of the lumbar facet joints there

Transcription

1 Skeletal Radiol (1999) 28:215±219 International Skeletal Society 1999 A R T I C L E Dominik Weishaupt Marco Zanetti Norbert Boos Juerg Hodler MR imaging and CT in osteoarthritis of the lumbar facet joints Received: 12 June 1998 Revision requested: 20 October 1998 Revision received: 29 December 1998 Accepted: 14 January 1999 D. Weishaupt, M.D. M. Zanetti, M.D. J. Hodler, M.D. ()) Department of Radiology, Orthopedic University Hospital Balgrist, Forchstrasse 340, CH-8008 Zurich, Switzerland N. Boos, M.D. Department of Orthopedic Surgery, Orthopedic University Hospital Balgrist, Zurich, Switzerland Abstract Objective. To test the agreement between MR imaging and CT in the assessment of osteoarthritis of the lumbar facet joints, and thus to provide data about the need for an additional CT scan in the presence of an MR examination. Design and patients. Using a fourpoint scale, two musculoskeletal radiologists independently graded the severity of osteoarthritis of 308 lumbar facet joints on axial T2-weighted and on sagittal T1- and T2-weighted turbo-spin-echo images and separately on the corresponding axial CT scans. Kappa statistics and percentage agreement were calculated. Results. The weighted kappa coefficients for MR imaging versus CT were 0.61 and 0.49 for readers 1 and 2, respectively. The weighted kappa coefficients for interobserver agreement were 0.41 for MR imaging and 0.60 for CT, respectively. There was agreement within one grade between MR and CT images in 95% of cases for reader 1, and in 97% of cases for reader 2. Conclusion. With regard to osteoarthritis of the lumbar facet joints there is moderate to good agreement between MR imaging and CT. When differences of one grade are disregarded agreement is even excellent. Therefore, in the presence of an MR examination CT is not required for the assessment of facet joint degeneration. Key words Magnetic resonance (MR) imaging, comparative studies Spine, diseases Spine, facet joints Introduction Magnetic resonance (MR) imaging is the imaging method of choice for degenerative disk disease, spinal stenosis, infection, and neoplasia of the lumbar spine [1]. Computed tomography (CT) has been largely replaced by MR imaging for these diagnoses. The role of MR imaging in the evaluation of facet joint degeneration is less clear. Osteoarthritis of these joints may be demonstrated in patients who present with back pain with or without pain radiating into the legs [2±4]. Grenier et al. [5] have shown that MR imaging accurately demonstrates abnormalities of the posterior spinal structures. However, for routine clinical use MR imaging is not usually considered to be equivalent to CT for the evaluation of facet joints [6]. Due to its more precise demonstration of bony details [7±9] CT is still preferred for this indication [6]. As health care resources are limited, imaging should be directed to obtaining the maximum information in as few steps as possible [10]. The purpose of this study was to test the agreement between MR imaging and CT in the assessment of osteoarthritis of the lumbar facet joints, and thus to provide data about the need for an additional CT scan in the presence of an MR examination. Materials and methods Patient characteristics Fifty consecutive patients referred for MR imaging of the lumbar spine at our institution who had an additional CT scan of the lumbar

2 216 spine within 6 months were included in this study. It was rarely possible to determine retrospectively the reason for referral for both studies. There were 28 women and 22 men. Their mean age was 59 years (range 25±85 years). Each patient included in the study had low back pain. None had undergone prior spinal surgery. The main indication for MR imaging of the 50 patients was suspected disk pathology in 40 cases (80%), suspected spinal canal stenosis in 7 cases (14%), vertebral fractures in 2 cases (4%) and spondylodiskitis in 1 case (2%). No patient was studied primarily to evaluate facet joint degeneration. Imaging parameters All MR studies were performed on the same 1.0 T imaging system (Siemens Impact, Siemens Medical Systems, Erlangen, Germany) with a dedicated receive-only spine coil. Each study included T1- weighted (TR 700 ms, TE 12 ms) and T2-weighted (TR 5000 ms, Fig. 1A, B A 52-year-old man with normal L4/5 facet joints. A CT demonstrates normal facet joints on both sides. B Axial T2-weighted turbo spin-echo image [3800/96 (TR/TE)]. Cartilage (arrows) is visualized as layer of intermediate signal intensity between the articular cortices of the superior and inferior articular processes. The clinical significance of the presence of joint fluid (slightly more intense than cartilage) is unknown [10] Fig. 2A, B A 48-year-old man with grade 1 osteoarthritis of the L5/ S1 facet joints. A Both articular joint spaces are narrowed on the CT. The