Indicators of lumbar zygapophyseal joint pain: survey of an expert panel with the Delphi technique.

Low back pain (LBP) is a common problem that causes substantial economic, social, and psychological stresses for both the community and the individual. (1,2) Effective treatment is one method of reducing the cost of LBP by accelerating recovery and minimizing recurfence. (3) Despite extensive research efforts to devise effective treatment, the extent of the LBP problem remains unchanged. (2,4) Furthermore, most randomized controlled trials (RCTs) investigating the efficacy of treatment for LBP show small effect sizes at best, and conflicting results are common. (5-8) A false assumption that people with LBP are a homogeneous population has been proposed as contributing to these RCT findings. (9,10) The inclusion of heterogeneous samples can lead to an intervention being applied inappropriately to a proportion of subjects, resulting in either failure to respond or exacerbation of LBP. This situation may diminish the chance of an RCT achieving a clinically and statistically significant treatment effect. (11)

Improving classification systems is one method of identifying subgroups of LBP that may be more responsive to a specific treatment approach. (12-14) However, a validated and reliable classification system for LBP currently does not exist. (8,15,16) The notion that the lumbar zygapophyseal joints (LZJs) are a source of LBP has significant biological plausibility; therefore, LZJ pain may be considered a potential subgroup. These joints are well innervated by the medial branches of the lumbar dorsal rami, receiving a branch at the same level and a branch originating from the level above. (17-19) They were identified previously as a source of clinical pain by injection of isotonic saline or contrast medium into the joint or by electrical stimulation of the medial branches of the lumbar dorsal rami. (17,20-23) As synovial joints, the LZJs potentially are subject to a variety of pathologies that could result in LBP, (12) and morphological evidence of this potential has been found in postmortem studies. (24-27) Estimates of the prevalence of LZJ pain range as high as 75% among people reporting LBP. (28)

Despite the biological plausibility for LZJ pain as a subgroup of LBP, the identification of features indicative or diagnostic of this condition remains problematic. (28-31) A significant proportion of people who are asymptomatic for LZJ pain have positive radiological changes. (8,32,33) Diagnostic anesthetic injections into the purportedly symptomatic LZJ have been investigated (28,30,31,34-41) but have not been subject to the same degree of scrutiny as other diagnostic injections, (42-44) and controversy surrounds their methodology. (29,34-36) Diagnostic injections in general are prone to false-positive results because of the multifactorial neurophysiological, social, and psychological aspects of back pain. (29,45)

The use of multifactorial indicators has been accepted for diagnosing lumbar pathologies such as disk herniation with associated radiculopathy. (46,47) This method also has been used for diagnosing other pathological conditions, such as myocardial infarction and associated chest pain. (48)

Indicators or clinical features, such as provocative loading of the LZJ by extension, lateral flexion, or rotation, are commonly used in clinical practice to identify people with LZJ pain. However, there is no consensus in the literature as to what these features are. (28,30,31,37) A common method of validating indicators is by comparison with an established gold or reference standard. (49) Studies to date have attempted to validate clinical features against single-anesthetic blocks (30,37,38) and double-anesthetic blocks (41) of the LZJ. However, given the controversy surrounding the validity of such procedures, the suitability of their use as reference standards against which to compare the validity of indicators of LZJ pain is questionable. (29)

In the absence of suitable reference standards for validating indicators of LZJ pain, alternative methodology is required. The Delphi technique is a method of systematically surveying a group of experts in order to reach a consensus on specific questions or issues. (50) This method has been used successfully in other areas of musculoskeletal and medical research in which similar difficulties relating to diagnosis or classification exist. (51-54) More recently, the Delphi technique has been used to achieve a consensus on the diagnosis of musculoskeletal conditions, including clinical cervical spine instability (54) and carpal tunnel syndrome (55); therefore, its use for LZJ pain is appropriate.

Researchers have investigated potential indicators of LZJ pain, (30,37,38,41) but none has attempted to base these features on pathoanatomical mechanisms. There is empirical evidence in the literature regarding pathological changes and biomechanical factors that may produce LZJ pain. No study to date has attempted to associate indicators of LZJ pain with this empirical evidence. We believe that doing so will increase the face validity (56) of these indicators.

Given the importance of identifying subgroups of LBP, the biological plausibility for the LZJ as a source of LBP, issues with suitable reference standards for validating indicators, and the absence of readily available and affordable clinical tests indicative of LZJ pain, the opinion of an expert panel was sought. The aim of this study was to provide preliminary evidence for the validity of indicators of LZJ pain for the purposes of future clinical research by assembling an expert panel of Australian and New Zealand practitioners of medicine and physical therapy who have extensive experience in the management of LBP, using the Delphi technique (57,58) to develop a consensus regarding indicators of LZJ pain, and identifying pathoanatomical mechanisms underpinning each indicator in order to strengthen face validity.

Method

Ethics Approval

Following study approval, potential members of the expert panel from Australia and New Zealand were invited to participate in the study. Information regarding the nature of the study was provided, and written informed consent was gained prior to commencement.

Delphi Technique

The Delphi technique was used to obtain a consensus on the indicators of LZJ pain. The Delphi technique is a method for systematically collecting informed judgments from a group of experts on specific questions or issues. (50) The Delphi technique is used to allow free discussion of views without the influence of personal status, to enable the alteration of personal views without embarrassment, and to allow the combination of many opinions into a collective response. (58) The approach is useful in situations in which a consensus is lacking (55) and when uncertainty surrounds the area being investigated, (59) as in the controversy surrounding the diagnosis of LZJ pain. There are no guidelines on the optimal size of expert panels. Linstone and Turoff, (60) who were pioneers of the technique, asserted that the Delphi technique can be used for "anywhere from 10 to 50 people" but provided no further justification. It has been suggested that the most reliable panels should include 20 or fewer people in order to retain all of their members. (50,61,62) A panel consisting of 20 experts in the field of LBP was selected for the present study.

