OUP user menu

Long-term assessment of a novel biodegradable paclitaxel-eluting coronary polylactide stent

Felix Vogt, Andreas Stein, Gösta Rettemeier, Nicole Krott, Rainer Hoffmann, Jürgen vom Dahl, Anja-Katrin Bosserhoff, Walter Michaeli, Peter Hanrath, Christian Weber, Rüdiger Blindt
DOI: http://dx.doi.org/10.1016/j.ehj.2004.06.010 1330-1340 First published online: 1 August 2004


Aim The aim of this study was to assess technical feasibility, biocompatibility, and impact on coronary stenosis of a new biodegradable paclitaxel-loaded polylactide stent. Due to high rates of in-stent restenosis and permanent nature of metal stent implants, synthetic polymers have been proposed as surrogate materials for stents and local delivery systems for drugs. Paclitaxel was shown to inhibit vascular smooth muscle cell proliferation and migration.

Methods and results A novel biodegradable double-helical stent was manufactured using controlled expansion of saturated polymers (CESP) for the moulding of a bioresorbable poly(d,l)-lactic acid (PDLLA). A modified balloon catheter for stent deployment was developed according to the mechanical stent properties. Twelve paclitaxel-loaded (170 μg) polylactide stents, 12 unloaded polylactide stents, and 12 316L bare metal stents were deployed in porcine coronary arteries of 36 animals. Six pigs of each group were sacrificed after 3 weeks and 3 months, respectively, for every setting. Drug release kinetics as well as histomorphometrical and histopathological analyses were performed. A slow paclitaxel release kinetic for more than 2 months and therapeutic tissue concentrations were demonstrated. Coronary stenosis after implantation of paclitaxel-loaded stents (30±5% or 49±4%) was significantly inhibited compared to unloaded PDLLA stents (65±10%, Math or 71±4%, Math) and metal stents (53±6% or 68±8%, Math and Math) after 3 weeks or 3 months. Early complete endothelialisation was shown. Nevertheless, a local inflammatory response to the polylactide as a result of the polymer resorption process was observed.

Conclusions This novel polylactide stent showed sufficient mechanic stability, and by incorporation of paclitaxel, a significant potential to reduce restenosis development after vascular intervention was seen.

  • Stents
  • Polymer
  • Restenosis
  • Paclitaxel


Although metal stents have clearly improved the outcome after percutaneous coronary interventions (PCI), stent restenosis remains the major challenge.1,2 Due to limitations of metallic implants as artefacts in in-stent lumen visualisation by standard magnetic resonance and computed tomography angiography, problems of post-PCI coronary artery bypass grafting, and unknown long-term effects, bioresorbable polymers appear to be interesting candidate materials for alternative stent concepts. Several different polymers have been examined as candidate materials for stents and stent coatings,3–5 though most materials induced a marked inflammatory and thrombogenic response.6 One candidate material, polylactic acid, was proposed to combine good mechanical stability, biodegradability, and sufficient blood compatibility.7 Generally, polylactide polymers are widely and safely used as suture and osteosynthesis materials, where they are hydrolytically degraded to lactic acid. Nevertheless, a potential to induce inflammatory responses by the resulting acidic degradation products was described.8,9 Biodegradability of polymers provides a means for controlled release of incorporated drugs. Therefore, these materials were suggested to have the potential advantage to combine the scaffolding properties of stents with local drug delivery.

The pathophysiology of restenosis constitutes a complex interaction between cellular and acellular elements of the vessel wall and the blood.10 Some anti-proliferative or anti-inflammatory agents have been shown to elute slowly from polymer coatings and to be associated with reduced neointima formation in animal models. Two anti-proliferative agents, paclitaxel11,12 and sirolimus,13 have been used in humans with promising early results. Results of the RAVEL and SIRIUS trial demonstrated that sirolimus-eluting stents effectively inhibit restenosis in humans.14–15 The recent TAXUS I–IV trials revealed a significant inhibition of coronary stenosis by paclitaxel.16–19

The aim of the present study was to develop and evaluate a newly designed poly(d,l)-lactic acid (PDLLA) stent with an adapted double helical geometry that can release the anti-proliferative substance paclitaxel over a period of at least 4 weeks. For this reason, a new manufacturing technique using controlled expansion of saturated polymers (CESP) for the moulding of biodegradable PDLLA was used which allows the incorporation of bioactive biological substances due to low process temperatures of ≈37 °C. The stents were implanted in a porcine restenosis model to assess technical feasibility, long-term biocompatibility, and coronary stenosis.


