Abstract
Background The effects of exercise on osteoarthritic cartilage remain elusive.
Objective The objective of this study was to investigate the effect of dynamic in vivo squatting exercise on the magnitude and spatial pattern of acute cartilage responses in people with tibiofemoral osteoarthritis (ie, Kellgren-Lawrence grades 1 and 2).
Design This investigation was a case-control study.
Methods Eighteen people with radiographic signs of doubtful to mild medial tibiofemoral osteoarthritis were compared with 18 people who were middle-aged and healthy (controls). Three-dimensional magnetic resonance imaging was used to monitor deformation and recovery on the basis of 3-dimensional cartilage volume calculations (ie, total volume and volumes in anterior, central, and posterior subregions) before and after a 30-repetition squatting exercise. Three-dimensional volumes were estimated after semiautomatic segmentation and were calculated at 4 time points (1 before and 3 after scans). Scans obtained after the exercise were separated by 15-minute intervals.
Results In both groups, significant deformation was noted in the medial compartment (−3.4% for the femur and −3.2% for the tibia in people with osteoarthritis versus −2.8% for the femur and −3.8% for the tibia in people in the control group). People with osteoarthritis had significant deformation in the lateral femur (−3.9%) and a tendency toward significant deformation in the lateral tibia (−3.1%). From 15 minutes after exercise cessation onward, volume changes were no longer significantly different from the baseline. At all time points, no significant between-group differences were revealed for volume changes. People with osteoarthritis showed a tendency toward slower recovery preceded by larger deformations in entire cartilage plates and subregions. Spatial subregional deformation patterns were similar between groups.
Limitations Generalizability is limited to people with doubtful to mild osteoarthritis and low levels of pain.
Conclusions Tibiofemoral cartilage deformation appeared similar in magnitude and spatial pattern in people who were middle-aged and either had or did not have tibiofemoral osteoarthritis (ie, Kellgren-Lawrence grades 1 and 2). Restoration of volumes required a 15-minute recovery, especially in the presence of osteoarthritic cartilage degeneration.
Clinical guidelines for osteoarthritis (OA) management indicate that exercise is an important component of first-line treatment strategies because of its potential to diminish pain and improve physical function.1–6 However, a weak correlation exists between clinical presentation and structural joint health, especially in the early stages of the OA disease process (ie, Kellgren-Lawrence [K/L] grades 1 and 2).7,8 Because most trials have focused on symptom-related outcomes, the effects of exercise on structural outcomes in joints with OA remain subject to disparity.
Although exercise appears to beneficially affect cartilage integrity in young adults who are healthy, the protective effects of light to moderate (therapeutic) exercise may persist with increasing age in people without radiographic signs of OA or people at risk for progressive radiographic OA (eg, K/L grade 1, previous knee injury or surgery, and occasional knee symptoms).9–17 In people with established radiographic OA (ie, K/L grades 2–4), single-event and long-term intervention trials (alone or combined with diet or glucosamine supplementation) showed beneficial changes or stability in cartilaginous biomarkers for ultrastructural compounds or anti-inflammatory responses.18–22 In contrast, Woollard et al23 reported small cartilage volume changes (up to a loss of 3.8%) in the central medial femur after treatment that included aerobic, strengthening, and flexibility exercises alone or with agility and perturbation. Although the disparity in treatment effects may be attributable to grouping of patients with various radiographic disease stages, characteristics of patients (such as body mass index [BMI] and lower-limb alignment), and differences in cartilage measures or exercise modes,2,23 a concern is that weight-bearing exercise may lead to acceleration of cartilage degradation instead of deceleration.19
Degraded cartilage shows proteoglycan loss and disruption of the collagen fiber network.24–26 These ultrastructural changes affect the mechanical behavior of cartilage. Fibrillation of the collagen network induces a loss of tensile strength and causes decreased cartilage compressive stiffness and increased tissue permeability.25–27 In people at risk for the development of radiographic OA displaying ultrastructural cartilage degeneration (ie, collagen disruption and water accumulation), cartilage showed delayed recovery of volumes after an in vivo running event.28 Maintained deformation and dehydration of cartilage tissue after loading were suggested to increase the vulnerability of cartilage to accelerated degeneration in the presence of repetitive (high)-impact loads.28,29
Although moderate therapeutic exercise that included weight-bearing neuromuscular control and strength exercises, such as a squatting exercise, was shown to beneficially affect physical function and cartilage integrity in people with the early stages of OA development (ie, K/L grades 1 and 2),10,30 these people, in turn, also had an increased risk of accelerated OA progression.31 Therefore, insight into the recovery times required after in vivo weight-bearing exercise in these people may be a first step toward the appropriate design of treatment programs to positively affect cartilage structural integrity and retard disease progression.
