A NEW METHOD FOR DETECTION

OF TOTAL HIP REPLACEMENT LOOSENING

Development and First Results of a Novel Mechano-acoustical Sensor

Cathérine Ruther, Daniel Kluess, Andreas Fritsche, Wolfram Mittelmeier, Rainer Bader

Department of Orthopaedics, University of Rostock

Doberaner Str. 142, 18057 Rostock, Germany

Hartmut Ewald

Department of General Electronics, University of Rostock, Justus-von-Liebig-Weg 2, 18059 Rostock, Germany

Keywords: Total hip replacement, Implant loosening, Sensor, Diagnosis tool.

Abstract: Currently applied diagnostic methods for loosening of total hip replacements often result in imprecise

identification of implant fixation and in the worst case unnecessary revision surgery. Developed sensors

integrated in implants require adequate energy supply, which in most cases is achieved by inductive

coupling and complex data telemetry. In order to avoid a telemetric apparatus, we developed a passive

concept of a novel in vivo method to improve diagnostic investigations of total hip replacement loosening.

A mechano-acoustical sensor, attached on small membranes inside the femoral hip stem, is proposed and

enables osseous anchorage detection. The sensor is excited and detected by extracorporeal coils. First

functional models show significant differences between different material layers located at the membrane.

The novel in vivo sensor system has a promising potential to detect implant loosening.

1 INTRODUCTION

Prediction of total hip replacement (THR) loosening

using ex vivo techniques such as imaging often fails

to deliver reliable and localized results

(Temmerman, 2005). Moreover, monitoring of

implant anchorage is an important factor in

determining the longterm success rate of implants

(Pastrav, 2009). Currently diagnostic methods used

to identify THR loosening do not provide satisfying

results. Hence, researchers strive for supporting

techniques or even attempt to replace ex vivo

techniques by novel in vivo sensors such as

accelerometers in vibrometry (Li, 1995), which

showed that implant instability can be identified by

several harmonics in the output signal of the sensors.

However, the low signal-to-noise ratio of presently

available accelerometers needs to be addressed

(Marschner, 2007).

Besides miniaturized design and

biocompatibility, a reliable power supply is a

challenge in development of in vivo sensors. The

technology of choice for vibrometry is wireless

powering and data telemetry (Puers, 1999). With this

system, energy can be provided externally and

transcutaneous. This is achieved by a primary coil,

used outside the human body, and a secondary coil

integrated in the implant in vivo.

Inductive powering and passive data telemetry is

effective for implants where both coils are

positioned with small separation distance and fixed

orientation. In many cases this cannot be achieved

because of possible changes in the patients´ weight.

The influence of variable distances between both

coils leads to lower coupling factors and therefore

poor signal transmission. Additionally, interferences

of the inductive telemetry during patients´

movements could occur. Another problem is the use

of only low frequencies, which affect the damping of

the signals caused by eddy currents, especially in

metals. This is an evident drawback for orthopaedic

implants, especially for THR where biocompatible

metal alloys are used.

Due to the aforementioned reasons we propose a

novel and passive sensor concept without inductive

Ruther C., Ewald H., Mittelmeier W., Kluess D., Fritsche A. and Bader R..

A NEW METHOD FOR DETECTION OF TOTAL HIP REPLACEMENT LOOSENING - Development and First Results of a Novel Mechano-acoustical

Sensor.

DOI: 10.5220/0003165900700073

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2011), pages 70-73

ISBN: 978-989-8425-37-9

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

coupling of two coils. The concept is characterized

by extracorporeal coils and permanent magnets

inside the THR acting as loosening sensors. In the

present work, we demonstrate the development of a

custom trigger circuit as well as first results gained

by functional models of a sensor prototype.

2 METHODS

2.1 Principle of the Loosening Sensor

The proposed method is based on the conservation

of momentum including impingement of a magnetic

body on a membrane placed inside the femoral hip

stem (Figure 1). The magnetic body, fixed on a flat

spring, is part of the sensor, which is designed as a

simple mechanical oscillator for impinging in the

middle of a membrane (Ruther, 2010). The magnetic

poles allow oscillations using an external magnetic

field impulse. Thus, motion of the oscillator is

initiated using an external coil.



