If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Functional adaptation of knee cartilage in asymptomatic female novice runners compared to sedentary controls. A longitudinal analysis using delayed Gadolinium Enhanced Magnetic Resonance Imaging of Cartilage (dGEMRIC)
Address correspondence and reprint requests to: A. Van Ginckel, Department of Rehabilitation Sciences and Physiotherapy, Ghent University, De Pintelaan 185, UZ campus, 3B3, BE-9000 Ghent, Belgium. Tel: 32-(0)9-332-53-74.
Department of Rehabilitation Sciences and Physiotherapy, Ghent University, Ghent, BelgiumDepartment of Rehabilitation Sciences and Physiotherapy, Artevelde University College, Ghent, Belgium
To longitudinally estimate the change in glycosaminoglycan content of knee cartilage in asymptomatic untrained female novice runners participating in a Start To Run program (STR) compared to sedentary controls.
Method
Nine females enrolling in a 10-week STR and 10 sedentary controls participated voluntarily. Prior to and after the 10-week period, both groups were subjected to dGEMRIC imaging. dGEMRIC indices of knee cartilage were determined at baseline and for the change after the 10-week period in both groups. Based on a self-reported weekly log, physical activity change during the study was depicted as decreased, unchanged or increased. The Mann–Whitney U and Kruskal–Wallis tests were applied to test the hypotheses that dGEMRIC changes occurred between groups and according to physical activity changes respectively.
Results
No significant differences were established between groups for dGEMRIC indices at baseline (P=0.541). A significant positive change of the median dGEMRIC index in the runners group was demonstrated when compared to the controls [+11.66 ms (95% CI: −25.29, 44.43) vs −9.56 ms (95% CI: −29.55, 5.83), P=0.006]. The change in dGEMRIC index differed significantly according to physical activity change (P=0.014), showing an increase in dGEMRIC index with increasing physical activity.
Conclusion
Since cartilage appears to positively respond to moderate running when compared to a sedentary lifestyle, this running scheme might be considered a valuable tool in osteoarthritis prevention strategies. Caution is warranted when applying these results to a wider population and to longer training periods.
. Additionally, an athletic lifestyle has been associated with a reduced risk of type II diabetes mellitus and of cancer to the reproductive system, breast and colon
, highly repetitive loading, in time, was generally thought to deplete the joint of lubricating glycoproteins, disrupt the collagen network and to slowly break down the cartilage causing microfractures in the underlying bones
. However, several studies have already investigated the association in prolonged running and osteoarthritis (OA) of the knee and hip showing conflicting results
Risk of osteoarthritis associated with long-term weight-bearing sports. A radiologic survey of the hips and knees in female ex-athletes and population controls.
Risk of osteoarthritis associated with long-term weight-bearing sports. A radiologic survey of the hips and knees in female ex-athletes and population controls.
. Furthermore, an extensive cohort of community-dwelling older adults could not associate recreational physical activity (e.g., walking, jogging) with increased nor decreased risk of OA
. The disparity in outcomes can be suggested being attributed to mixed subject characteristics or analysis methods insensitive to cartilage tissue itself (e.g., X-ray)
Positive effects of moderate exercise on glycosaminoglycan content in knee cartilage. A four-month, randomized, controlled trial in patients at risk of osteoarthritis.
Developments in Magnetic Resonance Imaging (MRI) allow monitoring cartilage macroscopic (morphology: e.g., volume and thickness) and ultra-structural changes (biochemical composition: e.g., glycosaminoglycan (GAG) content) accurately and precisely over time
. Recently, however, a longitudinal study could not show cartilage morphology changes in middle aged women after a 3-month endurance or strength program compared to autogenic training
. Since this observation concurs with the hypothesis that human adult cartilage is not likely to increase in thickness in response to an exercise regime
, one might suggest that the possible benefits of (running) exercise occur at an ultra-structural, qualitative level; the GAG content.