Stent design and mechanical properties

A double helical stent geometry was developed according to the material properties and based on the finite element method20,21 (Fig. 1(a) ). The stent was manufactured using the CESP-process.22 This method is characterised by a low process temperature that enables the processing of thermally sensitive polymers and the incorporation of biologically active substances. PDLLA polymer granules (R207, molecular weight 240–250 kDa, Roche, Germany) and paclitaxel powder (Sigma–Aldrich, Germany) were filled in the autoclave of a high-pressure gas loading equipment. The polymer was plasticised by applying a carbon dioxide pressure of about 120 bar at ≈37 °C. By controlled pressure release, the microcellular foaming of the polymer was initiated (average bubble diameter of 30 μm, Fig. 1(b)). The stent characteristics are: wall thickness 250 μm, stent diameter 3 mm, and stent length 20 mm. The total stent surface area is 87 mm2 and the expanded stent covers 46% of the vessel wall. 170 μg paclitaxel was loaded per stent. Complete expansion of the stents is reached at 6 bar insufflation pressure. The radial strength resists an outside pressure of 0.5 bar without relevant diameter shrinkage that corresponds to pressures resulting from physiological vascular tones.

Fig. 1

PDLLA stent: (a) The double-helical geometry of the PDLLA stent was designed based on FEM results. (b) Electron microscope photography of a stent cross-section with the characteristic foam structure.

Animal model

A porcine model of coronary arterial injury was used as described previously.23,24 The animal study was approved by the Institutional Animal Care and Use Committee and conformed to the tenets of the American Heart Association on research animal use. The number of necessary animals was estimated by a power calculation. For all experiments, the pigs were randomly assigned to the three different stent groups before implantation. After establishing arterial access by Doppler guided puncture of the femoral artery, 7500 IU heparin were administered and the right coronary artery (RCA) was engaged with a right Multipurpose 3.3 mm guiding catheter. After coronary angiography, the vessel diameter was calculated using the guiding catheter for calibration. A specially developed modified balloon catheter (diameter 3.0 mm) was placed in the RCA. Six unloaded PDLLA stents, 6 loaded stents with 170 μg paclitaxel, and 6 316L bare metal stents were deployed in the proximal RCA of 18 pigs by balloon inflation for 2×30 s (10–12 atm). The animals were sacrificed after 3 weeks. An additional long-term follow-up group consisted of 18 animals with 6 unloaded PDLLA stents, 6 paclitaxel-loaded stents, and 6 316L bare metal stents that were sacrificed after 3 months. The stent-to-artery ratio was 1.3, the ratios were similar in all groups. The animals were maintained on 75 mg clopidogrel p.o. and on 100 mg aspirin p.o. until sacrifice.

In vitro and in vivo pharmacokinetic studies

In vitro elution studies were carried out at room temperature in gently shaking phosphate buffered saline (PBS). Expanded paclitaxel-loaded (170 μg/stent) stents (Math) were placed in 1.5 ml vials filled with 500 μl PBS. After defined intervals, the stents were removed and placed in fresh PBS. Samples were collected, stored at −20 °C, and subjected to paclitaxel enzyme linked immunosorbent assay (ELISA) measurements. For in vivo pharmacokinetic studies, at each time point (i.e., 14 days, 28 days, and 3 months) 3 stents from 3 different randomly selected pigs after implantation of loaded stents were excised. Three randomly selected unloaded PDLLA stents and their surrounding tissue served as a negative control. The vascular segments were sliced longitudinally and stents were carefully separated from medial and neointimal tissue. The vascular tissue and the stents were each subjected to ELISA measurements. For paclitaxel measurements, a commercial quantitative ELISA was used (Hawaii Biotechnology Group, Hawaii, USA).


