**3.2 Vibrometry**

372 Recent Advances in Arthroplasty

Fig. 10. Comparison of different diagnostic imaging methods with regard to sensitivity and

These results prove that, considering the background of up to one million total hip replacements conducted worldwide every year, an advanced method for diagnosing implant loosening is needed. Intelligent implants with integrated sensors have the potential to revolve the diagnostic methods to raise sensitivity and specificity of loosening diagnostics

Intelligent implants are highly complex systems incorporating sensors, actuators and signal procession and have potential for advanced therapy or diagnosis. A main issue of intelligent implants is miniaturization. Smaller implants and sensors have lower area of influence to their ambient environment. Referring to the femoral component of THRs, it could be equipped with small batteries and a wireless signal transmission and power consumption. Common sensors exhibit at least one sensing unit, a processing unit, a transceiver unit and an energy unit. In the field of microsystems some diagnostic systems reached the status of implantable passive and telemetric prototypes (Bergmann et al., 2001, Marschner et al., 2009). Most notably the application of strain gages used in fracture healing implants are the system of the highest development status compared to other orthopaedic and trauma implants (Wilson et al., 2009). In case of THR, sensors are subject to stringent clinical requirements, like gamma sterilization, which may affect the functionality of the sensors inside the implant. Other requirements are the stability of the sensor and the electronic equipment during impaction of the THR and at least the same lifetime than the THR has. There are different working principles to realize a sensor, which can be used to characterize

The preferred detected properties, which were mainly considered, are the acoustical in combination with mechanical properties. The detection of micromotion with an in vivo sensor unit at the distal femoral component was proposed by Hao et al. (Hao et al., 2010). The differential variable reluctance transducer is mounted in the bone cavity and has no

**3. New diagnostic techniques under research – intelligent implants** 

specificity of loosening diagnosis of the femoral component of THRs

**3.1 Sensors for characterization of the implant-bone interface** 

the implant-bone interface to detect loosening of the THR (Fig. 11).

to a higher level.

Diagnostic investigation involving measuring the resonance frequencies of implants has its roots in the field of dentistry (Glauser et al., 2003; Huang et al., 2003). This field of research extended to determining the primary stability of THR during surgery (Lannocca et al., 2007). The eigenfrequency present the governing nature of the vibration motion of a structure. A disturbed structure vibrates in a way that is determined by the structures mass, cross sections, geometric relationships in its assembly, Young's modulus of its components and dampening factors.

The THR integrated in the femoral bone can be described as a system which vibrates in its multiple axial, lateral, and torsional eigenmodes after excitation (Qi et al., 2003). These eigenmodes change with the loosening state and the localization of loosening of the THR. Different eigenmodes are associated with the corresponding eigenfrequency. The changing of the eigenfrequencies of the THR-femur-system due to the loosening state can be measured as vibrations or structure-borne sound in a resonance frequency analysis. By means of a finite-element analysis, Qi et al. determined which minimal sensitivity of an in vivo accelerometer is required in order to detect differences in eigenfrequencies. Therefore,

Current Possibilities for Detection of Loosening of Total Hip

from noise.

the acoustic sound waves.

rehabilitation time (Shao et al., 2007).

**3.3 A new excitation method to detect aseptic loosening** 

**3.3.1 Principle of the novel excitation concept** 

outside the patient's hip opposite to the excitation coil.

application in clinical studies.

et al., 2001).