Therefore, the purpose of this study was to investigate the effect of a dynamic in vivo weight-bearing squatting exercise on acute cartilage responses in people with K/L grades 1 and 2. To this end, we evaluated in vivo cartilage deformation (ie, magnitude and spatial pattern) and time to recovery in both people with radiographic signs of doubtful to mild OA (ie, K/L grades 1 and 2) and people who were middle-aged and healthy (controls). Although we expected the spatial patterns to be similar between the groups,27 we hypothesized that knee cartilage in people with radiographic signs of doubtful to mild OA would exhibit increased deformation27,32 followed by a slower recovery after the exercise.28,29
Method
Study Design Overview
In this case-control study, in vivo cartilage deformation and recovery after a squatting exercise in people with radiographic signs of OA (ie, K/L grades 1 and 2 and with cartilage defects on magnetic resonance imaging [MRI]) were compared with those in people who were middle-aged and healthy (controls) (ie, K/L grade 0 and without cartilage defects).
Participants
Participants with OA were 18 people (12 men and 6 women) recruited from the Department of Physical Medicine and Orthopedic Surgery, Ghent University, Hospital Campus. Eligibility to participate was based on clinical assessments, medical imaging, and standard questionnaires. Inclusion criteria were clinical and radiographic signs of doubtful to mild medial tibiofemoral OA (ie, K/L grades 1 and 2)33,34 and medial tibiofemoral cartilage defects on MRI (ie, whole-organ MRI score of ≥2).35 All participants had degenerative meniscal tears on MRI. Additionally, participants had to be able to perform the exercise correctly at the time of the study, without substantial discomfort (ie, visual analog scale [VAS] score of <5 cm for pain during the exercise and active knee flexion range of motion of ≥90°). Exclusion criteria were history of knee surgery, including meniscal procedures, arthroplasty, or both; corticosteroid or hyaluronan injections within the 3 months preceding the study; MRI contraindications; and other known joint or bone pathologies. For participants with unilateral disease, the affected knee was investigated. For participants with bilateral radiographic disease, the more affected knee (within K/L grades 1 and 2) was included; when both knees were affected to similar extents, the dominant leg was investigated. Leg dominance was defined as the limb the participant would choose to kick a ball.36–38
Control participants were 18 people who were middle-aged and recruited from the community or university campus. Eligibility was verified with medical imaging and standard questionnaires. Inclusion criteria were no radiographic signs of OA and no cartilage defects on MRI. Additionally, control participants were selected on the basis of similar physical activity levels (ie, scores on the Baecke questionnaire28,37–39) and in similar proportions with regard to sex and limb dominance. Exclusion criteria were a history of knee pain, knee injury, or both, including a previous diagnosis of cartilage defects; previous knee surgery; BMI of greater than 30 kg/m2; and age younger than 40 years and older than 60 years. In this way, the risk of cartilage abnormalities on MRI with increasing age (even in the presence of normal radiographic appearances)40 was reduced. Additional exclusion criteria were known bone pathologies, joint pathologies, or both (eg, presence of bone marrow lesions or displaced meniscal tears or complete degeneration on MRI41) and MRI contraindications.
Informed consent was obtained from all participants. Participant characteristics are shown in Table 1.
Participant Characteristicsa
Setting and Experimental Procedures
All experimental procedures were performed during one test appointment. All participants were instructed to not practice sports on the day before testing or the day of testing and to avoid running, lifting heavy weights, and taking stairs for 4 hours preceding the actual experimental procedures.28,36–38,42 The procedures were performed on the hospital campus at the same time of day for all participants.28,36,38 The protocol comprised MRI evaluation for in vivo deformation and recovery, evaluation of lower limb function and knee alignment, and questionnaires.