Immuno-staining was performed on 5 μm paraffin embedded tissue sections as described previously.24,25 Anti-Ki-67 (1:10, Dako, Germany) and anti-von Willebrand factor (vWF, 1:600, Dako) were used as primary antibodies, and a horse radish peroxidase labeled secondary antibody (1:50, Dako) was applied for detection. Negative controls were carried out with non-immune IgG (Sigma). For immuno-stainings, the cell nuclei were counterstained with Mayer's hemalaun (Merck, Germany).

Histomorphometrical and histopathological evaluation

The hearts were harvested and the epicardial coronary arteries were removed. The arteries were sectioned into 3–5 mm segments and embedded in gelatin. 10–12 μm sections from the proximal, mid and distal part of the stents were performed. An alcohol-free thionin staining (Feyrter stain: 1% thionin solution, 0.5% tartaric acid) was performed from fresh gelatin sections to prevent dissolving of the polymer material. Metal stents were embedded in Technovit 9100 new (Kulzer, Germany) and stained with Giemsa. Lumen area, as well as internal elastic lamina area, was measured. Coronary stenosis was defined as 1-lumen area/internal elastic lamina×100%. Mean delta neointimal area was calculated as the difference between neointimal areas of each group after 3 weeks and 3 months. Inflammatory response to the polymer was analysed by counting of chronic inflammatory cells in multiple fields around the stent and near the lumen. Proliferation was analysed using Ki-67 immuno-staining. For each tissue section, 5 high-power fields located either at the luminal and abluminal stent margin or near the lumen were randomly chosen and the ratio of stained cells to total cells was assessed. Fibrin deposition was analysed by Ladewig staining. For this reason, the nuclei were stained with Weigert's iron haematoxylin after fixation and hydration. Then, tissue was stained with 1% phosphotungstic acid and after rinsing with water, further stained with a solution containing 0.5% methyl blue, 2% orange G, and 1% acid fuchsin. Endothelialisation was evaluated by staining with vWF. The analysis was performed using a microscope (Leica DM RX, Leica, Germany) equipped with a video documentation and software analysis program (Quantimed Q600, Vers. 01.06).

Statistical analysis

All results are expressed as means±SEM. Kruskal–Wallis test was used to determine overall statistical significance followed by the Mann–Whitney Math test with Bonferroni correction for subsequent pair-wise comparison if the overall significance was Math0.05. A Math value of Math0.05 was considered to indicate statistical significance.


In vitro and in vivo pharmacokinetics

In vitro drug release of paclitaxel stents (Math) was measured by quantitative ELISA. The stents were incubated in rotating PBS buffer that was changed in defined intervals over a period of 3 months. Drug release can be described by a slow release profile with an exponential function starting with a daily release between 5 and 8 μg that decreased to 1 μg after 4 weeks and was approaching zero after 3 months. After 3 months, all paclitaxel (162±43 μg) was eluted (Fig. 2(a) ). Furthermore, the in vivo paclitaxel concentration per gram wet weight of arterial tissue was determined. The tissue concentration was 0.5±0.05 μg/g after 14 days, 3.2±1.8 μg/g after 28 days, and below the detection limit after 3 months. This inversely corresponded with residual paclitaxel amounts per stent in vivo of 128±40 μg after 14 days, 63±2 μg after 28 days, and zero after 3 months (Fig. 2(b)). Unloaded stents served as the negative control where no paclitaxel was detectable.

Fig. 2

In vitro and in vivo pharmacokinetics: (a) In vitro drug elution was measured over a time span of 3 months. The figure shows accumulated drug amounts with accumulated standard errors. (b) ELISA measurements of paclitaxel doses in the vascular wall of stented segments and the corresponding PDLLA stents.

Histomorphometrical measurements

All vessels that were analysed were angiographically and histologically patent. After 3 weeks, coronary stenosis was significantly reduced in paclitaxel-loaded stents (30±5%) compared to unloaded PDLLA stents (65±10%, Math) and metal stents (53±6%, Math, Fig. 3(a) ). No difference could be observed between unloaded PDLLA stents vs. metal stents (Math). Thus, coronary stenosis after paclitaxel-loaded stent implantation was inhibited by 53% compared to unloaded PDLLA stents and by 44% compared to metal stents. The mean neointimal area was 2.7±0.4, 0.9±0.2, and 3.5±0.4 mm2 for the unloaded PDLLA, loaded PDLLA and metal stent group (Math for unloaded vs. loaded PDLLA stents, Math for loaded vs. metal stents and Math for unloaded vs. metal stents). The mean internal elastic lamina area after 3 weeks was 4.5±1.0, 3.0±0.2, and 6.7±0.4 mm2 (Math), the mean external elastic lamina area was 9.8±1.1, 7.0±0.7, and 9.5±0.5 mm2 for unloaded PDLLA, loaded PDLLA and metal stents, respectively (Math).

Fig. 3

Histomorphometrical analysis. Photomicrographs and statistical evaluation of histomorphometric analysis of coronary stenosis after implantation of unloaded PDLLA stents (left), paclitaxel-loaded PDLLA stents (middle) or metal stents (right, magnification 40-fold) after 3 weeks (a) and after 3 months (b). S=stent, L=lumen, N=neointima; the arrows indicate neointimal area.

After 3 months, coronary stenosis in the paclitaxel-loaded stent group was 49±4% compared to unloaded PDLLA stents (71±4%, Math) and metal stents (68±8%, Math). No difference could be observed between unloaded PDLLA stents vs. metal stents (Math). Thus, coronary stenosis was reduced by 31% and 28% compared to unloaded and metal stents (Fig. 3(b)). The mean neointimal area was 3.6±0.5, 2.5±0.3, and 3.6±0.5 mm2 for the unloaded PDLLA, loaded PDLLA and metal stent group, respectively (Math for unloaded vs. loaded PDLLA stents, Math for loaded vs. metal stents and Math for unloaded PDLLA vs. metal stents). Mean delta neointimal area was 0.9±0.7, 1.6±0.4 and 0.1±0.6 mm2 for the unloaded PDLLA, loaded PDLLA and metal stent group, respectively (Math for all groups). The mean internal elastic lamina area after 3 months was 5.1±0.5, 5.0±0.4, and 5.2±0.4 mm2 (Math for all groups), the mean external elastic lamina area was 10.4±0.3, 10.8±0.3, and 7.8±0.7 mm2 for unloaded PDLLA, loaded PDLLA and metal stents, respectively (Math for all groups). All histomorphometric data are summarised in Table 1. Stent malapposition was not detectable in the investigated specimens and by some IVUS examinations after stent deployment, although IVUS was not routinely performed in order to avoid vessel injury.

View this table:
Table 1

Histomorphometrical data

Unloaded PDLLA stentsPDLLA stents+paclitaxelMetal stents
3 weeks
Lumen area1.7±0.7 mm22.0±0.2 mm23.2±0.5 mm2
Neointimal area2.7±0.4 mm20.9±0.2 mm23.5±0.4 mm2
Internal elastic lamina area4.5±1.0 mm23.0±0.2 mm26.7±0.4 mm2
External elastic lamina area9.8±1.1 mm27.0±0.7 mm29.5±0.5 mm2
Coronary stenosis65±10%30±5%53±6%
3 months
Lumen area1.5±0.1 mm22.5±0.2 mm21.7±0.4 mm2
Neointimal area3.6±0.5 mm22.5±0.3 mm23.6±0.5 mm2
Internal elastic lamina area5.1±0.5 mm25.0±0.4 mm25.2±0.4 mm2
External elastic lamina area10.4±0.3 mm210.8±0.3 mm27.8±0.7 mm2
Coronary stenosis71±4%49±4%68±8%

