| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 All authors: Department of Radiology, University of Wisconsin Hospital, Clinical Science Center-E3/311, 600 Highland Ave., Madison, WI 53791-3252.
Received June 29, 2005;
accepted after revision August 8, 2005.
Address correspondence to R. Kijowski
(r.kijowski{at}hosp.wisc.edu).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. The study group consisted of 125 patients with symptomatic osteoarthritis of the tibiofemoral joint. For all patients, standing anteroposterior radiographs of the knee were obtained before arthroscopic knee surgery. Each articular surface of the tibiofemoral joint was graded at arthroscopy. Two radiologists retrospectively reviewed the knee radiographs without knowledge of the arthroscopic findings to determine the presence and severity of osteoarthritis of the tibiofemoral joint using the Kellgren-Lawrence, Ahlback, and Brandt grading scales. Correlation coefficients describing the relation between grade of osteoarthritis and severity of articular cartilage degeneration were calculated for each grading scale.
RESULTS. The correlation coefficients for the Kellgren-Lawrence, Ahlback, and Brandt grading scales were 0.49, 0.41, and 0.56, respectively. The differences between the correlation coefficients for the Kellgren-Lawrence and Ahlback grading scales and the correlation coefficients for the Brandt and Ahlback grading scales were statistically significant (p < 0.05). Many patients with no radiographic findings of osteoarthritis had significant articular cartilage degeneration within the tibiofemoral joint.
CONCLUSION. The Kellgren-Lawrence and Brandt grading scales were equally effective in defining the presence of and estimating the severity of osteoarthritis of the tibiofemoral joint but had only a moderately strong correlation with the actual degree of articular cartilage degeneration.
Keywords: cartilage • knee • MRI • musculoskeletal system • osteoarthritis
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Several radiographic grading scales for osteoarthritis of the tibiofemoral joint have been developed. These scales include the Kellgren-Lawrence, Ahlback, and Brandt scales [3-5]. In several studies these grading scales have been used to correlate radiographic findings of osteoarthritis with arthroscopic findings of articular cartilage degeneration within the tibiofemoral joint [4, 6-9]. To our knowledge, however, in no study has statistical analysis been used to compare the effectiveness of these grading scales in defining the presence of and estimating the severity of articular cartilage degeneration. This study was performed to correlate the radiographic grade of osteoarthritis according to the Kellgren-Lawrence, Ahlback, and Brandt grading scales with the actual degree of articular cartilage degeneration within the tibiofemoral joint in patients with chronic knee pain.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study group consisted of 125 patients (66 men and 59 women; age range, 35-76 years; average age, 51 years) with osteoarthritis of the tibiofemoral joint. A diagnosis of osteoarthritis was made when a patient complained of chronic knee pain and was found during arthroscopic knee surgery to have articular cartilage degeneration within the tibiofemoral joint. The same diagnostic criteria were used in previous studies correlating radiographic findings with arthroscopic findings in patients with osteoarthritis of the tibiofemoral joint [4, 6-9].
The patients in the study group were selected from a database of all MR examinations of the knee performed at our institution between January 1999 and June 2004. The MR database was used to identify 1,554 patients who had osteoarthritis of the tibiofemoral joint. The medical records of these 1,554 patients were reviewed to identify 125 patients who underwent subsequent arthroscopic surgery on the symptomatic knee. The presence of articular cartilage degeneration within the tibiofemoral joint in all 125 patients was confirmed at arthroscopic surgery. All 125 patients with osteoarthritis of the tibiofemoral joint complained of chronic knee pain that had lasted more than 2 months and had no history of recent knee trauma, previous knee surgery, inflammatory arthritis, septic arthritis, or crystalline-induced arthritis.
Radiographic Examination
An anteroposterior radiograph of the knee was obtained for all 125 patients
in the study group. All radiographs were obtained with the patient in the
upright standing position with the knee fully extended. All radiographs were
obtained with standardized technique. The X-ray cassette was placed posterior
to the knee at a tube-film distance of 40 inches (102 cm) with the X-ray beam
projecting parallel to the tibial plateau and centered on the patella.
