Strut Thickness and Species-Specific Healing are Key Considerations in Developing Zinc-Based Biodegradable Arterial Stents

Abstract

Zinc-based biodegradable stents offer a promising solution for temporary vascular support, aiding in vessel healing while reducing long-term complications such as thrombosis and restenosis. This study examined three zinc-based alloys: AMZ (Zn-4.0Ag-0.6Mn-0.1Zr), ACMZ (Zn-3.5Ag-0.5Cu-0.7Mn-0.1Zr), and CMM (Zn-1.0Cu-0.6Mn-0.05Mg), all featuring reduced wire diameters of 0.12 mm. The focus was on their microstructure, mechanical performance, thrombogenicity, and in vivo biocompatibility.

The alloys were processed through cold drawing and heat treatment, resulting in ultra-fine-grained structures measuring less than 1 μm. Energy Dispersive Spectroscopy (EDS) and X-ray Diffraction (XRD) analyses confirmed the presence of strengthening intermetallic phases (AgZn₃, CuZn₄, MnZn₁₃), while a uniform elemental distribution indicated solid solution strengthening. Heat treatment enhanced ductility, especially in the AMZ and ACMZ alloys, without compromising tensile strength.

In vitro hemocompatibility assays showed low thrombogenicity for these alloys, with lower fibrin and FXIIa generation compared to clinical control materials. Ex vivo assessments of platelet and fibrin attachment revealed similar results to those observed with pure zinc controls.

In vivo implantation of the stent materials into the abdominal aortas of mice and rats produced distinct species-specific immune responses. Mice exhibited significant inflammation, extracellular matrix degradation, and necrosis, raising potential concerns about their biocompatibility in this model. In contrast, implants in rats resulted in only mild inflammation and limited neointimal growth, with no signs of necrosis. This suggests a more favorable and controlled healing response in rats.

Morphometric analysis of Verhoeff Van Gieson-stained cross-sections indicated that the 0.12 mm wires produced significantly less neointimal growth compared to previously tested 0.25 mm implants in rats, highlighting the advantages of smaller implant dimensions. Overall, these findings support the use of refined zinc-based alloys with smaller struts for next-generation biodegradable stents. However, the inflammatory response observed in mice emphasizes the necessity for careful selection of models in preclinical testing.

1. Introduction

Biodegradable cardiovascular stents represent a promising advancement in the treatment of vascular occlusions. Unlike permanent metallic stents, these devices are designed to provide temporary support to the vessel, allowing for natural healing and reducing the need for long-term antithrombotic therapies [1, 2]. This approach mitigates the risk of late-stage complications such as thrombosis and restenosis, aligning with the growing trend towards bioresorbable materials in medical device design [3].

Among biodegradable metals, zinc (Zn) and its alloys have garnered considerable attention due to their favorable mechanical properties, biocompatibility, and controlled degradation behavior [4]. Unlike magnesium, zinc's corrosion does not produce hydrogen gas, minimizing the risk of gas pocket formation and the associated adverse reactions [5, 6]. Additionally, zinc's degradation rate can be tailored through alloying, allowing for a controlled release of Zn²⁺ ions, which is crucial for minimizing cytotoxicity and promoting endothelialization [7–9].

Recent research has significantly advanced the development of Zn-based biodegradable stents by optimizing both alloy composition and microstructure. For instance, binary Zn–Cu alloys such as Zn-1.5Cu have demonstrated improved hemocompatibility and reduced neointimal growth compared to lower copper content alloys like Zn–0.8Cu [10]. In another study, a 65 μm ultrathin-strut scaffold created by microalloying Zn with lithium achieved excellent radial strength, promoted endothelialization, and limited intimal hyperplasia in porcine arteries [11]. Zn–Cu–Mn stents implanted in porcine models for 18 months showed favorable biosafety, rapid endothelialization, and positive vascular remodeling, maintaining approximately six months of mechanical support and 26% structural retention at 18 months [12]. Furthermore, a Zn–Al–Cu alloy reinforced with titanium carbide (TiC) nanoparticles exhibited improved strength, ductility, and fatigue resistance while preserving favorable corrosion behavior and biocompatibility. TiC also contributed to stabilizing the microstructure during aging, highlighting the potential of nanoparticle-reinforced Zn alloys [13].

As the complexity of zinc alloys increases, the transition from binary and ternary systems to more advanced quaternary and quinary Zn-based alloys provides a superior balance of mechanical performance, corrosion uniformity, and biological response. Each alloying element contributes uniquely to the overall material behavior. Silver (Ag) additions, for example, enhance mechanical properties and corrosion behavior by refining the microstructure through the controlled formation and dissolution of AgZn₃ precipitates [14]. Silver also exhibits high solubility in Zn. However, excessive Ag can lead to over-precipitation, resulting in significant tensile strength loss and pronounced strain softening. Copper (Cu) plays a dual role by refining the microstructure, often promoting equiaxed grain formation, and improving antibacterial performance. Manganese (Mn) and zirconium (Zr) serve as effective grain refiners, enabling ultra-fine microstructures that improve ductility and promote uniform corrosion behavior [15]. Although magnesium (Mg) has limited solubility in Zn (less than 0.1 wt.%), it can act as a grain refiner and contribute to strength enhancement through particle dispersion strengthening mechanisms [16].

In addition to alloying strategies, microstructural engineering and thermal processing are critical approaches to optimize the performance of Zn-based stent materials. The benefits of Mn, Zr, and Cu are further amplified when combined with advanced processing techniques, such as rapid solidification, which has been shown, particularly in the development of the CMM alloy [17], to produce fine, uniform microstructures that achieve a balance of strength, ductility, and controlled degradation. To further refine these properties, thermal treatments are applied to dissolve secondary phase precipitates, restore mechanical integrity, and reduce strain-rate sensitivity. These treatments also promote a more uniform distribution of particles, enabling particle-dispersion strengthening and supporting a well-balanced combination of tensile strength and ductility critical for stent applications.

Despite these advancements, the inflammatory response to Zn-based implants remains a concern. While typically mild, inflammation can persist over time, potentially compromising the long-term performance of the stent [18, 19]. Additionally, the relatively low mechanical strength of pure zinc may limit its efficacy in vascular applications [20]. To address these limitations, alloying strategies that incorporate silver and copper have been shown to not only improve mechanical properties but also offer antibacterial benefits that further support biocompatibility [10, 14]. Another emerging approach to decrease inflammatory responses involves the design of stents with thinner struts. Reducing strut thickness has been shown to mitigate inflammatory responses, enhance healing by improving tissue integration, and lessen the mechanical burden on the vessel wall [21]. This design philosophy aligns with current trends in cardiovascular device development, which emphasize biocompatibility and personalized patient solutions.

