Background The long-term results after second generation everolimus eluting bioresorbable vascular scaffold (Absorb BVS) placement in small vessels are unknown. Therefore, we investigated the impact of vessel size on long-term outcomes, after Absorb BVS implantation.
Methods In ABSORB Cohort B Trial, out of the total study population (101 patients), 45 patients were assigned to undergo 6-month and 2-year angiographic follow-up (Cohort B1) and 56 patients to have angiographic follow-up at 1-year (Cohort B2). The pre-reference vessel diameter (RVD) was <2.5 mm (small-vessel group) in 41 patients (41 lesions) and ≥2.5 mm (large-vessel group) in 60 patients (61 lesions). Outcomes were compared according to pre-RVD.
Results At 2-year angiographic follow-up no differences in late lumen loss (0.29±0.16 mm vs 0.25±0.22 mm, p=0.4391), and in-segment binary restenosis (5.3% vs 5.3% p=1.0000) were demonstrated between groups. In the small-vessel group, intravascular ultrasound analysis showed a significant increase in vessel area (12.25±3.47 mm2 vs 13.09±3.38 mm2 p=0.0015), scaffold area (5.76±0.96 mm2 vs 6.41±1.30 mm2 p=0.0008) and lumen area (5.71±0.98 mm2 vs 6.20±1.27 mm2 p=0.0155) between 6-months and 2-year follow-up. No differences in plaque composition were reported between groups at either time point. At 2-year clinical follow-up, no differences in ischaemia-driven major adverse cardiac events (7.3% vs 10.2%, p=0.7335), myocardial infarction (4.9% vs 1.7%, p=0.5662) or ischaemia-driven target lesion revascularisation (2.4% vs 8.5%, p=0.3962) were reported between small and large vessels. No deaths or scaffold thrombosis were observed.
Conclusions Similar clinical and angiographic outcomes at 2-year follow-up were reported in small and large vessel groups. A significant late lumen enlargement and positive vessel remodelling were observed in small vessels.
- Coronary Artery Disease
Statistics from Altmetric.com
The first generation Absorb everolimus eluting bioresorbable vascular scaffold (Absorb BVS, Abbott Vascular, Santa Clara, California, USA) has been previously tested in the First-In-Man ABSORB Cohort A study, in a series of 30 patients1 ,2 with excellent long-term clinical results reported up to 4 years.3 In the second generation Absorb BVS, further modifications in the manufacturing processes and scaffold design resulted in a more durable and uniform support and drug delivery to the vessel wall.4 ,5 This second generation Absorb BVS has subsequently been tested in the ABSORB Cohort B trial, which enrolled 101 patients.
Due to the single size availability of this second generation scaffold (3.0 mm in diameter), in ABSORB Cohort B Trial the angiographic inclusion criteria were appropriately restrictive, in that the target lesion must have been located in a native coronary artery with pre-procedural reference vessel diameter (RVD) of 3.0 mm. However within the study, it was found that the visual estimation of the RVD was often incorrect, with a significant number of Absorb BVS being deployed in vessels with a pre-RVD smaller than 2.5 mm by quantitative coronary angiography (QCA). Among the 102 treated-lesions, 41 lesions had a RVD<2.5 mm (small-vessel group) and 61 lesions had a RVD mm≥2.5 mm (large-vessel group).6 Comparable clinical and angiographic outcomes in small and large vessels have been previously reported 6 months post implantation.6
A theoretical concern for the implantation of the Absorb BVS in small coronary arteries was the thick struts of this device, a characteristic that was reported in previous studies to be associated with an increased risk of restenosis especially with small vessel diameter.
