Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print October 26, 2005, 10.1148/radiol.2373041989
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2373041989v1
237/3/998    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hayes, C. W.
Right arrow Articles by Sowers, M. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hayes, C. W.
Right arrow Articles by Sowers, M. R.
(Radiology 2005;237:998-1007.)
© RSNA, 2005

Musculoskeletal Imaging

Osteoarthritis of the Knee: Comparison of MR Imaging Findings with Radiographic Severity Measurements and Pain in Middle-aged Women1

Curtis W. Hayes, MD, David A. Jamadar, MB, BS, Gavin W. Welch, PhD, Mary L. Jannausch, MS, Laurie L. Lachance, PhD, Diana C. Capul, BA and MaryFran R. Sowers, PhD

1 From the Department of Epidemiology, School of Public Health, University of Michigan, Room 2624, School of Public Health I, 109 Observatory St, Ann Arbor, MI 48109 (G.W.W., M.L.J., L.L.L., D.C.C., M.R.S.); Department of Radiology, University of Michigan Health Systems, Ann Arbor (D.A.J.); and Department of Radiology, Virginia Commonwealth University, Medical College of Virginia, Richmond, Va (C.W.H.). Received November 23, 2004; revision requested January 11, 2005; revision received February 1; accepted February 25. M.R.S. supported by grants S1092 from Association of Schools of Public Health, RO1-40888 from National Institute of Arthritis and Musculoskeletal and Skin Diseases, U01-04041 from National Institute of Nursing Research, and 17104 from AG. Address correspondence to M.R.S. (e-mail: mfsowers{at}umich.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
PURPOSE: To prospectively compare magnetic resonance (MR) imaging–defined abnormalities of osteoarthritis (OA) of the knee with radiographic severity measurements of OA of the knee and self-reported pain.

MATERIALS AND METHODS: This study was approved by the institutional review board of University of Michigan. Informed consent was obtained for this HIPAA-compliant study. Knee MR imaging was performed in 117 women (mean age, 46 years; range, 32–56 years) from a community-based arthritis study (n = 1053) with 30 women in each of four categories: (a) no pain and no OA of the knee, (b) no pain and OA of the knee, (c) pain and no OA of the knee, and (d) pain and OA of the knee. OA of the knee was defined from radiographs. Two hundred thirty-two eligible knees had Kellgren-Lawrence scores for OA of the knee as follows: grade 0, 115 (49.6%); grade 1, 33 (14.2%); grade 2, 66 (28.4%); grade 3, 17 (7.3%); and grade 4, one (0.4%). MR images were assessed for location and severity of defects of cartilage, bone marrow edema (BME), osteophytes, subchondral cysts, sclerosis, meniscal and/or ligamentous tears, joint effusion, synovial cysts, and synovitis. MR imaging findings were compared with radiographic severity of OA of the knee (Kellgren-Lawrence scale) and self-reported pain with analysis of variance, t tests, and contingency table analyses.

RESULTS: Defects of cartilage (higher than grade IIA) were found in 75% of knees; BME was found in 57% of knees (<1 cm, 41%; >1 cm, 16%). Large BME lesions were common in the pain and OA of the knee group (P = .001); this group was significantly more likely to have defects of cartilage (P = .001); meniscal tears (P = .001); and osteophytes, subchondral cysts, sclerosis, joint effusion, and synovitis (P < .001). Defects of cartilage, osteophytes, sclerosis, meniscal or ligamentous tears, joint effusion, and synovitis were strongly related to increasing Kellgren-Lawrence grade (P < .001).

CONCLUSION: In middle-aged women, there were significant associations between pain, radiographic severity of OA of the knee, and seven MR imaging–identified parameters.

© RSNA, 2005


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Osteoarthritis (OA) is a highly prevalent disease, and the prevalence is expected to increase substantially as a greater proportion of the population exceeds 60 years of age (1,2). It is estimated that there are in excess of 17 million cases of OA in the United States, and OA-related complaints are the most common reasons for visits to health care providers (1,2).

Historically, the primary modality for imaging evaluation of OA has been radiography, although the limitations of radiographic parameters for OA evaluation are well-documented (3–6) At best, radiographically determined joint space narrowing is an indirect measure of articular cartilage status, because radiographic evaluation includes only a small percentage of the total articular surface. Because radiographs are a two-dimensional composite of complex three-dimensional structures, the sensitivity for observation of features such as osteophytes, bone eburnation, and other subchondral bone abnormalities on them is limited. Lastly, radiographically observed abnormalities in OA often correlate imperfectly with clinical symptoms (3,6).

Recent focus on the development of disease-modifying therapeutic agents for OA has emphasized the need for imaging techniques capable of depicting relevant early abnormalities of OA over relatively short times. Simultaneously, there is increasing acknowledgment that OA, particularly in the knee, can be regarded as a whole-organ degenerative process (7), with an emphasis on the contribution of multiple articular and periarticular abnormalities in the clinical expression of the disease. The cross-sectional image display, spatial resolution, and tissue contrast of magnetic resonance (MR) imaging provide a unique capability to display bone, cartilage, and soft-tissue abnormalities that is necessary for whole-organ assessment of the knee. Early reports of MR imaging related to OA of the knee primarily were about small cross-sectional studies that were focused on a limited number of MR imaging–defined parameters (8–13). Investigators in several larger cross-sectional or longitudinal studies evaluated multiple parameters, with encouraging results (7,14–20), although the populations typically consisted of elderly subjects, with relatively advanced OA of the knee. Further validation of MR imaging–defined parameters for OA of the knee is necessary if MR imaging is to replace radiography in clinical trials.

