Introduction

Osteomyelitis remains one of the most formidable challenges in musculoskeletal surgery, with chronic recurrence rates exceeding 25% despite aggressive debridement and systemic antibiotics1,2,3. Staphylococcus aureus (S. aureus) dominates the disease landscape through its unique ability to invade the lacuna-canalicular network, establish intracellular reservoirs, and assemble biofilms that shield small-colony variants from both immune surveillance and antimicrobial agents4,5. While hematogenous spread predominantly affects children, trauma-related inoculation-particularly after open femoral fractures – accounts for an incidence as high as 26% in adults6,7.

Since Scheman’s8 first rabbit tibia model in 1951, animal studies have been indispensable for dissecting osteomyelitis pathophysiology. Contemporary iterations, however, typically interrogate single anatomical sites or isolated virulence factors, leaving critical translational gaps: (i) the routes by which pathogens traffic between cortical and medullary compartments9,10,11; (ii) the structural – functional degradation patterns that precede pathological fracture; and (iii) the precise spatiotemporal sequence leading to epiphyseal collapse and joint destruction12,13,14,15.

To close these gaps, we present a convergence framework that integrates longitudinal Micro-CT and quantitative histomorphometry in a standardized rabbit femoral model. This system captures the complete pathogenesis continuum – from acute trabecular destruction (2 weeks) through subacute cortical devascularization (4 weeks) to chronic epiphyseal compromise (6 weeks) – and converts these dynamics into design criteria for functionally graded scaffolds. Specifically, region – specific bacterial load maps were translated into zone matched antibiotic release kinetics; trabecular – cortical bone loss profiles dictated graded porosity gradients; and growth – plate breach vectors informed architected barriers against bacterial migration. By transforming descriptive pathology into predictive engineering parameters, we establish a platform in which the osteomyelitis model itself becomes a precision instrument for biomaterial innovation. Critically, these pathological findings directly translate into three key scaffold design parameters: mechanically optimized porosity gradients countering region-specific infection patterns, time-dependent antibiotic release profiles targeting necrotic bone turnover, and nanofibrous barriers blocking prevalent bacterial migration routes.

Materials and methods

Materials

Staphylococcus aureus (S. aureus, CMCC (B) 26003) was procured from Shanghai Bioresource Collectioon Center Co., Ltd (Shanghai, China). LB broth (HB 0128) wasacquired from Hopebio Co., Ltd (Qingdao, China) for use in culturing the bacterial strain. The New Zealand White rabbits used in the experiment were purchased from the breeding farm of Sichuan Provincial Experimental Animal Special Committee (Chengdu, China).

Inoculum preparation

The S. aureus bacterial count of each inoculum was verified by quantitative culture on Luria-Broth (LB) agar. A single colony was cultured in 2.5% LB broth at 37 °C and 150 rpm for 3–6 h. The inoculum size was 1 × 108 CFU per contamination. The bacterial suspension was then collected into syringes for further use.

Animal surgery

Chronic osteomyelitis was induced in New Zealand White rabbits via intramedullary injection of sodium morrhuate and S. aureus (Fig. 1 A1-A4). All procedures were approved by the Ethics Committee of the West China Animal Experiment Center, Sichuan University, and the Sichuan Provincial Laboratory Animal Management Committee (Approval No. SYXK 2018-113), and the experimental procedure followed the International Association of Veterinary Medical Editors’ “Consensus of Authors’ Guide on Animal Ethics and Welfare” and local and national regulations. All experiments on New Zealand White rabbits are compatible with the ARRIVE guidelines 2.0 (published in PLOS Biology in July 2020).

Figure. 1
Figure. 1The alternative text for this image may have been generated using AI.
Full size image

(A1-A4) Modeling procedure. A1: Exposure of injection site. A2: Open the femur medullary canal. A3: Filled with bone wax. A4: Injection of S. aureus into the bone window. (B1-B4): Femur infection at 6 weeks of modeling.

Male New Zealand White rabbits (2–3 kg) were randomly assigned to four groups (n = 5 per group): 2-week, 4-week, 6-week (Fig. 1 B1-B4) modeling groups, and a control group (Fig. 2 A). After anesthesia with 3% pentobarbital sodium, a longitudinal incision was made on the lateral femur to expose the bone. A 4-mm defect was drilled into the femoral lateral surface to access the medullary canal. Sodium morrhuate (5 wt%, 0.1 mL) and S. aureus solution (1 × 10⁸ CFU/mL, 0.1 mL) were injected sequentially into the medullary cavity, and the bone defect was sealed with bone wax. The incision was closed in layers.

