Assessment of structure integrity, corrosion behavior and microstructure change of AZ31B stent in porcine coronary arteries

1. Introduction

Coronary artery stents have been well used to be implanted into the stenotic arteries segments to treat the coronary arteriosclerosis. The coronary stent needs to have enough mechanical support, bio-compatibility performances. Currently, stainless steel and Co-based alloy are used as stent materials widely. However, a lot of problems have appeared on stent implanted lumen during long-term clinical applications. For example, late thrombosis and resistance occur after drug-eluting stents implantation, which was recognized as the result of the stent fracture [1,2] and the stents physical force [3]. Therefore, a degradable coronary stent is needed to replace the conventional one. Magnesium (Mg) alloys used for the intravascular stent is a new kind of biodegradable material with better biocompatibility and better mechanical strength than a bio-degradable polymer stent, which is expected to avoid the late intravascular restenosis and late thrombosis happening after stent implantation [[4], [5], [6], [7]]. Meanwhile, some disadvantages for the Mg alloy have to be fixed, and currently, the most important one is uncontrollable degradation rate. Although studies on Mg alloy stent have been carried out for more ten years, however, only “Magmaris” developed by Biotronik company has obtained the CE approval.

Many previous studies indicated that fracture failure of Mg alloy stents was a common issue after implantation in animal test and even in clinical application. Waksman et al. [8] implanted Mg alloy stent and stainless-steel stent into the porcine coronary arteries. It was found that the optical microscopic pictures of the Mg alloy stent were intact at 3 days, but the continuity of the stent struts was lost at 28 days at many areas. Shi et al. [9] showed the Micro-CT 3D reconstruction images of the JDBM Mg alloy stent with surface modification. It was shown that there were many fractures on the stent after 2-month implantation. The AZ31B Mg alloy stents with surface coating were implanted into white rabbit aorta abdominals, and many strut fractures had been observed after 1-month implantation [[10], [11], [12]]. Yang et al. [13] reported that restenosis in Magmaris stents was due to significant collapse. A 54-year-old man with a history of hypertension, diabetes, and tobacco use was admitted for unstable angina. Angiography revealed a 90% diameter stenosis in the proximal left anterior descending coronary artery. Stenting with Magmaris 3.0 mm × 25 mm and 3.5 mm × 15 mm bioresorbable scaffolds (Biotronik, Berlin, Germany) achieved satisfactory results by angiography and intravenous ultrasound. Nine months later, the patient developed recurrent angina at rest. Repeated angiography showed significant“in-stent” restenosis. Intravascular ultrasound and optical coherence tomography demonstrated collapse at the site of Magmaris implantation and a resultant small lumen. The narrowing in luminal dimensions after Mg alloy stent implantation was caused by three factors: 45% of adverse effects were due to neointimal hyperplasia, 42% due to negative remodeling, and 13% due to the expansion of the plaque area at the stent surface. Negative remodeling is responsible for insufficient metallic support for the collapse [14]. However, the reason leading to fracture failure and the happening on different magnesium alloy series are few investigated.

Whether the biodegradable Mg alloy stents can maintain the structural integrity undergoing both multiple deformations and degradation during the fabrication, implantation and service process is an important issue to be studied before their clinical application. In addition, the stent structural integrity will be another reason leading to restenosis for a vascular stent. Therefore, if a Mg alloy intravascular stent is implanted into the blocked vessel segment, the structural integrity of the stent needs to be evaluated.

When a Mg alloy intravascular stent is implanted into the blocked vessel segment, it will be embraced by neointima during one or a half month as a result of migration of vascular smooth muscle cells [12]. This phenomenon makes difficult to detect the stent structural integrity directly by using the routine monitoring methods, because of the lower density of Mg-based materials compared to conventional steel ones. Quantitative coronary angiography (QCA) and optical coherence tomography (OCT) could reveal the degree of in-stent restenosis in living body, which could reflect the possibility of intimal hyperplasia and the location of intravascular stent indirectly [15,16].