The Southeast Michigan Osteoarthritis cohort is an ongoing study that includes a population of more than 1053 women who are being followed up in regard to multiple clinical, biochemical, and imaging characteristics related to osteoarthritis, bone mineralization, and overall health status (21). Within this population, there is substantial variation in the manifestation of OA of the knee that ranges from none to severe. Thus, the purpose of our study was to prospectively compare MR imaging–defined parameters for OA of the knee with radiographic measurements of severity of OA of the knee and self-reported pain.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Subjects
Study subjects were from the Southeast Michigan Osteoarthritis cohort, which includes 543 women from the Study of Women's Health Across the Nation and 510 women from the Michigan Bone Health Study (21). The Southeast Michigan Osteoarthritis cohort women were classified into four groups on the basis of the presence or absence of radiographically determined OA of the knee and self-reported pain in one or both knees. OA of the knee was characterized by a Kellgren-Lawrence score of higher than 2 from anteroposterior weight-bearing knee radiographs of both knees obtained concurrently with the interview for evaluation of pain. This sample has been described previously (22). The frequency of knee pain and radiographically defined OA of the knee among the entire Southeast Michigan Osteoarthritis cohort is shown in Table 1. Because the Southeast Michigan Osteoarthritis cohort is a population-based study cohort of middle-aged women rather than a convenience sample from a clinical practice, the population had less osteoarthritis than did clinical groups of the elderly, and only 10% of the population were classified in the group with pain and OA of the knee. One-third of the Southeast Michigan Osteoarthritis cohort population with OA had self-reported knee pain without radiographically determined OA of the knee, a finding that supported the emphasis on the well-known discrepancy between the occurrence of pain and the presence of radiographically determined abnormalities in OA of the knee.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Absolute Frequency of Knee Pain and OA in Four Subgroups of Women from the Southeast Michigan Osteoarthritis Cohort

 
From the four subgroups of the Southeast Michigan Osteoarthritis cohort, 30 women per group were randomly selected for MR imaging evaluation from each of the four groups in regard to pain and OA of the knee (group 1, no pain plus no radiographically determined OA of the knee; group 2, no pain plus radiographically determined OA of the knee; group 3, pain plus no radiographically determined OA; and group 4, pain plus radiographically determined OA). Both knees of each woman within each group were evaluated. Eight knees with prior severe trauma (both knees in three women and one knee in two women) were excluded at the time of data analysis, and this exclusion resulted in the inclusion of 117 women and 232 knees in the final analysis. In these 232 knees, the radiographically determined Kellgren-Lawrence scores were as follows: grade 0 (normal), 115 knees (49.6%); grade 1 (doubtful and minimal OA of the knee), 33 knees (14.2%); grade 2 (mild OA of the knee), 66 knees (28.4%); grade 3 (moderate OA of the knee), 17 knees (7.3%); and grade 4 (severe OA of the knee), one knee (0.4%).

Additional characteristics of the 117 women in the study are described in Table 2. The mean age of subjects was 46.2 years (range, 32–56 years), and the mean body mass index was 30.3 kg/m2, as shown in Table 2. Body mass index was significantly greater in both groups with OA of the knee (P = .005).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Characteristics of Subgroups of Southeast Michigan Osteoarthritis Cohort Defined according to Radiographically Determined OA of the Knee and Self-reported Pain

 
This study, which was compliant with the Health Insurance Portability and Accountability Act, was approved by the institutional review board of the University of Michigan, and participants provided informed consent.

Radiographic Assessment
Anteroposterior radiographs of the knee were obtained in a weight-bearing extended position by using a standard radiographic technique. One experienced musculoskeletal radiologist (D.A.J.) and one experienced rheumatologist (each with at least 15 years of experience) independently assigned scores to all radiographs by using the Kellgren-Lawrence scoring system shown in the Atlas of Standard Radiographs of Arthritis (23) and agreed on a score in consensus if there was disagreement in score assignment. This summary Kellgren-Lawrence score was based on osteophyte formation, joint space narrowing, sclerosis, and joint deformity characteristics according to the five-level scale defined as follows: grade 0, normal; grade 1, doubtful osteoarthritis; grade 2, minimal osteoarthritis; grade 3, moderate osteoarthritis; or grade 4, severe osteoarthritis. The anteroposterior views characterize OA of the knee in the medial and lateral femorotibial compartments, with exclusion of the patellofemoral compartment.

MR Imaging Assessment
Approximately 1 year after radiography, MR imaging of each knee was performed separately by using a 1.5-T MR imager (Signa; GE Medical Systems, Milwaukee, Wis) equipped with a commercial knee surface coil. Imaging was performed with sagittal, coronal, and transverse fast spin-echo intermediate-weighted (IW) sequences with fat saturation (repetition time msec/echo time msec, 4000/15; section thickness, 4 mm), sagittal IW spin-echo sequences (1000/14; section thickness, 3 mm), and sagittal three-dimensional spoiled gradient-recalled acquisition in the steady state sequences with fat saturation (38/6.9; flip angle, 45°; effective section thickness, 2 mm). The field of view was optimized from 12 to 14 cm. IW fast spin-echo sequences with fat saturation and T2-weighted sequences with fat saturation have been validated by previous investigators for inherent contrast between articular cartilage, bone, and fluid and high signal-to-noise ratio for evaluation of periarticular soft tissues (24–27). Images obtained with spoiled gradient-recalled acquisition in the steady state sequences with fat saturation were included for additional assessment of articular cartilage (28).

MR images were interpreted by two musculoskeletal radiologists (C.W.H., with at least 15 years of experience in subspecialized musculoskeletal imaging, and D.A.J., with at least 5 years of experience in the same type of imaging) who were blinded to the radiographic assessment findings. At an initial calibration session (n = 40 knees), interreader variability ({kappa} = 0.85) was determined (29). After this calibration session, all interpretations were conducted by both radiologists, in consensus, and they were blinded as to the group in regard to pain and OA of the knee.

Each knee was assessed globally and according to compartment for 13 MR imaging–defined parameters: defects of cartilage, subchondral bone marrow edema (BME), osteophytes, subchondral cysts, sclerosis, meniscal abnormalities, ligamentous (anterior cruciate ligament, posterior cruciate ligament, medial collateral ligament, and lateral collateral ligament) abnormalities, joint effusions, synovial cysts, and synovitis.