In this respect, no study has yet been published investigating functional adaptation of human knee cartilage due to running by means of changes in GAG content in a longitudinal design. Hence, these results might contribute in understanding the value of moderate running in view of OA prevention strategies. A commonly used technology to estimate GAG content is the delayed Gadolinium Enhanced Magnetic Resonance Imaging of Cartilage (dGEMRIC) technique using the anionic contrast agent gadolinium diethylene triamine penta-acetic acid (Gd-DTPA2−)
. When injected intravenously, and given sufficient time, the anionic contrast agent distributes inversely to the fixed negative charge associated with the GAG content. Gd-DTPA2− therefore distributes in relatively higher concentrations in regions of low GAG, and vice versa. Since Gd-DTPA2− has a concentration dependent effect on the MRI parameter T1, T1 imaging in the presence of Gd-DTPA2− (T1Gd or dGEMRIC index) reflects the cartilage Gd-DTPA2− concentration and, hence, GAG concentration
Applying the dGEMRIC technique, the objective of this study was to investigate the change in dGEMRIC index over time in a cohort of untrained asymptomatic female novice runners participating in a Start To Run program (STR) compared to sedentary controls. It was hypothesized that the group of runners experienced chondroprotective effects of running exercise on knee cartilage when compared to the sedentary controls. This beneficial effect was expected to be shown by a positive dGEMRIC index change in the novice runners when compared to the controls.
Materials and methods
Prior to and after a 10-week STR, asymptomatic female novice runners were subjected to a dGEMRIC analysis of knee cartilage. Accordingly, sedentary controls were tested prior to and after a 10-week period. Consequently, for each subject the change in dGEMRIC index of knee cartilage was calculated and compared between groups.
Subjects
Two groups were recruited on a voluntary basis: (1) nine novice runners and (2) 10 sedentary controls. This study was approved by the relevant local Ethics Committees and all subjects granted their consent to participate. Ethics procedures followed were in accordance with the Helsinki Declaration.
For both groups, the inclusion criteria at baseline were a sedentary lifestyle (i.e., not being regularly involved in sports activities for the last 3–5 years), a sedentary occupation (e.g., desk work), age 20–40 years, Body Mass Index (BMI) 20–30 kg/m2 and female gender. Exclusion criteria were a history of knee complaints, knee internal derangements, surgical and arthroscopic procedures on the knee joint, known presence of degenerative knee pathology, contra-indications for MRI and the dGEMRIC technique in particular. All subjects used contraceptives.
For the runners, this STR program was their first attempt to engage in recreational running activities
. All runners recruited were enrolled to participate in the same STR organized in April 2009 in the same Track and Field club. Sedentary controls were recruited from the local community or university campus by oral and written advertisement according to similar Physical Activity Scores. On recruitment, eligibility of the subjects was verified using a standard questionnaire. Physical activity score in particular, was determined using the reliable and valid Baecke Questionnaire
. This questionnaire measures physical activity level by quantifying ‘work’, ‘sports’ and ‘leisure’ activities using a five-point scale (1=never and 5=always). By counting up the scores of the three distinct dimensions each subject’s total physical activity score was calculated. The sedentary controls were not individually matched to the novice runners. Subject demographics are listed in Table I.
Table IMedians (95% CIs) and P-values of the baseline characteristics of the novice runners compared to the control group
Prior to and after the STR subjects were invited to an MRI session. Four hours prior to the MRI appointment, subjects were instructed to restrain from taking stairs, running and lifting heavy weights
A 1.5 T magnet (Siemens Medical Solutions, Erlangen, Germany) and a dedicated 8-channel knee coil were used for cartilage imaging. At the start of each session, the subjects were subjected to 30–45 min physical rest. For the dGEMRIC technique, a double dose (0.2 mmol/kg) of Gd-DTPA2− (Magnevist, Bayern Schering, Germany) was administered slowly into the right antecubital vein followed by a saline flush with the subject lying supine
, two-dimensional sagittal single slice dGEMRIC images were obtained for the medial knee compartment. These dGEMRIC images consisted of sets of inversion recovery (IR) images with different inversion times (TR=1800 ms, TE=14 ms, TI=50-100-200-400-800-1800 ms, matrix 256×256, FOV 130×130, slice thickness 3 mm). Sagittal slices were centred on the medial femoral condyl using a standard series of localizer images in the three planes. Along with the IR sequence, sagittal proton density images with a similar voxel size were acquired for the purpose of visual guidance during image processing (i.e., T1 calculation for the dGEMRIC index)