Arthroscopic Knee Surgery
All 125 patients in the study group underwent arthroscopic knee surgery
within 2 months of the radiographic examination. All arthroscopic knee
operations were performed at our institution by one of three experienced
orthopedic surgeons specializing in sports medicine. The indications for
surgery were débridement or repair of a meniscal tear in 56 patients,
débridement or repair of a meniscal tear and débridement of an
articular cartilage defect in 42 patients, débridement of an articular
cartilage defect in 22 patients, and removal of intraarticular loose bodies in
five patients. At arthroscopic knee surgery the articular cartilage of the
medial femoral condyle, lateral femoral condyle, medial tibial plateau, and
lateral tibial plateau of each patient was graded according to the Noyes
classification system [10]
(Table 1). At arthroscopic knee
surgery the orthopedic surgeons were aware of the radiographic and MRI
findings of all patients.
|
Radiographic Grading of Osteoarthritis
The standing anteroposterior radiographs of the knee of all 125 patients
were retrospectively reviewed in consensus by two fellowship-trained
musculoskeletal radiologists. The radiologists were unaware of the
arthroscopic findings when viewing the knee radiographs. The radiologists used
the Kellgren-Lawrence, Ahlback, and Brandt radiographic grading scales to
determine the presence and severity of osteoarthritis of the tibiofemoral
joint in each patient (Tables
2,
3,
4). The images from the
original articles describing the radiographic grading scales were used as
references during the study
[3-5].
|
|
|
Estimation of the Severity of Osteoarthritis of the Tibiofemoral Joint
Two methods were used to estimate the severity of osteoarthritis of the
tibiofemoral joint for each patient in the study group. Both the highest
arthroscopic grade of articular cartilage degeneration and the total
arthroscopic grade of articular cartilage degeneration were used to represent
the severity of osteoarthritis. When both methods were used, higher numeric
values represented more severe articular cartilage degeneration within the
tibiofemoral joint.
In the first estimation method, the highest arthroscopic grade of articular cartilage degeneration was used to represent the severity of osteoarthritis of the tibiofemoral joint. Articular cartilage defects of grades 1A, 1B, 2A, 2B, 3A, and 3B were given numeric values of 1, 2, 3, 4, 5, and 6, respectively. The highest grade of articular cartilage degeneration was equivalent to the highest numeric value of the grades of the cartilage defects identified on the medial and lateral femoral condyles and medial and lateral tibial plateau at arthroscopy. For example, a highest grade of 5 was used to represent the degree of osteoarthritis within the tibiofemoral joint in a patient with a grade 2B defect (numeric value, 4) within the medial femoral condyle, normal cartilage (numeric value, 0) in the lateral femoral condyle, a grade 3A defect (numeric value, 5) of the medial tibial plateau, and normal cartilage (numeric value, 0) in the lateral tibial plateau.
In the second method, the total arthroscopic grade of articular cartilage degeneration was used to represent the severity of osteoarthritis of the tibiofemoral joint. Articular cartilage defects of Noyes grades 1A, 1B, 2A, 2B, 3A, and 3B were given numeric values of 1, 2, 3, 4, 5, and 6, respectively. The total grade of articular cartilage degeneration was calculated by summing the numeric values of the grades of the cartilage defects identified on the medial and lateral femoral condyles and medial and lateral tibial plateau at arthroscopy. For example, a total grade of 9 was used to represent the degree of articular cartilage degeneration within the knee joint in a patient with a grade 2B defect (numeric value, 4) within the medial femoral condyle, normal cartilage (numeric value, 0) in the lateral femoral condyle, a grade 3A defect (numeric value, 5) of the medial tibial plateau, and normal cartilage (numeric value, 0) in the lateral tibial plateau.
Statistical Analysis
Statistical analysis was performed to correlate the radiographic grade of
osteoarthritis according to the Kellgren-Lawrence, Ahlback, and Brandt grading
scales with the degree of articular cartilage degeneration within the
tibiofemoral joint identified at arthroscopy. Mean numeric values representing
the severity of articular cartilage degeneration within the tibiofemoral joint
were calculated for each radiographic grade of osteoarthritis according to the
Kellgren-Lawrence, Ahlback, and Brandt grading scales. Correlation
coefficients and squared correlation coefficients were calculated for the
Kellgren-Lawrence, Ahlback, and Brandt radiographic grading scales.
Differences between the correlation coefficients of the three radiographic
grading scales were formally tested
[11]. Differences were
considered statistically significant if the p value was less than
0.05. In the statistical analysis, both the highest arthroscopic grade of
articular cartilage degeneration and the total arthroscopic grade of articular
cartilage degeneration were used to represent the severity of osteoarthritis
of the tibiofemoral joint. The statistical software program S-Plus (version
3.4, Mathsoft) was used to perform the statistical analysis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
|
In general, higher grades of osteoarthritis were associated with more severe articular cartilage degeneration in all three radiographic grading scales. For the Ahlback and Brandt grading scales, the mean numeric values representing the severity of articular cartilage degeneration increased with each increasing radiographic grade of osteoarthritis. However, for the Kellgren-Lawrence radiographic grading scale, the mean numeric value representing the severity of articular cartilage degeneration for patients with grade 1 or doubtful osteoarthritis was higher than the mean numeric value for patients with grade 2 or definite osteoarthritis.