Replacements and How Intelligent Implants Could Improve Diagnostic Accuracy 375

and the THR was instrumented with a coil, amplifier, microcontroller and telemetry unit in the distal end of the hip stem. The lock-in amplifier separates the measurement signal

The availability of results which compare different loosening states of the THR and therefore sensitivity of the proposed sensor systems is rare. Indeed, the presented devices are available, but only few results verify the possibility to detect a difference between a loose and a well fixed implant. Currently sensitivity and specificity of vibration analysis in THR can only be estimated. Georgiou et al. stated a 20 % higher sensitivity measured with vibration analysis (80 %) than with radiographs (60 %) (Georgiou et al., 2001). The finite element analysis of Qi et al. showed that a loose THR can be identified reliable if 1/3 of the femoral component has no direct bone contact (Qi et al., 2003). One of the main problems of the resonance frequency analysis of THR is the soft tissue damping of the output vibration signal. Furthermore, connection to the tibia and the pelvis can decrease the energy rate of

In another in vitro study the accelerometer was replaced by a blood flow ultrasound probe which enabled consistent detection of the vibrations for a loose and a fixed THR (Rowlands et al., 2008). The ultrasound probe lead to increased amplitudes of the output vibrations compared to the accelerometer signal. Dahl et al. applied ultrasound to detect osseous integration of the total ankle replacement (Dahl et al., 2010). The cadaveric testing demonstrated that the ultrasound technique could distinguish between loose and fixed implant components. In a further study, a trans-femoral implant was measured with accelerometers. It could be shown that the eigenfrequency increases with the weight bearing

Paech et al investigated the natural frequency of four different types of THR and determined these sound patterns in further cadaver tests with bovine and human bone (Paech et al., 2007). The excitation of the THR was reached by using an acoustically neutral steel ball of 120 g as activating impact for sound emission. However, this is not reproducible for

Besides problems in signal analysis and soft tissue dampening, functional sustainment of the integrated electronics during sterilization and impaction of the THR seems to be a significant challenge. Nevertheless detailed research into simulation of different loosening states and significant test results are expected. Furthermore the excitation has to be optimized in order to avoid pressure pain induced by the electrodynamic shaker (Georgiou

In the proposed concept, an oscillator integrated in the THR is used. The oscillator consists of a magnetic body which is fixed on a flat steel spring (Ruther et al., 2010) (Fig. 13). This oscillator is excited by a coil placed outside the patient. The oscillator impinges inside the THR and excites the THR to vibrate in its eigenfrequency. This can be considered as an alternative to the electrodynamic shaker, which was used by different research groups. The excitation in the implant bending modes leads to a sound emission to the surrounding bone and soft tissue. The sound waves can be detected by a vibration sensor, which is applied

they considered three different situations: distal loosening, proximal loosening and central loosening. Because soft tissue dampening and connections to the upper/ lower joints will absorb acoustic energy, sensitivity of the sensors may have to be higher to acquire the signals for further processing. On the other hand, high sensitivity sensors have a greater chance to be affected by various noises.

A milestone in the linking of sensors to THR was reached by Rosenstein (Rosenstein et al. 1989), who introduced analysis of resonance frequencies involving an accelerometer. By inducing a sinusoidal signal at the femoral condyles with an electrodynamic shaker, the implant bone interface was excited. The shaker was placed perpendicular to the femur in order to induce bending eigenmodes of the THR-femur-system. Bending modes have the maximum influence on the sound emission into the adjacent soft tissue. The accelerometer is placed at the major trochanter for detection of the resulting vibrations. Applying this technology, Li et al. suggest that a well fixed THR behaves linear, while a loosened THR shows a non-linear acoustic behaviour (Li et al., 1996). In general, the signal amplitude decreases as the interface failure increases due to the decreasing structural stiffness of the implant-bone-compound (Qi et al., 2003).

Puers et al. instrumented the THR with a miniaturized accelerometer placed in the ball-head of the THR. They were the first research group who developed an implantable telemetry unit for the monitoring of the vibration output signal (Puers et al., 1999) (Fig. 12). In cadaver experiments they chose an excitation frequency between 100-200 Hz, which has to be considered critical in view of the findings, that the sensitive frequency band begins at 1.5 kHz (Qi et al., 2003). Further performances of the cadaver experiments are not described.

