Higher target lesion failure rates at 2 years with Absorb BVS vs Xience stents

Pearl Toh
30 Mar 2017
Dr Stephen Ellis
Dr Stephen Ellis
The everolimus-eluting bioresorbable vascular scaffold (BVS), Absorb, was noninferior to Xience metal stent for target lesion failure (TLF) by 1 year and between 1–2 years, but was associated with significantly higher cumulative TLF rates by 25 months, according to the ABSORB III trial presented at the ACC.17 in Washington, DC, US.    

The researchers attributed the difference in TLF rates at 25 months to the placement of Absorb in vessels with a reference vessel diameter (RVD) of <2.25 mm in some patients, which is smaller than the recommended size of 2.5–3.75 mm. 

“A key take-home is that this device shouldn’t be used in very small vessels,” said lead author Dr Stephen Ellis of Cleveland Clinic in Cleveland, Ohio, US. “This isn’t necessarily surprising because the Absorb BVS is considerably larger than the Xience stent, but it would have been preferable to have more consistency with regard to the vessel sizing and utilization of the device in the study.”

The multicentre, prospective, single-blind study randomized 2,008 patients (mean age 63 years) with stable ischaemic heart disease or stabilized acute coronary syndrome, who had significant plaque buildup in up to two areas of separate vessels, to receive Absorb BVS (n=1,322) or Xience stent (n=686). [ACC.17, LB-16265]

At 1 year, Absorb was noninferior to Xience for TLF, comprising a composite of target-vessel myocardial infarction (TV-MI), cardiac death, or ischaemia-driven target lesion revascularization (ID-TLR) (7.8 percent vs 6.1 percent, hazard ratio [HR], 1.3; p=0.15).

Similarly, between 1–2 years, TLF rates were comparable between the two groups (3.7 percent vs 2.5 percent, HR, 1.45; p=0.19), with a scaffold thrombosis rate of 0.3 percent (number needed to harm, 317) for Absorb.

However, the cumulative TLF rates by 25 months were significantly higher among patients receiving Absorb (11.0 percent vs 7.9 percent, HR, 1.42; p=0.03), driven mainly by a significantly higher rate of TV-MI in the Absorb group (7.3 percent vs 4.9 percent; p<0.04).   

Noting that an increased risk was associated with treating vessels with RVD <2.5 mm in a post hoc subgroup analysis, and that 19 percent of the ABSORB III cohort had stents placed in vessels <2.25 mm (as this correlates with a visual estimate of about 2.5 mm when vessels were assessed by coronary angiogram), Ellis and co-authors excluded patients with vessels <2.25mm (by quantitative coronary analysis) in a separate subanalysis.   

Subanalysis of only vessels of ≥2.25 mm showed a smaller and nonsignificant difference in TLF rates between Absorb and Xience (9.4 percent vs 7 percent, HR, 1.35; p=0.11).

“These results show that this device is generally comparable with the drug-eluting metal stent when the device is placed in appropriately-sized vessels and placed using appropriate procedural techniques,” said Ellis, adding that BVS is a first generation device and was used for the first time by most operators within this trial, and that the optimal implantation technique was still evolving during the initiation and enrolment of the study.

“Superiority of Absorb BVS is not likely to emerge before the bioresorption process is complete (approximately 3 years),” he suggested. “Longer-term data from the ABSORB III/IV program will determine whether better patient selection and technique improves short-term outcomes.”

Following the presentation, the FDA published a letter sent to healthcare providers to caution them on the increased risk with Absorb and to avoid using it in very small vessels.  

Higher target lesion failure rates at 2 years with Absorb BVS vs Xience stents

Editor's Recommendations