Table 8 shows the correlation coefficients with 95% CIs describing the relation between the radiographic grade of osteoarthritis according to the Kellgren-Lawrence, Ahlback, and Brandt grading scales and the degree of articular cartilage degeneration within the tibiofemoral joint identified at arthroscopy. The Brandt grading scale had the strongest correlation and the Ahlback grading scale had the weakest correlation with actual degree of osteoarthritis of the tibiofemoral joint identified at arthroscopy. The difference between the correlation coefficients of the Kellgren-Lawrence and Brandt grading scales was not statistically significant (p = 0.39 for the highest arthroscopic grade and p = 0.065 for the total arthroscopic grade). However, the differences between the correlation coefficients of the Kellgren-Lawrence and Ahlback grading scales and the correlation coefficients of the Brandt and Ahlback grading scales were significantly significant (p < 0.05).
|
Table 9 lists the squared correlation coefficients describing the relation between the radiographic grade of osteoarthritis according to the Kellgren-Lawrence, Ahlback, and Brandt grading scales and the degree of articular cartilage degeneration within the tibiofemoral joint identified at arthroscopy. The radiographic grading scales had at best a moderately strong correlation with the actual degree of articular cartilage degeneration. For example, the Brandt grading scale, which had the highest correlation with the actual severity of osteoarthritis, had a squared correlation coefficient value of only 0.36. This finding means that differences in the grade of osteoarthritis with the Brandt grading scale could account for only 36% of the variation in the severity of articular cartilage degeneration within the tibiofemoral joint in our patient population.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In our study, joint space loss occurred after osteophyte formation in most patients with osteoarthritis of the tibiofemoral joint. Only two of the 125 patients with osteoarthritis in our study group had joint space narrowing but no associated osteophytes on knee radiographs. These two patients could not be classified according to the Kellgren-Lawrence grading system.
According to the Kellgren-Lawrence radiographic grading scale, grade 1 osteoarthritis is defined by the presence of minute osteophytes of doubtful significance, and grade 2 osteoarthritis is defined by the presence of definite osteophytes with no associated joint space loss [3]. In our study, many more patients had Kellgren-Lawrence grade 1 osteoarthritis than had Kellgren-Lawrence grade 2 osteoarthritis. Furthermore, patients with grade 1 osteoarthritis had, on average, more severe articular cartilage degeneration within the tibiofemoral joint than did patients with grade 2 osteoarthritis. In our study, the minute osteophytes in patients with Kellgren-Lawrence grade 1 osteoarthritis were not of doubtful significance. Instead, the clinical significance of these tiny osteophytes as a radiographic finding of osteoarthritis was the same as that of larger osteophytes in patients with Kellgren-Lawrence grade 2 osteoarthritis.
Brandt and associates [4] described a radiographic grading scale of osteoarthritis of the tibiofemoral joint that emphasized joint space narrowing rather than osteophyte formation. According to the Brandt grading scale, grade 1 osteoarthritis is defined by the presence of osteophytes with minimal associated joint space narrowing and corresponds to Kellgren-Lawrence grades 1 and 2 osteoarthritis. Brandt grade 2 osteoarthritis is defined as joint space narrowing with no associated osteophytes and has no equivalent in the Kellgren-Lawrence grading system. Brandt and associates concluded in their study that their new radiographic grading scale had no advantage over the Kellgren-Lawrence grading scale in assessment of the severity of articular cartilage degeneration within the tibiofemoral joint. We came to a similar conclusion in our study. There was no statistically significant difference between the Kellgren-Lawrence and Brandt grading scales in correlation coefficients describing the relation between grade of osteoarthritis and degree of articular cartilage degeneration. Furthermore, the addition of a new grade of osteoarthritis in the Brandt grading system to describe patients with joint space narrowing but no associated osteophytes was of little clinical significance. Only two of the 125 patients in the study group had Brandt grade 2 osteoarthritis.