The current study builds upon these developments by evaluating the in vitro, in vivo, and ex vivo responses of zinc-based alloys with reduced strut dimensions. By combining select alloying elements and processing methods, this work aims to overcome the limitations of individual components while enhancing the overall performance of Zn-based stents. Specifically, three Zn-based alloy systems, AMZ, ACMZ, and CMM, were investigated, including both Ag-containing and Ag-free compositions. This allowed for direct comparison of mechanical behavior, hemocompatibility, inflammatory response, and tissue integration under clinically relevant conditions. The thin-strut geometry and ultra-fine microstructure used in this study more closely resemble current stent designs than many previous reports.

By comprehensively evaluating the mechanical properties, microstructure, biocompatibility, and degradation behavior of these alloys, this research seeks to identify material formulations that strike a clinically applicable balance between strength and biological performance. The findings have the potential to significantly advance the development of next-generation biodegradable stents, offering improved outcomes and fewer complications for patients undergoing vascular interventions.

2. Methods

Material Preparation, Processing and Characterization

Casting and Processing:

Ternary and quaternary alloys of Zn - 4.0Ag - 0.6Mn - 0.1Zr (AMZ), Zn - 3.5Ag - 0.5Cu - 0.7Mn - 0.1Zr (ACMZ), and Zn - 1.0Cu - 0.6Mn - 0.05Mg (CMM) were formulated and investigated. Our recently reported melt-spun CMM alloy (CMM-MS) with high strength, ductility and uniform corrosion [17] was also prepared for in vivo implantation to investigate its biocompatibility relative to that of the conventionally fabricated CMM alloy counterpart. The as-cast alloys in the shape of cylindrical billets with a diameter of 28 mm were produced by melting stoichiometric quantities of Zn shots (99.995%), and pure alloying metal shots (99.9% purity- Alfa Aesar) in sealed graphite crucibles at 700°C inside a resistance furnace. Annealing at 400°C was conducted for 2h to homogenize the cast structure, followed by water quenching. Annealed samples were subsequently extruded at 310°C with an extrusion ratio of 39:1 to obtain cylindrical rods. The extruded rods of 4 mm in diameter were centerless ground to provide a pristine surface finish. The rods were then multi-pass drawn at room temperature into wires with a diameter of 0.12 mm by Fort Wayne Metals (Fort Wayne, IN) at an extrusion rate of 20% initial strain rate per minute and constitute the as-received condition.

Microstructure Analysis:

The wire samples embedded in resin were ground and finely polished as per the standard metallographic procedures and then etched with a 3% Nital solution (3 mL HNO₃ + 100 mL ethanol). The samples were etched for less than 30 s to avoid over-etching. The microstructure was recorded using Olympus PME 3 inverted metallurgical microscope (Center Valley, PA, USA) with brightfield observation The embedded and polished wires were also carbon coated and then imaged with the Philips XL 40 ESEM (Amsterdam, Netherlands) for microstructure analysis.

Phase Identification:

X-ray diffraction (XRD) was carried out on the Zn alloy wires for phase identification using a ScintagXDS 2000 diffractometer (College Park, MD, USA). The samples were wound around the zero-background holder for the XRD analysis. A Cu Kα radiation of wavelength 1.540562 Å was used to scan in the range of 30° to 75° with a step time of 20 s and step size of 0.03°.

Mechanical Testing:

The tensile testing of the zinc alloys was performed using a Dynamic Mechanical Analyzer (DMA) Q800 DMA with TA Instruments Universal Analysis 2000 data analysis system. The tensile tests were carried out on wire segments with diameter of 0.12 mm and a variable gauge length of 10 ± 2 mm at a strain rate of 0.01 s⁻¹. The tensile properties of the alloy wires were evaluated before and after heat treatment at 320 °C for 120 s to mitigate the effect of work hardening during cold drawing. The heat treatment duration of the zinc wires was limited to 120 s to effectively suppress strain softening. Previous work on Zn alloys annealed at 320 °C for varying durations shows an abrupt transition at ~120 s, where strain softening and strain rate sensitivity are significantly reduced due to dissolution of the secondary phases and grain coarsening [15]. Short-time annealing up to 120 s is therefore sufficient to achieve the desired reduction in strain softening, whereas longer durations can lead to further grain growth and degradation in strength and ductility without additional benefit. Accordingly, 120 s was selected as the optimal duration to minimize strain softening while retaining acceptable mechanical properties.

Elemental Analysis:

The elemental analysis for the samples was carried out by energy-dispersive X-ray spectroscopy (EDS) in PHILIPS XL 40 ESEM. The scanning was done at 500x magnification using an acceleration voltage of 20 kV at a working distance of 12 mm. The elemental composition and distribution in each sample were quantified by scanning different cross-sectional areas.

MicroCT imaging of wires

The region of most extensive corrosion for each wire was imaged with microCT (Phoenix NanoTom M, GE Measurement and Control Solutions, Germany) after 10 days of in vitro corrosion in Hank’s solution at physiological temperature and oxygen/CO₂ conditions. The NanoTom M equipped with a 180-kV/15 W energy micro/nanofocus X-ray tube and a diamond-coated tungsten target was used for all acquisitions. Images were acquired at an 1 μm isotropic voxel size, with a source voltage of 100 kV and a tube current of 40 μA. The exposure time was 500 ms. Imaging was done in mode 0, and 1800 projections were acquired with an average of 1 and skip 0. All acquisitions took ~15 min in fast scan mode. A copper filter of 0.1 mm thickness was placed between the source and the sample. The microCT datasets were reconstructed with the Datos|x software (GE Measurement and Control Solutions). A beam hardening correction (BHC) of 9 was applied. Three different post-processing filters were applied in the reconstruction software: inline median (projections are filtered during reconstruction using a 3 × 3 median filter, which reduces the noise in the volume), ROI-CT filter (increases the resolution of the CT dataset for a selected ROI of the object), and filter volume (volume is filtered in slices after reconstruction using the Unsharp Masking Filter, enhancing object edges). Slices were exported as 16-bit XY slices (.tiff) and then converted to 8-bit (.bmp) images. The micro-CT image-based visualization of the corrosion site was generated using Avizo software (Thermo Fisher Scientific, Bordeaux, France).

In Vitro Thrombogenicity

Plasma was acquired from the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis (SSC/ISTH). SSC plasma was reconstituted with 1 mL of 37°C DI water and diluted with 1 mL of 37°C Dulbecco’s phosphate buffered saline (dPBS). After reconstitution, SSC plasma was kept on a 37°C heat block to maintain physiological temperature. The as-received zinc alloy wires (AMZ, ACMZ, CMM) were tested. Controls of pure Zn wire (0.125mm diameter), CoCr (Co40-Cr20-Fe15-Ni15-Mo7-Mn2-C-Be, 0.25 mm diameter) and stainless steel (0.125mm diameter) were acquired from Goodfellow (Cambridge, England). S-2302 Factor (F)XIIa chromogenic substrate was acquired by Fisher Scientific (Waltham, MA, USA). dPBS was acquired from Sigma (St Louis, MO, USA).