In the present study we evaluate the impact of vessel size on clinical and angiographic outcomes at 1- and 2-year follow up after implantation of the second generation Absorb BVS. This follow-up time period is critical due to the expected loss of mechanical support of the Absorb BVS with the potential return of vasomotion properties of the treated vessel at 1 year,7 ,8 and the expected substantial scaffold polymer bioresorption at 2 years.5
Study design and population
The study design and the study device have previously been described.9 In brief, the ABSORB Cohort B Trial is a multicentre single-arm, prospective, open-label trial assessing the safety and performance of the Absorb BVS (Abbott Vascular, Santa Clara, California, USA). All patients were older than 18 years, and had a diagnosis of stable or unstable angina or silent ischaemia. As per-protocol, treated-lesions were a maximum of two, de-novo lesions in separate native coronary arteries with a visually estimated diameter of 3.0 mm, a length shorter than 14 mm and a percentage diameter stenosis greater than or equal to 50% and less than 100%. Major exclusion criteria were patients presenting with an acute myocardial infarction or unstable arrhythmias, or those who had left ventricular ejection fraction less than 30%, restenotic lesions, lesions located in the left main coronary artery, lesions involving a side branch more than or equal to 2 mm in diameter, and the presence of thrombus or other clinically significant stenoses in the target vessel. The ethics committee at each participating institution approved the protocol and each patient gave written informed consent prior to inclusion. A total of 101 patients were enrolled in the ABSORB Cohort B study. Out of the total population, 45 patients (45 lesions) were assigned to undergo 6-month and 2-year angiographic follow-up (ABSORB Cohort B1), and 56 patients (57 lesions) 12-month angiographic follow-up (ABSORB Cohort B2). In the present investigation, evaluation of the impact of vessel size (RVD<2.5 mm or RVD≥2.5 mm) on angiographic outcomes was undertaken at 1 year in the ABSORB Cohort B2 population and at 2 years in ABSORB Cohort B1 population; at the same time points, clinical outcomes in the entire ABSORB Cohort B population (i.e. ABSORB Cohort B1 plus B2) were also evaluated.
Quantitative coronary angiography (QCA) analyses were performed with the Coronary Angiography Analysis System (Pie Medical Imaging, Maastricht, Netherlands). The 37 μm platinum radio-markers located at each end of the Absorb BVS aided in the localisation of the non-radio-opaque scaffold for QCA. The following QCA parameters were computed: pre-procedural RVD calculated with interpolated method,10 minimal luminal diameter (MLD), percentage diameter stenosis (%DS), in-scaffold acute gain and binary restenosis. Late loss was defined as the difference between the post-procedural and follow-up minimal luminal diameter.
Furthermore, to adjust the absolute change in MLD to the vessel size (pre-RVD) post-procedurally and at follow-up, the reporting of relative gain and relative loss—previously demonstrated to be more representative of the real injury to the vessel wall and the subsequent neointimal response11 ,12—were undertaken. The absolute net gain, the net gain index and the loss index were also computed.
Those parameters were defined as:
Relative gain: (post-procedure MLD minus pre-procedure MLD) divided by vessel size.
Relative loss: (post-procedure MLD minus MLD at follow-up) divided by vessel size.
Absolute net gain: MLD at follow-up minus pre-procedure MLD.
Net gain index: net gain normalised for the vessel size and calculated as (MLD at follow-up minus pre-procedure MLD) divided by vessel size.
Loss index: this is the relation of late loss to acute gain and is calculated as (MLD at follow-up minus post-procedure MLD) divided by (post-procedure MLD minus pre-procedure MLD).
IVUS and IVUS-VH analysis
Scaffolded segments were analysed post-procedurally and at follow-up with phased array intravascular ultrasound (IVUS) catheters (EagleEye; Volcano Corporation, Rancho Cordova, California, USA). An automated pullback of 0.5 mm per second was utilised. IVUS images were analysed off-line with semi-automatic contour detection provided by dedicated software. The vessel area, scaffold area, lumen area, plaque area, in-scaffold neointimal area and lumen area stenosis were measured using a computer based contour detection program. IVUS-VH provides tissue characterisation using an autoregressive spectral analysis classification system. Each tissue component is quantified and colour-coded as follow: fibrous tissue in green, fibrofatty in greenish yellow, necrotic core in red and calcium in white. For each frame the absolute and the percentage amount of each tissue component is assessed.
