Original contributions
Low back pain, the stiffness of the sacroiliac joint: A new method using ultrasound

https://doi.org/10.1016/j.ultrasmedbio.2004.09.014Get rights and content

Abstract

Abnormal biomechanical properties of the sacroiliac joints are believed to be related to low back and pelvic pain. Presently, physiotherapists judge the condition of the sacroiliac joints by function and provocation tests, and palpation. No objective measuring device is available. Research is ongoing to identify the biomechanical properties of the sacroiliac joints from the dynamic behaviour of the pelvic bones. A new concept based on ultrasound (US) for the measurement of bone vibration is under investigation. The objective of this study was to validate this concept on a physical model and to assess the applicability in vivo. A model consisting of a piezo shaker covered by a layer of US transmission gel (representing bone and soft tissue) has been used. A packet of US detection signals is directed onto the shaker and correlation-based processing is used to estimate the difference in time-of-flight of their echoes. These variations of time are used to compute the displacement of the shaker at each pulse reflection. To assess the validity of our US technique, we compared the obtained measurements with the readings of the built-in strain gauge sensor. The experimental procedure has been tested on a volunteer where low-frequency excitation was provided through the ilium and vibration detected on the sacrum and ilia. The results demonstrated that the correlation-based approach is capable of reproducing the piezo shaker displacements with high accuracy (± 7%). Vibration amplitudes from 0.25 μm to 3 μm could be measured. The US technique was able to detect bone vibration in vivo. In conclusion, the principle based on US waves can be used to develop a new measurement tool, instrumental in studying the relation between the biomechanical properties of the sacroiliac joints and low back pain. (E-mail: [email protected])

Introduction

Low back pain is one of the major causes of discomfort in the world. The origin of low back pain is the focal point of many investigations: in some cases, pathology can be identified during medical screening but, in most other cases, there is no medical evidence of the cause of the complaints (Kovacs et al. 2004; Mofidi et al. 2003).

Abnormal biomechanical properties of the sacroiliac joints are considered as a potential source of low back pain (Pool-Goudzwaard et al. 2003; Sims and Moorman 1996; Snijders et al. 1993). Function tests, provocation tests and palpation are among the techniques that are frequently used in physiotherapy and orthopaedic therapy to obtain information on possible manifestation of these joint abnormalities (hypermobility or hypomobility) (Bickley and Szilagyi 2002; Greene and Heckman 1994; Hoppenfeld 1976). However, in specific joints, such as the sacroiliac, carpal and tarsal joints, these clinical examinations are impaired by limited excursion of the joints and by restricted accessibility to the region-of-interest (ROI). To describe these movements in the sacroiliac joints under different static loads, an x-ray technique as well as a conventional light photography technique have previously been explored (Jacob and Kissling 1995; Sturesson et al. 1989). In the x-ray technique, radiopaque markers can be inserted into the sacrum and ilium, and displacement of the markers after joint manipulation is measured on x-ray photographs. In the conventional photography technique, external markers are introduced percutaneously. During different physiological exercises, rotations and translations of the ilium in relation to the sacrum can be measured using either of these techniques.

However, there are several problems associated with the techniques described above for joint mobility assessment. In function and provocation tests, neither the applied stimulus nor the resulting response can be quantified. The diagnosis depends upon the perception and experience of the clinician. Function and provocation tests are, therefore, highly subjective and their diagnostic value is generally quite low, particularly in joints with limited excursions (Dreyfuss et al. 1996; van der Wurff et al. 2000a, 2000b). In addition, any diagnostic procedure using X rays is subject to the well-established problems, hazards and restrictions of such techniques, and can cause secondary effects to the subjects under examination. The positioning of markers for assessment of range of motion in joints with limited accessibility requires at least minimally invasive procedures, which increase the costs and risks of the procedures and discomfort to the patient.