Subjects

The authors identified potential expert panel members on the basis of their substantial clinical, research, and educational expertise in LBP. An Australian and New Zealand multidisciplinary panel of clinicians who diagnose and treat LZJ pain was chosen to enable the development of indicators that would be relevant for international practitioners in physical therapy and medicine. A heterogeneous sample was chosen because it is widely believed that "if a disparate group ... achieves consensus, it is reasonable to conclude that [fills] consensus has ... merit." (63(p11)) We chose to recruit 5 physical therapist academic program leaders of postprofessionalentry specialist musculoskeletal physical therapy courses, 5 expert physical therapists, 5 musculoskeletal physicians, and 5 spinal orthopedic surgeons or neurosurgeons to make up the 20-member expert panel. Physical therapist academic program leaders of postprofessional-entry specialist musculoskeletal physical therapy courses were included because of their high levels of training and teaching in the assessment and treatment of LZJ problems. (64) Expert physical therapists were included to ensure that opinions outside of musculoskeletal physical therapy (eg, sports physical therapy) were included. Musculoskeletal physicians were included because they are specialist physicians who perform specific diagnostic and treatment procedures for LZJ problems. (65) Surgeons were selected because they commonly see people with the most recalcitrant types of LBP.

There is limited consensus in the literature as to the definition of an expert. (66) For the purpose of this study, inclusion criteria for the experts were as follows. Coordinators of postgraduate musculoskeletal physical therapy courses were located at Australian and New Zealand universities and had more than 10 years of academic and clinical experience. Expert physical therapists had more than 10 years of clinical experience in musculoskeletal or sports physical therapy and were considered experts by people in academic fields and peers. Musculoskeletal physicians had more than 10 years of clinical experience, academic experience, or both. Surgeons were neurosurgeons and orthopedic surgeons with a special interest in LBP and had more than 10 years of clinical and academic experience in managing LBP.

The experts were identified by the following processes. Every coordinator of postgraduate musculoskeletal physical therapy courses in Australia and New Zealand was identified by Internet-based searching. Expert physical therapists were identified by the authors and through recommendations from coordinators of physical therapy courses. Every musculoskeletal physician in Australia and New Zealand was identified through the College of Musculoskeletal Physicians. Surgeons were identified by the authors and through the Spine Society of Australia. All experts were invited to participate via telephone or e-mail. Experts who consented most promptly were selected until a panel of 20 people who satisfied all selection criteria was filled. If a particular geographic location was not represented in the panel, then experts from this region were preferentially selected to ensure an appropriate distribution across Australia and New Zealand.

Process

Expert panel. This Delphi survey consisted of 3 rounds of questionnaires that the expert panel answered consecutively (Fig. 1). In round 1 of this study, a questionnaire was mailed or e-mailed to all experts with the following instructions: "Please list the criteria [indicators] (and corresponding justifications) that you believe to be diagnostic of lumbar zygapophyscal joint pain." Justifications were requested in order to identify potential pathoana tomical mechanisms underpinning each indicator. The indicators listed in the returned questionnaires were collated and refined into common language by the authors (author panel) by using qualitative analytical techniques (sec "Author panel" below). This list of indicators then formed the questionnaire for round 2. Justifications for the selection of each indicator were recorded by the author panel but not returned to the expert panel.

[FIGURE 1 OMITTED]

The purposes of round 2 were to allow the experts to review all of the responses from round 1, rank the most relevant indicators, and provide additional justifications for their choices. The experts were given the following instructions: "Please rank (in order of importance) a maximum of fifteen (15) [indicators] that you believe to be indicative of lumbar zygapophyseal joint pain. Please provide justifications for your answers." The use of 15 indicators was an arbitrary decision made by the author panel in order to prevent experts from ranking all indicators and to encourage them to be selective in their choices. The experts effectively accepted or rejected indicators by using this ranking process. An indicator that was ranked was considered to be accepted; conversely, an indicator that was not ranked was rejected.

Upon receipt of the round 2 responses, the justifications from the experts were tabulated and analyzed by the author panel (see "Author panel" below); indicators that were accepted by fewer than 25% of the experts (5 panelists) were omitted from the list. (67) The remaining indicators were redistributed to the experts for round 3 along with additional information, including the average rank and range of ranks for each indicator as well as the percentage of experts who had selected each indicator in round 2. The experts were given the following instructions: "Please re-rank (in order of importance) a maximum of fifteen (15) [indicators] that you believe to be indicative of lumbar zygapophyseal joint pain." No further justifications for the selection of each indicator were requested. Subsequent rounds were to take place if necessary to achieve a consensus.

Author panel. The author panel comprised 3 physical therapists: 1 professor of physical therapy with 27 years of clinical experience and 30 years of academic experience; 1 therapist with a PhD and a postprofessional master's degree in musculoskeletal physical therapy, 20 years of clinical experience, and 15 years of academic experience; and 1 candidate for a PhD with 6 years of clinical experience and 3 years of academic experience.

A qualitative thematic analysis (68) was performed by the author panel following round 1 in order to eliminate overlap between the indicators listed by the expert panel. (58) The indicators from round 1 were tabulated by use of a Microsoft Excel * spreadsheet. The author panel met and grouped the listed indicators with similar meanings but variable wording and phrasing into mutually exclusive categories. Key themes in each category were identified and highlighted, and the author panel reached a consensus on clear and consistent wording for each indicator. In selecting appropriate wording, whenever possible, the author panel replicated the exact phrases used by the majority of the experts.

The author panel performed a similar process of refining the list of justifications for each indicator at the end of round 2. The experts were requested to articulate a clear mechanism based on empirical research for each indicator. The list of justifications was tabulated by use of an Excel spreadsheet and disseminated to the author panel. A qualitative thematic analysis similar to that used for the list of indicators was performed. If there were multiple justifications for the same criterion, then the justification supported by the largest number of experts was selected. References supporting the justifications for each indicator were frequently provided by the experts. A MEDLINE search with cross-referencing also was performed in an attempt to identify any additional relevant supportive literature for the justifications.

Results

Twenty experts consented to participate in the Delphi process (3 orthopedic surgeons, 2 neurosurgeons, 5 musculoskeletal physicians, 5 musculoskeletal physical therapists, and 5 coordinators of postgraduate musculoskeletal physical therapy programs), with 18 participating in all rounds. The sequence of the Delphi rounds and the process of moving toward a consensus are summarized in Figure 2. Two experts were unable to complete round 3 because of work commitments. However, their views expressed in rounds 1 and 2 were consistent with those of the other panel members; therefore, exclusion of their results in round 3 should not have altered the data.

