Development of a Wirelessly Powered Smart Implant to Monitor Spinal Fusion

Abstract

Lumbar spinal fusion surgery is performed on patients in whom non-operative treatments have failed to relieve chronic lower back pain (LBP) and restore functionality. The procedure involves inserting titanium alloy rods adjacent to two or more vertebrae on each side of the spine to support spinal fusion. Currently, clinicians rely upon periodic x-ray radiographic images to track fusion progress and determine whether patients can resume normal activities or to assess if the fusion has failed. However, the reliability of imaging evaluation techniques is questionable and leads to either very conservative (and prolonged) restrictions on activity or additional exploratory surgeries. The definitive criteria for a successful fusion remain ambiguous, and determining the progress of spinal fusion remains a challenge for orthopaedic surgeons and clinicians. Observing strain variations on a spinal fusion rod post-implantation has been demonstrated to correlate with changes in bony mass stiffness as fusion progresses, indicating the state of fusion. The challenge with strain measurements relates to having a reliable implant which aligns with existing clinical workflows and provides new data on the state of healing. If the existing titanium alloy rod could be made "smart", i.e. the strain measurement capabilities are embedded into the rod, then the existing clinical, surgical workflow could be maintained. This research focuses on the design and development of a smart spinal fusion implant with the potential to measure strain without complication in the surgical procedure. To meet this aim, two key research questions were addressed. First, a fully implantable wireless spinal rod was developed to support animal trials of spinal fusion. The implant was constructed by mounting a semiconductor strain gauge sensor into a housing machined into a custom spinal rod. A miniaturised electronic module was developed to measure the strain and transmit the data to an external wireless receiver. The module consisted of a strain gauge signal conditioning which was controlled by a microcontroller, and a custom wireless power and data transfer application-specific integrated circuit (ASIC) developed previously at the Auckland Bioengineering Institute (ABI). The electronics module was mounted into the housing, and a printed circuit board (PCB) coil was placed on top of it. This was sealed under a liquid crystal polymer (LCP) lid. Wireless power was transferred to the implant from an external coil at 6.78MHz for 980ms, over which 10 samples of strain were measured. The data was then transmitted using phase-shift keying at a data rate of 678kbps at 6.78MHz. Data was received at an external coil, demodulated and logged to a computer with a measurement cycle taking one second. The implant was characterised on a test rig, and it was confirmed that the 24-bit strain values could be wirelessly measured using the smart spinal implant designed to achieve 1με resolution. This showed that the device was ready for animal trials to quantify strain as fusion occurs in a sheep model. Second, to make the implant clinically relevant, it would be preferable to replace the LCP lid with titanium. LCP is an appropriate seal for animal trials with a lifespan of around several months before water permeates through it, and the device becomes unreliable. Titanium can be welded to the rod to achieve a hermetic seal (gas-tight) with a lifespan of many years, which leads to a smaller device and eases reliable manufacturing as welding is possible. However, this would require transferring inductive power through the conductive titanium lid, which has not been achieved in a spinal implant. Thus, inductive power transfer through metal sheets was investigated via a combination of numerical and experimental tests. A simple test set-up based on hand-wound, cylindrical 10-turn primary (inner radius of 30mm) and 10-turn secondary coils (inner radius of 5mm) was created into which metal sheets could be introduced to allow study their impact on wireless power transfer. The equivalent 2D axisymmetric FEM models were developed to analyse inductive link principles and validate experimental studies. The hand-wound coils were also used to investigate the impact of a titanium enclosure on IWPT system parameters through both simulations and experiments. The simulation results matched experimental results reasonably well, validating the approach; thus, in the future, the validated FEM simulations could be used to investigate power transfer to a miniaturised titanium-packaged smart spinal fusion implant. The impact of the titanium spinal fusion implant, consisting of a titanium spinal rod, housing, and lid, on an IWPT system and an optimum frequency for maximum power transfer was determined. The maximum transferred power was dependent on the titanium alloy, lid thickness, implant size, implant coil location, frequency of power transmission, magnitude of the primary field, and primary and secondary coils dimensions and configurations. FEM simulation results revealed that a maximum power of 1.84mW, at 1A primary current and an operating frequency of approximately 400kHz, could be transferred through a 110μm-thick Grade-5 titanium lid used to seal a 5.5mm-thick, 50mm-long Grade-5 titanium rod, and 0.5m-thick, Grade-5 housing with an internal volume of 18 x 8 x 5mm (L x W x H) for this spinal fusion application. The maximum link potential of 0.035 at 199kHz could be achieved for the same set-up. These results indicated that an acceptable amount of power could be transferred through titanium to power the implanted electronics, supporting the future development of titanium packaged smart spinal fusion rods. This research supports the hypothesis that it is feasible to construct a smart spinal fusion implant that includes the function of measuring strain, can ultimately be employed in clinical practices of spinal fusion, detection of the onset of fusion, non-union or other complications, determination of the efficiency of various bone treatments, and the design of rehabilitation protocols.