Author + information
- Received October 14, 2013
- Revision received January 23, 2014
- Accepted January 30, 2014
- Published online July 1, 2014.
- Alessio Mattesini, MD∗,†,
- Gioel G. Secco, MD∗,‡,
- Gianni Dall'Ara, MD∗,
- Matteo Ghione, MD∗,
- Juan C. Rama-Merchan, MD∗,
- Alessandro Lupi, MD‡,
- Nicola Viceconte, MD∗,
- Alistair C. Lindsay, MD, PhD∗,
- Ranil De Silva, MD, PhD∗,
- Nicolas Foin, PhD§,
- Toru Naganuma, MD‖,
- Serafina Valente, MD†,
- Antonio Colombo, MD, PhD‖ and
- Carlo Di Mario, MD, PhD∗∗ ()
- ∗NIHR Biomedical Research Unit, Royal Brompton and Harefield NHS Foundation Trust, London, United Kingdom
- †Department of Heart and Vessels, AOUC Careggi, Florence, Italy
- ‡Department of Clinical and Experimental Medicine, University of Eastern Piedmont, Novara, Italy
- §National Heart Centre Singapore, Singapore
- ‖Interventional Cardiology Unit, EMO-GVM Centro Cuore Columbus, Milan, Italy
- ↵∗Reprint requests and correspondence:
Dr. Carlo Di Mario, Royal Brompton Hospital, Sydney Street, London SW3 6NP, United Kingdom.
Objectives The aim of this study was to compare the acute performance of the PLLA ABSORB bioresorbable vascular scaffold (BVS) (Abbott Vascular, Santa Clara, California) with second-generation metal drug-eluting stents (DES) in complex coronary artery lesions.
Background Thick polymer-based BVS have different mechanical properties than thin second-generation DES. Data on the acute performance of BVS are limited to simple coronary lesions treated in trials with strict inclusion criteria.
Methods Fifty complex coronary lesions (all type American College of Cardiology/American Heart Association B2-C) treated with a BVS undergoing a final optical coherence tomography (OCT) examination were compared with an equal number of matched lesions treated with second-generation DES. The following stent performance indexes were assessed with OCT: mean and minimal area, residual area stenosis (RAS), incomplete strut apposition (ISA), tissue prolapse, eccentricity index, symmetry index, strut fracture, and edge dissection.
Results One hundred lesions from 73 patients were analyzed. A higher balloon diameter/reference vessel diameter ratio was used for predilation in the BVS group (p < 0.01). Most of the BVS and DES were post-dilated with short noncompliant (NC) balloons of similar diameter. OCT showed in the BVS group a higher tissue prolapse area (p = 0.08) and greater incidence of ISA at the proximal edge (p = 0.04) with no difference in the overall ISA. The RAS was 20.2% in the BVS group and 21.7% in the DES group (p = 0.32). There was no difference in the eccentricity index. The minimal and mean lumen areas were similar in the 2 groups. Two cases of strut fractures occurred after the BVS, whereas none was observed in the DES.
Conclusions Based on OCT, the BVS showed similar post-procedure area stenosis, minimal lumen area, and eccentricity index as second-generation DES. The different approach for lesion preparation and routine use of OCT guidance during BVS expansion may have contributed to these results.
In daily clinical practice interventional cardiologists face complex coronary lesions that are often calcified and long and involve bifurcations. Optimal mechanical stent performance is crucial in the treatment of these lesions, and the alloy-based thin-strut second-generation drug-eluting stents (DES) are considered the gold standard for the treatment of these complex coronary artery lesions (1,2). The conformability and radial strength of second-generation DES allow optimal deployment in long tapered lesions and bifurcations because of the possibility to post-dilate the stent and improve expansion and apposition (3). Furthermore, the high radial strength of modern DES is crucial to counteract the acute plaque recoil frequently observed in fibrocalcific lesions.