Round 1 returned 135 indicators which, following qualitative thematic analysis, (63) were condensed into 31 indicators for round 2. Table 1 shows the 31 indicators, the number of experts accepting each indicator, and the average rank of each indicator. Round 1 also highlighted problems with the nomenclature used among the disciplines. Some experts (in the medical disciplines) believed that the term "diagnostic criteria" referred to a gold standard diagnosis. After acknowledging the different definitions of the word "diagnostic" understood by the experts, it was decided that it would be more appropriate to use the phrase "criteria indicative of lumbar zygapophyseal joint pain" to avoid such confusion. An indicator was defined as "a special symptom or the like which points out a suitable remedy or treatment or shows the presence of a disease." (69(p724)) The Delphi technique enables the provision of controlled feedback between rounds68; this feedback can widen knowledge, stimulate new ideas, and be educational for participants. (68,70)s We view the clarification of terminology described above as an example of the Delphi technique working effectively. The experts were informed of the terminology changes via a letter distributed with the round 2 questionnaire. Following round 2, the 31 indicators were condensed into 18 via the ranking process (Tab. 2); 13 indicators were omitted because they were selected by fewer than 25% of the experts.

[FIGURE 2 OMITTED]

The remaining 18 indicators were redistributed for round 3, in which the experts were asked to re-rank 15 indicators of LZJ pain. Round 3 maintained the 18 responses because no indicator was selected by fewer than 25% of the experts.

There are no firm rules for establishing when agreement or consensus is reached. (60) In a selection of Delphi studies reviewed by Powell, (68) consensus was defined in a variety of ways. Setting a percentage level for the inclusion of items appears to be a common practice, although variable levels, ranging from 51% to 100%, have been noted. (68) For the purpose of this study and to limit the number of successive rounds required for the survey, consensus was defined as greater than 56% agreement between experts on all indicators. Because the aim of this study was to provide preliminary validation, we thought it more prudent to set a lower level of consensus to minimize the risk of useful indicators being erroneously omitted. Twelve indicators with a consensus of 56% or more are shown in Table 3.

The experts were requested to list their justifications for each indicator of LZJ pain in both round 1 and round 2. The most frequently reported mechanisms for each indicator and supportive references (71-85) identified by the experts or authors are shown in Table 3.

Discussion

The present study identified 12 indicators of LZJ pain selected by a panel of experts in medicine and physical therapy. Recent reviews in the literature have identified the need for more targeted therapy and further research into subgroups of people who respond best to particular treatments, (10) and the present study is an important preliminary step in achieving this goal. These indicators may be useful for the clinical identification of LZJ pain and for the selection of homogeneous samples for future RCTs on the efficacy of target-specific treatment.

The validity of clinical features indicative of LZJ pain has been investigated in several studies with a variety of methodologies. (28,30,31,37,38,41,86-88) Concurrent validity is a methodological approach in which the ability of a test (for example, indicators determining subgroup membership) to predict the result of a criterion or reference standard is evaluated. (89) In studies evaluating the validity of subgroup membership, the reference standard test represents an absolute measure of truth by which subgroup membership can be determined. (89)

To date, several researchers have evaluated the concurrent validity of clinical indicators of LZJ pain against a reference standard of diagnostic injections and have obtained conflicting results. Revel and colleagues (37,38) identified indicators that could predict a positive response to single diagnostic blocks but stated that these criteria were not to be used for diagnostic purposes because of the high false-positive rates of single blocks compared with double blocks. (38,39) Double or "comparative" blocks refer to a series of 2 diagnostic injections in which anesthetics of various durations are administered. (65) Following the first injection, an individual who experiences a reduction in pain consistent with the duration of action of the anesthetic agent undergoes a second, confirmatory injection of an anesthetic with a different duration of action. (65) If the results of this confirmatory block are also positive, then the individual is deemed to have LZJ pain.

Several investigators have been unable to demonstrate the concurrent validity of the indicators identified by Revel and colleagues (37,38) against double blocks. (28,31) Schwarzer et al (30) investigated an array of clinical features in 176 subjects receiving double LZJ blocks. They concluded, "No combination of historical or examination features could be used to predict pain of [LZJ] origin" (30(p1136)); however, their chosen indicators were a limited clinical set. The results of these studies are in conflict with those of a recent study in which a clinical prediction rule consisting of 5 clinical features found to predict a positive response to double LZJ blocks was developed. (41) Perhaps the variable findings in the concurrent validity research are attributable partially to diagnostic injections being an insufficient reference standard. (29,34-36,90)

The validity of diagnostic LZJ blockage, both as an appropriate reference standard and as a diagnostic test in its own right, is based on 3 premises, (90) each of which is subject to confounding factors that increase the chances of false-positive findings. These premises surmise the following:

* The LZJ is the only pain generator, and there is no centralized pain modulation. However, LZJ pain is often a chronic condition, and extensive literature suggests that chronic LBP is associated with central sensitization and other neurophysiological changes. (91-94)

* Anesthetic blockage of the LZJ affects only the LZJ, and there is no overlap with other nerves or tissues. (90) However, studies have shown that one does not have to block the actual painful site of pathology directly to obtain subjective pain relief. (95,96) There also are potential issues with the delivery of anesthetic and uptake by surrounding structures. (29)

* The relief experienced is attributable solely to the anesthetized LZJ and is unaffected by central processing (placebo effect) or motivational reporting. However, the placebo effect is difficult to control, and motivational reporting is impossible to measure given the subjective nature of pain relief. Some investigators (35) have suggested that controlled triple LZJ blocks (consisting of a preliminary short-acting anesthetic and then a series of 2 masked injections--1 with anesthetic and the other with extraarticular saline) can control for placebo or motivational issues; however, these techniques have not been used in the research literature, presumably because of logistical and ethical issues.

Furthermore, the results obtained with diagnostic blocks are based on the degree of relief, a measure of questionable reliability because of the variability in people's abilities to differentiate between significant pain relief and insignificant pain relief. (29) On the basis of the above-described literature, we believe that a concurrent validity methodology may provide some information regarding preliminary validity for indicators of LZJ pain but that this information cannot be considered absolute. (29,90)

Evidence for indicators of LZJ pain, obtained with methodology other than concurrent validity methodology, has been presented in several other articles. Kent and Keating (88) surveyed 651 primary care clinicians regarding methods used to classify nonspecific LBP (including LZJ pain). Their results showed that most clinicians identified pathoanatomical subgroups (including lumbar facet syndrome) but that consensus was poor, with only 10% of clinicians agreeing on the indicators for each subgroup.

