Abstract
Purpose
To assess the effects of local paclitaxel delivery using the Remedy catheter on neointimal hyperplasia in a porcine model and compare these results to commercially available BMS and biodegradable polymer-coated paclitaxel-eluting stents (BP-PES).
Methods and materials
A total of 31 stents were implanted into coronary arteries of 15 domestic swine including eight BMS, six BP-PES, and 17 BMS after intravasal paclitaxel delivery at doses of 250 μg (LPD250; n =9) and 500 μg (LPD500, n =6). All stents were implanted under quantitative coronary angiography (QCA) guidance to achieve a balloon/artery diameter ratio of 1.15:1.0. Twenty-eight days after the procedure, follow-up coronary angiography was performed, the animals were euthanized, and the coronary arteries harvested for histopathological analysis.
Results
At follow-up, QCA analysis revealed that lumen loss was significantly worse in BMS and in both LPD groups in comparison to BP-PES stents ( P =.02). Histomorphometric analysis showed that the LPD500 group presented the highest percentage of area stenosis, achieving a statistically significant difference in comparison to BMS and BP-PES stents.
Conclusion
Our study demonstrates that local paclitaxel delivery using the Remedy transport catheter in the two studied doses (250 and 500 μg) is not effective at neointimal hyperplasia inhibition.
1
Introduction
Drug-eluting stent (DES) implantation carries an increased risk of late and very late stent thrombosis thereby necessitating prolonged dual antiplatelet therapy . This is exacerbated by polymer- or drug-induced delayed endothelialization, chronic inflammatory reactions, late hypersensitivity reactions, and necrosis within vascular smooth muscle cells .
There are numerous proposals to avoid the suboptimal consequences of DES. These include biodegradable polymers, direct drug bindings to specially modified stent surfaces with a microporous structure, or biomimetic coatings utilizing endogenous or biocompatible materials . Furthermore, recent advances in drug-eluting balloon technology, an interesting alternative to situations unfavorable for stent usage, have renewed interest in local delivery of antiproliferative drugs to the arterial wall. Examples of such situations include peripheral vascular disease, in-stent restenosis, and unfavorable coronary anatomy (small vessels, distal, or tortuous segments, etc.). Despite this renaissance of non–stent-based drug delivery to arterial sites, the optimal choice of drug, vehicle, and delivery mode for nonstent arterial drug delivery remains controversial, even for thoroughly characterized drugs like paclitaxel. Therefore, we aim to assess neointimal hyperplasia levels with local paclitaxel delivery using the Remedy transport catheter (Boston Scientific) before bare metal stent (BMS) implantation in a porcine model and then compare to commercial BMS and paclitaxel-eluting stents.
2
Material and methods
2.1
Remedy transport catheter
The Remedy catheter (Boston Scientific) is a triple lumen, dual-purpose, angioplasty and drug delivery catheter ( Fig. 1 ). The noncompliant dilatation balloon allows for high-pressure angioplasty and is decoupled from the infusion lumen. The outer surface of the balloon consists of multiple distinct channels, each with several 100-μm holes through which infusate with therapeutic agents is delivered directly against the arterial wall . This permits high local drug concentration in the vessel wall, facilitates homogenous distribution of a solution against the vessel wall, and eliminates the risk of endothelial or vessel wall injury (when the pressure applied is under 4 atm and the infusate volume is under 5 ml) .
The Remedy catheter has been shown to be safe and effective in delivering therapeutic agents into the arterial wall .
2.2
Paclitaxel and its preparation
Because our objective was to investigate a commercial paclitaxel formulation without any unusual vehicle or special preparation, Poltaxel (Polfa Tarchomin, Poland) was selected. Current literature shows that vessels treated with the paclitaxel vehicle only (ethyl alcohol and castor oil in saline) showed no histopathological differences from paclitaxel . A concentrate for intravenous infusion was diluted in normal saline before usage. One milliliter of concentrate contains 6 mg of paclitaxel dissolved in a vehicle consisting of ethyl alcohol and chromatographically pure castor oil. In the present study, paclitaxel in two different doses (250 and 500 μg) was dissolved in 2 ml of 0.9% NaCl. This volume was administered to the arterial wall using the Remedy catheter.
2.3
Comparator stents
All tested stents were made of 316L alloy and had a closed cell design and strut thickness of 0.14 mm . The biodegradable polymer-coated paclitaxel-eluting stents (BP-PES; LUC-Chopin 2 , Balton, Poland) were covered with a multilayer structure containing a copolymer of lactic and glycolic acids blended with paclitaxel . Both stents used the same platform. Total polymer mass on a 3.0×15-mm stent did not exceed 360 μg. In vitro evaluations performed under shear conditions (system rinsing with physiological salt solution at 37°C) revealed that the coating degraded almost entirely in 8 weeks. Microscopic examination of BP-PES stent surface structure before and after expansion showed uniform coating without peeling or fractures. Paclitaxel at 1.0 μg/mm 2 was chosen to inhibit neointimal proliferation.
2.4
Study design
The study was approved by the local Animal Care and Use Committee of Silesian Medical University in Katowice (Poland). Fifteen domestic swine weighing 35–40 kg were included in the study. Three days prior to stent implantation, antiplatelet therapy of acetylsalicylic acid (300 mg loading dose, 150 maintenance dose throughout the remainder of the study) and ticlopidine (2×250 mg/day) was initiated and continued until study termination. Thirty-one stents (3.0–3.5×12 mm) were implanted into coronary arteries (left anterior descending, left circumflex, right coronary artery) including eight BMS, six BP-PES, and 17 BMS following local paclitaxel delivery in doses of 250 μg (LPD250, n =9) and 500 μg (LPD500, n =8) using Remedy catheters (3.0–3.5×20 mm). Only one stent per artery was permitted. The drug delivery balloon catheters were significantly longer than the studied stents to avoid “geographical miss” which can occur when approximately the same length devices are deployed. All efforts were made to balance and randomize the treatment groups between consecutive animals and different arterial locations. Individual animals received different treatments in different vessels such that the distribution of treatments by vessel and animal was randomized and balanced. Animals were enrolled based on their assigned identification number, in ascending order. Segments were chosen for the avoidance of side branches ensuring uniform interaction of the balloons and stents with the arterial wall. Efforts were made to inflate the Remedy catheter in the segment which ensured a slight oversize of 10% and proper balloon apposition was confirmed by angiographic visual assessment (no contrast flow past the inflated balloon). The paclitaxel, diluted in 2 ml of 0.9% NaCl solution, was given over 60 s under 2–3 atm of pressure. All stents were implanted using quantitative angiography guidance at an inflation pressure sufficient to ensure a balloon/artery diameter ratio of approximately 1.15:1.0.
Twenty-eight days after stent implantation, follow-up coronary angiography was performed. Subsequently, animals were euthanized; their coronary arteries were perfused at 100 mmHg with 5% formalin solution and prepared for histopathology.
The quantitative coronary angiography (QCA) was performed by an experienced analyst, utilizing CMS-QCA software (Medis, Leiden, Netherlands). Two orthogonal projections were chosen for stent assessment. Minimal lumen diameter (MLD) and reference vessel diameter (RVD) were measured at baseline, post-stent implantation, and 28 days’ follow-up. Post and follow-up MLD was taken from a segment including ±5 mm outside the stent (balloon-treated zone). The diameter of the inflated stent balloon was also measured to assess the stent-to-artery ratio. Percent diameter stenosis (%DS) at follow-up was calculated as: [100−(MLD/RVD)]×100%, and late lumen loss (LL) was defined as the difference between MLD in the angiogram obtained after stent implantation and at the 28 days’ follow-up.
For histopathological analysis, resin-embedded (glycol methacrylate) arterial segments containing the stents were cut into 10-μm slices using a Leica microtome (Leica RM2265). The specimens were stained with hematoxylin–eosin. Based on the histomorphometric analysis, the following parameters were assessed: lumen cross-sectional area (lumen CSA), stent cross-sectional area (stent CSA), and internal elastic lamina area (IEL area). Neointimal area (NA) was defined as the difference between IEL area and lumen CSA. The percentage of vascular lumen obstruction (%Area Stenosis) was calculated according to the following formula: 100×(IEL area−lumen CSA)/IEL area. Each of the aforementioned parameters was measured in a proximal, medial, and distal stent portion, and a mean value was calculated.
Finally, qualitative analysis to assess arterial wall integrity and the presence of endothelium, fibrin, thrombi, or focal necrosis was conducted. Injury score was evaluated as previously described by Schwartz et al. , and the inflammation score for each individual strut was graded as described by Kornowski et al. . The injury score and the inflammatory score for each cross-section were calculated as the sum of the individual injury and inflammatory scores, divided by the number of struts in the examined section.
2.5
Statistical analysis
Results are expressed as mean±S.D. Normal distribution of variables was verified by the Shapiro–Wilk and Kolmogorov–Smirnov test. Variance uniformity was verified with the Levene test. Angiographic and histomorphometric data were analyzed using ANOVA and post hoc (Newman–Keulus) tests. In any cases of skewed distribution or nonuniformity of variance, a nonparametric Kruskal–Wallis and Mann–Whitney U tests were used. P values of <.05 were considered statistically significant. Sample sizes of a minimum of six vessels per group were calculated to be substantial to give a power of 0.85 with a standard deviation of 0.2 mm and to detect a 0.4-mm difference in late lumen loss between tested groups. This assumption was based on a previous experimental study in which paclitaxel-coated balloons were compared to DES and BMS .
2
Material and methods
2.1
Remedy transport catheter
The Remedy catheter (Boston Scientific) is a triple lumen, dual-purpose, angioplasty and drug delivery catheter ( Fig. 1 ). The noncompliant dilatation balloon allows for high-pressure angioplasty and is decoupled from the infusion lumen. The outer surface of the balloon consists of multiple distinct channels, each with several 100-μm holes through which infusate with therapeutic agents is delivered directly against the arterial wall . This permits high local drug concentration in the vessel wall, facilitates homogenous distribution of a solution against the vessel wall, and eliminates the risk of endothelial or vessel wall injury (when the pressure applied is under 4 atm and the infusate volume is under 5 ml) .
The Remedy catheter has been shown to be safe and effective in delivering therapeutic agents into the arterial wall .
2.2
Paclitaxel and its preparation
Because our objective was to investigate a commercial paclitaxel formulation without any unusual vehicle or special preparation, Poltaxel (Polfa Tarchomin, Poland) was selected. Current literature shows that vessels treated with the paclitaxel vehicle only (ethyl alcohol and castor oil in saline) showed no histopathological differences from paclitaxel . A concentrate for intravenous infusion was diluted in normal saline before usage. One milliliter of concentrate contains 6 mg of paclitaxel dissolved in a vehicle consisting of ethyl alcohol and chromatographically pure castor oil. In the present study, paclitaxel in two different doses (250 and 500 μg) was dissolved in 2 ml of 0.9% NaCl. This volume was administered to the arterial wall using the Remedy catheter.
2.3
Comparator stents
All tested stents were made of 316L alloy and had a closed cell design and strut thickness of 0.14 mm . The biodegradable polymer-coated paclitaxel-eluting stents (BP-PES; LUC-Chopin 2 , Balton, Poland) were covered with a multilayer structure containing a copolymer of lactic and glycolic acids blended with paclitaxel . Both stents used the same platform. Total polymer mass on a 3.0×15-mm stent did not exceed 360 μg. In vitro evaluations performed under shear conditions (system rinsing with physiological salt solution at 37°C) revealed that the coating degraded almost entirely in 8 weeks. Microscopic examination of BP-PES stent surface structure before and after expansion showed uniform coating without peeling or fractures. Paclitaxel at 1.0 μg/mm 2 was chosen to inhibit neointimal proliferation.
2.4
Study design
The study was approved by the local Animal Care and Use Committee of Silesian Medical University in Katowice (Poland). Fifteen domestic swine weighing 35–40 kg were included in the study. Three days prior to stent implantation, antiplatelet therapy of acetylsalicylic acid (300 mg loading dose, 150 maintenance dose throughout the remainder of the study) and ticlopidine (2×250 mg/day) was initiated and continued until study termination. Thirty-one stents (3.0–3.5×12 mm) were implanted into coronary arteries (left anterior descending, left circumflex, right coronary artery) including eight BMS, six BP-PES, and 17 BMS following local paclitaxel delivery in doses of 250 μg (LPD250, n =9) and 500 μg (LPD500, n =8) using Remedy catheters (3.0–3.5×20 mm). Only one stent per artery was permitted. The drug delivery balloon catheters were significantly longer than the studied stents to avoid “geographical miss” which can occur when approximately the same length devices are deployed. All efforts were made to balance and randomize the treatment groups between consecutive animals and different arterial locations. Individual animals received different treatments in different vessels such that the distribution of treatments by vessel and animal was randomized and balanced. Animals were enrolled based on their assigned identification number, in ascending order. Segments were chosen for the avoidance of side branches ensuring uniform interaction of the balloons and stents with the arterial wall. Efforts were made to inflate the Remedy catheter in the segment which ensured a slight oversize of 10% and proper balloon apposition was confirmed by angiographic visual assessment (no contrast flow past the inflated balloon). The paclitaxel, diluted in 2 ml of 0.9% NaCl solution, was given over 60 s under 2–3 atm of pressure. All stents were implanted using quantitative angiography guidance at an inflation pressure sufficient to ensure a balloon/artery diameter ratio of approximately 1.15:1.0.
Twenty-eight days after stent implantation, follow-up coronary angiography was performed. Subsequently, animals were euthanized; their coronary arteries were perfused at 100 mmHg with 5% formalin solution and prepared for histopathology.
The quantitative coronary angiography (QCA) was performed by an experienced analyst, utilizing CMS-QCA software (Medis, Leiden, Netherlands). Two orthogonal projections were chosen for stent assessment. Minimal lumen diameter (MLD) and reference vessel diameter (RVD) were measured at baseline, post-stent implantation, and 28 days’ follow-up. Post and follow-up MLD was taken from a segment including ±5 mm outside the stent (balloon-treated zone). The diameter of the inflated stent balloon was also measured to assess the stent-to-artery ratio. Percent diameter stenosis (%DS) at follow-up was calculated as: [100−(MLD/RVD)]×100%, and late lumen loss (LL) was defined as the difference between MLD in the angiogram obtained after stent implantation and at the 28 days’ follow-up.
For histopathological analysis, resin-embedded (glycol methacrylate) arterial segments containing the stents were cut into 10-μm slices using a Leica microtome (Leica RM2265). The specimens were stained with hematoxylin–eosin. Based on the histomorphometric analysis, the following parameters were assessed: lumen cross-sectional area (lumen CSA), stent cross-sectional area (stent CSA), and internal elastic lamina area (IEL area). Neointimal area (NA) was defined as the difference between IEL area and lumen CSA. The percentage of vascular lumen obstruction (%Area Stenosis) was calculated according to the following formula: 100×(IEL area−lumen CSA)/IEL area. Each of the aforementioned parameters was measured in a proximal, medial, and distal stent portion, and a mean value was calculated.
Finally, qualitative analysis to assess arterial wall integrity and the presence of endothelium, fibrin, thrombi, or focal necrosis was conducted. Injury score was evaluated as previously described by Schwartz et al. , and the inflammation score for each individual strut was graded as described by Kornowski et al. . The injury score and the inflammatory score for each cross-section were calculated as the sum of the individual injury and inflammatory scores, divided by the number of struts in the examined section.
2.5
Statistical analysis
Results are expressed as mean±S.D. Normal distribution of variables was verified by the Shapiro–Wilk and Kolmogorov–Smirnov test. Variance uniformity was verified with the Levene test. Angiographic and histomorphometric data were analyzed using ANOVA and post hoc (Newman–Keulus) tests. In any cases of skewed distribution or nonuniformity of variance, a nonparametric Kruskal–Wallis and Mann–Whitney U tests were used. P values of <.05 were considered statistically significant. Sample sizes of a minimum of six vessels per group were calculated to be substantial to give a power of 0.85 with a standard deviation of 0.2 mm and to detect a 0.4-mm difference in late lumen loss between tested groups. This assumption was based on a previous experimental study in which paclitaxel-coated balloons were compared to DES and BMS .
3
Results
All stents were easily introduced into the selected coronary segments and implanted without any complications. Local drug delivery with the Remedy catheter was also uneventful. One animal (6.7%) with three stents (one BMS, two LPD500) died 6 h postprocedure without signs of stent thrombosis upon postmortem examination. All other animals remained in good physical condition, and steady increases in body weight were observed. At 28 days’ follow-up necropsy, no animal displayed macroscopic signs of myocardial infarction or inflammation.
3.1
Imaging outcomes at 28 days
There were no significant differences in the QCA results between all tested groups post stent implantation. Reference diameters were similar (BMS: 3.05±0.26; BP-PES: 3.00±0.35; LPD250: 3.31±0.22; LPD500: 3.02±0.13; P =ns) as well as MLD after stent expansion (BMS: 2.86±0.26; BP-PES: 2.82±0.35; LPD250: 3.13±0.24; LPD500: 2.86±0.15; P =ns). The stent diameter to reference diameter ratio was also similar for all types of stents (BMS: 1.17±0.11; BP-PES: 1.15±0.13; LPD250: 1.10±0.06; LPD500: 1.13±0.15; P =ns) thereby minimizing expected differences in the degree of arterial injury between groups.
At termination, angiographically examined stenosis severity expressed by late lumen loss and percent diameter stenosis was significantly reduced in BP-PES stents compared to all other groups ( Table 1 ). Of note, the reference diameter in the LPD500 group was the lowest of all groups and statistically lower than in the LPD250 at the follow-up. Interestingly, in this group it was also lower than its own baseline value, raising suspicion of negative remodeling.