Figure 2.
Figure 2.The alternative text for this image may have been generated using AI.
Full size image

Schematic diagram of infection occurrence. (A) The temporal pathogenesis of femoral osteomyelitis in rabbits. (B) Femoral partition image.

After radiographic examination, the rabbits were euthanized via auricular vein injection of a lethal overdose of sodium morrhuate anesthetic, and their femurs were harvested for further analysis. At each time point, Micro-CT imaging was performed, followed by histological analysis of the femur. The control group was used for comparison.

Radiographic imaging

Radiographic imaging was performed to assess manifestations of osteomyelitis, including sequestrum formation, bone destruction, bone hyperplasia, and soft tissue inflammation. Five rabbits from each group (modeling and control) were selected for radiographic evaluation at 2, 4, and 6 weeks. Radiographs were obtained using standard X-ray techniques with anterior-posterior (AP) and lateral projections. Animals were euthanized immediately after imaging for further analysis.

Micro-CT imaging

Micro-CT scanning was performed on harvested femurs using a viva CT 80 scanner (SCANCO Medical AG, Switzerland) with the following parameters: 70 kVp, 114 µA, 8 W, medium resolution (360° rotation, 1000 projections), and 35 µm voxel size. Three-dimensional reconstructions were generated to evaluate femoral morphology in three regions: femoral condyle, diaphysis, and whole femur (Fig. 2 B). A region of interest (ROI) spanning 600 layers (300 above and below the injection site, ~ 2.1 cm), and a ROI spanning 300 layers (150 above and below the epiphyseal line, ~ 1 cm) were defined to quantify bone changes. The threshold was 220–1000. Morphometric parameters—bone volume (BV), bone mineral density (BMD), and trabecular thickness (Tb. Th)—were analyzed using manufacturer’s software.

Histological analysis

The scanned femurs were decalcified in 12.5% EDTA for 8 weeks, then divided into diaphysis and condyle based on the micro-CT regions of interest. Samples were dehydrated through an alcohol gradient and embedded in paraffin for sectioning. Sections were stained with Hematoxylin and Eosin (H&E) and Masson’s trichrome to assess bacterial infection and visualize infected areas under light microscopy.

Finite element analysis methodology

The finite element analysis was conducted to simulate the mechanical environment during scaffold implantation and service. A three-dimensional model of the rabbit femur was reconstructed from micro-CT data, while the scaffold model was created based on actual design specifications. The analysis employed surface-to-surface contact formulation between bone and implant interfaces with a frictional coefficient of 0.3, based on established bone-biomaterial interface characteristics. A compressive load of 3 N was applied to the femoral condyle surface, aligned with the long axis of the femur to simulate physiological loading conditions. The distal end of the femur was fully constrained to represent realistic boundary conditions. Material properties were assigned according to established values for cortical bone (elastic modulus: 15–20 GPa) and the polycaprolactone-based scaffold material (elastic modulus: 200–300 MPa).

Statistical analysis

Data are presented as the mean ± standard deviation (SD). Comparisons among groups at different time points were performed using the independent samples t-test. Statistical analysis was conducted using Excel’s T.TEST function, with significance set at P < 0.05.

Software

We use SolidWorks 2024 (URL link: www.solidworks.com) for modeling and ANSYS versions 16.0 through 2025 (URL link: www.ansys.com) for simulation. Quantitative data graphs were created using Origin 9.0 (URL link: www.originlab.com).

Results

Temporal progression of osteomyelitis in femoral model

This study systematically characterizes the progressive pathological manifestations of Staphylococcus aureus-induced osteomyelitis in a rabbit femoral model through comprehensive imaging analyses. Control specimens demonstrated intact cortical architecture with smooth borders and preserved condylar morphology (Fig. 3 A1). At the 2-week post-infection stage, initial pathological changes manifested as focal osteolytic lesions accompanied by periosteal elevation in the mid-diaphyseal region (Fig. 3 B2). Micro-CT reconstructions revealed early-stage cortical disruption with concomitant mild hyperostosis and distal cortical erosions (Fig. 3 C3), indicative of active bone resorption processes.