MRI evaluation of cartilage.
Cartilage deformation and recovery were registered by monitoring cartilage quantitative morphology (ie, 3-dimensional [3D] volumes) before and after an in vivo weight-bearing exercise.28,36,38 High-resolution images of cartilage morphology were acquired by means of a sagittal 3D double-echo steady-state sequence with water excitation (3D DESS WE). Additionally, to determine eligibility for inclusion, a fat-saturated turbo–spin-echo (TSE) sequence with intermediate weighting was included next to the 3D DESS WE sequence at the baseline, allowing for grading of cartilage with the whole-organ MRI score.35 Finally, a T2 map (MapIt, Siemens Medical Solutions, Erlangen, Germany) was included. T2 relaxation times depict ultrastructural changes in the collagen and water contents of the cartilage matrix. Higher T2 values are associated with early degeneration even before macroscopic changes are present and were investigated to estimate the presence of insidious cartilage disease in conjunction with the macromorphological appearance of the cartilage surface.43
T2 maps were centered on the tibiofemoral compartments and were reconstructed online with a pixel-wise, monoexponential, nonnegative least squares fit analysis (MapIt), enabling instant T2 quantification after image acquisition. All images were obtained with a dedicated 8-channel knee coil and a 3-T Trio Tim magnet (Siemens Medical Solutions). Knee joints were scanned in extension, and neutral rotation was ensured by placement of rigid foam around the lower leg. Supine positioning of participants was standardized on the basis of the position of the knee joint according to the reference points on the knee coil.37 The sequence parameters for 3D DESS WE, the TSE sequence with intermediate weighting, and the T2 map were previously described.28
To reduce interference from residual deformation preceding the experiment, the MRI protocol started with a 1-hour physical rest period with the participants in a supine position.28,36,38,44 After the rest period, baseline scans (tpre: baseline sagittal 3D DESS WE, T2 map, and TSE sequence with intermediate weighting) were obtained, and then the weight-bearing exercise under study was performed. Sagittal 3D DESS WE scans were obtained within 90 seconds after exercise cessation (tpostt0),36,38 at 15 minutes after tpostt0 (tpostt15), and at 30 minutes after tpostt0 (tpostt30). Deformation was expressed as the 3D volume change measured at tpostt0 relative to the baseline: [(3D volume at tpostt0 − 3D volume at tpre)/3D volume at tpre] × 100. The morphological changes measured at tpostt15 and tpostt30 relative to the baseline were considered to represent recovery.28,38 The sequence of events is displayed in Figure 1.
Schematic overview of the sequence of events during the magnetic resonance imaging experimental protocol. 1–3=postexercise scans obtained within 90 seconds after exercise cessation (tpostt0), at 15 minutes after tpostt0, and at 30 minutes after tpostt0, respectively; bpm=beats per minute; 3D DESS WE=3-dimensional double-echo steady-state sequence with water excitation; TSE=turbo–spin-echo (sequence). Adapted with permission from Van Ginckel et al.28,38
The exercise consisted of 30 bilateral knee bends until the upper leg was lowered to a horizontal position (referenced to the seat of a chair) in 1 minute. To ensure correct and standardized performance, the exercise was carried out under a researcher's supervision and performed barefoot next to the scanner magnet.36,38,44 The exercise speed was set to the pace of a metronome (60 bpm). Visual analog scale scores were collected for the extent of knee pain experienced during the exercise (on a 10-cm scale, with 0 cm representing “no pain at all” and 10 cm representing “extremely painful”). The effect of 30 knee bends on cartilage in adults was previously evaluated with MRI.44–46
Evaluation of lower limb function and knee alignment.
Functional lower limb performance was evaluated with the Five-Times-Sit-to-Stand (FTSTS) Test.47 The FTSTS Test was performed twice, and both the mean time and the best time were used for analysis.