High-resolution transmission X-ray tomography (HRTXRT) enables us to nondestructively examine the internal structure of the biodegradable stent at the micron level, even dozens of nanometers [17]. It could detect the holes, cracks and corrosion products formed in/on the samples [18,19]. In this study, HRTXRT was used to detect the structural integrity and local details of Mg alloy intravascular stent during its implantation periods. Furthermore, the degradation of Mg alloy stent was also demonstrated by both HRTXRT and scanning electron microscope (SEM). Then the degradation behaviors of the stent were analyzed by the initial location and composition of degradation product.

Because undergoing multiple deformations cycles on implanted Mg alloy stent, the microstructure change may happen in the stent material especially after implantation in coronary arteries for a set period. However, there were few investigations on microstructure change of Mg alloy stent in the previous literature. To assess the structural integrity, corrosion behavior and microstructure change of magnesium-based stent after implantation into the normal porcine arteries ought to be examined. Al‒Zn series Mg alloy intravascular stents were implanted in the single arm design. The stent was coated by a complex coating, consisting a chemical conversion coating and a biodegradable polymer (PDLLA) coating with a molecular weight of 100,000 in order to improve the corrosion resistance of the stent [20]. The intimal hyperplasia, the structural integrity, the degradation behavior and the microstructures of Mg stent with different implantation periods were analyzed in this study. Under the circumstances, the possible reason leading to the loss of the structural integrity was also inferred. The relevance established in this study can provide important guidance for the design of Mg alloy coronary stent.

2. Experimental

2.1. Stent fabrication

The Mg alloy stent system consisted of a stent crimped on a fast exchange delivery system. The stent was made of a biodegradable Mg alloy (AZ31B) containing 3% aluminum and 1% zinc. This balloon expandable stent was fabricated using a laser cutting facility from φ2.4 mm × 0.2 mm Mg alloy tube which was entrusted to be processed by EUROFLEX GmbH. The strut thickness was approximately 140 μm. It was necessary to obtain a stent with the smooth surface through electrochemical polishing to remove the machining defects on the stent surface. The original configuration and the configuration after crimping are shown in (Fig. 1 (a) and (b)), respectively. The stent was coated by a complex coating to improve the corrosion resistance [20]. In addition, a rapamycin (supplied by North China pharmaceutical) drug-eluting poly (lactide-co-glycolide) (PLGA (75:25), supplied by DURECT Corporation) coating with 100 μg drug loading was fabricated on the outer layer to inhibit the hyperplasia. The coated stent was able to keep the structural integrity after immersion in a phosphate-buffered saline (PBS) solution at 37 ℃ for 28 days (as shown in Fig. 1(d)). The drug-releasing and surface morphology of the Mg alloy stent previously obtained are shown in Fig. 1 (c). The delivery system was based on a fast exchange percutaneous transluminal coronary angioplasty (PTCA) catheter. The expansion of the stent could be accomplished by a balloon at the distal end of the system. Two radiopaque markers were located at both ends of the balloon, and the stent was centered on the expansion balloon between the markers. For a 3-mm diameter stent, the elastic recoil was ˜4% with a radial support force of 11 N.

View original graphic| Download| PPT slide

Fig. 1.   Configurations of Mg alloy stent: (a) original configuration, (b) after premounted onto the balloon, (c) drug release curve, (d) morphology after 28 days immersing.

2.2. Animals

A series of experimental studies were conducted using 6 Chinese miniature porcine of either sex (each two as a group for 1-month, 3-months and 6-months studies) with 25-30 kg weight and 6 to 8 months old, supplied from China Agricultural University, China. The juvenile porcine studies were approved by the Institutional Animal Care and Use Committee of Peking University First Hospital and conformed to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All the animals were fed a standard laboratory chow diet without lipid throughout the course of the study.