Fig. 12. Total system for the detection of hip prosthesis loosening

One problem of the vibration analysis with in vivo accelerometer was the low signal-tonoise ratio of available accelerometers. In a further study this principle was continued by Marschner et al. (Marschner et al., 2010). Wireless lock-in measurement was introduced

they considered three different situations: distal loosening, proximal loosening and central loosening. Because soft tissue dampening and connections to the upper/ lower joints will absorb acoustic energy, sensitivity of the sensors may have to be higher to acquire the signals for further processing. On the other hand, high sensitivity sensors have a greater

A milestone in the linking of sensors to THR was reached by Rosenstein (Rosenstein et al. 1989), who introduced analysis of resonance frequencies involving an accelerometer. By inducing a sinusoidal signal at the femoral condyles with an electrodynamic shaker, the implant bone interface was excited. The shaker was placed perpendicular to the femur in order to induce bending eigenmodes of the THR-femur-system. Bending modes have the maximum influence on the sound emission into the adjacent soft tissue. The accelerometer is placed at the major trochanter for detection of the resulting vibrations. Applying this technology, Li et al. suggest that a well fixed THR behaves linear, while a loosened THR shows a non-linear acoustic behaviour (Li et al., 1996). In general, the signal amplitude decreases as the interface failure increases due to the decreasing structural stiffness of the

Puers et al. instrumented the THR with a miniaturized accelerometer placed in the ball-head of the THR. They were the first research group who developed an implantable telemetry unit for the monitoring of the vibration output signal (Puers et al., 1999) (Fig. 12). In cadaver experiments they chose an excitation frequency between 100-200 Hz, which has to be considered critical in view of the findings, that the sensitive frequency band begins at 1.5 kHz (Qi et al., 2003). Further performances of the cadaver experiments are not described.

chance to be affected by various noises.

implant-bone-compound (Qi et al., 2003).

Fig. 12. Total system for the detection of hip prosthesis loosening

One problem of the vibration analysis with in vivo accelerometer was the low signal-tonoise ratio of available accelerometers. In a further study this principle was continued by Marschner et al. (Marschner et al., 2010). Wireless lock-in measurement was introduced and the THR was instrumented with a coil, amplifier, microcontroller and telemetry unit in the distal end of the hip stem. The lock-in amplifier separates the measurement signal from noise.

The availability of results which compare different loosening states of the THR and therefore sensitivity of the proposed sensor systems is rare. Indeed, the presented devices are available, but only few results verify the possibility to detect a difference between a loose and a well fixed implant. Currently sensitivity and specificity of vibration analysis in THR can only be estimated. Georgiou et al. stated a 20 % higher sensitivity measured with vibration analysis (80 %) than with radiographs (60 %) (Georgiou et al., 2001). The finite element analysis of Qi et al. showed that a loose THR can be identified reliable if 1/3 of the femoral component has no direct bone contact (Qi et al., 2003). One of the main problems of the resonance frequency analysis of THR is the soft tissue damping of the output vibration signal. Furthermore, connection to the tibia and the pelvis can decrease the energy rate of the acoustic sound waves.

In another in vitro study the accelerometer was replaced by a blood flow ultrasound probe which enabled consistent detection of the vibrations for a loose and a fixed THR (Rowlands et al., 2008). The ultrasound probe lead to increased amplitudes of the output vibrations compared to the accelerometer signal. Dahl et al. applied ultrasound to detect osseous integration of the total ankle replacement (Dahl et al., 2010). The cadaveric testing demonstrated that the ultrasound technique could distinguish between loose and fixed implant components. In a further study, a trans-femoral implant was measured with accelerometers. It could be shown that the eigenfrequency increases with the weight bearing rehabilitation time (Shao et al., 2007).

Paech et al investigated the natural frequency of four different types of THR and determined these sound patterns in further cadaver tests with bovine and human bone (Paech et al., 2007). The excitation of the THR was reached by using an acoustically neutral steel ball of 120 g as activating impact for sound emission. However, this is not reproducible for application in clinical studies.