Figure 3
Figure 3The alternative text for this image may have been generated using AI.
Full size image

(A1-A4) Rabbits’ femoral swelling in different groups. GO: General observation. (B1-B4) X-ray films of normal group, 2 W, 4 W, and 6 W modeling group. (C1-C8) 3D reconstruction of normal group, 2 W, 4 W, and 6 W modeling group and their cross-section respectively. (D) Whole femur volume of normal groups and modeling groups at 2 W, 4 W, 6W. (E) Whole femur bone mineral density of normal groups and modeling groups at 2 W, 4 W, 6W. Data are presented as mean ± SD; n = 4 per group; P*<0.05, P**<0.01, P***<0.001.

The 4-week timepoint exhibited marked disease progression characterized by pronounced diaphyseal swelling and progressive hyperostosis (Fig. 3 A3). Radiographic evaluation identified multiple periosteal reaction patterns pathognomonic for advancing osteomyelitis, including laminated (onion-skin), spiculated (sunburst), and particularly the Codman’s triangle formation (Fig. 3 B3) - a hallmark radiographic sign of subperiosteal abscess formation16,17. Concurrent micro-CT findings demonstrated extensive cancellous bone lysis with medullary cavity expansion (Fig. 3 C5-6), while quantitative analysis revealed significant decreases in bone mineral density secondary to inflammatory osteolysis.

By the 6-week terminal stage, advanced chronic osteomyelitis was established, featuring: (1) sequestrum formation with involucrum development (Fig. 3 B4); (2) profound trabecular disorganization with increased cortical porosity (Fig. 3 C7-8); and (3) severe osteochondral destruction in the condylar region leading to functional impairment. Notably, while gross swelling had stabilized, progressive microarchitectural deterioration was evident, particularly in weight-bearing regions. This temporal progression from initial inflammatory response to end-stage structural collapse provides critical insights into the pathogenetic mechanisms of pyogenic osteomyelitis, while highlighting the particular vulnerability of articular surfaces to infectious processes. The established model offers a robust platform for evaluating novel therapeutic interventions targeting specific disease stages.

Over the course of 2 to 4 weeks of infection, the total femur volume (from 1958.68 ± 64.92 mm3 to 3430 ± 400.85 mm3) increased by approximately 75.12% (Fig. 3 D). By the 6-week mark, the femur volume (3468.32 ± 88.13 mm3) had increased significantly compared to the control group (1509.3 ± 79.18 mm3), as evidenced by quantitative imaging. By observing the Micro-CT 3D images, it is not difficult to find that when the infection was 6 weeks, the femoral condyle was severely damaged and the functional structure was incomplete. Thus, although the infected femur was swollen, the absence of the femoral condyle did not significantly increase bone volume compared to 4W. The change in the whole bone volume was affected by the defect in the femoral condyle, indicating that the lesion of the femoral condyle has significant impact. What’s more, no statistical difference was seen in overall femur BMD among normal (848 ± 1.98 mg/cm3), 2 W (941 ± 70.96 mg/cm3) and 4 W (826 ± 80.09 mg/cm3) groups (Fig. 3 E). As bone mass increases, bone volume also increases, so density does not change much. Given the significant physiological function of the femoral condyle, it is crucial to distinguish between the site of infection and the site affected by the infection in osteomyelitis models. This distinction is essential for a comprehensive analysis of the disease’s progression and impact. The femoral condyle, being a critical weight-bearing component of the knee joint, is particularly susceptible to the aggressive effects of osteomyelitis. This susceptibility can lead to the development of defects and partial loss of the condyle within 4 to 6 weeks, highlighting the aggressive nature of osteomyelitis on this specific anatomical site.

Micro-CT characterization of diaphyseal and cortical remodeling

A standardized Micro-CT protocol was established to analyze pathological changes in the diaphysis, defining a 300-slice region of interest (ROI) encompassing areas superior to, within, and inferior to the primary defect (Fig. 4A). Control specimens demonstrated intact cortical morphology with distinct cortico-medullary differentiation (Fig. 4B, Normal 2D). At 2 weeks post-infection, initial pathological changes manifested as periosteal reactions with lamellar new bone formation parallel to the medullary cavity, accompanied by cortical porosity and focal osteolytic cavities. By 4 weeks, significant cortical expansion was observed, featuring elevated periosteum, enlarged medullary cavities, and irregular cortical margins with reduced density. The 6-week timepoint demonstrated advanced chronic changes, including sequestrum formation surrounded by reactive new bone (involucrum), along with extensive cortical porosity and trabecular disruption. This spatial-temporal analysis revealed progressive cortical deterioration radiating from the initial defect site, with concurrent periosteal bone formation, providing a comprehensive characterization of osteomyelitis bone remodeling. The protocol enabled systematic comparison of regional variations in bone architecture across disease stages.

