SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. ADDITIONAL DISCLOSURES
  9. Acknowledgements
  10. REFERENCES

Objective

To assess several baseline risk factors that may predict patellofemoral and tibiofemoral cartilage loss during a 6-month period.

Methods

For 177 subjects with chronic knee pain, 3T magnetic resonance imaging (MRI) of both knees was performed at baseline and followup. Knees were semiquantitatively assessed, evaluating cartilage morphology, subchondral bone marrow lesions, meniscal morphology/extrusion, synovitis, and effusion. Age, sex, and body mass index (BMI), bone marrow lesions, meniscal damage/extrusion, synovitis, effusion, and prevalent cartilage damage in the same subregion were evaluated as possible risk factors for cartilage loss. Logistic regression models were applied to predict cartilage loss. Models were adjusted for age, sex, treatment, and BMI.

Results

Seventy-nine subregions (1.6%) showed incident or worsening cartilage damage at followup. None of the demographic risk factors was predictive of future cartilage loss. Predictors of patellofemoral cartilage loss were effusion, with an adjusted odds ratio (OR) of 3.5 (95% confidence interval [95% CI] 1.3–9.4), and prevalent cartilage damage in the same subregion with an adjusted OR of 4.3 (95% CI 1.3–14.1). Risk factors for tibiofemoral cartilage loss were baseline meniscal extrusion (adjusted OR 3.6 [95% CI 1.3–10.1]), prevalent bone marrow lesions (adjusted OR 4.7 [95% CI 1.1–19.5]), and prevalent cartilage damage (adjusted OR 15.3 [95% CI 4.9–47.4]).

Conclusion

Cartilage loss over 6 months is rare, but may be detected semiquantitatively by 3T MRI and is most commonly observed in knees with Kellgren/Lawrence grade 3. Predictors of patellofemoral cartilage loss were effusion and prevalent cartilage damage in the same subregion. Predictors of tibiofemoral cartilage loss were prevalent cartilage damage, bone marrow lesions, and meniscal extrusion.

Magnetic resonance imaging (MRI) is the only noninvasive method that is able to directly visualize articular cartilage and has superior sensitivity to radiography in detecting progressive cartilage damage (1, 2). Longitudinal studies of knees with osteoarthritis (OA) have shown that MRI-detected tibiofemoral cartilage loss is associated with older age, female sex, higher body mass index (BMI), African American ethnicity, varus malalignment, a high degree of synovitis, large bone marrow lesions, anterior cruciate ligament (ACL) tears, meniscal tears, and meniscal extrusion (3–11). Risk factors for patellofemoral cartilage loss are partially overlapping, but also seem to be distinct from those for tibiofemoral disease (12–14).

Due to the slowly progressive course of the disease, epidemiologic OA studies require large cohorts that must be followed up over relatively long periods. It would be useful to identify a subgroup of subjects who have no or early disease but who are at high risk of more rapid cartilage loss, as such subjects should be ideal for testing new treatments and should have the greatest need for preventive maneuvers or treatments. Changes that might be induced by a specific therapy on a surrogate end point, such as reduction in cartilage loss, ideally need to also reflect changes in a clinically meaningful end point. Clinical benefit could include improvement in pain or function or a delay in the need for a surgical intervention, such as a total joint arthroplasty.

Using a semiquantitative approach and focusing only on MR image features, Biswal et al found in a small cohort that baseline ACL and meniscal tears and cartilage lesions in the central weight-bearing regions of the medial tibiofemoral compartment were risk factors for more rapid cartilage loss, which the authors did not define in detail, but they found that lesions in these regions were significantly more likely to progress than lesions in other joint compartments (15). Using volumetric cartilage morphometry as the outcome in a patient population with symptomatic OA, Raynauld and coworkers differentiated 3 different subgroups according to the rate of global cartilage volume loss after 24 months. Baseline predictors of faster cartilage volume loss were severe meniscal extrusion, severe medial meniscal tears, bone marrow lesions, high BMI, and age (4). Recently, several demographic and structural risk factors for rapid tibiofemoral cartilage loss over a 30-month period were described in the Multicenter Osteoarthritis Study (MOST) (16). Baseline predictors of rapid cartilage loss included high BMI, prevalent meniscal damage and meniscal extrusion, and presence of synovitis and effusion.

Despite their potential utility, there are few additional data on radiographic or clinical predictors of rapid cartilage loss that would allow investigators to identify high-risk populations. An alternative approach to defining rapid cartilage loss would be to assess knees at short study intervals. It has to be assumed that knees exhibiting any cartilage loss during such a short observational period are losing cartilage at a faster rate than knees not exhibiting any change. Few data assessing knees over short intervals are currently available. A recent study by Hunter and colleagues using quantitative methodology and assessing knees at high risk of progression over a 3–6 month period failed to provide evidence of cartilage loss (17).

The aim of the present study was to identify predictors of patellofemoral and tibiofemoral cartilage loss as defined semiquantitatively by MRI in a cohort of subjects with frequent knee pain over a period of 6 months. We examined a set of putative risk factors for cartilage loss that included age, sex, BMI, and several MR image features.

SUBJECTS AND METHODS

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. ADDITIONAL DISCLOSURES
  9. Acknowledgements
  10. REFERENCES

Study design and subjects.

Subjects were participants in the Joints On Glucosamine (JOG) study, a randomized controlled clinical trial of 201 persons ages 35–65 years with a goal of investigating the possible structural benefit of oral glucosamine hydrochloride. The sample size calculations were based on the primary outcome of decreased development of MRI-detected cartilage damage (18). A flow chart of the included knees is shown in Figure 1.

thumbnail image

Figure 1. Flow chart of the included knees.