2.3. Stenting procedure

A series of eighteen biodegradable Mg alloy intravascular stents were randomly assigned and placed in the left anterior descending (LAD), circumflex (LCX) and right coronary artery (RCA) with precise positioning determined by pre-deployment angiographic measurements to achieve a 1.1:1 to 1.2:1 stent (dilated diameter: 3.0 mm)/artery diameter ratio. During the stents implantation process, the dilatation process was without post-dilatation. The anesthesia and stent implantation procedures were described previously [16].

2.4. Quantitative coronary angiography and optical coherence tomography

Quantitative coronary angiography and optical coherence tomography were carried out at different time points after stent implantation under anesthesia. The coronary artery lumen images and in-stent restenosis were detected by CAAS 5.9 QCA Software (PIEMEDICAL IMAGINE, Netherlands). In addition, an optical coherence tomography (C7-XR Dragonfly System, St. Jude Medical, USA) was used to reflect the cross-sectional lumen area, in-stent restenosis and stent degradation degree. OCT was performed after Mg alloy stents implanted at 28 days, 3 months and 6 months following stenting. To clear blood from the imaging site, the occlusion balloon was inflated and Lactated Ringer's solution was infused in the coronary artery from the distal tip of the occlusion balloon. OCT images were analyzed using proprietary offline software provided by LightLab Imaging.

2.5. Neointima coverage

At the periods of 1, 3 and 6 months, a couple of animals from either sex were sacrificed, the hearts were immediately harvested and the coronary arteries were perfusion-fixed with 10% neutral buffered formalin at 60-80 mm Hg for 30 min via the aortic stump. The stented coronary artery segments were carefully dissected from the epicardial surface of the heart and fixed in a tube with 2.5% glutaraldehyde solution and absorbent cotton added. These segments were dehydrated in 30%, 50%, 70%, 80%, 90% and 100% ethyl alcohol successively and dried to analyze the neointima coverage condition of coronary artery wall with stent implantation by scanning electron microscopy (SEM, Hitachi S-3400 N).

2.6. Stent structural integrity

The structural integrity and local details were examined by using HRTXRT. The HRTXRT examination was performed using an Xradia “Versa XRM-500’’desktop system with the acceleration voltage of 50 kV. The sample for structural integrity analysis and local details was placed between the X-ray source and a 2048 × 2048 pixel array CCD detector equipped with a lens of 0.4 ×(4×). The exposure time was about 5 s (2 s) for each of 1600 projections while the sample was rotated 360× along its vertical axis. The projection data were reconstructed by a filtered back projection algorithm, and then processed and visualized with the software Avizo Fire. Both the geometrical magnification and the optical magnification resulted in a voxel size of 20.752 μm (3.2152 μm).

2.7. Element distribution

The intravascular tissue samples with AZ31B stent for neointima coverage analysis were embedded in epoxy resin to analyze the element distributions in the cross-section of the stent strut by SEM and energy dispersion spectrum (EDS) analysis. Some mesh cross-sections were chosen, and line-scan was carried out on the cross-section. Then the degradation states of the stent after implantation for different periods were evaluated.

2.8. Microstructure analysis

The implanted stents were ground, polished, and etched by an etching agent containing ethanol, picric acid and ethylic acid for 30 s. Microstructures were observed by optical microscopy.

3. Results

3.1. Quantitative coronary angiography and optical coherence tomography imaging

Three AZ31B stents with surface coating were implanted in the left anterior descending (LAD), circumflex (LCX), and right coronary artery (RCA) randomly. The lumen loss (LL) is a quantitative detection index to evaluate the in-stent stenosis of the vessel. The lumen loss was detected by quantitative coronary angiography (QCA). Stent/artery diameter ratio was 1.11:1 actually detected by QCA. Besides that the lumen loss was 0.63 ± 0.42 mm (1-month), 0.73 ± 0.43 mm (3-month) and 0.56 ± 0.30 mm (6-month), respectively (n = 6). Optical coherence tomography (OCT) measurement are shown in Table 1 (n = 2). It was found that the neointimal area became thicker as the implantation time went on.