facets are slightly hypertrophied on both sides. B The corresponding axial T2-weighted MR image shows similar findings TE 112 ms) sagittal turbo-spin-echo images. Image matrix was , field of view 240 mm, slice thickness 4 mm, interslice gap 0.5 mm, and the echo-train length 3 for the T1-weighted sequence and 7 for the T2-weighted sequence. In addition, angled axial T2-weighted turbo-spin-echo images of all abnormal intervertebral spaces were obtained (TR 3800 ms, TE 96 ms, image matrix , field of view 180 mm, slice thickness 4 mm, interslice gap 0.5 mm, echo train length 7). CT was performed on four different types of scanners at six different institutions with 120±130 kv and 420±600 mas. Lateral scout images were obtained in all patients for alignment of the sections to the respective disk level. Slice thickness was 3±5 mm, and overlapping or contiguous angled axial sections were acquired. Images were printed with bone windowing (level, 450 HU; width, 1800 HU in most cases). Only those disk levels for which both CT and MR images were available were included in the investigation. Image evaluation Two experienced musculoskeletal radiologists reviewed all MR images independently. They were masked to any clinical and prior imaging data. All imaging planes of the MR examination were available for readout. The readers were asked to grade the facet joints at each of the included disk levels. Four grades of osteoarthritis of the facet joints were defined using criteria similar to those published by Pathria et al. [9]: grade 0, normal; grade 1, mild degenerative disease; grade 2, moderate degenerative disease; and grade 3, severe degenerative disease (Figs. 1±4). The criteria for grading are given in Table 1. According to Haughton [6] the normal joint space width 1A 1B 2A 2B

3 217 3A 3B 4A 4B Fig. 3A, B A 46-year-old woman with grade 1 and grade 2 osteoarthritis of the L4/5 facet joints. A CT scan demonstrates grade 1 osteoarthritis of the right facet joint with a small osteophyte (arrow). On the left side grade 2 osteoarthritis with narrowing of the facet joint and moderate osteophytes is present (arrowheads). B Axial T2-weighted MR image also reveals a small osteophyte on the right side (arrow). A moderate osteophyte (arrowheads) is visible on the left. Cartilage is replaced by fluid Fig. 4A, B A 50-year-old woman with grade 3 osteoarthritis of the L5/S1 facet joints. A Severe degeneration of the facet joint with joint space narrowing, hypertrophy of the articular processes, large osteophytes (arrows), and subarticular bone erosions (arrowheads) is demonstrated on the CT scan. B Axial T2-weighted MR image also reveals severe degeneration of both facet joints with joint space narrowing, hypertrophy of the articular processes, large osteophytes (arrows), and subarticular bone erosions (arrowheads) Table 1 Criteria for grading osteoarthritis of the facet joints (adapted from [9]) Grade Criteria 0 Normal facet joint space (2±4 mm width) 1 Narrowing of the facet joint space (< 2 mm) and/or small osteophytes and/or mild hypertrophy of the articular process 2 Narrowing of the facet joint space and/or moderate osteophytes and/or moderate hypertrophy of the articular process and/or mild subarticular bone erosions 3 Narrowing of the facet joint space and/or large osteophytes and/or severe hypertrophy of the articular process and/or severe subarticular bone erosions and/or subchondral cysts was defined as 2±4 mm, which represents the thickness of the two cartilaginous layers. A total of 308 facet joints were analyzed: 8 facet joints at the L1/2 level, 24 at the L2/3 level, 90 at the L3/4 level, 96 at the L4/5 level, and 90 at the L5/S1 level. Two weeks later, the same radiologists independently evaluated the facet joints on CT scans using the same grading. Percentage agreement and weighted kappa statistics [11] were used to describe interobserver agreement for both imaging modalities and agreement between MR imaging and CT for both observers. To assess intraobserver variability, an additional interpretation of the CT and MR images was performed by the same two radiologists 8 months after the initial readout. Results On CT images reader 1/reader 2 graded 56/98 facet joints (18%/32%) as grade 0, 143/136 facet joints (46%/44%) as grade 1, 64/42 facet joints (21%/14%) as grade 2, and 45/ 32 facet joints (15%/10%) as grade 3. Interobserver agreement for MR imaging The weighted kappa coefficient for interobserver agreement was This coefficient represents moderate agree-