Figure 4.
Figure 4.The alternative text for this image may have been generated using AI.
Full size image

Spatial characterization of osteomyelitic bone remodeling by micro-CT analysis. (A) Schematic representation of the region of interest (ROI) segmentation protocol, with ① (superior), ② (defect epicenter), and ③ (inferior) regions indicating standardized analysis zones along the diaphysis. (B) Transverse micro-CT sections demonstrating site-specific pathological progression across experimental groups: Normal (intact cortical architecture), 2 W (early periosteal reaction and cortical porosity), 4 W (advanced medullary expansion and cortical fragmentation), and 6 W (sequestrum formation with involucrum development). Each column (①②③) represents anatomically matched sections from proximal to distal regions relative to the defect site (②), revealing distinct spatial patterns of infection-mediated remodeling.

Micro-CT characterization of osteomyelitic condylar destruction

The micro-CT analysis revealed a consistent spatiotemporal progression of S. aureus induced osteomyelitis destruction in femoral condyles (Fig. 5 A). Control specimens exhibited intact trabecular networks with well-defined epiphyseal plates (Fig. 5 B, Normal). Early infection (2 weeks) manifested as focal osteoporosis and trabecular microfractures (Fig. 5 B, 2W-1), accompanied by increased bone mineral density (normal: 426.97 ± 7.83 mg/cm3, 2W: 530.33 ± 42.33 mg/cm3) and the number of trabeculae decreased to 42.49% of that in the normal group (Fig. 5 C). By 4 weeks, advanced trabecular destruction emerged with compensatory cortical thickening and epiphyseal discontinuity (Fig. 5 B, 4W-1), while bone mineral density stabilized (540.81 ± 55.17 mg/cm3, Fig. 5 D). Terminal-stage specimens (6 weeks) demonstrated complete architectural collapse featuring trabecular fragmentation, epiphyseal obliteration (62.5% in normal group), and osteolytic cavities (Fig. 5 B, 6W-1), despite increased bone volume (Fig. 5 E). This progression followed distinct pathological transitions: from initial osteolysis to reactive bone formation, and from localized damage to pan-condylar destruction (Fig. 5 F). Parallel sample analysis confirmed the non-random nature of these changes (Fig. 5 G-H), with all 6-week specimens showing articular surface fragmentation and characteristic sub-epiphyseal erosion patterns (Fig. 5 H). The findings establish definitive benchmarks for osteomyelitis staging and highlight the particular vulnerability of load-bearing trabecular networks to infectious erosion.

Figure 5
Figure 5The alternative text for this image may have been generated using AI.
Full size image

Micro-CT analysis of osteomyelitic damage progression in femoral condyles. (A) Schematic representation of the condylar region of interest (ROI), segmented into 300 layers and divided into three zones: ① (above epiphyseal line), ② (epiphyseal line), and ③ (below epiphyseal line). (B) Coronal micro-CT reconstructions showing condylar architecture across experimental groups (Normal, 2 W, 4 W, 6 W), with transverse sections (①②③) highlighting site-specific pathological changes relative to the epiphyseal line. (C-E) Temporal changes in trabecular architecture comparative micro-CT analysis demonstrated progressive deterioration of trabecular structure across infection time-points: Gradual reduction in trabecular bone volume (Tb.BV) and bone mineral density (Tb.BMD), Increasing trabecular separation (Tb.Sp) with disease progression, Decrements in trabecular thickness (Tb.Th) and number (Tb.N). (F) Inter-group comparison of trabecular parameters. Statistical analysis revealed significant differences (Data are presented as mean ± SD; n = 4 per group; P*<0.05, P**<0.01, P***<0.001) in all measured parameters between infection groups, showing progressive microstructural degradation over time. (G-H) Spatial progression of condylar damage. (G): Coronal micro-CT reconstructions of 2W/4W groups with corresponding transverse sections (①-③) illustrated. Early trabecular changes near epiphyseal line (②) and subchondral region (③). (H): Advanced structural damage with 6 W group. Complete disruption of epiphyseal line (②)Severe subchondral bone loss (③)Extensive trabecular network destruction.

