Strain and Bone Fracture Healing: Image-Based Mechanics Models to Redefine the Rules

PROJECT SUMMARY The long-term goal of this research is to understand the mechanical factors that influence bone fracture healing in large animals and humans. In the early stages of bone healing, the fragments of a broken bone can move relative to one another. These small movements stretch the soft tissues that are involved in early fracture repair,

producing a mechanical effect known as strain. Since the 1970s, strain has been strongly linked with the biology of fracture repair, but the conceptual framework for explaining strain in the context of bone healing has not evolved in four decades. Today, orthopaedic surgeons are keenly aware that strain regulates fracture healing,

but they cannot measure it in their patients. Authoritative clinical textbooks are riddled with nonspecific, alarming, and impractical advice about the risks of fixing a fracture with a bad strain environment. In the absence of clear guidance, surgeons learn to rely on biomechanical rules of thumb for how to select the right implant for certain

types of fractures. Decades of mixed messaging and indirect discussion about strain and bone healing have created significant barriers to innovation in clinical training and implant design. There is now a major unmet need to develop innovative new research tools that can provide insights on how mechanical strain regulates bone

healing. To address this need, we will bring together a suite of sophisticated physics-based models and image analysis techniques to do what has been impossible until now: directly assess strain at the tissue level and show its association with the processes of fracture healing. This research has two technical aims. For the first aim, we

will use micro-computed tomography (µCT) scans to create 3D virtual reconstructions of the shinbones of sheep with fractures that healed after surgery. We will simulate gait-induced loads on the bones and use the models to measure strain in and around the fracture line. Strain measured from the models will be spatially correlated with

the new bone formation and a threshold for allowable strain will be determined. In the second aim, the focus will be on adaptive changes that occur in old bone near a healing fracture. The image-based models will again be used to measure strain, but now spatial cross-correlation of high-resolution data from the images will be used to

assess whether strain on the outer surface of the old bone is associated with an internal loss of bone mineral density compared to before the injury. The results from this project will pave the way for a new paradigm of thinking about strain and bone healing. Although we will be studying sheep, the groundbreaking methodologies

developed for this project have high translational potential for use in clinical research. The same types of modeling and image data-mining techniques could be used to study fracture healing in human patients. This will ultimately help improve clinical decision-making for treatment of complex fractures, where there is still

considerable debate among surgeons about how much strain is biomechanically optimal for fracture healing.