THE SACROILIAC JOINT (SIJ) can be a source of low

Transcription

1 10 ORIGINAL ARTICLE A Multitest Regimen of Pain Provocation Tests as an Aid to Reduce Unnecessary Minimally Invasive Sacroiliac Joint Procedures Peter van der Wurff, PT, PhD, Evert J. Buijs, MD, Gerbrand J. Groen, MD, PhD ABSTRACT. van der Wurff P, Buijs EJ, Groen GJ. A multitest regimen of pain provocation tests as an aid to reduce unnecessary minimally invasive sacroiliac joint procedures. Arch Phys Med Rehabil 2006;87:10-4. Objective: To compare the diagnostic accuracy of a multitest regimen of 5 sacroiliac joint (SIJ) pain provocation tests with fluoroscopically controlled double SIJ blocks using a short- and long-acting local anesthetic in order to reduce the exposure of patients to unnecessary invasive SIJ procedures. Design: Prospective, observational study. Setting: Hospital setting. Participants: Sixty patients with chronic low back pain. Interventions: Not applicable. Main Outcome Measures: Visual analog scale score and receiver operating characteristic (ROC) curve. Results: Twenty-seven patients responded positively to the blocks, of whom 23 were found positive after the multitest regimen and 4 were negative. For the nonresponders (n 33), these figures were 7 positive and 26 negative. The calculated sensitivity and specificity were.85 (95% confidence interval [CI],.72.99) and.79 (95% CI,.65.93), respectively. Positive and negative predictive values were.77 (95% CI,.62.92) and.87 (95% CI,.74.99), respectively. The positive likelihood ratio was 4.02 (95% CI, ); the negative likelihood ratio was.19 (95% CI,.07.47). The area under the ROC curve was.799. Conclusions: The test regimen with 3 or more positive tests is indicative of SIJ pain. It can be used in early clinical decision making to reduce the number of unnecessary minimally invasive diagnostic SIJ procedures. Key Words: Intra-articular injections; Pain assessment; Rehabilitation; Sacroiliac joint by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation From the Division of Perioperative Medicine and Emergency Medicine, Department of Anesthesiology and Pain Treatment, University Medical Centre Utrecht, Utrecht, The Netherlands (van der Wurff, Buijs, Groen); Department of Physical Therapy, Military Rehabilitation Centre Aardenburg, Doorn, The Netherlands (van der Wurff); and Department of Anesthesiology and Pain Treatment, Gelre Ziekenhuizen Apeldoorn, Apeldoorn, The Netherlands (Buijs). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Peter van der Wurff, PT, PhD, Div of Perioperative Medicine and Emergency Care, Dept of Anesthesiology and Pain Treatment, University Medical Center Utrecht, HP E03.511, PO Box 85500, 3508 GA Utrecht, The Netherlands, P.vanderWurff@ok.azu.nl /06/ $32.00/0 doi: /j.apmr THE SACROILIAC JOINT (SIJ) can be a source of low back pain (LBP) and referred pain to the lower limb, with a reported prevalence rate varying from 7% to 30%. 1,2 To confirm the clinical diagnosis of SIJ pain, intra-articular injection is reported to be the most accepted method of diagnosing or excluding the SIJ as a source of pain. 2-4 When a specific diagnosis cannot be made and the SIJ is suspected to be the source of pain, diagnostic intra-articular SIJ blocks with local anesthetics are considered to be the definitive test. In recent studies, 5-8 various invasive procedures were introduced as treatment options for SIJ pain when specific pathologic conditions such as spondylarthritis, infection, malignancy, or fracture have been excluded. A prerequisite for SIJ treatment procedures is the availability of appropriate diagnostic information; however, a review of the literature shows insufficient reliable data. 9,10 In the selection process, reliable clinical features that can minimize the potential hazard of unnecessarily including too many patients for invasive SIJ procedures are still lacking. The decision as to which patient would be eligible for a diagnostic SIJ block is mainly based on the outcome of SIJ pain provocation tests, pain projection, and the pain referral area. A major problem is to differentiate between SIJ pain and other sources of LBP. 2,3,11 There are numerous clinical tests to identify SIJ pain, but none is reliable or validated. 