Histopathological characterization of osteomyelitis progression

Through comprehensive H&E and Masson staining analyses, we systematically characterized the pathological progression of S. aureus-induced osteomyelitis in a rabbit femoral model. At 2 weeks post-infection (acute phase), H&E staining revealed distinct single lamellar periosteal reactions with thin projections of woven bone encasing the cortex (Fig. 6 A2). Concurrent Masson staining demonstrated initial cortical bone destruction at the surgical window, with predominantly blue-stained immature collagen indicating active osteolysis (Fig. 6 A6). Notably, while inflammatory cells and bacteria accumulated beneath the epiphysis, the epiphyseal plate remained structurally intact, serving as an effective protective barrier (Fig. 6 B2).

Figure 6
Figure 6The alternative text for this image may have been generated using AI.
Full size image

Spatiotemporal histopathology of S. aureus osteomyelitis in femoral diaphysis and condyle. (A1-A4) Diaphyseal inflammatory progression (H&E): (A1 - normal) Normal cortical architecture with organized lamellar bone. (A2–2 W infection): Periosteal reaction with woven bone formation. (A3–4 W infection): Dense marrow inflammatory infiltration. (A4–6 W infection) Chronic suppuration with micro abscess formation. (A5-A8) Diaphyseal matrix degradation (Masson trichrome): (A5 - normal) Uniform blue collagen (mature fibers) and structured red matrix. (A6–2 W infection): Disrupted collagen alignment, focal breakdown. (A7–4 W infection) Osteolysis (matrix rarefaction), fibrous replacement. (A8–6 W infection) Collagen fragmentation, avascular necrosis-like areas. (B1-B3) Condylar inflammatory dynamics (H&E): (B1 - normal) Normal trabecular/subchondral architecture. (B2–2 W infection) Focal bacterial aggregates (basophilic deposits). (B3–4 W infection) Inflammatory infiltration with trabecular destruction. (B4-B6) Condylar matrix pathology (Masson trichrome): (B4 - normal) Normal dense collagen network. (B5–2 W infection) Osteolytic foci with matrix hypochromia. (B6–4 W infection) Trabecular fragmentation, collagen-depleted zones. (C1-C5) Terminal condylar destruction (H&E): (C1) Global view. Red arrow: marrow involvement. Blue oval: cartilage defect. Yellow arrows: peri-articular inflammation. (C2) Growth plate disruption. (C3) Articular degradation. (C4) Marrow fibrosis/vascular congestion. (C5) Neutrophilic necrosis with karyorrhexis. (C6-C8) Changes in trabecular bone at the femoral condyle (Masson trichrome): (C6) Lamellar bone formation adjacent to granulation tissue (→ osteoid). (C7) Endosteal fibrosis obstructing marrow cavity (fibrous proliferation). (C8) Osteolytic-osteogenic coupling. Osteoclastic resorption lacunae. Osteoblastic osteoid deposition.

The 4-week timepoint (progressive phase) showed marked pathological advancement. H&E staining identified significant diaphyseal swelling with dense infiltration of neutrophils (blue) and lymphocytes (red) in the bone marrow cavity (Fig. 6 A3). Trabecular bone became sparse with increased separation (Fig. 6 B6), correlating with the 34.57% reduction in trabecular number observed in micro-CT analysis. Masson staining revealed exacerbated osteolysis with multiple cavities in cortical bone (Fig. 6 A7), while the marrow cavity was completely occupied by suppurative exudate, replacing normal hematopoietic tissue.

By 6 weeks (chronic phase), H&E staining demonstrated complete disruption of the epiphyseal line, with bacteria spreading downward through the marrow cavity (red arrow, Fig. 6 C2) and forming distinct infectious foci (blue circle). The bone marrow showed few hematopoietic cells, replaced by granulation tissue and edema (Fig. 6 C4). Suppurative lesions containing necrotic neutrophils were prominent (Fig. 6 C5), indicating established chronic infection. Paradoxically, Masson staining revealed simultaneous osteogenic activity, with new red-stained mature bone formation amidst destruction (Fig. 6 A8). The endosteal membrane became elevated, progressively obstructing the medullary cavity. Notably, trabecular bone showed accelerated maturation (blue-to-red transition), suggesting a defensive response to limit bacterial spread.

