**2.4 FDG-PET**

Based on the principle of scintigraphy, fluorodeoxyglucose-positron emission tomography (FDG-PET) uses radionuclides (FDG) which visualize the transport and metabolic rate of

Fig. 8. (A) 18F-FDG-PET coronary slices 7 mm thick showing right-sided artificial hip joint in the patient. Increased glucose metabolism particularly around prosthesis neck and polyethylene inlay (arrowheads) is seen. (B) 18FFDG-PET of a patient with a clinically inconspicuous right-sided hip prosthesis without suggestive signs concerning the implant. (C) 18F- DGPET showing bilateral hip joint implants of a patient without any clinical discomforts of the hips. There are no signs of enhanced glucose consumption in the joint capsules as well as around both prostheses (Kisielinski et al., 2003)

image the diagnosis of infection and septic loosening is positive. Due to a normal bone scan, the existence of a sepsis can be excluded and an additional white blood cell imaging should

Mean sensitivity and specificity of scintigraphy are at 85 % and 72 % (Temmermann et al. 2005) (Fig. 7). Reinartz found a mean sensitivity/specificity of 78 % /84 % for diagnosing THR loosening with triple-phase bone scans (Reinartz 2009). It was also concluded, that bone scans of THR's are highly sensitive but not specific (Segura et al., 2004). The lowest sensitivity was found by Zilkens et al. with a value of 50 % (Zilkens et al., 1988), while the

Based on the principle of scintigraphy, fluorodeoxyglucose-positron emission tomography (FDG-PET) uses radionuclides (FDG) which visualize the transport and metabolic rate of

Fig. 8. (A) 18F-FDG-PET coronary slices 7 mm thick showing right-sided artificial hip joint in the patient. Increased glucose metabolism particularly around prosthesis neck and polyethylene inlay (arrowheads) is seen. (B) 18FFDG-PET of a patient with a clinically inconspicuous right-sided hip prosthesis without suggestive signs concerning the implant. (C) 18F- DGPET showing bilateral hip joint implants of a patient without any clinical discomforts of the hips. There are no signs of enhanced glucose consumption in the joint

capsules as well as around both prostheses (Kisielinski et al., 2003)

be taken into account.

**2.4 FDG-PET** 

lowest specificity was 38 %(Ovesen et al., 2003).

glucose by emitting positrons for the generation of cross-sections (Fig. 8). FDG-PET is one of the most expensive imaging modalities and has already been documented to be a precise method to diagnose infection particularly (Chryssikos et al., 2008). Active cells like leukocytes and macrophages show a higher energy demand compared to other cells and therefore preferred ingest the FDG (Reinartz, 2009). Due to this fact, infected tissue with a high number of active cells revealed an increased uptake of FDG, which leads to a positive finding in the PET scan (Stumpe & Strobel, 2006). Furthermore, higher FDG uptake in patients with aseptic loosening of the THR can be explained by wear debris and the development of granulomatous tissue in the periprosthetic membrane around the THR (DeHeer et al., 2001; Kadoya et al., 1998).

The criteria for the interpretation of a PET scan differ between different studies (Reinartz et al., 2005). In the area of the femoral component an increased uptake of FDG can be interpreted as loosening or infection, while an increased uptake at the femoral head and neck cannot be determined as loosening or unspecific (Zhuang et al., 2007). In contrast the increased uptake of FDG at the distal tip of the stem is unspecific. Furthermore, a high uptake at the implant-bone interface of the acetabular component can be identified as a pathologic process or a sign of loosening (Mayer-Wagner et al., 2009).

Based on the great variety of interpretation criteria to analyse PET scans, sensitivities of PET to implant loosening differ from 33 % (Stumpe et al., 2004) to 100 % (Zhuang et al., 2002), while specificity varies from 78 % (Vanquickenborne et al., 2003) to 100 % (Zhuang et al., 2002) with an average of 85 %/90 % and an accuracy of 89 % (Reinartz, 2009) (Fig. 9).

Fig. 9. Literature comparison of the results determining sensitivity and specificity of loosening diagnosis of THR using FDG-PET

### **2.5 Summary of actual loosening diagnostics using imaging techniques**

Figure 10 shows the comparison of the mean sensitivities and specificities of the presented imaging methods to diagnose loosening of the THR. FDG-PET has both, the highest sensitivity and specificity and therefore is the most accurate method to diagnose loosening exactly. Although the high accuracy, FDG-PET is highly cost-intensive, for that reason it has not reached the status of a standard examination.

Current Possibilities for Detection of Loosening of Total Hip

parameter for additional reference measurements.

status of loosening of the THR

**3.2 Vibrometry** 

dampening factors.

Replacements and How Intelligent Implants Could Improve Diagnostic Accuracy 373

direct contact with the THR. The resistance inductance bridge based technique enables small axial displacements of a rod to be measured about a pre-determined zero reference point. Detection of acoustical properties includes measuring of the resonance frequency which correlates with the stiffness of the implant-bone interface (Li et al., 1996). Resonance frequency and dampening are the most frequently measured properties and reached the highest development status compared to other detection methods using in vivo sensors. This method is called vibrometry. Measuring the biological properties e.g. the temperature at the border of the implant, can only be used in cases of infection and is even more a

Fig. 11. Overview of the possible parameters which can be detected in order to identify the

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

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,

Fig. 10. Comparison of different diagnostic imaging methods with regard to sensitivity and specificity of loosening diagnosis of the femoral component of THRs

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 to a higher level.