9,10 In a recent study, Kokmeyer et al 12 found that a multitest regimen of 5 pain provocation tests was reliable if 3 of 5 tests are positive. However, 5 studies used a diagnostic SIJ block to determine the value of SIJ pain provocation tests with different study designs. In the study of Maigne et al, 1 all patients received a short-acting anesthetic SIJ block (ie, a screening block) and only 19 of 54 patients underwent a second (confirmatory) block. A relation between any of the pain provocation tests and SIJ dysfunction could not be established. Dreyfuss et al, 3 using only a single anesthetic SIJ block, concluded that none of the 12 SIJ tests was reliable and acceptable for clinical practice. Broadhurst and Bond, 13 applying 3 SIJ provocation tests (Patrick sign, posterior shear, resisted abduction), claimed to achieve sensitivity values ranging from 77% to 87% and specificity values of 100% for each test. However, they constructed only positive predictive values. Their conclusion that the combination of these 3 tests has a high predictive value for SIJ dysfunction is not justified by the methodology they used, because 1 group of patients received an SIJ block with a local anesthetic and the other group a sham injection. Slipman et al, 14 also using multiple pain provocation tests and single SIJ blocks, found a positive predictive value of 60% for the physical examination; however, negative predictive values, specificity, and sensitivity and likelihood ratios could not be calculated. Finally, only 1 study has been performed to validate a cluster of SIJ pain provocation tests after excluding diskogenic complaints with use of a McKenzie evaluation. 15 However, in that study, not all patients received a double intra-articular SIJ block, thus increasing the possibility of bias. The primary aim

2 SACROILIAC JOINT TESTS TO REDUCE INVASIVE PROCEDURES, van der Wurff 11 of the present study was to determine the diagnostic accuracy of this multitest regimen compared with an SIJ double block applied to all patients, which may aid in reducing the number of unnecessary intra-articular SIJ blocks. METHODS The Medical Ethical Approval Committee of the Gelre Hospital Apeldoorn, the Netherlands, approved this study. The nature of this prospective, observational study and the potential hazards of the diagnostic procedures were explained and informed consent was obtained from all patients before their participation. Patients for this study were selected from among those with chronic LBP (CLBP) referred to the pain clinic of the Gelre Hospital Apeldoorn for further invasive diagnostic and therapeutic procedures including radiofrequency treatment. Patients were included in the study if their reported pain was below L5, over the posterior aspect of 1 SIJ, with or without leg pain. The pain measured on the visual analog scale (VAS) had to be at least 45mm and the pain had to be chronic ( 50d). Exclusion criteria were age less than 18 or over 80 years, ankylosing spondylitis, leg-length discrepancies of more than 2cm, Waddell 16 score more than 2, osteoporosis, tumors, infection and recent fractures in the lumbar spine region, osteoarthritis of the hip with clinical symptoms, radicular pain with neurologic signs, pregnancy, anticoagulants, and the presence of liver and/or kidney failure. Disk abnormalities as well as bulging disks were not excluded from the study except if nerve root compression with corresponding clinical signs were present. Each patient underwent 2 diagnostic SIJ injections on separate occasions. An independent investigator (manual therapist) examined the patient approximately 30 to 45 minutes before the injection procedure. The patient filled in 2 questionnaires: the Oswestry Disability Index (ODI) and the Symptoms Check List 90 (SCL-90) questionnaire. Each patient was asked to complete a pain diagram with a grid of 1mm 2 and was interviewed by the examiner to confirm the correlation of the pain diagram with the complaints. The average level of pain on the day of the injection was indicated on a 0- to 100-mm VAS (0mm, no pain; 100mm, worst imaginable pain ever). Before the first injection, the patient underwent 5 SIJ provocation tests conducted by the independent investigator. The anesthesiologist was unaware of the results of the provocation tests and the independent investigator was unaware of the medical history of the patient. All patients received diagnostic injections with short- and long-acting local