Optical Coherence Tomography (OCT) analysis
As additional information an OCT data comparison between large and small vessels was performed in the subgroup of patients undergoing OCT imaging at baseline and follow-up. In ABSORB Cohort B Trial OCT imaging was performed at baseline and follow-up in selected centres as an optional investigation. The OCT M3 (Time Domain-OCT) and C7 (Fourier Domain-OCT) systems (LightLab Imaging Inc., Westford, Massachusetts, US) were used. The vessels were imaged using an automated pullback system at 20 mm/s (C7 system) and 3.0 mm/s (M3 system). During image acquisition, coronary blood was replaced by continuous flushing of contrast at 3.0–4.0 ml/s. Cross-sectional images were acquired at 100 frames/sec for C7 and at 20 frames/sec for M3. The analysis of continuous cross-sections was performed at each 1 mm longitudinal interval within the treated segment. Quantitative measurements were performed as previously described7 ,9 ,13 ,14
The composite end point of ischaemia-driven major adverse cardiac events (ID-MACE) comprised cardiac death, myocardial infarction (MI, classified as Q-wave and Non-Q wave) and ischaemia-driven target lesion revascularisation (ID-TLR) by coronary artery bypass graft or percutaneous coronary intervention. Angiographic restenosis was defined as a diameter stenosis greater than or equal to 50% at the treated site at follow-up angiography. Cardiac death was defined as any death in which a cardiac cause could not be excluded. A diagnosis of Q-wave MI was made with evidence of new pathological Q waves on electrocardiogram (ECG). A diagnosis of non-Q-wave MI was made with an elevation of the creatinine kinase (CK) greater than or equal to twice the upper limit of normal, with an elevated CK-MB and the absence of new pathological Q waves. ID-TLR was defined by either of the three following definitions: (I) revascularisation of the target lesion associated with a positive functional ischaemia study, (II) ischaemia symptoms and angiographic lumen diameter stenosis greater than or equal to 50% by core laboratory assessed quantitative coronary angiography (QCA), (III) revascularisation of a target lesion with diameter stenosis greater than or equal to 70% by core laboratory assessed QCA, without either ischaemic symptoms or a positive functional study.
The ARC (Academic Research Consortium) definitions15 for stent thrombosis were utilised for this study. The CEC (clinical events committee) reviewed and adjudicated all cases of safety endpoint events.
Categorical variables are reported as counts and percentages, continuous variables as mean±SD; p values were calculated with Fisher's Exact test for binary variables, Wilcoxon's Rank Sum test for continuous variables and Wilcoxon Signed Rank test for paired comparison between the different time points. A p value<0.05 was considered statistically significant. Statistical analyses were performed using SAS V.9.1.3.
Pre-procedural mean RVD was significantly smaller in vessels with RVD<2.5 mm; percentage diameter stenosis was similar between large and small vessels (table 1).
Post-procedural MLD was significantly greater in large vessels; however, the mean acute gain was similar between groups. After adjustment for vessel size, the relative gain was found to be significantly greater in the small-vessel group (table 1).
At 1- and 2-year follow-up no differences were reported in in-scaffold late loss, %DS and binary restenosis. Also after adjustment for vessel size, the relative loss, the absolute net gain, the net gain index and the loss index were similar between the small- and large-vessel groups (table 2).
IVUS and IVUS-VH analysis
In the large vessel group, paired per lesion IVUS grey scale analysis demonstrated a constant vessel area growth at each time point that became significant at 2-year follow-up associated with a similar plaque area enlargement. Scaffold area also showed a significant enlargement between 6 months and 2 years. Lumen area numerically increased in ABSORB cohort B 1 between 6-month and 2-year follow-up, but without reaching any statistical significance, with a borderline significant decrease in ABSORB Cohort B 2 between baseline and 1-year follow-up (table 3).