Defects of cartilage.—Location, severity, and approximate size of defects of cartilage in three compartments (medial femorotibial, lateral femorotibial, patellofemoral compartments) and seven specific surfaces (medial tibial, medial femoral, lateral tibial, and lateral femoral surfaces; trochlea; and medial and lateral patellar facets) were assessed. Severity scoring was based on the Noyes arthroscopic system, which was modified for MR imaging (30). Cartilage was assigned grades as follows: grade 0, normal; grade I, internal signal intensity alteration only; grade IIA, defect of cartilage of less than 50%; grade IIB, defect of cartilage of 50%–99%; grade IIIA, 100% defect of cartilage with no bone ulceration; or grade IIIB, 100% defect of cartilage with subjacent bone ulceration.

Subchondral trabecular BME.—BME was defined as noncircumscribed areas of abnormally high signal intensity on IW fast spin-echo images obtained with fat saturation in a subchondral location and verified in at least two imaging planes (27,31). BME lesions were recorded according to site and severity in the following manner: grade 0, normal; grade 1, largest diameter of less than 10 mm; or grade 2, largest diameter of greater than 10 mm.

Marginal osteophytes.—Osteophytes were defined as any abnormal bone growth that arose from the margin of the involved compartment. Tibial spine growths were considered as osteophytes in their respective compartments only if there were definite excrescences, as opposed to mere "pointedness" of the tibial spines. Osteophytes in each compartment were assigned grades as follows: grade 0, no osteophytes; grade 1, osteophytes of less than 5 mm; or grade 2, osteophytes of greater than 5 mm (32).

Subchondral cysts.—Subchondral cysts were defined as marginated circular or oval areas of hyperintensity (similar to that of fluid) on IW fast spin-echo images obtained with fat saturation in a subchondral location (33). Cysts were assigned grades according to compartment as follows: grade 0, normal; grade 1, less than 10 mm in diameter; or grade 2, greater than 10 mm in diameter.

Subchondral sclerosis.—Definite focal thickening of the subchondral bone, consisting of low signal intensity on both IW spin-echo images and IW fast spin-echo images obtained with fat saturation, was classified as sclerosis (11). Sclerosis was assigned grades according to each compartment as follows: grade 0, normal; grade 1, sclerosis with depth extending less than 5 mm; or grade 2, sclerosis with depth extending more than 5 mm.

Joint effusion.—Joint effusion was defined as the presence of greater than expected physiologic amounts of synovial fluid in the lateral or medial patellar recesses that exceeded 10 mm in width (34). Joint effusion was assigned grades as follows: grade 0, normal; grade 1, small effusion; or grade 2, moderate or large effusion.

Synovial cysts.—Synovial cysts were fluid-filled collections that arose from the joint, extended between the semimembranosis and semitendinosis tendons and the medial head of the gastrocnemius muscle, or were in a less typical site, if confirmed to be in continuity with the joint (34). Meniscal cysts, ganglionic cysts, and other nonspecific periarticular cysts were not included in the analysis. Synovial cysts were assigned grades according to a semiquantitative scale as follows: grade 0, normal; grade 1, small; or grade 2, large.

Synovitis.—Synovitis was defined as increased linear striations within the infrapatellar (Hoffa) fat pad and irregular thickening at the margin of the fat pad with the articular cartilage (35,36). These abnormalities consisted of low signal intensity on IW spin-echo images and intermediate to high signal intensity on IW fast spin-echo images obtained with fat saturation. Synovitis was assigned grades thus: grade 0, normal; grade 1, mild; or grade 2, moderate to marked.

Meniscal abnormalities.—By using accepted MR imaging criteria described in a study of Crues et al (37), intrameniscal abnormalities and frank tears were assigned grades that were recorded as follows: grade 0, normal or globular intrasubstance abnormalities (grade 1 of Crues et al); grade 1, linear intrasubstance meniscal abnormality (grade 2 of Crues et al); grade 2, nondisplaced tear (grade 3 of Crues et al); or grade 3, displaced, complex, or macerated tears.

Ligamentous abnormalities.—Cruciate ligamentous abnormalities were assigned grades as follows: grade 0, normal; grade 1, increased edema within or surrounding the ligament, with a normal course, and at least some intact fibers present; or grade 2, complete tear, acute or chronic. Collateral ligamentous abnormalities were assigned grades as follows: grade 0, normal ligament; grade 1, substantial periligamentous edema, with the ligament intact; grade 2, partial tear; or grade 3, complete tear.

Pain Assessment
Pain was assessed at the 1998 annual examination concurrently with evaluation of radiographs, which were used to define group assignment. For this assessment, the question, which was derived from the Tecumseh Community Health Study, inquired if there was pain in either of the knee joints and if it lasted for more than half the month (38).

Other Measures
Age was assessed from the self-reported date of birth. Body mass index was calculated as weight divided by height squared, where weight was measured in kilograms by using a balance-beam scale and height was measured in meters by using a stadiometer.

Data and Statistical Analysis
For Table 2, median Kellgren-Lawrence scores and the interquartile ranges were estimated for each group for OA of the knee and pain. The associations between the group for OA of the knee and pain and the continuous variables, age and body mass index, were estimated from a general linear model analysis of variance after log transformation of body mass index; t tests from this procedure were used to assess pairwise age or body size differences in groups.

For the reporting of abnormalities at MR imaging according to compartment and surface (Tables 3–5), simple counts (numbers and percentages) were estimated. In the analysis of MR imaging findings versus the group assignment for OA of the knee and pain (Tables 6, 7), the signal knee (the most painful knee at inclusion in the study) for each subject was analyzed (n = 117). For subjects in whom both knees equally fit the categoric criteria, the self-described dominant knee was chosen as the signal knee. For the comparison between MR imaging findings and Kellgren-Lawrence score (Tables 8, 9), each imaged knee was analyzed (n = 232). Contingency tables were used to relate the frequency of ordered categoric MR imaging findings to group assignment for OA of the knee and pain, Kellgren-Lawrence scores, or compartments of the knee. Differences in frequencies were assessed with {chi}2 tests of homogeneity by using the Mantel-Haenszel test of general association or the Fisher exact test of general association in Tables 6–9. For all statistical tests, a statistically significant difference was demonstrated with P values less than .05 and 95% confidence intervals. Analyses were undertaken by using software (SAS, version 8.0, 2001; SAS Institute, Cary, NC) that included the following procedures: Univariate, Frequency, Npar1way, and General Linear Models.


