Bone fractures

Musculoskeletal injuries are among the most disabling conditions in the United States; with the total number of bone fractures ranging from 12 to 15 million per year. Treatment of these fractures represents a significant burden on the U.S. healthcare system; with hospitalization costing an estimated $23.4 billion in 2004. Determining how well a fracture is healing is crucial to making correct clinical decisions for patients, but multiple studies cite the lack of standardized methods for assessing fracture union.
Radiographic imaging and physical evaluation are two common methods of monitoring bone fracture healing in the clinic. Plain X-ray radiographs are using, but studies show that these correlate poorly with bone strength; do not define a union with enough accuracy, and are unreliable for determining the stage of fracture repair. Computed tomography (CT), dual-energy X-ray absorptiometry (DEXA).

Improved diagnostic capabilities

Ultrasound can offer improved diagnostic capabilities but have limited use in the clinic primarily due to cost and high radiation dose. So physical examinations by a physician are often, but these are subjective and can result in imprecise assessments. To measure the impedance of a changing fracture callus over the course of healing, we implanted commercially-fabricated sensors into broken tibias stabilized in an external fixator model for initial validation.

These 250 µm diameter FR4 (a glass-reinforced epoxy laminate material) sensor pins were placed into either 0.5 mm (N = 6) or 2 mm (N = 5) defects to determine if they could distinguish EIS trends between previously established healing versus non-healing fracture models, respectively. Measurements taken twice weekly from 20 Hz to 1 MHz with mice sacrificed at scattered time points up to day 28.

Healing response with fracture

With 0.5 mm defects initiated a healing response with fracture calli containing a heterogeneous mixture of cartilage and new trabecular bone 14 days post-injury. However, mice with 2 mm defects contained primarily undifferentiated fibrous tissue; indicating the minimal healing response. Histology images have false-colored to aid in interpretation of tissue composition.
Therefore linear regression analyses indicate a significant positive relationship between electrical resistance (R) and time with 0.5 mm defects (p < 0.0001), and no correlative relationship between R and time in mice with 2 mm defects. So by day 28, histology for the 0.5 mm defects shows new bone formation; while 2 mm critical-sized defects result in atrophic nonunions with the absence of bony bridging. Longitudinal impedance data to the 2nd measurement time point to allow the injury to stabilize.
Univariate linear regression analyses to compare end-point impedance measurements to stereology results (quantified tissue composition); to a number of µCT indices (bone volume/total volume, bone mineral density, trabecular number, trabecular thickness, and trabecular separation). Meanwhile Two-tailed t-tests determined whether regression slopes were significantly different than zero, and significance was set at p < 0.05.