It is our view that the goal of developing a set of features for identifying LBP subgroups is best served by an expert panel whose higher levels of knowledge and experience would be more likely to result in consensus and valid indicators. (97) Despite the potential limitations and general lack of consensus found by Kent and Keating, (88) the features most commonly agreed upon in their survey were similar to those identified by our expert panel. George and Delitto (98) evaluated the validity of LBP subgroups by investigating the responses of these subgroups to specific treatment techniques. Although their approach reportedly was unrelated to pathoanatomical subgroups, the lumbar mobilization subgroup had features similar to those identified by our expert panel (eg, unilateral LBP and a "closing pattern" with movement testing). Using similar methods, Fritz et al (14) defined a manipulation subgroup that also had similar features (eg, pain upon palpation and LBP without radiation below the knee). Therefore, it appears that multiple investigators, using a variety of methods, have identified similar sets of signs and symptoms indicative of what we refer to as LZJ pain.

The concurrent validity studies described above usually did not take into consideration the pathoanatomical cause of pain in the selection of indicators to be validated. We believe that the face validity of indicators is increased if they relate directly to pathoanatomical mechanisms. For example, 94% of the expert panel agreed that localized unilateral LBP was an indicator of LZJ pain because the nociceptive supply to the LZJ is unilateral and pain is felt over the affected joint. (17-20,22,79) This mechanism is supported by anatomical studies of the LZJ nerve supply (17-19) and LZJ pain provocation studies. (20,22,79) All 12 indicators of LZJ pain identified by the expert panel were supported by pathoanatomical justifications, thereby increasing their face validity. (56)

Given the inherent complexity of LBP, the validity of indicators of LZJ pain cannot be established through one study or with one type of design. Instead, evidence supporting or refuting a variety of indicators should be gathered from different sources and from the use of different methods. (98) In the best-case scenario, these sources converge and identify similar indicators. (97,98) Our Delphi survey of experts, the previously reported studies on concurrent validity and predictive validity, and a large-scale survey of clinicians demonstrate the beginnings of such a convergence, providing support for the indicators of LZJ pain described in the present study.

The proposed indicators of LZJ pain require further validation before they can be endorsed for use in clinical practice. Studies worthy of consideration include predictive validity studies and RCTs. Radiofrequency denervation (neurotomy) is an invasive technique that is used to treat LZJ pain and that may be more target specific than controlled nerve blocks. (65) During this procedure, the nerve supply of the LZJ (the medial branches of the lumbar dorsal rami) is denervated, thereby preventing pain generation from the LZJ for approximately 10.5 months. (99) Future studies could investigate whether the indicators of LZJ pain identified in the present study predict a positive outcome from this long-acting and target-specific treatment. Anecdotal evidence suggests that manual therapy has a large effect when used in carefully selected cases with the indicators of LZJ pain. However, there is minimal empirical evidence to support this observation. The indicators described in the present study may be appropriate for use as selection criteria for future RCTs investigating the efficacy of manual therapy in the management of LZJ pain. The observation of a large effect size would provide further evidence for the validity of our indicators. Subsequently, once validated, these indicators may be used to screen people who may benefit from diagnostic injections, thereby reducing the costs of potentially ineffective procedures and improving the poor clinical indicators currently used in clinical practice.

Limitations

The present study has several limitations. The experts selected were from Australia and New Zealand; therefore, their views may not be consistent with international opinions. However, given that all have access to international publications and are regular attendees and presenters at international conferences, it is reasonable to assume that they are familiar with the current international literature.

The word "diagnostic" had to be altered to "indicative" after round 1 because of differences in the nomenclature used among the disciplines. Although the Delphi technique enables the use of controlled feedback between rounds, it is not known whether the experts might have answered the surveys differently had this information been provided from the beginning. However, given that subsequent rounds followed this nomenclature change and a consensus was reached, we believe that this issue did not have a negative impact on the outcome of our Delphi survey.

Opponents of the Delphi technique argue that Delphi findings should not be judged with the same validity as research derived by more established scientific methods (55); rather, they argue that the findings should be considered a process for making the best use of available information in the absence of a reference standard and in the presence of ambiguity. (53) With regard to this opinion, our Delphi survey provides preliminary validation of the described indicators and contributes to the convergence of opinions that this pathoanatomical subgroup of LBP exists and can be identified by clinicians. However, further validation is required before these indicators can be recommended for use in clinical practice.

Although 2 experts were unable to complete the final rounds of the survey because of time commitments, their initial views were consistent with those of the other experts; therefore, their exclusion or inclusion should not have altered the findings.

The expert panel identified diagnostic injections as being the most important indicators of LZJ pain. As discussed previously, we believe that the nature of LBP is multifactorial; therefore, the diagnosis of LZJ pain should not be based on one indicator alone. Diagnostic injections should not be used as a reference standard because no single indicator is sufficient for identifying LZJ pain. Instead, diagnostic injections in conjunction with clinical features based on pathoanatomical mechanisms are more likely to increase the probability that pain is arising from the LZJ.

Conclusion

The LZJ is thought be a source of LBP; however, indicators of LZJ pain have yet to be validated. Our Delphi survey of 20 experts in medicine and physical therapy identified 12 indicators of LZJ pain. All indicators were based on pathoanatomical mechanisms. The findings of this research converge with those of much of the research already published on identifying LZJ pain; however, further validation is required. Future studies should examine the ability of these indicators to predict responses to LZJ-specific treatments in prospective cohort studies and RCTs. Further validation is required before these indicators can be recommended for use in clinical practice.

Ms Wilde and Dr Ford provided concept/ idea/research design. All authors provided writing and project management. Ms Wilde provided data collection and analysis and subjects. Ms McMeeken provided institutional liaisons. Dr Ford and Ms McMeeken provided consultation (including review of manuscript before submission).

This study was approved by the University of Melbourne Human Research Ethics Committee.

This article was submitted October 27, 2006, and was accepted May 16, 2007.

DOI: 10.2522/ptj.20060329

References

(1) Andersson G. Epidemiological features of chronic low-back pain. Lancet. 1999; 354:581-585.

(2) Walker B. The prevalence of low back pain: a systematic review of the literature from 1966 to 1998. J Spinal Disord. 2000;13:205-217.

(3) Fletcher R, Fletcher S, Wagner E. Clinical Epidemiology: The Essentials. Baltimore, Md: Williams & Wilkins; 1996.

(4) Cassidy JD, Cote P, Carroll LJ, Kristman V. Incidence and course of low back pain episodes in the general population. Spine. 2005;30:2817-2823.

(5) Assendelft WJJ, Morton SC, Yu EI, et al. Spinal manipulative therapy for low back pain: a meta-analysis of effectiveness relative to other therapies. Ann Intern Med. 2003;138:871-881.