View this table:
[in this window]
[in a new window]

  		TABLE 3. MR Imaging–defined Defects of Cartilage and BME in 232 Knees according to Compartment and Surface

 

View this table:
[in this window]
[in a new window]

  		TABLE 4. Frequency of MR Imaging–related Findings, Osteophytes, Subchondral Cysts, Sclerosis, and Meniscal Tears according to Surface and Compartment in 232 Knees

 

View this table:
[in this window]
[in a new window]

  		TABLE 5. Frequency of Noncompartment-Specific MR Imaging Findings in 232 Knees

 

View this table:
[in this window]
[in a new window]

  		TABLE 6. Frequency of MR Imaging–defined Defects of Cartilage and BME according to Subgroups of OA of the Knee and Pain Identified at Radiography and Interview for Evaluation of Pain

 

View this table:
[in this window]
[in a new window]

  		TABLE 7. MR Imaging–defined Abnormalities of the Knee according to Subgroups of OA of the Knee and Pain Identified at Radiography and Interview for Evaluation of Pain

 

View this table:
[in this window]
[in a new window]

  		TABLE 8. Frequency of Defects of Cartilage and BME according to Radiographically Determined Severity of OA Defined with Kellgren-Lawrence Score

 

View this table:
[in this window]
[in a new window]

  		TABLE 9. Frequency of MR Imaging–defined Abnormalities versus Radiographically Determined Severity of OA Defined with Kellgren-Lawrence Score

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
MR Imaging Findings
Seventy-five percent of knees showed at least one cartilage defect, and the defects were classified as grade IIA or higher (Tables 3–5). The medial, lateral, and patellofemoral compartments showed grade IIA or higher defects of cartilage in 30%, 12%, and 66% of knees, respectively. The medial, lateral, and patellofemoral compartments showed full-thickness defects of cartilage in 5.6%, 3.5%, and 30%, respectively. Full-thickness defects of cartilage with cortical disruption (grade IIIB) were more common in the patellofemoral compartment (12.5% of knees). Of all articular surfaces, the trochlea was the most common location for grade IIIB lesions (9.9%).

BME lesions were present in 57% of all knees, with 16.4% of knees showing lesions greater than 1 cm in diameter. The distribution of BME in the medial, lateral, and patellofemoral compartments was 13%, 7%, and 47%, respectively. The frequency of large BME lesions of greater than 1 cm was 3%, 2%, and 11.6% of knees in the medial, lateral, and patellofemoral compartments, respectively.

According to compartment, osteophytes were present in the medial, lateral, and patellofemoral compartments in 24%, 19%, and 43% of knees, respectively. Subchondral cysts and subchondral sclerosis were found in 19% and 43% of all knees, respectively, and the majority of subchondral cysts (15%) and sclerosis (36%) were in the patellofemoral compartment.

Meniscal tears were present in 27% of all knees. Meniscal tears, both simple and complex, were equally likely with medial meniscus (18%) and lateral meniscus (15%). Eight percent of knees showed a complex or macerated meniscal tear. Cruciate and collateral ligamentous injuries were uncommon. Tears of the cruciate ligament were identified in just two (1%) knees, and there was only one partial tear of the medial collateral ligament. Joint effusions were detected in 24 (11%) knees. Baker cysts were more frequent than were effusions; in 79 (34%) knees, the Baker cyst was small, and in 12 (5%) knees, the cyst was moderate or large. MR imaging evidence of synovitis was present in 36 (15%) knees, and in three (1%) knees, synovitis was considered marked.

MR Imaging Findings in Subgroups for OA of the Knee and Pain
In the groups with radiographically determined OA of the knee (ie, OA of the knee and pain, OA of the knee and no pain), there was a greater frequency of increasingly severe defects of cartilage (P = .001) and greater size of the BME lesion (P = .001) (Tables 6, 7). The MR imaging parameters for which a stronger statistical association (P < .001) with the subgroups of radiographically determined OA of the knee was demonstrated included osteophytes, subchondral cysts, sclerosis, meniscal tears, joint effusion, and synovitis. There were no statistically significant differences in the frequency of Baker cysts or abnormalities of the cruciate and collateral ligaments among the subgroups for OA of the knee and pain.

MR Imaging Findings in Relation to Kellgren-Lawrence Scores
There was a strong association between the Kellgren-Lawrence score and the likelihood that defects of cartilage were present in the knee (P < .001) (Tables 8, 9). Defects of cartilage that were classified as grade IIB (50%–100%) or higher were found in 35% of knees that were assigned Kellgren-Lawrence grade 0, in 54% of knees that were assigned Kellgren-Lawrence grade 1, in 83% of knees that were assigned Kellgren-Lawrence grade 2, in 88% of knees that were assigned Kellgren-Lawrence grade 3, and in one knee that was assigned Kellgren-Lawrence grade 4. In knees classified as normal at radiography, defects of focal cartilage were common (defects classified as higher than grade IIA, 60% of knees; those classified as IIIA or IIIB, 19% of knees). For BME, there also was a significant association with the radiographically determined Kellgren-Lawrence score (P = .005). Large BME lesions of greater than 1 cm in diameter were increasingly likely with increasing Kellgren-Lawrence grade and were found in 9% of knees classified as Kellgren-Lawrence grade 0, in 6% of knees classified as Kellgren-Lawrence grade 1, in 28% of knees classified as Kellgren-Lawrence grade 2, in 29% of knees classified as Kellgren-Lawrence grade 3, and in one knee classified as Kellgren-Lawrence grade 4.

There was an increased likelihood of MR imaging–detected osteophytes with increasing Kellgren-Lawrence grades (P < .001). Twenty-three (35%) of the knees assigned a Kellgren-Lawrence score of grade 2 at radiography, which requires the presence of at least one osteophyte, had no osteophytes at MR imaging.

