Đánh giá về đau lưng ppsx

Diskography in the Evaluation of Low Back Pain Abstract Diskography is evolving to play a crucial role in the evaluation of axial low back pain, especially in regard to surgical decision making. Despite advances in other forms of imaging, diskography remains unique in that it is the only test that seeks to provoke a pain response during the study. It has been suggested that patients with axial low back pain who experience a concordant pain response during diskography are more likely to respond favorably to surgical intervention. However, the efficacy of using this potential correlation is dependent on the technical application and interpretation. The validity of diskography remains controversial, in part because postdiskography surgical outcomes have been inconsistent. Therefore, in select patients with recalcitrant back pain, diskography remains a second-line diagnostic modality that is used to clarify surgical indications. Despite well-defined guidelines, the technical aspects of diskography and its interpretation are still evolving. L ow back pain (LBP) occurs fre- quently and, in some cases, may become chronic. 1 Many causes of LBP remain poorly understood. Al- though the disk is thought to be a pain generator, establishing a discogenic origin for LBP in a specific patient can be challenging. Evidence supports the concept that the intra-vertebral disk is a potential pain generator. The nucleus pulposus, the central core of the disk, is composed of type II collagen fibers, along with various muco- polysaccharides and glycosami- noglycans. This composition supports the high water content of the nucleus, which behaves biomechan- ically as a fluid cushion that trans- mits its loading forces to the outer anulus fibrosus as well as to the vertebral end plate. 2,3 The anulus is distinct from the nucleus and is prima- rily composed of type I collagen arranged in multiple concentric lay- ers. This fiber arrangement allows the anulus to optimally resist ten- sile, radial, and torsional forces. With acute trauma or the degenerative changes associated with time and repetitive microtrauma, the fibers of the anulus may be disrupted. Although the nucleus pulposus has no nerve supply, the outer third of the anulus is innervated. It receives supply from both the sinuvertebral nerve, which innervates the posterior and posterolateral regions, and the gray ramus, which is distributed pri- marily laterally and anteriorly. 4 Pro- liferation of nerve fibers containing substance P have been found in the anuli of degenerative disks and in the disks of patients undergoing spinal fusion for back pain. 5,6 These findings suggest that the anulus fibrosus may play a role in neuromodulation of pain and, more specifically, that the Spiros G. Pneumaticos, MD, PhD Charles A. Reitman, MD Ronald W. Lindsey, MD Dr. Pneumaticos is Assistant Professor, Department of Orthopedic Surgery, Baylor College of Medicine, Houston, TX. Dr. Reitman is Assistant Professor, Department of Orthopedic Surgery, Baylor College of Medicine. Dr. Lindsey is Professor and Chair, Department of Orthopedic Surgery, University of Texas Medical Branch, Galveston. None of the following authors or the departments with which they are affiliated has received anything of value from or owns stock in a commercial company or institution related directly or indirectly to the subject of this article: Dr. Pneumaticos, Dr. Reitman, and Dr. Lindsey. Reprint requests: Dr. Lindsey, Department of Orthopaedic Surgery, University of Texas Medical Branch, Rebecca Sealy Hospital, 2316 University Boulevard, Galveston, TX 77555. J Am Acad Or thop Surg 2006;14: 46-55 Copyright 2006 by the American Academy of Orthopaedic Surgeons. 46 Journal of the American Academy of Orthopaedic Surgeons mechanical irritation of the outer anulus can stimulate these nerve endings. 7 Additionally, annular fis- sures may allow nuclear cytokines and other inflammatory substrates to communicate with the nerve endings and cause a chemical irritation. 8-10 Discogenic pain may occur secondary to acute disk disruption, from chronic degenerative processes, or from conditions that result in seg- mental spine instability. Although episodes of acute LBP are prevalent in the general US population, most symptomatic episodes resolve within a short time. Only a small percentage of these patients develop chronic symptoms; of this group, most do not have a clear diagnosis. It is important for the clini- cian to differentiate patients with axial pain or possibly referred pain syndromes from those who have true radicular pain. In the latter, it i s usually easier to establish specific diagnosis because the workup, management, and prognosis of these patients greatly differ from those of patients with axial pain syndromes. Individuals with true radicular pain often have positive neurologic signs; the goal of treatment, therefore, is to relieve the cause of the neurologic disturbance. A thorough history and physical examination are essential for all patients. Those with discogenic pain syndromes will have established axial pain complaints of >3 months. Symptoms typically are aggravated by weight-bearing activity, and relative relief can be achieved from re- cumbency. The physical examination is often nonspecific. Lumbar spine range of motion is usually re- stricted secondary to pain. Paraspi- nal muscle spasm may be present, nerve-root tension signs are negative, and the neurologic examination is usually normal. Plain radiography also typically is nonspecific and may demonstrate disk space narrowing, end plate scle- rosis, or osteophyte formation. Dy- namic flexion and extension radiographs may detect elements of instability; computed tomography (CT) is rarely enlightening. Magnet- ic resonance imaging (MRI) may be appropriate for acute disk derange- ment; however, it does not consistently demonstrate abnormalities in the disk. In advanced degenerative conditions, MRI more routinely de- tects degenerative disk morphology. However, the relationship between MRI-depicted disk degenerative changes and LBP remains poorly understood. 11 Although MRI is highly sensitive in depicting disk disease, its specificity or ability to correlate symptoms with these changes is low. This low MRI sensitivity in establishing a discogenic origin for LBP is best reflected by the high incidence of disk abnormalities associated with lumbar spine MRI studies in asymptomatic volunteers. 12 Diskography may be considered to further evaluate the condition of patients diagnosed with discogenic LBP that is not responsive to an ex- tended course of nonsurgical treatment. Unlike other tests for LBP, diskography has a provocative com- ponent. It not only demonstrates abnormal morphology, but it can also assist in localizing the disk as a pain generator. Diskography can be ap- plied to determine the presence of a painful disk despite normal noninvasive imaging studies, to identify the painful disk or disks amid multiple degenerative disks, and to assist in defining the extent of surgery when multiple levels of potential pain gen- erators exist. However, the efficacy of diskography is controversial, and appropriate interpretation requires an in-depth appreciation of the patient, the pathologic condition or conditions, and the proper technique for its application. History Lindblom, 13 stimulated by the earli- er work of Schmorl and Lindgren, first reported a novel technique for the direct evaluation of potentially symptomatic lumbar disks. The test consisted of a diagnostic disk puncture and injection, which was called diskography. Lindblom noted that the use of diskography was indicated based on its ability to distinguish between normal and ruptured disks and to reproduce pain concordant with symptoms in symptomatic disks. This was corroborated by Hirsch, 14 who successfully provoked pain in 16 patients by injecting sa- line solution into the disk. Although diskography initially was sparingly used, its popularity further decreased in 1968, when Holt 15 demonstrated a 37% incidence of positive diskograms in a group of asymptomatic prison volunteers. Almost two decades later, Simmons et al 16 deemed Holt’s study to be flawed in its methodology. However, when Walsh et al 17 later duplicated Holt’s study with improved methodology, they reported no false-positive results. The sensitivity of diskography has been considerably enhanced by the advent of CT. Bernard 18 demonstrated that plain diskography in combination with postdiskography CT improved diagnostic accuracy. More recently, interest has developed in the degree of intradiskal pressure and its relationship to provocation during diskography. The use of manometry allows for the measurement and correlation of incre- mental injectable pressure with the resultant pain response. There is early evidence that this may permit more specific interpretation and diagnosis and thus may more accu- rately guide treatment. 19,20 Indications The indications for diskography in the management of LBP have been outlined in a position statement from the Executive Committee of the North American Spine Society Diagnostic and Therapeutic Com- mittee 21 and are described in Table 1. Patients considered for diskogra- Spiros G. Pneumaticos, MD, PhD, et al Volume 14, Number 1, January 2006 47 phy ideally should have low back symptoms for a minimum of 4 to 6 months that are nonresponsive to nonsurgical management, including physical therapy and medical treatment. The potentially painful level should be consistent with the findings on physical examination, plain radiographs, and MRI. In these select discogenic back pain patients, diskography serves as a test to confirm, not determine, the need for surgical intervention. Procedure Anesthesia and Positioning Because the procedure requires the patient’s response to pain provocation, it is imperative that only a mild sedative be administered intravenously, such as midazolam 2 mg with fentanyl 25 µg. The merits of prophylactic antibiotic use currently is controversial. 22 When administered, antibiotics can consist of 1 g cefazolin given intravenously within 1 hour before the procedure, as well as cefazolin 0.5 mL added into the injectate for each disk. Two standard approaches are ad- vocated for access to the lumbar disks: the direct posterior or intra- pedicular approach and the posterolateral or extrapedicular approach. The posterolateral approach more commonly is used. 23 The patient is positioned prone on a radiolucent table in a manner to accommodate anteroposterior and lateral fluoroscopic images. The preferred side for injection is that contralateral to the patient’s dominant pain. The levels to be evaluated are determined by the clinical presentation of the patient and the results of the imaging studies. A legitimate study always should include at least one control level consisting of a normal-appearing, asymptomatic disk adjacent to the level or levels being evaluated. Posterolateral Techniques The two-needle technique is preferred to lower the incidence of postdiskography infection. The skin is anesthetized with local anesthesia approximately 5 to 6.5 cm from the midline of the level of the intervertebral disk. The initial larger, cannu- lated needle (18-gauge) is first passed through the skin and soft tissue to the anulus of the disk; this needle guides the second, smaller needle (22- or 25-gauge) that actually punc- tures the anulus. Fluoroscopy is used to confirm the disk level and to direct the larger needle from the skin puncture toward the anterior margin of the superior articular process and into the posterolateral aspect of the disk (Figure 1, A). The smaller needle is then directly inserted through the larger needle to puncture each disk and terminate within the nucleus pulposus. Proper needle placement in the disk is confirmed by vi- sualizing the tip of the needle in the inner one third of the disk in both the anteroposterior and lateral radiographs. Confirmation of optimal needle placement can be challenging at the L5-S1 level, where significant caudal angulation is required for adequate visualization. Anatomic landmarks of the safe zone for needle insertion are the inferior end plate of L5, the superior articulating process of S1, and the iliac crest (Figure 1, B). Most of the technical difficulties encoun- tered when injecting the L5-S1 space can be avoided by selecting a higher puncture site compared with the other disk spaces. When the iliac crest obstructs the approach despite adjusting the angle of trajectory, then alternative approaches, such as a midline transthecal L5-S1 puncture, might be warranted. However, this approach clearly carries a higher risk for complications and theo- retically may provoke more pain. Assessment After all needles have been inserted into the levels to be studied, a water-soluble contrast solution is injected with the patient unaware of the precise moment of injection, the level or levels injected, and the exact amount of contrast infiltrated. Disk pressurization should first be performed at the control level to establish the patient’s pain tolerance. As other levels are injected, the patient is asked to determine whether the injection provoked pain and to compare these symptoms with his or her usual pain level in both distribution and quality. Digital manometry should be used to measure intradiskal pressure at the time of the injection and at 0.5-mL injectable increments. The static disk pressure (relative to the opening pressure) associated with a pain response is noted, as are the level of pain intensity and the concordance of the provoked symptoms. When the test is done properly, a Table 1 Indications for Diskography in a Position Statement From the Executive Committee of the North American Spine Society 21 Unremitting spinal pain, with or without extremity pain, of >4 months’ duration, not responsive to all standard methods of conservative treatment Persistent disk-related pain, suspected when other evaluation modalities are equivocal Persistent pain in the postoperative period as a result of suspected intervertebral disk degeneration, recurrent herniation, or a pseudarthrosis Disk space evaluation in a spine segment considered for fusion to determine whether it is a pain generator Determination of the primary symptom-producing level or levels when chemonucleosis or other intradiskal procedures are being contemplated Diskography in the Evaluation of Low Back Pain 48 Journal of the American Academy of Orthopaedic Surgeons morphologically normal disk should not produce pain. It is imperative to avoid excessively high intradiskal pressures, which may precipitate false-positive results. 20 The normal disk accepts a fluid volume of between 1.5 and 2.5 mL; more than 3 mL is abnormal. Increased intradiskal contrast volume often suggests a complete annular tear with leakage of fluid into the epidural space. Once the disk has been injected, images are recorded with high-quality bipla- nar radiographs, followed by CT axial scans. Interpreting the Results An accurate diskogram interpretation requires the documentation and analysis of several vital pieces of information: (1) the disk pressure and the volume of fluid accepted by the disk; (2) the pattern of the contrast distribution or morphology of the injected disks; and (3) the subjective pain response of the patient to the injection, compared with the absence of a pain response in an adjacent control level. 21 One classification for morphologic pattern of progressive degeneration is shown in Figure 2. Additional information on disk morphology and stage of degeneration can be obtained from the postdiskography axial CT images (Figure 3). The nature and extent of the pathology are better defined, and some lesions, such as annular fissuring and lateral disk herniations, are more clearly delineated. A diskogram is considered positive when it depicts an abnormal disk and concordant pain is reproduced during an injection with reasonable disk pressures (Table 2). O’Neill and Kurgansky 20 concluded that disk pressures >50 psi were likely to result in a false-positive pain response. Furthermore, they identified two distinct subgroups of true-positive disks on diskography: contact-sensitive and pressure- sensitive. In contact-sensitive disks, simple contact between the contrast and the disk was sufficient to provoke pain. True-positive pressure- sensitive disks demonstrated a concordant pain response between and 10 and 50 psi. The investigators were unable to differentiate false-positive from true-positive disks when the disk pain threshold was <10 psi. Derby et al 19 also identified a low- pressure–sensitive group of patients with positive diskography. They noted that pressures <15 psi were insuf- ficient to cause significant mechanical deformation; patients who had positive diskography with pressures <15 psi were termed to have chemically sensitive disks. The authors postulated that the level of pain- provoking pressure may be related to specific surgical indications. Figure 1 A, Proper positioning for L4-5 diskogram. The superior articular process of L5 (S) is positioned midway in the superior end plate of L5 (SEP). Correct needle placement is shown by the black target. I = inferior articular process of L4; P = pedicle of L4, L5. B, Proper positioning for L5-S1 diskogram. The small inverted triangle defined by the superior articular process of S1 (SAP), the iliac crest (IC), and the inferior end plate of L5 (IE) marks correct needle placement. Spiros G. Pneumaticos, MD, PhD, et al Volume 14, Number 1, January 2006 49 Complications Complications resulting from diskography are rare; among the most serious is diskitis. The reported incidence of diskitis is 0.1% to 0.2% 24 and reflects a steady decrease in oc- currence since 1980. This decline can be attributed to the use of the two-needle technique, prophylactic antibiotics, improved fluoroscopic equipment, and use of smaller needles. 25 Other reported complications, although less prevalent, include tem- porary headache, nausea, meningi- tis, epidural abscess, arachnoiditis, intrathecal hematoma, and intradu- ral injection of contrast. 26 In addition, there have been isolated reports of urticaria, retroperitoneal hematoma, cauda equina syndrome, and acute disk herniation. Perhaps the most frequent complaint is pain ex- acerbation for 1 to 2 weeks, which resolves with short-term administration of analgesics and muscle re- laxants. Early concerns regarding the long-term adverse effects of diskography on disk viability have not ma- terialized. In an up to 20-year clinical follow-up study, Flanagan and Chung 27 reported no radiographic evidence of progressive disk degeneration following diskography. Validity Diskography initially was developed as a provocative test for assess- ing patients with disk herniation. As experience with this modality increased, it became evident that diskography also could provide morphologic details about the internal Figure 2 Classification of stages of disk degeneration. (Reproduced with permission from Adams MA, Dolan P, Hutton WC: The stages of disc degeneration as revealed by diskograms. J Bone Joint Surg Br 1986;68:36-41.) Figure 3 A, Lateral CT image of a lumbar spine immediately following a diskogram at the L3-4, L4-5, and L5-S1 segments. The upper level (L3-4) is normal, with a cottonball appearance. Contrast material remains central. The middle level (L4-5) is irregular, with the majority of the contrast in the central nucleus and a focal area of contrast extension in the anulus. The lower level (L5-S1) is an example of an annular fissure, with contrast extending to the posterior annular margin. The upper level is normal, the middle level shows some degeneration, and the lower level is incompetent with advanced degeneration. B, Axial CT image of the middle level in panel A, showing an annular tear with dye tracking to the posterolateral anulus. Table 2 Criteria for Establishing a Positive Diskogram Abnormal disk morphology, including posterior annular disruptions Pain concordant to the patient’s usual pain Pain limited to 1 or 2 disk levels Negative control Diskography in the Evaluation of Low Back Pain 50 Journal of the American Academy of Orthopaedic Surgeons architecture of the disk. Although the association of diskography findings with symptoms is not well understood, some early investigators did manage to correlate back pain with disk morphology. The issue of disk morphologic change and its association with pain became extremely controversial when Holt 15 in 1968 performed mul- tilevel lumbar diskograms in 30 asymptomatic volunteer inmates and reported a 37% false-positive rate (pain provocation in asymptomatic participants). In this study, a positive disk was designated by either provocation of pain or the presence of a morphologically abnormal disk. When Holt’s study was later reas- sessed by Simmons et al, 16 it was noted that diatrizoate (the contrast medium used) was more irritating than currently available contrast me- dia. Furthermore, Holt claimed that his patients were asymptomatic volunteers, yet the selection process was not detailed. Additionally, the imaging equipment and lumbar puncture technique employed were substan- dard, and at least some of the injec- tions were done through inaccurately placed needles. Finally, a positive diskogram was not clearly defined; this was most evident in the diskography interpretation made in patients with a positive response of concordant pain without initial symptoms. Walsh et al 17 evaluated the results of diskography in 10 asymptomatic volunteers as well as in 7 symptomatic patients. Morphologically abnormal disks were observed in 17% of the injected disks in the asymptomatic volunteers and in 65% of the symptomatic patients. None of the diskograms in the asymptomatic group was associated with significant pain, whereas typical pain was reproduced during the injection in six of the seven symptomatic patients. The authors concluded that the single most important determinant of accurate sensitivity was a subjective patient pain response. In three recent diskography studies, Carragee and colleagues 28-30 made several important observations (Table 3). In the first study, a group of patients (mean age, 43 years) without LBP was studied by diskography. 28 Ten were asymptomatic, 10 had chronic neck pain, and 6 had a somatization disorder. The authors determined that disks with annular disruption were more likely to be painful. However, this association was considerably more com- mon in the group with an associated chronic pain syndrome. In this study, 10% of the asymptomatic group, 40% of the chronic neck pain group, and 83% of the somatization group experienced painful lumbar diskograms. In the second study, Carragee et al 29 studied 24 disks in 8 patients who had previously undergone iliac crest bone grafting for a non-thora- columbar–related disorder. Disko- grams were performed in all disks 2 to 4 months following iliac crest surgery. Because these patients had no lumbar spine symptoms, concordance was determined as reproduc- tion of their familiar bone graft site pain. Four of the eight patients demonstrated concordant iliac crest pain, and the authors concluded that dif- ferentiating between spinal and non- spinal origins of pain was difficult. Again, intensity of pain response was greater in disks with annular disruption compared with disks clas- sified as normal or intermediate. In the last study, 25 patients with mild persistent LBP were evaluated with diskography. 30 None of the subjects was limited functionally, nor were any seeking medical attention for their symptoms. Nine of 25 patients (36%) had fully concordant diskography with negative control levels. These results imply that diskography may not be specific in dif- ferentiating clinically relevant pathology. Fifty-two patients being considered for surgery had diskography as part of their preoperative evaluation and were used as control subjects; in this group, 73% had positive diskograms. Low-pressure in- jections made no significant differ- ence in the percentage of positive diskograms among the two groups; in the experimental group, a positive response was more likely in patients with a history of chronic pain. 30 Table 3 Cumulative Results of Diskography 28,30-32 Patient Group Patients Reporting Pain (%) Mild low back pain (n = 25) 30 36 Chronic neck pain (n = 10) 28 40 Somatization disorder (n = 6) 28 83 Severe low back pain (n = 52) 30 73 Posterior iliac bone graft (n = 8) 29 50 HIZ lesion: asymptomatic with normal psychometrics (n=4) 32 50 HIZ lesion: asymptomatic with abnormal psychometrics (n=6) 32 100 Posterior disk surgery: asymptomatic (n = 20) 31 40 Posterior disk surgery: symptomatic with normal psychometrics (n = 7) 31 43 Posterior disk surgery: symptomatic with abnormal psychometrics (n = 20) 31 70 HIZ = high-intensity zone Spiros G. Pneumaticos, MD, PhD, et al Volume 14, Number 1, January 2006 51 Other diskography studies by Car- ragee and colleagues 31,33,34 have conclusively demonstrated that assessment of the patient’s psychometric profile is critical in the comprehensive evaluation of LBP. These charac- teristics are shown to be predictors o f the development of LBP as well as of the patient’s response to diskography. These studies challenge the ability of structural variables such as diskography (or even MRI) to detect clinically relevant disk pathology. 33 However, clinically useful information can be obtained with the careful analysis of specific pain-response patterns in subgroups of patients with LBP. Chronic LBP patients with abnormal psychometric evaluations and/or somatization features demonstrated notably higher painful responses. 31,33,34 These facts should be determined, understood, and appro- priately factored into the interpretation of provocative diskography. Diskography Versus Magnetic Resonance Imaging Prior to the advent of magnetic resonance imaging (MRI) in the 1980s, diskography was the only means available of depicting the degenerative quality of the disk. Ini- tially, MRI was considered to be at least as accurate as diskography in the assessment of an abnormal disk while maintaining all of the benefits of a noninvasive procedure that did not require ionizing radiation. 35 This position was supported by Schnei- derman et al, 36 who reported MRI to be accurate in predicting normal versus abnormal disk morphology in 100 of 101 disks compared with diskography. Later studies, however, demonstrated that diskography may detect abnormal disks in the presence of a normal MRI and that many abnormal disks on MRI are not painful with provocative diskography. 18,37 Furthermore, it has been suggested that there are no MRI features that can reliably predict positive provocation during diskography. 38 Aprill and Bogduk 39 were the first to describe the high-intensity zone (HIZ) (Figure 4), which was detected in the MRI scans of 28% of 500 patients with back symptoms without radiculopathy. In addition, they reviewed a subset of 41 patients with HIZs who also were studied with CT diskography. The presence of an HIZ correlated notably with high-grade annular tears and was strongly related to concordant pain patterns. The authors suggested that the HIZ was pathognomonic of an internally disrupted and symptomatic disk. Oth- ers have since verified the high sensitivity of an MRI-depicted HIZ in predicting an annular tear. 40 Several other studies, however, have not supported the high correlation of an HIZ to a diskographically concordant disk. 41,42 Horton and Daftari 43 studied 63 disks in 25 patients with LBP; they evaluated specific MRI patterns and their association with results of diskography . The authors identified consistently be- nign patterns that had a strong negative correlation, as well as other advanced degenerative or disrupted patterns that had a strong positive correlation with concordant provoked pain. Carragee et al 32 also studied the prevalence and significance of the HIZ in an asymptomatic population prone to degenerative changes, as well as in a group of patients with LBP. In 50% of disks with an HIZ and with normal psychometrics (based on a Modified Zung Depres- sion Test and a Modified Somatic Pain Questionnaire), the diskogram was positive, whereas it was positive in all of the disks with abnormal psychometrics. The authors concluded that an HIZ does not reliably reflect the presence of internal disk Figure 4 A, Sagittal T2-weighted MRI scan of the lumbar spine demonstrating high-intensity zones at the two lower disk levels (arrows). B, Lateral radiograph of the lumbar spine with a high-intensity zone at L4-5 following an L3-4, L4-5, and L5-S1 diskography. Note the leakage of the contrast material at L4-5, the segment corresponding with the high-intensity zone. Diskography in the Evaluation of Low Back Pain 52 Journal of the American Academy of Orthopaedic Surgeons disruption or a concordantly painful disk. Currently, an HIZ appears to be most sensitive for an outer annular disruption or fissure. However, the significance of this finding in regard to diskogenic LBP is still in dispute. Diskography and Treatment Outcome In most studies that have ad- dressed outcomes related to the treatment of discogenic back pain as evaluated by diskography, inclusion criteria required the presence of chronic LBP—that is, at minimum, LBP of more than 1 year’s duration. Furthermore, all patients will have failed comprehensive nonsurgical treatment, with many unable to work as a result of back pain. Also, all patients will have had preoperative diskography to assist with surgical decision making. The traditional surgical treatment of discogenic pain has been lumbar fusion, and several uncontrolled, ret- rospective studies have evaluated the results of fusions (anterior interbody , posterolateral, posterior interbody, and/or 360°) for diskographically confirmed discogenic pain. 1,44-46 Clini- cally successful results varied from approximately 40% to 90%; these authors noted a 74% functional success rate (return to work, no medica- tion requirements) at approximately 2.5 years postoperatively. 44 Blumen- thal et al 44 analyzed 34 patients with LBP confirmed by abnormal diskography and treated with anterior interbody fusion. Wetzel et al 46 reported 66% satisfactory clinical outcomes in 48 patients fused anteriorly or posteriorly; successful outcome was strongly correlated with achiev- ing a solid arthrodesis. Linson and Williams 47 reviewed 51 patients with chronic LBP (duration >1 year) treated with anterior lumbar interbody fusion or 360° fusion and compared their outcomes with those of a similar group of 10 patients who could not undergo surgery because of state insurance reg- ulations. According to the Oswestry Disability Index and patient reports of pain, the study demonstrated an 80% measurable diminution of preoperative pain at follow-up of 15 to 36 months. The results of anterior versus 360° fusion were similar. In the nonsurgical group, by compari- son, only 1 of the 10 improved after a mean follow-up of 24 months. Colhoun et al 48 reviewed 162 patients who had preoperative diskography before anterior or posterior fusions. This is the only study to date to compare outcomes in patients undergoing fusion for chronic LBP who had normal, nonprovocative diskograms with those who had concordant diskograms. In the study, 137 patients had a least one concordant pain level on diskography; 25 had morphologically abnormal disks but were asymptomatic with provocation (diskogram negative). The diskogram-positive group had an 89% clinical success rate compared with only 52% for the diskogram- negative group; follow-up ranged from 2 to 10 years. In the study of Derby et al, 19 78 subjects were treated for chronic LBP. Based on diskographic charac- teristics, several subcategories were defined; these included a group of patients who experienced concordant pain with low-pressure injection. Chemically sensitive disks were demonstrated in 36 patients; treatment consisted of anterior interbody or combined fusion (9 patients), posterior intertransverse fusion (10), and nonsurgical management (17). In this subgroup, the most favorable outcome (89% good or excellent results) was observed in the patients treated with fusion (either anterior or combined). Patients treated non- surgically demonstrated only 12% good or excellent results. Based on these observations, the authors concluded that precise categorization of diskogram responses may improve surgical decision making and outcome. The efficacy of diskography in treatment outcomes is still in dispute because class I evidence (prospective, randomized clinical trials) is still lacking. Although the current diskography literature is extensive, no standardization exists in the assessment of clinical outcomes, the choice of surgical approach, patient selection, and psychometric evaluation. Furthermore, satisfactory clinical outcomes have varied signifi- cantly. Nonetheless, for studies in which reasonable comparisons have been possible, the clinical outcomes in patients with positive diskograms were more favorable than in patients with nonconcordant provocative studies. 19,47,48 Other surgical strategies for discogenic pain, such as intradiskal electrothermal therapy (IDET), have recently been developed. IDET is a percutaneous technique that in- volves the placement of a heat probe into the disk space. The mechanism by which this affects pain is unclear, but it is thought to be the thermal ablation of pain-sensitive fibers in the outer anulus and/or the change in the biomechanical properties of the anulus as the structure of collagen is altered during heating and subsequent cooling. In general, IDET is considered to be beneficial in a very select group of patients, for which inclusion criteria include positive diskography. Saal and Saal 49 reported on 58 patients treated with IDET and as- sessed at 2-year follow-up by visual analog scale (VAS) and the Medical Outcomes Study 36-Item Short Form (SF-36). These patients demonstrated significant improvement in pain (SF-36, P = 0.0017; VAS, P = 0.4960), quality of life (SF-36, P = 0.0001), and physical functioning (SF-36, P = 0.0001). Symptomatic improvement was maintained at 2 years. Bogduk and Karasek 50 followed a group of 53 patients for 2 years after failure of nonsurgical treatment of chronic LBP. Although the study was not randomized, 36 patients were treated with IDET, and 17 patients served as control sub- Spiros G. Pneumaticos, MD, PhD, et al Volume 14, Number 1, January 2006 53 jects. None of the control patients improved, whereas >50% of the treatment group experienced at least a 50% pain reduction; 20% were pain free. Both studies 49,50 concluded that IDET can be beneficial in a care- fully selected group of patients with discogenic pain. More recently, disk arthroplasty has been employed for the treatment of discogenic pain. Although the procedure has been employed in Europe for many years, it has only recently been approved for use in the United States. Inclusion criteria for the US Food and Drug Administration study were very strict and included positive diskography. Two centers reported on their combined 2-year results as part of the FDA trial, 51 which consisted of a prospective randomized study comparing 100 patients with disk arthroplasty with 44 patients undergoing anterior lumbar interbody fusion with threaded cag- es. Improvement was reported in both treatment groups, and their re- spective outcomes and complications were similar. Disk replace- ment offers the hypothetical advantage of maintaining the motion segment, but how it compares in the long term with the traditional spine fusion remains to be learned. Summary Considerable controversy exists in reporting on the efficacy of diskography. Proponents conclude that diskography is essential for the workup of patients with persistent back pain because of its ability to designate a specific disk as a potential pain generator. However, the mechanism by which pain is reproduced is not well understood, and the prognostic value of the symptomatic levels identified has not been conclusively demonstrated. Opponents of diskography also conclude that this diagnostic modality results in unnecessary surgery. Diskography should be considered when all other treatment modalities have failed and surgery is being contemplated. A position statement of the North American Spine Society advocates the use of diskography within strict guidelines. Finally, a positive diskography evaluation consists of an abnormal disk with a concordant pain response in a patient with an adjacent normal, asymptomatic disk or disks. Even in this scenario, however, diskography results must be considered as an ad- junct in the assessment of the indi- vidual patient. References Evidence-based Studies: Carragee et al (reference 31) is a level I study. Carragee et al (references 33 and 34) and Collins et al (reference 38) are level III studies. 1. Parker LM, Murrell SE, Boden SD, Horton WC: The outcome of posterolateral fusion in highly selected patients with discogenic low back pain. Spine 1996;21:1909-1917. 2. Heggeness MH, Doherty BJ: Discogra- phy causes end plate deflection. Spine 1993;18:1050-1053. 3. Reitman CA, Hipp JA, Kirking BC, Haas S, Esses SI: Posterior annular strains during discography. J Spinal Disord 2001;14:347-352. 4. Bogduk N, Tynan W, Wilson AS: The nerve supply to the human intervertebral discs. J Anat 1981;132(pt1):39- 56. 5. Coppes MH, Marani E, Thomeer RT, Groen GJ: Innervation of “painful” lumbar discs. Spine 1997;22:2342- 2349. 6. Freemont AJ, Peacock TE, Goupille P, Hoyland JA, O’Brien J, Jayson MIV: Nerve ingrowth into diseased intervertebral disc in chronic back pain. Lancet 1997;350:178-181. 7. Lee SH, Derby R, Chen Y, Seo KS, Kim MJ: In vitro measurement of pressure in intervertebral discs and annulus fibrosus with and without annular tears during discography. Spine J 2004;4: 614-618. 8. Aoki Y, Rydevik B, Kikuchi S, Ol- marker K: Local application of disc- related cytokines on spinal nerve roots. Spine 2002;27:1614-1617. 9. Kawakami M, Tamaki T, Hashizume H, Weinstein JN, Meller ST: The role of phospholipase A2 and nitric oxide in pain-related behavior produced by an allograft of intervertebral disc material to the sciatic nerve of the rat. Spine 1997;22:1074-1079. 10. Saal JS: The role of inflammation in lumbar pain. Spine 1995;20:1821- 1827. 11. Nachemson A: Lumbar discography—where are we today? Spine 1989;14:555-557. 12. Boden SD, Davis DO, Dina TS, Patronal NJ, Wiesel SW: Abnormal magnetic-resonance scans of the lumbar spine in asymptomatic subjects. J Bone Joint Surg Am 1990;72:403- 408. 13. LindblomK: Diagnostic disc puncture of intervertebral discs in sciatica. Acta Orthop Scand 1948;17:231-239. 14. Hirsch C: An attempt to diagnose the level of a disc lesion clinically by disc puncture. Acta Orthop Scand 1949; 18:132-140. 15. Holt EP: The question of lumbar discography. J Bone Joint Surg Am 1968; 50:720-726. 16. Simmons JW, Aprill CN, Dwyer AP, Brodsky AE: A reassessment of Holt’s data on: “The question of lumbar discography″. Clin Orthop 1988;237: 120-124. 17. Walsh TR, Weinstein JN, Spratt KF, Lehmann TR, Aprill C, Sayre H: Lum- bar discography in normal subjects. J Bone Joint Surg Am 1990;72:1081- 1088. 18. Bernard T: Lumbar discography followed by computed tomography: Re- fining the diagnosis of low-back pain. Spine 1990;15:690-707. 19. Derby R, Howard MW, Grant JM, Let- tice JJ, Van Peteghem PK, Ryan DP: The ability of pressure-controlled discography to predict surgical and nonsurgical outcomes. Spine 1999;24: 364-372. 20. O’Neill C, Kurgansky M: Subgroups of positive discs on discography. Spine 2004;29:2134-2139. 21. Guyer RD, Ohnmeiss DD: Lumbar discography: Position statement from the North American Spine Society Di- agnostic and Therapeutic Commit- tee. Spine 1995;20:2048-2059. 22. Willems PC, Jacobs W, Duinkerke ES, DeKleuver M: Lumbar discography: Should we use prophylactic antibiotics? A study of 435 consecutive disco- grams and a systematic review of the literature. J Spinal Disord Tech 2004; 17:243-247. 23. Konings JG, Veldhuizen AG: Topo- graphic anatomical aspects of lumbar disc puncture. Spine 1988;13:958- 961. Diskography in the Evaluation of Low Back Pain 54 Journal of the American Academy of Orthopaedic Surgeons 24. Guyer RD, Ohnmeiss DD, Mason SL, Shelokov AP: Complications of cervi- cal discography: Findings in a large se- ries. J Spinal Disord 1997;10:95-101. 25. Osti OL, Fraser RD, Vernon-Roberts B: Discitis after discography: The role of prophylactic antibiotics. J Bone Joint Surg Br 1990;72:271-274. 26. Tehranzadeh J: Discography 2000. Radiol Clin North Am 1998;36:463- 495. 27. Flanagan MN, Chung BU: Roentgeno- graphic changes in 188 patients 10-20 years after discography and chemonu- cleolysis. Spine 1986;11:444-448. 28. Carragee EJ, Tanner CM, Khurana S, et al: The rates of false-positive lumbar discography in select patients without low back symptoms. Spine 2000;25:1373-1381. 29. Carragee EJ, Tanner CM, Yang B, Brito JL, Truong T: False-positive findings on lumbar discography: Reliability of subjective concordance assessment during provocative disc injection. Spine 1999;24:2542-2547. 30. Carragee EJ, Alamin TF, Miller J, Grafe M: Provocative discography in volunteer subjects with mild persistent low back pain. Spine J 2002;2: 25-34. 31. Carragee EJ, Chen Y, Tanner CM, Truong T, Lau E, Brito JL: 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. 32. Carragee EJ, Paragioudakis SJ, Khura- na S: Lumbar high-intensity zone and discography in subjects without low back problems. Spine 2000;25 (6 suppl): 2987-2992. 33. Carragee EJ, Alamin TF, Miller JL, Carragee JM: Discographic, MRI and psychosocial determinants of low back pain disability and remission: A prospective study in subjects with be- nign persistent back pain. Spine J 2005;5:24-35. 34. Carragee EJ, Barcohana B, Alamin T, van den Haak E: Prospective controlled study of the development of lower back pain in previously asymptomatic subjects undergoing experimental discography. Spine 2004;29: 1112-1117. 35. Gibson MJ, Buckley J, Mawhinney R, Mulholland RC, Worthington BS: Magnetic resonance imaging and discography in the diagnosis of disc degeneration. J Bone Joint Surg Br 1986; 68:369-373. 36. Schneiderman G, Flannigan B, King- ston S, Thomas J, Dillin WH, Watkins RG: Magnetic resonance imaging in the diagnosis of disc degeneration: Correlation with discography. Spine 1987;12:276-281. 37. Zucherman J, Derby R, Hsu K, et al: Normal magnetic resonance imaging with abnormal discography. Spine 1988;13:1356-1359. 38. Collins CD, Stack JP, O’Connell DJ, et al: The role of discography in lumbar disc disease: A comparative study of magnetic resonance imaging and discography. Clin Radiol 1990;42: 252-257. 39. Aprill C, Bogduk N: High-intensity zone: A diagnostic sign of painful lumbar disc on magnetic resonance imaging. Br J Radiol 1992;65:361- 369. 40. Saifuddin A, Braithwaite I, White J, Taylor BA, Renton P: The value of lumbar spine magnetic resonance imaging in the demonstration of annular tears. Spine 1998;23:453-457. 41. Ito M, Incorvaia KM, Yu SF, Fredrick- son BE, Yuan HA, Rosenbaum AE: Predictive signs of discogenic lumbar pain on magnetic resonance imaging with discography correlation. Spine 1998;23:1252-1260. 42. Ricketson R, Simmons JW, Hauser BO: The prolapsed intervertebral disc: The high-intensity zone with discography correlation. Spine 1996;21: 2758-2762. 43. Horton WC, Daftari TK: Which disc as visualized by magnetic resonance imaging is actually a source of pain? A correlation between magnetic resonance imaging and discography. Spine 1992;17 (suppl 6): S164-S171. 44. Blumenthal S, Baker J, Dossett A, Sel- by DK: The role of anterior lumbar fusion for internal disc disruption. Spine 1988;13:566-569. 45. Loguidice VA, Johnson RG, Guyer RD, et al: Anterior lumbar interbody fusion. Spine 1988;13:366-369. 46. Wetzel FT, LaRocca SH, Lowery GL, Aprill CN: The treatment of lumbar spinal pain syndromes diagnosed discography. Spine 1994;19:792-800. 47. Linson MA, Williams H: Anterior and combined anteroposterior fusion for lumbar disc pain. Spine 1991;16:143- 145. 48. Colhoun E, McCall IW, Williams L, Pullicino VNC: Provocation discography as guide to planning operations on the spine. J Bone Joint Surg Br 1988; 70:267-271. 49. Saal JA, Saal JS: Intradiscal electrothermal treatment for chronic discogenic low back pain. Spine 2002;27: 966-974. 50. Bogduk N, Karasek M: Two-year follow-up of a controlled trial of intradiscal electrothermal anuloplasty for chronic low back pain resulting from internal disc disruption. Spine J 2002;2:343-350. 51. Guyer RD, McAfee PC, Hochschuler SH, et al: Prospective randomized study of the Charite artificial disc: Data from two investigational centers. Spine J 2004;4(6 suppl):252S- 259S. Spiros G. Pneumaticos, MD, PhD, et al Volume 14, Number 1, January 2006 55