Knee alignment (genu varum or genu valgum) was determined by measuring the intercondylar (IC) or intermalleolar (IM) distance with an inside caliper as previously described.48 The IM distance was subtracted from the IC distance, and the resulting value was considered to be the absolute IC distance. Quantification of the absolute IC or IM distance attained high intertester and intratester reliability values (intraclass correlation coefficients of .95 and .96, respectively)48 and was shown to be valid when compared with full limb radiographs (Bland-Altman plot: R2=.98, P<.001; no correlation between BMI and absolute IC distance [r=−.03, P=.85]).
Questionnaires.
All participants completed the Baecke questionnaire to quantify general physical activity level on the basis of a work, sports, and leisure index39; the Factor Occupational Rating System Scale to rate knee joint load during work situations in particular49; a Likert-scale version of the Western Ontario and McMaster Universities Arthritis Index (WOMAC) to quantify pain, stiffness, and physical function (activities of daily living)50; and the RAND 36-Item Health Survey (RAND-36) to measure quality of life.51 Visual analog scale scores (out of 10) were used to describe the amount of pain experienced during the preceding week, and self-reported duration of knee complaints (in months) was recorded.
Data Analysis
Image analysis: 3D volume calculation.
Three-dimensional reconstruction, volume calculation, and model registration were performed with a commercial modeling software package (Mimics, version 14.0, Materialise NV, Leuven, Belgium).28,36,38
Three-dimensional double-echo steady-state sequence image stacks were segmented to generate a 3D reconstruction of lateral femur, medial femur, lateral tibia, and medial tibia cartilage. A semiautomated segmentation procedure was implemented with a 3D LiveWire algorithm52 and slice-by-slice manual correction to digitize cartilage plates by masking. A region-growing algorithm to dispose of abundant voxels was applied before manual correction. Three-dimensional cartilage plates were reconstructed, and absolute 3D volumes (in mm3) were calculated for baseline and postexercise scans.28,36,38 In addition to the calculation of total volumes at all time points, subregional tibiofemoral volumes were determined to investigate spatial deformation patterns (eg, at tpostt0).27 As defined in the cartilage whole-organ MRI score system,35 femoral and tibial cartilage plates were divided into anterior, central, and posterior subregions (anteromedial femur, centromedial femur, posteromedial femur, anterolateral femur, centrolateral femur, posterolateral femur, anteromedial tibia, centromedial tibia, posteromedial tibia, anterolateral tibia, centrolateral tibia, and posterolateral tibia). An illustration of the division into subregions is shown in Figure 2.
Illustration of the subregions used in this study, as defined in the whole-organ MRI score system.35 The femoral and tibial surfaces were divided into anterior (A), central (C), and posterior (P) regions. Region A of the femur corresponds to the patellofemoral articulation, region C corresponds to the weight-bearing surface, and region P corresponds to the posterior convexity that articulates only in deeper flexion. Region C of the tibial surface corresponds to the uncovered portion between the anterior and posterior horns of the meniscus centrally and the portion covered by the body of the meniscus peripherally.35 Images of the 3-dimensional reconstructions are screen shots taken from the Mimics software interface.
All image analyses were performed by a single researcher who had 4 years of experience at the time of analysis and who was unaware of the time sequence of scanning.28,36,38,53 On the basis of 3 repetitions for all cartilage plates, the intratester reliability values (intraclass correlation coefficients) for the 3D volumetric measurements were .96 to .99 in 3 control participants and .92 to .99 in 3 participants with OA, and the precision errors (root-mean-square coefficients of variation) were .02 to .03 in both groups of participants.
Power analysis.
For participants with various K/L grades and ultrastructural cartilage degeneration, the mean (standard deviation) morphological changes after an in vivo load ranged from −1.8% (3.0%) to −7.9% (11.0%).27,28,32 Attaining the smallest difference with a statistical significance (α) of less than .05 and standard power required the inclusion of at least 24 participants in the entire group. However, the between-group differences were expected to range from 0.1% to approximately 4.5%.27,28,32 In view of our precision errors (which were consistent with precision errors reported in the relevant literature27,54), the between-group differences needed to reach approximately 3% to be relevant in the present study. Detecting this difference required the inclusion of at least 16 participants in each group. The power analysis was performed with Gpower (version 3.1.5, Universität Kiel, Kiel, Germany).
