Length Matters: More Adverse Events With Absorb BVS at 1 Year in Long Lesions
Registry results show a threefold higher rate of TLF in lesions ≥ 60 mm, far exceeding the recommended length, compared with shorter lesions.
In patients with very long coronary lesions, use of Absorb BVS (Abbott Vascular) is associated with a high rate of TLF at 1 year, driven by MI and clinically driven TLR, according to a subanalysis from the GHOST-EU registry.
The findings should hammer home to physicians the importance of appropriate lesion selection and optimized technique when using the new devices, according to Salvatore Geraci, MD (San Giovanni di Dio Hospital, Agrigento, Italy), and colleagues.
Although this technology has been available in Europe for more than half a decade, the everolimus-eluting Absorb BVS became the first such device introduced in the United States just last summer following US Food and Drug Administration approval. Absorb BVS is indicated for use in lesions less than or equal to 24 mm in length, with a reference vessel diameter of ≥ 2.5 mm and ≤ 3.75 mm. The approval was based on data from the ABSORB III study, a head-to-head comparison of Absorb and a conventional everolimus-eluting cobalt-chromium stent (Xience, Abbott Vascular). Both devices showed similar 1-year rates of the primary endpoint of TLF (a composite of cardiac death, target-vessel MI, and ischemia-driven TLR).
Still, some concern has been raised by data showing a signal of more target-vessel MI and scaffold thrombosis with Absorb BVS. It has also been suggested by the ABSORB researchers that the potential advantages of a bioresorbable scaffold in comparison with a standard DES may not be apparent for 2 to 5 years—after the scaffold has completely disappeared.
Triple Rate of TLF for Longest Lesions
In a paper published online this week in JACC: Cardiovascular Interventions, Geraci and colleagues looked at use of the Absorb BVS in patients with varying lengths of coronary lesions. The substudy of the all-comers GHOST-EU registry included a group of 1,468 consecutive patients enrolled between November 2011 and September 2014 at 11 European centers. All had lesions that were treated with at least one Absorb BVS.
Lesions were categorized as: short (< 30 mm), moderate (30-60 mm), and long (≥ 60 mm). Long lesions were more likely than short or moderate lesions to include chronic total occlusions, bifurcations, and ostial lesions, and these patients were more likely than those in the other categories to undergo postdilatation and intravascular imaging.
At 1 year, clinical outcomes were “acceptable,” the authors noted, and comparable between patients with short and moderate lesions. However, those with lesions ≥ 60 mm had about triple the rate of the primary endpoint of TLF than patients in the short- or moderate-lesion group (14.3% vs 4.8% and 4.5%, respectively; P = 0.001). When individual components of the primary endpoint were analyzed, patients in the long lesion group had more MI and clinically driven TLR than the other groups, with no between-group differences in rates of cardiac death.
The rate of TVF also was higher in the long lesion group, while rates of scaffold thrombosis were numerically higher but not statistically different compared with the other two groups (P = 0.29).
On multivariable analysis, diabetes, ACS, and ostial lesions were found to be independent predictors of TLF at 1 year, as was total scaffold length (per 30-mm increase).
‘Breakthrough Potential’ Yet to Emerge
Importantly, the “long” lesions in the current study were more than twice as long as the maximum length recommended in the Absorb label. In an email, ABSORB III principal investigator Gregg Stone, MD (Columbia University Medical Center, New York, NY), said the finding of higher event rates at 1 year in those with ≥ 60 mm lesions was not surprising. Metallic DES, he noted, face the same higher 1-year event rates in such long lesions. As such, Stone also echoed a point made by the study authors, namely that the theoretical advantages of a bioresorbable scaffold over DES in longer lesions—recovery of vasomotor and endothelial function, with no permanent metallic cage to complicate future surgical interventions—remain to be proven with longer follow up.
“The breakthrough potential for BVS in long lesions would not be expected to emerge until after 3 years, once the scaffold is completely resorbed, which compared to metallic DES may reduce very late (> 3 year) restenosis and stent thrombosis, and preserve bypass graft options,” he said. “Moreover, the BVS implant techniques used in the early GHOST-EU registry were rudimentary compared to our present knowledge. Greater attention to lesion preparation, proper vessel, and scaffold sizing and routine high-pressure postdilatation may further improve outcomes.”
Geraci and colleagues point out that unlike the short or intermediate lesions, those ≥ 60 mm were more often associated with use of 2.5-mm Absorb BVS. In the ABSORB III trial, these small-diameter devices were associated with worse outcomes. Additionally, the greater lesion complexity of the longer lesions may have played a role in outcomes, they write.
With regard to technique, the investigators say the physical properties of the Absorb BVS must be taken into account, including the need to implant the rectangular-shaped struts “with minimal overlap and without gaps to minimize thrombogenicity and delayed neointimal coverage of the struts.”
Overall, they note, the composite endpoint seen in the long-lesion subgroup compares favorably with first- and second-generation DES, with the exception of the numerically higher rates of rates of thrombosis, TLR, and TVR.
Worrisome Signals Call for Restriction
In an editorial accompanying the study, Christian Spaulding, MD, PhD, and Nicole Karam, MD, MPH (European Georges Pompidou, Paris, France), note that some recent trials have suggested that “the BVS bullet may be missing the target.”
Based on the findings from GHOST-EU, they further conclude that the newer devices “do not seem to be the miracle solution” to the challenge of long lesions and, in fact, add “another piece to the worrisome picture on outcomes after BVS implantation with high rates of repeat revascularization and most of all numerically higher rates of stent thrombosis.”
As for the suggestion that optimizing technique is “the Holy Grail” to solving the problems of bioresorbable scaffolds, Spaulding and Karam say long procedures and costly imaging devices “are unlikely to speed the adoption of BVS in an era where costs restraints and improvement in efficiency have replaced patient and physician preference, especially if there is no clear benefit, or even worse, higher rate of adverse outcomes.”
While acknowledging that coronary devices from the BMS through the DES era required several iterations to optimize outcomes, Spaulding and Karam conclude that until better results can be demonstrated, the worrisome signals coming from randomized trials and large registries suggest that BVS use should be restricted to highly selected patients and lesions.
Photo Credit: Gregg Stone. Extracted from Perspectives on the ABSORB Randomized Trials: Take-home Messages for the Clinical Practitioner
Geraci S, Kawamoto H, Caramanno G, et al. Bioresorbable everolimus-eluting vascular scaffold for long coronary lesions: a subanalysis of the international, multi-center GHOST-EU registry. J Am Coll Cardiol Intv. 2017; Epub ahead of print.
Spaulding C, Karam N. The scaffold disappears but the GHOST remains. J Am Coll Cardiol Intv. 2017; Epub ahead of print.
- Geraci and Karam report no relevant conflicts of interest.
- Spaulding reports research grants from the French Ministry of Health; consulting fees from Abiomed, Medtronic, Medpass, and Zoll; and speaker fees from AstraZeneca, Bayer, Cordis, Eli Lilly, Lead-Up, The Medicines Company, Servier, and WebMD.