The frequency of subchondral cysts was only moderately (P = .02) associated with the Kellgren-Lawrence score, which is in contrast to the strong correlation between subchondral sclerosis, joint effusion, synovitis, meniscal tears, and cruciate ligament abnormalities and the Kellgren-Lawrence score (P < .001). Sixteen percent of knees classified as normal at radiography showed meniscal tears, although none of these tears were complex or macerated.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
Three major findings are noteworthy. First, our results confirm that there are increasingly more frequent, as well as more severe, abnormalities at MR imaging in multiple tissues as the Kellgren-Lawrence grade of OA of the knee at radiography increases. Second, MR imaging–defined abnormalities were more frequent in the patellofemoral compartment than they were in the medial and lateral femorotibial compartments. Third, the group with radiographically determined OA of the knee and pain had more frequent and more severe abnormalities at MR imaging than did the other three subgroups. The group with radiographically determined OA of the knee and no pain appeared to reflect an intermediate stage with respect to the OA process.

Frequent defects of focal cartilage, BME, and other soft-tissue abnormalities and the frequent patellofemoral findings contributed substantially to the burden of abnormalities in our subjects and may be critical in consideration of the clinical expression of early OA of the knee. Defects of cartilage of grade IIA or higher and BME lesions were evident on MR images, even among those knees classified as normal at radiography. The majority of the advanced defects of cartilage and BME lesions were located in the patellofemoral compartment in our population. The disparity in sensitivity between MR imaging and standard radiography for detection of abnormalities, especially in the patellofemoral compartment, may partly explain the observation that clinically symptomatic OA of the knee can long precede radiographic evidence of OA of the knee. Since lateral or skyline views of the knee were not used in the categorization of the radiographic severity of OA in our subjects, radiographically determined patellofemoral involvement was undiagnosed. Our data confirm the high prevalence of patellofemoral compartment disease, however, even among women with relatively minimal OA of the femorotibial compartment (39), and these data indicate that future studies should most certainly include assessment of all three knee compartments.

Our data show that there were more frequent and more severe abnormalities detected at MR imaging as the radiographically determined grades of OA of the knee increased. Increasing Kellgren-Lawrence scores were associated with more frequent and more severe defects of cartilage, BME, osteophytes, subchondral sclerosis, effusion, synovitis, and meniscal tears. This finding reinforces the evolving concept that OA of the knee is a whole-organ disease and that MR imaging is capable of showing the bone and soft-tissue evidence of OA of the knee.

MR imaging–depicted synovitis was strongly correlated with the Kellgren-Lawrence grade. The capability of MR imaging to depict evidence of synovitis with abnormal signal intensity in the Hoffa fat pad was suggested by Schweitzer et al (35) and confirmed by Fernandez-Madrid et al (36). Although OA of the knee is not considered an inflammatory arthritis per se, it clearly may have associated synovitis that may be depicted by using MR imaging. In support of this idea, we previously reported that C-reactive protein, a well-established measurement of chronic inflammation, is elevated among those with OA of the knee (22).

It is not surprising that there was a strong correlation between MR imaging–depicted osteophytes and radiographically determined OA of the knee, since the Kellgren-Lawrence system is driven primarily by radiographically determined osteophyte size. Interestingly, some knees with Kellgren-Lawrence scores of grade 2 were classified as having no osteophytes with MR imaging. This discrepancy suggests that radiography may be more sensitive for the detection of tiny osteophytes than are some MR imaging sequences, particularly those with robust fat saturation. Future protocols may benefit from a sequence more specifically targeted for osteophyte detection.

The study design enabled comparison of MR imaging–defined abnormalities among women with and without radiographically determined OA of the knee and pain. Clearly, the group with radiographically determined OA of the knee and pain consistently had more frequent and more severe MR imaging–defined abnormalities than did the other three subgroups. The group with radiographically determined OA of the knee but no pain, however, appeared to be an intermediate group between that with normal knees and that with OA of the knee with pain, particularly with respect to the frequency and severity of defects of cartilage, BME, effusion, subchondral cysts, Baker cysts, and MR imaging–defined osteophytes. When we compared the frequencies of sclerosis and synovitis in the group with OA of the knee and no pain and in the group with OA of the knee and pain, we found that they were similar. The group with no OA of the knee and pain was much less likely to have sclerosis and BME, and this finding suggested that the origin of the pain among women in this group might involve different processes that are not related to OA.

In our report, there are findings that are parallel with findings of other investigators. Fernandez-Madrid et al (12) published a study of 97 knees in 52 subjects with OA of the knee in which MR imaging findings and radiographic evidence of OA of the knee were compared. They found a significant correlation between the MR imaging parameters of synovial proliferation, meniscal abnormalities, defects of cartilage, effusion, subchondral lesions, and osteophytes and the radiographically determined Kellgren-Lawrence score. Only effusion, however, showed a positive correlation with pain. Link et al (15) assessed knees in 50 subjects and compared MR imaging findings with results of radiographic and clinical assessment for OA of the knee. Of the parameters examined, defects of cartilage, BME, cysts, effusion, and ligamentous abnormalities showed a positive association with radiographic severity of OA of the knee. Although several parameters approached a statistically significant difference with P < .05, only defects of cartilage showed a statistically significant correlation with pain. Felson et al (14,18) and Bhattacharyya et al (40) also described the association between several MR imaging–defined parameters and radiographic severity of OA of the knee, as well as pain, in men and women. These authors reported that they found a statistically significant association between pain and BME lesions, especially larger lesions (14), just as we did.

Meniscal tears were present in about one-quarter of knees examined, a frequency that is lower than that reported in several other studies (40,41) involving older subjects whose occupations and lifestyles may have predisposed them to having a different frequency of meniscal injuries than did those of the women in our sample.

There were few ligamentous abnormalities in our sample. This may be explained, in part, because women with known severe trauma were excluded from data analyses, and unlike the typical clinical cohort that commonly reflects acute or chronic trauma, our study participants more accurately reflect the population at large, in which severe knee injuries may be less common. Indeed, the most important risk factor for OA in our population was obesity.

Synovial or Baker cysts were common in our population and occurred in more than one-third of knees evaluated, but there was no correlation with radiographic severity of OA or pain status. Joint effusion, however, showed a strong correlation with both radiographic severity of OA and pain, a finding that was consistent with the findings of Felson et al (14).

Several additional points and differences between the results of this study and those in prior reports reflect some strengths and limitations of our study. Our participants were younger than those included in the study of Felson et al (14) or in that of Link et al (15), and they had less severe OA of the knee as a group. By design, we included a younger population in the hope of authenticating MR imaging–defined parameters for detection of early OA of the knee. As opposed to older patients with advanced OA of the knee, younger persons with early OA of the knee may benefit most from the development of structure-modifying therapies.

Our study sample was limited to women. Some results, such as the high prevalence of patellofemoral abnormalities, may be sex specific. Therefore, our results should not be extrapolated to men, who are likely to have different body size characteristics and occupational demands.

In the selection of MR imaging sequences for whole-organ assessment, a trade-off exists among signal-to-noise ratio, fluid sensitivity, and efficiency. We believe that IW fast spin-echo MR imaging sequences with fat saturation represent a balance of these characteristics. More heavily fluid-weighted sequences, such as T2-weighted fast spin-echo MR imaging with fat saturation, typically have lower signal-to-noise ratios and lack the versatility we desired. We included a three-dimensional spoiled gradient-recalled acquisition in the steady state sequence with fat saturation for additional assessment of cartilage, although volume measurements were not performed. Quantitative analysis of the volume of cartilage has been validated (42–44), and findings in some studies show that there is promise in the demonstration of change in the volume of cartilage related to OA of the knee, but findings in these reports are inconsistent (45–50). The modified Noyes scale used in assigning grades to defects of focal cartilage does not capture the area of the defect or give weighting on the basis of location. Ultimately, a scoring system that incorporates all these factors may prove useful.

Last, the evaluation of meniscal abnormalities in this study is based on tears and internal degeneration and does not address meniscal extrusions. This parameter and others such as varus and valgus alignment and femorotibial subluxations (51–53) may prove valuable in future studies.

In summary, findings of this study of 232 knees in 117 middle-aged women showed that the MR imaging parameters of defects of cartilage, BME, osteophytes, subchondral sclerosis, meniscal abnormalities, joint effusion, and synovitis have a strong correlation with radiographic severity measurements of osteoarthritis of the knee and pain. Our data establish the relevance of these MR imaging–defined parameters in a population of relatively young women with generally mild OA of the knee. These findings should prove useful in assignment of parameter-specific relative weighting factors for achievement of a sensitive whole-organ assessment scale for severity of OA of the knee.


   FOOTNOTES
 

Abbreviations: BME = bone marrow edema • IW = intermediate weighted • OA = osteoarthritis

Authors stated no financial relationship to disclose.