4 218 ment (<0.00=poor, 0.00±0.20=slight, 0.21±0.40=fair, 0.41±0.60=moderate, 0.61±0.80 substantial, 0.81±1.00= almost perfect) [11]. There was perfect agreement between the readers in 158 of 308 facet joints (51%) and agreement within one grade in an additional 139 facet joints (45%). In this subgroup with disagreement of one grade, 89% of the differences were between grade 0 or grade 1 or grade 1 and 2, respectively. In only 11 of the 308 facet joints (4%) there was a difference of two grades. Interobserver agreement for CT The weighted kappa coefficient for interobserver agreement was There was perfect interobserver agreement in the grading of 189 of the 308 facet joints (61%). Agreement within one grade was present in an additional 110 facet joints (36%). In this subgroup with disagreement of one grade, 87% of the differences were between grade 0 or grade 1 or grade 1 and 2, respectively. In 9 of the 308 facets joints (3%) there was a difference of two grades. Agreement between CT and MR imaging The weighted kappa coefficients for agreement between MR and CT grading were 0.49 and 0.61 for readers 1 and 2, respectively. MR grading of facet degeneration was identical to the CT grading in 160 joints (52%) for reader 1 and in 194 joints (63%) for reader 2. In an additional 133/107 facet joints (43%/35%), the grading differed by one grade between readers 1 and 2. In this subgroup with disagreement of one grade 83%, and 93% of the differences were between grade 0 or grade 1 or grade 1 and 2, respectively. In only 15/7 joints (5%/2%) did the MR and CT grading differ by two grades. Agreement between CT and MR imaging in the assessment of osteoarthritis within one grade was 95% and 97% for readers 1 and 2, respectively. Intraobserver agreement for CT and MR imaging Intraobserver agreement was higher than interobserver agreement. The weighted kappa coefficients for intraobserver agreement for CT and MR imaging were substantial for both readers (0.70/0.70 for reader 1; 0.77/0.76 for reader 2). Discussion Osteoarthritis of the facet joints is common. It may be related to repeated stress or trauma and spinal deformity with secondary overload [12]. The facet joints are true synovial joints with hyaline articular cartilage, a synovial membrane and a joint capsule. Facet joint osteoarthritis does not differ from degenerative changes found in other synovial articulations [5]. It is characterized by fibrillation and later fissuring and ulceration of articular cartilage which progress from the superficial to the deep cartilage layers. There is commonly a proliferative response with formation of osteophytes and sclerosis of subchondral bone. In addition, subchondral cysts and synovial inflammation may be present. Abnormal stress across the facet joint can cause hypertrophy of the articular process which is not identical to osteophyte formation [7]. The fact that pain can originate from the facet joints is widely accepted in the radiologic and orthopedic literature [2, 3, 7, 8, 13±15]. This is supported by investigations employing successful intra-articular or periarticular joint blocks [2, 15]. The cardinal role of facet joint abnormalities in patients with low back pain is still debated [13± 15]. Lewinnek et al. [2] found that 96% of patients with CT-documented osteoarthritis of the facet joints responded to the injection technique. This indicates that detection and quantification of osteoarthritis involving the facet joints may aid in selecting patients for facet block therapy. However, there are no outcome studies demonstrating that patient management is positively affected by demonstration of facet joint osteoarthritis. Schwarzer et al. [16] even questioned the clinical importance of facet joint osteoarthritis. They were not able to demonstrate a significant correlation between the degree of osteoarthritis seen on CT and the pain score during the facet block. Although the clinical relevance of osteoarthritis of the facet joints remains unclear, radiologists are commonly asked by clinicians to determine the degree of the facet joint osteoarthritis. The diagnostic methods used for the evaluation of facet joint degeneration are standard radiographs, CT and MR imaging [6]. Standard radiographs, especially without oblique views, are of limited value, however. The facet joints are in an oblique position and have a curved configuration. Even on oblique views only the portion of each joint that is oriented parallel to the X- ray beam is visible [9]. Therefore, standard radiographs can only be used for screening for facet joint osteoarthritis [9]. In comparison with standard radiographs, CT improves delineation of the facet joints due to its capability to image the joint in the axial plane and the high contrast between bony structures and the surrounding soft tissue [7]. The abnormalities associated with osteoarthritis can be demonstrated and categorized by CT [7]. In a retrospective study of the MR images of the lumbar spines of 13 healthy subjects and 30 patients with degenerative changes Grenier et al. [5] have demonstrated the MR appearances of normal and degenerative posterior spinal structures. Leone et al. [17] have compared CT and MR findings of lumbar facet joint degeneration in nine autopsy specimens and correlated their findings with the