The stented coronary artery imaging of coronary angiography and optical coherence tomography are shown in Fig. 2. The QCA imaging of coronary artery stenting for 1 month is shown in Fig. 2(a-1). The section of LAD with the stent was stenosis more than those of LCX and RCA. There was neointima coverage on the AZ31B stent, and meanwhile, the vessel wall was uneven and smooth (Fig. 2(a-2)). Combined the results of QCA and OCT, the apparent stenosis was attributed to the neointima.

View original graphic| Download| PPT slide

Fig. 2.   Stenosis degree followed coronary stenting were severe generally. (a-1), (b-1), (c-1) were the QCA imaging of AZ31B stents implanted for 1, 3 and 6 months. Three AZ31B stents implanted into coronary artery. Positions of stents were indicated by yellow marks. As long with implanting period, the degree of stenosis became severe. In-stent stenosis rate was almost 40%-50% at most narrow section. (a-2), (b-2), (c-2) were the OCT imaging of stents implanted for 1, 3 and 6 months. The neointimal thickness and expansion uniformity of the stent were exhibited. The neointimal proliferation degree had little change as vessel stenting prolonged.

3.2. Neointima morphologies

Fig. 3 shows the neointima morphologies of coronary arteries with stent implantation periods. As shown in Fig. 3(a), there was complete neointima coverage on the surface of a stent after 1-month implantation, and each stent strut was covered by neointima. After 3-month and 6-month implantation, neointima was smooth and uniform as shown in Fig. 3(b) and (c)

View original graphic| Download| PPT slide

Fig. 3.   Neointima morphologies after implantation of AZ31B stents in porcine coronary arteries characterized by SEM: (a) 1-month implantation; (b) 3-month implantation; (c) 6-month implantation; (d) 1-month implantation; (e) 3-month implantation; (f) 6-month implantation.

3.3. Structure integrity of the stent

The structural integrity of Mg alloy stent after implantation was characterized by HRTXRT. As shown in Fig. 4(a), the mesh structure of the Mg alloy stent was mostly in integrity. It is indicated that Mg alloy stent kept its structural integrity after deformation and implantation. However, there were three fractural positions on strut mesh after stent implantation for 2 days, which induced reduction of the stent diameter from 3.0 mm at immediate postoperative to 2.8 mm. But there was no obvious corrosion product on Mg alloy substrate and the fractured position. While after 1-month implantation, more fractures were found on Mg alloy stent struts, with a ratio of more than 50%, as shown in Fig. 4(b). A lot of corrosion products could be detected on some fractured positions at 3-month implantation, which was pointed by red arrows. It means that corrosion of the stent struts processed after cracking formation. When the stent was implanted for 6 months, more Mg alloy substrates were converted to the corrosion products showing a change of strut color, as pointed by red arrows in Fig. 4(d). There were fractures distributed on almost all the struts of the stent, however, stent diameter was not reduced further. Stent lumen is also shown in Fig. 4. As the implantation time went on, more degradation products appeared learning from color change.

View original graphic| Download| PPT slide

Fig. 4.   Mesh structures of AZ31B stents after implantation for 2 days (a), 1 month (b), 3 months (c) and 6 months (d). In the HRTXRT imaging, the Mg alloy substrate shows as green color and corrosion product shows as blue color pointed by red arrows. The color difference is attributed to the density difference of substances.

3.4. Crack demonstration

Fig. 5 presents the fracture evolution on Mg alloy stent after implanted into a coronary artery of porcine. As shown in Fig. 5(a), there was no obvious corrosion happened on the stent strut after implantation for 1 month, with a crack initiating on the free outer surface of peak strut. As implantation time went to 3 months, crack initiation processed to crack propagation, leading to fracture. The surface protected coating became invalid as long with a crack formation, and crevice corrosion happened at the site of the fracture, as shown in Fig. 5(b). The Mg alloy substrate on both sides of crack was corroded seriously, as shown by the red circle.