FXIIa Generation Assay:

6mm length wires were cut and sonicated for 10 minutes in 100% EtOH to reduce physical debris. The wires were added to a transparent polystyrene 96-well plate and rinsed twice with room temperature dPBS before the addition of SSC plasma. All conditions were given 2.4 mM S-2302. The color change from the Factor (F)XIIa chromogenic substrate (Enzyme Research Laboratories, South Bend, IN, USA) in each well was read by a TECAN (Männedorf, Switzerland) infinite series plate reader at 405 nm each minute for a total of 1 hr. The max slope of the change in absorbance was calculated by the Magellan 7.2 software. Each assay was repeated 3 times with 3 different wires resulting in n=9.

Fibrin Generation Assay:

6mm length wires were cut and sonicated for 10 minutes before addition to the plate. Wires were rinsed 2 times with room temperature dPBS as described above. SSC plasma was added to the plate, and afterwards, 25mM CaCl₂ was added to all wells, such that the final concentration was 8.3 mM CaCl₂. The plate was read by Tecan infinite series plate reader, taking a 405nm reading at each minute for one hour. In this assay, 405nm represents the turbidity of the solution. The take-off time for fibrin clotting was calculated by Magellan 7.2 software, based on a 10% increase in solution absorbance compared to baseline. This assay was repeated three times in triplicate, resulting in 3 total trials and n=9 samples.

Ex Vivo Thrombogenicity testing in non-human primate (NHP)

Wire lengths of 25 cm with diameter of 0.12 – 0.125mm were coiled on a 4 mm outer diameter rod to a length of 2 cm and covered with a fluorinated ethylene propylene (Zeus Inc., Orangeburg, SC) heat shrink tubing that was shrunk to a 4mm outer diameter rod. Samples were placed in a chronic arteriovenous shunt loop, as described previously [22] and summarized here (see Supplemental Figure 1). A single, juvenile male baboon (papio anubus, 12kg) was used in these studies under the care of the veterinary staff at the Oregon National Primate Research Center (ONPRC), which follows guidelines set by the National Research Council and maintains Association for Assessment and Accreditation of Laboratory Animal Care certification. Studies were approved by the ONPRC Institutional Animal Care and Use Committee. Blood flowed over the wire coil for 1 hr. Flow was measured with an upstream flow probe (Transonics, Inc) and maintained at 100 mL/min with a downstream clamp. No anticoagulant or antiplatelet therapies were used throughout this study. Autologous platelets were labelled with In-111, and quantitative platelet attachment was measured dynamically every 1 min on a Brivo NM615 (General Electric, Boston, MA). Homologous fibrinogen was labelled with I-125 and quantified as an endpoint value with a gamma counter (PerkinElmer 1480 Wizard, Waltham, MA). Five separate wire coils were tested on each sample type (AMZ, ACMZ, CMM, and pure Zn) for platelet and fibrin quantification. Platelet data were analyzed with a 1-way repeated measures ANOVA. Fibrin data were measured with a 1-way ANOVA. To obtain qualitative thrombus data, 2 straight wires from each group were placed in the shunt loop downstream of the coil (see Supplemental Figure 1). SEM samples were fixed in 4% PFA overnight, rinsed with PBS, and dehydrated in increasing ethanol/water concentrations. After dehydration in hexamethyldisilazane and coating with gold on an ACE600 (Leica), thrombi were imaged with a Volumescope 2 SEM (Thermo Scientific, Waltham, MA) using high vacuum. Whole blood draws were taken from the animal before the study as a bulk draw, immediately downstream of the coil continuously from 0–30 min (labelled 30min), and immediately downstream of the coil continuously from 0–60 min (labelled 60min). These downstream blood samples provide a more concentrated content of blood factors from closer interaction with the wire [22, 23].

Downstream Shunt Assays:

Plasma collected from the downstream blood sampling was frozen at −80°C immediately after centrifugation (3min centrifugation at 10000 rpm). Plasma was thawed on wet ice before being analyzed by the myeloperoxidase (MPO), platelet factor 4 (PF4), and thrombin anti-thrombin (TAT) assays. MPO was measured through manufacturer’s instructions using a fluorescent detection kit from Cell Technology (Hayward, California). PF4 and TAT were measured by Enzyme-Linked Immunosorbent Assay (ELISA) kits, according to manufacturer’s instruction. PF4 was measured with the PF4 ELISA Kit from R&D Systems (Minneapolis, Minnesota) while TAT was measured with the Siemens Healthineers (Erlangen, Germany) TAT ELISA kit. Plasma for the MPO assay was diluted 1:1 with buffer, while plasma for the PF4 ELISA was diluted 1:500 with buffer. Finally, TAT plasma was diluted 1:4, 1:16, and 1:128 depending on timepoint. Half of the samples at the 60 minute time point were higher than the assay can register. Those samples have been denoted as the max concentration possible for the TAT assay, 60 ug/L.

In Vivo Studies on Biodegradation and Biocompatibility

Wire samples were implanted into C57BL6 mice (n=20 per group) and Sprague Dawley rats (n=6 per group) for a period of three months. Samples were sterilized in ethylene oxide and maintained under sterile conditions until implantation. All animal experiments were approved by the Michigan Technological University Institutional Animal Care and Use Committee (IACUC) board.

Implantation and Removal:

0.12mm diameter wires were implanted into the abdominal aortas of C57BL/6 mice and Sprague Dawley rats, as previously reported [8, 24]. After three months, the animals were euthanized, and the aortas were removed. All samples were flash frozen in optimal cutting temperature (OCT) medium (Thermo Fisher Scientific) and liquid nitrogen before being stored at −80℃ until cross-sectioning.

All explants were cross-sectioned at a temperature of −25℃ and a thickness of 9 μm using an HM525 NX Thermo Fisher Scientific cryo-microtome. Cross-sections were collected once the implant had advanced through the arterial elastic lamina. The cross-sections were placed on Histobond slides (VWR) and stored at −80℃ to await staining.

Hematoxylin and Eosin (H&E) staining/Cell Morphology:

Hematoxylin and eosin (H&E) staining was performed to qualitatively assess neointimal cell composition and distribution. Tissue cross-sections were fixed in 10% neutral buffered formalin (Sigma Aldrich), then rinsed three times in phosphate-buffered saline (PBS) for 5 minutes each. Following a 5-minute rinse in deionized water, sections were stained with Gill’s Hematoxylin #3 (Sigma Aldrich) for 5 minutes. Excess stain was regressed using diluted hydrochloric acid (pH 1.8–2.0), followed by a 5-minute bluing step under running tap water. Slides were then washed in two changes of 95% ethanol (5 minutes each), counterstained with Eosin Y working solution for 30 seconds, and regressed again with 95% ethanol. Sections were subsequently dehydrated in two changes of 100% ethanol (5 minutes each), cleared in xylene substitute, and then incubated in two fresh changes of xylene substitute (Sigma Aldrich) for 5 minutes each. Finally, slides were mounted using Eukitt quick-hardening mounting medium and imaged at 40× magnification using a Zeiss Axioscan Z1 (Oberkochen, Germany).

