Characterisation of Cardiovascular Stents
Employing the imaging techniques described in this article for characterisation, problem solving and product development offers companies significant advantages, says Dr Alan Brown of CERAM.
For the past twenty years, the outlook for patients suffering from coronary arterial blockage has improved enormously. This is a result of four distinct revolutions in treatment, starting with balloon angioplasty, followed by bare metal stent (BMS) implants, drug-eluting stents (DES) and most recently, successful clinical trials of Abbott’s ABSORB, one of the first bioresorbable vascular scaffolds (BVS). Today, the worldwide stent industry has an annual turnover exceeding US$10 billion.
For DES and BVS products, the drug delivery systems consist of a polymer matrix with an anti-proliferation agent (for example, paclitaxel, sirolimus or everolimus) to minimise restenosis, a reaction of the vessel wall, which results in the build-up of smooth muscle cells at the stent implantation site. The controlled release of the drug is crucial to the success of these formulations. In particular, the early phases of elution, the ongoing magnitude of the release rate and any temporal variations in rate are critical in providing maximum patient benefit while minimising risk of systemic toxicity.
In vitro dissolution studies monitoring drug elution by, for example, high performance liquid chromatography (HPLC) techniques are commonly used to monitor the process for quality control purposes. However, the development of surface and in-depth characterisation methods based on techniques such as secondary ion mass spectrometry (SIMS) and white light interferometry (WLI) now provide medical device manufacturers with a set of tools that give a detailed insight into drug distribution through the coating system, layer thickness and the cleanliness of substrate materials. This article discusses the application of these essential characterisation techniques to a range of BMS and DES products, based on real-life product development issues.
Monitoring stent cleanliness
For BMS and DES one of the critical factors in manufacture is the uniformity and cleanliness of the metal substrate. The presence of processing residues and impurities, the thickness of metal oxides from the alloy components and the surface topography may all play a role in the long term behaviour of the device.
The ultra low energy SIMS (ULESIMS) depth profile of a 316 stainless steel stent shows the significant compositional changes occurring within the first 15 nm of the metal surface (Figure 1). The outer surface contains carbon (probably from organic residues) with an underlying oxide layer approximately 3 nm thick. This oxide layer is enriched with chromium and iron oxide species, at the expense of the nickel component of the alloy. The carbon-rich region below the oxide layer is a result of the wire-forming process or polishing stage.
Figure1: ULESIMS depth profile of a 316L
Design of the DES coating structure
DES typically consist of a stent platform and a coating system that contains a polymer or copolymer containing an active constituent (drug). In some cases a primer layer such as Parylene C is initially applied to the stent platform to aid adhesion of the subsequent layers. A capping layer may also be added to induce delayed or controlled delivery of the drug. With complex layer structures like this there is a clear need for analytical techniques that provide molecularly specific in-depth information. In conventional dynamic SIMS (DSIMS) depth profile analysis any molecular information is rapidly lost from the detectable secondary ions produced as a result of ion beam-induced radiation damage. However, in favourable cases, the drug, polymer system and primer layers may be uniquely distinguishable by the presence of hetero-atoms within their structure.
In a DES development study, designed to investigate the effect of capping layers on drug release, a 316L stainless steel platform was coated with Parylene C primer approximately 1-µm thick, monitored by chlorine (Cl); a drug-loaded polymer system approximately 4-µm thick; followed by an optional approximately 1-µm polymer cap. The polymer systems contained butyl methacrylate and polyvinylpyrrolidone monitored as butoxy and CN species, respectively. The drug's chemical structure contained a thiazole system, hence sulphur was used as a unique monitoring species.
The normalised DSIMS depth profile of the uncapped system in Figure 2 shows a relatively steady composition for drug and polymer constituents throughout the coating. For the capped sample, the concentration of drug is significantly lower in the capping layer, as expected. There is, however, some drug still present within the cap, which will be eluted into the body immediately following stent placement.
Further studies on differing drug–polymer combinations have illustrated how the temporal elution of drug can be controlled or phased using the varying miscibility of drug within the capping-polymer system. In these studies, the in-depth composition of the DES layer as determined by DSIMS has been compared with the in vitro drug elution profile monitored by HPLC. In all cases there is good agreement between the two techniques.
Figure 2:Normalised DSIMS depth profilesof drug-eluting stents with (right) and without (left) capping layers.
Characterisation of the drug elution process
An important factor for success in stent design is dosage, that is, the release rate of the drug into the vessel wall, which relates to maintaining concentrations at a therapeutic level while avoiding toxic levels. Stent placement often induces restenosis together with chronic inflammatory response to the presence of a foreign body. The incorporation of cytostatic drugs into DES structures has shown dramatic reduction in the occurrence of restenosis. In a study of cytochalasin D, the in vitro release of the drug from coated stents in buffered saline solution has been studied using the effective combination of HPLC and DSIMS analysis to assess the distribution of drug remaining in the coating system as a function of release time.1
The HPLC data shown in Figure 3 illustrate the cumulative elution of cytochalasin D over a six-week period. There is an initial rapid release of drug over the first few days, followed by slower release over time. After six weeks only approximately 50% of drug had been released into solution. In the DSIMS study, stents were left in the elution medium for time periods of up to six weeks.