Author contributions: Guarantors of integrity of entire study, C.W.H., M.R.S.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, C.W.H., M.R.S.; clinical studies, C.W.H., M.R.S.; statistical analysis, M.L.J., M.R.S.; and manuscript editing, C.W.H., M.L.J., D.C.C., M.R.S.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. From the Centers for Disease Control and Prevention. Prevalence and impact of arthritis among women—United States, 1989–1991. JAMA 1995;273:1820–1821.[CrossRef][Medline]
  2. Praemer A, Furner S, Rice DP. Musculoskeletal conditions in the United States. Park Ridge, Ill: American Academy of Orthopedic Surgeons, 1992.
  3. Spector TD, Hart DJ, Huskisson EC. The use of radiographs in assessing the severity of knee osteoarthritis. J Rheumatol Suppl 1991;27:38–39.[Medline]
  4. Altman RD. Criteria for classification of clinical osteoarthritis. J Rheumatol Suppl 1991;27:10–12.[Medline]
  5. Brandt KD, Fife RS, Braunstein EM, Katz B. Radiographic grading of the severity of knee osteoarthritis: relation of the Kellgren and Lawrence grade to a grade based on joint space narrowing, and correlation with arthroscopic evidence of articular cartilage degeneration. Arthritis Rheum 1991;34:1381–1386.[Medline]
  6. Hannan MT, Felson DT, Pincus T. Analysis of the discordance between radiographic changes and knee pain in osteoarthritis of the knee. J Rheumatol 2000;27:1513–1517.[Medline]
  7. Peterfy CG. Imaging of the disease process. Curr Opin Rheumatol 2002;14:590–596.[CrossRef][Medline]
  8. McAlindon TE, Watt I, McCrae F, Goddard P, Dieppe PA. Magnetic resonance imaging in osteoarthritis of the knee: correlation with radiographic and scintigraphic findings. Ann Rheum Dis 1991;50:14–19.[Abstract/Free Full Text]
  9. Chan WP, Lang P, Stevens MP, et al. Osteoarthritis of the knee: comparison of radiography, CT, and MR imaging to assess extent and severity. AJR Am J Roentgenol 1991;157:799–806.[Abstract/Free Full Text]
  10. Broderick LS, Turner DA, Renfrew DL, Schnitzer TJ, Huff JP, Harris C. Severity of articular cartilage abnormality in patients with osteoarthritis: evaluation with fast spin-echo MR vs arthroscopy. AJR Am J Roentgenol 1994;162:99–103.[Abstract/Free Full Text]
  11. Bergman AG, Willen HK, Lindstrand AL, Pettersson HT. Osteoarthritis of the knee: correlation of subchondral MR signal abnormalities with histopathologic and radiographic features. Skeletal Radiol 1994;23:445–448.[Medline]
  12. Fernandez-Madrid F, Karvonen RL, Teitge RA, Miller PR, Negendank WG. MR features of osteoarthritis of the knee. Magn Reson Imaging 1994;12:703–709.[CrossRef][Medline]
  13. Drape JL, Pessis E, Auleley GR, Chevrot A, Dougados M, Ayral X. Quantitative MR imaging evaluation of chondropathy in osteoarthritic knees. Radiology 1998;208:49–55.[Abstract/Free Full Text]
  14. Felson DT, Chaisson CE, Hill CL, et al. The association of bone marrow lesions with pain in knee osteoarthritis. Ann Intern Med 2001;134:541–549.[Abstract/Free Full Text]
  15. Link TM, Steinbach LS, Ghosh S, et al. Osteoarthritis: MR imaging findings in different stages of disease and correlation with clinical findings. Radiology 2003;226:373–381.[Abstract/Free Full Text]
  16. Boegard TL, Rudling O, Petersson IF, Jonsson K. Magnetic resonance imaging of the knee in chronic knee pain: a 2-year follow-up. Osteoarthritis Cartilage 2001;9:473–480.[CrossRef][Medline]
  17. Pessis E, Drape JL, Ravaud P, Chevrot A, Dougados M, Ayral X. Assessment of progression in knee osteoarthritis: results of a 1 year study comparing arthroscopy and MRI. Osteoarthritis Cartilage 2003;11:361–369.[CrossRef][Medline]
  18. Felson DT, McLaughlin S, Goggins J, et al. Bone marrow edema and its relation to progression of knee osteoarthritis. Ann Intern Med 2003;139:330–336.[Abstract/Free Full Text]
  19. Peterfy CG, Guermazi A, Zaim S, et al. Whole-Organ Magn Reson Imaging Score (WORMS) of the knee in osteoarthritis. Osteoarthritis Cartilage 2004;12:177–190.[CrossRef][Medline]
  20. Kornaat PR, Ceulemans RY, Kroon HM, et al. MRI assessment of knee osteoarthritis: Knee Osteoarthritis Scoring System (KOSS)—inter-observer and intra-observer reproducibility of a compartment-based scoring system. Skeletal Radiol 2005;34:95–102.[CrossRef][Medline]
  21. Sowers M, Lachance L, Hochberg M, Jamadar D. Prevalence of radiographically defined osteoarthritis of the hand and knee in a population of pre- and perimenopausal women. Osteoarthritis Cartilage 2000;8:69–77.[CrossRef][Medline]
  22. Sowers MF, Hayes C, Jamadar D, et al. Magnetic resonance-detected subchondral bone marrow and cartilage defect characteristics associated with pain and x-ray-defined knee osteoarthritis. Osteoarthritis Cartilage 2003;11:387–393.[CrossRef][Medline]
  23. Kellgren JH, Lawrence JS. Epidemiology of chronic rheumatism. Philadelphia, Pa: Davis, 1963.
  24. Harned EM, Mitchell DG, Burk DL, Vinitski S, Rifkin MD. Bone marrow findings on magnetic resonance images of the knee: accentuation by fat suppression. Magn Reson Imaging 1990;8:27–31.[CrossRef][Medline]
  25. Potter HG, Linklater JM, Allen AA, Hannafin JA, Haas SB. Magnetic resonance imaging of articular cartilage in the knee: an evaluation with use of fast-spin-echo imaging. J Bone Joint Surg Am 1998;80:1276–1284.[Abstract/Free Full Text]
  26. Bredella MA, Tirman PF, Peterfy CG, et al. Accuracy of T2-weighted fast spin-echo MR imaging with fat saturation in detecting cartilage defects in the knee: comparison with arthroscopy in 130 patients. AJR Am J Roentgenol 1999;172:1073–1080.[Abstract/Free Full Text]
  27. Lal NR, Jamadar DA, Doi K, et al. Evaluation of bone contusions with fat-saturated fast spin-echo proton-density magnetic resonance imaging. Can Assoc Radiol J 2000;51:182–185.[Medline]
  28. Recht MP, Piraino DW, Paletta GA, Schils JP, Belhobek GH. Accuracy of fat-suppressed three-dimensional spoiled gradient-echo FLASH MR imaging in the detection of patellofemoral articular cartilage abnormalities. Radiology 1996;198:209–212.[Abstract/Free Full Text]
  29. Gordis L. Assessing the validity and reliability of diagnostic and screening tests. In: Gordis L, ed. Epidemiology. Philadelphia, Pa: Saunders, 1996; 63–81.
  30. Noyes FR, Stabler CL. A system for grading articular cartilage lesions at arthroscopy. Am J Sports Med 1989;17:505–513.[Abstract/Free Full Text]
  31. Zanetti M, Bruder E, Romero J, Hodler J. Bone marrow edema pattern in osteoarthritic knees: correlation between MR imaging and histologic findings. Radiology 2000;215:835–840.[Abstract/Free Full Text]
  32. Boegard T, Rudling O, Petersson IF, Jonsson K. Correlation between radiographically diagnosed osteophytes and magnetic resonance detected cartilage defects in the tibiofemoral joint. Ann Rheum Dis 1998;57:401–407.[Abstract/Free Full Text]
  33. Tschirch FT, Schmid MR, Pfirrmann CW, Romero J, Hodler J, Zanetti M. Prevalence and size of meniscal cysts, ganglionic cysts, synovial cysts of the popliteal space, fluid-filled bursae, and other fluid collections in asymptomatic knees on MR imaging. AJR Am J Roentgenol 2003;180:1431–1436.[Abstract/Free Full Text]
  34. Hill CL, Gale DG, Chaisson CE, et al. Knee effusions, popliteal cysts, and synovial thickening: association with knee pain in osteoarthritis. J Rheumatol 2001;28:1330–1337.[Medline]
  35. Schweitzer ME, Falk A, Pathria M, Brahme S, Hodler J, Resnick D. MR imaging of the knee: can changes in the intracapsular fat pads be used as a sign of synovial proliferation in the presence of an effusion? AJR Am J Roentgenol 1993;160:823–826.[Abstract/Free Full Text]
  36. Fernandez-Madrid F, Karvonen RL, Teitge RA, Miller PR, An T, Negendank WG. Synovial thickening detected by MR imaging in osteoarthritis of the knee confirmed by biopsy as synovitis. Magn Reson Imaging 1995;13:177–183.[CrossRef][Medline]
  37. Crues JV, Mink J, Levy TL, Lotysch M, Stoller DW. Meniscal tears of the knee: accuracy of MR imaging. Radiology 1987;164:445–448.[Abstract/Free Full Text]
  38. Butler WJ, Hawthorne VM, Mikkelsen WM, et al. Prevalence of radiologically defined osteoarthritis in the finger and wrist joints of adult residents, Tecumseh, Michigan, 1962–1965. J Clin Epidemiol 1988;41:467–473.[CrossRef][Medline]
  39. McAlindon TE, Snow S, Cooper C, Dieppe PA. Radiographic patterns of osteoarthritis of the knee joint in the community: the importance of the patellofemoral joint. Ann Rheum Dis 1992;51:844–849.[Abstract/Free Full Text]
  40. Bhattacharyya T, Gale D, Dewire P, et al. The clinical importance of meniscal tears demonstrated by magnetic resonance imaging in osteoarthritis of the knee. J Bone Joint Surg Am 2003;85-A:4–9.[Abstract/Free Full Text]
  41. Berthiaume MJ, Raynauld JP, Martel-Pelletier J, et al. Meniscal tear and extrusion are strongly associated with progression of knee osteoarthritis as assessed by quantitative magnetic resonance imaging. Ann Rheum Dis 2005;64:556–563.[Abstract/Free Full Text]
  42. Burgkart R, Glaser C, Hyhlik-Durr A, Englmeier KH, Reiser M, Eckstein F. Magnetic resonance imaging-based assessment of cartilage loss in severe osteoarthritis: accuracy, precision, and diagnostic value. Arthritis Rheum 2001;44:2072–2077.[CrossRef][Medline]
  43. Raynauld JP, Kauffmann C, Beaudoin G, et al. Reliability of a quantification imaging system using magnetic resonance images to measure cartilage thickness and volume in human normal and osteoarthritic knees. Osteoarthritis Cartilage 2003;11:351–360.[CrossRef][Medline]
  44. Graichen H, von Eisenhart-Rothe R, Vogl T, Englmeier KH, Eckstein F. Quantitative assessment of cartilage status in osteoarthritis by quantitative magnetic resonance imaging: technical validation for use in analysis of cartilage volume and further morphologic parameters. Arthritis Rheum 2004;50:811–816.[CrossRef][Medline]
  45. Gandy SJ, Dieppe PA, Keen MC, Maciewicz RA, Watt I, Waterton JC. No loss of cartilage volume over 3 years in patients with knee osteoarthritis as assessed by magnetic resonance imaging. Osteoarthritis Cartilage 2002;10:929–937.[CrossRef][Medline]
  46. Cicuttini FM, Wluka AE, Forbes A, Wolfe R. Comparison of tibial cartilage volume and radiologic grade of the tibiofemoral joint. Arthritis Rheum 2003;48:682–688.[CrossRef][Medline]
  47. Cicuttini FM, Wang YY, Forbes A, Wluka AE, Glisson M. Comparison between patella cartilage volume and radiological assessment of the patellofemoral joint. Clin Exp Rheumatol 2003;21:321–326.[Medline]
  48. Jones G, Ding C, Scott F, Glisson M, Cicuttini F. Early radiographic osteoarthritis is associated with substantial changes in cartilage volume and tibial bone surface area in both males and females. Osteoarthritis Cartilage 2004;12:169–174.[CrossRef][Medline]
  49. Raynauld JP, Martel-Pelletier J, Berthiaume MJ, et al. Quantitative magnetic resonance imaging evaluation of knee osteoarthritis progression over 2 years and correlation with clinical symptoms and radiologic changes. Arthritis Rheum 2004;50:476–487.[CrossRef][Medline]
  50. Wluka AE, Wolfe R, Stuckey S, Cicuttini FM. How does tibial cartilage volume relate to symptoms in subjects with knee osteoarthritis? Ann Rheum Dis 2004;63:264–268.[Abstract/Free Full Text]
  51. Sharma L, Song J, Felson DT, Cahue S, Shamiyeh E, Dunlop DD. The role of knee alignment in disease progression and functional decline in knee osteoarthritis. JAMA 2001;286:188–195.[Abstract/Free Full Text]
  52. Adams JG, McAlindon T, Dimasi M, Carey J, Eustace S. Contribution of meniscal extrusion and cartilage loss to joint space narrowing in osteoarthritis. Clin Radiol 1999;54:502–506.[CrossRef][Medline]
  53. Gale DR, Chaisson CE, Totterman SM, Schwartz RK, Gale ME, Felson D. Meniscal subluxation: association with osteoarthritis and joint space narrowing. Osteoarthritis Cartilage 1999;7:526–532. [CrossRef][Medline]