Histopathological analysis

The neointima of all stents mainly consisted of spindle-shaped cells and extracellular matrix. After 3 weeks, fibrin deposition was observed to be mild and to be preferentially located at the abluminal side of PDLLA stents. No difference in fibrin deposition was found between unloaded and paclitaxel-loaded stents (Fig. 4(a) ). After three months, no fibrin could be observed (Fig. 4(b)). VWF staining indicated complete endothelialisation of the lumen after 3 weeks and after 3 months (Fig. 4(c) and (d)). There was no thinning of the media layer detectable indicating no paclitaxel-induced media necrosis. Medial cell counts per field after 3 weeks were 101±5, 95±5, and 110±10, for unloaded PDDLA stents, loaded stents and metal stents, respectively (Math). After 3 months there were 102±6, 119±7, and 111±12 medial cells per field for unloaded PDDLA stents, loaded stents and metal stents, respectively (Math, Fig. 4(e)).

Fig. 4

Histopathological analysis of fibrin deposition, endothelialisation, and medial cellular content. Photomicrographs of PDLLA stent tissue sections, S=stent, L=lumen, N=neointima, F=fibrin, EC=endothelial cells. Ladewig-stained tissue sections with mild fibrin deposition after 3 weeks (a, indicated by arrows) and none after 3 months (b, magnification 100-fold for both). Complete endothelialisation after 3 weeks (c) and after 3 months (d) was indicated by vWF staining (magnification 100-fold and 400-fold for c and d). Medial cell counts per field were similar between all groups (e).

Infiltration with chronic inflammatory cells was observed around the stent in response to the polymer resorption process. After 3 weeks, tissue sections of unloaded stents showed significantly more chronic inflammatory cells (42±5 cells/field) than paclitaxel-loaded and metal stents (5±1 and 6±1 cells/field, Math for both) around the stent, while no significant inflammation was observed near to the lumen (1±1, 1±1 and 1±0 cells/field, Math). After 3 months, both PDLLA stents induced significantly more inflammation compared to metal stents due to resorption of the stent (93±12 and 97±10 vs. 5±1 cells/field, Math for both). Mild foreign body reaction with scarce giant cells and some macrophages was visible around the PDDLA stents as well as around a few stent struts in the metal stent group. Again, in the neointimal area distant from the stent significantly less inflammation was found (2±1, 1±1, and 1±0 cells/field, Math, Fig. 5(a)–(e) ).

Fig. 5

Histopathological analysis of inflammation, proliferation, and stent stability. Photomicrographs of PDLLA stent tissue sections, S=stent, L=lumen, N=neointima, CIC=chronic inflammatory cells. Counts of chronic inflammatory cells around stents and near the lumen after 3 weeks or 3 months (a). PDLLA stents with infiltration of chronic inflammatory cells after 3 weeks (b and c, magnification 100 and 400-fold) or 3 months (d and e, magnification 100-fold for both). Proliferation indexes around the stent and near the lumen after 3 weeks or 3 months (f). Ki-67 staining of unloaded PDLLA stents (g and h, magnification 100 and 1000-fold) and paclitaxel-loaded stents (i and j, magnification 100 and 1000-fold). The arrows indicate Ki-67 positive cells.

Proliferation of neointimal cells was compared in paclitaxel-loaded and unloaded stents. As response to the anti-proliferative effect of paclitaxel, there was a strong decrease in proliferative activity around the stent in drug-loaded, compared to unloaded, PDDLA stents after 3 weeks (4.5±1.2% vs. 9.9±1.4%, Math), while after 3 months both proliferation indexes had decreased (1.5±1.3% vs. 3.4±0.9%, Math). The proliferation indexes near the lumen were 1.5±1.1% vs. 2.0±1.0% and 0.7±0.4% vs. 0.7±0.3% (Math and Math) in the drug-loaded compared to the unloaded PDLLA stent group after 3 weeks and after 3 months, respectively (Fig. 5(f)–(j)).