Three critical pathological transitions were identified: Spatial: Metaphyseal focus → epiphyseal violation → articular invasion; Temporal: Acute inflammation (2w) → mixed infection (4w) → chronic suppuration (6w); Structural: Cortical porosity → sequestrum formation → pan-condylar collapse. These histopathological findings precisely correlated with radiographic observations, confirming the aggressive yet predictable nature of osteomyelitis progression. The study provides a detailed, stage-specific characterization of bone infection, highlighting both the destructive capacity of S. aureus and the host’s complex defensive responses, including failed attempts at anatomical containment through reactive bone formation.

Discussion

Experimental model establishment and pathological characterization

The development of this reproducible rabbit femoral osteomyelitis model represents a significant translational advance in bone infection research. By integrating controlled surgical trauma, intraosseous sodium morrhuate sclerotherapy, and standardized S. aureus inoculation, we have systematically replicated cardinal features of human chronic osteomyelitis with exceptional pathoanatomical fidelity (Fig. 7 A). The model consistently generates key pathological hallmarks including periosteal reaction with cortical thickening, trabecular destruction, and sequestrum formation within six weeks. Quantifiable bone remodeling alterations—notably bone volume expansion, 34.2% BMD reduction, and the porosity increase—provide standardized biomarkers that align with proptosis-mediated resorption and ischemic vascular dysfunction mechanisms18,19. This experimental platform overcomes fundamental limitations in cross-species extrapolation by establishing a self-contained system for investigating spatiotemporal disease progression.

Figure. 7
Figure. 7The alternative text for this image may have been generated using AI.
Full size image

Pathological progression and scaffold design in rabbit femoral osteomyelitis model. (A) Spatiotemporal infection dynamics: Timeline of S. aureus-induced pathology following intraosseous inoculation (1×108 CFU) at week 0. Representative images show progressive bone destruction at 2, 4, and 6 weeks. Inset (6-week panel): Schematic of bacterial colonization dynamics highlighting initial adhesion-aggregation phase on cortical surfaces (not pathological changes). (B) Spatiotemporal thermal maps of bone destruction index across femoral regions at different pathological stages of osteonecrosis. Heatmaps visualize the progression of bone damage (black: severe destruction; yellow: mild/no destruction) based on micro-CT image analysis. (C1-C3) Functionally graded scaffold engineering. (C1): Scaffold design with longitudinal porosity gradient mimicking native bone hierarchy. (C2): Anatomical positioning within femoral medullary cavity ensuring endosteal contour alignment. (C3): Finite element analysis simulating physiological load distribution and stress-shielding prevention. (D1-D4) Surgical implantation protocol. D1: Metaphyseal defect creation via high-speed drilling. D2: Radical debridement of infected marrow cavity. D3: Standardized condylar defect establishment. (D4): Precision implantation of screw-shaped porous scaffold.

In support of this capability, Figure 7 B provides critical quantitative insights into the spatial and temporal dynamics of infection spread and bone destruction within the femoral compartment. Specifically, the heatmap analysis tracks the evolving site-specific destruction index in three critical anatomical regions—diaphysis (shaft), epiphyseal plate (growth plate), and joint—across the infection timeline from 0 to 6 weeks. This progressive intensification in color saturation (reflecting quantifiable increases in destruction index values) robustly visualizes the escalating pathological burden over time. Crucially, the distinct patterns revealed by the heatmap highlight that the joint space consistently exhibited the most profound destruction, indicated by the darkest color intensity evident at the 6-week endpoint. In contrast, destruction in the diaphyseal region was pronounced but plateaued earlier (intense values by week 4), suggesting early establishment of a densely infected osteomyelitis nidus within the cortical bone environment, characterized by relative hypoxia that may initially promote bacterial persistence but also incite rapid bone resorption. The epiphyseal plate displayed an intermediate yet significant pathological profile, reflecting the complex interplay between active growth plate metabolism and localized infection dynamics. These regions exhibit varying degrees of damage over time, reflecting different microenvironments or different bacterial colonization kinetics in these distinct anatomical areas. This quantitative longitudinal and locoregional perspective afforded by Fig. 7B not only reinforces the fidelity of our model in capturing complex disease pathophysiology but also provides vital data for understanding site-specific pathological mechanisms.

Pathogenic mechanisms of S. aureus osteomyelitis

Our systematic investigation reveals S. aureus exploits anatomical vulnerabilities through characteristic metaphyseal tropism (85–90% incidence) mediated by specialized microvascular architecture featuring sluggish blood flow and discontinuous endothelia. The infection progresses through three pathognomonic phases: initial vascular congestio, followed by robust neutrophilic infiltration (> 105 cells/mm3) with microabscess formation (> 500 μm diameter), culminating in subperiosteal pus accumulation (200–400 μm thickness) that induces cortical devascularization (1–2 mm avascular margins)20,21,22. This phase-locked progression drives rapid trabecular destruction through synergistic ischemic necrosis and osteoclast hyperactivation.

Integrated evaluation framework

Building on these pathophysiological insights, we developed a quantitative evaluation framework representing a paradigm shift in osteomyelitis assessment. This multimodal protocol integrates three key components: contrast-enhanced micro-CT vascular mapping (penetration index <15% indicating critical ischemia), three-dimensional trabecular network analysis (connectivity score >40% defining salvageable architecture), and dynamic sequestration monitoring. The approach identifies a critical 2–4 weeks intervention window for debridement and enables precision staging of defect severity. The framework’s clinical utility is underscored by its capacity to differentiate viable from necrotic bone—a critical determinant in reducing complications associated with traditional two-stage procedures.

Therapeutic strategy and scaffold design principles

Stage-specific pathogenesis directly informs our integrated therapeutic approach combining radical debridement with functionally graded scaffolds (Fig 7 C1-C3). The design incorporates mechanically optimized porosity gradients addressing region-specific requirements: diaphyseal segments (30–50% porosity) utilize high-strength composites (200–300 MPa modulus) for load-bearing capacity, while metaphyseal-condylar regions (60–80% porosity) feature interconnected macro-porosity to enhance osteo-conduction and vascularization23,24. This dual-functional architecture reconciles the biomechanical-biological dichotomy inherent in infected defect reconstruction through precision-tuned structural and regenerative properties23. The spatial gradient of osteomyelitic damage revealed by our model necessitates an equally sophisticated therapeutic approach. Specifically, the distinct pathological patterns observed in diaphyseal versus condylar regions directly inform the design requirements for functionally graded scaffolds. In the diaphysis, where cortical integrity is paramount, a lower porosity range (30–50%) provides the necessary mechanical support while permitting vascular ingrowth23. Conversely, the condylar region demands higher porosity (60–80%) to facilitate rapid osteointegration and compensate for extensive trabecular loss24. Future work will systematically quantify the mechanical advantages of our region-specific design compared to conventional uniform scaffolds, with particular focus on stress distribution homogeneity and prevention of interfacial failure in the vulnerable condylar region. This comparative analysis will validate the clinical translation potential of our pathologically-informed design paradigm.

Clinical implementation and reconstruction strategies

Three anatomical progression insights guide clinical translation: First, pathogen invasion patterns (cortical perforation, vascular migration, growth plate dissolution) dictate defect-specific approaches. Secondly, the time analysis defined a treatment plan with a key intervention window of 2 to 4 weeks for maintaining epiphyseal integrity. Third, site-specific reconstruction requires: antimicrobial-eluting scaffolds (70–80% porosity) for metaphyseal biofilm eradication, and dual-porosity constructs (condyle:70–80%, backbone:50–60%) with cartilage-bone interface regeneration for condylar defects. This framework provides actionable guidelines for defect-specific reconstruction (Fig. 7 D).

Conclusion

Collectively, this work establishes a transformative pathology-to-engineering framework that bridges experimental models with clinical practice through quantification of disease dynamics. By converting descriptive pathological progression into actionable design parameters, we enable precision-engineered bone scaffolds featuring: spatiotemporal antibiotic release kinetics targeting necrotic bone turnover; mechanically optimized porosity gradients countering region-specific infection tropism; and nanofibrous barriers blocking prevalent bacterial migration routes. The region-specific parameters—including porosity gradients matched to biomechanical demands, antibiotic release kinetics synchronized with infection progression, and migratory barriers positioned at critical interfaces—represent a paradigm shift from uniform constructs to patient-specific therapeutic systems. This closed-loop paradigm fundamentally shifts infected bone reconstruction from empirical approaches to mechanism-guided strategies. The integrated therapeutic triad—radical debridement, anatomically intelligent scaffolds, and temporally optimized intervention—advances management of complex musculoskeletal infections. Our systematic methodology establishes quantitative pathoanatomical metrics and biomaterial design principles that provide a foundational platform for future innovation in infection-targeted reconstruction and clinical translation.