In the small vessel group, vessel area was observed to grow from baseline to 6-month and 2-year follow-up and was associated with a parallel growth in plaque area over time (table 3). The scaffold area at 6-month follow-up did not show any enlargement but was observed to grow significantly from 6-month to 2-year follow-up. The lumen area followed a similar evolution over time with a significant late lumen enlargement from 6-month to 2-year follow-up. (table 3).
Comparison of small and large vessels at 1- and 2-year follow-up showed similar minimal lumen area, percentage lumen area stenosis and neointimal hyperplasia area between groups at each time point. (table 3).
Paired per-lesion IVUS-VH analyses demonstrated a consistent reduction in dense calcium content over time that reached the significant level in the cohort B2 at 1-year follow-up (large vessels 28.50±9.05% vs 24.13±7.84%, p=0.0041; small vessels 32.96±11.65% vs 24.77±7.75%, p=0.0047) but was not significant at 2-year in cohort B1. Fibrous and fibro-fatty tissue significantly increased at 1-year follow-up (Fibrous tissue: large vessels 36.14±10.89% vs 41.52±9.75% p=0.0148; small vessels 31.28±10.01% vs 39.98±9.25%, p=0.0104. Fibro-fatty large vessels 2.90±2.40% vs 4.35±2.20% p=0.0035; small vessels 2.50±1.90% vs 4.56±2.79%, p=0.0120). No changes in percentage necrotic core were observed at either time point. No further differences in plaque compositional characteristics were detected between small and large vessels at 1-and 2-year follow-up. (figures 1 and 2, online supplementary table).
OCT data at baseline were available for the present analysis in 50 patients, 20 patients in the small vessel group and 30 in the large vessel group. Post-procedurally the incomplete stent apposition (ISA) area was similar in both groups. Mean scaffold area, minimal scaffold area, mean lumen area and minimal lumen area were significantly larger in the large vessel group and this was observed also at 2-year follow-up. No differences in percentage uncovered struts, percentage lumen area stenosis and mean neointimal hyperplasia (NH) area were observed over time. (table 4)
At 1-year follow-up, comparisons of clinical outcomes between small and large vessels, demonstrated no differences in MACE (small-vessel group: 3/41 cases, 7.3% vs large-vessel group 4/60 cases, 6.7%, p=1.0000), MI (small-vessel group: 2/41 cases, 4.9% vs large-vessel group 1/60 cases, 1.7%, p=0.5645) or ischaemia driven target lesion revascularisation (TLR) (small-vessel group: 1/41 cases, 2.4% vs large-vessel group 3/60, 5.0%, p=0.6445). (table 5, online supplementary figure S1).
At 2-year follow-up only 2 further events were reported, namely 2 ischaemia-driven TLR, both in the large vessels group. (see online supplementary figure S2). No deaths were reported in either group up to 2 years. No episodes of definite, probable, or possible scaffold thrombosis were observed up to 2 years (table 5).
The present study investigated the long-term clinical and angiographic outcomes following the implantation of the second generation Absorb everolimus eluting bioresorbable vascular scaffold in small and large vessels. The major findings are: (I) similar angiographic results were observed up to 2-year follow-up between the two groups; (II) IVUS grey scale analyses in the small vessel group demonstrated expansive vessel remodelling associated with a significant late lumen enlargement; (III) Both the small- and large- vessel groups, treated with a 3.0 mm Absorb BVS, showed a similar and low rate of cardiovascular events.
In the ABSORB Cohort B Trial, vessels with a RVD<or ≥2.5 mm were both treated with a 3.0 mm device. Both groups demonstrated a similar pre-procedural MLD and post-procedural in-scaffold acute gain. Following adjustment for vessel size, it was demonstrated that the relative gain was significantly greater in the small-vessel group compared with the large-vessel group.