Moreover, all the methods used to determine joint movements discussed above are only suitable for the determination of the static response of the joint. One important source of information for the determination of biomechanical properties is the dynamic response of the bones. For in vivo conditions, this dynamic response is expected to be of much lower amplitude (in the range of micrometres), than in a static test (typically in the order of a few millimetres) (Sturesson et al. 1989). The methods of measurement of static response are not sufficiently sensitive to measure displacements in the lower order of magnitude required for dynamic response.

Measurement of minimal movements or displacements of structures in the human body is a common issue in the medical field; estimation of blood flow velocity (Ferrara and DeAngelis 1997), tissue motion, and elasticity (Ophir et al. 1991) are based on the estimation of displacement through echo time-delay analysis. A recent technology that takes advantage of time-delay estimations is intravascular ultrasound (US) elastography (de Korte et al. 2003). In this technique, local strain in the artery wall and plaques is assessed by looking at the deformation of the tissue. Snijders et al. (1993) proposed mechanical excitation of the pelvic girdle to assess joint laxity, whereas Buyruk et al. (1995) introduced an US technique to measure the induced vibrations of the pelvic bones. These measurements were used to categorise laxity of the pelvic joints. The subject was examined while lying in a prone position on a physiotherapy table. The pelvis was unilaterally supported by a vibrating surface, located at the spina iliaca anterior superior (SIAS). This vibrating surface transmits vibrations to the ilium. To pick up the vibrations of the bones, a linear-array transducer was positioned over the dorsal side of the sacroiliac joint. Detection of the vibrations was performed by means of an US scanner operating in colour Doppler imaging mode. The transmission of the vibrations from the ilium to the sacrum was believed to be deducible from the obtained colour flow map and was then used to rank the laxity of the intermediate sacroiliac joint. However, vibration parameters that characterise the dynamic response of the bones lie outside the range that can be quantified using this method. In addition, the correlation between the colour maps obtained and the actual vibration of the pelvis was not validated.

A further development in the investigation described above was to introduce a computer model of the human pelvis for a quantitative interpretation of the measured vibration. Unlike the work of Buyruk and colleagues, in this approach, the identification of the biomechanical properties of the sacroiliac joints (i.e., stiffness and damping) is based on a mathematical algorithm that uses the measured vibrations as an input (Conza and Rixen 2004). Relevant information in the dynamic response is expected within the frequency range of 10 up to 500 Hz and amplitudes in the interval of 0.1 up to 100 μm. The overall goal of the investigation is to develop an appropriate laboratory setup to be downscaled further to equipment that can be used in daily practice of physiotherapy, orthopaedics and rehabilitation medicine.

One of the challenges of the investigation described above is the noninvasive measurements of the vibration of the pelvic bones. Instead of exploiting the frequency shift of the returning US waves (Doppler effect), in our approach we base the measurement of the bone vibration on the time-delay estimation of the US waves. Although theoretically solid, the principle has to be verified and validated in practice by a series of experiments where the measurements obtained are compared to calibrated reference values.

The aim of this study is to validate the proposed US technique within the relevant working range and to demonstrate the feasibility of the experimental approach in vivo.

Section snippets

Materials and methods

A physical model (Fig. 1) consisting of a piezo shaker (P-841, Physik Instruments (PI) GmbH & Co., KG, Karlsruhe/Palmbach, Germany) surmounted by a layer of US transmission gel represents a layered structure of bone and tissue. The piezo shaker, equipped with a strain gauge position sensor was operated in a closed loop. The control input was sinusoidal with preset frequencies of 80, 180 and 280 Hz and amplitudes from 250 nm up to 3 μm. An US transducer (Aerotech, Agfa, Rijswijk, the

Results

Figure 5 shows an example of the measured displacement as a function of time. The piezo shaker was vibrating at 180 Hz with displacement amplitude of 2.5 μm. The measured displacement curve as calculated using time-delay measurements reproduces adequately both amplitude and frequency of the vibrations.

