Introduction

Percutaneous treatment of atherosclerotic lesions with use of metallic stents leads to a stretching of coronary plaques by baro-trauma [1]. The implantation of a foreign body, such as a metallic stent, is also known to induce an inflammatory response adjacent to the struts [2–4]. However, serial analyses by intravascular ultrasound virtual histology (IVUS-VH) have demonstrated no changes at 10 months in the VH components of coronary plaques after bare metal or drug eluting stent implantation [5, 6].

The new everolimus-eluting bioresorbable vascular scaffold (ABSORB, Abbott Vascular, Santa Clara, CA) has been developed with the intention to provide temporary lumen scaffolding and, in contrast to metallic platform stents, to allow late lumen enlargement and restoration of normal vasomotion post bioresorption [7]. In the ABSORB Cohort A trial a non significant reduction in the VH-derived necrotic core (NC) was demonstrated between 6 month and 2 year follow-up [7, 8]. However, as the polymeric struts are mistakenly recognized by IVUS-VH as dense calcium and necrotic core and are progressively bioresorbed, a concomitant result of scaffold bioresorption and plaque modification has to be taken into account for the correct interpretation of VH changes over time [9, 10].

Conversely, in the ABSORB Cohort B trial, a customized software, allowing for the exclusion of the dense calcium and necrotic core due to the temporary presence of the scaffold from the quantification of the VH-plaque behind the polymeric struts, was used. At 6-month follow-up a slight increase in the plaque area, in particular in its relative NC content, was evident [11]. The 12 month follow-up of this trial recently demonstrated a further increase in the plaque size but with clear signs of pharmacologically induced vasomotion of the scaffolded segment, suggesting loss of the mechanical integrity and radial forces of the scaffold and favourable changes in the composition of the plaque [12].

The aim of this study was therefore to analyze the 12 month changes in the VH composition of only the plaque behind the polymeric struts.

Methods

Study population

The ABSORB Cohort B trial enrolled patients older than 18 years, with a diagnosis of stable, unstable or silent ischemia, subdivided in two subgroups: the first group (Cohort B1) underwent invasive imaging such as quantitative coronary angiography, IVUS, IVUS-VH and optical coherence tomography at 6 months whereas the second group (Cohort B2) underwent the same invasive imaging at 12 months. For the present analysis, we screened patients from ABSORB Cohort B2 with paired post-implantation and 12 month follow-up IVUS-VH.

Briefly, all lesions were de novo, in a native coronary artery with a reference vessel diameter of 3.0 mm, with a percentage diameter stenosis ≥50 and <100% and a thrombolysis in myocardial infarction flow grade of ≥1. All lesions were treated by implantation of the ABSORB scaffold 1.1 (3.0 × 18 mm) [7, 13]. Major exclusion criteria were: patients presenting with an acute myocardial infarction, unstable arrhythmias or patients who had left ventricular ejection fraction ≤30%, restenotic lesions, lesions located in the left main coronary artery, lesions involving a side branch >2 mm in diameter, and the presence of thrombus or another clinically significant stenosis in the target vessel. The trial was approved by the ethics committee at each participating institution and each patient gave written informed consent before inclusion.

Imaging acquisition and analysis

IVUS-VH post-implantation and at 12 month follow-up were acquired with a phased array 20 MHz intravascular ultrasound catheter (EagleEye™; Volcano Corporation, Rancho Cordova, CA, USA) using an automated pullback of 0.5 mm per second. The baseline and one-year follow-up region of interest were matched by use of anatomical landmarks. The radiofrequency data, required for VH analysis, were acquired during the IVUS pullback and raw radiofrequency data capture gated to the R wave (In-Vision Gold, Volcano). These files were stored on DVD and sent to an independent core laboratory for analyses (Cardialysis, BV, Rotterdam, The Netherlands).

The data were analyzed by the QIVUs software (Medis, Leiden, The Netherlands). This allowed the user to draw, in a semi-automatic fashion, a third contour (i.e. the scaffold contour). One limitation of this software, however, is its ability to read only IVUS-VH data acquired using a specific commercially available IVUS-VH console (s5i system, Volcano Corporation, Rancho Cordova, CA, USA).