This article has been cited by other articles:


Home page
 Ann Rheum DisHome page
C Boileau, J Martel-Pelletier, F Abram, J-P Raynauld, E Troncy, M-A D'Anjou, M Moreau, and J-P Pelletier
Magnetic resonance imaging can accurately assess the long-term progression of knee structural changes in experimental dog osteoarthritis
Ann Rheum Dis, July 1, 2008; 67(7): 926 - 932.
[Abstract] [Full Text] [PDF]

Home page
 Ann Rheum DisHome page
J-P Raynauld, J Martel-Pelletier, M-J Berthiaume, F Abram, D Choquette, B Haraoui, J F Beary, G A Cline, J M Meyer, and J-P Pelletier
Correlation between bone lesion changes and cartilage volume loss in patients with osteoarthritis of the knee as assessed by quantitative magnetic resonance imaging over a 24-month period
Ann Rheum Dis, May 1, 2008; 67(5): 683 - 688.
[Abstract] [Full Text] [PDF]

Home page
 Ann Rheum DisHome page
G Hernandez-Molina, T Neogi, D J Hunter, J Niu, A Guermazi, S Reichenbach, F W Roemer, C E McLennan, and D T Felson
The association of bone attrition with knee pain and other MRI features of osteoarthritis
Ann Rheum Dis, January 1, 2008; 67(1): 43 - 47.
[Abstract] [Full Text] [PDF]

Home page
 Ann Rheum DisHome page
M Grunke and H Schulze-Koops
Successful treatment of inflammatory knee osteoarthritis with tumour necrosis factor blockade
Ann Rheum Dis, April 1, 2006; 65(4): 555 - 556.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2373041989v1
237/3/998    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hayes, C. W.
Right arrow Articles by Sowers, M. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hayes, C. W.
Right arrow Articles by Sowers, M. R.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE