Answers are grounded in this paper and cite their sources.
A 46-year-old healthy man presents with an 8-week history of axial low back pain without precipitating trauma. He was initially seen in the physiatry spine clinic, and at that time he described right lower back pain without radiation into the legs. Clinical examination was positive for sharp, localized right lower back pain that was exacerbated with lumbar extension and right quadrant loading and relieved with lumbar flexion. Pertinent negative findings included a normal neurologic examination and no neural tension signs, including the straight leg raise and slump test. Flexion/extension radiographs revealed mild disk height loss at L4-L5, no pars interarticularis fractures, and no evidence of dynamic instability. The patient was prescribed nonsteroidal anti-inflammatory drugs (meloxicam, 15 mg) and muscle relaxants (cyclobenzaprine, 5 mg) at night to help with sleep. He has since completed 8 weeks of physical therapy with a combination of Williams flexion exercises, deep tissue massage, and pelvic traction; however, his pain persists. He presents to discuss further diagnostic imaging and possible interventional treatments. What imaging examination is most appropriate to identify the likely pain generator? Dr Jason Talbott will take the position that functional nuclear medicine imaging is the appropriate examination. Dr J. Levi Chazen will take the position that magnetic resonance imaging is most appropriate at this stage. This case illustrates the all too frequent clinical scenario of a patient with axial low back pain that is refractory to conservative interventions. Although the clinical symptoms and physical examination findings are suggestive of symptomatic osteoarthritis (OA) of the lumbar zygapophyseal (facet or z joint), localizing the precise level remains a major challenge. Confirming z-joint OA as the primary pain generator is elusive even with a thorough history, physical examination, and anatomic imaging. It is estimated that 15%-40% of all low back pain is related to z-joint disease 1. Although characteristic clinical examination findings including morning stiffness and pain exacerbated by extension, bending, and rotation but relieved with flexion have been described, these indicators of z-joint arthropathy are nonspecific 2, 3. In fact, with use of diagnostic blocks of the z joint, Schwarzer and colleagues 2 found that no combination of history or physical examination features can accurately predict pain arising from the z joint, calling into question the notion of a clinical facet joint syndrome. Although diagnostic blocks remain the gold standard for identifying symptomatic z-joint–mediated pain, a diagnostic block is an invasive and time-consuming procedure that is not feasible as a screening tool for all patients presenting with nonspecific low back pain. Clearly, the need exists for accurate, noninvasive biomarkers of symptomatic lumbar z-joint arthropathy. I believe that bone scintigraphy with single proton emission computed tomography (SPECT) and positron emission tomography (PET) are the ideal imaging choices for confirming z-joint–related low back pain. With the widespread utilization of computed tomography (CT) scanning and magnetic resonance imaging (MRI) in the 1980s, extensive literature now exists with respect to cross-sectional imaging characterization of z-joint degeneration. Anatomic and structural features of z-joint arthropathy on CT and MRI have naturally been pursued as potential biomarkers of z-joint–mediated pain, and many MRI and CT-based grading schemes for z-joint arthropathy have been devised 4-6. However, the diagnostic specificity of these morphologic findings is lacking. Both CT and conventional MRI sequences routinely used in clinical practice are limited to demonstrating structural alterations in the z joint, such as cartilage loss with joint space narrowing, bony hypertrophy, synovial and capsular thickening, and intra-articular fluid accumulation 4. It is well known that these same structural alterations of the z joint may occur in asymptomatic persons. Thus structural degenerative changes are common in the asymptomatic population and may be part of the normal aging process 7. In a large prospective longitudinal trial including 148 asymptomatic study participants, Jarvik and colleagues 8 found that new low back pain symptoms developed in 67% of participants during a 3-year interval, but MRI features of z-joint arthropathy had no correlation with clinical symptoms. Many studies have corroborated these findings, showing a lack of specificity for anatomic changes of the z joint with respect to symptomatic pain and disability 3, 9, 10. Despite the lack of data supporting MRI for accurate identification of symptomatic lumbar z-joint disease, MRI evaluation for suspected z-joint arthropathy contributes significantly to the rapidly increasing cost of low back pain care 11. As an alternative to structural imaging techniques such as conventional MRI, physiologic imaging modalities, including bone scintigraphy with SPECT and PET, have shown great promise as biomarkers for z-joint–related pain 12. Although these techniques initially were established to evaluate bone tumors and malignant disease, their role in benign degenerative disease is being increasingly recognized 12. With bone scintigraphy, a radioactive tracer such as technetium-99m methylene diphosphonate (Tc-99m MDP) is injected intravenously into a patient, who then undergoes imaging with a gamma camera that detects and localizes the accumulated radiotracer in the body. Acquisitions may be performed as planar 2-dimensional projections (similar to radiographs) or as 3-dimensional images (similar to CT) with SPECT. Tc-99m MDP is incorporated into hydroxyapatite of bone matrix and reflects the degree of osteogenic activity in bone. When combined with CT, precise localization of radiotracer uptake can be achieved to detect a specific bony structure with high physiologic turnover and hyperemia related to inflammation 13. Thus SPECT provides metabolic physiology data that are not achieved with conventional MRI. The radiation dose is relatively low, with a typical bone scan equaling approximately 2-5 mSv (similar to a noncontrast head CT scan). Several studies have validated the diagnostic capacity for bone scintigraphy with symptomatic z-joint OA 13-15. In a prospective, randomized controlled trial, Pneumaticos et al 13 reported that 87% of patients obtained symptomatic relief when MDP-SPECT was used to target lumbar z-joint injections. Conversely, among patients with clinical symptoms suggesting z-joint arthropathy but no suspicious z-joint uptake on SPECT, only 12% had relief when injections were targeted on the basis of clinical symptoms alone. SPECT also was shown to increase the specificity of injections, resulting in a decrease in the number of targeted injections by 50%. Cost analysis was favorable when compared with treatment without SPECT 13. The high accuracy of SPECT for diagnosing symptomatic z-joint OA has been demonstrated by other investigators 14, 15. These data are supported by a larger literature showing the value of SPECT for identifying active synovial inflammation in a variety of synovial joints 16-18. In summary, I support the recommendation for Tc-99m MDP SPECT in our patient with clinically suspected lumbar z-joint arthropathy. Although a head-to-head study comparing conventional MRI and SPECT for targeting lumbar z-joint–mediated pain has not been performed, the present literature suggests that the metabolic physiology data afforded with SPECT more accurately diagnoses symptomatic joints and predicts a successful treatment response. Available data suggest that conventional MRI will nicely characterize static, structural abnormalities that may be part of the normal aging process but will not accurately localize the site of "facetogenic" pain with high confidence. Increased acceptance of SPECT and other emerging physiologic imaging techniques will improve the diagnosis and treatment of z-joint–related low back pain. In doing so, we may improve care by increasing the efficacy of targeted treatments and lower the impact of imaging costs relating to our growing low back pain epidemic. Low back pain is a common health problem and the leading cause of disability for persons younger than 45 years, with a lifetime prevalence exceeding 80% 1, 2. The patient in the aforementioned clinical scenario presents with symptoms and physical examination findings strongly suggestive for z-joint–mediated pain. However, history and physical examination alone are unreliable in isolating the z joint as the underlying pain generator 3. Although this patient's radiographs are essentially unrevealing, aside from mild L4-L5 disk height loss, a myriad of underlying causes may be occult on lumbar spine plain films. Collaborative clinical practice guidelines from the American College of Physicians (ACP) and American Pain Society (APS), as well as Imaging Appropriateness Criteria from the American College of Radiology (ACR), suggest conservative treatment in patients with acute low back pain of less than 4-6 weeks' duration without "red flags" 4, 5. Red flags include findings suspicious for underlying malignancy, infection, or significant trauma. Given the absence of these concerning findings in this scenario, MRI is recommended by ACP/APS guidelines after 4 weeks of unremitting symptoms and by ACR guidelines after 6 weeks. In this clinical context, the ACR Appropriateness Criteria rates a noncontrast lumbar spine MRI scan as the most appropriate imaging study, with an 8 out of 9 rating on the appropriateness scale ("usually appropriate"). CT is rated 5 out of 9 ("may be appropriate"), and lumbar radiographs and nuclear bone scan with SPECT are both rated 4 out of 9. Appropriateness was determined by subspecialty panels of clinical and imaging experts and graded on a 1 through 9 (most appropriate) scale, where "the expected health benefit exceeds the expected negative consequences by a sufficiently wide margin that the procedure is worth doing, exclusive of cost" 6. These expert society statements suggest MRI as the most appropriate imaging modality for this patient. Z-joint–mediated lumbar pain accounts for 15%-30% of axial low back pain but may have protean clinical symptoms 7. Although classically associated with back pain, a "pseudoradicular" pattern of pain has been described that can mimic a disk herniation 8. The lumbar z-joints are considered 2 parts of the "3-joint complex" consisting of the intervertebral disk and bilateral z-joint articulations 9. The components of this interconnected lumbar motion segment degenerate in concert; however, intervertebral disk degeneration almost always occurs before that of the z joints. In a study including 41 patients with z-joint arthropathy, 40 of 41 patients had MRI findings of disk degeneration at the same level 10. Z-joint arthropathy without concomitant disk degeneration is sufficiently unusual that some authors recommend an additional workup to exclude underlying inflammatory arthropathy 11. This coexistence of disk degeneration along with z-joint arthropathy is one reason a specific diagnosis of z-joint–mediated pain can be challenging upon clinical evaluation. MRI offers additional important information in the imaging workup by evaluating for coexisting or masquerading disease. Entities that frequently coexist with z-joint disease include degenerative spondylolisthesis, synovial cysts, and z-joint tropism. Masquerading spinal disease may present with a confounding clinical picture that can overlap with z-joint OA, including disk herniation, transitional lumbosacral anatomy, spinal fracture, neoplasm, and a variety of infectious and inflammatory conditions. The broad overlap in clinical manifestations between these varied causes further reinforces the value of MRI in the workup of this patient. Standard lumbar spine imaging protocols for low back pain include sagittal T1, T2, and fat-suppressed T2 imaging along with axial T2-weighted sequences 12. The sagittal fat-suppressed T2 sequence is of particular importance when evaluating patients with low back pain, including z-joint pain, as fracture, bone marrow edema, and soft tissue edema become particularly conspicuous. The fat-suppressed T2 sequence can be achieved with short tau inversion recovery T2-weighted fat suppression, frequency selective chemical fat suppression, or newer Dixon-type chemical shift methods 13. Some authors advocate for the addition of a wide field-of-view coronal T2 fat-suppressed sequence to evaluate for extraspinal disease (for example, sacroiliac joint edema) that may have overlapping clinical symptoms 14. Deciding between superconducting 1.5- or 3-T magnets traditionally has been less important with degenerative spine imaging versus spinal cord or brain evaluation. In the setting of spinal hardware, 1.5-T imaging historically has been associated with decreased artifact; however, higher 3-T field strength imaging can be optimized by MR parameter optimization, including increased bandwidth, reduced field of view, and increased echo train length 15. Open MR scanners may be preferable for patients with claustrophobia but are associated with suboptimal imaging quality and prolonged imaging times 16. Dynamic flexion/extension (kinetic) MRI and axial load imaging may offer additional information with increased severity of symptomatic spinal stenosis upon axial loading and unmasking of spinal instability on kinetic MRI 17, 18. A number of MRI findings have been associated with z-joint–specific disease. A more sagittal orientation of the lumbar z joints has been strongly correlated with the development of degenerative spondylolisthesis 19. The joint angle can be readily calculated on axial images as described by Boden et al 20. T2-hyperintense fluid signal within the z joint measuring greater than 1 mm is predictive of same-level spondylolisthesis on standing lumbar radiographs, even if the spondylolisthesis is not apparent on the recumbent MRI 21. Periarticular T2-hyperintense signal, bone marrow lesions, and z-joint effusions are furthermore suggestive of a painful z joint and positive response to injection or surgical therapy 22. It is important to be mindful of the limitations of noninvasive imaging; percutaneous image-guided z-joint medial branch block with anesthetic remains the most specific test for isolating the z joint as the pain generator 23. Ideally, significant pain relief should be achieved for the half-life of the injected anesthetic. Existing evidence suggests that a single pain generator can be isolated and a combination of both z-joint and diskogenic pain or z-joint and sacroiliac pain is seen in fewer than 5% of patients 24, 25. The clinical identification of lumbar z-joint syndrome remains problematic 26. History and physical examination are of limited specificity because most maneuvers simultaneously stress multiple structures, including ligaments, muscles, disks, and joints 27. MRI is the most prudent imaging choice in this patient to exclude concomitant spinal disease, evaluate the anatomic integrity of the lumbar spine, and guide subsequent therapy. Dr Chazen supports the position that MRI of the lumbar spine is the most appropriate imaging study to further evaluate our patient with clinically suspected z-joint–mediated low back pain. I acknowledge that this opinion represents the more mainstream approach. In support of this stance, Dr Chazen cites the ACR Appropriateness Guidelines for low back pain as a primary reference 1. These guidelines recommend MRI as the most appropriate modality (with a rating of 8 out of 9) for evaluation of low back pain or radiculopathy in a patient who is a candidate for surgery and/or intervention. Such guidelines are incredibly useful for radiologists and referring clinicians alike. However, these recommendations must be interpreted and closely scrutinized in the appropriate clinical context. Thus, I believe it is worth highlighting a couple of important details related to this reference in the context of our patient with suspected z-joint pain, as outlined below. Most importantly, the ACR recommendations are written broadly for any patient with low back and/or radicular pain. Further, the recommendations do not specifically exclude patients with red flags. The case presented, however, is clearly a patient without red flags and without radicular symptoms. Radiographic and clinical findings for our patient are all highly suggestive of z-joint–mediated pain. Therefore, the pretest probability for z-joint disease is high. This information is important to consider, because the ACR rating of 4 out of 9 for a SPECT bone scan largely relates to its lower specificity for diagnosing radicular and red-flag diseases such as tumors, infection, or traumatic fractures 1. Relevant to our patient, the ACR guidelines specifically state that "bone scintigraphy with single photon-emission CT can be useful to identify symptomatic facet disease in patients treated with facet injection" 1. The rating table also lists the following comment related to bone scan: "may be particularly useful for facet arthropathy, stress fracture, and spondylolysis" 1. For these reasons, I believe the ACR guidelines strongly support the use of bone scan with SPECT for initial evaluation of our patient with a high pretest probability of z-joint disease. The ACR guidelines for low back pain were published in 2009, with no included references more recent than 2006 directly relevant to imaging low back pain. Since 2006, a number of clinical studies and review papers have been published in support of nuclear scintigraphy for workup of z-joint–mediated low back pain 2-5. These studies (and others) largely substantiate the findings of Pneumaticos and colleagues 6, further supporting the value of nuclear scintigraphy for evaluation of z-joint arthropathy as cited in my initial response. The ACR guidelines acknowledge that uncomplicated acute low back pain and/or radiculopathy do not warrant use of imaging studies and that indiscriminate use of expensive imaging procedures, such as MRI, lead to significant increases in cost without apparent benefit 1, 7, 8. Further, nonspecific lumbar abnormalities that are seen in painful z-joint disease on MRI are common and readily demonstrated in asymptomatic patients 1. Abundant literature suggests MRI may lead to many false-localizing