Verhoeff van Gieson (VVG) Staining:

Verhoeff-Van Gieson (VVG) staining was performed to quantitatively assess wire-to-lumen thickness (WLT) and neointimal area (NA). Tissue cross-sections were fixed in 10% neutral buffered formalin for 15 minutes, followed by three rinses in phosphate-buffered saline (PBS) for 5 minutes each. Sections were then stained with Verhoeff’s solution for 8 minutes and individually differentiated in ferric acid. Slides were rinsed in deionized water for 5 minutes, treated with sodium thiosulfate for 1 minute, and “blued” in running tap water for 5 minutes. A counterstain with van Gieson solution was applied for 4 minutes, followed by regression in 95% ethanol. Sections were dehydrated in two changes of 100% ethanol (5 minutes each), cleared in xylene substitute, and then incubated in two fresh changes of xylene substitute. Finally, slides were cover-slipped and imaged at 20× magnification using a Zeiss Axioscan Z1.

Statistical Analysis:

All data are presented as mean ± standard deviation. In vitro thrombogenicity tests and ex vivo fibrin data were tested for (ANOVA) assumptions using Brown-Forsythe tests and quantile-quantile plots in GraphPad Prism (v. 10.4.1). Data sets that satisfied ANOVA assumptions were tested using a 1-way ANOVA with Greenhouse–Geisser correction and a Tukey’s post hoc was used when the main effects test was significant (p<0.05). TAT and PF4 data did not meet ANOVA assumptions and were log-transformed to allow for valid statistical analysis and are shown in the figures after transformation. Data from TAT, PF4, and MPO assays were excluded if they fell outside the assay’s standard range or if plasma discoloration indicated invalid results. Ex vivo platelet data were analyzed with repeated measures 1-way ANOVA in SPSS (IBM, version 27). For in vivo experiments, statistical analyses were performed in R (v. 2023.06.0). Data were tested for normality using the Shapiro-Wilk test and for equal variance using Bartlett’s test A Kruskal Wallis test followed by Dunn’s post hoc test was used for non-normal data, while analysis of variance (ANOVA) was used for normally distributed data.

3. Results

Materials Analysis

Microstructure

Figure 1 shows the cross-sections of the different alloys before and after etching. The CMM alloys exhibited minor central defects in the cross-section. The optical micrographs of the wires after etching exhibited ultra-fine grain sized microstructure as a result of the grain refinement due to severe plastic deformation during cold-drawing. The ultra-fine-grained features can be attributed to the severe plastic deformation imposed during cold drawing. Such deformation processes are well known to induce significant grain refinement, dislocation accumulation, and crystallographic texture evolution during cold drawing in Zn-based alloys [25, 26].

Figure 1.

Optical micrographs of Zn wires before and after etching for (a) AMZ, (b) ACMZ, and (c) CMM. Multiple wires of each alloy were imaged prior to etching to ensure reproducibility.

Elemental composition by EDS analysis

The quantitative analysis from the EDS results for each alloy is shown in Figure 2, with the corresponding data summarized in Table 1. The analysis was performed on two marked regions in the SEM images of each alloy. The results closely matched the theoretical alloy compositions, with the exception of minor alloying elements such as Zr in the AMZ and ACMZ alloys, and Mg in the CMM alloy. These discrepancies are likely due to the very low concentrations of these elements (approximately 0.1 wt.% or less), which are near the detection limit of EDS.

Figure 2.

SEM images of alloys and corresponding EDS layered image and elemental maps: a) AMZ, b) ACMZ, and c) CMM. The circular markers represent the specific regions where point EDS measurements were carried out to confirm the elemental composition (results summarized in Table 2).

Table 1.

EDS elemental composition of alloys.

Alloy		Cu	Mn	Zr	Ag	Mg	Zn
AMZ	EDS result	-	0.5	<0.1	4.1	-	Balance
	Intended composition	-	0.6	0.1	4.0	-	Balance
ACMZ	EDS result	0.6	0.7	<0.1	3.9	-	Balance
	Intended composition	0.5	0.7	0.1	3.8	-	Balance
CMM	EDS result	0.9	0.6	-	-	0.02	Balance
	Intended composition	1.0	0.6	-	-	0.05	Balance

The EDS elemental maps of all the alloys show that all elements are uniformly distributed throughout the AMZ and ACMZ alloys. This uniform dispersion of alloying elements can contribute to solid solution strengthening, thereby enhancing the mechanical strength. In contrast, the CMM alloy exhibited defects at the center, which were also evident in the corresponding SEM images, and are attributed to drawing-induced damage factors. Specifically, inhomogeneities such as segregation or clustering of second-phase particles at the billet center act as preferential sites for void nucleation, which under triaxial stress conditions during wire drawing can evolve into central-burst defects [27, 28].

XRD

The XRD patterns of all samples are presented in Figure 3. The analysis confirms the presence of the Zn phase in all alloys, along with the formation of additional phases corresponding to specific alloying elements. Interestingly, a prominent peak near 36° (dashed arrow), associated with the (002) plane of the Zn phase, was absent in all samples. This plane is typically more exposed in the direction of plastic deformation, likely explaining its suppression due to the cold-drawing process. The AMZ alloy shows additional peaks corresponding to where AgZn₃ was observed. No peaks related to Mn were observed in AMZ alloys, likely because the Mn content in the ternary alloy is very low and the expected MnZn₁₃ intermetallic has relatively weak diffraction peaks compared to other phases. However, compositional analysis confirmed the presence of Mn in the alloy. Similarly, the ACMZ alloy also exhibited peaks corresponding to AgZn₃, along with additional peaks identified as CuZn₄. No peaks related to Zr were detected in ACMZ alloys, which is expected due to its very low concentration (0.1 wt.%). For the CMM alloy, the XRD pattern shows additional peaks near 41°, which correspond to the high-intensity peaks of MnZn₁₃. Peaks corresponding to CuZn₄ were also observed in CMM alloys. Again, no phase related to Mg was identified in CMM alloys, as its concentration is extremely low (0.05 wt.%).

Figure 3.

XRD pattern of AMZ, ACMZ, and CMM alloys.

From a materials performance perspective, the presence of AgZn₃ and CuZn₄ intermetallics is likely to enhance strength through precipitation hardening. However, these phases can also introduce galvanic couples, potentially decreasing corrosion resistance, especially in physiological environments. Similarly, MnZn₁₃, although strengthening, may lead to localized corrosion if not uniformly distributed. The absence of a detectable Zr phase suggests that Zr remains in solid solution, potentially contributing to solid solution strengthening without significantly impacting corrosion behavior.

