Evaluation of factors affecting bonding rate of calcium phosphate ceramic coatings for in vivo strain gauge attachment.

The aim of this study was to compare the bone-bonding rates of eight calcium phosphate ceramic (CPC) coatings attached to strain gauges, alone and in conjunction with an OP1 device (Creative BioMolecules, Hopkinton, MA) and autologous concentrated pericyte cells. These coatings were studied to develop faster bone bonding to long-term in vivo strain sensors. Characterization of the CPC powders using electron microscopy and X-ray diffraction showed that they had shapes ranging from spherical to rocklike and properties ranging from highly crystalline to amorphous. CPC coated gauges were placed on the femora of young male dogs during aseptic surgery and were initially held in place using resorbable sutures. Test groups were euthanized after 3, 9, and 12 weeks. Both femora of the dogs were explanted and cantilever loaded. Response of the implanted hydroxyapatite (HA) coated gauges were compared to the response of bench-top glued sets of gauges (controls) attached to the contralateral femur and reported as a percentage of the control values. One CPC coating type showed an average response of 30% of controls after 3 weeks, four showed average responses higher than 75% after 9 weeks, and three showed averages higher than 82% after 12 weeks in vivo. Amorphous CPC coatings bonded more quickly than crystalline ones and particle shape had less effect than crystal structure on bonding rates. When either OP1 or autologous concentrated pericyte cells were placed on selected CPC coated gauge surfaces, the CPC5 coated gauges bonded best after 3 weeks with a response of 59%. After the same time period in vivo, CPC3 and CPC7 provided responses of 40 and 16%, respectively. Comparison of a soluble calcium-coated CPC with an uncoated one that had identical crystal structure and similar particle shape indicated that the calcium coating slowed bone bonding substantially in the young dog model. Optical microscopy of stained undecalcified bone sections and backscattered electron imaging indicated bone formation at all bone-HA interfaces and an increase in the number of areas of bone remodeling adjacent to the gauge at all time periods. Gross bone remodeling due to strain gauge placement was only observed near the distalmost cell-seeded strain gauges. Selection of the type of coating and enhancement system can accelerate bone bonding to strain sensors but must be tailored to the bone of the model in which it is being used. Augmentation of CPC coatings with cells or OP1 resulted in variable enhancement of the bonding rate and depended on the CPC and the enhancement system.