Bioabsorbable drug-eluting scaffolds (BVS) (ABSORB, Abbott Vascular, Santa Clara, California) have emerged as a potential major breakthrough for treatment of coronary artery lesions. In principle, the need for vessel scaffolding and drug delivery is temporary, rendering a permanent stent superfluous once the vessel has healed and the processes of recoil and hyperplasia have ended. Conventional permanent stent implantation precludes future surgical revascularization, complicates recrossing into side branches, eliminates reactive vasomotion, impairs noninvasive imaging, and exposes patients to the risk of very late thrombosis. These long-term limitations of conventional stents may be overcome to a degree by using biodegradable scaffolds (4). However, the mechanical properties of polymer-based scaffolds substantially differ from those of metal stents, and thus far, available data on the acute mechanical performance of bioresorbable vascular scaffolds (BVS) are limited to the treatment of relatively simple coronary lesions in the context of early-stage clinical trials (5,6).
The aim of this study was to compare the acute performance of BVS with that of second-generation DES in the treatment of complex coronary artery lesions using optical coherence tomography (OCT) to assess appropriate stent deployment (7).
The study population comprised consecutive patients undergoing percutaneous coronary intervention (PCI) of complex coronary lesions with stent optimization under OCT guidance, which is our routine for complex lesion stenting (8). From September 2012 until May 2013, patients treated with a BVS at the Royal Brompton Hospital (London, United Kingdom) and Columbus Hospital (Milan, Italy) were prospectively enrolled. Of 148 patients with complex lesions treated with second-generation DES at the Royal Brompton Hospital between January 2009 and May 2013 and optimized using post-deployment OCT examination (DES group), we selected an equal number of lesions with angiographic characteristics matched to those in the BVS group. The 1:1 selection without replacement has been performed according to the following stepwise selection criteria: lesion length, vessel reference diameter, ostial position, bifurcation involvement, severe or moderate calcifications, and chronic total occlusion.
All patients provided signed informed consent for stent deployment and OCT guidance. The devices used in the DES group were the everolimus-eluting Xience Pro and Prime (Abbott Vascular), Promus Element and Premiere stents (Boston Scientific, Natick, Massachusetts), and the zotarolimus-eluting Resolute Integrity stent (Medtronic Vascular, Santa Rosa, California). BVS were not used in patients presenting with acute ST-segment elevation myocardial infarction, coronary bifurcations with a default 2-stent strategy, a target lesion in a vessel with a reference diameter <2.5 mm, and, because of the impossibility of performing serial OCT examinations, an estimated glomerular filtration rate <30 ml/min, or aorto-ostial lesions. The main inclusion criteria in the BVS group used to define the lesion complexity were length >24 mm, moderate to heavy calcification, ostial position (different from aorto-ostial RCA and LM ostium which were excluded), bifurcation involvement, and chronic total occlusion.
Quantitative coronary angiography analysis and lesion characterization
Quantitative coronary angiography (QCA) was performed using a computer-based QCA system (CAAS QCA-2D system, Pie Medical Imaging BV, Maastricht, the Netherlands) with the dye-filled catheter used for calibration. For each lesion, the following QCA parameters were measured: minimal lumen diameter, reference vessel diameter (RVD), percentage of area stenosis, and lesion obstruction length. The largest balloon diameter and maximal inflation pressure during lesion predilation were recorded and used to calculate the balloon/artery ratio (mean inflated balloon diameter/mean reference vessel diameter). In addition, we assessed the presence of angiographic calcification.
In both groups, lesions were treated with pre-dilation using conventional semicompliant or NC balloons. The use of additional devices (cutting balloons or rotablator) was left to the operator's discretion. Unlike for DES, deployment of the BVS was performed using slow balloon inflation (i.e., 2 atm per 10 s) without exceeding the rated pressure indicated in the product instructions for use. Post-dilation with short NC balloons was systematically performed both for the BVS and DES, using OPN NC balloons (SIS Medical AG, Winterthur, Switzerland) when pressures >30 atm were required (9). Attention was paid to avoid reaching a maximal balloon diameter beyond the recommended rupture point of the BVS by strictly following the NC balloon compliance chart. In case of lesions involving a bifurcation, final optimization with sequential dilation was preferentially adopted for BVS. Conversely, for DES final kissing balloon was the default strategy. OCT assessment was performed in most cases before stent deployment and repeated when stent expansion was considered optimal angiographically. In the event of suboptimal deployment as assessed with OCT, further post-dilation was performed or additional BVS/DES were implanted, after which a final OCT scan was performed and used for the study analysis.