Statistical analysis.
The Shapiro-Wilk test revealed a parametric distribution (P>.05) for all included variables except age, WOMAC total scores, all RAND-36 items, VAS scores for pain during the preceding week and pain during the study exercise, and FTSTS Test best and mean times. Parametric and nonparametric statistics were executed, and data are presented as means (standard deviations) and medians (25th percentile, 75th percentile, and interquartile range), respectively. To investigate baseline differences in group characteristics, we applied the t test for independent samples or the Mann-Whitney U test. To test the hypothesis that the morphology of all cartilage plates changed significantly over time within and between groups, we applied a general linear model for repeated measures with time and cartilage plate as the within-subject factors and participant group allocation (participants with OA and control participants) as the between-subject factor. The model corrected for the main confounding factors BMI and age as covariates. Bonferroni corrections were used to adjust P values for multiple comparisons of main effects. The level of significance (α) was set at less than .05, and SPSS (version 21, IBM Statistics, Armonk, New York) was used for all analyses.
Role of the Funding Source
This study was funded by the Research Foundation of Flanders (FWO Vlaanderen).
Results
Group Characteristics: Demographics, Symptoms, and Function
The group characteristics are shown in Table 1. No significant between-group differences were found for the Baecke physical activity level (P=.93), the Factor Occupational Rating System Scale (P=.10), the FTSTS Test mean time (P=.12) and best time (P=.06), and knee alignment (P=.26). The WOMAC standardized total scores were significantly lower in participants with OA than in control participants (P<.001), as were scores for all RAND-36 items except for role limitations emotional health (P=.32), general health (P=.21), and health change (P=.44). Control participants were younger and had a lower BMI than participants with OA (P<.001 and P=.02, respectively). For participants with OA, VAS scores for pain during the preceding week revealed mild to moderate discomfort; the median (25th percentile, 75th percentile, interquartile range) score was 2.8 (0.0, 5.0, 5.0) cm. The mean (standard deviation) duration of self-reported knee complaints was 40.36 (31.8) months.
In Vivo Cartilage Deformation and Recovery: Percent 3D Volume Changes After Exercise
For the entire sample (N=36), the squatting exercise effected significant deformation relative to the baseline; the mean (standard deviation) reductions in 3D volumes at tpostt0 were −3.3% (3.6%) for the lateral femur (P<.001), −3.1% (4.0%) for the medial femur (P<.001), −2.2% (4.5%) for the lateral tibia (P=.02), and −3.5% (3.6%) for the medial tibia (P<.001). None of the plates showed significant volume decreases at the recovery time points (tpostt15 and tpostt30).
For the control participants (n=18), relative to the baseline, none of the morphological changes at all postexercise time points differed significantly in the lateral femur (P=.10, P=1.00, and P=1.00 for tpostt0, tpostt15, and tpostt30, respectively) or the lateral tibia (P=.73, P=1.00, and P=1.00, respectively). In the medial femur and the medial tibia, only changes measured at tpostt0 (ie, deformation) were significantly different from the baseline; the mean (standard deviation) changes were −2.8% (4.6%) (P=.04) and −3.2% (3.9%) (P=.01), respectively.
For the participants with OA (n=18), changes measured at tpostt0 differed significantly from the baseline for all plates, except for the lateral tibia; the mean (standard deviation) changes were −3.9% (3.5%) (P=.001), −3.4% (3.2%) (P=.02), and −3.8% (3.3%) (P=.01) for the lateral femur, medial femur, and medial tibia, respectively. There was a tendency toward significance for the lateral tibia; the change was −3.1% (4.6%) (P=.05). After completion of the squatting exercise, the participants with OA reported no to mild knee pain; the median (25th percentile, 75th percentile, interquartile range) VAS score was 1.0 (0.4, 3.3, 2.9) cm.
No significant between-group differences were revealed. For all plates and all time points, percent changes and confounding factor–adjusted P values are shown in Table 2.
In Vivo Cartilage Deformation and Recovery Revealed by Three-Dimensional Volume Changes After Exercisea
In Vivo Cartilage Spatial Deformation Patterns: Subregional Analysis of Percent 3D Changes at tpostt0
In both groups of participants, 3D volumes were significantly smaller in all subregions at tpostt0 than at the baseline, with the largest deformation being noted in the posterior femoral condyles and anterior tibial plateaus. On the basis of the magnitude of the mean subregional volume decreases, similar spatial deformation patterns were observed in both groups (posteromedial femur > anteromedial femur = centromedial femur, posterolateral femur > centrolateral femur > anterolateral femur, anteromedial tibia > posteromedial tibia > centromedial tibia, and anterolateral tibia > posterolateral tibia > centrolateral tibia in participants with OA and posteromedial femur > anteromedial femur = centromedial femur, posterolateral femur = centrolateral femur > anterolateral femur, anteromedial tibia > posteromedial tibia > centromedial tibia, and anterolateral tibia > posterolateral tibia > centrolateral tibia in control participants). For all plates, subregional percent changes and confounding factor–adjusted P values are shown in Table 3.
In Vivo Cartilage Deformation Patterns Revealed by Subregional Analysis of Three-Dimensional Volume Changesa
Discussion
The main purpose of the present study was to investigate tibiofemoral cartilage deformation and recovery after a 30-repetition squatting exercise in participants with osteoarthritic cartilage degeneration (ie, up to radiographic signs of mild OA; K/L grades 1 and 2) and participants who were middle-aged and healthy (ie, no radiographic signs of OA and no cartilage defects on MRI). The principal finding was that, despite a tendency toward more deformation in the participants with OA, no significant differences between the groups in volume decreases immediately after the exercise were revealed. Additionally, similar spatial deformation patterns were observed in both groups. Interestingly, recovery tended to occur more slowly in participants with OA, requiring at least 15 minutes after exercise cessation for all cartilage plates to return to baseline volumes.
In the present study, mean cartilage deformation in the tibiofemoral compartments in participants with OA ranged from −3.1% to −3.9%. To the best of our knowledge, this is the first report on the effects of an in vivo weight-bearing dynamic exercise on the deformation behavior of human osteoarthritic cartilage. Two previous studies of patient populations with K/L grades 2 to 427 and K/L grades 2 and 332 examined tibiofemoral morphological changes after a static load was applied to a knee flexed 20 degrees. Relative changes ranged from +1.92% to −7.85%. Static loading has been described as conveying more deformation than dynamic loading; this factor may explain the broader range of outcomes observed in the static loading experiments. Gradually applied static loads allow cartilage deformation responses to adapt more easily to the imposed load, leading to larger deformations of tissue without a considerable pressure surge within its matrix.36,44,45,55,56 In vitro experiments with healthy and osteoarthritic cartilage revealed that dynamic intermittent loading protocols may upregulate matrix synthesis; in contrast, static and injurious impacts tend to decrease the production of matrix compounds and to stimulate protease activity, exerting a deleterious effect on cartilage quality.56–58 Therefore, in view of clinical practice, we preferred to investigate dynamic exercise in the present study.
In the control participants, mean 3D volume decreases of −1.4% to −3.2% were observed. In young adults, a similar exercise yielded mean 3D volume changes of +0.1% to −3.9% in the tibiofemoral compartments.44–46 Interestingly, the deformation outcomes for both control participants and participants with OA were within the ranges established in young adults. In the present study, the mean difference between the groups at deformation was 1.7%, which did not meet the required difference of approximately 3%. However, we noted a tendency toward more deformation in the participants with OA, especially in the lateral femur and the lateral tibia.
Although not involved on radiography, baseline biochemical T2 maps showed higher T2 values in the lateral femur and the lateral tibia: 37.4 (SD=4.0) milliseconds in the lateral femur and 27.4 (SD=4.8) milliseconds in the lateral tibia in control participants versus 40.1 (SD=5.9) milliseconds in the lateral femur and 32.3 (SD=6.2) milliseconds in the lateral tibia in participants with OA. Higher T2 values are associated with early degeneration even before macroscopic changes are present.43 An ex vivo study of unicompartmental OA confirmed that cartilage in unaffected compartments was mechanically inferior to normal cartilage despite sound clinical, radiographic, and morphological appearances.59 Hence, the tendency toward ultrastructural deterioration in these lateral compartments may have brought about the larger volume decreases immediately after the exercise.