anesthetics. The sequence of the injections was at random and known by the anesthesiologist only. If the patients with a negative result from the first injection had been excluded for the second injection, then bias would have occurred because the independent investigator indirectly would have had knowledge of the result of the first injection. Approximately 30 minutes after the injection, the patient was asked to complete a postinjection VAS and pain drawing. Injection Procedure For the intra-articular SIJ block, we adopted the standards for facet joint injections as accepted by the International Spinal Injection Society, which include that the patient has SIJ pain if radiographically guided injections with a short- and a longacting anesthetic medium relieve the patient s characteristic pain. 2,17 Bogduk and McGuirk 18 argued that the injection procedures for the SIJ should be similar to that of the facet joints. We followed the injection technique for the SIJ as described by Schwarzer et al. 2 The patient lies in prone position with a pillow placed under the abdomen at the iliac crests. With a C-arm, a the SIJ is visualized on the side where the pain is located. The place of the injection is marked and the skin is prepared in a regular manner. An RCN 10 needle b is inserted and positioned in the SIJ cavity. Next, 1mL of contrast medium (iohexol [Omnipaque]) is injected and the position of the needle and contrast spread is documented by fluoroscopy (lateral view and a three-quarter view). Additionally, 2mL of 2% lidocaine or.25% bupivacaine is injected. Afterward, the patient is transferred to the outpatient recovery department and allowed to move freely. Provocation Tests Five SIJ pain provocation tests (distraction test, compression test, thigh trust test, Patrick sign, Gaenslen test) were found to be reliable in a previous study. For details about the execution of these tests, see Kokmeyer et al. 12 Scoring Criteria For all provocation tests, scoring criteria were dichotomous: positive or negative. The test result was positive if the patient experienced their familiar pain. The test was negative if the patient experienced pain before full range was obtained, if the evoked pain was not characteristic, or if no pain was evoked. For the SIJ blocks, a reduction in the patient s characteristic pain of 50% or more on the VAS remaining for at least 1 hour (short-acting local anesthetic [ie, lidocaine]) or 4 hours (longacting local anesthetic [ie, bupivacaine]) was required. When a patient showed this VAS reduction after both intra-articular SIJ blocks, this was considered a positive response (responders). Any other outcome was considered a negative response (nonresponders). Statistical Methods First, a receiver operating characteristic (ROC) curve was produced to determine if the cutoff point of 3 of 5 provides acceptable sensitivity, specificity, negative and positive predictive values, and likelihood ratios. The ROC area can range from 0.5 (useless model) to 1.0 (perfect discrimination). A value greater than 0.7 can be interpreted as reasonable or fair, and greater than 0.8 as good. A2 2 contingency table was constructed to determine the association between results of the multitest regimen and the response to the blocks. The sensitivity, specificity, predictive values, likelihood ratios, and 95% confidence intervals (CIs) were calculated. RESULTS Between January 2001 and April 2002, a total number of 140 patients with CLBP visited the pain clinic of the Gelre Hospital Apeldoorn, The Netherlands. Sixty patients entered the study. The characteristics of the study population, sex, age, duration of complaints, ODI score, SCL-90 score, and VAS score are shown in table 1. There were no statistically significant differences between 27 positive (45%) and 33 negative responders (55%) to the diagnostic SIJ blocks. Leakage of the anesthetic fluid, mostly due to a ventral position of the needle, occurred in 5 patients, resulting in a temporary sciatic palsy. The results of these injections were rejected and the procedure was redone at a later time. In the renewed attempt, a less anterior position of the needle revealed excellent spread of the contrast fluid with no ventral leakage or sciatic palsy. Three of these patients could be categorized as true positive, 1 as false positive, and 1 as true negative.

3 12 SACROILIAC JOINT TESTS TO REDUCE INVASIVE PROCEDURES, van der Wurff Table 1: Summary of Patients Characteristics (N 60) Characteristic Total Group Positive SIJ Blocks Negative SIJ Blocks Age (y) No. of patients (men/ women) 60 (13/47) 27 (4/23) 33 (9/24) Duration of complaints (mo) Right side/left side (n) 29/31 16/11 13/20 ODI score at intake VAS score* at intake SCL-90 Global Severity Index NOTE. Values are mean standard deviation or n. *VAS 0 100mm scale. Figure 1 shows the relative reduction in VAS for responders and nonresponders. After the lidocaine injection, the mean VAS reduction was 67.6% (range, 50% 100%); after the bupivacaine injection, the mean VAS reduction was 71.4% (range, 52% 100%). Thirty-three (55%) patients did not respond to the blocks. The mean VAS reduction after the lidocaine injection was 6.0% (range, 38% 56%); after the bupivacaine injection, the mean VAS reduction was 0.4% (range, 36% 59%). None of the provocation tests or a specific combination of them was favorable. Of the patients with 3 or more positive tests, 29 scored positive after the first block, 23 also scored positive (true positives) after the second block, 6 patients responded positive to only 1 block, and 1 patient did not respond to either block. Seven patients had to be categorized as false positive. Four of the 23 true positive patients obtained complete relief of their pain after the blocks. Figure 2 presents the ROC curve expressing the sensitivity against 1 the specificity of positive pain provocation tests in patients with a positive SIJ block. The suggested cutoff point, Fig 1. The relative reductions of pre-post VAS scores (%, VAS) of patients after intra-articular SIJ injection of short- and long-acting anesthetics. Data are rearranged to separate nonresponders (left side) and responders (right side). Dashed line represents 50% level of the VAS reduction. Fig 2. The ROC curve of the positive pain provocation tests with positive SIJ blocks (>50% VAS reduction). The number of positive provocation tests (1 5) is shown at the relevant cutoff points of the curve. NOTE. AUC.799; standard error,.059; 95% CI, of 5 tests, 12 results in a maximal area under the curve (AUC) of.799 (fig 2, point 3). Table 2 presents the relation between the number of positive pain provocation tests and the positive likelihood ratio. It shows that at the cutoff point of 3 or more positive provocation tests the balance between sensitivity and specificity is optimal. This is related with the highest positive likelihood ratio (4.02). Table 3 shows the final results of the multitest regimen compared with the reference test in a 2 2 table at the level of 3 or more positive tests, revealing 23 true positives, 7 false positives, 4 false negatives, and 26 true negatives. DISCUSSION An evident correlation was found between the outcome of 3 or more positive SIJ pain provocation tests and the outcome of double intra-articular SIJ blocks. For 3 or more positive tests, the ROC curve shows optimal results for sensitivity (85%), specificity (79%), negative (87%) and positive (77%) predictive values, and likelihood ratios. The sensitivity and specificity values are in the range of being acceptable. Sensitivity and specificity rates cannot be used to quantify the shift in probability of SIJ pain. Likelihood ratios overcome the aforementioned problems by combining the information obtained in sensitivity and specificity values into a ratio. This ratio indicates by how much a given test result will increase or decrease the pretest probability. 19 In our study, we achieved a positive

4 SACROILIAC JOINT TESTS TO REDUCE INVASIVE PROCEDURES, van der Wurff 13 Table 2: Cutoff Points of the ROC Criterion Sensitivity 95% CI Specificity 95% CI LR LR PPV NPV Abbreviations: LR, negative likelihood ratio; LR, positive likelihood ratio; NPV, negative predictive value; PPV, positive predictive value. likelihood ratio of 4.02 and a negative likelihood ratio of.18, indicating a small but likely important shift in probability. 20 Furthermore, according to Sackett et al, 21 the negative predictive value (ie, 87%) ruling out SIJ pain is more useful than the positive predictive value (ie, 77%) because it avoids requiring the patients to undergo unnecessary treatments. One might argue, from table 2, that compared with using only 1 positive test, using 3 positive tests as the threshold for performing an SIJ block would mean that 15 of 100 patients would not learn their true diagnosis or receive the treatment they need, while only 37% would be spared an unnecessary block. However, ROC cutoff points should have the highest values of sensitivity and specificity (ie, at 3 positive tests). The proportion of false negative results, that is, 15% (95% CI, 1% 28%), indicates that one would miss 4 in 30 