(6) Assendelft WJJ, Morton SC, Yu EI, et al. Spinal Manipulative Therapy for Low Back Pain. Maastricht, the Netherlands: Cochrane Back Group; 2005.

(7) de Nelemans PJ, de Bie RA, de Vet HCW, Sturmans F. Injection Therapy for Subacute and Chronic Benign Low Back Pain. Maastricht, the Netherlands: Cochrane Back Group; 2001.

(8) Koes B, van Tudler M, Thomas S. Diagnosis and treatment of low back pain. BMJ. 2006;332:1430-1434.

(9) Fritz J, George S. The use of a classification approach to identify subgroups of patients with acute low back pain: interrater reliability and short-term treatment outcomes. Spine. 2000;25:106-114.

(10) Harvey N, Cooper C. Physiotherapy for neck and back pain: we need to know who will benefit from which intervention. BMJ. 2005;330:53-54.

(11) Flynn T, Fritz J, Whitman J, et al. A clinical prediction rule for classifying patients with low back pain who demonstrate short-term improvement with spinal manipulation. Spine. 2002;27:2835-2843.

(12) Bogduk N. Clinical Anatomy of the Lumbar Spine and Sacrum. 3rd ed. New York, NY: Churchill Livingstone; 1997.

(13) Dankaerts W, O'Sullivan P, Burnett A, Straker L. Differences in sitting postures are associated with nonspecific chronic low back pain disorders when patients are subclassified. Spine. 2006;31:698-704.

(14) Fritz JM, Brennan GP, Clifford SN, et al. An examination of the reliability of a classification algorithm for subgrouping patients with low back pain. Spine. 2006; 31:77-82.

(15) Petersen T. Classification of non-specific low back pain: a review of the literature on classification systems relevant to physical therapy. Phys Ther. 1999;4:265-281.

(16) Riddle DL. Classification and low back pain: a review of the literature and critical analysis of selected systems. Phys Ther. 1998;78:708-737.

(17) Cavanaugh JM, Ozaktay AC, Yamashita HT, King AI. Lumbar facet pain: biomechanics, neuroanatomy and neurophysiology. J Biomech. 1996;29:1117-1129.

(18) Bogduk N. The innervation of the lumbar spine. Spine. 1983;8:286-293.

(19) Jackson H, Winkelmann R, Bickel W. Nerve endings in the human lumbar spinal column and related structures. J Bone Joint Surg Am. 1966;48:1272-1281.

(20) Marks R. Distribution of pain provoked from lumbar facet joints and related structures during diagnostic spinal infiltration. Pain. 1989;39:37-40.

(21) Schwarzer AC, Derby R, Aprill CN, et al. The value of the provocation response in lumbar zygapophyseal joint injections. Clin J Pain. 1994;10:309-313.

(22) McCall I, Park W, O'Brien J. Induced pain referral from posterior lumbar elements in normal subjects Spine. 1979;4:441-446.

(23) Mooney V, Robertson J. The facet syndrome. Clin Orthop. 1976;115:149-156.

(24) Farfan HF, Cossette JW, Robertson GH, et al. The effects of torsion on the lumbar intervertebral joints: the role of torsion in the production of disc degeneration. J Bone Joint Surg Am. 1970;52:468-497.

(25) Yang K, King A. Mechanism of facet load transmission as a hypothesis for low-back pain. Spine. 1984;21:538-543.

(26) Twomey L, Taylor J, Taylor M. Unsuspected damage to lumbar zygapophyseal (facet) joints after motor-vehicle accidents. Med J Aust. 1989;151:210-217.

(27) Taylor J, Twomey L, Corker M. Bone and soft tissue injuries in post-mortem lumbar spines. Paraplegia. 1990;28:119-129.

(28) Manchikanti L, Pampati V, Fellows B, Ghafoor Baha A. The inability of the clinical picture to characterize pain from facet joints. Pain Physician. 2000;3:158-166.

(29) Saal J. General principles of diagnostic testing as related to painful lumbar spine disorders: a critical appraisal of current diagnostic techniques. Spine. 2002;27: 2538-2545.

(30) Schwarzer AC, Aprill CN, Derby R, et al. Clinical features of patients with pain stemming from the lumbar zygapophysial joints: is the lumbar facet syndrome a clinical entity? Spine. 1994;19:1132-1137.

(31) Laslett M, Oberg B, Aprill CN, McDonald B. Zygapophysial joint blocks in chronic low back pain: a test of Revel's model as a screening test. BMC Musculoskelet Disord. 2004;5:43.

(32) Murtagh F. Computed tomography and fluoroscopy guided anaesthesia and steroid injection in facet syndrome. Spine. 1988;13:686-689.

(33) Schwarzer AC, Wang S, O'Driscoll D, et al. The ability of computed tomography to identify a painful zygapophysial joint in patients with chronic low back pain. Spine. 1995;20:907-912.

(34) Manchikanti L, Pampati V, Fellows B, Bakhit C. The diagnostic validity and therapeutic value of lumbar facet joint nerve blocks with or without adjuvant agents. Curr Rev Pain. 2000;4:337-344.

(35) Slipman CW, Bhat AL, Gilchrist RV, et al. A critical review of the evidence for the use of zygapophysial injections and radiofrequency denervation in the treatment of low back pain. Spine J. 2003;3:310-316.

(36) Hildebrandt J. Relevance of nerve blocks in treating and diagnosing low back pain: is the quality decisive? Schmerz. 2001; 15:474-483.

(37) Revel ME, Poiraudeau S, Auleley GR, et al. Capacity of the clinical picture to characterize low back pain relieved by facet joint anesthesia: proposed criteria to identify patients with painful facet joints. Spine. 1998;23:1972-1976.

(38) Revel ME, Listrat VM, Chevalier XJ, et al. Facet joint block for low back pain: identifying predictors of a good response. Arch Phys Med Rehabil. 1992;73:824-828.

(39) Schwarzer AC, Aprill CN, Derby R, et al. The false-positive rate of uncontrolled diagnostic blocks of the lumbar zygapophysial joints. Pain. 1994;58:195-200.

(40) Bogduk N. International Spinal Injection Society guidelines for the performance of spinalinjection procedures. Part 1. Zygapophysial joint blocks. Clin J Pain. 1997;13: 285-302.

(41) Laslett M, McDonald B, Aprill CN, et al. Clinical predictors of screening lumbar zygapophyseal joint blocks: development of clinical prediction rules. Spine J. 2006;6:370-379.

(42) Carragee EJ. Is lumbar discography a determinate of discogenic low back pain: provocative discography reconsidered. Curr Rev Pain. 2000;4:301-308.