5 219 histopathologic findings. CT was superior to MR imaging in the depiction of joint space narrowing and subchondral sclerosis. The authors concluded that CT represents the examination of choice in the diagnosis of facet joint degeneration. To our knowledge, the agreement between MR imaging and CT in the assessment of facet joint osteoarthritis has not been investigated in a larger series of patients with low back pain. In our investigation, interobserver agreement between CT and MR imaging with regard to facet joint osteoarthritis was only moderate to good. However, the majority of disagreements were only by one grade, and these disagreements were most commonly related to mild degenerative disease. Disagreement with regard to advanced osteoarthritis was rare. This indicates that the diagnosis of facet joint osteoarthritis will be similar but not identical for CT and MR imaging. On the basis of these results, there is probably no need to add a CT scan to an existing MR examination for the assessment of the facet joints. This is supported by the fact that a significant proportion of the disagreements may be due not to the imaging methods themselves but to the difficulties the readers had in applying the grading system consistently. Interobserver agreement for grading facet joint osteoarthritis was moderate for CT and MR imaging (0.6 and 0.41, respectively) but intraobserver agreement was notably higher for both imaging methods (kappa 0.70±0.77). In our investigation CT appeared to be more reproducible in grading osteoarthritis than in a previous study conducted by Coste et al. [18]. However, we confirm that there is substantial interobserver variability for both CT and MR imaging. Although we did not investigate the role of the MR imaging plane in grading facet joint osteoarthritis, we have observed that both readers primarily used the axial plane for this purpose. Sagittal images were used only for confirmation of equivocal findings. In conclusion, there is moderate to good agreement between MR imaging and CT in the assessment of lumbar facet joint osteoarthritis. When differences of one grade are disregarded agreement is excellent. Therefore, in the presence of an MR examination CT is not required for the assessment of facet joint degeneration. References 1. Boden SD. The use of radiographic imaging studies in the evaluation of patients who have degenerative disorders of the lumbar spine. J Bone Joint Surg Am 1996; 78:114± Lewinnek GE, Warfield CA. Facet joint degeneration as a cause of low back pain. Clin Orthop 1986; 213:216± Helbig T, Lee CK. The lumbar facet syndrome. Spine 1988; 13:61± Resnick D, Niwayama G. Degenerative diseases of the spine. In: Resnick D, ed. Diagnosis of bone and joint disorders. Philadelphia: Saunders, 1995:1396± Grenier N, Kressel HY, Schiebler ML, Grossman RI, Dalinka MK. Normal and degenerative posterior spinal structures: MR imaging. Radiology 1987; 165:517± Haughton V. Imaging techniques in intraspinal diseases. In: Resnick D, ed. Diagnosis of bone and joint disorders. Philadelphia: Saunders, 1995: 237± Carrera GF, Haughton VM, Syvertsen A, Williams AL. Computed tomography of the facet joints. Radiology 1980; 134:145± Raskin SP. Degenerative changes of the lumbar spine: assessment by computed tomography. Orthopedics 1981; 4:186± Pathria M, Sartoris DJ, Resnick D. Osteoarthritis of the facet joints: accuracy of oblique radiographic assessment. Radiology 1987; 164:227± Boos N, Lander PH. Clinical efficacy of imaging modalities in the diagnosis of low-back pain disorders. Eur Spine J 1996; 5:2± Landis RJ, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33:159± Taylor JR, Twomey LT. Age changes on lumbar zygapophyseal joints: observations on structure and function. Spine 1986; 11:739± Badgley CE. The articular facets in relation to low-back pain and sciatic radiation. J Bone Joint Surg 1941; 23:481± Nachemson AL. Newest knowledge of low-back pain: a critical look. Clin Orthop 1992; 179:8± Schwarzer AC, Aprill C, Derby R, Fortin J, Kine G, Bogduk N. Clinical features of patients with pain stemming from the lumbar zygapophyseal joints. Is the lumbar facet syndrome a clinical entity? Spine 1994; 10:1132± Schwarzer AC, Wang SC, O'Driscoll D, Harrington T, Bogduk N, Laurent R. The ability of computed tomography to identify a painful zygoapophyseal joint in patients with chronic low back pain. Spine 1995; 20:907± Leone A, Aulisa L, Tamburrelli F, Lupparelli S, Tartaglione T. Ruolo della tomografia computerizzata e della risonanza magnetica nella valutazione della artropatia degenerativa delle facette articolari lombari. Radiol Med 1994:88:547± Coste J, Judet O, Barre O, Siaud JR, Cohen de Lara A, Paolaggi JP. Interand intraobserver variability in the interpretation of computed tomography of the lumbar spine. J Clin Epidemiol 1994; 47:375±381.