A Single-Channel Telemetric Intramedullary Nail for In Vivo Measurement of Fracture Healing

Objective: The objective of this study was to develop a single-channel telemetric intramedullary nail that measures anterior-posterior bending strains and determine whether these forces decrease sigmoidally when normalized to the ground reaction force during fracture healing. Methods: A transverse midshaft femoral osteotomy (1 mm) was stabilized using a customized TriGen intramedullary nail incorporating a strain gauge in the anterior-posterior plane. Fourteen skeletally mature sheep (2-3 years old) were treated in two pilot studies (n = 3/pilot) and a pivotal study (n = 8). Three animals were excluded as a result of welfare issues. Static strain measurements were acquired at approximately 130 Hz during leg stance. In vivo gait analysis was carried out weekly to assess ground reaction forces and biweekly x-rays to assess stability and fracture healing. Animals were euthanized 12 weeks postoperatively. Callus formation was assessed by microcomputed tomography and histomorphometry. The degree of load share between bone and the nail was determined postmortem by three-point bending. Results: A significant preload was generated during implantation, most notably during placement of the four interlocking screws and by the action of attached soft tissues. Eight animals showed evidence of bone healing by x-ray, microcomputed tomography, and histology. However, a reduction in implant load was only observed with two of the eight. The degree of load sharing observed in vivo in these animals (50%-75%) compared favorably with the in vitro observations (approximately 50%). In the nonhealing ambulating animals, nail forces did not change over time. Three-point bend tests carried out on "healed" femurs suggested that load sharing between the bone and nail could be detected more easily in the absence of soft tissues. Conclusion: No clear correlation between implant strain and fracture healing was observed using the single-channel system when subjected to one external loading regime (leg stance phase). However, ex vivo biomechanical testing demonstrated that load share changes could be detected when loads were directly applied to the bone in the absence of muscle and ligament forces. These data emphasize the need to fully characterize the complex biomechanical environment of the limb to determine the load changes resulting from fracture healing.