(43) Carragee EJ, Chen Y, Tanner CM, et al. Provocative discography in patients after limited lumbar discectomy: a controlled, randomized study of pain response in symptomatic and asymptomatic subjects. Spine. 2000;25:3065-3071.

(44) Carragee EJ, Tanner CM, Yang B, et al. False-positive findings on lumbar discography. Spine. 1999;24:2542-2547.

(45) Carragee EJ, Lincoln T, Parmar VS, et al. A gold standard evaluation of the "discogenic pain" diagnosis as determined by provocative discography. Spine. 2006;31: 2115-2123.

(46) Vroomen P, de Krom M, Knotmerus J. Diagnostic value of history and physical examination in patients suspected of sciatica due to disc herniation: a systematic review. J Neurol. 1999;246:899-906.

(47) Vucetic N, Astrand P, Gunter P, Svensson O. Diagnosis and prognosis in lumbar disc herniation. Clin Orthop. 1999;361: 116-122.

(48) Solomon CG, Lee TH, Cook EF, et al. Comparison of clinical presentation of acute myocardial infarction in patients older than 65 years of age to younger patients: the Multicenter Chest Pain Study experience. Am J Cardiol. 1989;63:772-776.

(49) Sackett DL, Haynes RB. Evidence base of clinical diagnosis: the architecture of diagnostic research. BMJ. 2002;324:539-542.

(50) Reid N. Professional Competence and Quality Assurance in the Caring Professions. New York, NY: Chapman & Hall; 1988.

(51) Leone M, D'Amico D, Grazzi L, et al. Cervicogenic headache: a critical review of the current diagnostic criteria. Pain. 1998;78:1-5.

(52) Olesen J, Lipton R. Migraine classification and diagnosis: International Headache Society criteria. Neurology. 1994;44:S6-S10.

(53) The Non-Hodgkin's Lymphoma Classification Project. A clinical evaluation of the International Lymphoma Study Group classification of non-Hodgkin's lymphoma. Blood. 1997;89:3909-3918.

(54) Cook C, Brismee J-M, Fleming R, Sizer PS Jr. Identifiers suggestive of clinical cervical spine instability: a Delphi study of physical therapists. Phys Ther. 2005;85:895-906.

(55) Graham B, Regehr G, Wright JG. Delphi as a method to establish consensus for diagnostic criteria. J Clin Epidemiol. 2003; 12:1150-1156.

(56) Portney L, Watkins M. Foundations of Clinical Research: Applications to Practice. Norwalk, Conn: Appleton & Lange; 1993.

(57) Jones J, Hunter D. Consensus methods for medical and health services research. BMJ. 1995;311:376-380.

(58) Beech R. Go the extra mile: use the Delphi technique. J Nurs Manag. 1999;7: 281-288.

(59) Hardy D, O'Brien A, Gaskin C, et al. Practical application of the Delphi technique in a bicultural mental health nursing study in New Zealand. J Adv Nurs. 2004;46: 95-109.

(60) Linstone H, Turoff M. The Delphi Method: Techniques and Applications. Reading, Mass: Addison-Wesley Publishing; 1975.

(61) Jeffery D, Ley A, Bennun I, McLaren S. Delphi survey opinion on interventions, service principles and service organisation for severe mental illness and substance misuse problems. J Ment Health. 2000; 9:371-384.

(62) Mullen PM. Delphi: myths and reality. J Health Organ Manag. 2003;17:37-52.

(63) Mead D, Moseley L. The use of the Delphi as a research approach. Nurs Res. 2001;8: 4-23.

(64) Maitland GD, Hengeveld E, Banks K. Maitland's Vertebral Manipulation. 7th ed. Oxford, United Kindgom: Elsevier Butterworth-Heinemann; 2005.

(65) Hooten WM, Martin DP, Huntoon MA. Radiofrequency neurotomy for low back pain: evidence-based procedural guidelines. Pain Med. 2005;6:129-138.

(66) Baker J, Lovell K, Harris N. How expert are the experts? An exploration of the concept of 'expert' within Delphi panel techniques. Nurs Res. 2006;14:59-70.

(67) Roddy E, Zhang W, Doherty M, et al. Evidence-based recommendations for the role of exercise in the management of osteoarthritis of the hip or knee: the MOVE consensus. Rheumatology (Oxford). 2005;44:67-73.

(68) Powell C. The Delphi technique: myths and realities. J Adv Nurs. 2003;41:376-382.

(69) Yallop C, Bernard JRL, Blair D, et al, eds. The Macquarie Dictionary. 4th ed. North Ryde, New South Wales, Australia: The Macquarie Library Pty Ltd; 2005.

(70) Stokes F. Using the Delphi technique in planning of a research project on the occupational therapists' role in enabling people to make vocational choices following illness/injury. British Journal of Occupational Therapy. 1997;60:263-267.

(71) Marks R, Houston T, Thulbourne T. Facet joint injection and facet nerve block: a randomised comparison in 86 patients with chronic low back pain. Pain. 1992;49:325-328.

(72) Dreyfuss PH, Dreyer SJ, Herring SA. Lumbar zygapophysial (facet) joint injections. Spine. 1995;20:2040-2047.

(73) Dreyfuss P, Schwarzer AC, Lau P, Bogduk N. Specificity of lumbar medial branch and L5 dorsal ramus blocks: a computed tomography study. Spine. 1997;22:895-902.

(74) Kaplan M, Dreyfuss P, Halbrook B, Bogduk N. The ability of lumbar medial branch blocks to anesthetize the zygapophysial joint: a physiologic challenge. Spine. 1998;23:1847-1852.

(75) Fritz J, Piva S. Physical impairment index: reliability, validity and responsiveness in patients with acute low back pain. Spine. 2003;28:1189-1194.

(76) May S, Littlewood C, Bishop A. Reliability of procedures used in the physical examination of non-specific low back pain: a systematic review. Aust J Physiother. 2006;52:91-113.

(77) Adams MA, Hutton WC. The effect of posture on the role of the apopbyseal joints in resisting intervertehrai compressive force. J Bone Joint Surg Br. 1980;62:358-362.

(78) Adams MA, Hutton WC. The mechanical function of the lumbar apophyseal joints. Spine. 1983;8:327-329.

(79) Fukui S, Ohseto K, Shiotani M, et al. Distribution of referred pain from the lumbar zygapophyseal joints and dorsal rami. Clin J Pain. 1997;13:303-307.