Mechanical properties

The representative stress-strain curves obtained from tensile testing are shown in Figure 4. A noticeable change in mechanical behavior is observed for CMM after heat treatment, characterized by a significant increase in ductility accompanied by a reduction in strength. This change is attributed to the mitigation of work hardening effects introduced during continuous cold drawing, as a result of the applied heat treatment. The tensile test results, summarized in Table 2, confirm that heat treatment improved the ductility of the CMM alloy while reducing its overall strength. In contrast, heat treatment of the AMZ and ACMZ alloys led to an improvement in tensile performance. The as-drawn alloys underwent severe plastic deformation during cold drawing, which resulted in significant strain hardening and grain refinement. Furthermore, the higher silver content may have led to the formation of intermetallic phases, such as AgZn₃, along grain boundaries, potentially degrading the tensile properties. Heat treatment can improve mechanical behavior by alleviating strain hardening, promoting grain growth, and enabling more uniform dissolution of intermetallic phases throughout the alloy matrix. Additionally, the uniform distribution of these intermetallic phases may contribute to particle dispersion strengthening. As a result, the heat-treated alloys resulted in superior tensile performance for the AMZ and ACMZ alloys.

Figure 4.

Examples of mechanical stress-strain curves for all alloys before and after 120 s heat treatment..

Table 2.

Engineering the tensile properties of the CMM alloy after heat treatment

Alloy	Heat treatment time (seconds)	Yield strength at 0.2 % strain (MPa)	Ultimate tensile strength (MPa)	Elongation (%)
AMZ	0 s	96 ± 1	102 ± 1	15 ± 3
	120 s	92 ± 1	125 ± 1	32 ± 6
ACMZ	0 s	103 ± 1	113 ± 1	16 ± 1
	120 s	116 ± 4	163 ± 3	19 ± 3
CMM	0 s	259 ± 13	300 ± 4	6 ± 2
	120 s	206 ± 3	211 ± 1	24 ± 7

It is evident that heat treatment improved the ductility across all alloys. This improvement can be attributed to the fact that cold drawing induced work hardening and produced fine, highly deformed grains, while heat treatment alleviated the effects of work hardening, promoted grain growth, and facilitated a more uniform distribution of secondary phases. However, the CMM alloy showed a notable reduction in tensile strength following heat treatment. This decrease may be attributed to the presence of internal defects in the CMM alloy, as observed in the SEM image.

In vitro corrosion

Wire samples underwent in vitro corrosion testing for 10 days, after which micro-CT was used to scan the most corroded region along each wire. At this time point, the CMM alloy exhibited the most extensive degradation, characterized by localized surface corrosion and deep internal penetration. In contrast, the ACMZ and AMZ wires showed localized corrosion but no evidence of internal attack throughout the wire cross-section (Figure 5).

Figure 5.

Micro-CT images showing the most severely corroded region along a ~50 mm wire segment for each alloy after 10 days of in vitro corrosion. Each panel includes a 3D reconstruction at a fracture site (top of each wire) and a corresponding optical cross-section highlighting internal corrosion features. A) ACMZ, B) AMZ, C) CMM. All wires are 0.12mm in diameter.

Thrombogenicity Testing of Zinc Alloys

In vitro hemocompatibility testing of the alloys quantified the material thrombogenicity through the intrinsic pathway FXIIa generation rate (Fig. 6A) and fibrin clotting time (Fig. 6B). FXIIa generation (Fig. 6A) was not significantly different between zinc alloys, indicating no effect on FXII activation from the alloying elements. All materials tested had a significantly lower FXIIa generation rate than stainless steel (SS). Both AMZ and ACMZ did not have a significantly higher FXIIa generation rate than plasma alone, while Zn, CMM, and CoCr did (p < 0.05, p < 0.005, and p<0.01, respectively). No materials caused any significant different times to the initiation of fibrin formation compared to each other or plasma alone (Fig. 6B).

Figure 6.

In vitro analysis of zinc alloy thrombogenicity compared to clinical controls. A) Zinc alloys, plasma only, and CoCr had significantly reduced FXIIa substrate cleavage when compared to SS. There were no significant differences between zinc alloys and pure zinc. Plasma alone had significantly reduced FXIIa compared to Zinc (p < 0.05), CMM (p < 0.005), and CoCr (p < 0.01). Significance of materials to SS ranges from p < 0.05 (CoCr, CMM), p < 0.01 (Zn, AMZ, ACMZ), to p < 0.001 (plasma only). B) There were no significant differences for any materials compared to plasma alone in the fibrin clotting time assay. Alloying elements had no effect on Zn materials in this context and Zn alloys performed similarly to clinical controls. Data were analyzed with one-way ANOVA and Tukey post-hoc test.

Ex vivo thrombogenicity (Figure 7 and Supplemental Figure 2) showed thrombus developing on all Zn alloys and pure Zn without any anticoagulant or antiplatelet therapies. Ex vivo results supported these findings, with thrombus formation observed on all samples, but no significant differences in platelet attachment or fibrin content were detected between any of the metals tested. SEM imaging (Fig 7B) confirmed the presence of typical thrombus components, including platelets, fibrin, red blood cells, and white blood cells, on all materials. Platelet attachment increased over time on the wire coils (Fig 7C) with no significant differences between the metals. Fibrin quantification (Fig 7D) was also not significantly different between the wires. Sampling blood downstream of the zinc alloys indicated similar thrombogenicity (Fig. 8A, Fig. 8B) and immune activity (Fig. 8C) of alloys after exposure to the ex vivo shunt loop. Thrombogenicity, including platelet activation, measured via PF4 release, and plasma TAT levels, showed no significant material-dependent differences, although both markers increased significantly over time (Fig. 8A and Fig. 8B). The immune response, measured via myeloperoxidase activity were similar between materials with a significant increase over time (Fig. 8C). Together with the in vitro results, data suggest that while zinc alloys elicit some degree of thrombotic and immune activation, their performance is comparable across compositions and generally more favorable than SS in terms of initiating coagulation. These findings for zinc alloys are similar to our recent findings comparing pure zinc across a broad range of biodegradable and conventional metals [25].

Figure 7.

Wire coils (n=5) after 1hr exposure to whole, flowing, non-anticoagulated blood in a non human primate ex vivo shunt (A) Cross-section images show proximal (left) and distal (right) ends. SEM imaging (B) showed the presence of considerable platelets, fibrin, and red blood cells (n=2). Platelet quantification over time (C) and fibrin attachment (D) showed no significant difference between the tested Zn alloys and pure zinc controls. Platelet data were analyzed with 1-way repeated measures ANOVA, and fibrin data with 1-way ANOVA. SEM scale bar = 30um

Figure 8.

Downstream shunt sample analysis for thrombogenicity and immune activation. A) PF4 concentration of plasma analyzed with ELISA between materials at different time points. No material was significantly different than other materials, indicating a shared thrombogenicity response of zinc alloys. PF4 concentration increased significantly at each time point, with significant differences between 0 min and 30 min, and between 30 min and 60 min. B) TAT concentration from shunt plasma samples. There was no significant difference between materials in the TAT ELISA. TAT concentration increased significantly over time, with significant differences between the 0 min and 30 min time points and between the 30 min and 60 min time points. Values in the TAT that were over the calibration curve were noted as the maximum possible value, due to the high concentration of TAT in multiple samples. C) MPO activity of plasma from shunt loop. There was no significant difference between materials in the MPO activity assay. MPO activity increased significantly from 0 min to 30 min and from 0 min to 60 min. Data was analyzed with two-way ANOVA and mixed effects analysis. *** = p < 0.01. **** = p < 0.001

In Vivo Studies on Biodegradation and Biocompatibility

Structural Changes

Hematoxylin and eosin (H&E) staining was used to evaluate arterial remodeling in response to biodegradable metal implantation. In mouse arterial implants (Figure 9), all samples exhibited pronounced inflammatory responses, marked by extensive infiltration of inflammatory cells and severe disruption of the native extracellular matrix. High magnification views showed that inflammatory cell infiltration was predominantly localized to the adventitial region, with comparatively limited involvement of the medial and neointimal layers, consistent with a localized foreign body-type response rather than a luminal inflammatory process. Regions of necrosis were consistently observed, and in many cases, neointimal tissue overgrowth extended into the arterial lumen, resulting in partial obstruction. All three alloys elicited the presence of small, round, closely packed nuclei with high uniformity (Figure 9) and, together with the absence of foreign body giant cells and negative CD68 and iNOS staining (data not shown), pointing towards a lymphocyte infiltrate.

Figure 9:

Representative Hematoxylin and Eosin (H&E) stained images for each alloy type A-C, taken three months post-implantation of 120 μm diameter wires in mice, along with magnified views (from blue square) of representative inflammatory infiltrates E-F. The asterisk indicates the location of the wire implant. Inflammatory responses were comparable across panels A&D - ACMZ, B&E-AMZ, and C&F-CMM. Scale bar in panel C represents 200 μm. n = 20 per group.

In stark contrast, the same zinc alloys and dimensions implanted in rat arteries (Figure 10) elicited only mild inflammatory cell infiltration and limited neointimal formation. In contrast to the severe lymphocyte-dominated inflammation observed in mice, the rat model exhibited a more moderate and localized response, characterized by a dense accumulation of inflammatory cells at the implant interface. Based on morphological assessment and without direct immunophenotypic characterization, this pattern is more consistent with a typical macrophage-driven foreign body response [29–31] and has been observed in our previous studies with zinc implants in rats [8, 14, 18, 32, 33]. Critically, no necrotic regions were present in the rat arterial cross-sections. Instead, a pro-healing and integrative tissue response predominated in the rat model.

Figure 10:

(A - D) Representative Hematoxylin and Eosin (H&E) stained images for each alloy type, three months post-implantation of 120 μm diameter wires in rats. The asterisk indicates the location of the wire implant. Blue boxes indicate regions shown at higher magnification in panels E - H. Scale bar in panel D represents 200 μm and applies to panels A - D. (E - H) High-magnification images highlighting potential inflammation within the artery. Inflammation is observed in the adventitia for AMZ, CMM, and CMM-MS, while both adventitial and neointimal inflammation are present in ACMZ and CMM. Scale bar in panel H represents 100 μm and applies to panels E - H. n = 6 per group.

This was also the case for the CMM-MS implant, with neointimal characteristics similar to those of the conventional CMM counterpart. Pure Zn controls, in the form of thicker (0.25 mm) struts, have been extensively characterized in previous studies [8, 14]. The reported tissue response to pure zinc demonstrated a pro-healing and integrative profile with limited tissue overgrowth, comparable to the responses observed for the zinc alloys in the rat model. The biological responses observed for the CMM alloy were comparable to those of the two Ag-containing alloys, suggesting that differences in metallic ion release did not significantly influence the overall biocompatibility in this study. While metallic ions are known to affect biological outcomes, the similar tissue responses across all alloys indicate that potential differences in ion release were not a dominant factor differentiating their in vivo behavior.

The melt-spun variant of the CMM alloy (CMM-MS), previously reported in detail [17], was produced using rapid solidification followed by compaction and extrusion, resulting in a markedly refined and homogeneous microstructure compared to conventionally processed CMM. Melt spinning generates cooling rates on the order of 10⁴–10⁶ K/s, which suppress microsegregation and produce a fine-grained structure with uniformly distributed nanoscale precipitates. In contrast to the bimodal grain structure and central defects observed in the conventionally processed CMM alloy, the CMM-MS material exhibits equiaxed grains on the order of ~1 – 5 μm with minimal porosity and a more uniform phase distribution. These microstructural features result in improved mechanical performance, with reported ultimate tensile strength exceeding 320 MPa and elongation approaching 25–30%, reflecting a favorable strength-ductility synergy. Additionally, the homogeneous microstructure promotes more uniform corrosion behavior and reduces susceptibility to localized internal degradation. These characteristics motivated the inclusion of CMM-MS in the present in vivo study as a defect-minimized counterpart to the conventionally processed CMM alloy.

Morphometrics

Neointimal area (NA) and wire-to-lumen thickness (WLT) were quantified from Verhoeff-Van Gieson (VVG) stained images (Figure 11A–D). Across all materials, neointimal tissue growth in the rat model was comparable (Figures 11E and 11F). However, implants with a 0.12 mm diameter exhibited a significantly reduced neointimal area compared to previously tested 0.25 mm diameter implants [8, 14] (Figure 11H) (p<0.05), highlighting the enhanced biocompatibility associated with reduced implant dimensions. The results also demonstrate a stronger effect of implant dimension vs. zinc alloy composition or processing conditions on the neointimal area and thickness. The neointimal morphometrics are summarized in Table 3.

Figure 11:

(A–D) Representative Verhoeff–Van Gieson (VVG) stained images for each alloy type, three months post-implantation of 120 μm diameter wires in rats. The asterisk indicates the location of the wire implant. Scale bar in panel D represents 200 μm and applies to panels A–D. (E–F) Quantitative analysis of wire-to-lumen thickness (WLT) and neointimal area (NA), shown as mean ± standard deviation. No statistically significant differences were observed between groups (Kruskal–Wallis test, n = 6 per group). (G–H) Comparison of WLT among pooled previously published alloys with 250 μm diameter wires, pure zinc (250 μm), and pooled alloys reported here with 120 μm diameter wires. No statistical difference was found between the 120 μm groups; however, the previously reported pooled alloys (250 μm) differed significantly from both pure zinc (250 μm) and the pooled 120 μm alloys (p < 0.05, Kruskal–Wallis test). Group sizes: Pooled previously published [8, 14] Zn Alloys (0.25 mm), n = 118; Pure Zn (0.25 mm), n = 10; Pooled current Zn Alloys (0.12 mm), n = 24.