The Ahlbach radiographic grading scale of osteoarthritis of the tibiofemoral joint is based exclusively on the presence of joint space narrowing [5]. Our study showed that the Ahlbach grading scale is inferior in relation to the Kellgren-Lawrence and Brandt grading scales in defining the presence of and estimating the severity of articular cartilage degeneration within the tibiofemoral joint. The inferior performance of the Ahlbach grading scale probably is the result of its use of joint space narrowing as the only radiographic feature of osteoarthritis. Previous studies have shown that joint space narrowing is an insensitive radiographic finding of osteoarthritis of the tibiofemoral joint [6, 13].
Our study raised serious questions about the usefulness of radiographic grading scales in defining the presence of and estimating the severity of osteoarthritis of the tibiofemoral joint. Many patients in our study with no radiographic findings of osteoarthritis had significant articular cartilage degeneration within the tibiofemoral joint. Furthermore, the Kellgren-Lawrence, Ahlback, and Brandt grading scales had at best a moderately strong correlation with the actual degree of osteoarthritis. The results of our study suggest that imaging methods more sensitive than knee radiography are needed to define the presence of and estimate the severity of osteoarthritis of the tibiofemoral joint in epidemiologic and clinical studies.
In several previous studies, investigators correlated radiographic findings of osteoarthritis according to the Kellgren-Lawrence, Ahlback, and Brandt grading scales with arthroscopic findings of articular cartilage degeneration within the tibiofemoral joint. To our knowledge, however, in no previous study has statistical analysis been used to compare the usefulness of these radiographic grading scales in defining the presence of and estimating the severity of articular cartilage degeneration. Lysholm and associates [6] concluded that the Ahlback grading scale was insensitive in detecting early articular cartilage degeneration in 63 patients with osteoarthritis of the tibiofemoral joint. Blackburn and associates [8] found that use of the Kellgren-Lawrence grading scale led to significant underestimation of the degree of articular cartilage degeneration in 36 patients with osteoarthritis of the tibiofemoral joint. Brandt and associates [4] concluded that their newly developed grading scale had no advantage over the Kellgren-Lawrence grading scale in assessment of the severity of articular cartilage degeneration in 92 patients with osteoarthritis of the tibiofemoral joint. Those authors also found that both radiographic grading scales were insensitive in detection of early articular cartilage degeneration within the tibiofemoral joint. Wada and associates [9] found that the Kellgren-Lawrence grading scale and the Ahlback grading scale had a sensitivity of 98% and 91%, respectively, in the detection of articular cartilage degeneration in 173 patients with osteoarthritis of the medial compartment of the tibiofemoral joint. However, most patients in that study had advanced osteoarthritis of the tibiofemoral joint characterized by deep partial-thickness or full-thickness articular cartilage defects within the medial femoral condyles and medial tibial plateau.
The major limitation of our study was the presence of selection bias. All patients in our study group were selected from a database of MR examinations of the knee performed at our institution over the past several years. This selection method allowed us to identify a large number of patients with arthroscopically confirmed osteoarthritis of the tibiofemoral joint. The large number of patients with osteoarthritis allowed performance of statistical analysis on the data collected during the study. In addition, all patients in our study group had symptomatic osteoarthritis. Because of inherent selection bias, our patient population represented only a small subset of persons in our community with osteoarthritis of the tibiofemoral joint. Another limitation of our study was the absence of patients in the study group with arthroscopically confirmed normal articular cartilage within the tibiofemoral joint. For this reason, the correlation coefficients describing the relation between the radiographic grade of osteoarthritis according to the Kellgren-Lawrence, Ahlback, and Brandt grading scales and the actual degree of articular cartilage degeneration within the tibiofemoral joint did not take into account the specificity of these grading systems in the detection of articular cartilage degeneration.
In conclusion, in our study the Kellgren-Lawrence and Brandt radiographic grading scales were equally effective in defining the presence of and estimating the severity of articular cartilage degeneration within the tibiofemoral joint. These grading scales, however, did not define the presence of osteoarthritis in a large number of patients with the disease and had only a moderately strong correlation with the actual degree of articular cartilage degeneration within the tibiofemoral joint. The results of our study suggest that imaging methods more sensitive than knee radiography are needed to define the presence of and estimate the severity of osteoarthritis of the tibiofemoral joint in epidemiologic and clinical studies.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  W. F. M. Jackson, W. M. van der Tempel, L. J. Salmon, H. A. Williams, and L. A. Pinczewski Endoscopically-assisted single-bundle posterior cruciate ligament reconstruction: RESULTS AT MINIMUM TEN-YEAR FOLLOW-UP J Bone Joint Surg Br, October 1, 2008; 90-B(10): 1328 - 1333. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