In previous studies, a larger relative gain was observed to be an independent predictor of restenosis, with a direct relationship between relative gain and the subsequent relative loss at follow-up.11 ,16
Schwartz et al17 previously demonstrated that neointimal growth is proportional to the degree of vessel injury during the index procedure, and more specifically to the disruption of the internal elastic membrane. Consequently if we consider the relative gain as a surrogate measurement of the vessel injury induced by treatment, a greater luminal re-narrowing in vessels with a greater luminal relative gain is not surprising.18
Theoretically a similar relative gain and injury in small and large vessels would produce a similar quantitative neointimal response. In this situation small vessels would be less able to accommodate the same absolute amount of neointimal tissue compared with large vessels.19 ,20
These considerations suggest that optimal acute and long-term outcomes after coronary interventions may result from the appropriate balance between luminal gain and vascular injury.
In the present study the implantation of a 3.0 mm Absorb BVS in vessels with RVD<2.5 mm produced a significantly larger angiographic relative gain in small vessels compared with large vessels; however, at 1-and 2-year follow-up no differences in relative loss and percentage diameter stenosis were evident. Furthermore the net gain index (describing the outcome of intervention relating the pre MLD to the MLD at follow-up),11 and the loss index (relating late loss to acute gain)12 were also similar between the small- and large-vessel groups.
These data suggest that the slightly higher relative gain in the present series of small vessels, was associated with a favourable balance between vessel injury and luminal expansion, followed by comparable long-term clinical and angiographic outcomes between groups.
IVUS grey scale analyses
IVUS geometrical analyses demonstrated that at 1- and 2-year follow-up, minimal lumen area, percentage lumen area stenosis and neointimal hyperplasia area were similar between small and large vessels. Specifically, the neointimal hyperplasia area in both the small and large vessel groups and at both time points was limited and at the resolution limit of the imaging technique,21 thus making even a statistically significant difference poorly relevant. IVUS analyses demonstrated also a consistent growth in vessel area over time irrespectively of vessel size; this observation was associated with an increase in plaque area but also with an increase in scaffold area, that was observed to occur in both large and small vessel at 2-year follow-up. These findings are consistent with the planned scaffold bioresorption process. During the first months after implantation the Absorb BVS plays a role similar to conventional metallic DES, providing essentially a vessel mechanical scaffolding and drug elution. Between 6 months and 1 year the gradual polymer degradation leads to a progressive loss in scaffold structural integrity, and as recently demonstrated, to the return of the vasomotion properties of the treated segment at 12-month follow-up.8
Furthermore, whilst from baseline to 6-months lumen area slightly decreased as would be expected after conventional metallic stent implantation, at 2-year follow-up a late lumen enlargement was observed, that reached statistical significance in the small vessel group.
A largely accepted theory for the higher rates of restenosis and clinical events in small vessels is that a smaller lumen area is less able to accommodate the same amount of neointimal tissue compared with a larger lumen area.19 Therefore, in this scenario a late lumen enlargement may play a key role in the prevention of restenosis in the small vessels lesions subset.
These data also suggest that the potential concern related to the implantation of the present thick strut BVS (strut thickness 157 μm) in vessels with small lumen area may be unfounded.
The dense calcium content has been previously used as a surrogate marker of polymer bioresorption22–24 and in the present study was observed to decrease in parallel to a mild increase in fibro-fatty and fibrous tissue over time in both groups reaching the statistically significant level at 1-year follow-up in the Cohort B2 population. Compositional analyses also disclosed no significant changes in necrotic core content in both groups over time, consistently with previous histological studies reporting no relevant inflammatory response associated with Absorb BVS implantation.5 (figures 1 and 2, online supplementary table).
The OCT analysis showed a larger mean scaffold area, minimal scaffold area, mean lumen area and minimal lumen area, in large vessels compared with small vessels with similar post-procedural percentage lumen area stenosis, demonstrating a larger stent deployment and vessels size in the large vessel group.