Besides problems in signal analysis and soft tissue dampening, functional sustainment of the integrated electronics during sterilization and impaction of the THR seems to be a significant challenge. Nevertheless detailed research into simulation of different loosening states and significant test results are expected. Furthermore the excitation has to be optimized in order to avoid pressure pain induced by the electrodynamic shaker (Georgiou et al., 2001).

### **3.3 A new excitation method to detect aseptic loosening 3.3.1 Principle of the novel excitation concept**

In the proposed concept, an oscillator integrated in the THR is used. The oscillator consists of a magnetic body which is fixed on a flat steel spring (Ruther et al., 2010) (Fig. 13). This oscillator is excited by a coil placed outside the patient. The oscillator impinges inside the THR and excites the THR to vibrate in its eigenfrequency. This can be considered as an alternative to the electrodynamic shaker, which was used by different research groups. The excitation in the implant bending modes leads to a sound emission to the surrounding bone and soft tissue. The sound waves can be detected by a vibration sensor, which is applied outside the patient's hip opposite to the excitation coil.

Current Possibilities for Detection of Loosening of Total Hip

linear-elastic materials.

sensor.

studies.

coupled accelerometer

Replacements and How Intelligent Implants Could Improve Diagnostic Accuracy 377

The material data of the components in terms of density, Young's modulus and Poisson ratio were collected from the manufacturer information. All components were modelled as

Fig. 14. Left: Prototype of a femoral component with the embedding heights tested in a cylindrical setup with epoxide resin. Middle: Model of the femoral component for finiteelement analysis exemplarily with an embedded height of 65 mm. Right: Experimental setup with embedded hip stem and the applied water reservoir with external vibration

For post-mortem investigation of a porcine leg, a small cuboid-formed implant prototype was manufactured (Fig. 15). This implant may be applied for animal experiments in further

Fig. 15. Left: Porcine femur embedded within epoxide resin. Middle: Experimental

prototype for the feasibility study in porcine bone, implanted in the femoral condyle. Right: Experimental setup with ground porcine muscle tissue around the porcine femur with the

Fig. 13. Functional principle of the novel in vivo excitation method inside the femoral component of the THR

### **3.3.2 Material and methods**

For testing of the new in vivo excitation principle a custom straight total hip stem with an integrated oscillator in the proximal part of the implant was used (Fig. 14). By embedding a femoral component of a present THR system (CBH Schaft Size 3, MATHYS European Orthopaedics AG, Bettlach, Switzerland) in epoxide resin with different embedding heights, the differences in resonance frequency were recorded. The experimental set up was placed in a water reservoir, which was the first simulation of soft tissue. This was performed according to the procedure of Paech et al. (Paech et al., 2007). The excitation was realized by a coil outside the water reservoir with a distance of 50 mm to the oscillator. The coil excited the oscillator in its first eigenfrequency of 70 Hz. A piezoelectric vibration sensor for broadband measurements between 1 and 24 kHz (Metra, Radebeul, Germany) with a high sensitivity of 100 mV/g was fixed outside the water reservoir to detect the transmitted oscillations.

In order to determine how the eigenfrequency changes with different embedding heights, we conducted a modal analysis using the finite-element method. Therefore, a model of the femoral component was constructed with a gap in the proximal part where the oscillator is placed. The FE-solver Abaqus V6.9 (Dassault Systémes, Simulia, Providence, RI, USA) was used to calculate the eigenfrequencies of the components with the given material data in table 1.


Table 1. Material properties of the straight CBH femoral component used in the numerical simulation

Fig. 13. Functional principle of the novel in vivo excitation method inside the femoral

For testing of the new in vivo excitation principle a custom straight total hip stem with an integrated oscillator in the proximal part of the implant was used (Fig. 14). By embedding a femoral component of a present THR system (CBH Schaft Size 3, MATHYS European Orthopaedics AG, Bettlach, Switzerland) in epoxide resin with different embedding heights, the differences in resonance frequency were recorded. The experimental set up was placed in a water reservoir, which was the first simulation of soft tissue. This was performed according to the procedure of Paech et al. (Paech et al., 2007). The excitation was realized by a coil outside the water reservoir with a distance of 50 mm to the oscillator. The coil excited the oscillator in its first eigenfrequency of 70 Hz. A piezoelectric vibration sensor for broadband measurements between 1 and 24 kHz (Metra, Radebeul, Germany) with a high sensitivity of 100 mV/g was fixed outside the water reservoir to detect the transmitted

In order to determine how the eigenfrequency changes with different embedding heights, we conducted a modal analysis using the finite-element method. Therefore, a model of the femoral component was constructed with a gap in the proximal part where the oscillator is placed. The FE-solver Abaqus V6.9 (Dassault Systémes, Simulia, Providence, RI, USA) was used to calculate the eigenfrequencies of the components with the given material data

Poisson ratio Density

(g/cm3)

component of the THR

oscillations.

in table 1.

simulation

Material Young's

modulus (MPa)

Ti6Al4V 114000 0.342 4.432

Table 1. Material properties of the straight CBH femoral component used in the numerical

**3.3.2 Material and methods** 

The material data of the components in terms of density, Young's modulus and Poisson ratio were collected from the manufacturer information. All components were modelled as linear-elastic materials.

Fig. 14. Left: Prototype of a femoral component with the embedding heights tested in a cylindrical setup with epoxide resin. Middle: Model of the femoral component for finiteelement analysis exemplarily with an embedded height of 65 mm. Right: Experimental setup with embedded hip stem and the applied water reservoir with external vibration sensor.

For post-mortem investigation of a porcine leg, a small cuboid-formed implant prototype was manufactured (Fig. 15). This implant may be applied for animal experiments in further studies.

Fig. 15. Left: Porcine femur embedded within epoxide resin. Middle: Experimental prototype for the feasibility study in porcine bone, implanted in the femoral condyle. Right: Experimental setup with ground porcine muscle tissue around the porcine femur with the coupled accelerometer

Current Possibilities for Detection of Loosening of Total Hip

and 65 mm embedded hip stem.

advanced loosening is obvious.

Replacements and How Intelligent Implants Could Improve Diagnostic Accuracy 379

 Fig. 17. Left: Characteristic voltage signal of the produced structure-borne sound of the 65 mm embedded CBH hip stem proportional to acceleration recorded by the detection coil. Right: Frequency spectrum using Fast Fourier Transformation. Comparison between 20 mm

The characteristic progression of the small cuboid-formed prototype is presented in figure 18. As seen in the experiments with the femoral component, the fixed prototype reveals a higher dampening of the sensor signal compared to the simulation of complete loosening. The displacement of the balance point in the area of lower frequency in case of

Fig. 18. Characteristic voltage signal of the produced structure-borne sound in the press

fitted implant prototype proportional to acceleration.

In a first experimental in vitro set-up, the implant was inserted in the femoral condyles with press fit fixation. In order to simulate an advanced loosening the bore whole was reamed, filled with water and then the implant was reinserted. For simulation of soft tissue, ground porcine muscle tissue was arranged around the femur. The vibration sensor was placed perpendicular to the femur to detect bending eigenfrequencies.

### **3.3.3 Results and discussion**

The results of the numerical modal analysis show that the higher the embedding height of the femoral component, the higher the eigenfrequency due to an increased system stiffness (Fig. 16). According to Qi et al., the highly sensitive frequency band is over 2.5 kHz due to higher differences between the eigenfrequencies (Qi et al., 2002).

Characteristic progressions of the structure borne sound are shown in figure 17 exemplarily in case of the 65 mm embedding height (well fixed THR) and the 20 mm embedding representative for a loose THR. The signal of the 65 mm embedded THR resulted in a higher dampening than the 20 mm embedded component simulating loosening. As a result the time in which the vibration signal achieved the complete deactivation of the excited state varies. The well fixed femoral component resulted in an oscillation time of 1.6 ms. In contrast to the stable component, the THR with an unstable seating (20 mm) showed an oscillation time of 3.3 ms. After Fast Fourier Transformation (FFT), frequency shifts and dampening of the spectral amplitude were evaluated. In the frequency spectrum the system with the loose THR had its first eigenfrequency at 328 Hz, while the first eigenfrequency of the system with the well fixed THR was at 926 Hz. The amplitudes in the frequency graph are dampened by about 75 % by increasing the embedding height from 20 to 65 mm.