After 3 weeks, no relevant stent degradation such as fragmentation, disruption of stent struts or complete disintegration by hydrolysis could be observed. Only a few cell nests, consisting of spindle-shaped cells and of chronic inflammatory cells in small cavities of the stent indicated the onset of stent replacement by the surrounding tissue (Fig. 5(b)). By contrast, after 3 months, hydrolysation of the stent had occurred and more than 50% of the stent mass was replaced by fibrotic tissue (Fig. 5(e)). Nevertheless, no complete disruption of stents could be observed.


The present study demonstrates that, in an animal model, the novel biodegradable double-helical stent (a) can reliably be placed in coronary arteries and shows sufficient mechanical stability, (b) releases the anti-proliferative substance paclitaxel over a period of more than 2 months with a slow release profile, (c) induces less coronary stenosis after incorporation of 170 μg paclitaxel compared to unloaded PDLLA stents and bare metal stents over 3 months, but (d) effects a local inflammatory reaction due to the resorption of the polylactide in a long-term course.

Stent concept, bio- and haemocompatibility

In prior studies, the use of vascular polymer stents has been limited due to minor mechanical stability and higher inflammatory potential of different polymer materials.6,7,26 Hence, in this study a new manufacturing process (CESP-process) was used to produce the polylactide stents. With this manufacturing technique, and based on the stent design, the radial force resistance reached by the polymer stent was 0.5 bar. Compared to currently available metal and biodegradable stents,3 the polymer stents show sufficient resistance against the radial forces of the coronary vascular tone.

In contrast to conventional stent coatings, this technology allows the incorporation of biological substances at low process temperatures with a reliable slow release profile over more than 2 months. Pronounced degradation of PDLLA stents was not observed after 3 weeks, but after 3 months highly advanced resorption of the polymer with replacement by fibrotic tissue was observable. Currently, there are no available data which analyse the long-time resorption of bioresorbable polymers in coronary arteries. The only study that investigated long-term biocompatibility of polylactide vascular implants in rabbit aortas was performed by Hietala et al.27 The authors observed stent hydrolysis onset at 12 months and disintegration at 24 months with mild replacement by fibrosis accompanied by only a mild foreign body reaction. By contrast, the data from this study indicate that polylactide degradation already starts after 3 weeks. Resorption was accompanied by inflammatory reaction and increased proliferation rates. Importantly, this response was locally limited to the polylactide and did not spread to other parts of the neointima or media. Paclitaxel significantly inhibited the inflammatory response. Lincoff et al.28 observed that the inflammatory response to poly-l-lactic acid depends on the molecular weight of the substance and that low molecular weight polymers (80 kDa) cause a significant inflammatory reaction of the vascular wall, while high weight polymers (321 kDa) do not induce a significant response. Although a high molecular weight polymer was used in this study, these observations could only be confirmed by the 3 weeks results. There were increased stenosis rates and increased delta neointimal areas in the paclitaxel-loaded stent group after 3 months compared to the 3 week data indicating that loaded stents might suffer from a late catch-up phenomenon after complete release of pacitaxel. Thus, according to the long-term results of this study, in the loaded stent group the inflammatory response to the polymer might partially be delayed by paclitaxel. Nevertheless, because polymer resorption was highly advanced after 3 months and paclitaxel was already fully eluted after 2 months, the long-term data indicate persisting stenosis reduction after drug-loaded stent implantation compared to both control groups. Drug loading of the polymer stent with paclitaxel seems to be the main contributor to these favourable effects. The only study that revealed detailed clinical data on polymer stent implantation in men was conducted by Tamai et al.3 The Igaki–Tamai stent was a self-expanding coil stent made of PLLA monofilaments (molecular mass 183 kDa) with a helical design. The thickness of the stent struts was 0.17 mm and the stent covered 24% of the vessel area. The radial force resistance was 11.9%/0.06 bar resulting in a complete compression at 0.5 bar. Compared to the PDLLA stent the Igaki–Tamai stent shows a thinner wall thickness and covers less total surface area of the vessel wall. The thinner wall thickness might contribute to the lower radial force resistance of the Igaki–Tamai stent compared to the PDLLA stent. Nevertheless, after implantation of 25 Igaki–Tamai stents in men, no stent recoil was demonstrated and the restenosis rate of the unloaded stents was moderate with 10.5% demonstrating sufficient mechanical stability and biocompatibility of the polymer.

