MR imaging and CT in osteoarthritis of the lumbar

Skeletal Radiol (1999) 28:215±219
International Skeletal Society 1999
Dominik Weishaupt
Marco Zanetti
Norbert Boos
Juerg Hodler
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
ARTICLE
MR imaging and CT in osteoarthritis
of the lumbar facet joints
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
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 dem-
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
onstration 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
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 T1weighted (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
256180, 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
256210, 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
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)
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.
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
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 0.41. This coefficient represents moderate agree-
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 0.60. 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 sy-
novial 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 Xray 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
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 im-
aging 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±124.
2. Lewinnek GE, Warfield CA. Facet joint
degeneration as a cause of low back
pain. Clin Orthop 1986; 213:216±222.
3. Helbig T, Lee CK. The lumbar facet
syndrome. Spine 1988; 13:61±64.
4. Resnick D, Niwayama G. Degenerative
diseases of the spine. In: Resnick D, ed.
Diagnosis of bone and joint disorders.
Philadelphia: Saunders, 1995:1396±
1462.
5. Grenier N, Kressel HY, Schiebler ML,
Grossman RI, Dalinka MK. Normal and
degenerative posterior spinal structures:
MR imaging. Radiology 1987; 165:517±
525.
6. Haughton V. Imaging techniques in intraspinal diseases. In: Resnick D, ed.
Diagnosis of bone and joint disorders.
Philadelphia: Saunders, 1995: 237±276.
7. Carrera GF, Haughton VM, Syvertsen
A, Williams AL. Computed tomography
of the facet joints. Radiology 1980;
134:145±148.
8. Raskin SP. Degenerative changes of the
lumbar spine: assessment by computed
tomography. Orthopedics 1981; 4:186±
195.
9. Pathria M, Sartoris DJ, Resnick D. Osteoarthritis of the facet joints: accuracy
of oblique radiographic assessment.
Radiology 1987; 164:227±230.
10. Boos N, Lander PH. Clinical efficacy of
imaging modalities in the diagnosis of
low-back pain disorders. Eur Spine J
1996; 5:2±22.
11. Landis RJ, Koch GG. The measurement
of observer agreement for categorical
data. Biometrics 1977; 33:159±174
12. Taylor JR, Twomey LT. Age changes on
lumbar zygapophyseal joints: observations on structure and function. Spine
1986; 11:739±745.
13. Badgley CE. The articular facets in relation to low-back pain and sciatic radiation. J Bone Joint Surg 1941;
23:481±496.
14. Nachemson AL. Newest knowledge of
low-back pain: a critical look. Clin
Orthop 1992; 179:8±20.
15. 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±1137.
16. 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±912.
17. 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±552.
18. 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.

Download Report