sites of pain and potentially increased utilization of ineffective invasive interventions 6. Finally, Dr Chazen emphasizes the value of MRI for identification of clinically significant disease that may masquerade as z-joint pain. However, for this patient with no red flags and clinical symptomatology highly suggestive of z-joint disease, I believe the risk of missing clinically significant disease is relatively low. Certainly, in the setting of a negative bone scan and persistent or progressive symptoms, additional imaging with MRI may be indicated. However, as a first-line screening tool, I believe the current literature favors bone scintigraphy with SPECT for most effective targeting of z-joint intervention in our patient. In the context of spine imaging, different imaging modalities often offer complementary information. For example, a concerning lesion noted on MRI can have a characteristic trabeculated pattern on CT, sealing the diagnosis of a benign intraosseous hemangioma 1. Similarly, in the current clinical scenario, although the nuclear medicine imaging studies Dr Talbott discusses may offer useful diagnostic information, I do not believe that these studies are the most appropriate first-line imaging selection in this clinical context for 2 principle reasons. The first is that the diagnosis of z-joint–mediated pain is still uncertain. The second is that MRI is a noninvasive imaging study without exposure to ionizing radiation, and intravenous contrast material is not required in this setting. A variety of clinical entities can present with axial low back pain, as previously discussed, and MRI provides the most thorough diagnostic assessment to exclude underlying conditions, including potential infection or inflammatory or neoplastic entities 2. Furthermore, degenerative conditions involving the intervertebral disk, including herniations, will be occult at the time of SPECT imaging but can be accurately characterized on MRI. SPECT and PET have the promise of revealing important physiologic information but lack the soft tissue contrast and spatial resolution of MRI. These studies therefore may overlook critical disease that can alter the patient's clinical treatment algorithm. According to the National Council on Radiation Protection 3, ionizing radiation from medical imaging is estimated to account for nearly half of the radiation exposure experienced by the U.S. population. CT and nuclear medicine imaging are responsible for the majority of medical imaging exposure, representing 24% and 12%, respectively 4. National Council on Radiation Protection data reflect a 6-fold increase in the average individual exposure from medical imaging sources over a 25-year period, while background (naturally occurring) exposure has remained stable. These data have raised concerns for radiation-induced cancer risks from medical imaging. Although the majority of robust experimental data are derived from high-exposure atomic bomb survivors, a conservative linear no-threshold model has been widely adopted to mitigate exposure. The "as low as reasonably achievable" principle is generally applied to radiation exposure from medical imaging 5. The effective dose from Tc-99m MDP administration is estimated at 6.3 mSv, which will increase with CT co-registration 6; these studies should be limited to cases in which the diagnostic information gained is expected to be particularly informative. Conversely, MRI has no known adverse tissue effects in healthy subjects 7. The aforementioned nuclear imaging studies all require the intravenous administration of a radiotracer. Although rare, reports have been published of adverse effects from Tc-99m MDP administration 8. Patients also require placement of an intravenous line with the associated inconvenience. The ACR and ACP/APS guidelines suggest that a lumbar spine MRI be performed in this setting without use of intravenous contrast material. The widely cited study by Pneumaticos et al 9 on the predictive value of MDP-SPECT in patients with low back pain indicates the patients had already undergone "other diagnostic techniques, such as … MR imaging." Although nuclear medicine imaging offers the promise of added functional information and may help accurately identify the problematic z joint in a patient with multilevel disease, the existing literature most strongly support MRI as the most appropriate and diagnostically useful first-line examination in this clinical context. To cast your vote, visit www.pmrjournal.org
No takes yet. Share an insight, caveat, or question.
Chazen et al. (2016) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: