SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

Objective

To identify structural differences in total subchondral bone area (tAB) and cartilage thickness between healthy reference knees and knees with radiographic osteoarthritis (OA).

Methods

Baseline magnetic resonance images from 1 knee of 1,003 Osteoarthritis Initiative participants were studied: 112 healthy reference knees without radiographic OA, symptoms, or risk factors; 70 preradiographic OA knees (calculated Kellgren/Lawrence [K/L] grade 0/1); and 821 radiographic OA knees (calculated K/L grade ≥2). Means and standard (Z) scores (SD unit differences compared with normal subjects) of the tAB and regional cartilage thickness were assessed in the weight-bearing femorotibial joint and compared between groups.

Results

In men, tAB was 8.2% larger in preradiographic OA knees and 6.6%, 8.1%, and 8.5% larger in calculated K/L grade 2, 3, and 4 radiographic OA knees, respectively, than in reference knees. In women, the differences were +6.8%, +7.3%, +9.9%, and +8.1%, respectively. The external medial tibia showed the greatest reduction in cartilage thickness (Z scores −5.1/−5.6 in men/women) with Osteoarthritis Research Society International medial joint space narrowing (JSN) grade 3, and the external lateral tibia (Z scores −6.0 for both sexes) showed the greatest reduction with lateral JSN grade 3. In all subregions of end-stage radiographic OA knees, ≥25% of the average normal cartilage thickness was maintained. An overall trend toward thicker cartilage was found in preradiographic OA and calculated K/L grade 2 knees, especially in the external central medial femur.

Conclusion

tABs were larger in preradiographic OA and radiographic OA knees than in healthy reference knees, and the difference did not become larger with higher calculated K/L grades. Specific subregions with substantial cartilage thickening or thinning were identified in pre-, early, and late radiographic OA.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

Osteoarthritis (OA) is a multi-tissue joint disease causing disabling symptoms and is characterized by pathology and progressive failure of multiple joint tissues (1). Radiographs have commonly been used to grade the severity of radiographic OA, and joint space narrowing (JSN) has historically been used as a surrogate of cartilage thinning (2–5). Magnetic resonance imaging (MRI) is, however, increasingly used to obtain quantitative end points of morphology in OA (6–8). Using MRI, 3 measures (total area of subchondral bone [tAB], cartilage thickness, and denuded areas of subchondral bone) have been shown to provide independent information and were thus recommended as preferred parameters for a comprehensive description of the cross-sectional and longitudinal variation in healthy and osteoarthritic cartilage (9).

Using MRI, one study reported that knees with medial JSN displayed a lower cartilage volume, but no differences in tibial tAB in the medial tibia (MT) and lateral tibia (LT), than did knees without medial JSN, whereas the presence of osteophytes was related to larger tibial tABs but not to differences in cartilage volume (10). A recent study reported that knees of women with a high body mass index (BMI) and Kellgren/Lawrence (K/L) grade 3 had thinner cartilage and those with a high BMI and K/L grade 2 had thicker cartilage than healthy reference knees (10). However, both K/L grade 2 and K/L grade 3 knees displayed larger tABs than reference knees (11).

An increase of tibial tAB has been suggested to represent an early structural feature of knee OA (12). In OA knees, an annual increase of tibial tAB of 2.2% medially and 1.5% laterally was reported (13), but similar rates of tAB expansion have also been observed in healthy (non-OA) knees (14, 15). No large-sample studies to our knowledge, however, have previously compared tABs in healthy knees and in knees with specific radiographic OA stages.

The objective of this cross-sectional study was to describe and compare differences in femorotibial tAB and regional cartilage thickness between healthy control knees of men and women without radiographic OA, symptoms, or OA risk factors, and knees with preradiographic OA or different specific stages of definite radiographic OA using data from the Osteoarthritis Initiative (OAI). Specifically, we asked whether tABs are larger already in the earliest stages of the disease, which specific femorotibial subregions display cartilage thickness differences in early and later disease stages, and whether there is a relationship between the differences in tAB and those in regional cartilage thickness.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

Knees were selected from the OAI (online at www.oai.ucsf.edu), which includes participants ages 45–79 years with or at risk for symptomatic femorotibial OA. Exclusion criteria have been described previously (16, 17). The OAI also includes a healthy (nonexposed) reference subcohort, which had no symptoms at baseline (answered “No” to the question: “In the past 12 months, have you had any pain, aching or stiffness in or around either knee?”), no radiographic signs of OA (Osteoarthritis Research Society International [OARSI] grade 0 for osteophytes and JSN), and none of the other OAI screening risk factors (i.e., overweight, history of knee injury or surgery, family history of knee replacement, Heberden's nodes, and frequent knee-bending activities). Fixed flexion radiographs and 3T MRIs were acquired and provided for public use (18–20). A sagittal 3-dimensional double-echo steady-state sequence (DESS) with water excitation was available for both knees, and a coronal fast low-angle shot (FLASH) with water excitation was available for 1 (usually the right) knee (Figure 1).

thumbnail image

Figure 1. A, Reconstruction of the weight-bearing parts of the central medial femur (cMF) and central lateral femur (cLF) displaying the various subregions. B, Magnetic resonance (MR) image acquired with a double-oblique coronal fast low-angle shot sequence with water excitation showing the different femorotibial cartilage plates.C, MR image acquired with a sagittal double-echo steady-state (DESS) with water excitation. The lines in the DESS image indicate the analyzed femoral regions of interest and how the central weight-bearing region is divided from the trochlea and from the posterior region. D, Reconstruction of the medial tibia (MT) and lateral tibia (LT) displaying the various subregions. ecLF = external cLF; ccLF = central cLF; icLF = internal cLF; icMF = internal cMF; ccMF = central cMF; ecMF = external cMF; aLT = anterior LT; eLT = external LT; cLT = central LT; iLT = internal LT; pLT = posterior LT; aMT = anterior MT; iMT = internal MT; cMT = central MT; eMT = external MT; pMT = posterior MT.

Download figure to PowerPoint

Radiographic grading.

The radiographic grading used here relied on the baseline calculated K/L grades, derived from osteophyte and JSN grades. These were assigned by centrally trained and certified readers at the clinical sites. Readers assessed each knee for the presence/absence of definite marginal osteophytes (OARSI atlas grade 1–3: any medial and lateral tibial and femoral osteophytes) and medial and lateral JSN grades 1 (OARSI atlas grades 1–2) or 2 (OARSI atlas grade 3) (21, 22). The calculated K/L grade was defined as 0 = grade 0 for both osteophytes and JSN scores, 1 = questionable osteophyte and grade 0–1 or grade 0 osteophyte and grade 1 JSN, 2 = definite osteophyte and grade 0 JSN or no/questionable osteophyte and grade 2 JSN, 3 = definite osteophyte and grade 1 JSN, and 4 = definite osteophyte and grade 2 JSN.

Sample selection.

This cross-sectional analysis was performed in a sample of OAI participants analyzed by a consortium of pharmaceutical industry partners, the OAI coordinating center, and an image analysis company (Chondrometrics). Baseline MRI data from 1,003 knees (956 right and 47 left; 907 FLASH with water excitation and 96 DESS with water excitation) from 1,003 subjects were studied. There were 112 of 122 knees with usable baseline acquisitions from the healthy, nonexposed cohort (public-use data set 0.F.1), constituting references for comparative analyses in this study. There were 158 knees from an age- and sex-stratified sample of the progression subcohort (public-use data set 0.B.1) (16, 23). This sample included participants with frequent symptoms and a calculated K/L grade ≥2 in at least 1 knee. As right knees were studied (16), knees with and without symptoms spanning all calculated K/L grades were included. There were 495 knees with calculated K/L grade 2–3, selected by ascending OAI identification numbers from the first half of the OAI cohort (public-use data set 0.C.1). No other inclusion or exclusion criteria were applied. There were 90 knees from participants with calculated K/L grade 4 (public-use data set 0.C.1). There were 52 knees with a calculated K/L grade 0–1 (public-use data set 0.C.1) that had a calculated K/L grade ≥2 in the contralateral knees. Finally, there were 96 knees with calculated K/L grade 3–4 selected by the OAI coordinating center for a “core image assessment sample” of the progression subcohort. These had baseline radiographic OA, frequent pain, and acceptable knee radiographs and knee MRIs from both the baseline and the 2-month visits. Sagital DESS with water excitation acquisitions were used, whereas for all other subsamples, the coronal FLASH with water excitation was used. A high agreement between the coronal FLASH with water excitation and the sagittal DESS with water excitation was recently shown (24, 25).

Baseline clinical and radiographic data were obtained from public-use data set 0.2.2. Knees with a calculated K/L grade 0–1 (calculated K/L grade ≥2 contralaterally) were considered preradiographic OA, and knees with a calculated K/L grade ≥2 were considered radiographic OA knees. Calculated K/L grade 2 was considered early radiographic OA, calculated K/L grade 3 late-stage radiographic OA, and calculated K/L grade 4 end-stage radiographic OA. For the calculated K/L grade comparisons, 18 calculated K/L grade 1 knees and 6 calculated K/L grade 2 knees with JSN >0 were excluded post hoc from the analysis (for calculated K/L grade definitions, see above) so that all preradiographic OA and early radiographic OA knees were without JSN.

MRI analysis.

MR images were acquired as described previously (16, 17, 20, 26). After quality control at the image analysis center, segmentation of the femorotibial cartilages was performed by 7 operators. The tAB (6) and the cartilage surface area were traced manually in each section for the MT and LT, and for the weight-bearing central medial femur (cMF) and central lateral femur (cLF) (16, 24, 27), excluding osteophytes (see Supplementary Table A, available in the online version of this article at http://www3.interscience.wiley.com/journal/77005015/home). The region of interest of the femoral condyles was defined by a 60% distance between the intercondylar notch and the posterior end of the femoral condyles in the coronal and sagittal images (24). All segmentations were quality controlled by 1 expert who, as well as the operators, was blinded to the clinical and radiographic data. The tAB and cartilage thickness over the entire tAB, including denuded areas with 0-mm cartilage thickness, were computed (16 femorotibial subregions) using Chondrometrics software as described previously (16, 24, 27).

Precision (test–retest) errors at the cartilage plate level have been explored in the OAI pilot studies. For cartilage thickness, they were reported to range from 2.3% (MT) to 4.5% (cMF) with the sagittal DESS for unpaired analysis (24), from 1.8% (LT) to 2.8% (cLF) with the sagittal DESS for paired analysis (28), from 3.1% (MT) to 5.4% (cMF) with the coronal FLASH for unpaired analysis (24), and from 1.1% (LT) to 2.3% (cMF) with the coronal FLASH for paired analysis (28). Precision tests for tAB ranged from 2.6% (MT) to 9.2% (LT) with the sagittal DESS for unpaired analysis (24), from 1.2% (MT) to 2.5% (LT) with the sagittal DESS for paired analysis (28), from 2.8% (MT) to 5.4% (cMF) with the coronal FLASH for unpaired analysis (24), and from 0.9% (LT and cMF) to 1.5% (MT) with the coronal FLASH for paired analysis (28). Cartilage thickness in the central, external, and internal subregions of MT, LT, cMF, and cLF, and in the anterior and posterior subregions of MT and LT, were computed as described previously (27, 29) (Figure 1). Precision (test–retest) error of regional cartilage thickness were reported to range from 19 μm (1.5%) in the external subregion of MT to 84 μm (4.7%) in the posterior subregion of LT (29).

Statistical analyses.

All analyses were stratified by sex, as men and women showed significant differences in tABs and cartilage thickness (30, 31). Distributions of these parametric variables in the healthy reference knees were used to calculate standardized differences (Z scores) in each quantitative measure for knees grouped according to calculated K/L grade and JSN grades: (mean in calculated K/L grade or JSN group − mean of reference cohort)/SD of reference cohort (32). A medial/lateral ratio of differences in tAB ([MT + cMF]/[LT + cLF]) between healthy reference knees and knees with JSN was computed and presented.

Statistical tests were performed using SPSS, version 15.0 (SPSS). Crude values of tAB and regional cartilage thickness were presented as means and SDs. Due to the relatively small number of knees with calculated K/L grade 0 (n = 32) or 1 (n = 38) and because all calculated K/L grade 0 knees were from participants with definite radiographic OA (i.e., calculated K/L grade ≥2) in the contralateral knee, the calculated K/L grade 0/1 cases were combined into 1 group of preradiographic OA knees. No differences in age, sex, or BMI were found between participants with calculated K/L grade 0 or grade 1 knees (P > 0.161). Because calculated K/L grade is not compartment specific, we also used medial and lateral JSN grades. First, univariate comparisons between knees grouped according to calculated K/L grade and JSN grades and healthy reference knees were performed using t-tests, followed by multivariate analyses adjusting for covariates using general linear models. Because age has been suggested to be related to tAB but not to cartilage thickness (13, 14), multivariate comparisons of tAB were adjusted for age and BMI, whereas cartilage thickness comparisons were adjusted for BMI only. Exploratively, and on a cartilage plate level only, we did adjust cartilage thickness comparisons for age in addition to BMI, but these results remained basically unchanged (data not shown).

In knees with medial and lateral JSN and for both men and women, we also calculated the medial/lateral ratio for the tAB difference between radiographic OA and healthy knees in each cartilage plate. A Bonferroni-adjusted significance level of 0.0125 was applied to correct for multiple testing of tAB (4 cartilage plates), and a Bonferroni-adjusted significance level of 0.0025 to correct for multiple testing of cartilage thickness (4 plates and 16 subregions). However, P values were not adjusted for multiple comparisons between various calculated K/L grade and JSN groups or for the exploratory age- and BMI-adjusted model of cartilage thickness comparisons (33). We assessed the relation between the difference in tAB bone area and regional cartilage thickness (Z scores) in subjects with preradiographic OA and radiographic OA (n = 867) using Spearman's rho.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

Demographics and crude values for tAB and cartilage thickness are summarized in Table 1 (crude regional cartilage thickness values are presented in Supplementary Table A, available in the online version of this article at http://www3.interscience.wiley.com/journal/77005015/home). Men and women with preradiographic OA and radiographic OA knees were significantly older and showed a significantly higher BMI than healthy subjects. No clinically relevant differences in body height were found between groups (Table 1).

Table 1. Subject demographics: crude means and SDs for the tAB (mm2) and regional cartilage thickness (mm) for the healthy reference knees and for preradiographic OA and radiographic OA by each calculated K/L grade*
 Healthy references (n = 112)Preradiographic OARadiographic OA
Grade 0/1 (n = 70)Grade 2 (n = 315)Grade 3 (n = 387)Grade 4 (n = 119)
  • *

    Values are the mean ± SD unless otherwise indicated. tAB = total area of subchondral bone; OA = osteoarthritis; K/L = Kellgren/Lawrence; BMI = body mass index; MT = medial tibia; cMF = central medial femur; LT = lateral tibia; cLF = central lateral femur; OARSI = Osteoarthritis Research Society International; JSN = joint space narrowing; OAI = Osteoarthritis Initiative.

  • Fourteen of 37 men and 4 of 33 women had no osteophytes but an OARSI JSN grade of 1–2 and were classified as calculated K/L grade 1 in the OAI database, version 0.2.2. These cases were excluded post hoc from the calculated K/L grade analysis.

  • Two of 115 men and 4 of 201 women had no osteophytes but an OARSI JSN grade of 3 and were classified as calculated K/L grade 2 in the OAI database, version 0.2.2. These cases were excluded post hoc from the calculated K/L grade analysis.

  • §

    P < 0.05 compared with the healthy reference group (statistical comparison only for demographic variables).

  • Data missing for 9 men and 11 women.

Men, no. (n = 402)432311314760
 Demographics     
  Age, years57.0 ± 9.661.3 ± 9.7§61.1 ± 9.0§63.4 ± 9.8§62.0 ± 9.6§
  Height, cm175 ± 6.9175 ± 5.9175 ± 6.2177 ± 6.7176 ± 6.1
  Weight, kg79.3 ± 8.291.8 ± 12.4§90.0 ± 13.1§93.5 ± 15.0§90.5 ± 14.5§
  BMI, kg/m226.1 ± 3.029.8 ± 3.9§29.2 ± 3.8§29.8 ± 3.9§29.1 ± 4.2§
 tAB     
  MT12.8 ± 1.213.9 ± 1.514.0 ± 1.613.8 ± 1.614.2 ± 1.8
  cMF6.0 ± 0.76.6 ± 0.96.4 ± 0.86.8 ± 1.06.8 ± 1.1
  LT11.4 ± 1.112.0 ± 1.112.0 ± 1.512.2 ± 1.512.0 ± 1.7
  cLF6.9 ± 0.87.2 ± 0.87.2 ± 0.87.4 ± 0.97.3 ± 1.2
 Cartilage thickness     
  MT1.8 ± 0.21.8 ± 0.21.9 ± 0.21.8 ± 0.31.5 ± 0.4
  cMF2.0 ± 0.32.0 ± 0.32.1 ± 0.41.9 ± 0.41.3 ± 0.7
  LT2.3 ± 0.32.2 ± 0.32.3 ± 0.32.0 ± 0.41.8 ± 0.6
  cLF1.9 ± 0.31.9 ± 0.32.0 ± 0.32.0 ± 0.31.9 ± 0.5
Women, no. (n = 601)692919724058
 Demographics     
  Age, years53.8 ± 6.059.4 ± 9.9§60.5 ± 9.2§64.2 ± 8.9§65.1 ± 8.0§
  Height, cm164 ± 6.5163 ± 4.7162 ± 5.8§163 ± 6.5163 ± 5.3
  Weight, kg61.9 ± 8.277.6 ± 13.3§76.6 ± 14.4§78.3 ± 15.7§80.7 ± 14.4§
  BMI, kg/m223.1 ± 2.529.1 ± 4.9§29.4 ± 5.0§29.7 ± 5.4§30.5 ± 5.3§
 tAB     
  MT10.1 ± 0.910.7 ± 0.910.7 ± 1.110.8 ± 1.210.7 ± 1.1
  cMF4.9 ± 0.65.3 ± 0.75.3 ± 0.75.6 ± 0.75.5 ± 0.7
  LT8.5 ± 0.99.3 ± 0.89.0 ± 1.09.2 ± 1.29.1 ± 1.0
  cLF5.4 ± 0.65.7 ± 0.75.7 ± 0.75.9 ± 0.75.9 ± 0.7
 Cartilage thickness     
  MT1.6 ± 0.21.6 ± 0.21.7 ± 0.21.6 ± 0.21.4 ± 0.4
  cMF1.7 ± 0.21.7 ± 0.21.8 ± 0.31.6 ± 0.31.3 ± 0.6
  LT2.0 ± 0.21.9 ± 0.31.9 ± 0.31.7 ± 0.41.4 ± 0.5
  cLF1.6 ± 0.21.7 ± 0.21.8 ± 0.31.7 ± 0.31.6 ± 0.4

Relationship with calculated K/L grades.

Compared with healthy reference knees, the femorotibial tAB was larger both in preradiographic OA and radiographic OA knees (both sexes), with the MT showing the largest difference (Table 2). The cartilage was generally thicker in preradiographic OA (albeit with relatively small differences) and early radiographic OA than in healthy reference knees, but thinner in late- or end-stage radiographic OA (Table 3). Substantially thinner cartilage was found in specific plates and subregions of knees with JSN (Table 4). Calculated K/L grade 4 knees displayed substantially thinner cartilage than healthy reference knees in MT, LT, and cMF, but not in cLF (Table 3). The subregional pattern of thinner cartilage was similar in men and women, with the greatest standardized cartilage loss (smallest Z scores) observed in the external subregion of MT in men (−3.8) and women (−3.0). In calculated K/L grade 3 knees, the greatest reduction of cartilage thickness at the total plate level was in LT for both sexes, followed by cMF. The smallest Z scores were −1.3 (posterior subregion of LT) in men and −1.1 (the posterior and central subregions of LT) in women (Table 3), closely followed by the internal subregion of LT (−1.2 in men and −1.0 in women). Calculated K/L grade 2 knees displayed an overall trend toward thicker cartilage (albeit of a generally small magnitude) compared with healthy knees, with the highest Z scores for the external subregion of cMF (+0.9 for both sexes) (Table 3). In preradiographic OA knees, the external subregion of cMF also showed the most positive values in both men and women, although the differences were smaller. Of the 4 subregions displaying a trend toward thinner cartilage in preradiographic OA knees of men, 3 (the external, internal, and posterior subregions of LT) also displayed substantial cartilage thinning at calculated K/L grade 4 (Table 3).

Table 2. Differences in tAB (mm2) per calculated K/L grade*
RegionGrade 0/1Grade 2Grade 3Grade 4
Difference, %Mean Z scoreDifference, %Mean Z scoreDifference, %Mean Z scoreDifference, %Mean Z score
  • *

    Combined grades 0 and 1 indicate preradiographic OA and grades 2, 3, and 4 indicate radiographic OA. Mean differences in percentage and standard scores (Z scores) were determined in comparison with the means and SDs of the reference cohort (see Table 1 for crude values). FT = femorotibial; MFTC = medial FT compartment; LFTC = lateral FT compartment. See Table 1 for additional definitions.

  • Fourteen of 37 men and 4 of 33 women had no osteophytes but had OARSI JSN grade 1–2 and were classified as calculated K/L grade 1 in the OAI database, version 0.2.2. These cases were excluded post hoc from the calculated K/L grade analysis.

  • Two of 115 men and 4 of 201 women had no osteophytes but OARSI JSN grade 3 and were classified as calculated K/L grade 2 in the OAI database, version 0.2.2. These cases were excluded post hoc from the calculated K/L grade analysis.

  • §

    For grade 0/1, n = 23, for grade 2, n = 113, for grade 3, n = 147, and for grade 4, n = 60.

  • No statistical comparisons were made.

  • #

    P < 0.0125 in univariate analysis and after adjustment for age and BMI.

  • **

    P < 0.0125 in univariate analysis.

  • ††

    For grade 0/1, n = 29, for grade 2, n = 197, for grade 3, n = 240, and for grade 4, n = 58.

Men (n = 343)§        
 Total FT joint+8.2+0.9+6.6+0.7+8.1+0.9+8.5+0.9
  MFTC+10+1.1+8.3+0.9+9.4+1.0+12+1.2
  LFTC+5.9+0.6+4.9+0.5+6.9+0.7+5.3+0.6
 MT+10#+1.1+9.0#+0.9+7.7#+0.8+11#+1.1
 cMF+11#+0.9+6.7#+0.6+13#+1.1+14#+1.1
 LT+5.7+0.6+5.2+0.6+6.6**+0.7+5.1+0.5
 cLF+6.2+0.6+4.3+0.4+7.3#+0.6+5.8+0.5
Women (n = 524)††        
 Total FT joint+6.8+0.8+7.3+0.8+9.9+1.1+8.1+0.9
  MFTC+6.4+0.7+7.8+0.9+11+1.1+8.2+0.9
  LFTC+7.2+0.7+6.8+0.7+9.2+0.9+8.0+0.8
 MT+6.1#+0.7+6.8**+0.7+7.8**+0.8+6.2**+0.7
 cMF+6.7+0.6+9.9#+0.8+16#+1.3+12#+1.0
 LT+9.2#+0.8+6.6**+0.6+9.1#+0.8+7.2**+0.7
 cLF+4.4+0.4+7.1#+0.7+9.5#+0.9+8.9**+0.9
Table 3. Differences in regional cartilage thickness per calculated K/L grade*
 Grade 0/1Grade 2Grade 3Grade 4
Difference, %Mean Z scoreDifference, %Mean Z scoreDifference, %Mean Z scoreDifference, %Mean Z score
  • *

    Combined grades 0 and 1 indicate preradiographic OA and grades 2, 3, and 4 indicate radiographic OA. Mean differences in percentage and standard scores (Z scores) were determined in comparison with the means and SDs of the reference cohort (crude values presented in Table 1 and Supplementary Table A, available in the online version of this article at http://www3.interscience.wiley.com/journal/77005015/home). See Table 1 for definitions.

  • Fourteen of 37 men and 4 of 33 women had no osteophytes but had OARSI JSN grade 1–2, and were classified as calculated K/L grade 1 in the OAI database, version 0.2.2. These subjects were excluded from the analyses.

  • Two of 115 men and 4 of 201 women had no osteophytes but OARSI JSN grade 3, and were classified as calculated K/L grade 2 in the OAI database, version 0.2.2. These subjects were excluded from the analyses.

  • §

    For grade 0/1, n = 23, for grade 2, n = 113, for grade 3, n = 147, and for grade 4, n = 60.

  • P < 0.0025 in univariate analysis and after adjustment for BMI.

  • #

    P < 0.0025 in univariate analysis.

  • **

    For grade 0/1, n = 29, for grade 2, n = 197, for grade 3, n = 240, and for grade 4, n = 58.

Men (n = 343)§        
 Plate        
  MT+4.2+0.4+4.8+0.4−2.0−0.2−18−1.6
  cMF+2.8+0.2+4.8+0.3−6.4−0.5−34−2.5
  LT−0.7−0.1−1.9−0.2−12−1.0−23−1.9
  cLF+6.3+0.5+6.5+0.5+3.6+0.3+0.60
 Subregion        
  Central MT+7.9+0.5+6.1+0.4−3.8−0.3−22−1.5
  External MT+0.9+0.1+1.9+0.1−14−1.0−53−3.8
  Internal MT+5.3+0.5+7.6+0.8+6.9+0.7+2.1+0.2
  Anterior MT+6.5+0.5+5.0+0.400−5.6−0.5
  Posterior MT−1.4−0.1+3.4+0.2+1.0+0.1−16−1.1
  Central cMF+2.4+0.1+3.9+0.2−13#−0.8−46−2.8
  External cMF+7.4+0.5+13+0.900−41−2.9
  Internal cMF+0.50+1.4+0.1−3.9−0.3−18−1.2
  Central LT+1.8+0.1+0.50−13−0.8−25−1.6
  External LT−2.3−0.2−0.10−8.2−0.7−20−1.7
  Internal LT−2.1−0.2−4.1−0.3−15−1.2−30−2.4
  Anterior LT00+2.7+0.2−0.9−0.1−7.3−0.5
  Posterior LT−2.4−0.2−8.4−0.6−19−1.3−28−1.9
  Central cLF+7.4+0.4+6.9+0.4+3.3+0.2+0.70
  Exterior cLF+5.5+0.4+6.3+0.4+2.6+0.2−0.8−0.1
  Interior cLF+6.1+0.5+6.7+0.5+5.4+0.4+1.7+0.1
Women (n = 524)**        
 Plate        
  MT+0.40+2.6+0.2−2.1−0.2−12−1.1
  cMF+0.20+3.9+0.3−7.6−0.6−23−1.9
  LT+0.90−0.10−13−1.1−28−2.5
  cLF−0.3+0.1+8.1#+0.7+5.2+0.4−5.3−0.4
 Subregion        
  Central MT+1.6+0.1+3.5+0.3−4.4−0.3−16−1.2
  External MT+0.30+2.7+0.2−7.1−0.6−36−3.0
  Internal MT−0.8−0.1+3.2+0.3+0.9+0.1+2.4+0.2
  Anterior MT−1.0−0.1+5.0+0.4+1.8+0.1−2.4−0.2
  Posterior MT+1.3+0.1−1.1−0.1−1.5−0.1−12−1.0
  Central cMF−1.4−0.1−0.8+0.1−14−1.0−32−2.3
  External cMF+5.5+0.4+12#+0.9+1.0+0.1−27−1.9
  Interior cMF−1.3−0.1+2.7+0.2−6.0−0.5−11−0.8
  Central LT+2.1+0.1−0.10−18−1.1−37−2.3
  External LT−2.2−0.2+0.50−11−0.9−31−2.7
  Internal LT+1.8+0.1−2.8−0.2−15−1.0−34−2.3
  Anterior LT+0.8+0.1+6.8+0.5+0.50−5.2−0.4
  Posterior LT+0.9+0.1−3.2−0.2−15−1.1−26−1.8
  Central cLF00+8.9#+0.6+4.8+0.3−4.8−0.3
  External cLF+2.0+0.2+10#+0.8+6.9+0.6−11−0.9
  Internal cLF−2.7−0.2+5.6+0.4+4.8+0.4+0.20
Table 4. Differences in cartilage thickness per grades of JSN of the medial/lateral compartment*
 Medial JSN grade 1Medial JSN grade 2Lateral JSN grade 1Lateral JSN grade 2
Difference, %Mean Z scoreDifference, %Mean Z scoreDifference, %Mean Z scoreDifference, %Mean Z score
  • *

    Standard scores (Z scores) determined in comparison with the means and SDs of the reference cohort (crude values presented in Table 1). See Table 1 for definitions.

  • For medial JSN grade 1, n = 126, for medial JSN grade 2, n = 46, for lateral JSN grade 1, n = 60, and for lateral JSN grade 2, n = 18.

  • P < 0.0025 in univariate analysis and after adjustment for BMI.

  • §

    P < 0.0025 in univariate analysis.

  • For medial JSN grade 1, n = 172, for medial JSN grade 2, n = 36, for lateral JSN grade 1, n = 116, and for lateral JSN grade 2, n = 26.

Men        
 Plate        
  MT−4.5−0.4−26−2.3−1.6−0.1+3.3+0.3
  cMF−9.9§−0.7−50−3.6−5.4−0.4+7.0+0.5
  LT−9.6−0.8−11§−0.9−21−1.8−55−4.7
  cLF+3.0+0.2+12+0.9+1.1+0.1−29−2.1
 Subregion        
  Central MT−7.1−0.5−32−2.2−2.8−0.2+6.4+0.4
  External MT−19−1.4−71−5.1−12−0.9−3.6−0.3
  Internal MT+6.1+0.6+1.0+0.1+5.3+0.5+5.6+0.6
  Anterior MT−2.0−0.2−9.5−0.8+2.1+0.2+7.7+0.6
  Posterior MT−0.8−0.1−22−1.5−1.8−0.1−1.4−0.1
  Central cMF−17−1−64−3.9−12−0.700
  External cMF−3.2−0.2−64−4.5+1.2+0.1+23+1.6
  Internal cMF−6.7−0.5−26−1.8−2.8−0.2+4.0+0.3
  Central LT−9.0−0.6−8.3−0.5−29−1.8−68−4.3
  External LT−5.6−0.500−15−1.3−71−6
  Internal LT−14−1.1−23−1.8−27−2.2−51−4.1
  Anterior LT−1.9−0.1−2.1−0.1−2.2−0.1−24−1.6
  Posterior LT−15−1−18−1.3−27−1.9−53−3.6
  Central cLF+2.9+0.2+11+0.700−28−1.8
  Exterior cLF+2.0+0.1+18+1.2+1.0+0.1−51−3.4
  Interior cLF+4.5+0.4+6.2+0.5+2.7+0.2−10−0.8
Women        
 Plate        
  MT−4.6−0.4−25−2.4−1.0−0.1+5.3+0.5
  cMF−12−1−48−4−2.7−0.2+8.7+0.7
  LT−8.6−0.8−9.2§−0.8−23−2−54−4.8
  cLF+5.8+0.5+5.3+0.4+2.8+0.2−22−1.8
 Subregion        
  Central MT−7.4−0.6−32−2.5−2.8−0.2+6.8+0.5
  External MT−12−1−66−5.6−4.3−0.4+5.6+0.5
  Interior MT0000+1.5+0.1+4.0+0.3
  Anterior MT−1.1−0.1−14−1.1+4.3+0.3+11§+0.9
  Posterior MT−2.6−0.2−18−1.5−3.1−0.3−1.3−0.1
  Central cMF−20−1.4−63−4.6−7.7−0.6+8.0+0.6
  External cMF−3.2−0.2−58−4.2+5.7+0.4+15§+1.1
  Interior cMF−9.0−0.7−24−1.9−2.5−0.2+6.4+0.5
  Central LT−10§−0.6−8.8−0.6−34−2.2−75−4.8
  External LT−7.1−0.6−2.9−0.3−18−1.6−69−6
  Internal LT−12§−0.8−26−1.8−25−1.7−47−3.2
  Anterior LT+2.4+0.2+9.6§+0.7−3.8−0.3−27−2.1
  Posterior LT−14−1−16§−1.1−24−1.7−40−2.8
  Central cLF+5.5+0.4+6.5+0.4+1.9+0.1−23−1.6
  Exterior cLF+8.2+0.7+11§+0.9+4.6+0.4−45−3.7
  Interior cLF+4.8+0.4−1.2−0.1+3.1+0.200

Medial and lateral JSN.

The tAB was generally larger in knees with JSN than in healthy reference knees (Table 5). In knees with medial JSN, the medial/lateral ratio for the tAB difference between radiographic OA and healthy knees became larger, with higher medial JSN grades in men (1.5 and 3.3 for medial JSN grades 1 and 2, respectively) and in women (1.3 and 1.6, respectively), whereas it was unchanged or lower at higher lateral JSN grades (1.2 and 1.2 for men and 1.0 and 0.5 for women, respectively).

Table 5. Differences in tAB (mm2) per medial/lateral JSN grade*
RegionMedial JSN grade 1Medial JSN grade 2Lateral JSN grade 1Lateral JSN grade 2
Difference, %Mean Z scoreDifference, %Mean Z scoreDifference, %Mean Z scoreDifference, %Mean Z score
  • *

    Mean differences in percentage and standard scores (Z scores) were determined in comparison with the means and SDs of the reference cohort (see Table 1 for crude values). FT = femorotibial; MFTC = medial FT compartment; LFTC = lateral FT compartment. See Table 1 for additional definitions.

  • For medial JSN grade 1, n = 126, for medial JSN grade 2, n = 46, for lateral JSN grade 1, n = 60, and for lateral JSN grade 2, n = 18.

  • No statistical comparisons were made.

  • §

    P < 0.0125 in univariate analysis.

  • P < 0.0125 in univariate analysis and after adjustment for age and BMI.

  • #

    For medial JSN grade 1, n = 172, for medial JSN grade 2, n = 36, for lateral JSN grade 1, n = 116, and for lateral JSN grade 2, n = 26.

Men (n = 343)        
 Total FT joint+7.4+0.8+7.7+0.8+9.4+1.0+11+1.2
  MFTC+8.8+0.9+12+1.1+10+1.1+12+1.3
  LFTC+5.9+0.6+3.6+0.4+8.5+0.9+9.9+1.0
 MT+7.1§+0.7+11+1.2+8.7+0.9+11+1.2
 cMF+12+1.0+13+1.1+14+1.1+14+1.2
 LT+5.6§+0.6+3.7+0.4+8.2+0.9+9.3+1.0
 cLF+6.4§+0.6+3.4+0.3+8.9+0.8+11+1.0
Women (n = 524)#        
 Total FT joint+8.5+1.0+7.9+0.9+11+1.3+7.6+0.9
  MFTC+9.5+1.0+9.7+1.1+11+1.2+5.3+0.6
  LFTC+7.5+0.8+6.0+0.6+11+1.2+10+1.0
 MT+6.9§+0.7+8.2§+0.9+8.6§+0.9+2.8+0.3
 cMF+15+1.2+13+1.1+16+1.4+10§+0.9
 LT+7.2+0.6+5.2+0.5+11+1.0+9.8§+0.9
 cLF+7.9+0.8+7.2§+0.7+12+1.1+11§+1.0

Medial JSN grade 2 knees displayed considerably thinner cartilage in medial plates than healthy reference knees, with cMF being more strongly affected than MT; cLF (but not LT) displayed a trend toward thicker cartilage (Table 4). In the medial compartment, 6 of the 8 subregions in men and 7 of the 8 subregions in women showed substantially thinner cartilage than healthy reference knees (minimal Z scores −5.1 and −5.6 for external subregion of MT). The internal and posterior subregions of LT, but none of the other lateral subregions, showed considerably thinner cartilage in medial JSN grade 2 knees. Lateral JSN grade 2 knees displayed thinner cartilage in lateral cartilage plates, with LT being more strongly affected than cLF (Table 4). Seven of 8 lateral subregions displayed considerably thinner cartilage in lateral JSN grade 2 (minimal Z scores −6.0 in the external subregion of LT in men and women). In medial JSN grade 1 knees, the regional cartilage was thinner in similarly as many medial as lateral subregions. Interestingly, LT (but not MT) showed substantially lower cartilage thickness in both men and women with medial JSN grade 1 compared with healthy reference knees. In lateral JSN grade 1 knees, 4 of the 8 lateral but none of the medial subregions displayed considerably thinner cartilage than healthy reference knees (Table 4).

Relation between differences in tAB and regional cartilage thickness.

The relationship between the difference in tAB and regional cartilage thickness (Z scores of same regions) was weak in both the medial and lateral compartment (Figure 2) as well as for plates and subregions (−0.03< r >0.25). Although statistically significant in some regions of the knee, the difference in tAB explained <6% of the variation (maximum R2 = 0.06) in the difference in regional cartilage thickness.

thumbnail image

Figure 2. The correlation between the difference in total subchondral bone area and regional cartilage thickness (expressed as Z scores) in A, the medial and B, the lateral compartment for all subjects with preradiographic osteoarthritis (OA) and radiographic OA (calculated Kellgren/Lawrence grade 0–4, n = 867).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

This cross-sectional study compared femorotibial tABs and regional cartilage thickness between healthy reference knees, preradiographic OA knees, and early, late-stage, and end-stage radiographic OA knees of 1,003 participants of the OAI cohort. Compared with healthy reference knees, we observed larger tABs already at the preradiographic OA stage, with the differences not becoming larger with increasing calculated K/L grades. The medial/lateral ratio of tABs became higher with increasing medial JSN grades (in both men and women), and was unchanged (in men) or lower (in women) with increasing lateral JSN grades compared with healthy reference knees. These findings extend previous results showing a relationship between the ratio of medial/lateral tAB and knee alignment (34). In calculated K/L grade 3–4 knees, cartilage thickness in cMF showed greater reductions than in MT, and greater reductions in LT than in cLF, which agrees well (medial compartment) with a previous report, although that report was of a much smaller sample of cases of radiographic OA (35). The smallest Z scores were observed in the external subregion of MT and the central and external subregions of cMF in medial radiographic OA, and in the external and central subregions of LT in lateral radiographic OA. In preradiographic OA and early radiographic OA, however, substantially thicker cartilage was observed in the majority of subregions (specifically in the external subregion of cMF). With medial JSN, cMF (particularly the central subregion of cMF), MT (particularly the external subregion of MT), and LT (particularly the posterior and internal subregions of LT), but not cLF showed significantly thinner cartilage than healthy reference knees, whereas with lateral JSN, only the lateral regions (particularly the external and central subregions of LT) displayed thinner cartilage. The early differences in tAB and later-stage differences in regional cartilage thickness as well as the lack of correlation between the two variables found in this study support the previous suggestion that these variables provide independent information in OA (9).

A limitation of the current study is the cross-sectional design, in which assumptions about the course of OA disease only can be inferred indirectly. Longitudinal trials investigating the structural development of radiographic knee OA, ranging from a state of healthy unaffected to end-stage knees, however, require extensive observation periods (>10 years) and an enormous sample size, as it is unclear who will develop radiographic OA. A strength of this study is the large sample size (except for preradiographic OA cases, in which calculated K/L grades 0 and 1 were grouped together) and the inclusion of all radiographic OA stages. Another limitation is the use of calculated K/L grades assigned by different readers; however, central K/L readings are not yet available for the entire sample studied here.

After adjusting for BMI and age, we found that tABs were substantially larger in both preradiographic OA and radiographic OA knees than in healthy knees, with the magnitude of differences being in agreement with previous results (11). A recent report suggested that lower JSN grades (specifically, grade 1) were not associated with increased subchondral bone size, but higher JSN grades (grade 3 in the tibia and grades 2 and 3 in the femur) were (36). However, the latter report compared 1 knee without JSN with the contralateral knee with JSN in subjects with radiographic knee OA, whereas the current study used healthy reference knees for comparison. We are not aware of any reports exploring the relationship between tAB changes and bone structure; however, a loss of mineralized bone was recently found to be related to cartilage degeneration (37).

Our findings agree with others in indicating that an increased tAB may be an early indicator of OA disease (12), and in the context of other previous reports (34, 38), this could be due to a higher challenge of mechanical load absorption early in the radiographic OA process (or even before the development of radiographic OA). Still, this needs to be confirmed in properly designed longitudinal trials. Increases in tAB do not appear to be dependent on the presence of osteophytes, because differences were similar for knees with (calculated K/L grade 2) and without (calculated K/L grade 0/1) definite osteophytes. Preradiographic OA knees in this study were, however, from participants with definite radiographic OA in the contralateral knee, and enlargement of the tAB may have been driven by systemic “cross talk” between both knees. Alternative explanations are altered loading conditions due to pain in the contralateral knee, or that both knees shared larger tABs before OA developed in one of them.

Cartilage swelling was suggested to precede cartilage breakdown in anterior cruciate ligament (ACL)–transected dogs (39). Further, regional increases in cartilage thickness were reported as a potential early sign of OA in human knees with ACL rupture and in knees with early radiographic OA (40, 41). At the total plate level, the difference in cartilage thickness between preradiographic OA/early radiographic OA and healthy reference knees was <1 SD in this study, although most plates showed a consistent trend toward thicker cartilage. The external subregion of cMF displayed substantially (and statistically significant) thicker cartilage in calculated K/L grade 2 knees in both sexes, and a trend toward greater thickness was observed already at the preradiographic OA stage. Interestingly, this specific location agrees with other recent cross-sectional and particularly with recent longitudinal observations of cartilage swelling in early radiographic OA (11, 35, 41).

