Hydroxyapatite-coated strain gauges for long-term in vivo bone strain measurements

The aim of this study was to examine the bonding process between hydroxyapatite-coated strain gauges and bone in order to continue development of a long term in vivo strain measurement device. Two types of commercially available hydroxyapatite (HA) particles were applied to the sensing surface of uniaxial strain gauges using a polysulfone solution as an adhesive. Characterization by scanning electron microscopy and x-ray diffraction (XRD) was used to determine materials property differences between the two powders. Interfacial strengths between the HA coatings and the strain gauges were tested and found comparable to interfacial strength obtained by a plasma sprayed HA coating on the surface of a titanium implant. Gauges were surgically placed on the periosteal surface of greyhound femora. Three groups of dogs were implanted with gauges for periods of 3, 6, and 12 weeks using cyanoacrylate, resorbable sutures, and cable ties to initially hold the gauge against the surface of the bone. Following euthanasia, the femora of the dogs were explanted and subjected to cantilever loading. Response of the implanted HA-coated gauges were compared to a control set that had been freshly glued onto the contralateral femur. Full response, that is, 100% of the strain measurement with respect to the control, was obtained after 12 weeks in vivo. Attachment of HA-coated gauges with circumferential suture showed bonding, while HA-coated gauges attached with cyanoacrylate did not bond to bone. After mechanical testing, femora were embedded in polymethylmethacrylate, cut, ground, and polished. Sections were stained using mineralized bone stain (MIBS) and optical microscopy was performed using transmitted and fluorescent light to allow analysis of remodeling occurring in the region of the strain gauges. Bone formation occurred at the HA surface of sutured gauges, and a fibrous tissue layer developed between the bone and HA coating when the tissue adhesive was used to initially bond the gauge. Fluorescence microscopy indicated an increase in the number of areas of bone remodeling adjacent to the gauge but a normal rate of remodeling of 0.93 +/- 0.07 mum/day was observed. No gross bone remodeling due to strain gauge placement was observed. Backscattered electron imaging (BSE) indicated new bone apposition at all time periods.