Study Devices

The basic characteristics of each device are visualized in Fig. 1. Additionally, more detailed information about e.g. drug load or strut thickness can be found in supplementary Table 1. Implanted stent sizes are depicted in supplementary Table 2.

Fig. 1

Overview of the different study devices. Devices differed in following main stent components: metallic backbone, drug carrier matrix and anti-proliferative drug

CC-EEPFS (Cobalt Chromium Everolimus Eluting Polymer Free Stent, Test Device, ISAR Summit, n = 24) consisted of a novel thin-strut CoCr based everolimus-eluting stent (2.6 µg/mm2 everolimus) using probucol as excipient for delayed drug release.

SS-SEPBS (Stainless Steel Sirolimus Eluting Polymer Based Stent, Control 1, Yukon® Choice PC, biodegradable polymer-based DES n = 24, commercially available), a well-established older generation DES, differed in all three major components and consisted of a SS sirolimus-eluting stent (2.6 µg/mm2 sirolimus) using biodegradable polymer (PLA, polylactic acid) as matrix for delayed drug release.

SS-SEPFS (Stainless Steel Sirolimus Eluting Polymer Free Stent, Control 2, n = 23, designed specifically for this study and not commercially available) differed in two components and consisted of a SS sirolimus-eluting stent (2.6 µg/mm2 sirolimus) using probucol as excipient for delayed drug release.

CC-SEPFS (Cobalt Chromium Sirolimus Eluting Polymer Free Stent, Control 3, Vivo ISAR, n = 24, commercially available) differed in only one component and consisted of a CoCr backbone and probucol matrix with sirolimus (2.6 µg/mm2 sirolimus).

The strut thickness was 68 µm (devices for small vessels) or 79 µm (devices for medium vessels) for the CoCr-groups (CC-EEPFS and CC-SEPFS) and 90 µm/87 µm (devices for small vessels) or 95 µm/96 µm (devices for medium vessels) for SS-SEPBS/SS-SEPFS (SS-groups).

Study Design

Ten stents from each of the four different DES types were randomly implanted in the coronary arteries of 28 healthy, juvenile domestic farm pigs for histopathological analysis at 28- and 90- days follow-up. Additionally, three stents from each group were randomly implanted in the coronary arteries of four healthy, juvenile domestic farm pigs for scanning electron microscopy at 14-days follow-up (see Fig. 2). Study pathologists were blinded until raw data collection was finished.

Fig. 2

Schematic overview of the study design. 92 stents were successfully implanted in the coronary arteries of 32 conventional farm pigs. SEM analysis of n = 3 stents per group was performed 14 days after stent implantation. QCA and histopathological analysis was performed at short- (28 days) and mid-term (90 days) time points. LAD = left anterior descending, LCx = left circumflex, RCA = right coronary artery, SEM = scanning electron microscopy, QCA = quantitative coronary angiography

Porcine Model of Coronary Stent Implantation

A total of 33 juvenile domestic farm pigs (castrated males, about 3 months old) were included in the study. Animals were received from Sauenhaltung Tierbach (Uthmöden, Germany) and were acclimatized for 14 days prior to use in the study. All animal experiments were performed in accordance with the requirements and guidelines of the directive 2010/63/EU and the German Animal Protection Act (in its current version).

Interventional Procedures and Tissue Harvest

A loading dose of clopidogrel (75 mg) and aspirin (100 mg) was administered orally two days before treatment and long-acting Verapamil hydrochloride was given within 24 h prior to the procedure to prevent vascular spasm during catheterization.

After premedication with ketamine (20 mg/kg, i.m.) and xylazinhydrochloride 2% (2 mg/kg i.m.), pigs were put under general anesthesia using propofol 1% (3 mg/kg, i.v.). All animals received butorphanol (0.1 mg/kg, i.v.), meloxicam (0.4 ml/10 kg, i.m.) before intubation and ursocyclin as prophylactic antibiosis (1 ml/5 kg). Maintenance of anesthesia was performed by ventilation with a mixture of 30–60 vol% of pure oxygen, 40–70 vol% air and 1–2 vol% of Isoflurane. After sheath insertion in the carotid artery, heparin-natrium (5000 IU) and D,L-Lysinacetylsalicylat (250 mg) were administered intraarterially. Additional heparin doses were given as needed.

Angiographic imaging was performed before and after stent implantation, after application of glyceroltrinitrat (200 µg, i.c.). At day 0, the randomly assigned devices were implanted in the left anterior descending (LAD), left circumflex (LCx) and right coronary artery (RCA) of each animal with an overstretch of approximately 10% relative to the reference diameter of the vessel (overstretch ratio of 1.1:1). After stent implantation, coronary angiography was performed to document vessel patency and absence of residual dissection. Animals were sacrificed at 14-, 28- and 90-days following stent implantation after final angiography and complete necropsy was performed. Immediately after explantation, the hearts as well as each coronary artery were flushed with isotonic NaCl and fixed with 10% buffered formalin (pressure fixation; infusion of the fixative at a constant fluid pressure (~ 100 mm Hg) in each coronary artery). After fixation, the coronaries were explanted with a short segment of native (non-stented) vessel at the proximal and distal end. Vessel segments were marked with a suture tag at the proximal end of the vessel and photographed before processing.

Quantitative Coronary Angiography (QCA)

Angiographic images were acquired with contrast media to identify an appropriate location for the stent implantation site. Computer-assisted software for offline analysis (Qangio XA7.3, Medis Medical Imaging Systems, Leiden, Netherlands) was used for angiographic analysis. The following parameters were measured or calculated: percentage diameter stenosis, mean lumen diameter, minimum lumen diameter. Late lumen loss was calculated as the difference in minimum lumen diameter between post-procedure and follow-up.

Histopathological Evaluation

Vessels were embedded in methyl methacrylate (MMA) resin and divided into proximal (non-stented), proximal middle, middle, distal middle and distal (non-stented) blocks. Tissue blocks were then cut at 10 μm using a laser microtome (TissueSurgeon, LLS ROWIAK LaserLabSolutions GmbH, Hannover, Germany). Sections were stained with hematoxylin and eosin (H&E) and Verhoeff van Gieson (VVG). Histology slides were photographed and digitalized using an Olympus microscope (BX 41, Olympus, Tokyo, Japan) with associated camera (DP 74, Olympus, Tokyo, Japan). Morphometric measurements were taken in VVG images using cellSens Imaging Software (cellSens Standard 1.17, Olympus, Tokyo, Japan). All sections were examined by light microscopy. Vascular injury following stent implantation was graded as described previously [9].

The different scores are summarized in supplementary tables 3, 4, 5. Fibrin deposition (identified as intense, homogenous pink stain) and inflammatory response were assessed as previously described [3]. Cross sectional measurements (Lumen area and internal elastic lamina (IEL) area) were performed in stented cross sections and used to calculate percentage stenosis with following formula: percentage stenosis = [1- (Lumen Area/IEL Area)] *100.

Scanning Electron Microscopy (SEM)

Specimens were fixed in 10% buffered formalin for 48 h, followed by fixation in glycerol. Subsequently, they were bisected longitudinally, post-fixed in 100% glycerol, and gently rinsed with 0.1 M sodium phosphate buffer solution (pH 7.4). Afterwards, the samples were dehydrated in a graded series of ethanol and critical point dried. The luminal surface was exposed en face and sputter-coated with gold. Specimens were visualized using a Zeiss EVO MA 15 scanning electron microscope. Stent areas and uncovered areas were measured using ImageJ 1.53 k (National Institutes of Health, USA). Measurements were performed on consecutive stent strut rows identified on bisected stent halves. The area of discernable stent strut rows was measured using the polygon selection tool. The expected number of strut rows for each stent was provided by the manufacturer; rows where stent struts were buried under a dense layer of tissue were confirmed to exhibit complete endothelialization on high-power (200x) magnification and then classified as completely covered, and those with visible processing artifacts were excluded from analysis. Visible connector areas between struts were measured separately. Uncovered areas were measured accordingly. The percentage of uncovered stent strut area was first calculated per strut row and connector, and then summed up to represent the entire bisected stent half.

Statistical Methods

For QCA analysis, mean and standard deviation (SD), as well as the median and first- and third quartile were determined for each parameter. Continuous variables with normal distribution were compared using an unpaired student t-test, while variables showing non-parametric distribution were compared using Wilcoxon Rank-Sum test (Mann–Whitney u test).

For histopathological assessment, continuous data are presented as median (25th – 75th percentiles). Categorical data are presented as absolute and relative frequencies. Hypothesis testing of differences between the groups was performed using the Wilcoxon rank-sum test for continuous variables and the Pearson χ2 test (or Fisher’s exact test where any expected cell count of the contingency table was < 5) for categorical variables. To account for the clustered nature of the data, a linear mixed model was used for the analysis of histology data. The model contained a fixed-effects term for the variable of interest and a random intercept as random-effects term for animals in case of cross-sectional analysis and as nested random-effects term for animals and cross-section for strut-level analysis.

A p-value less than 0.05 was considered statistically significant. Statistical analysis was performed using the R 4.1.0 Statistical Package (The R Foundation for Statistical Computing, Vienna, Austria).