View original graphic| Download| PPT slide

Fig. 5.   Fracture evolution on AZ31B stent characterized by HRTXRT after implantation for 1 month (a) and 3 months (b). Crack initiating crowed by a red circle in (a); Crack propagation and fractured by a red circle in (b).

3.5. Element distribution in stent strut

For studying the corrosion behavior of AZ31B stent, Mg and other elements such as O, Ca and P in degradation products were studied. As shown in Fig. 6, the morphologies of the cross-section of the stent strut and corresponding elements distribution were detected by SEM and the line-scan EDS. Mg, O, Ca and P distributions in the cross-section of the stent strut were analyzed. After the stent was implanted into the coronary artery for 1 month (Fig. 6(a)), the external surface of the stent strut began to degrade. The Mg alloy matrix was transformed into the Mg oxide and the compound containing Ca and P as the degradation products. As the implantation time went on, O, Ca and P distributions expanded to the internal section gradually. As shown in Fig. 6(b), the stent strut has degraded nearly 40%. In this study, the coronary artery samples with stent inside were embedded in epoxy resin, which contained C element, so C was not within the scope of research. In spite of degradation of the stent strut, the macrostructure of the stent struts kept the same as the origin.

View original graphic| Download| PPT slide

Fig. 6.   Cross-section images and EDS line-scan of AZ31B intravascular stents: (a) after 3 months implantation; (b) after 6 months implantation.

3.6. Microstructure change in stent strut

The initial microstructures of the stent before deformation and implantation are shown in Fig. 7(a) and (b). There was no crack and twins in the stent struts. The microstructures of the stent after implantations for 1 month and 3 months are presented in Fig. 7(c)‒(f). After the stent was implanted into coronary arteries for 1 month, there were a lot of twins growing on the edge of the inner strut, as shown in Fig. 7(c) and (d). Meanwhile, the outside of the stent began to degrade on account of the degradation product dissolved by the etching agent. As the implantation time went to 3 months, the Mg alloy stent further degraded, as shown by the red dotted line in Fig. 7(e) and (f).

View original graphic| Download| PPT slide

Fig. 7.   The initial microstructure of a stent unit without deformation and implantation (a, b), and microstructures after implantations for 1 month (c, d) and 3 months (e, f). A lot of twins were found on the stent strut after deformation and implantation, and the twin numbers became larger as long with the implantation went on. At the meanwhile, the grain size became smaller. Degradation products were highlighted by red dotted line. Degradation began from the external to the internal of the stent matrix.

4. Discussion

Coronary stent is transported to the stenosis site in coronary arteries by a balloon dilation catheter in order to prevent the restenosis happening post-percutaneous transluminal coronary angioplasty (PTCA). There were a lot of investigations on the drug-eluting stent mitigating the restenosis rate after stent implantation [11]. Rapamycin is a kind of hyperplasia inhibiting drug, which was used on Mg alloy stent surface in this study. For improving the corrosion resistance, PDLLA was sprayed on the surface of Mg alloy stent with about 10 μm thickness. The spray parameters were selected according to the previous research [21]. Fig. 1(d) presents that there was no breakage on the stent structure after deformation and immersion for 28 days. This indicates that the coating prepared in this study could improve the corrosion resistance and control the drug released rate of the Mg alloy stent.

Eighteen Mg alloy drug-eluting stents were implanted into six young porcine. During implantation and follow up, all the tested animals survived and were healthy. The conventional detection methods used to evaluate the effectiveness of the stent were quantitative coronary angiogram (QCA) and optical coherence tomography (OCT) in the living body. The QCA results reveal that the lumen loss (LL) rose gradually during 3-month implantation and fell later at 6-month implantation, the relative morphologies were shown in Fig. 2 (a)‒(c). OCT results offered neointima area, luminal restenosis and stent position. The results indicate that the intimal hyperplasia was gradually increased during stent implantation in vivo, and stents were still at the post-implanted positions, i.e., no movement (Fig. 2 (a)‒(c)). The coronary artery stent is going to be covered by neointima after implantation in vascular, which contains proliferative smooth muscle cells and extracellular matrix [22]. The stent needs to be covered by re-endothelialization or neointima rapidly, which is an effective way to prevent the restenosis [[23], [24], [25]]. In this study, the stent was completely embraced by neointima after 1-month implantation, and a slightly thicker intima was found along with the implantation (Fig. 2 and Table 1). There was no thrombus on the internal surface of coronary artery lumen. QCA and OCT results are at the same levels with the results obtained by Liu et al [26]. It is turned out that the surface modified Mg alloy stent showed good blood compatibility and histocompatibility.

