Total Knee Arthroplasty - part 8 pps

283 Chapter 45 · The Electronic Knee – C W Colwell, Jr., D D D’Lima cylinders The inner cylinder was instrumented with strain gages while the outer cylinder was fixed to bone [25] This transducer was used to measure tibial forces in normal cadaver knees and was inserted through a 1-cm resection of the proximal tibia.A prototype electronic tibial replacement system was also developed, using radiofrequency transmission of data, which demonstrated comparable accuracy in measuring the location and magnitude of tibiofemoral compressive forces [26] This telemetry system was found to accurately transmit the signal through bone,cement,and soft tissues with a range of approximately m Polyethylene insert Transducer strain gauges Titanium shell Location of multichannel transmitter Glass feed through antenna Technical Details a Protective polyethylene cap Transducers A titanium alloy revision tibial prosthesis design was instrumented with force transducers, a microtransmitter, and an antenna The tibial-force transducers were manufactured by NK Biotechnical and have been described previously [3] and have been used in cadaveric studies [24, 27] The tibial tray consists of upper and lower halves separated by support posts,below which lie the load cells, located at the four corners of the tibial tray By measuring the force on each load cell and the total axial load, the location of center of pressure can be determined.In addition, the distribution of forces between the medial and the lateral compartments can be calculated Figure 45-1 demonstrates the location of the multichannel transmitter and the hermetic feed-through antenna (⊡ Fig 45-1) Telemetry The microtransmitter receives the analog signal through leads from the load cells and is connected to the transmitting antenna The microtransmitter was developed by Microstrain (Burlington, VT) entirely from off-the-shelf surface-mount integrated circuits The requisite bridge signal conditioning,the multiplexer,the A/D (analog-to-digital) converter,and the programmable gain and filter functions were combined on a single chip (AD7714, Analog Devices, Norwood, MA) The AD7714 features three true differential bridge inputs, five pseudo differential inputs, a maximum resolution of 22 bits, and a software programmable gain of 1-128 The microprocessor (PIC16C, MicroChip Technologies, Chandler Arizona) allows the AD7714 to be reprogrammed through the serial port of a PC Once programmed, the configuration is stored in nonvolatile, electrically erasable, programmable, read-only memory (EEPROM) on the PIC16C On power up, the PIC16C reads the EEPROM to configure the AD7714 for the appropriate channel-specific gain, filtering, and sample rate parameters The microprocessor also performs pulse-code modulation (PCM) of a surface acoustic wave (SAW) radio-frequency (RF) oscillator (RF Monolithics, Dallas, TX) PCM modulation is less prone to interference than pulse-width modulation b ⊡ Fig.45- 1a, b a Cross-sectional diagram demonstrates the location of the multichannel transmitter and the hermetic feed-through antenna b Exploded view of the tibial tray sections and the polyethylene cap protecting the transmitting antenna (PWM) and pulse-interval modulation (PIM) Another advantage is the ability for error detection in RF transmission This is accomplished by sending a checksum byte, which is the sum of the preceding data bytes.A mismatch between the checksum and the number of bytes received generates an error flag in the data Transmitting Antenna Since RF signals not pass through the sealed titanium shell in the current configuration, a single-pin hermetic feed-through transmitting antenna is located at the distal tip of the stem and is connected to the microtransmitter The antenna is manufactured by Hermetic Seal Technology (Cincinnati, OH) and 45 284 V Technology is certified medical grade The feed-through wire is tantalum (ASTM B-365) with a glass insulator (Sandia Cabal12) and a titanium-retaining ring (Ti-6AL-4V, ASTM F136) Receiver The PCM receiver contains a receiving antenna, a matched RF SAW oscillator,and a level converter to generate RS-232 signals from the incoming data stream The RS-232 signal is read by custom PC software and can be integrated with data from other instruments 45 Power Powering an implantable system can be demanding Although batteries have been used in the past, they contain toxic chemicals, have a limited life span, and occupy valuable space Remote powering can be achieved with the use of magnetic near-field coupling Briefly, an external coil driven with AC current generates an AC magnetic field,which in turn generates AC voltage in a receiving coil placed within the magnetic field.This voltage is then rectified and filtered to provide the DC power required for the telemetry system Remote powering through a metallic orthopedic implant presents a further challenge, since the efficiency of coupling is significantly reduced by the presence of metal between the transmitting and receiving coils.Although high frequencies (>125 kHz) increase the coupling efficiency, unacceptable shielding and eddy currents are produced at these frequencies With the current device the optimum frequency of operation is found to be approximately 3.5 kHz An external (primary) coil, inches in diameter, is housed in a custom-padded ring that can be contoured to the subject’s leg The external coil has an impedance of 10 Ohms at 1.6 kHz drive frequency.A function generator provides a 3-V AC signal at 3.45 kHz to a 100-W power amplifier, which amplifies the signal and drives the external coil The implant contains the internal (secondary) coil,which consists of 2000 turns of magnet wire wound on a highpermeability ferrite core Approximately 40 mW of power can be generated in the internal coil, which is adequate to power the telemetry system In Vitro Testing Following assembly and sealing,the prosthesis was recalibrated and checked for thermal drift between 15° and 45°C The integrity of the radiofrequency signal was then tested with the prosthesis cemented in five cadaver knees The device was tested in three knees that were mounted on a dynamic quadriceps-driven knee-extension simulator (Oxford knee rig) The external coil was secured around the upper tibia and the transducer data were measured during knee flexion-extension The receiving antenna was progressively moved away from the rig,and the range of transmission was measured as the maximum distance between the transmitter and the antenna without loss of signal To test the measurement of the magnitude and the location of applied forces, compressive loads of different magnitudes were applied at various locations The implant was cemented in a custom fixture and was mounted on a multiaxial testing machine (Force 5, AMTI, Watertown, MA) The machine can apply flexion-extension, axial-tibial rotations,and translations in the anteroposterior direction.Vertical load was applied perpendicular to the tibial tray To test the effect of the presence of an insert on the tray, loads were applied with and without an 8-mm polyethylene insert that matched the size of the tibial tray (⊡ Fig 45-2) The magnitude and the location of the externally applied loads were compared with those measured by the tibial transducers In the normal knee, loads are typically bicondylar rather than through a single point To determine whether the implant could detect bicondylar loading,loads were applied through a femoral prosthesis at different flexion angles and in liftoff The loads measured by the individual transducers were then analyzed to determine center of contact and load distribution In vitro cadaver tests demonstrated that the implant transmitted reliably through bone, cement, and soft tis- Loading through a cobalt chrome ball Tibial Tray External Coil Transmitting antennae within protective polyethylene cap Polyethylene Insert ⊡ Fig 45-2a, b The accuracy of the calibration was checked on an AMTI Force multiaxial testing machine a Dynamic loads were applied through a cobalt-chrome ball directly to the tibial tray b The same loads were applied through an 8-mm-thick polyethylene insert 285 Chapter 45 · The Electronic Knee – C W Colwell, Jr., D D D’Lima sue After a stable signal was obtained (based on the error detection of the checksum byte), the receiving antenna could be moved between and m before loss of signal was noted Accuracy tests were first performed without a polyethylene insert (loads applied directly to the tibial tray) Results comparing the magnitude of externally applied loads with measured loads of up to 2000 N showed close agreement with a mean absolute error of 1.2% When the location of center of load along the mediolateral direction was considered, the mean absolute error between the applied load location and the measured load location was