Three different contours were drawn (lumen, scaffold and external elastic membrane) by experienced IVUS analysts blinded to the time of acquisition (e.g. baseline or 12 month follow-up) [14]. Reproducibility of these measurements using this software has been previously demonstrated to be good [11]. Four tissue components (necrotic core NC—red; dense calcium DC—white; fibrous FT—green; and fibrofatty FF—light green) were identified with autoregressive classification system. Each individual tissue component was quantified and colour coded in all IVUS cross sections [7, 15]. As previously described, the scaffold contour was drawn behind the polymeric struts of the ABSORB scaffold, excluding the polymeric struts and their ultrasonic signature from the quantification of the VH components of the plaque behind the struts (PBS). The PBS was calculated as the sum of all IVUS-VH components, excluding the grey media stripe seen with IVUS-VH.

In addition, we qualitatively evaluated the presence of necrotic core in contact with the lumen, defined as a confluent necrotic core >10% of plaque area without evidence of overlying non-necrotic core tissue in the treated segment and in the edges (5 mm length adjacent to each ABSORB scaffold edge), as previously reported [6, 11]. Importantly, as the polymeric struts are recognized as dense calcium surrounded by necrotic core, this surrounding necrotic core, with the help of the corresponding grey-scale image, was not interpreted as necrotic core in contact with the lumen [5, 6, 9, 11].

Statistical analysis

Discrete variables are presented as counts and percentages. Continuous variables are presented as means ± standard deviation (SD) or median and interquartile range, according to their normal or not normal distribution. Normal distribution of continuous variables was tested by the Kolgomorov-Smirnov test. The ratio between the number of frames with necrotic core in contact with the lumen and the total number of frames analyzed has been defined as the incidence of necrotic core in contact with the lumen for the treated segment and the edges. Paired comparisons between post-procedure and 12-month follow-up were done by paired t-test or Wilcoxon signed rank test, where appropriate. Correlation between parameters was performed by the Spearman test. A two-side P value of less than 0.05 indicated statistical significance. Statistical analyses were performed with the use of SPSS 16.0 software (SPSS Inc., Chicago IL, USA).

Results

Baseline clinical and angiographic characteristics

Overall, forty-six patients had paired post-ABSORB implantation and 12 month follow-up IVUS-VH data. Of these patients, only 17 patients (18 lesions) were imaged using the s5i system and therefore were included in the present study. Table 1 shows their clinical and angiographic data. No differences were found in the clinical and angiographic data between patients included and excluded from the analysis.

Table 1 Baseline clinical and angiographic characteristics
Full size table

Grey-scale IVUS and IVUS-VH analyses (Table 2)

PBS tended to increase from baseline to 12 months follow-up (from 2.39 ± 1.85 to 2.76 ± 1.79 mm2; P = 0.078) with a significant reduction in lumen area (from 6.37 ± 0.90 to 5.98 ± 0.97 mm2; P = 0.006). No changes were found in the EEM area (from 14.08 ± 3.14 to 14.32 ± 3.13 mm2; P = 0.349).

Table 2 IVUS analysis in the scaffold segment at baseline and follow-up (n = 18 lesions)
Full size table

In the IVUS-VH analysis of the PBS, there was a significant increase in the absolute and relative content of fibrous and fibrofatty tissue. Conversely, there was a significant decrease in the relative content of necrotic core and dense calcium (Fig. 1).

Fig. 1
figure 1

Example of reduction of necrotic core and dense calcium in the plaque behind the polymeric struts (white arrows). Note also the increase of plaque size with predominant content of fibrous tissue. Yellow contour is drawn behind the ABSORB polymeric struts in a semi-automatic way by the dedicated software, excluding struts from the VH quantification of the plaque behind

Full size image

Whilst there was no relationship between the changes in NC and DC areas and in the PBS area between baseline and follow-up (Spearman rho = 0.24, P = 0.336 for NC; Spearman rho = 0.15, P = 0.553 for DC), there was a significant relationship between the changes in FT and FF area and in the PBS area (Spearman rho = 0.87, P < 0.001 for FT; Spearman rho = 0.78, P < 0.001 for FF).

Incidence of necrotic core in contact with the lumen

Overall there was a reduction in necrotic core in contact with the lumen not only at the scaffold site, but also at the proximal and distal reference segments (Fig. 2). In particular the reduction of NC in contact with the lumen approached a statistical significance in the proximal reference segment (P = 0.09) (Fig. 3).

Fig. 2
figure 2

Incidence of segments with necrotic core in contact with the lumen at baseline and follow-up

Full size image
Fig. 3
figure 3

Representative IVUS-VH image of necrotic core in contact with the lumen in the proximal reference segment at baseline and which has disappeared at follow-up

Full size image

Intra-observer and inter-observer variabilities yielded good concordance for necrotic core in contact with the lumen (κ = 0.86 and κ = 0.80, respectively).