Table 3.

Average and standard deviations for in vivo neointimal morphometrics (WLT and NA)

Diameter	Alloy	WLT (μm)	NA (mm2)
0.12 mm	AMZ	53 ± 17	0.054 ± 0.006
	ACMZ	46 ± 30	0.053 ± 0.003
	CMM	43 ± 31	0.044 ± 0.017
	CMM-MS	63 ± 40	0.049 ± 0.025
0.25 mm	Pure Zn	67 ± 36	0.074 ±0.026
	Zn Alloys	54 ± 35	0.092 ± 0.067

4. Discussion

The present study provides a comprehensive examination of the structural, mechanical, and biological performance of three Zn-based alloys (AMZ, ACMZ, and CMM) intended for biodegradable vascular implants. While all three alloys exhibited refined grain structures due to cold drawing, their performance diverged significantly based on microstructural integrity, phase composition, and alloying strategies. Beyond reporting these findings, it is essential to delve into the mechanisms and broader implications they reveal.

One of the central findings is the impact of alloy design and processing on both grain structure and mechanical behavior. The ultra-fine grains observed in AMZ and ACMZ (∼750 nm) reflect effective grain boundary pinning and recrystallization suppression from combined alloying and plastic deformation. The higher silver content likely contributed to precipitation of AgZn₃ along grain boundaries, a phenomenon known to enhance strength via precipitation hardening but has been known to compromise ductility and corrosion performance [21]. The absence of detectable Zr and Mg phases supports the hypothesis that these elements remain in solid solution, enabling solid-solution strengthening without the deleterious effects of galvanic corrosion. By contrast, the CMM alloy, with elongated grains and central defects, highlights the trade-offs between alloy complexity and processing-induced defects. These defects likely act as stress concentrators, weakening mechanical performance and highlighting the necessity for additional processing routes. The findings suggest that AMZ and ACMZ alloys in their as-drawn conditions are better suited for biomedical applications due to their refined microstructures and lack of defects, while the CMM alloy requires further optimization to improve its mechanical properties and address defect-related issues, as we have done in a recent work [17].

The presence of central defects in the CMM alloy, likely introduced during processing, limits direct comparison with the defect-free AMZ and ACMZ systems and highlights the importance of microstructural integrity in Zn-based implants. While these defects influenced mechanical and corrosion behavior, all alloys exhibited similar biological responses in the rat model over 3 months, suggesting that early host response is relatively robust to these differences.

However, this study does not capture late-stage degradation or long-term thrombogenic risk. Although prior work on zinc implants suggests sustained biocompatibility during gradual resorption, extended-duration and large-animal studies are required to determine whether these early favorable responses translate to safe terminal degradation without thrombotic complications.

Micro-CT imaging after 10 days of in vitro corrosion testing revealed substantial differences in corrosion morphology among the three alloys. The CMM alloy showed the most pronounced degradation, including internal corrosion features likely related to processing-induced voids or heterogeneities that act as localized initiation sites. In contrast, AMZ and ACMZ exhibited more uniform, surface-localized corrosion with no evidence of internal degradation, consistent with their refined grain structures and lack of structural defects. While these differences highlight the importance of processing and microstructural integrity in influencing corrosion behavior, they did not translate into major differences in biocompatibility across the alloys in vivo. All three materials elicited comparable biological responses in both mouse and rat models, suggesting that within the scope of this study, localized corrosion severity was not the dominant driver of tissue response. These findings underscore the importance of decoupling degradation morphology from biological compatibility, at least in early-stage evaluations, and point to the need for further studies linking degradation kinetics, corrosion byproducts, and immune activation over longer time scales.

The study also provides insight into the interplay between material properties and thrombogenic behavior. All zinc alloys demonstrated reduced FXIIa activation relative to stainless steel, suggesting inherent hemocompatibility. This is likely tied to zinc’s known role in modulating platelet activity and inflammation. However, subtle differences between alloys, such as the slightly elevated FXIIa levels in CMM, may stem from differences in surface roughness, intermetallic phase exposure, or defect density, all of which can alter protein adsorption and cellular interactions. While heat treatment improved ductility and intermetallic phase distribution, it may also influence surface chemistry in ways that affect biological response, a topic warranting further mechanistic exploration. Together, these results suggest that while zinc alloys elicit some degree of thrombotic and immune activation, their performance is comparable across compositions and generally more favorable than SS in terms of initiating coagulation. These findings for zinc alloys are similar to our recent findings comparing pure zinc across a broad range of biodegradable and conventional metals [22].

The in vivo evaluation of zinc-based stents revealed distinct differences in biocompatibility and structural changes between mouse and rat models. This is the first report, to our knowledge, of zinc implants in the arteries of mice. H&E staining demonstrated that while all materials triggered a pronounced inflammatory response in mouse arteries (characterized by extensive immune cell infiltration, extracellular matrix degradation, and necrosis), the same implants elicited a far more favorable response in rats. In the rat model, inflammation was minimal, with limited neointimal formation and no evidence of necrosis, suggesting species-specific differences in vascular reactivity and healing potential. The present findings are also broadly similar to other reports covering a broad spectrum of zinc alloys in the rat [4, 8, 10, 14, 32]. This pro-healing response in rats indicates a more suitable physiological environment for assessing the long-term performance of zinc-based biodegradable implants. Furthermore, morphometric analysis of neointimal area (NA) and wire-to-lumen thickness (WLT) from VVG-stained sections confirmed that reduced implant dimensions correlate with improved tissue compatibility. Specifically, the 0.12 mm diameter implants showed a significant reduction in neointimal growth compared to earlier 0.25 mm designs, reinforcing the importance of strut minimization in mitigating vascular injury and promoting favorable remodeling. Collectively, these findings underscore the potential of thinner, zinc-based stents to achieve enhanced biocompatibility and support the ongoing development of next-generation biodegradable vascular implants.

The differences in inflammatory responses observed in the mouse and rat arterial implantation models raise important considerations for selecting animal models in studies of the biocompatibility of zinc-based devices. The more robust lymphocyte-predominant inflammation, extracellular matrix degradation, and necrosis seen in the mouse model appear to reflect a species-specific immune reaction. In contrast to the response we report here in C57BL/6 mice, rats, rabbits, and pigs exhibited generally mild-to-moderate, macrophage-predominant responses [18, 34, 35], indicating a more consistent and potentially more clinically relevant biocompatibility profile. There is some literature supporting an increased role of lymphocytes in the response to biomaterial implants in mice compared with rats. For instance, mice, but not rats, showed lymph node activation in response to dextran-hydrogel implants [36]. Additionally, T cell-deficient mice, but not rats, exhibited significant alterations in the foreign body response to implanted materials [37]. Together, these results underscore the importance of selecting appropriate animal models to provide a more accurate picture of host-material interactions, particularly for vascular implants, where immune modulation is crucial for integration and function. While further optimization of material properties and surface characteristics may still help refine the immune profile of zinc-based devices, the overall body of evidence suggests that the severe response in mice should be interpreted with caution. Ultimately, this study reinforces the need for careful model selection in preclinical testing when evaluating the translational potential of zinc-based vascular implants.

However, the pronounced species-specific differences in inflammatory responses between mice and rats also raise critical concerns that go beyond the need for careful model selection. While the rat model demonstrated a pro-healing tissue response consistent with prior zinc alloy studies, the severe inflammation, necrosis, and lymphocyte-dominated reaction in mice suggest a fundamentally different host-material interaction. These discrepancies highlight the potential for misleading conclusions when extrapolating findings from a single animal model. Therefore, it is essential not only to validate the biocompatibility of zinc-based materials across multiple preclinical models but also to investigate the underlying immunological and molecular mechanisms driving these divergent responses. Future studies should incorporate both mechanistic immunoprofiling and longitudinal degradation analysis in various species, including large animals with closer physiological relevance to humans, to better predict clinical outcomes. Furthermore, incorporating genetically modified mouse models or immunosuppressive interventions could help elucidate the contribution of specific immune pathways to the observed responses. These efforts are critical to refining our understanding of host responses to bioresorbable zinc alloys and ensuring the translational fidelity of preclinical evaluations.

The stark differences in inflammatory response between mice and rats observed in this study may also be driven by species-specific physiological factors. Mice possess smaller arteries and lower blood flow velocities, which may hinder the clearance of zinc ions and degradation products, leading to localized accumulation, pH changes, and heightened tissue irritation. Additionally, mice have been shown to mount more lymphocyte-driven responses to foreign materials, whereas rats typically exhibit macrophage-predominant inflammation, consistent with our findings. The identical implant diameter used across species further compounds these effects, as the implant-to-vessel ratio is substantially larger in mice, potentially increasing mechanical injury and altering local hemodynamics. These observations underscore the importance of considering vascular scale, immune context, and physiological buffering capacity when selecting preclinical models for degradable metal implants.

A major finding of our study is that reducing the strut thickness improves the biocompatibility of zinc alloy arterial implants. Strut thickness is becoming recognized as a critical factor influencing the biocompatibility of arterial stents [38–40], with significant implications for vascular healing, blood flow dynamics, and long-term clinical outcomes. Thinner struts tend to promote faster endothelialization, by reducing deployment injury and the surface area re-coverage required for regenerating cells. Rapid, successful healing reduces inflammation and minimizes the risk of late stent thrombosis. In contrast, thicker struts cause more injury to the vessel wall during deployment, which can delay healing and provoke stronger inflammatory responses, ultimately increasing the likelihood of thrombotic or restenosis events. The geometry of the struts also impacts blood flow. Thicker struts protrude further into the arterial lumen, disrupting laminar flow and shear stress [41–43]. These disturbances can activate platelets and encourage neointimal hyperplasia. Thinner struts, by contrast, are more streamlined and cause less flow disruption, resulting in improved hemocompatibility. However, there is a necessary trade-off between biocompatibility and mechanical performance. Thinner struts must still provide sufficient radial strength to support the vessel [21], a challenge that is being addressed in conventional metals by using high-strength materials such as cobalt-chromium and platinum-chromium alloys. Biodegradable metals, even with advanced alloying and processing, often fail to reach the mechanical performance of advanced conventional metals. A challenge remains in balancing reduced dimensions with mechanical integrity, particularly in biodegradable systems where strength degrades over time. Alloying and processing strategies must continue to evolve to support ultra-thin geometries without sacrificing radial strength.

This study relies on static histological endpoints and does not include direct hemodynamic measurements such as QFR. These assessments are not feasible in small animal models due to anatomical and imaging constraints. However, reduced neointimal growth and preserved lumen geometry are established structural correlates of improved vascular patency. Future work in large-animal models will be needed to directly evaluate flow dynamics and to confirm these translational implications.

Looking forward, the study’s findings point to several critical directions for future research. First, deeper mechanistic work is needed to map how alloying elements, particularly trace additions like Zr and Mg, influence not just structure and strength, but degradation behavior and immune modulation. Advanced analytical techniques, such as atom probe tomography or synchrotron X-ray analysis, could resolve nanoscale phase distributions and elemental partitioning. Second, surface modification strategies, including passivation layers, nanostructuring, or bioactive coatings, may mitigate adverse biological responses while preserving mechanical performance. Finally, translational studies must include more representative large-animal models to capture the long-term interplay between degradation, vascular remodeling, and immune tolerance in clinically relevant settings.

In summary, while AMZ and ACMZ alloys exhibit promising mechanical and biological profiles, their future as viable implant materials depends not only on refining their composition and microstructure, but also on elucidating how they interact with the complex vascular environment. The integration of materials science with vascular biology and immunology will be crucial to advancing zinc-based alloys from promising concepts to clinical reality.

5. Conclusion

This study highlights the significant influence of alloy composition, microstructure, and strut thickness on the mechanical performance and biocompatibility of Zn-based biodegradable arterial stents. The AMZ and ACMZ alloys exhibited refined, defect-free microstructures and superior mechanical properties after processing, positioning them as strong candidates for biomedical applications. Conversely, the CMM alloy’s microstructural defects adversely affected its mechanical behavior, indicating a need for further optimization. In vitro blood component and ex vivo whole-blood testing showed little change in responses to Zn alloys compared with pure Zn metal. These blood responses to Zn alloys were equivalent to or less thrombogenic compared to the responses to clinical, biostable materials like stainless steel, suggesting that the pharmacological standard of care for maintaining hemocompatibility in patients will likely translate to Zn alloys. Notably, in vivo evaluations revealed species-specific differences in tissue response, with rats showing markedly improved healing and reduced inflammation compared to mice, underscoring the importance of model selection in preclinical testing. A key finding is that reducing the implant thickness significantly improves vascular compatibility. This enhancement is likely due to reduced vessel injury, accelerated endothelialization, reduced inflammation and localized toxicity from degradation products, and diminished flow disturbance. However, maintaining mechanical integrity remains a critical challenge when optimizing for thinner struts. Overall, these results support the continued development of thin-strut, zinc-based biodegradable stents as a promising strategy to improve long-term vascular outcomes and advance the field of bioresorbable cardiovascular implants.

Data Availability Statement

All data generated or analyzed during this study are included in this published article and its supplementary information files. Additional raw data are available from the corresponding author upon reasonable request.