Notably at baseline the mean incomplete stent apposition (ISA) area was similar between the two groups reducing the possibility that the increased positive vessel remodelling in the small vessel group was due to an artificial improved stent apposition and drug elution in this population.
At 2-year follow-up similar neointimal growth was observed between small and large vessels and in only in one patient for each group was detectable incomplete stent apposition.
A small vessel diameter is a well-recognised independent predictor of a higher rate of in-stent restenosis and poorer clinical outcomes25 ,26 with bare metal stents and first and second generation drug eluting stents.27–29
In the present study, the long-term clinical outcomes after implantation of the second generation Absorb BVS in small vessels were similar to that observed in large vessels, with low and comparable rates of MI and ischaemia-driven TLR. Notably, no death or in-scaffold thrombosis events were reported at 2-year follow-up.
The two clinical events reported between 6 and 12 months and the further 2 between 1 and 2 years all occurred in the large vessel group. (see online supplementary figure S1 and S2) Furthermore the three events that occurred in the small vessel group up to 6 months were likely to have been related to baseline procedural complications, rather than device failure per se.6
However, it should be highlighted that due to the limited patients population and low incidence rate in this study, caution should be made in making conclusions with regards to the safety.
In conclusion, treatment of small vessels with second generation Absorb BVS was observed in the present report to be associated with similar angiographic, IVUS and long-term clinical outcomes compared with large vessels. The expected loss of structural integrity of the Absorb BVS device, and the return of the vasomotor properties of the treated vessel, unhindered by a metallic cage, translated into further beneficial positive remodelling and late lumen enlargement, which was particularly relevant in the small vessel group.
The present study is a post-hoc analysis of the ABSORB Cohort B Trial. Due to the limited patient population, the p values were calculated for descriptive purposes and should be considered as exploratory and hypothesis generating. Further investigations are needed to fully investigate the impact of vessel size on clinical and angiographic outcomes after Absorb BVS implantation.
In Absorb Cohort B the bioresorbable vascular scaffold was implanted in discrete lesions located in native coronary arties excluding restenotic lesions, lesions located in the left main coronary artery, lesions involving a side branch more than 2 mm in diameter, and lesions with presence of thrombus or another clinically significant stenosis in the target vessel. We therefore cannot exclude that the implantation of BVS in different and more complex lesions could have been associated with different results.
During the ABSORB Cohort B study only the 3 mm BVS was available. A novel BVS 2.5 mm is currently under clinical evaluation in two separate studies (the multicenter ABSORB Extend Single-Arm Study and the ABSORB II Randomised Controlled Trial) possibly aiding in further understanding on the performance of this device in the small vessels scenario. Quantitative coronary angiography that in the present paper was used to define vessel size is characterised by the detection of the angiographic lumen contour without any evaluation of the actual vessel wall size.
In the present study the implantation of the 3.0 mm ABSORB everolimus eluting bioresorbable vascular scaffold in coronary arteries with RVD<2.5 mm appears safe, with angiographic and clinical outcomes similar to those in large vessels at 1- and 2-year follow-up. A favourable balance between acute luminal gain and vascular injury and the observed significant late lumen enlargement may be the possible explanation for the encouraging long-term results in small vessels.
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online appendix
XL, KM-H and SV are employed by Abbott Vascular
Funding The present study was funded by Abbott Vascular.
XL, KM-H and SV, that are coauthors of the present study, are employed by Abbott Vascular. Detailed disclosures for all authors are reported at the end of the manuscript.
All authors made substantial contributions to the conception and design, acquisition of data or analysis of data; or drafted the article or revised it critically for important intellectual content or gave final approval of the version to be published.
Contributors RD and PWS interpreted the data and drafted the manuscript. XL performed the statistical analysis. All the remaining authors revised the manuscript critically for important intellectual content and gave final approval of the version to be published.
Competing interests Obtained.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.