Fig. 16. Results of the eigenfrequencies with increasing embedding height of the CBH femoral component using finite-element-analysis

In a first experimental in vitro set-up, the implant was inserted in the femoral condyles with press fit fixation. In order to simulate an advanced loosening the bore whole was reamed, filled with water and then the implant was reinserted. For simulation of soft tissue, ground porcine muscle tissue was arranged around the femur. The vibration sensor was placed

The results of the numerical modal analysis show that the higher the embedding height of the femoral component, the higher the eigenfrequency due to an increased system stiffness (Fig. 16). According to Qi et al., the highly sensitive frequency band is over 2.5 kHz due to

Characteristic progressions of the structure borne sound are shown in figure 17 exemplarily in case of the 65 mm embedding height (well fixed THR) and the 20 mm embedding representative for a loose THR. The signal of the 65 mm embedded THR resulted in a higher dampening than the 20 mm embedded component simulating loosening. As a result the time in which the vibration signal achieved the complete deactivation of the excited state varies. The well fixed femoral component resulted in an oscillation time of 1.6 ms. In contrast to the stable component, the THR with an unstable seating (20 mm) showed an oscillation time of 3.3 ms. After Fast Fourier Transformation (FFT), frequency shifts and dampening of the spectral amplitude were evaluated. In the frequency spectrum the system with the loose THR had its first eigenfrequency at 328 Hz, while the first eigenfrequency of the system with the well fixed THR was at 926 Hz. The amplitudes in the frequency graph

are dampened by about 75 % by increasing the embedding height from 20 to 65 mm.

Fig. 16. Results of the eigenfrequencies with increasing embedding height of the CBH

femoral component using finite-element-analysis

perpendicular to the femur to detect bending eigenfrequencies.

higher differences between the eigenfrequencies (Qi et al., 2002).

**3.3.3 Results and discussion** 

Fig. 17. Left: Characteristic voltage signal of the produced structure-borne sound of the 65 mm embedded CBH hip stem proportional to acceleration recorded by the detection coil. Right: Frequency spectrum using Fast Fourier Transformation. Comparison between 20 mm and 65 mm embedded hip stem.

The characteristic progression of the small cuboid-formed prototype is presented in figure 18. As seen in the experiments with the femoral component, the fixed prototype reveals a higher dampening of the sensor signal compared to the simulation of complete loosening. The displacement of the balance point in the area of lower frequency in case of advanced loosening is obvious.

Fig. 18. Characteristic voltage signal of the produced structure-borne sound in the press fitted implant prototype proportional to acceleration.

Current Possibilities for Detection of Loosening of Total Hip

No.9, p.648, ISSN 0147-7447

0009-921X

0268-0033

Replacements and How Intelligent Implants Could Improve Diagnostic Accuracy 381

Bergmann G, Deuretzbacher G, Heller M, Graichen F, Rohlmann A, Strauss J, Duda GN

Böhler, M., Knahr, K., Riegler, M., & Salzer, M. (1994). Ten-year results of cemented Weller-

*of Orthopaedic and Trauma Surgery*, Vol.113, No.2, pp. 57–60, ISSN 0936-8051 Chryssikos, T., Parvizi, J., Ghanem, E., Newberg, A., Zhuang, H., & Alavi, A. (2008). FDG-

Cuckler, J. M. (2010). Unexplained pain after THR: what should i do? *Orthopedics*, Vol.33*,* 

Dahl, M. C., Kramer, P. A., Reinhall, P. G., Benirschke, S. K., Hansen, S. T., & Ching, R. P.