Download figure to PowerPoint

Eligible participants had mild-to-moderate knee pain (Western Ontario and McMaster Universities Osteoarthritis Index pain subscale score ≥125 and ≤500 [19]). Study participants were enrolled through a community-based recruitment. Subjects were excluded from the JOG study if they had a Kellgren/Lawrence (K/L) grade of 4 in both knees on fixed-flexion knee radiographs, screened positive for rheumatoid arthritis, had ankylosing spondylitis, psoriatic arthritis, or chronic reactive arthritis, had renal insufficiency that required hemo- or peritoneal dialysis, were taking bisphosphonates or dietary supplements for knee pain in the 6 months prior to study entry, had a history of cancer (except for nonmelanoma skin cancer), had or planned to have bilateral knee replacement surgery, or were unable to walk without assistance. Sample size estimates were based on the primary outcome of cartilage damage as assessed by knee MRI and analysis of urine C-telopeptide of type II collagen. We calculated that 88 subjects would be required in either of the treatment groups (glucosamine or control) to detect a correlation of r = 0.30 between these 2 outcome measures with an alpha level of 0.05 and with 90% statistical power. Allowing for an attrition rate of ∼10–12%, 100 subjects per group would be required.

The baseline and followup MRI examinations of both knees of the 177 subjects who underwent the imaging at both time points were included. Due to previous total knee arthroplasty or the presence of severe joint space narrowing with bone-to-bone contact in either the medial or lateral tibiofemoral compartment in 1 knee on the anteroposterior radiograph (K/L grade 4), 8 participants had only 1 knee scanned, leaving 346 knees that were included in these analyses (not shown).

Radiography.

At screening, all subjects underwent weight-bearing posteroanterior (PA) fixed-flexion knee radiographs. Those with severe joint space narrowing (Osteoarthritis Research Society International grade 3 [20]) in both knees were excluded from study enrollment (n = 40). One experienced musculoskeletal radiologist (AG) graded all fixed-flexion PA radiographs according to the K/L scale (21). Radiographic tibiofemoral OA was considered present if the K/L grade was ≥2.

MRI acquisition.

MRI of both knees was performed on a 3T system (Trio; Siemens) at the Pittsburgh Osteoarthritis Initiative (OAI) clinical site (University of Pittsburgh, Pittsburgh, PA). The MRI pulse sequence protocol was identical to the OAI protocol without the specific sagittal T2 mapping (multiecho spin-echo) and coronal 3-dimensional (3-D) T1-weighted fast low-angle shot water excitation sequences.

MR images were acquired with a dedicated quadrature transmit/receive knee coil using a coronal intermediate weighted 2-D turbo spin-echo sequence (slice thickness 3.0 mm, no slice gap, repetition time [TR] 3,700 msec, time to echo [TE] 29 msec, flip angle 180°, field of view [FOV] 140 mm, matrix 384 × 307 pixels, echo train length [ETL] 7 msec, number of slices 35, 352-Hz pixel bandwidth, number of excitations [NEX] averaged = 1, right/left phase encoding axis, acquisition time 3 minutes 25 seconds), a sagittal 3-D double-echo steady-state (DESS) sequence (slice thickness 0.7 mm, no slice gap, TR 16.3 msec, TE 4.7 msec, flip angle 25°, FOV 140 mm, matrix 384 × 307 pixels, ETL 1 msec, number of slices 35, 185-Hz pixel bandwidth, NEX averaged = 1, anterior/posterior phase encoding axis, acquisition time 10 minutes 23 seconds), coronal and axial multiplanar reformations of the 3-D DESS sequence, and a sagittal intermediate-weighted fat-suppressed fast spin-echo sequence (slice thickness 3 mm, no slice gap, TR 30 msec, TE 3,200 msec, flip angle 180°, FOV 160 mm, matrix 313 × 448 pixels, ETL 5 msec, number of slices 37, 248-Hz pixel bandwidth, NEX averaged = 1, anterior/posterior phase encoding axis, acquisition time 4 minutes 42 seconds). Additional details of the full OAI pulse sequence protocol and the sequence parameters have been published in detail (22).

MRI interpretation.

One musculoskeletal radiologist (FWR) with 9 years experience of standardized semiquantitative assessment of knee OA, blinded to clinical data, read the MR images according to the Whole-Organ Magnetic Resonance Imaging Score (WORMS) method (23) taking into account all 5 available sequences. The following joint structures were assessed in the present study: cartilage morphology and signal, subchondral bone marrow lesions, meniscal status, meniscal extrusion, synovitis, and effusion. Baseline and followup MR images were read paired and with the chronological order known to the reader, an approach that attempts to maximize sensitivity to change as suggested from studies in rheumatoid arthritis and osteoporosis (24). Cartilage signal and morphology were scored according to the WORMS system from 0 to 6 in each of the 5 subregions in the medial and lateral tibiofemoral compartments and 4 subregions of the patellofemoral joint, for a total of 14 subregions. A modification of the WORMS system that has been employed in other longitudinal cohort studies was also used in this study (16).

Limitations of the WORMS scoring system for the assessment of knee MRI have been described (25). In the longitudinal comparison of MRIs from 2 time points, there were instances where the 2 MRIs would fall under the same cartilage grade reading, but a within-grade change between the MRIs was noted. To address this limitation, a partial grade was introduced to reflect a within-grade change that does not account for a full-grade change in cartilage damage. Any partial grade change in at least 1 subregion was defined as the minimum requirement for cartilage loss. Cartilage loss further included incident damage defined as a change from a score of 0 to a score of at least 2 (a superficial focal defect) at followup (Figure 2).

thumbnail image

Figure 2. Patellofemoral cartilage loss. Left, Axial reformatted double-echo steady-state image shows normal cartilage morphology of the anterior femur at baseline. Right, At 6-month followup there is evidence of a full-thickness focal defect in the lateral anterior femoral trochlea (arrow).

Download figure to PowerPoint

Bone marrow lesion size was scored from 0 to 3 based on the extent of subregional involvement. Meniscal morphology was graded from 0 to 4 in the anterior horn, the body segment, and the posterior horn of the medial and lateral meniscus. In addition, meniscal extrusion of the medial and lateral meniscal body was scored on the coronal plane according to previous publications as this feature is not part of the original WORMS score (9, 10). Signal alterations in the infrapatellar and intercondylar regions of Hoffa's fat pad were scored from 0 to 3 as a surrogate for synovial thickening according to the literature as this feature is also not part of the original WORMS system (6). Joint effusion was graded from 0 to 3 in terms of the estimated maximal distention of the synovial cavity (23). The published interreader reliability (weighted kappa) for the readings of the different features for 2 experienced readers (including the reader in the present study) was 0.62 for bone marrow lesions, 0.65 for both synovitis and joint effusion, 0.65 for meniscal extrusion, 0.78 for cartilage morphology, and 0.80 for meniscal status (16).

Statistical analysis.

All MRI-based morphologic predictors were dichotomized into present (≥1) versus absent (0), with the exception of cartilage morphology where absence of cartilage included grades 0 and 1, as grade 1 cartilage damage represents an intrachondral signal change with intact cartilage surface. Logistic regression models were applied to assess the risk of cartilage loss (i.e., the outcome) for knees exhibiting risk factors when compared to knees without the risk factor at baseline. We assessed age, BMI, and sex as possible demographic risk factors (i.e., predictors). Age and BMI were treated as continuous variables. Males were used as the reference group for sex. As MRI-based risk factors (predictors) for both patellofemoral and tibiofemoral joints, we included baseline presence of cartilage damage, subchondral bone marrow lesions, synovitis, and effusion. For the tibiofemoral joint, we also analyzed presence of baseline meniscal damage and extrusion. Risk of cartilage loss in the patellofemoral joint was assessed separately from that in the tibiofemoral joint.

We performed a subregion-based logistic regression analysis using generalized estimating equations (GEEs) to account for the clustering of subregions within a knee and knees within an individual. The odds ratios (ORs) reported can be heuristically interpreted as average ORs for cartilage loss in the presence of a risk factor—averaged over all the relevant subregions and compartments. The bone marrow lesions and prevalent cartilage damage risk factors were subregion specific, and meniscal damage and extrusion were compartment specific, while effusion and synovitis were tibiofemoral or patellofemoral joint-level risk factors in which ORs were taken over the relevant subregions. Multivariate models were adjusted for age, sex, treatment (oral glucosamine), and BMI. In addition, all MRI-based risk factors assessed at baseline were included in the multivariate model. We considered a 2-tailed P value less than 0.05 statistically significant. Joint-specific Bonferroni-adjusted P values for multiple comparisons were also evaluated to control for false positives (4 risk factors for the patellofemoral joints and 6 risk factors for the tibiofemoral joints). All statistical calculations were performed using SAS software for Windows, version 9.2.

RESULTS

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. ADDITIONAL DISCLOSURES
  9. Acknowledgements
  10. REFERENCES

The mean ± SD age at enrollment of the 177 participants who completed the study was 52.3 ± 6.2 years (range 35–65 years). There were slightly more men than women (51.2%), and patients were, on average, overweight (mean ± SD BMI 29.1 ± 4.1 kg/m2). Using the worst K/L grade for either the left or the right knee, at baseline 37 knees (20.9%) had K/L grade 0, 14 knees (7.9%) had K/L grade 1, 26 knees (14.7%) had K/L grade 2, 81 knees (45.8%) had K/L grade 3, and 19 knees (10.7%) had K/L grade 4. The mean ± SD followup period was 172 ± 11.6 days (median 168 days, range 160–269 days). Only 5% of subjects underwent the followup examination after more than 188 days.

The majority of knees exhibited baseline bone marrow lesions and meniscal extrusion, while meniscal tears were seen in ∼15% of knees. Approximately onethird of knees showed any synovitis, and almost 15% exhibited joint effusion of any degree at baseline. The detailed overview of the maximum grades for the different features is presented in Table 1.

Table 1. Prevalence of magnetic resonance imaging features (maximum grades; n = 346 knees)*
 Bone marrow lesionsCartilage damageSynovitisMeniscal tear§EffusionMeniscal extrusion
  • *

    Values are the number (%) of knees. NA = not applicable.

  • Maximum grade of 14 subchondral bone marrow lesion subregions.

  • Maximum grade of 14 cartilage subregions.

  • §

    Maximum grade of 6 subregions.

  • n = 345, as extrusion was not readable for 1 knee due to complete meniscal maceration medially and laterally.

  • #

    For grade 2.0, n = 40 (11.6%); for grade 2.5, n = 14 (4.0%).

Maximum grade      
 092 (26.6)42 (12.1)138 (39.9)203 (58.7)195 (56.4)138 (40.0)
 1110 (31.8)0 (0.0)136 (39.3)22 (6.4)85 (24.6)172 (49.9)
 2100 (28.9)54 (15.6)#63 (18.2)36 (10.4)56 (16.2)35 (10.1)
 344 (12.7)123 (35.5)9 (2.6)77 (22.3)10 (2.9)NA
 4NA14 (4.0)NA8 (2.3)NANA
 5NA84 (24.3)NANANANA
 6NA29 (8.4)NANANANA
Total of knees showing ≥1 lesions254 (73.4)304 (87.9)108 (31.2)43 (12.4)51 (14.7)207 (60.0)

Of the 346 knees, 304 knees (87.9%) and 1,153 subregions (23.8%) exhibited prevalent cartilage damage at baseline. Seventy-nine subregions (1.6%) in 65 knees showed incident or worsening cartilage damage at 6-month followup. Of these, 27 subregions (34.2%) showed full-grade changes and 52 subregions (65.8%) showed within-grade progression. Fifty subregions (63.3%) exhibiting change were in the tibiofemoral joint and 29 subregions (36.7%) exhibiting change were in the patellofemoral joint. In regard to radiographic OA, the majority of subregions exhibiting any cartilage loss were observed in knees with more advanced OA, but knees without radiographic OA also showed cartilage loss. The following distribution of knees with cartilage loss at followup was found for the different K/L grades: for K/L grade 0, 7 knees (10.8%); for K/L grade 1, 5 knees (7.7%); for K/L grade 2, 2 knees (3.1%); for K/L grade 3, 44 knees (67.7%); for K/L grade 4, 7 knees (10.8%). Figure 3 shows an illustrative example of incident tibiofemoral cartilage loss at 6 months that is clearly visualized by 3T MRI and may be categorized by semiquantitative methodology.

thumbnail image

Figure 3. Progressive cartilage damage in the medial tibia. Left, Baseline image shows normal cartilage (Whole-Organ Magnetic Resonance Imaging Score [WORMS] grade 0) in the anterior part of the medial tibial plateau (arrowhead). Right, Six-month followup image depicts diffuse full-thickness cartilage loss (WORMS grade 5) in the same subregion (arrows) and superficial damage (WORMS grade 3) in the central subregion of the medial tibial plateau.

Download figure to PowerPoint

None of the demographic risk factors were predictive of future cartilage loss. Age showed an adjusted OR of 0.99 per 5 years (95% confidence interval [95% CI] 0.72–1.36). For BMI, the adjusted OR for each unit increase was 1.05 (95% CI 0.98–1.13). Women had an adjusted OR of 1.25 (95% CI 0.72–2.18) for cartilage loss when compared to men.

Predictors of patellofemoral cartilage loss at 6 months were baseline presence of effusion (adjusted OR 3.5 [95% CI 1.3–9.4]) and prevalent cartilage damage in the same subregion (adjusted OR 4.3 [95% CI 1.3–14.1]) (Table 2). Age, sex, BMI, and treatment condition (glucosamine) were not associated with cartilage loss in the patellofemoral joint.

Table 2. Risk factors for patellofemoral cartilage loss at 6 months*
Risk factorReferenceAdjusted OR (95% CI) including prevalent cartilage damage in the modelPAdjusted OR (95% CI) omitting prevalent cartilage damage from the modelP
  • *

    OR = odds ratio; 95% CI = 95% confidence interval; WORMS = Whole-Organ Magnetic Resonance Imaging Score; NA = not applicable.

  • Multiadjusted generalized estimating equation model accounting for correlations within and between knees, and adjusted for age, sex, body mass index (BMI), and treatment (age and BMI as continuous variables).

  • Cartilage loss at 6 months in any of 4 patellofemoral subregions.

  • §

    Cartilage loss at 6 months in same subregion.

Effusion (WORMS grade ≥1)Effusion absence (WORMS grade 0)3.54 (1.33–9.37)0.0113.94 (1.55–10.05)0.004
Synovitis (modified WORMS grade ≥1)Synovitis absence (modified WORMS grade 0 in both synovitis subregions)0.84 (0.32–2.21)0.7260.99 (0.38–2.60)0.990
Prevalent cartilage damage in subregion (WORMS grade ≥2)§No cartilage damage in subregion (WORMS grade 0 or 1)4.31 (1.32–14.09)0.016NA
Bone marrow lesions in subregion (WORMS grade ≥1)§No bone marrow lesions in subregion (WORMS grade 0)1.63 (0.67–3.92)0.2802.98 (1.32–6.72)0.008

Risk factors for tibiofemoral cartilage loss were baseline ipsicompartmental meniscal extrusion (adjusted OR 3.6 [95% CI 1.3–10.1]), prevalent bone marrow lesions (adjusted OR 4.7 [95% CI 1.1–19.5]), and cartilage damage in same subregion (adjusted OR 15.3 [95% CI 4.9–47.4]) (Table 3). Age, sex, BMI, and treatment condition (glucosamine) were also not associated with cartilage loss for this region.

Table 3. Risk factors for tibiofemoral cartilage loss at 6 months*
Risk factorReferenceAdjusted OR (95% CI) including prevalent cartilage damage in the modelPAdjusted OR (95% CI) omitting prevalent cartilage damage from the modelP
  • *

    See Table 2 for definitions.

  • Multiadjusted generalized estimating equation model accounting for correlations within and between knees, and adjusted for age, sex, BMI, and treatment (sex and BMI as continuous variables).

  • Cartilage loss in any of 10 tibiofemoral subregions.

  • §

    Cartilage loss in same compartment as meniscal damage or extrusion (5 subregions medial or lateral).

  • Cartilage loss in same subregion.

Effusion (WORMS grade ≥1)Effusion absence (WORMS grade 0)1.76 (0.75–4.13)0.1492.28 (1.02–5.11)0.045
Synovitis (modified WORMS grade ≥1)Synovitis absence (modified WORMS grade 0)0.68 (0.32–1.44)0.3100.86 (0.43–1.77)0.700
Meniscal damage§No meniscal damage1.92 (0.74–4.97)0.1773.72 (1.56–8.89)0.003
Meniscal extrusion§No meniscal extrusion3.60 (1.29–10.07)0.0155.45 (1.91–15.56)0.002
Prevalent cartilage damage in subregion (WORMS grade ≥2)Absence of cartilage damage in subregion (WORMS grade 0 or 1)15.25 (4.90–47.43)<0.001NA
Bone marrow lesions (WORMS grade ≥1)No bone marrow lesions in subregion (WORMS grade 0)4.71 (1.14–19.50)0.03215.64 (3.92–62.47)<0.001

DISCUSSION

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. ADDITIONAL DISCLOSURES
  9. Acknowledgements
  10. REFERENCES

In a cohort of subjects with frequent knee pain and with mixed radiographic disease severity, we identified the presence of baseline effusion and prevalent cartilage damage (in the same subregion) as risk factors for patellofemoral cartilage loss at the 6-month followup. Predictors of tibiofemoral cartilage loss were prevalent cartilage damage, prevalent bone marrow lesions, and meniscal extrusion. Neither age nor BMI nor sex predicted cartilage loss at followup.

Many available studies investigating the progression or incidence of OA have applied joint space width measurements from radiographs as a surrogate outcome for cartilage loss (26, 27). However, as these measures are indirect and reflect not only cartilage status, but also pathology of other intrinsic joint structures, MRI has become the method of choice for cartilage assessment (4, 9, 16, 28–30). Furthermore, joint space narrowing is only detectable over long periods of time, as the annual rate of joint space decrease is estimated to be only ∼0.1 mm per year (26, 31). Although the superiority of MRI is established, the majority of knees will show either no or subtle MRI-detectable cartilage loss in a short followup period, which was confirmed by our results (17, 32–34).

We assessed age, sex, and BMI as baseline demographic risk factors. None of these showed any significant associations with cartilage loss at 6 months. Despite the fact that obesity is one of the few established demographic risk factors for incident radiographic OA and for fast tibiofemoral cartilage loss, we did not find an association between obesity and MRI-detected cartilage loss in our cohort (16, 35, 36). One possible explanation for this discrepant finding could be the fact that our followup period was very short and that demographic factors are more relevant for long-term outcome. Thus, it seems that preexisting morphologic pathology on the joint level might be more influential than demographic risks.

To date, few studies have assessed MRI-defined predictors of longitudinal cartilage loss, and most of those studies investigated knees with established radiographic OA (6, 7, 11, 37, 38). Raynauld and colleagues reported correlations between subchondral bone marrow lesion change and cartilage volume loss in the same compartment (38). Investigators from our group did not find an association of baseline bone marrow lesions with an increased risk of slow or rapid cartilage loss in the MOST study, probably due to a knee-based rather than a subregional approach that takes into account associations between bone marrow lesions and directly adjacent cartilage status in the same subregion as we used in the current analysis (16). A strong association of bone marrow lesions and cartilage loss has been shown longitudinally in previous work using subregional semiquantitative approaches (7, 37). As none of our subjects exhibited a complete ACL tear at baseline, we were not able to include ACL status as a possible predictor as has been reported in other studies (11, 15).

We found a very strong association of baseline meniscal extrusion with consequent cartilage loss in the same compartment, as has been shown previously in subjects with OA or at high risk of OA, but surprisingly, not for baseline meniscal damage (9, 13, 16). There was a trend toward meniscal damage in the fully adjusted multivariate model; however, this was not statistically significant. Not including prevalent cartilage damage as a predictor resulted in an OR of 3.72 for meniscal damage predicting cartilage loss (P = 0.003). Further adjustment by adding prevalent cartilage damage to the model reduced the meniscal damage OR to 1.92 (P = 0.177). This 52% attenuation in the OR gives evidence for mediation, in the sense that cartilage damage is one of the pathways by which meniscal damage affects rates of cartilage loss. Adding baseline cartilage damage “explains” some of the effect of meniscus damage on rates of cartilage loss. Subjects without OA but with baseline meniscal tears show an increased risk of progression of cartilage lesions as compared to a control group without meniscal tears (15). Several other large studies have shown a highly increased risk of incident radiographic OA for knees with baseline meniscal damage or with partial meniscectomy (39, 40). Our results confirm that meniscal extrusion seems to be an independent risk factor for cartilage loss (9, 16) despite the fact that meniscal extrusion is also present in non-OA, asymptomatic knees (41).

To assess the degree of synovial activation of the joint, we used a surrogate of signal alterations in Hoffa's fat pad and assessed joint effusion separately (6, 23). Hill et al recently reported an association of moderate and severe baseline synovitis with an increased risk of cartilage loss in the tibiofemoral compartment at followup (6). In the MOST study, these findings were also observed using a composite measure of synovitis and effusion, demonstrating an increased risk of rapid cartilage loss. However, in that study, the association was not statistically significant after fully adjusting for other risk factors (16).

Risk factors for patellofemoral and tibiofemoral cartilage loss differed, with joint effusion predicting cartilage loss in the patellofemoral joint but not in the tibiofemoral joint. There is no simple explanation for this finding, as joint effusion also seems to predict cartilage loss in the tibiofemoral joint, at least in knees without any prevalent cartilage damage at baseline, over a 30-month period (42). Synovitis was not a significant risk factor, although a definite shortcoming is the application of the synovitis surrogate that we used because only nonenhanced sequences were available. Contrast-enhanced imaging might show different results as the surrogate is only a nonspecific, although sensitive, measure of synovitis (43). Prevalent bone marrow lesions strongly predicted future cartilage loss in the tibiofemoral joint, but not in the patellofemoral joint. We hypothesize that this discordance may be due to the differences in loading, as the weight-bearing tibiofemoral joint seems to be exposed to the effects of loading to a greater extent than the patellofemoral joint (44). This fact appears to be relevant at least for the short observational period of 6 months. Preexisting cartilage damage was one of the strongest predictors of subsequent cartilage loss in the same subregion, which suggests that once cartilage damage is established, further progression is to be expected. One could also hypothesize that prevalent cartilage damage represents an intermediate step for future widespread cartilage loss, with other joint pathology (such as malalignment or meniscal pathology) acting as the primary triggers (13, 39).

We only analyzed each MRI risk factor dichotomized into absence and presence, as limited numbers of subregions showing any cartilage loss at followup did not allow for a detailed subanalysis of high-grade versus low-grade lesions. However, several studies showed a direct association between baseline lesion grade and risk of subsequent cartilage loss (6, 7, 37).

We are not aware of any study that tried to define a time period during which any cartilage loss is to be expected and may be detectable by MRI. We decided to analyze knees that were imaged at baseline and at 6 months. Whether this is the minimum time interval to detect any changes remains to be shown. A recent study by Hunter et al did not find any change from baseline in a 3–6-month interval using similar technology (17). Those investigators attributed this to large test–retest variability that exceeded the sensitivity required to detect change. We note that the results presented by Hunter et al pertain primarily to knees with mean cartilage thickness and not necessarily to knees having higher OA progression at baseline. In knees with more advanced OA progression, the mean changes in cartilage thickness may be larger than the test-retest variability. Our findings support the notion of higher rates of cartilage loss when there is prevalent cartilage damage at baseline. This is an example of the horse-racing effect: the baseline is related to the rate of change (45). In addition, our logistic regression analysis GEE method pooled over many subregions, whereas Hunter et al focused on each specific region. Consequently, our approach potentially had higher power to detect cartilage loss, but at the expense of identifying which subregions were the location of most likely cartilage loss due to the risk factors. The OAI reported only small changes during a 12-month interval using morphometric measurements (32, 33).

A limitation is that we were not able to analyze the full array of possible baseline risk factors or predictors of cartilage loss over short observational periods, as they were not available to us at the time of analysis. Genetic predisposition is probably an important factor to be considered (46, 47); serum biomarkers might also play a relevant role in predicting progression (48). Other factors that were not included but need mentioning are occupational and additional nutritional risk factors (49, 50). Baseline and followup MR images were read paired and with the chronological order known to the reader to maximize sensitivity to change. The intent is to reduce error in the measurement of change. We acknowledge as a limitation the statistical possibility of spurious correlation due to the inclusion of baseline cartilage damage as a predictor when cartilage loss (essentially change in cartilage) is the outcome. In addition, we analyzed each risk factor as a separate predictor together in the multivariate model. In a much larger study sample, a highly increased risk of cartilage loss might be proven for subjects with several concomitant MRI and non-MRI risk factors. Furthermore, although our study sample was chosen based on the presence of knee pain and not radiographic OA or the presence of cartilage lesions, there may be unmeasured risk factors for knee pain (i.e., study eligibility) and cartilage loss that are not related to the semiquantitative 3T MRI–derived risk factors we evaluated, and these may act as confounders in the analysis (51). As a result, some of our reported associations between the measured risk factors and cartilage loss may have been reduced.

In summary, we were able to identify baseline effusion and prevalent cartilage damage as risk factors for cartilage loss in the patellofemoral joint and meniscal extrusion, bone marrow lesions, and prevalent cartilage damage as risk factors for cartilage damage in the tibiofemoral joint over a 6-month period. By including knees with baseline meniscal extrusion (and possibly prevalent meniscal structural damage), bone marrow lesions, and prevalent cartilage lesions, a subpopulation at high risk of progressive cartilage loss in a short time interval may be identified. In our cohort, the majority of knees exhibiting cartilage loss had a radiographic disease severity of K/L grade 3. For patellofemoral disease, joint effusion seems to play an important role in future cartilage loss—a finding that needs further exploration. Our results are another step toward characterizing subjects at risk of progressive cartilage loss, which may be relevant to defining eligibility in clinical trials and epidemiologic studies.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. ADDITIONAL DISCLOSURES
  9. Acknowledgements
  10. REFERENCES

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 published. Dr. Roemer 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. Roemer, Kwoh, Hannon, Jakicic, Moore, Guermazi.

Acquisition of data. Roemer, Hannon, Green, Jakicic, Moore, Guermazi.

Analysis and interpretation of data. Kwoh, Hannon, Jakicic, Boudreau, Crema, Moore, Guermazi.

ROLE OF THE STUDY SPONSOR

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. ADDITIONAL DISCLOSURES
  9. Acknowledgements
  10. REFERENCES

The sponsor of the JOG study, the Beverage Institute for Health & Wellness, the Coca Cola Company, had no role in the study design, data collection, data analysis, or writing of the manuscript. The sponsor agreed to submit the manuscript for publication and gave approval of the content of the submitted manuscript. Publication of the manuscript was not contingent upon the sponsor's approval.

ADDITIONAL DISCLOSURES

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. ADDITIONAL DISCLOSURES
  9. Acknowledgements
  10. REFERENCES

Authors Roemer and Guermazi are vice president and president, respectively, of Boston Imaging Core Lab.

Acknowledgements

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. ADDITIONAL DISCLOSURES
  9. Acknowledgements
  10. REFERENCES

We wish to acknowledge the staff of the Arthritis Research Center, Division of Rheumatology and Clinical Immunology, Department of Medicine, University of Pittsburgh, and the staff of the Magnetic Resonance Research Center, University of Pittsburgh Medical Center. We further wish to thank all participants in the JOG study, without whom this study would not have been possible.

REFERENCES

  1. Top of page
  2. Abstract
  3. SUBJECTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. ADDITIONAL DISCLOSURES
  9. Acknowledgements
  10. REFERENCES
  • 1
    Amin S, LaValley MP, Guermazi A, Grigoryan M, Hunter DJ, Clancy M, et al. The relationship between cartilage loss on magnetic resonance imaging and radiographic progression in men and women with knee osteoarthritis. Arthritis Rheum 2005; 52: 31529.
  • 2
    Bruyere O, Genant HK, Kothari M, Zaim S, White D, Peterfy C, et al. Longitudinal study of magnetic resonance imaging and standard X-rays to assess disease progression in osteoarthritis. Osteoarthritis Cartilage 2007; 15: 98103.
  • 3
    Davies-Tuck ML, Wluka AE, Wang Y, Teichtahl AJ, Jones G, Ding C, et al. The natural history of cartilage defects in people with knee osteoarthritis. Osteoarthritis Cartilage 2008; 16: 33742.
  • 4
    Raynauld JP, Martel-Pelletier J, Berthiaume MJ, Beaudoin G, Choquette D, Haraoui B, et al. Long term evaluation of disease progression through the quantitative magnetic resonance imaging of symptomatic knee osteoarthritis patients: correlation with clinical symptoms and radiographic changes. Arthritis Res Ther 2006; 8: R21.
  • 5
    Cicuttini FM, Wluka AE, Wang Y, Stuckey SL. Longitudinal study of changes in tibial and femoral cartilage in knee osteoarthritis. Arthritis Rheum 2004; 50: 947.
  • 6
    Hill CL, Hunter DJ, Niu J, Clancy M, Guermazi A, Genant HK, et al. Synovitis detected on magnetic resonance imaging and its relation to pain and cartilage loss in knee osteoarthritis. Ann Rheum Dis 2007; 66: 1599603.
  • 7
    Hunter DJ, Zhang Y, Niu J, Goggins J, Amin S, LaValley MP, et al. Increase in bone marrow lesions associated with cartilage loss: a longitudinal magnetic resonance imaging study of knee osteoarthritis. Arthritis Rheum 2006; 54: 152935.
  • 8
    Roemer FW, Guermazi A, Hunter DJ, Niu J, Zhang Y, Englund M, et al. The association of meniscal damage with joint effusion in persons without radiographic osteoarthritis: the Framingham and MOST osteoarthritis studies. Osteoarthritis Cartilage 2009; 17: 74853.
  • 9
    Hunter DJ, Zhang YQ, Niu JB, Tu X, Amin S, Clancy M, et al. The association of meniscal pathologic changes with cartilage loss in symptomatic knee osteoarthritis. Arthritis Rheum 2006; 54: 795801.
  • 10
    Hunter DJ, Zhang YQ, Tu X, LaValley M, Niu JB, Amin S, et al. Change in joint space width: hyaline articular cartilage loss or alteration in meniscus? Arthritis Rheum 2006; 54: 248895.
  • 11
    Amin S, Guermazi A, LaValley MP, Niu J, Clancy M, Hunter DJ, et al. Complete anterior cruciate ligament tear and the risk for cartilage loss and progression of symptoms in men and women with knee osteoarthritis. Osteoarthritis Cartilage 2008; 16: 897902.
  • 12
    Kalichman L, Zhu Y, Zhang Y, Niu J, Gale D, Felson DT, et al. The association between patella alignment and knee pain and function: an MRI study in persons with symptomatic knee osteoarthritis. Osteoarthritis Cartilage 2007; 15: 123540.
  • 13
    Sharma L, Eckstein F, Song J, Guermazi A, Prasad P, Kapoor D, et al. Relationship of meniscal damage, meniscal extrusion, malalignment, and joint laxity to subsequent cartilage loss in osteoarthritic knees. Arthritis Rheum 2008; 58: 171626.
  • 14
    McAlindon T, Zhang Y, Hannan M, Naimark A, Weissman B, Castelli W, et al. Are risk factors for patellofemoral and tibiofemoral knee osteoarthritis different? J Rheumatol 1996; 23: 3327.
  • 15
    Biswal S, Hastie T, Andriacchi TP, Bergman GA, Dillingham MF, Lang P. Risk factors for progressive cartilage loss in the knee: a longitudinal magnetic resonance imaging study in forty-three patients. Arthritis Rheum 2002; 46: 288492.
  • 16
    Roemer FW, Zhang Y, Niu J, Lynch JA, Crema MD, Marra MD, et al. Tibiofemoral joint osteoarthritis: risk factors for MR-depicted fast cartilage loss over a 30-month period in the multicenter osteoarthritis study. Radiology 2009; 252: 77280.
  • 17
    Hunter DJ, Bowes MA, Eaton CB, Holmes AP, Mann H, Kwoh CK, et al. Can cartilage loss be detected in knee osteoarthritis (OA) patients with 3–6 months' observation using advanced image analysis of 3T MRI? Osteoarthritis Cartilage 2010; 18: 67783.
  • 18
    Kwoh CK, Roemer FW, Hannon MJ, Moore CE, Jakicic JM, Guermazi A, et al. The Joints On Glucosamine (JOG) Study: a randomized, double-blind, placebo-controlled trial to assess the structural benefit of glucosamine in knee osteoarthritis based on 3T MRI [abstract]. Arthritis Rheum 2009; 60 Suppl: S7256.
  • 19
    Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988; 15: 183340.
  • 20
    Altman RD, Gold GE. Atlas of individual radiographic features in osteoarthritis, revised. Osteoarthritis Cartilage 2007; 15 Suppl A: A156.
  • 21
    Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis 1957; 16: 494502.
  • 22
    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.
  • 23
    Peterfy CG, Guermazi A, Zaim S, Tirman PF, Miaux Y, White D, et al. Whole-Organ Magnetic Resonance Imaging Score (WORMS) of the knee in osteoarthritis. Osteoarthritis Cartilage 2004; 12: 17790.
  • 24
    Felson DT, Nevitt MC. Blinding images to sequence in osteoarthritis: evidence from other diseases. Osteoarthritis Cartilage 2009; 17: 2813.
  • 25
    Felson DT, Lynch J, Guermazi A, Roemer FW, Niu J, McAlindon T, et al. Comparison of BLOKS and WORMS scoring systems part II. Longitudinal assessment of knee MRIs for osteoarthritis and suggested approach based on their performance: data from the Osteoarthritis Initiative. Osteoarthritis Cartilage 2010; 18: 14027.
  • 26
    Hellio Le Graverand MP, Vignon EP, Brandt KD, Mazzuca SA, Piperno M, Buck R, et al. Head-to-head comparison of the Lyon Schuss and fixed flexion radiographic techniques: long-term reproducibility in normal knees and sensitivity to change in osteoarthritic knees. Ann Rheum Dis 2008; 67: 15626.
  • 27
    Felson DT, Zhang Y, Hannan MT, Naimark A, Weissman BN, Aliabadi P, et al. The incidence and natural history of knee osteoarthritis in the elderly: the Framingham Osteoarthritis Study. Arthritis Rheum 1995; 38: 15005.
  • 28
    Cicuttini F, Hankin J, Jones G, Wluka A. Comparison of conventional standing knee radiographs and magnetic resonance imaging in assessing progression of tibiofemoral joint osteoarthritis. Osteoarthritis Cartilage 2005; 13: 7227.
  • 29
    Buck RJ, Wyman BT, Hellio Le Graverand MP, Hudelmaier M, Wirth W, Eckstein F, for the A9001140 Investigators. Does the use of ordered values of subregional change in cartilage thickness improve the detection of disease progression in longitudinal studies of osteoarthritis? Arthritis Rheum 2009; 61: 91724.
  • 30
    Eckstein F, Wirth W, Hudelmaier M, Stein V, Lengfelder V, Cahue S, et al. Patterns of femorotibial cartilage loss in knees with neutral, varus, and valgus alignment. Arthritis Rheum 2008; 59: 156370.
  • 31
    Hellio Le Graverand-Gastineau MP. OA clinical trials: current targets and trials for OA. Choosing molecular targets: what have we learned and where we are headed? Osteoarthritis Cartilage 2009; 17: 1393401.
  • 32
    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.
  • 33
    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.
  • 34
    Pelletier JP, Raynauld JP, Berthiaume MJ, Abram F, Choquette D, Haraoui B, et al. Risk factors associated with the loss of cartilage volume on weight-bearing areas in knee osteoarthritis patients assessed by quantitative magnetic resonance imaging: a longitudinal study. Arthritis Res Ther 2007; 9: R74.
  • 35
    Spector TD, Hart DJ, Doyle DV. Incidence and progression of osteoarthritis in women with unilateral knee disease in the general population: the effect of obesity. Ann Rheum Dis 1994; 53: 5658.
  • 36
    Felson DT, Zhang Y, Hannan MT, Naimark A, Weissman B, Aliabadi P, et al. Risk factors for incident radiographic knee osteoarthritis in the elderly: the Framingham Study. Arthritis Rheum 1997; 40: 72833.
  • 37
    Roemer FW, Guermazi A, Javaid MK, Lynch JA, Niu J, Zhang Y, et al. Change in MRI-detected subchondral bone marrow lesions is associated with cartilage loss: the MOST Study. A longitudinal multicentre study of knee osteoarthritis. Ann Rheum Dis 2009; 68: 14615.
  • 38
    Raynauld JP, Martel-Pelletier J, Berthiaume MJ, Abram F, Choquette D, Haraoui B, et al. 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 2008; 67: 6838.
  • 39
    Englund M, Guermazi A, Roemer FW, Aliabadi P, Yang M, Lewis CE, et al. Meniscal tear in knees without surgery and the development of radiographic osteoarthritis among middle-aged and elderly persons: the Multicenter Osteoarthritis Study. Arthritis Rheum 2009; 60: 8319.
  • 40
    Englund M, Lohmander LS. Risk factors for symptomatic knee osteoarthritis fifteen to twenty-two years after meniscectomy. Arthritis Rheum 2004; 50: 28119.
  • 41
    Puig L, Monllau JC, Corrales M, Pelfort X, Melendo E, Caceres E. Factors affecting meniscal extrusion: correlation with MRI, clinical, and arthroscopic findings. Knee Surg Sports Traumatol Arthrosc 2006; 14: 3948.
  • 42
    Roemer FW, Guermazi A, Felson DT, Niu J, Nevitt MC, Crema MD, et al. Presence of MRI-detected joint effusion and synovitis increases the risk of cartilage loss in knees without osteoarthritis at 30-month follow-up: the MOST study. Ann Rheum Dis 2011; 70: 18049.
  • 43
    Roemer FW, Guermazi A, Zhang Y, Yang M, Hunter DJ, Crema MD, et al. Hoffa's fat pad: evaluation on unenhanced MR images as a measure of patellofemoral synovitis in osteoarthritis. AJR Am J Roentgenol 2009; 192: 1696700.
  • 44
    Felson DT, McLaughlin S, Goggins J, LaValley MP, Gale ME, Totterman S, et al. Bone marrow edema and its relation to progression of knee osteoarthritis. Ann Intern Med 2003; 139: 3306.
  • 45
    Burrows B, Knudson RJ, Camilli AE, Lyle SK, Lebowitz MD. The “horse-racing effect” and predicting decline in forced expiratory volume in one second from screening spirometry. Am Rev Respir Dis 1987; 135: 78893.
  • 46
    Valdes AM, Loughlin J, Van Oene M, Chapman K, Surdulescu GL, Doherty M, et al. Sex and ethnic differences in the association of ASPN, CALM1, COL2A1, COMP, and FRZB with genetic susceptibility to osteoarthritis of the knee. Arthritis Rheum 2007; 56: 13746.
  • 47
    Kraus VB, Jordan JM, Doherty M, Wilson AG, Moskowitz R, Hochberg M, et al. The Genetics of Generalized Osteoarthritis (GOGO) study: study design and evaluation of osteoarthritis phenotypes. Osteoarthritis Cartilage 2007; 15: 1207.
  • 48
    Garnero P, Peterfy C, Zaim S, Schoenharting M. Bone marrow abnormalities on magnetic resonance imaging are associated with type II collagen degradation in knee osteoarthritis: a three-month longitudinal study. Arthritis Rheum 2005; 52: 28229.
  • 49
    Felson DT, Niu J, Clancy M, Aliabadi P, Sack B, Guermazi A, et al. Low levels of vitamin D and worsening of knee osteoarthritis: results of two longitudinal studies. Arthritis Rheum 2007; 56: 12936.
  • 50
    Felson DT, Hannan MT, Naimark A, Berkeley J, Gordon G, Wilson PW, et al. Occupational physical demands, knee bending, and knee osteoarthritis: results from the Framingham Study. J Rheumatol 1991; 18: 158792.
  • 51
    Zhang Y, Niu J, Felson DT, Choi HK, Nevitt MC, Neogi T. Methodologic challenges in studying risk factors for progression of knee osteoarthritis [review]. Arthritis Care Res (Hoboken) 2010; 62: 152732.