Drug release kinetics and effect of paclitaxel on coronary stenosis

Paclitaxel was chosen to be loaded into the polymer to evaluate the new stent concept. Dose finding revealed a total stent dose of 170 μg based on the inhibitory in vitro levels of paclitaxel for SMC proliferation projected on the stent area.29 In vivo pharmacokinetics demonstrated therapeutic paclitaxel tissue concentrations of 0.5 μg/g (corresponding to 0.7 μmol/l) after 14 days and 3.2 μg/g (corresponding to 3.7 μmol/l) after 28 days. After 3 months, tissue concentrations were close to zero. Also, concerning the in vitro elution of paclitaxel for more than two months, tissue concentrations over this period within the growth-inhibitory range for SMCs can be assumed.29 This was confirmed by decreased proliferative activity in the paclitaxel-loaded stent group and by significant stenosis reduction.

There are some animal studies and clinical studies available that have analysed the effect of paclitaxel on restenosis development. Polymer-based and non polymer-based local drug delivery approaches were chosen. Heldman et al.30 deployed paclitaxel-loaded (187 μg) metal stents without polymer coating in porcine coronary arteries which resulted in reduced neointima growth in the paclitaxel group. Results from implantation in men were received after implantation of paclitaxel-loaded, non-biodegradable polyacrylate QuaDS-QP2 stents (2400–4000 μg paclitaxel derivate) in native coronary lesions11 and in restenotic lesions.12 Analysis of the study data revealed a cross-sectional area increase of 14% in patients with primary lesions. After implantation in in-stent restenotic lesions, at 12 months 62% had angiographic restenosis. Delayed appearance of restenosis and a relatively high complication rate was explained by paclitaxel overloading of the QP2 stent or an inflammatory reaction to the polymer sleeve31 The SCORE trial was terminated because of a high major adverse cardiac event rate (10%) with an increased rate of stent thrombosis at 30 days in the QP2 group.32 The recently published TAXUS I–IV trials using a hydrocarbon-based polymer coated NIR-stent loaded with 1 μg paclitaxel/mm2 and with a 10 days slow biphasic paclitaxel release profile revealed low MACE rates after 12 months (3%) and restenosis rates (late loss 0.36 vs. 0.71).16–19

In summary, this study presents the first long-term data evaluating the effect of polylactides in coronary arteries alone, or in combination with an anti-proliferative agent such as paclitaxel. Paclitaxel was shown to effectively inhibit proliferation and coronary stenosis after vascular injury in a long-term course. In vitro pharmacokinetics with a very slow release pattern of paclitaxel over a time span of more than 2 months might contribute to this favourable outcome. The CESP manufacturing process using biodegradable polymers provides the possibility of long-term local drug delivery. Paclitaxel loading did not induce increased toxicity as reflected by media necrosis or delayed healing of endothelial cells. As the temporal response to injury is greatly accelerated in the pig model compared to humans, our data might correspond to about 12–18 months follow-up data in humans.33

Study limitations

Although paclitaxel was shown to effectively inhibit SMC proliferation and inflammatory responses to polylactide resorption, there was continuing local inflammation after 3 months due to polymer resorption. Even though a high injury approach with considerable overstretch during stent deployment was used in order to ensure sufficient neointima formation in the animal model, a contribution of the continuing polymer resorption and inflammation process to restenosis development cannot be excluded. As the data from this study are collected in an animal model, conclusions drawn from this study might apply only in physiological conditions because all studied parameters were not performed under disease states.


This research project was supported by the IZKF “BIOMAT.” – Interdisciplinary Center for Clinical Research in Biomaterials and Tissue–Material-Interaction in Implants (BMBF project No. 01 KS 9503/9). We thank Angela Freund from the university department of oral, maxillofacial and plastic facial surgery, Ute Müller from the BMP Aachen, and Tadeusz Stopinski from the university animal care unit for excellent technical assistance.


  • 1 Both authors contributed equally.


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
View Abstract