In longitudinal trials, cartilage loss was shown to be more prominent in cMF than MT, and in LT than in cLF (16, 17); the same spatial pattern was identified here. A recent meta-analysis identified the central and external subregions of cMF, the external subregion of MT, and cMT as the regions of the largest longitudinal cartilage thinning medially, and the central, internal, and posterior subregions of LT as the regions of the largest longitudinal cartilage thinning laterally (42). Our cross-sectional findings in late- and end-stage knees agree well with these patterns. At end-stage radiographic OA, we found small Z scores, indicating considerable loss of cartilage, but even in the leading subregion (the external subregion of MT), ≥25% of the assumed initial cartilage thickness was maintained. Surprisingly, internal MT did not display a reduction in cartilage thickness even with medial JSN grade 2, and internal cLF displayed only minor reductions with lateral JSN grade 2 compared with healthy reference knees. These findings suggest that cartilage thickness is almost fully maintained in some subregions of the ipsilateral compartment with end-stage JSN. This may provide an opportunity to measure further longitudinal cartilage loss with MRI, even after the “dynamic window” for measuring progression with radiographs has closed, provided that cartilage loss does occur in these regions after reaching end-stage radiographic OA. Consistent cartilage thinning across all radiographic OA stages was found in the internal and posterior subregions of LT in both sexes, and these may thus represent subregions of particular interest in longitudinal studies of early radiographic OA.

In conclusion, this cross-sectional study suggests that femorotibial tABs are larger in both preradiographic OA and radiographic OA than in healthy reference knees, with differences not becoming larger at higher calculated K/L grades. Reductions in cartilage thickness in radiographic OA, compared with healthy reference knees, were greater in the femur than the tibia medially, but greater in the tibia than the femur laterally. Specific subregions with substantial cartilage thickening or thinning were identified in preradiographic OA and radiographic OA knees, respectively. Medially, the external subregion of MT and the external and central subregions of cMF showed large reductions in cartilage thickness. Laterally, the external and central subregions of LT had the greatest reductions. Still, substantial average cartilage thickness was maintained in many (and ≥25% of average normal thickness was maintained in all) subregions at end-stage radiographic OA. Thicker cartilage in early radiographic OA was observed in the external subregion of cMF.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Frobell had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Frobell, Nevitt, Hudelmaier, Wirth, Wyman, Benichou, Dreher, Davies, Lee, Baribaud, Gimona, Eckstein.

Analysis and interpretation of data. Frobell, Nevitt, Wirth, Eckstein.

ROLE OF THE STUDY SPONSOR

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

As authors of this article, individual representatives from Chondrometrics, Pfizer, Eli Lilly, Merck Serono, GlaxoSmithKline, Wyeth Research, Centocor, and Novartis Pharmaceuticals were involved in the study conception and design for this analysis of imaging data from the selected subcohort of the OAI data, in writing and editing the manuscript, and in approving the content of the submitted manuscript. The publication of this article was contingent on the approval of these individual researchers (authors) and was also edited and approved by other representatives from Merck Serono, to whom the manuscript was circulated by the Merck Serono coauthor.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

The authors thank the following readers for their dedicated data segmentation: Gudrun Goldmann, Linda Jakobi, Manuela Kunz, Dr. Susanne Maschek, Jana Matthes, Sabine Mühlsimer, Annette Thebis, and Dr. Barbara Wehr.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information
  • 1
    Crema MD, Roemer FW, Marra MD, Guermazi A. Magnetic resonance imaging assessment of subchondral bone and soft tissues in knee osteoarthritis. Rheum Dis Clin North Am 2009; 35: 55777.
  • 2
    Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, et al. Development of criteria for the classification and reporting of osteoarthritis: classification of osteoarthritis of the knee. Arthritis Rheum 1986; 29: 103949.
  • 3
    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: 13816.
  • 4
    Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis 1957; 16: 494502.
  • 5
    Menkes CJ. Radiographic criteria for classification of osteoarthritis. J Rheumatol Suppl 1991; 27: 135.
  • 6
    Eckstein F, Ateshian G, Burgkart R, Burstein D, Cicuttini F, Dardzinski B, et al. Proposal for a nomenclature for magnetic resonance imaging based measures of articular cartilage in osteoarthritis. Osteoarthritis Cartilage 2006; 14: 97483.
  • 7
    Eckstein F, Cicuttini F, Raynauld JP, Waterton JC, Peterfy C. Magnetic resonance imaging (MRI) of articular cartilage in knee osteoarthritis (OA): morphological assessment. Osteoarthritis Cartilage 2006; 14 Suppl A: 4675.
  • 8
    Guermazi A, Burstein D, Conaghan P, Eckstein F, Hellio Le Graverand-Gastineau MP, Keen H, et al. Imaging in osteoarthritis. Rheum Dis Clin North Am 2008; 34: 64587.
  • 9
    Buck RJ, Wyman BT, Le Graverand MP, Wirth W, Eckstein F, for the A9001140 Investigators. An efficient subset of morphological measures for articular cartilage in the healthy and diseased human knee. Magn Reson Med 2010; 63: 68090.
  • 10
    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: 16974.
  • 11
    Hellio Le Graverand MP, Buck RJ, Wyman BT, Vignon E, Mazzuca SA, Brandt KD, et al. Subregional femorotibial cartilage morphology in women: comparison between healthy controls and participants with different grades of radiographic knee osteoarthritis. Osteoarthritis Cartilage 2009; 17: 117785.
  • 12
    Ding C, Cicuttini F, Jones G. Tibial subchondral bone size and knee cartilage defects: relevance to knee osteoarthritis. Osteoarthritis Cartilage 2007; 15: 47986.
  • 13
    Wang Y, Wluka AE, Cicuttini FM. The determinants of change in tibial plateau bone area in osteoarthritic knees: a cohort study. Arthritis Res Ther 2005; 7: R68793.
  • 14
    Eckstein F, Buck RJ, Burstein D, Charles HC, Crim J, Hudelmaier M, et al. Precision of 3.0 tesla quantitative magnetic resonance imaging of cartilage morphology in a multicentre clinical trial. Ann Rheum Dis 2008; 67: 16838.
  • 15
    Wang Y, Ding C, Wluka AE, Davis S, Ebeling PR, Jones G, et al. Factors affecting progression of knee cartilage defects in normal subjects over 2 years. Rheumatology (Oxford) 2006; 45: 7984.
  • 16
    Eckstein F, Maschek S, Wirth W, Hudelmaier M, Hitzl W, Wyman B, et al. One year change of knee cartilage morphology in the first release of participants from the Osteoarthritis Initiative progression subcohort: association with sex, body mass index, symptoms and radiographic osteoarthritis status. Ann Rheum Dis 2009; 68: 6749.
  • 17
    Hunter DJ, Niu J, Zhang Y, Totterman S, Tamez J, Dabrowski C, et al. Change in cartilage morphometry: a sample of the progression cohort of the Osteoarthritis Initiative. Ann Rheum Dis 2009; 68: 34956.
  • 18
    Nevitt MC, Peterfy C, Guermazi A, Felson DT, Duryea J, Woodworth T, et al. Longitudinal performance evaluation and validation of fixed-flexion radiography of the knee for detection of joint space loss. Arthritis Rheum 2007; 56: 151220.
  • 19
    Peterfy C, Li J, Zaim S, Duryea J, Lynch J, Miaux Y, et al. Comparison of fixed-flexion positioning with fluoroscopic semi-flexed positioning for quantifying radiographic joint-space width in the knee: test–retest reproducibility. Skeletal Radiol 2003; 32: 12832.
  • 20
    Schneider E, NessAiver M, White D, Purdy D, Martin L, Fanella L, et al. The Osteoarthritis Initiative (OAI) magnetic resonance imaging quality assurance methods and results. Osteoarthritis Cartilage 2008; 16: 9941004.
  • 21
    Altman RD, Gold GE. Atlas of individual radiographic features in osteoarthritis, revised. Osteoarthritis Cartilage 2007; 15 Suppl A: 156.
  • 22
    Altman RD, Hochberg M, Murphy WA Jr, Wolfe F, Lequesne M. Atlas of individual radiographic features in osteoarthritis. Osteoarthritis Cartilage 1995; 3 Suppl A: 370.
  • 23
    Eckstein F, Wirth W, Hudelmaier MI, Maschek S, Hitzl W, Wyman BT, et al. Relationship of compartment-specific structural knee status at baseline with change in cartilage morphology: a prospective observational study using data from the Osteoarthritis Initiative. Arthritis Res Ther 2009; 11: R90.
  • 24
    Eckstein F, Hudelmaier M, Wirth W, Kiefer B, Jackson R, Yu J, et al. Double echo steady state magnetic resonance imaging of knee articular cartilage at 3 tesla: a pilot study for the Osteoarthritis Initiative. Ann Rheum Dis 2006; 65: 43341.
  • 25
    Wirth W, Nevitt M, Hellio Le Graverand MP, Benichou O, Dreher D, Davies RY, et al. Sensitivity to change of cartilage morphometry using coronal FLASH, sagittal DESS, and coronal MPR DESS protocols: comparative data from the Osteoarthritis Initiative (OAI). Osteoarthritis Cartilage 2010; 18: 54754.
  • 26
    Peterfy CG, Schneider E, Nevitt M. The osteoarthritis initiative: report on the design rationale for the magnetic resonance imaging protocol for the knee. Osteoarthritis Cartilage 2008; 16: 143341.
  • 27
    Wirth W, Hellio Le Graverand MP, Wyman BT, Maschek S, Hudelmaier M, Hitzl W, et al. Regional analysis of femorotibial cartilage loss in a subsample from the Osteoarthritis Initiative progression subcohort. Osteoarthritis Cartilage 2009; 17: 2917.
  • 28
    Eckstein F, Kunz M, Schutzer M, Hudelmaier M, Jackson RD, Yu J, et al. Two year longitudinal change and test–retest-precision of knee cartilage morphology in a pilot study for the Osteoarthritis Initiative. Osteoarthritis Cartilage 2007; 15: 132632.
  • 29
    Wirth W, Eckstein F. A technique for regional analysis of femorotibial cartilage thickness based on quantitative magnetic resonance imaging. IEEE Trans Med Imaging 2008; 27: 73744.
  • 30
    Faber SC, Eckstein F, Lukasz S, Muhlbauer R, Hohe J, Englmeier KH, et al. Gender differences in knee joint cartilage thickness, volume and articular surface areas: assessment with quantitative three-dimensional MR imaging. Skeletal Radiol 2001; 30: 14450.
  • 31
    Otterness IG, Eckstein F. Women have thinner cartilage and smaller joint surfaces than men after adjustment for body height and weight. Osteoarthritis Cartilage 2007; 15: 66672.
  • 32
    Burgkart R, Glaser C, Hinterwimmer S, Hudelmaier M, Englmeier KH, Reiser M, et al. Feasibility of T and Z scores from magnetic resonance imaging data for quantification of cartilage loss in osteoarthritis. Arthritis Rheum 2003; 48: 282935.
  • 33
    Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology 1990; 1: 436.
  • 34
    Eckstein F, Hudelmaier M, Cahue S, Marshall M, Sharma L. Medial-to-lateral ratio of tibiofemoral subchondral bone area is adapted to alignment and mechanical load. Calcif Tissue Int 2009; 84: 18694.
  • 35
    Reichenbach S, Yang M, Eckstein F, Niu J, Hunter DJ, McLennan CE, et al. Does cartilage volume or thickness distinguish knees with and without mild radiographic osteoarthritis? The Framingham Study. Ann Rheum Dis 2010; 69: 1439.
  • 36
    Eckstein F, Wirth W, Hunter DJ, Guermazi A, Kwoh CK, Nelson DR, et al. Magnitude and regional distribution of cartilage loss associated with grades of joint space narrowing in radiographic osteoarthritis: data from the Osteoarthritis Initiative (OAI). Osteoarthritis Cartilage 2010; 18: 7608.
  • 37
    Bolbos RI, Zuo J, Banerjee S, Link TM, Ma CB, Li X, et al. Relationship between trabecular bone structure and articular cartilage morphology and relaxation times in early OA of the knee joint using parallel MRI at 3 T. Osteoarthritis Cartilage 2008; 16: 11509.
  • 38
    Eckstein F, Faber S, Muhlbauer R, Hohe J, Englmeier KH, Reiser M, et al. Functional adaptation of human joints to mechanical stimuli. Osteoarthritis Cartilage 2002; 10: 4450.
  • 39
    Brandt KD, Braunstein EM, Visco DM, O'Connor B, Heck D, Albrecht M. Anterior (cranial) cruciate ligament transection in the dog: a bona fide model of osteoarthritis, not merely of cartilage injury and repair. J Rheumatol 1991; 18: 43646.
  • 40
    Frobell RB, Le Graverand MP, Buck R, Roos EM, Roos HP, Tamez-Pena J, et al. The acutely ACL injured knee assessed by MRI: changes in joint fluid, bone marrow lesions, and cartilage during the first year. Osteoarthritis Cartilage 2009; 17: 1617.
  • 41
    Buck RJ, Wyman BT, Hellio Le Graverand MP, Hudelmaier M, Wirth W, Eckstein F. Osteoarthritis may not be a one-way-road of cartilage loss: comparison of spatial patterns of cartilage change between osteoarthritic and healthy knees. Osteoarthritis Cartilage 2009; 17: 117785.
  • 42
    Eckstein F, Guermazi A, Roemer FW. Quantitative MR imaging of cartilage and trabecular bone in osteoarthritis. Radiol Clin North Am 2009; 47: 65573.

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. ROLE OF THE STUDY SPONSOR
  9. Acknowledgements
  10. REFERENCES
  11. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
ACR_20262_sm_OnlineSuppl.doc97KSupplementary Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.