Discussion

The major finding of our study is that at 12 month follow-up there is an increase of plaque located behind the polymeric struts of the ABSORB scaffold, mainly due to an increase in its fibrous and fibrofatty content. Conversely, there is a significant decrease in necrotic core and dense calcium tissue.

The use of IVUS-VH in the study of the coronary plaque after stent/scaffold implantation is challenging, as metallic and polymeric struts are detected as dense calcium, surrounded by necrotic core and are accounted for the quantification of the plaque VH component [6, 9]. In addition, with bioresorbable polymeric struts, the weight of the scaffold in the quantification of the VH plaque components changes at various time points according to the scaffold bioresorption. Consequently, a reduction in the dense calcium and necrotic core content of the plaque may be an artifactual result of this process [7–9]. The semi-automatic introduction of a third contour, excluding the scaffold from the analysis of the VH plaque, allows us to overcome these problems.

Using this methodology, at 6 month follow-up we demonstrated an increase in plaque area and in all the VH components, without an increase of the necrotic core in contact with the lumen [11]. At 12 months, we conversely found a significant decrease in necrotic core and dense calcium, despite an increase in plaque area. The reduction of necrotic core was also accompanied by a reduction in the necrotic core in contact with the lumen either in the scaffold region or within the proximal/distal reference segments. Of note is that dense calcium is a frequent finding within the necrotic core region; in this case it is frequently “speckled” which can be due to calcification of a “nidus” of macrophages, a sign associated with plaque instability and rupture [16]. The reduction of dense calcium found in our analysis was probably mainly due to the reduction of this type of dense calcium within the necrotic core area (Fig. 1). However, as calcium tends to be stable over short follow-up, another possible explanation could be that scaffold bioresorption influences the radiofrequency backscattering signal behind the scaffold, producing a reduction in dense calcium. At 12-month follow-up there was also an increase in fibrotic tissue content of the plaque. Previous optical coherence tomography studies have shown that polymeric struts apposed and embedded in the vessel wall become covered by neointima hyperplasia that also intersperses between the polymeric struts [17, 18].

Everolimus is probably the major component in the reduction of VH-derived necrotic core. Verheye et al. showed in rabbit atherosclerotic plaques that a stent-based delivery of everolimus leads to a marked reduction in macrophage content without altering the amount of smooth muscle cells, namely inducing autophagy by mTOR inhibition. [19, 20] This autophagy of macrophages is classically described as a process of vacuolization of the cytoplasm with formation of auto-phagosome, digesting the surrounding atherosclerotic debris. This observation is important in the context of plaque stabilization, as it is generally assumed that the presence of macrophages triggers plaque destabilization [21]. In addition, the 12 month inflammatory reaction after ABSORB implantation in a porcine model has been shown to be lower than for metallic stents, in which the inflammatory process is prolonged [22]. These pathological findings from animal studies are in line with the present reduction in the necrotic core content of the atherosclerotic plaque at 12 month follow-up after ABSORB implantation, where the favourable effects of everolimus are combined with a bioresorbable platform [2–4]. It is noteworthy that the lack of everolimus effect on the smooth muscle cells may be the basis of the preserved pharmacologically induced vasomotion of the scaffold segment, as highlighted at the 12-month follow-up [12].

Statin treatment, that was used in most of the patients, should be considered as another possible explanation of our findings, in particular of the reduction in necrotic core, as previously shown [23]. Statin therapy could also explain the reduction in the necrotic core in contact with the lumen in the segment proximal to the device, as elution of everolimus is more frequent distal than proximal to the device [1].

Limitations

The present study included a small number of patients. IVUS-VH data pre-ABSORB implantation were not available. The interpretation of the backscattering signal from the plaque behind the polymeric struts by the IVUS-VH software may be influenced by the presence of the scaffold. IVUS-VH has not been validated to assess plaque composition behind scaffold struts and to identify bioresorbable struts. In addition struts bioresorption may account for the changes in the radiofrequency backscattering signal in the plaque behind.

Conclusions

At 12 months after ABSORB implantation, there was a slight increase in plaque area located behind the polymeric struts. Nevertheless, the necrotic core and dense calcium content of the plaque decreased significantly. These findings are compatible with the everolimus-induced autophagy of macrophages and subsequent reduction of inflammatory microenvironment of atherosclerotic plaque. The ability to decrease necrotic core content of coronary plaque without the permanent presence of metal may have important clinical implications in atherosclerosis treatment. Validation of the actual findings is needed and further studies with larger sample sizes and longer clinical follow-up are required before the impact of these observations can be fully understood.