Frequency domain OCT was performed using the C7 system or the Ilumien Optis system (St. Jude Medical, Minneapolis, Minnesota). For both systems, DragonFly or DragonFly 2 imaging catheters (St. Jude Medical) were used. Automatic pullbacks were performed at 20 mm/s during contrast injection at a rate of 3 to 5 ml/s using a power injector. The OCT catheter was inserted distal to the treated segment, and the pullback continued until either the guiding catheter was reached or the maximal pullback length (5.5 cm with the C7 system and 7.4 cm with the Ilumien Optis system) was completed. Two sequential pullbacks were combined to enable assessment of the entire stented segment when required.
OCT offline analysis
The OCT measurements were repeated offline using the LightLab Imaging workstation (St. Jude Medical). The analysis of contiguous cross sections was performed at 1-mm intervals within the entire stented segment and at 5 mm proximal and distal to the stent to measure the proximal and distal reference vessel area (RVA) and to identify dissections. RVA was calculated as the mean of the 2 largest luminal areas 5 mm proximal and distal to the DES/BVS edge (7). In case of the absence of a meaningful proximal or distal segment due to the ostial location of the lesion or the presence of a large side branch at the stent edge, only a proximal or distal reference cross section was used to calculate the RVA (10). Stent edge dissection was defined as a disruption of the vessel luminal surface at the stent edge with visible flap. Stent fracture was suspected in the presence of isolated struts lying unapposed in the lumen with no connection or overridden by the contiguous stent struts. For each cross section analyzed, the area, mean, and maximal and minimal diameter of the stent were automatically contoured and measured by the analysis system, with manual correction as appropriate (4).
For analysis of the BVS, which are transparent to the near-infrared light of the OCT catheter, incomplete strut apposition (ISA) was defined as presence of struts separated from the underlying vessel wall (7). For metal DES, inducing posterior dropout, struts were considered malapposed when the axial distance between the strut's surface and the luminal surface was greater than the strut thickness (11). Tissue prolapse was defined as the presence of tissue protruding between stent struts extending into the lumen as a circular arc connecting adjacent struts (4).
The following quantitative parameters were calculated for each stent (5,7,11,12): the percentage of the ISA, calculated as a ratio of the total number of struts observed at 1-mm intervals; the percentage of stents with ISA at the proximal and distal edges defined as the last 5 mm of the stent before the stent end; the ISA area (mm2) (only for BVS) measured as illustrated in Figure 1; the tissue prolapse area (mm2) calculated as the difference between the stent area and the lumen area as illustrated in Figure 1; the percentage of RAS calculated as: [1 − (minimal lumen area/RVA) × 100], as illustrated in Figure 2; the eccentricity index, the ratio between the minimal and the maximal diameter. For each stent both the mean and minimal eccentricity index were computed (illustrated in Fig. 2); the symmetry index, defined as: (maximal stent diameter − minimal stent diameter)/(maximal stent diameter).
Clinical follow-up was obtained at ∼1 month after the procedure and every 6 months after by direct clinical examination.
Descriptive statistics (means and SDs for continuous variables with normal distribution and frequency and relative frequency for categorical variables) were computed according to treatment type (BVS or DES). Comparison between groups for continuous variables was performed by an unpaired t test (in case of parametric distribution) or the Mann-Whitney U test (in case of nonparametric distribution), as appropriate. Univariate associations between treatment type and coronary lesion features were examined using 2-way contingency tables. Significance of associations were assessed using the chi-square test or the Fisher exact test, as appropriate. For all the statistical tests used, a p value <0.05 was required to reject the null hypothesis. The statistical analysis was performed using the SPSS statistical software package version 16.0 (IBM Corporation, Somers, New York).
Fifty lesions treated with 63 BVS in 35 patients were matched with 50 lesions treated with 61 second-generation DES in 38 patients. Baseline clinical characteristics of the patients are shown in Table 1. There were no significant differences in the 2 groups, with a minority of patients (4.1%) presenting with unstable angina as an indication for the PCI procedure.