(80) Eisenstein S, Parry C. The lumbar facet arthrosis syndrome: clinical presentation and articular surface changes. J Bone Joint Surg Br. 1987;69:3-7.

(81) Igarashi A, Kikuchi S, Konno S, Olmarker K. Inflammatory cytokines released from the facet joint tissue in degenerative lumbar spinal disorders. Spine. 2004;29: 2091-2095.

(82) Tournade A, Patay Z, Krupa P, et al. A comparative study of the anatomical, radiological and therapeutic features of the lumbar facet joints. Neuroradiology. 1992;24: 257-261.

(83) Strender L, Sjoblom A, Sundell K, et al. Interexaminer reliability in physical examination of patients with low back pain. Spine. 1997;22:814-820.

(84) Miller D. Comparison of electromyographic activity in the lumbar paraspinal muscles of subjects with and without chronic low back pain. Phys Ther. 1985;65:1347-1354.

(85) Hirayama J, Yamagata M, Ogata S, et al. Relationship between low-back pain, muscle spasm and pressure pain thresholds in patients with lumbar disc herniation. Eur Spine J. 2006;15:41-47.

(86) Fairbank JCT, Park WM, McCall IW, O'Brien JP. Apophyseal injection of local anaesthetic as a diagnostic aid in primary low-back pain syndrome. Spine. 1981;6:598-605.

(87) Helbig T, Lee C. The lumbar facet syndrome. Spine. 1988;14:61-64.

(88) Kent P, Keating JL. Classification in nonspecific low back pain: what methods do primary care clinicians currently use? Spine. 2005;30:1433-1440.

(89) Anastasi A, Urbina S. Psychological Testing. 7th ed. Englewood Cliffs, NJ: Prentice Hall; 1997.

(90) Carragee E, Hannibal M. Diagnostic evaluation of low back pain. Orthop Clin North Am. 2004;35:7-16.

(91) Butler DS, Moseley GL. Explain Pain. Adelaide, South Australia, Australia: Noigroup Publishing; 2003.

(92) Winkelstein BA. Mechanisms of central sensitization, neuroimmunology and injury biomechanics in persistent pain: implications for musculoskeletal disorders. J Electromyogr Kinesiol. 2003;14:87-93.

(93) CavanaughJM. Neural mechanisms of lumbar pain. Spine. 1995;20:1804-1809.

(94) Lidbeck J. Central hyperexcitability in chronic musculoskeletal pain: a conceptual breakthrough with multiple clinical implications. Pain Res Manag. 2002; 7:81-92.

(95) Siddall P, Cousins M. Spinal pain mechanisms. Spine. 1997;22:98-104.

(96) Kibler R, Nathan P. Relief of pain and paraesthesias by nerve block distal to a lesion. J Neurol Neurosurg Psychiatry. 1960;23:91-98.

(97) Ford J, Story I, O'Sullivan P, McMeeken J. A review of the methodology of classification development and validation for low back pain. Physical Therapy Reviews. 2007;12:33-42.

(98) George SZ, Delitto A. Clinical examination variables discriminate among treatmentbased classification groups: a study of construct validity in patients with acute low back pain. Phys Ther. 2005;85:306-313.

(99) Schofferman J, Kine G. Effectiveness of repeated radiofrequency neurotomy for lumbar facet pain. Spine. 2004;29: 2471-2473.

* Microsoft Corp, One Microsoft Way, Redmond, WA 9805243399.

VE Wilde, B Physio (Hons), is PhD candidate, School of Physiotherapy, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne, Melbourne, Victoria 3010, Australia. Address all correspondence to Ms Wilde at: v.wilde@pgrad. unimelb.edu.au.

JJ Ford, B App Sc (Physio), M Physio, PhD, Cred MDT, is Lecturer, School of Physiotherapy, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne.

JM McMeeken, Dip Physio, BSc (Hons), MSc, is Foundation Professor and Foundation Head, School of Physiotherapy, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne.

[Wilde VE, Ford JJ, McMeeken JM. Indicators of lumbar zygapophyseal joint pain: survey of an expert panel with the Delphi technique. Phys Ther. 2007;87:1348-1 361.]

Table 1.
Delphi Results: Round 2

Clinical Indicator of             % Expert     Average    Rank
Lumbar Zygapophyseal              Acceptance   Rank       Range
Joint (LZJ) Pain

 1. Positive response to          90            4         1-14
    intra-articular facet
    joint injection

 2. Pain relieved by              85            2         1-15
    fluoroscopically guided
    double-anesthetic blocks
    of the medial branch of
    the dorsal ramus
    supplying the LZJ

 3. Localized unilateral          80            4         1-10
    low back pain

 4. Lack of radicular             75            7         3-15
    features

 5. Replication or aggrava-       75            6         2-14
    tion of pain by unilateral
    pressure over the facet
    joint or transverse
    process

 6. Pain in extension, lateral    65            7         3-12
    flexion, or rotation to
    the ipsilateral side

 7. Unilateral muscle spasm       55           10         5-13
    over the affected LZJ

 8. Pain, if referred to the      40            5         3-10
    leg, is above the knee

 9. Pain in extension             40            7         3-15

10. Palpation: local unilateral   35           10         7-15
    passive movement shows
    reduced range of motion
    (ROM) or increased
    stiffness on the side of
    LZJ pain

11. The following information     30            7         2-13
    is unable to isolate an
    LZJ as the source of back-
    pain: history, including
    type of onset, and pattern
    of painful limitation of
    lumbar movements

12. Pain eased in flexion         30            9         5-13

13. Evidence of radiological      30           12         9-15
    degenerative changes at
    the LZJs

14. Pain: dull and deep ache      25            5         2-11

15. Pain relieved relatively      25            7         5-9
    quickly by joint
    mobilization

16. Incidence increases with      25            8         3-15
    age

17. Radiology is unreliable and   25            9         6-15
    cannot diagnose pain

18. Unpredictable sharp or        25           12        11-13
    catching pain

19. No clinical features or       15            2         1-3
    signs diagnostic of LZJ
    pain

20. Pain: worse in the morning,   15            6         1-12
    settles to an extent during
    the day, and may resume
    with stiffness by after-
    noon or evening

21. Apply sustained natural       15            8         5-13
    apophyseal glide and
    retest ROM: if better,
    then an LZJ is implicated;
    if not, then maybe a disk
    is responsible

22. Symptoms: worse in morning    15           11         5-15

23. Gradual and progressive       15           11        10-12
    degenerative history

24. Straight leg raising is not   15           11         7-14
    limited and does not
    reproduce symptoms early
    in the range

25. Absence of sacroiliac joint   15           12        12-15
    pain provocation signs or
    tests

26. Symptoms relieved by          15           12         8-14
    movement

27. Clinical (criteria of Revel   10            4        3 or 4
    and colleagues (37, 38):
    age of >65 y, pain not
    exacerbated by coughing,
    pain relieved by recum-
    bency, pain not worsened
    by forward flexion, pain
    not worsened when rising
    from forward flexion, pain
    not worsened by
    hyperextension, pain
    not worsened by
    extension-rotation

28. Habitual posture of the       10           11        10-12
    spine in end-range lumbar
    spine extension or
    hyperlordotic posture

29. "Inflammatory" rather than    10           14       13 or 14
    mechanical history

30. Mechanism of injury or         5           11          11
    history of trauma

31. Males > females                0            0          0

Table 2.
Delphi Results: Round 3

Clinical Indicator of Lumbar       % Expert     Average   Rank
Zygapophyseal Joint(LZJ) Pain      Acceptance   Rank      Range

 1. Positive response to intra-    100           3        1-11
    articular facet joint
    injection

 2. Localized unilateral low        94           4         1-6
    back pain

 3. Pain relieved by fluoros-       94           2        1-10
    copically guided double-
    anesthetic blocks of the
    medial branch of the dorsal
    ramus supplying the LZJ

 4. Replication or aggravation      89           6        3-11
    of pain by unilateral
    pressure over the LZJ or
    transverse process

 5. Lack of radicular features      89           7        2-15

 6. Pain eased in flexion           78          11        6-15

 7. Pain, if referred to the        72           7        2-11
    leg, is above the knee

 8. Palpation: local unilateral     61          11        6-15
    passive movement shows-
    reduced range of motion or
    increased stiffness on the
    side of LZJ pain

 9. Unilateral muscle spasm over    61          10        4-15
    the affected LZJ

10. Pain in extension               56           9        4-12

11. Pain in extension, lateral      56           6        3-10
    flexion, or rotation to
    the ipsilateral side

12. Radiology is unreliable         56          11        4-15
    and cannot diagnose LZJ
    pain

13. Pain: dull and deep ache        50           8        5-14

14. Incidence increases with        50           9        3-14
    age

15. The following information       44           7        1-15
    is unable to isolate an LZJ
    as the source of back pain:
    history, including type of
    onset,and pattern of
    painful limitation of
    lumbar movements

16. Pain relieved relatively        39          10        4-14
    quickly by joint
    mobilization

17. Evidence of radiological        39          10        3-14
    degenerative changes
    at the LZJs

18. Unpredictable sharp or          33          11        2-15
    catching pain

Table 3.
Evidence-Based Justifications for Indicators of Lumbar
Zygapophyseal joint (LZJ) Pain

Criterion                            Mechanism

1. Positive response to intra-      Anesthetic block of the painful LZJ
   articular facet joint injection    provides short-term pain relief.
                                      (40,71,72)

2. Localized unilateral back pain   The nociceptive supply to the LZJ
                                      is via the medial branches of the
                                      dorsal rami, and these nerves do
                                      not cross the midline to supply
                                      tissues on the contralateral side.
                                      (17-19) This unilateral back pain
                                      does not have to be felt in the
                                      low back but rather over the
                                      affected joint. 20-22

3. Pain relieved by                 Each LZJ sends its nociceptive
   fluoroscopically guided            input through the medial branches
   double-anesthetic blocks of the    of the 2 dorsal rami (the 1 above
   medial branch of the dorsal        and the 1 below), except for the
   ramus supplying the LZJ            L5-Sl joint. which receives only
                                      1 nerve (L5). (17-19) Thus,
                                      selective blocking of both nerves
                                      to a particular LZJ will isolate
                                      it as the source of pain. For
                                      accuracy of the technique, the
                                      anesthetic injection needs to be
                                      guided fluoroscopically. (73,74)

4. Replication or aggravation of    This criterion is a powerful
   pain by unilateral pressure        indicator for the origin of
   over the LZJ or transverse         symptoms under the area of
   process                            pressure and has been shown to be
                                      reliable. (75-76)

5. Lack of radicular features       Symptoms of nerve root irritation
                                      (dermatomal pain, paresthesia, or
                                      both, often worse distally) and
                                      signs of nerve root compression
                                      (dermatomal sensory loss, myotomal
                                      weakness, and loss of reflex)
                                      (46,47) are not signs of LZJ pain.

6. Pain eased in flexion            In standing, the LZJs are reported
                                      to carry 16% of the spinal
                                      compression load." In sitting
                                      (flexion), the LZJs are under
                                      relatively no load. (78)

7. Pain, if referred to the leg,    Typical of somatic referred pain
   is above the knee                  and further illustrated by LZJ
                                      pain provocation studies. (20,79)

8. Palpation: local unilateral      With increased age, disk narrowing
   passive movement shows reduced     results in as much as 70% of the
   range of motion or increased       intervertebral compression force
   stiffness on the side of LZJ       being transmitted across the LZJ.
   the affected LZJ                   Studies have shown increased
                                      degenerative changes in such
                                      specimens. (77) Degenerative
                                      changes. such as osteoarthritis
                                      and ligamentous damage, may
                                      contribute to LZJ stiffness.
                                      (80-82) This mechanism is similar
                                      to the compressive pain and
                                      stiffness associated with
                                      osteoarthritis of the knee
                                      joint. (67)

9. Unilateral muscle spasm over     Localized spasm can indicate
   the affected LZJ                   muscle guarding to prevent
                                      movement of the sympomatic
                                      LZJ. (83) Neurophysiological
                                      mechanism is reflex excitation
                                      of spasm. (84,85)

10. Pain in extension               In standing, the facets are
                                      reported to carry approximately
                                      16% of the spinal compressive
                                      load. (77) In extension, the
                                      compressive load on the facets
                                      increases; therefore, further
                                      pain is provoked. (17,78)

11. Pain in extension, lateral      Maximal compression of LZJ
    flexion, or rotation to the       surfaces occurs with these
    ipsilateral side                  movements. (25) This provocation
                                      may induce pain attributable to
                                      intra-articular or extra-articular
                                      pathology.

12. Radiology is unreliable and     Research has shown that radiology
    cannot diagnose LZJ pain          findings do not correlate
                                      with the presence or absence of
                                      symptoms. (8,32,33,80)