Figure 1: System concept of the novel sensors for

excitation of the THR inside the hip stem and wireless

detection of implant loosening.

The end velocity of the oscillator after

impingement varies depending on the

osseointegration of the implant. On the one hand, a

well osseous integrated implant leads to lower end

velocities of the oscillator after impingement caused

by the higher energy transfer of the oscillator due to

lower deformation energy. Therefore, the spring

back of the sensor will be lower. On the other hand,

a loosened THR with soft tissue on the external

surface of the membrane results in higher end

velocities caused by higher deformation energy.

Since the oscillator is magnetic, the read out of

the measured data can also be achieved by an

external coil.

2.2 Experimental Setup

In order to demonstrate the functional principle of

our loosening detection approach, a cylindrical setup

composed of non magnetic aluminium was built. A

0.25 mm thick titanium membrane was mounted to

the test device and tightened by screws to ensure

symmetrical oscillations of the membrane

(Figure 2). The oscillator was made of a steel spring

and an attached magnetic neodymium body (1.3T).

Figure 2: Experimental setup for the new loosening

sensor. Left: Fixing device with oscillator, which impinge

on a titanium membrane. Right: Cross section of the setup

with different material layers attached to the membrane.

The influence of different material layers on the

impingement behaviour of the oscillator was also

tested. Thus, artificial bone substitute material

(20 pcf, Sawbones, Malmö, Sweden) with a

thickness of 50 mm was used to simulate full

osseous implant integration. A further setup with

additional gelatine layers of 10 mm and 5 mm

thickness, simulating a loose implant was

investigated (Figure 2). Gelatine was used to

represent collagen and fluid outside the membrane

as in the case of implant loosening. In a final setup,

small gaps were included in the artificial bone in

order to simulate bone defects adjacent to the

membrane (partial osteolysis). Measurements for

each material setup were repeated 20 times.

2.3 Power Supply and Detection

2.3.1 Inductive Unit

The inductive unit is composed of two coils and

shall be placed outside the patients` leg in defined

distances to the oscillators.

In the experimental setup (Figure 3) the

excitation coil made of a ferrite core to concentrate

A NEW METHOD FOR DETECTION OF TOTAL HIP REPLACEMENT LOOSENING - Development and First Results

of a Novel Mechano-acoustical Sensor

the streamlines of the magnetic field on the

oscillator inside the implant. Ferrite cores allow

faster reduction of the magnetic field than iron cores.

The detection coil included an iron core with a

diameter of 2.3 mm for precise detection of the

oscillator signal.

Figure 3: Arrangement of the excitation and detection

coils around the oscillator.

2.3.2 Trigger circuit

For adequate configuration of the trigger circuit

(Figure 4), it was important that the magnetic field

of the excitation coil (L1) reduces very fast.

Otherwise, the signal of L1 overlaps the signal of the

oscillator. Therefore the external hardware included

a central processing unit (CPU) where the sequence

control could be achieved by different counters,

configured by a custom Labview program (NI 9.0,

TX, USA). Counter 0 defined past which time

counter 1 was closed. Counter 1 determined the time

how long switch 1 was closed. An overflow of

counter 0 resulted in an increasing edge (CTR 0

OUT), thus signalizing that switch 1 could be closed

due to the excitation time and displacement of the

oscillator. Simultaneously, the release of counter 1

was introduced (CTR 1 GATE). The signal for the

overflow of counter 1 was CTR 1 OUT.

Figure 4: Block diagram of the external hardware.

A recovery diode enabled the fast decrease of the

excitation magnetic field. Switch 2 was controlled

by a digital signal defining the time when the

detection coil (L2) could monitor the oscillator

signal without detection of the excitation magnetic

field. The signal of L2 was amplified and evaluated

in the Labview program. Parameters such as the

highest amplitude after impingement and amplitudes

in a frequency spectrum were evaluated.

3 RESULTS

The detection of the oscillation signal of the

loosening sensor resulted in a good differentiation of

the varying material layers at the external side of the

membrane. For evaluation the first amplitude in the

time signal proportional to the highest velocity of

the oscillator after impingement was considered

(Figure 5). The results are presented as mean values

± standard deviation in Figure 6.

Figure 5: Time signal of the oscillation of the loosening

sensor for artificial bone with partial osteolysis.

Figure 6: Results of the amplitude measurements with

different material layers attached to the external side of the

membrane.

Furthermore, the signal in the frequency domain

as spectral analysis with a Fast Fourier

Transformation was investigated (Figure 7). The

BIODEVICES 2011 - International Conference on Biomedical Electronics and Devices

first eigen frequency of the oscillator was identified

at 67 Hz. The peak of the second harmonic was at

134 Hz. Comparison of different material layers

revealed decreased amplitudes with a thicker

gelatine layer.

Figure 7: Frequency spectrum of the oscillations of the

loosening sensor.

4 DISCUSSION

With respect to the standard application of total hip

replacements, unsatisfying results in loosening

diagnosis increase the demand for more precise in

vivo techniques. Inductive coupling based on radio

frequency powering to provide energy supply of in

vivo sensors causes problems, especially with regard

to the coupling factor between the two coils. To

circumvent this problem, our approach was to show

the capability of a novel sensor principle using basic

research models.

The trigger circuit to control the excitation and

detection coils allows the identification of the

highest velocity after impingement of the oscillator

on the membrane. This leads to promising

experimental results, which show the usability of the

described in vivo method for detecting total hip

replacement loosening with extracorporeal coils and

a passive internal sensor.

The robustness of the oscillators due to the

simplicity of the assembly guarantees functionality

during intraoperative impaction and sterilization of

the implant. Moreover, the oscillators can be used in

experimental applications to determine the quality of

osseointegration of new coated implant materials.

5 CONCLUSIONS

In the present study, a new measurement method for

in vivo diagnosis of total hip replacement loosening

without inductive coupling based radio frequency

powering was demonstrated. The described

loosening sensors with two coils and a custom

trigger circuit shows results with good prospects in

preliminary tests.

Future work will include the implementation of

the loosening sensor in real implants. Furthermore,

solutions for continuous excitation and therefore the

optimization of the trigger circuit are aspired.

Enhancements of the inductive unit are

designated with respect to air-core coils as detection

coils, which are switched as differential coils for

better erasure of the excitation field during

detection.

ACKNOWLEDGEMENTS

This research project is granted by the German

research foundation (DFG) under Reg.No. KL 2327.

REFERENCES

Temmerman, O. P., Raijmakers, P. G., Berkhof, J.,

Hoekstra, O. S., Teule, G. J., Heyligers, I. C., 2005.

Accuracy of diagnostic imaging techniques in the

diagnosis of aseptic loosening of the femoral

component of a hip prosthesis. In: Journal of Bone

and Joint Surgery, 87, pp. 781-785.

Pastrav, L. C., Jaeques, S. V. N., Jonkers, I., Van der

Perre, G., Mulier, M., 2009. In vivo evaluation of a

vibration analysis technique for the per-operative

monitoring of the fixation of hip prostheses. In:

Journal of Orthopaedic Seurgery and Research. 4, pp.

1-10.

Marschner, U., Jettkant, B., Ruwisch, D., Zhu, Y., Grätz,

H., Fischer W. J., Clasbrummel, B., 2007. FEM

simulation and wireless measurement of hip prosthesis

vibrations for loosening detection. European

Symposium , Technical Aids for Rehabilitation, 25-26

January 2007, Berlin, pp. 87-88.

Puers, R., Catrysse, M., Vandevoorde, G., Collier, R. J.,

Louridas, E., Burny, F., Donkerwolcke, M., Moulart,

F.,1999. A telemetry system for the detection of hip

prosthesis loosening by vibration analysis. In: Sensors

and Actuators, 85, pp. 42-47.

Ruther, C., Ewald, H., Biemann, A., Nierath, H., Bader,

R., Kluess, D., 2010. A New Concept for Non-

Invasive Radiation-Free Detection of Implant

Loosening. 56

Annual Meeting of the Orthopaedic

Research Society, New Orleans, USA, 6-9 March

2010, #2413.

A NEW METHOD FOR DETECTION OF TOTAL HIP REPLACEMENT LOOSENING - Development and First Results

of a Novel Mechano-acoustical Sensor