Angiographic and QCA baseline lesion characteristics are summarized in Table 2. The left anterior descending artery (LAD) was the target vessel in a large proportion of cases in both groups (BVS, n = 34, 68%; DES, n = 25, 50%; p = 0.11). As expected, based on the inclusion criteria, all lesions met the American College of Cardiology/American Heart Association classification criteria for B2 or C lesions. There were no significant differences in the presence of calcification, ostial involvement, and bifurcation involvement. Reference vessel diameter, minimal lumen diameter, and lesion length, as assessed with QCA, were also similar (lesion length: BVS, 24.7 ± 14.2 mm; DES, 25.1 ± 10.6 mm; p = 0.86). Two chronic total occlusions were successfully treated in the BVS group and 4 in the DES group.
Sixty-three BVS and 61 DES were implanted with a similar number of stents per lesion in the 2 groups (BVS, 1.3 ± 0.6; DES, 1.2 ± 0.5; p = 0.28).
Xience Prime was the most frequently used DES (n = 35, 57.4%), whereas Promus Element or Premiere and Resolute Integrity stents were used in 16 (26.2%) and 10 (16.4%) of cases, respectively. The median stent length was 28.0 mm (interquartile range: 20.5 to 28.0 mm) in the BVS group and 28.0 mm (interquartile range: 20.0 to 38.0 mm) in the DES group (p = 0.42). As shown in Table 3, a higher balloon diameter/mean reference vessel diameter ratio was used for predilation in the BVS group (BVS, 1.1 ± 0.1; DES, 0.9 ± 0.1; p < 0.01), with significantly higher pressure inflation for both pre- and post-dilation. NC balloons were more frequently used for lesion preparation in the BVS group. Sequential dilation was the only technique used for bifurcation optimization in the BVS group, whereas kissing balloons were consistently used in the DES group (13).
OCT findings are summarized in Table 4. A total of 2,953 cross sections and 24,352 struts were analyzed. The mean and minimal lumen area were similar in the 2 groups. The incidence of RAS >20% was not statistically significant different in the BVS group (BVS: n = 25, 39.7%; DES: n = 26, 42.6%; p = 0.85), and there was no difference in the mean RAS (BVS, 20.2 ± 7.5%; DES, 21.7 ± 9.9%; p = 0.32). There was a higher incidence of ISA at the proximal edge in the BVS group (BVS: n = 25, 39.7%; DES: n = 14, 23.0%; p = 0.04) but no difference in the overall percentage of ISA (BVS, 1.7 ± 2.1%; DES, 1.9 ± 2.4%; p = 0.62) and number of stents with ISA (BVS: n = 33, 52.4%; DES: n = 39, 63.9%; p = 0.19) (Fig. 3).
The mean and minimal eccentricity index and the symmetry index were similar in the 2 groups.
In the BVS group, there was a trend toward a higher prolapse area (BVS, 1.5 ± 2.4 mm2; DES, 0.8 ± 1.2 mm2; p = 0.08), but this did not have a significant impact on the final lumen area, which was similar in both groups. OCT analysis showed 12 edge dissections (BVS: n = 5, 7.9%; DES: n = 7, 11.5%; p = 0.55), which were not apparent on the angiogram. None of these required further stent implantation. In the DES group, strut fractures were not observed, whereas in 2 patients in the BVS group, 2 stent fractures developed. In both cases, the lesions were localized in the LAD across the origin of a diagonal branch, and the scaffolds were recrossed to optimize the results with sequential dilation.
Clinical follow-up data were available for all BVS patients and for 31 patients (89.5%) in the DES group. The mean duration of follow-up was significantly different in the 2 groups (BVS, 8.5 ± 2.8 months; DES, 17.3 ± 8.7 months; p < 0.01). One patient treated with 2 BVS in the proximal, mid, and distal LAD presented 2 months later with acute coronary syndrome due to a critical lesion in a diagonal branch not treated with stenting during the index PCI, although the 3 previously implanted BVS were patent. One insulin-dependent diabetic patient with a proximal LAD BVS required a coronary artery bypass graft after 9 months because of diffuse distal disease progression (no in-stent restenosis). Further details provided in the Online Appendix. In the DES group, 2 patients (5.3%) underwent PCI due to in-stent restenosis at 14 and 24 months, respectively, after the baseline PCI.
This is the first OCT study to compare acute stent performance between a BVS and second-generation DES in complex coronary artery lesions.
The OCT indexes used in our study are widely accepted as criteria for determining optimal stent deployment (7,12). These were derived from intravascular ultrasound (IVUS) criteria used for the evaluation of metal stents and shown to correlate with 1-year clinical outcomes (stent thrombosis and restenosis) after implantation of a BMS and first-generation DES (14,15). In particular, a RAS >20% and an absolute minimal cross-sectional area <5.5 or 6.0 mm2 were previously correlated with acute/subacute stent thrombosis and restenosis (16,17). In our study, the mean RAS and absolute MLA in both the BVS and second-generation DES were close to these cutoff levels. The thresholds defined in these early IVUS evaluations of DES were derived from analyses of trials including short-type A-B1 lesions. Results for more complex lesions are limited and expected to be worse because of the higher plaque burden and resistance. In a previous study by our group, calcified lesions with OCT were analyzed, and a higher RAS and a greater ISA than in simple lesions were observed (18). The use of OCT rather than IVUS may explain part of the difference. By virtue of its higher resolution, OCT can define more precisely the lumen area contours and quantify plaque prolapse, which is frequently concealed by strut artifacts when using IVUS. Moreover, OCT has been shown to measure lower absolute areas than IVUS, both in vitro and in vivo (19). The areas observed in the complex lesions of our BVS group were very similar to those reported in an OCT substudy of the ABSORB trial cohort B (10) in which relatively simple lesions were treated. Also the type of metal stents used may play a role. The thick stainless steel struts of first-generation DES may create a rough cobblestone surface with a higher risk of thrombosis but have less recoil than the thin struts of second-generation stents constructed using alloys such as cobalt and platinum chromium (3,20).
The lack of difference between the BVS and second-generation DES in mean and minimal values of relative and absolute stent area is the true novel finding of the current study. This could be expected based on in vitro studies conducted by the industry for registration and on small comparative studies with IVUS between the BVS and Xience V stents (21,22). The observations made using OCT in the current study support the application of BVS beyond the current indications. These results suggest that a satisfactory BVS expansion can also be achieved in complex coronary lesions, at least when appropriate lesion preparation and deployment under OCT guidance is performed.
The clinical relevance of ISA is controversial. Previous IVUS studies using first-generation DES to treat simple lesions (23,24) showed no value in predicting late adverse events. However, an IVUS study performed at the time of acute stent thrombosis showed a higher incidence of ISA compared with controls (25). In our study, no difference was observed in the absolute number and percentage of malapposed struts between the BVS and DES, with the latter being considerably lower than shown in previous OCT studies of predominantly first-generation DES (26). Repeated OCT examinations in both groups, prompted additional dilations at high pressure with properly sized balloons, probably explain the low prevalence of ISA in the complex lesions treated in this study. Some malapposition was still observed in pre- or post-stenotic ectasic segments, in eccentric calcific lesions, and at bifurcations, although in many cases, this was impossible to correct completely despite serial or kissing balloon dilation (27,28). In the BVS group, we observed a more frequent malapposition at the proximal stent edge (39.7% of BVS vs. 23.0% of DES, p = 0.04), which may have procedural relevance because struts protruding into the lumen complicate the advancement of balloons or additional distal stents. The greater conformability of second-generation DES compared to the BVS may explain these findings because extreme attention was paid to post-dilation with appropriately sized balloons that covered the proximal edge of the BVS, which is angiographically well visualized by the radiopaque platinum proximal marker. A potential advantage of the bioresorbable technology over the metal device is that any acute ISA resolves after the process of strut degradation has been completed, although this process may require as long as 2 years (29).
Gomez-Lara et al. (10) performed an OCT substudy of the ABSORB trial Cohort B, in which only 3-mm diameter BVS devices were deployed. These investigators found a higher incidence of malapposed struts in vessels with a maximal diameter >3.3 mm. These data emphasize the importance of correct vessel size measurement to select the appropriate BVS diameter. Based on this observation, we recommend oversizing the BVS as much as 0.5 mm above the smallest reference vessel diameter, which allows better adaptation, especially in tapered vessels.