The causes leading to in-stent stenosis contain the intimal hyperplasia from the host and the structural integrity of the stent [14]. In this study, a drug-eluting coating was used to inhibit the intimal hyperplasia. The highlight of this study was to assess the structural integrity and degradation behaviors of Mg stent after stent implantation in vivo. The intravascular stent would be covered by neointima after implantation in vascular (Fig. 3). It could be difficult to peel off the neointima surrounding the stent strut. Furthermore, the Mg alloy is a kind of low-density material, leading to an invisible performance under X-ray inspection (Digital Subtraction Angiography, DSA) used in clinic researches, compared with a 316 L stainless steel stent [27]. HRTXRT is an effective method to characterize the structural integrity of intravascular stents, including Mg-based, iron-based and cobalt-based materials. Thus HRTXRT could be used to analyze the expansion uniformity, the location of the fracture, fracture mechanism, and the statistics fracture ratio of a stent strut. 3D structures of Mg alloy stents implanted for different periods are shown in Fig. 4. It can be found that there were few fractures on the stent strut after 2-day implantation, keeping good structural integrity after stent dilatation (in Fig. 4(a)). Following that, fracture rate of the stent strut was almost 40% after 1-month implantation (Fig. 4(b)). In addition, the fracture rate was increased continuously with the implantation period prolonging, showing serious lumen loss of Mg alloy stent. This should be the reason why the lumen lost seriously at 3 months than 1 month post-implantation. The same results were obtained in the study by Shi et al. [9]; even the different Mg alloy series were used for stents. So the fracture phenomenon was common for all Mg alloy stents. Further degradation of the strut happened from external to internal and the fracture location, after the stent was implanted for 6 months (Fig. 4(d)). More Mg alloy matrix converted into degradation product after fragments of the stent strut covered by vascular neointima. The bound effective of Mg alloy stent on coronary vessels was relieved at 6 months post-implantation, which decreased the lumen loss rate, showing positive remodeling. There is some difference between the results obtained from the animal test in this study and from clinical test [13], the stent struts collapsed and lumen loss seriously in the clinical test. That should be due to the condition of the coronary arteries vessel. In the clinical test, the lesion vessel needs a stent to support, when the stent fractured, showing lumen lost seriously. However, in the animal test, the vessel will recover elastic after stent fractured, showing negative remodeling.

For analyzing the main reason leading to so many fractures, crack developing progress, degradation progress and microstructure change on Mg alloy stents were further studied. Wu et al. produced the finite element analysis about Mg alloy stent degradation [28,29]. It found that degradation began at the location of maximum principal stress distribution for the bared Mg alloy stents. The location was concentrated mainly at the more deformed locations during stent expansion, which was a weakened section for Mg alloy stent in the static corrosive environment. According to the present results, the inner location with maximum stress should be prior to the outer one to degrade. A crack initiating on the stent strut was detected after 1-month implantation (Fig. 5(a)). There was no corrosion product surrounding the crack location, and the crack originated on the free outer surface of the peak strut. This result might be owing to the corrosion medium (flowing blood) and alternative stress from vascular pulsation simultaneously.