Interestingly, although the present study included people with radiographic signs of medial compartment OA, it did not reveal between-group differences for the medial cartilage plates. In contrast, in previous in vivo static loading experiments, the medial compartments in people with OA were driving the larger thickness decreases.27,32 However, the fact that those particular studies included people with K/L grades of at least 2, as opposed to the maximum K/L grade of 2 in the present study, and therefore with more advanced disease, may have enabled more evident differences to be established between groups.
In agreement with Cotofana et al,27 we found similar subregional spatial deformation patterns in healthy and diseased knees27; the largest deformation was observed in the posterior femoral condyles and anterior tibial plateaus. Kinematic analyses showed that during increasing knee flexion, tibiofemoral contact areas shifted to the posterior femur.60,61 Although this observation may explain the femoral spatial deformation patterns in the present study, the anteriorly directed deformation patterns on the tibial plateaus may have resulted from altered tibial rotation during knee flexion in the presence of increasing age and OA.62 In participants who are healthy, next to femoral roll-and-glide motion and tibial valgus, coupled tibial internal rotation accounts for increased anterior and posterior loads on the medial tibial cartilage and lateral tibial cartilage, respectively. In older participants with OA, decreased axial rotation with more apparent diminished rollback of the lateral femur over the tibial plateau has been observed.62 Therefore, tibiofemoral contact may have occurred more anteriorly during flexion movements, increasing the load on the anterior regions of both tibial cartilage plates.
Early recovery encompasses the most important and critical changes after pressure release.38,63 Recovery appeared to be similar in both groups (ie, the mean between-group differences of 0.2%–2.3% did not reach or exceed the required difference of ∼3%). However, the course of volume changes presented in Table 2 suggested a tendency toward slower recovery in participants with OA than in control participants. Recovery required at least 15 minutes for all tibiofemoral cartilage plates, including the lateral knee compartment, to return to baseline morphological status. Delayed recovery may induce a state of maintained deformation and dehydration, which may have deleterious effects on chondrocyte metabolism.28,29 Therefore, hasty load repetitions may induce a negative cycle toward progressive degeneration.
The results of the present study should be interpreted in view of the relatively limited sample size and limited generalizability of the findings. The recruited participants had radiographic signs of doubtful to mild OA, with low levels of pain, and the majority were men.
For the present study, we intentionally did not recruit participants with moderate to severe OA (ie, K/L grades 3 and 4). Because of heterogeneous symptomatic and structural OA presentations, the effects of exercise should be investigated in subgroups rather than in the aggregate group of people with OA.64 Although shifting the focus of OA management to include people at increased risk for OA development or progression as well as people with established disease has been suggested,65 rat models of experimentally induced OA showed that exercise initially led to the suppression of inflammation and the promotion of matrix synthesis. When OA progressed over time, exercise appeared to have effects similar to those in nonexercised joints or appeared to aggravate catabolic responses, promoting joint deterioration.66,67
As indicated by subregional analysis, the squatting exercise in the present study induced general dynamic joint loading, which may have facilitated matrix synthesis.56–58 Although from a clinical point of view this exercise is usually not included in exercise programs for people with advanced and severe OA, unstable knees, malaligned knees, patellofemoral arthritis, or a combination of these conditions because of pain aggravation, it is commonly incorporated in therapeutic programs to rehabilitate neuromuscular control and functional strength in people with meniscal degeneration (eg, the SCOPEX trial, including people with K/L grades 1 and 2)—like those recruited in the present study—or after partial meniscectomy.10,30,68,69 Although in these particular populations of people with doubtful to mild OA, weight-bearing exercises, such as squatting, were shown to improve physical function and potentially cartilage integrity,10,30,69,70 these people, in turn, had an increased risk of accelerated OA progression.31
At the time of the present study, however, participants did not exhibit considerable levels of pain. In view of the clinical presentation of people seeking treatment, the relevance of the investigated population may be questioned. The perception of disease does not correlate well with joint health status,7,8 and symptoms are known to fluctuate over time and to display large interindividual variations.71,72 The constructs for pain intensity (ie, VAS and WOMAC) in the present study were based on a 1-week history, at the most, and the data were collected at the time of the study. Hence, the pain intensity measures did not cover the participants' entire history of symptomatic knee OA. Although the participants were recruited from an outpatient setting at our university hospital—and, therefore, were seeking treatment for their condition—and the mean duration of self-reported knee symptoms was 40 months, the clinical relevance of the population in the present study is supported.
