions. Although the study was terminated early due to slow recruitment and funding issues, the study had already recruited sufficient participants to provide 86% post-hoc power to detect a clinically significant difference in the primary outcome (close to initial target power). As a result, use of BMS over DES would seem reasonable for vein graft PCI in circumstances in which BMS may otherwise be preferable (the shortest duration of dual antiplatelet therapy possible is preferred, resource availability). Notably, the VA trial setting does limit generalizability of study findings to other populations (100% of study population was male, and 85% were white).


As of November 2018, no guidelines have been published that reflect the results of this trial.


  • Multicenter, randomized, double-blind, controlled trial
  • N=597
    • Bare-metal stent (N=305)
    • Drug-eluting stent (N=292)
  • Setting: 25 VA centers in US
  • Enrollment: January 1, 2012 to December 31, 2015
  • Duration follow-up for primary outcome: 12 months
  • Median follow-up: 2.7 years
  • Analysis: Intention-to-treat
  • Primary Outcome: Target-vessel failure (cardiac death, target-vessel MI, target-vessel revascularization)


Inclusion Criteria

  • Age ≥ 18 years
  • ≥ 1 significant de-novo vein graft lesion (defined as 50-99% stenosis of 2.25-4.5mm graft) with plan for PCI
  • Intent to use embolic protection device during PCI
  • Agreed to participate and use medications as prescribed

Exclusion Criteria

  • Planned non-cardiac surgery within 12 months of screening
  • STEMI presentation
  • Target vein graft is last remaining vessel or left main equivalent
  • Any history of percutaneous treatment of the target vessel within previous 12 months
  • Bleeding diathesis
  • Refusal to receive blood transfusion
  • Requirement for warfarin administration in the following 12 months and is considered high risk of bleeding with triple antithrombotic therapy
  • Recent positive pregnancy test, breastfeeding, or possibility of future pregnancy
  • Life expectancy < 12 months due to co-existing conditions
  • Any history of allergy or sensitivity to any drug or metal included in DES
  • Allergic to clopidogrel and did not present with ACS at sites that used blinded study medication
  • Participation in another study

Baseline Characteristics

From the BMS group.

  • Demographics: Age 68.2 years, male 100%, white 85%
  • Comorbidities: BMI 30.4, HTN 97%, HLD 96%, DM 61%, smoker 24%, previous MI 50%, AF 19%, CHF 39%, LVEF 49.4, PAD 18%
  • Native coronary disease: Years since CABG 12.8, three vessel disease 89%
  • PCI indication: stable angina 35%, unstable angina 31%, NSTEMI 24%, other 10%
  • Lesion characteristics: ostial SVG lesion 25%, body SVG lesion 67%, anastomotic SVG lesion 9%, TIMI 0 1%, TIMI I 4%, TIMI II 12%, TIMI III 84%
  • Intervention characteristics: Staged PCI 11%, hemodynamic support during PCI 1%, embolic protection device 69%, IVUS 19%, angiographic success 95%, target SVGs per patient 1.0, number of stents in target SVG per patient 1.4, total length of stents in target lesion per patient 26.6mm, target lesion stent diameter 3.42mm
  • Intra-procedural medications: UFH 58%, bivalirudin 45%, IIb/IIIa inhibitor 14%


  • Patients randomized 1:1 using phone randomization system to BMS or DES
  • Each patient received as many stents as indicated based on operator judgment
  • Randomization stratified by diabetes status and number of target vein graft lesions requiring PCI
  • Operators could use any stent design within allocated type
  • The same type of stent was used for all lesions whenever possible
  • Patients, referring physicians, study coordinators, outcome assessors masked to group allocations (interventionalists were not)
  • An independent clinical events committee adjudicated all possible primary outcome events
  • Study angiographic and electrocardiographic core laboratories reviewed the angiograms and electrocardiographs in a blinded manner to verify that events reported by the sites as definite stent thrombosis met the definition used in the trial
  • To preserve study blinding, at sites where standard of care was to treat patients receiving BMS with 1 month of a P2Y12 inhibitor, patients who did not present with an ACS and who did not require at least 12 months of P2Y12 inhibitor for clinical reasons were given clopidogrel or matching placebo after the first month for 11 months
  • Patients were followed-up every 3 months during the first year and every 6 months thereafter


Comparisons are DES versus BMS

Primary Outcomes

Target-vessel failure (cardiac death, target-vessel MI, target-vessel revascularization)
51 (17%) vs. 58 (19%); HR 0.92 (95% CI 0.63-1.34); p = 0.70

Secondary Outcomes

Cardiac death
17 (6%) vs. 16 (5%); HR 1.15 (95% CI 0.58-2.27); p = 0.72
Target-vessel MI
16 (5%) vs. 20 (7%); HR 0.83 (95% CI 0.43-1.60); p = 0.54
Target-vessel revascularization
34 (12%) vs. 34 (11%); HR 1.04 (95% CI 0.65-1.67); p = 0.74
Any death
23 (8%) vs. 21 (7%); HR 1.17 (95% CI 0.65-2.12); p = 0.63
2 (1%) vs. 3 (1%); HR 0.72 (95% CI 0.12-4.32); p = 0.71
Definite stent thrombosis
6 (2%) vs. 7 (2%); HR 0.85 (95% CI 0.28-2.53); p = 0.81

Safety Outcomes

Serious adverse event
231 (79%) vs. 244 (80%); p = 0.79
Complications associated with device
25 (9%) vs. 28 (9%); p = 0.79
Procedural related injuries and complications NEC
15 (5%) vs. 14 (5%); p = 0.76

Subgroup Analyses

  • There were no significant interactions between differences in the primary outcome and demographic, clinical, lesion, and procedural characteristics.


  • Failure to mask treatment allocation to interventionalists allows for possible bias (e.g., interventionalists may perform procedure differently if knowingly using BMS versus DES)
  • Although median follow-up was 2.7 years, only 68% of patients followed for at least 2 years, limiting generalizability of study findings to very late events
  • VA study population limits generalizability to other populations (e.g., women and nonwhite)


  • Funder was involved in study design, data collection, site monitoring, data analysis, data interpretation, and writing of the report, but did not influence the interpretation of the trial results nor the decision to submit the paper for publication

Further Reading