When a Mg alloy coronary stent is implanted into the lesion in arteries, the nominal diameter of the stent is oversized to the lumen diameter with expanded diameter/vascular diameter of 1.1:1 to 1.2:1. The stent undergoes a radial pressure from the coronary artery, which is distributed on the outer side of peak struts. The Mg alloy stent matrix and its protective coating endure this radial pressure with a certain frequency. It is a weak spot on the outer side of the peak struts for the coating by the tensile stress from the coronary artery. Both corrosion medium and alternative stress act on that weak spot simultaneously, leading to the crack initiating on the outer surface (Fig. 5(a)). After the crack formation, the cyclic radial pressure coming from blood vessel pulsating imposed on crack initiation would process for the crack propagation and finally the fracture failure (Fig. 5(b)).

For the surface modified Mg alloy stent, the struts without a stress concentration tended to make a uniform corrosion start and evolve as immersed in vitro (Fig. 1(d)). The corrosion began from the external to the internal gradually on the locations without any cracks (Fig. 6). The Mg alloy matrix was transformed into a loose corrosion product containing Mg oxide and a compound with Ca and P during corrosion process. Under the cyclic radial pressure condition, the stress was concentrated on the outer surface of the peak struts of the stent, which was the sensitive location for the corrosion.

Mg and its alloys generally have a close-packed hexagonal (hcp) structure with poor deformability [30]. As a new stent material, Mg alloys need to undergo a complicated deformation and cyclic radial stress. There was no crack and twins in the stent unit before deformation and implantation. By contrast, after the Mg alloy stent was implanted for 1 month, it was found that there were a lot of twins distributed at the location of stress concentration. As the stent was implanted for 3 months, there were also some twins existed, at the meanwhile, grains became refined compared with 1-month implantation. There was no apparent crack on the selected peak struts (Fig. 7(c)‒(f)), but a certain amount of twins. This is mainly because that twins (or high-order twins) play an important role in plastic deformation of the Mg alloy, which could release the local stress, reduce the crack nucleation and block the crack propagation [31]. The twin, slippage and crack are three approaches to release the stress concentration, which are in competitions with each other [31,32]. For Mg and Mg alloys, there are fewer slip systems for a plastic deformation. So there were a lot of twins existed in the location of stress concentration after stent deformation (Fig. 7). However, the stress concentration location is under compressing stress after stent implantation, which could not lead to crack. On the contrary, the outer surface of peak struts of the stent is under tensile stress, which could lead to protective coating damage, crack formation and propagation.

When a stent is implanted into a coronary artery, the stent will be exposed to the flowing blood and suffer the cyclic pressure from cardiac impulse. Corrosion resistance and mechanical properties of the Mg alloy stent need to be further improved to maintain the stent structural integrity. The plastic deformation ability of surface protective coating and the complex deformation on the stent should be well matched with each other. In addition to this, the stress concentration in the stent after deformation should be reduced, and the radial support force should be strong enough to support the lesion vessel.

5. Conclusion

Above all, the AZ31B Mg alloy stent with surface modified coating possessed enough initial radial support force in the animal test and better biocompatibility. It was important to study the structural integrity, degradation behavior and microstructure change of Mg alloy coronary stent after implantation. After implantation, a large number of fractures happened on the struts of Mg alloy stent prepared in this study, i.e., loss of the structural integrity. Moreover, there was no degradation product at the crack site during the crack formation. However, there were more degradation products at the crack location as implantation time went on. After dilatation and implantation, the microstructure of Mg alloy with large deformation was changed, and the grain size became smaller. On the other hand, there are some limitations in this study: firstly, the single-arm design was a major limitation of this study, the degradable vascular stent was the study subject with the different properties from the conventional inert metal stent, so there was no control group stent chosen in this study; secondly, there were no direct morphologies of fracture, owing to the difficulty to peel off the stent struts from coronary arteries; lastly, the finite element analysis has not been produced to study the stress‒strain distribution after stenting in blood vessel. This study indicates that biodegradable Mg alloy coronary stent should be further optimized by reducing the stress concentration and improving the surface protection to inhibit failure in vivo.

Acknowledgement

This work was financially supported by the National Key Research and Development Program of China (No. 2016YFC1102404).