. Scanning and slice positioning were performed by a qualified and experienced musculoskeletal radiologist. Patient positioning was standardised using the position of the knee joint according to the reference points on the knee coil. Knee joints were scanned in extension with rigid foam placed around the lower leg and pads around the knee joint to prevent additional movement. In all subjects, the right dominant knee was scanned. Dominance of the lower leg was defined as the leg the subject would choose to kick a ball
During the 10-week period, all novice runners participated in a standardised STR. The STR coaches novice runners to achieve the goal of jogging 5 km (±30 min) within a training period of 10 weeks. This initiative is supervised by the Flemish Track and Field Association and is organized in qualified Track and Field clubs. Participants are trained in a group by a qualified STR coach three times a week. In this study, the coach was a qualified physiotherapist. The STR comprises a gradual build-up of interspersed running and walking units during which the participants are encouraged to jog at their own comfortable speed
To standardise cushioning properties of footwear, all runners wore the same type of neutral running shoe during training (Landreth Gel, Asics Benelux). Additionally, all runners filled out a weekly training log registering training compliance (i.e., participated training units per week/total amount of training units), running surface [grass, athletics track, (hard) woodland, asphalt, other (specify)], absence from training and reason, other concomitant sports/leisure activities (type of activity and duration), possible (knee) complaints. After 10 weeks, runners were subjected to a test during which they had to run laps continuously without resting for a distance of 5 km.
Accordingly, during a 10-week period, sedentary controls were instructed to carry on with their usual lifestyle. Concomitant leisure or sports activities were registered in a weekly log. If usual activity level was restricted controls were instructed to report this as well.
Based on the weekly logs change in physical activity during the 10-week period for each subject was depicted as unchanged, increased or decreased
Positive effects of moderate exercise on glycosaminoglycan content in knee cartilage. A four-month, randomized, controlled trial in patients at risk of osteoarthritis.
Subsequently, mean T1 values for the Region Of Interest (ROI) were determined in that ROIs were drawn on the T1 map on the medial femoral cartilage overlying the posterior horn of the meniscus as described previously
Positive effects of moderate exercise on glycosaminoglycan content in knee cartilage. A four-month, randomized, controlled trial in patients at risk of osteoarthritis.
(Fig. 1). This ROI covered cartilage full thickness and has been shown to present with low intra- and inter-observer variabilities. Additionally, since this region is known for encountering most of the weight-bearing and is one of the primary locations for knee OA onset
, this ROI was of particular interest. In the present study, intra-rater reliability and variability in drawing this ROI attained an intraclass correlation coefficient (ICC) of 0.98 and RMS CV of 0.02 respectively. T1 maps were manually processed in pairs by one researcher with 2 years of practice in cartilage segmentation at the time of analysis, and who was blinded to the time of scanning. Because of the range in BMI (min. 20 – max. 30), for all mean T1 values, the T1-corrected was determined as put forward by Tiderius et al.
Fig. 1Colour-coded maps displaying an example of the dGEMRIC change for the three categories of self-reported physical activity change: increased (novice runner), unchanged (control group), decreased (control group). Additionally, the ROI under study is illustrated.
between inactive and moderately active subjects can be expected clinically. Consequently, to attain such a difference and to reject the null hypothesis (i.e., no difference between groups exists) with a standard power of 80% and α<0.05, one needs to include at least six subjects in both groups.
Prior to the statistical analysis, all outcome variables (i.e., dGEMRIC index at baseline, dGEMRIC index change, subject demographics) were subjected to the Shapiro–Wilk test for normality testing revealing a non-parametric distribution (P<0.05). Consequently, the non-parametric Mann–Whitney U test was applied to test the hypothesis that significant differences in dGEMRIC index change occurred between the novice runners and controls
Positive effects of moderate exercise on glycosaminoglycan content in knee cartilage. A four-month, randomized, controlled trial in patients at risk of osteoarthritis.
. The Kruskal–Wallis test was used to test the hypothesis that differences in dGEMRIC change occurred between the three categories in physical activity change (unchanged, decreased, and increased). In this regard, Spearman’s rho correlation coefficients were calculated as well. Level of significance was set at α<0.05. PASW (version 18.0, Chicago, Illinois) was used for the statistical analyses.
Results
At baseline, no statistical significant differences were established between both groups for age (P=0.515), BMI (P=0.964), and physical activity score (P=0.965) (Table I). Similarly, no significant differences between groups were shown for the dGEMRIC indices at baseline (P=0.541) (Table II).
Table IIMedians (95% CIs) and P-values of the dGEMRIC indices at baseline and of the change in dGEMRIC indices after the 10-week period for the novice runners group and control group
At the end of the 10-week period, eight runners succeeded the final running test and were scanned a second time. One runner dropped out of the study because of sustained shin splints reported during the third week of the program whereas the other runners did not report any complaint. Compliance to the running scheme was 89%. During the 10 weeks, running surface consisted for 54% of participated training units of grass, for 23% of asphalt, for 19% of (hard) woodland, and for 4% of athletics track. In all eight runners, physical activity increased due to participation in the STR program. Based on the log, runners reported, next to the STR, no unusual change in their leisure time activities.
All controls met the second MRI appointment. Based on their weekly log, all of them reported no change in physical activity except for four subjects. These four controls reported decreased activity because of upcoming exams or sickness.
A significant difference between the runners and controls was found for the change in dGEMRIC index after the 10-week period [+11.66 ms (95% confidence interval (CI): −25.29, 44.43) vs −9.56 ms (95% CI: −29.55, 5.83), P=0.006, Table II]. The Kruskal–Wallis test revealed significant differences in dGEMRIC change according to physical activity change category (P=0.014). Median dGEMRIC index changes were −26.24 ms (95% CI: −29.55, −12.19), 4.34 ms (95% CI: −6.94, 5.83), 11.66 ms (95% CI: −25.29, 44.43) for the decreased, unchanged and increased category respectively. Spearman’s rho analysis revealed the relationship between self-reported physical activity change and dGEMRIC index change to display a positive significant correlation (rs=0.741, P<0.001). In Fig. 1, Fig. 2, the dGEMRIC index changes are stratified according to self-reported physical activity change by either using scatter plots or colour-coded maps.
Fig. 2Scatter plots showing individual data points and medians (bars) of the medial femoral cartilage dGEMRIC changes for both the runners (i.e., “Increased”) and controls (i.e., “Decreased” and “Unchanged”) stratified according to change in physical activity level. Additionally, Spearman’s rho outcome for the non-parametric statistical correlation between dGEMRIC change and the three categories of physical activity change is presented. This correlation coefficient reveals a good to strong positive significant correlation between dGEMRIC change and physical activity change.
The most important finding of the present study was that the change in dGEMRIC index after the 10-week period revealed a positive change in the novice runners when compared to the sedentary controls. Since the change in dGEMRIC indices registered was significantly different according to self-reported change in physical activity, these authors suggest that increasing physical activity was associated with positive dGEMRIC index changes, and vice versa.
This study to our knowledge is the first longitudinal design to address the ultra-structural response of cartilage to running in humans. The present results can be supported by the cross-sectional comparison of the dGEMRIC index between sedentary subjects, recreational runners and elite runners performed by Tiderius et al.
. Reporting mean indices (S.D.) of 382 (33) ms, 424 (22) ms and 476 (36) ms respectively, Tiderius et al. substantiate functional adaptation capacity of cartilage with increasing running level. Although not running, nor in asymptomatic subjects, Roos et al.
Positive effects of moderate exercise on glycosaminoglycan content in knee cartilage. A four-month, randomized, controlled trial in patients at risk of osteoarthritis.
similarly presented positive effects on the mean dGEMRIC index change (S.D.) of medial femoral cartilage in post-meniscectomized patients undergoing a 4-month weight bearing exercise program (+15 (45) ms in the exercise group vs −15 (32) ms in the control group) endorsing the notion that moderate exercise can positively affect the dGEMRIC index.