patients (95% CI, 1 8 patients) in which the SIJ blocks were positive, despite negative ( 3) pain provocation tests. The false positive proportion of 21% (95% CI, 7% 35%) implies that 7 of 30 patients (95% CI, 2 11 patients) would receive unnecessary SIJ blocks. Until now, clinical decision making was based on the assumption that SIJ pain is probable if pain was found below L5 overlying the SIJ and if SIJ blocks were positive. Each putative SIJ pain patient was (supposed to be) subjected to an SIJ block, disregarding the role of pain provocation tests. Implementing data of the present study into the clinical decision-making process, the following approach in patients with pain below L5 overlying the SIJ could be suggested. One, SIJ pain is probable if 3 or more positive provocation tests are found. This could be followed by SIJ blocks with therapy mainly aimed at the SIJ. If results of SIJ blocks are negative, other sources should be Table 3: Threshold of 3 Positive Tests (N 60) Positive Result Injection Reference Test Negative Result Injection Total Positive ( 3 SIJ tests) Negative ( 2 SIJ tests) Total Test Feature Value 95% CI Sensitivity Specificity Positive predictive value Negative predictive value Likelihood ratio positive Likelihood ratio negative False positive proportion False negative proportion Overall accuracy considered. Two, SIJ pain is less probable if fewer than 3 tests are positive. SIJ blocks are not primarily advocated and the clinician should focus on the identification of other sources and related therapy. SIJ blocks may re-emerge in a later stage of the diagnostic process. Thus, at least in the early stage of clinical decision making, the number of SIJ blocks is reduced substantially. Our results corroborate the findings of a recent study by Laslett et al, 15 although in that study double blocks had been applied to only a subset of the patients. The calculated sensitivity (91%), specificity (87%), and likelihood ratio (6.97) do not differ much, again underlining that SIJ pain provocation tests can be used by clinicians within a context of a clinical reasoning process. 15 In contrast to the present study, double blocks with short- and long-acting local anesthetics were used at random in all patients. Incorporation bias occurs when the reference test (ie, the SIJ blocks) includes the outcome measurement of the diagnostic test regimen (ie, the set of 5 provocation tests) under investigation. To avoid this pitfall, results of reference and diagnostic tests were assessed separately, and we used only the VAS as an outcome measurement for the judgment of the SIJ blocks. Following Fortin s stringent criteria in our selected group of 60 patients, a prevalence rate of only 7% to 10% would have been found and of 3% to 4% in the whole group of 140 consecutive patients with CLBP. Based on data from our study, prevalence rates would vary between 16% (23/140 patients) and 38% (23/60 patients), which are slightly higher than mentioned in literature (15% 20%). 18 A limitation of the present study could be our acceptance of a response of 50% or greater pain reduction, whereas in other studies, there are more stringent criteria of pain reduction. 1-3,13,15 Because our patients were referred for further invasive diagnostic and therapeutic procedures, including radiofrequency treatment, we adopted the criteria of Cohen and Abdi. 6 This threshold was also applied in recent lumbar facet joint trials The comparison with lumbar facet joint blocks is justified because the SIJ, with regard to the innervation pattern, can be considered as merged sacral facet joints. Within the responder group, responses to both local anesthetics showed a high correlation, which was not found for the nonresponder group. In our opinion, this strong correlation adds to the reliability of the result. Others have emphasized that intra-articular injection of local anesthetics blocks all pain from the SIJ. 25 If that were true, a 100% VAS reduction should be achieved, which rarely occurs. Probably more factors influence the outcome, and not all pain-generating structures related to the SIJ are blocked sufficiently. Accepting only 80% or more VAS reduction in our study for the diagnostic blocks disregards the results of the SIJ provocation tests. Patients did respond to this multitest regimen in a highly reliable manner and expressed each time that their familiar pain was provoked.

5 14 SACROILIAC JOINT TESTS TO REDUCE INVASIVE PROCEDURES, van der Wurff The responder and nonresponder patients clearly show differences in their responses to the provocation tests. CONCLUSIONS This study, designed to determine the relevance of a multitest regimen of SIJ provocation tests, shows that application of this regimen may be useful in reducing unnecessary intraarticular SIJ blocks in the early stage of clinical decision making. When 3 or more provocation tests are positive, the probability is between 65% and 93% that the pain is related to the SIJ, in which case confirming SIJ blocks are required. With fewer than 3 provocation tests positive, the probability is between 72% and 99% that the SIJ is unlikely to be the source of pain. In this case, other sources should be looked for primarily. SIJ blocks are not initially advocated. References 1. Maigne JY, Aivaliklis A, Pfefer F. Results of sacroiliac joint double block and value of sacroiliac pain provocation tests in 54 patients with low back pain. Spine 1996;21: Schwarzer AC, Aprill CN, Bogduk N. The sacroiliac joint in chronic low back pain. Spine 1995;20: Dreyfuss P, Michaelsen M, Pauza K, McLarty J, Bogduk N. The value of medical history and physical examination in diagnosing sacroiliac joint pain. Spine 1996;21: Elgafy H, Semaan HB, Ebraheim NA, Coombs RJ. Computed tomography findings in patients with sacroiliac pain. Clin Orthop 2001;Jan(382): Buijs E, Kamphuis ET, Groen GJ. Radiofrequency treatment of sacroiliac joint-related pain aimed at the first three sacral dorsal rami: a minimal approach. Pain Clin 2004;16: Cohen SP, Abdi S. Lateral branch blocks as a treatment for sacroiliac joint pain: a pilot study. Reg Anesth Pain Med 2003; 28: Ferrante FM, King LF, Roche EA, et al. Radiofrequency sacroiliac joint denervation for sacroiliac syndrome. Reg Anesth Pain Med 2001;26: Yin W, Willard F, Carreiro J, Dreyfuss P. Sensory stimulationguided sacroiliac joint radiofrequency neurotomy: technique based on neuroanatomy of the dorsal sacral plexus. Spine 2003; 28: van der Wurff P, Hagmeijer RH, Meyne W. Clinical tests of the sacroiliac joint. A systemic methodological review. Part 1: reliability. Man Ther 2000;5: van der Wurff P, Meyne W, Hagmeijer RH. Clinical tests of the sacroiliac joint. A systemic methodological review. Part 2: validity. Man Ther 2000;5: Saal JS. General principles of diagnostic testing as related to painful lumbar spine disorders: a critical appraisal of current diagnostic techniques. Spine 2002;27: ; discussion Kokmeyer DJ, van der Wurff P, Aufdemkampe G, Fickenscher TC. The reliability of multitest regimens with sacroiliac pain provocation tests. J Manipulative Physiol Ther 2002;25: Broadhurst NA, Bond MJ. Pain provocation tests for the assessment of sacroiliac joint dysfunction [see comments]. J Spinal Disord 1998;11: Slipman CW, Sterenfeld EB, Chou LH, Herzog R, Vresilovic E. The predictive value of provocative sacroiliac joint stress maneuvers in the diagnosis of sacroiliac joint syndrome. Arch Phys Med Rehabil 1998;79: Laslett M, Young SB, Aprill CN, McDonald B. Diagnosing painful sacroiliac joints: a validity study of a McKenzie evaluation and sacroiliac provocation tests. Aust J Physiother 2003;49: Waddell G, Main CJ, Morris EW, Di Paola M, Gray IC. Chronic low-back pain, psychologic distress, and illness behavior. Spine 1984;9: Bogduk N. International Spinal Injection Society guidelines for the performance of spinal injection procedures. Part 1: zygapophysial joint blocks. Clin J Pain 1997;13: Bogduk N, McGuirk B. Medical management of acute and chronic low back pain. Amsterdam: Elsevier Sci; Moons KG, Stijnen T, Michel BC, et al. Application of treatment thresholds to diagnostic-test evaluation: an alternative to the comparison of areas under receiver operating characteristic curves. Med Decis Making 1997;17: Jaeschke R, Guyatt G, Sackett DL. Users guides to the medical literature. III. How to use an article about a diagnostic test. A. Are the results of the study valid? Evidence-Based Medicine Working Group. JAMA 1994;271: Sackett DL, Haynes RB, Guyatt GH, Tugwell P. Clinical epidemiology: a basic science for clinical medicine. 2nd ed. Boston: Little Brown; Geurts JW, van Wijk RM, Wynne HJ, et al. Radiofrequency lesioning of dorsal root ganglia for chronic lumbosacral radicular pain: a randomised, double-blind, controlled trial. Lancet 2003; 361: Leclaire R, Fortin L, Lambert R, Bergeron YM, Rossignol M. Radiofrequency facet joint denervation in the treatment of low back pain: a placebo-controlled clinical trial to assess efficacy. Spine 2001;26:1411-6; discussion van Kleef M, Barendse GA, Kessels A, Voets HM, Weber WE, de Lange S. Randomized trial of radiofrequency lumbar facet denervation for chronic low back pain. Spine 1999;24: Fortin JD, Pier J, Falco FJ. Sacroiliac joint injection: pain referral mapping and arthrographic findings. In: Vleeming A, Mooney V, Snijders C, Dorman T, Stoeckart R, editors. Movement, stability and low back pain. 2nd ed. London: Churchill Livingstone; p Suppliers a. Ziehm Exposcop 7000; Ziehm Imaging GmbH, Donaustr 31, Nürnberg, Germany. b. Top Inc, Senju Nakai-cho-Adachi-ku, 120 Tokyo, Japan.