The clinical relevance of the present study is that when weight-bearing exercises, such as squatting, are considered to be clinically feasible and are applied in people who are middle-aged and have doubtful to mild OA, clinicians must be aware of the discordance between symptomatic responses and potentially disproportionate cartilage deformation behavior, which may incite a downward spiral toward accelerated cartilage degeneration. Hence, the results of the present study may have implications for the design of exercise therapy programs for these particular groups of people. Ideally, after a full-weight-bearing 30-repetition squat, people who are middle-aged should allow approximately 15 minutes for tibiofemoral cartilage volumes to recover, especially if they have radiographic signs of doubtful to mild OA. In this way, cartilage recovery can sufficiently protect against progressive deterioration.28,29 Translation of these findings to clinical practice may entail shorter exercise sessions at a higher frequency over the course of the day, alternating between weight-bearing and non–weight-bearing exercises, and alternating use of assisted weight-bearing exercises (such as seated leg presses, assisted weight-bearing squatting under a vertical pulley apparatus, and aquatic exercises). Nonetheless, future research should continue to investigate the long-term effects of structured therapeutic exercise regimens on cartilage structural integrity.
Finally, the sex distribution in the present study did not concur with the typical presentation of OA in the community, in which higher prevalences are recorded in women than in men.73 In people who are younger (<63 years old)—as in the present study (ie, 42–65 years old)—epidemiological reports have conversely described higher prevalences, higher incidences, or both in men, supporting the validity of the population included in the present study in this developmental OA stage.73–75 Nonetheless, the analyses in the present study took sex distribution into account, and the direction of the main results was in agreement with that in studies in which the participants with OA were all women.27
Conclusion
After a 30-repetition squatting exercise, tibiofemoral cartilage deformation appeared to be similar in magnitude (within the measurement error) and spatial pattern in participants who were middle-aged and either had or did not have radiographic signs of doubtful to mild tibiofemoral OA (ie, K/L grades 1 and 2). Restoration of volumes required a 15-minute recovery after the exercise, especially in participants with osteoarthritic cartilage degeneration. In terms of prevention of accelerated OA progression, these results may have implications for dosing and grading of exercise therapy in people who have doubtful to mild OA and for whom weight-bearing exercise is considered clinically feasible.
The Bottom Line
What do we already know about this topic?
Cartilage in joints with osteoarthritis (OA) shows altered mechanical behavior that may increase the vulnerability of the cartilage to accelerated degeneration because of repetitive impact loads.
What new information does this study offer?
After a 30-repetition squatting exercise, tibiofemoral cartilage deformation appeared to be similar in magnitude and spatial pattern in participants who were middle-aged and either had or did not have tibiofemoral OA. Restoration of cartilage volumes to baseline levels required a 15-minute recovery, especially in participants with OA.
If you're a patient, what might these findings mean for you?
After 30 repetitions of full weight-bearing squatting, middle-aged people should allow at least 15 minutes of rest from exercise to permit knee cartilage volumes to recover to pre-exercise levels.
Footnotes
Both authors provided concept/idea/research design, project management, and consultation (including review of manuscript before submission). Ms Van Ginckel provided writing, data collection and analysis, study participants, and facilities/equipment. Dr Witvrouw provided fund procurement and institutional liaisons. The authors gratefully acknowledge Greta Vandemaele, PhD, for implementation of parameters for the magnetic resonance imaging sequences used in this study.
This study was approved by the Ethical Committee of Ghent University Hospital.
This study was funded by the Research Foundation of Flanders (FWO Vlaanderen).
- Received December 7, 2012.
- Accepted April 2, 2013.
- © 2013 American Physical Therapy Association
References
Issue highlights