To estimate the accuracy of the US measurement system with respect to the strain gauge position sensor, we analysed a total of 66 measurement results obtained by changing the displacement

Discussion

Ultrasound with a frequency of 5 MHz has a wavelength of 0.3 mm. Using the phase of the reflected signal, an accurate position, down to 1 μm or less, of the reflector can be measured as shown in this study. If our identification method described in the introduction requires higher accuracy, then further improvements using signal processing and US parameter detection can be applied.

The amplitude measurement error of the US system increases with decreasing amplitude and increasing frequency of

Conclusion

In this paper, a new noninvasive method to measure bone vibrations was presented. In vitro evaluation of the experimental procedure showed that displacement amplitudes as low as 0.25 μm can be detected with a maximum error of ± 7%. Preliminary volunteer tests have demonstrated the ability of the technique to detect low-frequency vibrations of the pelvic bones. The reproducibility will be the focus of further investigation.

From these preliminary results, it is concluded that this approach based

Acknowledgements—

This research was supported by the Technology Foundation STW, Applied Science Division of NWO and the Technology Programme of the Ministry of Economic affairs of the Netherlands.

References (18)

There are more references available in the full text version of this article.

Cited by (26)

  • From conventional sensors to fibre optic sensors for strain and force measurements in biomechanics applications: A review

    2014, Journal of Biomechanics
    Citation Excerpt :

    As pointed by Ladd and Quick (2000) ferromagnetic based sensors should not be used in magnetic resonance imaging (MRI) because they will act as an antenna and generate significant heating effects, which might cause image artefacts. Different methods have been used to measure bone deformation under load, including SG (Szivek and Gharpuray, 2000), extensometer (Perusek et al., 2001), ultrasound (Vlannderen et al., 2005) and echo tracking (Matsuyama et al., 2006). The SG is the gold standard for measuring bone deformation in vivo (Lanyon et al., 1975; Szivek et al., 2005).

  • In the trail of a new bio-sensor for measuring strain in bone: Osteoblastic biocompatibility

    2011, Biosensors and Bioelectronics
    Citation Excerpt :

    Information regarding the adaptive response of bone to load is necessary to implement prevention programmes of bone loss and to design, evaluate and optimize treatment options of bone repair and regeneration, namely those involving bone engineering strategies. A variety of methods have been tried to measure bone deformation under load, including strain gauge (Szivek and Gharpuray, 2000), extensometer (Perusek et al., 2001), ultrasound (Vlannderen et al., 2005) and echo tracking (Matsuyama et al., 2006). The strain gauge, a device with electrical resistance varying in proportion to the amount of strain, is the gold standard for measuring bone deformation in vivo (Szivek and Gharpuray, 2000; Lanyon et al., 1975).

  • Influence of frequency-dependent properties on system identification: Simulation study on a human pelvis model

    2007, Journal of Sound and Vibration
    Citation Excerpt :

    The rationale behind this research project, which has been hypothesized and investigated in previous works [1,2] but which is still controversial, is that abnormal biomechanical properties of the sacroiliac joints can generate pain. Responding to the lack of objective diagnostics in this matter, the main goal of this research is the development of a diagnostic tool for an objective non-invasive measurement of those properties, comprising an apparatus for the bone vibration measurements by ultrasound [3,4] and an updating algorithm to process the measurements. A dynamic model of the human pelvis has been built to support the identification of the biomechanical properties of the sacroiliac joints.

  • Vibration testing of a fresh-frozen human pelvis: The role of the pelvic ligaments

    2007, Journal of Biomechanics
    Citation Excerpt :

    At this point the model is said to be updated, and its mechanical properties are derived. In the frame of our research, an apparatus intended for the introduction of vibrations in the human pelvis and for the ultrasound measurement of its response is being developed (Vlaanderen et al., 2005; Conza et al., 2006). The measured dynamic behaviour will be used as input for an identification algorithm based on a dynamic model of the human pelvis (Conza and Rixen, 2005).

View all citing articles on Scopus
View full text