DeLee, J. G., & Charnley, J. (1976). Radiological demarcation of cemented sockets in total hip

Dihlmann, W., Dihlmann, S. W., & Hering, L. (1991). Alloarthroplasty of the hip joint.

Gelman, M. I., Coleman, R. E., Stevens, P. M., & Davey, B. W. (1978). Radiography,

Georgiou, A. P., & Cunningham, J. L. (2001). Accurate diagnosis of hip prosthesis loosening

Ginai, A. Z., van Biezen, F. C., Kint, P. A., Oei, H. Y., & Hop, W. C. (1996). Digital

arthroplasty. *Skeletal Radiology*, Vol.25, No.4, pp. 357–363, ISSN 0364-2348 Glauser, R., Sennerby, L., Meredith, N., Ree, A., Lundgren, A. K., Gottlow, J., et al. (2004).

Gruen, T. A., McNeice, G. M., & Amstutz, H. C. (1979). "Modes of failure" of cemented stem-

Hardy, D. C., Reinus, W. R., Totty, W. G., & Keyser, C. K. (1988). Arthrography after total

Hendrix, R. W., Wixson, R. L., Rana, N. A., & Rogers, L. F. (1983). Arthrography after total

replacement. *Radiology*, Vol.128, No.3, pp. 677–682, ISSN 0033-8419

*and Related Research,* Vol. 466, No.6, pp.1338-1342, ISSN 0009-921X

*Biomechanics*, Vol.34, No.7, pp. 859-871, ISSN 1435-2443

*Research*, Vol.54, No.1, pp. 12–19, ISSN 0021-9304

*Radiologe*, Vol.31, No.10, pp. 496–505, ISSN 0033-832X

*Research*, Vol.15, No.4, pp. 428–434, ISSN 0905-7161

No.1, pp. 20–23, ISSN 0364-2348

Vol.148, No.3, pp. 647–652, ISSN 0033-8419

*and Related Research*, No.141, pp. 17–27, ISSN 0009-921X

(2001). Hip contact forces and gait patterns from routine activities. *Journal of* 

type total hip arthroplasties. Analysis using different definitions of failure. *Archives* 

PET imaging can diagnose periprosthetic infection of the hip, *Clinical Orthopaedics* 

(2010). The efficacy of using vibrometry to detect osteointegration of the Agility total ankle. *Journal of Biomechanics*, Vol.43, No.9, pp. 1840–1843, ISSN 0021-9290 DeHeer, D. H., Engels, J. A., DeVries, A. S., Knapp, R. H., & Beebe, J. D. (2001). In situ

complement activation by polyethylene wear debris. *Journal of Biomedical Materials* 

replacement. *Clinical Orthopaedics and Related Research*, No.121, pp. 20–32, ISSN

Radiologic diagnosis of loosening and infection in cemented total endoprostheses.

radionuclide imaging, and arthrography in the evaluation of total hip and knee

using a vibrational technique. *Clinical Biomechanics*, Vol.16, No.4, pp. 315–323, ISSN

subtraction arthrography in preoperative evaluation of painful total hip

Resonance frequency analysis of implants subjected to immediate or early functional occlusal loading. Successful vs. failing implants. *Clinical Oral Implants* 

type femoral components: a radiographic analysis of loosening. *Clinical Orthopaedics* 

hip arthroplasty: utility of postambulation radiographs. *Skeletal Radiology*, Vol.17,

hip arthroplasty: a modified technique used in the diagnosis of pain. *Radiology*,

These first test results demonstrate differences between two different states of fixation of both the femoral component and a prototype for latter animal experiments. Hence, this confirms the potential usability of the novel non-invasive approach for detecting implant loosening. In further tests, the influence of the femur-tibia and the femur-pelvis connection has to be investigated. Additionally, the femoral component will be integrated in a human cadaveric femur and measured with defined loosening zones. Moreover, the oscillators could be used in animal studies in order to determine the quality of osseointegration e.g. of new implant coatings and materials.