The amount of tissue prolapse was slightly higher with the BVS than the DES. This may be explained by differences in stent design and the lower number of struts per cross section in BVS compared with second-generation DES. In our population of stable lesions, the greater plaque prolapse was not clinically relevant because it did not alter the final mean and minimal lumen area. Moreover, the thicker struts of the BVS mean that the prolapsed plaque is always surrounded by the stent struts with less herniation into the vessel lumen and a smoother surface compared with thinner strut DES (Fig. 1). However, tissue prolapse is more prominent in unstable or thrombus-containing lesions (30), and in this setting, the scaffolding properties of the BVS might be insufficient to counteract plaque prolapse.
A very similar post-procedural stent geometry was observed in both BVS and DES. The mean and minimal values of the eccentricity index were similar in the 2 groups. Brugaletta et al. (22) reported a higher symmetry index value for the BVS, whereas the eccentricity index was significantly lower in the BVS compared with the DES (0.85 ± 0.08 vs. 0.90 ± 0.06, p < 0.01). In the current study, the minimal eccentricity index was lower for both the BVS and DES compared with the findings of Brugaletta et al. (22), likely a consequence of including more complex lesions.
The acute mechanical performance of the BVS observed in this study cannot be taken for granted unless implantation is performed with the same meticulous attention to lesion preparation, systematic sizing of the BVS to the proximal reference vessel diameter, and high pressure post-dilation. Scaffold implantation was performed under OCT guidance. Therefore, our results cannot be automatically applicable to conventional angiographic BVS deployment. The main limitation of this study is the use of a historical nonrandomized control group with a limited sample size and follow-up duration, which precludes any meaningful long-term clinical comparison of the rare long-term adverse events observed in patients treated with modern DES. A matched cohort as a control population is frequently used in the evaluation of novel devices and is certainly of value for the assessment of the mechanistic response. There was no significant difference observed in key parameters of acute stent performance in the 2 subgroups treated with a BVS or DES. However, adjustments for multiple correlated observations were not made, and definitive proof of noninferiority would require a prospective, randomized study with a larger study population. Prospective, randomized, controlled trials are required to determine whether the acute mechanical performance of a BVS observed in the current study can be translated into an improved long-term clinical outcome in patients with complex coronary lesions treated by PCI. The short follow-up duration provided in the BVS group may be inappropriate to observe disease progression and late target lesion failure. Finally, although no new Q waves or ST-segment elevation or prolonged chest pain were observed, we did not consistently measure post-procedural troponin in all patients.
We report our early experience of treating complex coronary lesions using a BVS. Systematic aggressive lesion preparation, sizing of a BVS to the proximal reference vessel diameter, high pressure post-dilation, and use of OCT for optimization enable us to achieve post-procedural area stenosis, minimal lumen area, and an eccentricity index similar to that of contemporary DES platforms.
The authors acknowledge Eng Cinzia Lazzara for support in the preparation of the illustrations of this paper.
For supplemental figures, please see the online version of this article.
The Royal Brompton Hospital Research Office is the recipient of a grant for the study EXCEL from Abbott Vascular. The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- bioresorbable vascular scaffold(s)
- drug-eluting stent(s)
- incomplete strut apposition
- intravascular ultrasound
- left anterior descending artery
- optical coherence tomography
- percutaneous coronary intervention
- quantitative coronary angiography
- residual area stenosis
- reference vessel area
- Received October 14, 2013.
- Revision received January 23, 2014.
- Accepted January 30, 2014.
- American College of Cardiology Foundation
- Serruys P.W.,
- Onuma Y.,
- Ormiston J.A.,
- et al.
- Tearney G.J.,
- Regar E.,
- Akasaka T.,
- et al.
- Gomez-Lara J.,
- Diletti R.,
- Brugaletta S.,
- et al.
- Prati F.,
- Guagliumi G.,
- Mintz G.S.,
- et al.
- Foin N.,
- Torii R.,
- Mortier P.,
- et al.
- Sonoda S.,
- Morino Y.,
- Ako J.,
- et al.
- Fujii K.,
- Carlier S.G.,
- Mintz G.S.,
- et al.
- Doi H.,
- Maehara A.,
- Mintz G.S.,
- et al.
- Hong M.K.,
- Mintz G.S.,
- Lee C.W.,
- et al.
- Hoffmann R.,
- Morice M.C.,
- Moses J.W.,
- et al.
- Cook S.,
- Wenaweser P.,
- Togni M.,
- et al.
- Onuma Y.,
- Serruys P.W.,
- Perkins L.E.,
- et al.