Figure 3: HPLC elutionprofileof cytochalasin D from stents
over a six-week period
Figure 4 shows a comparison of the depth profiles for stents with no-elution period and a six week elution period. For the six weeks elution sample, there is a significant reduction in drug level within the first 3 µm of the stent coating. The profile for the no-elution sample shows the as-manufactured concentration of cytochalasin D within the intact coating. DSIMS depth profiling of additional samples with elution times from a few hours to several weeks provided important information on the release behaviour of cytochalasin D in this coating system.
Figure 4: Comparison of cytochalasin D elution from two stents (no elution and six weeks elution), monitored by DSIMS
Imaging techniques for chemical characterisation, surface topography and film thickness
The manufacture of DES is achieved by spray processes and/or dipping to produce an ideally uniform coating of each of the distinct layers within the coating structure. However, the unique shape and morphology of stent platforms and the nature of the coating processes lead to possibilities of poor overall coating uniformity and the potential for non-equilibrium states at the outermost surface of the stent. Imaging techniques, involving SIMS, scanning electron microscopy (SEM) and WLI have been used extensively in the company’s laboratory to investigate the early stages of drug elution, coating uniformity, thickness and long-term failure of coatings.
In DES elution studies an initial rapid release of drug has often been noted. The reason for this lies in an excess drug level within the upper portion of the coating layer compared with deeper regions. The SIMS image from a DES development study (Figure 5) clearly shows evidence for localised high levels of drug, probably present as isolated crystallites that form during the drying stage after spray coating as solvent evaporates.
Figure 5: SIMS imgae of a DES. Bright areas are drug crystallites
monitored as CN.
These crystallites are typically up to approximately 10 µm in size. SEM techniques allow imaging of a range of morphological features on the outer surface of stent coatings. Figure 6 shows cracks that have appeared at stress points on the inner curve of a stent structure. The SEM image in Figure 6a shows the dimensions of cracks and their location around the inner curved region of the stent. The SIMS image in Figure 6b from a similar region of the same stent shows clear evidence of exposure of the Parylene C primer layer by the cracking process, detected as Cl and shown in red in the image.
Figure 6: (a) SEM image showng cracks in the coating structure; (b) SIMSS overlay image of a similar cracked coating region; CI (red), carbon (cyan)
Although SEM and SIMS imaging help to clarify defective structure and identify chemical features on the stent surface, the techniques do not convey any information regarding topography or thickness of the coating. For this WLI is the technique of choice. For materials with transparent coatings in the micron scale the white light illumination provides topographic information from the top surface of the stent coating and the highly reflective platform surface below. With knowledge of the refractive index of the coating material an accurate film thickness measurement can be conducted for research and development or quality assurance purposes.
In the development stent study example in Figures 7 and 8, the top surface of the stent coating exhibits an undulating topography with some evidence of circular features probably resulting from uneven drying after the spray coating process. The thickness profile, calculated from the marked line, shows significant variation in thickness over the range 3 µm to 5 µm and an overall mean value of 3.97 µm. Such wide variations in coating thickness will affect local elution characteristics of the stent. Hence, for quality assurance purposes a multi-point sampling routine is adopted for each sampled stent to provide a statistical representation of coating topography and film thickness.
Figure7: WLI pseudo-colour height map (left) and film thickness profile (right)
The benefit of any topographic technique that samples an x, y, z data array is that the data can be represented as 3D images, as shown in Figure 8, with suitable colour scaling. This allows clearer visualisation of topographic features such as defects and in this case drying spots.
Figure 8: WLI 3D image rom the coated stent showingdrying spot features
Reap the rewards
Techniques such as SIMS, WLI and SEM provide a detailed insight into drug distribution through the coating system, layer thickness and the cleanliness of substrate materials. Companies that regularly use these techniques for characterisation, problem-solving and product development gain significant advantages over competitors in terms of reduced failure rates and time to market. In the future it is hoped that ongoing developments in time of flight SIMS methods will provide the molecular specificity for depth profiling studies, which currently limit conventional SIMS approaches.
1. M.L.P.M. Verhoeven et al., "DSIMS Characterisation of a Drug-Containing Polymer-Coated Cardiovascular Stent,” Journal of Controlled Release, 96, 113–121 (2004).
Dr Alan Brown is Director of Development at CERAM Queens Road,
If you found this article interesting please
Sign Up Now for your free subscription to