GAGs are known for being important structural matrix compounds in regulating the cartilage tissue’s endo-osmotic swelling pressure and thus, the tissue’s compressive strength
. Therefore, GAG content could be put forward as a surrogate marker for cartilage quality. The positive change of the dGEMRIC index in the novice runners when compared to the sedentary controls allows conjecture about concordant ultra-structural adaptations of cartilage occurring in subjects withstanding higher mechanical demands during the 10-week period. However, one might argue that the difference in dGEMRIC indices observed in the runners group in this study does not appear to meet the expected changes in dGEMRIC index of 42 ms. The median difference in dGEMRIC index after the 10-week period between groups attained 47.69 ms (95% CI: 17.16, 102.96), hence, confirming the expected estimate. Since this difference, however, is driven by an imbalance in physical activity change between groups, the established significance cannot be considered solely in view of the runners group but always in relation to the sedentary lifestyle characterized by inactivity or even decreased activity. Consequently, combined with the positive significant correlation established between physical activity change and dGEMRIC index change, these results remain to endorse the possible chondroprotective effect of the running scheme.
The effect of physical exercise on knee joints is known to display inter-individual differences
. Despite the efforts to select a specific subset of individuals in the present study, the main outcome remains to display substantial variation (Table II). In this regard, Fig. 2 underlines the importance of physical activity change in – but does not entirely explain – the variance in index changes observed. Next to physical activity/sedentary lifestyle, (female) gender, BMI <30, age <40 (pre-menopausal), no known history of knee injury and cartilage degeneration, there are other factors defining a subject’s responsiveness to exercise.
The main limitations of this study comprise the reproducibility and validity of dGEMRIC technique in the long-term and the limited sample size. With reproducibility of T1 measurements within the range of 5–8%
, sources of long-term analysis in-accuracy are mainly patient and slice positioning faults and/or segmentation error due to the smaller areas (i.e., fewer pixels) of the ROI under study. Our segmentation precision (RMS CV=0.02) falls within ranges of those previously reported by Tiderius et al.
. In addition, segmentation and the scanning procedures for both groups were performed by the same tester and the same trained technicians respectively. An advantage of a statistical comparison between groups encompasses that the established differences in dGEMRIC change are prone to the same measurement errors and these errors are therefore counterbalanced
Positive effects of moderate exercise on glycosaminoglycan content in knee cartilage. A four-month, randomized, controlled trial in patients at risk of osteoarthritis.
. Additionally, T1 quantification is influenced by contrast agent distribution primarily regulated by extra-cellular water in the lean and adipose tissues
. Consequently, long-term evaluation might be confounded by alterations in body composition over time and due to the training regimen. These authors acknowledge that, next to BMI measurements during the two test appointments, no other measures were acquired (e.g., bio-electric impedance, DEXA scan) to evaluate body composition. Nonetheless, BMI has been shown to be associated with Gd-PTA2− plasma concentrations without changes in Gd-PTA2− kinetics
. Although circumspection remains warranted, no significant changes in BMI after the 10-week period were observed in this study (median BMI change=0.20; 95% CI: 0.00, 0.49; P=0.910). As the same dosage was administered to the subject twice, delivery at the cartilage plate was likely to be similar during the two test appointments. Finally, the study sample size was relatively limited. Larger sample sizes would have reduced variability or might have allowed taking confounding factors into account. Although confirmation in larger samples is needed, these results suggest similar (i.e., chondroprotective) effects of moderate physical activity as proposed by previous cross-sectional and longitudinal studies applying direct or indirect measures for cartilage status in larger study populations
These results suggest that a gradually built up running scheme causes a chondroprotective effect on the knee when compared to a sedentary lifestyle in a specific subset of asymptomatic subjects. This effect is shown by a positive change in dGEMRIC index (i.e., estimation of GAG content) in the novice runners when compared to the sedentary controls. Consequently, such a moderate running scheme might be proposed valuable in OA prevention strategies. Nonetheless, caution is advised when interpolating these results to a wider variety of individuals and to longer training periods.
Author contributions
Van Ginckel, Ans: conception and design, analysis and interpretation of the data, drafting of the article, critical revision of the article for important intellectual content, final approval of the article, provision of study materials or patients, collection and assembly of data.
Baelde, Nick: final approval of the article, provision of study materials or patients, collection and assembly of data, administrative, technical or logistic support.
Almqvist, Fredrik: conception and design, critical revision of the article for important intellectual content, final approval of the article.
Roosen, Philip: conception and design, critical revision of the article for important intellectual content, final approval of the article.
McNair, Peter: conception and design, final approval of the article.
Witvrouw, Erik: conception and design, critical revision of the article for important intellectual content, final approval of the article, obtaining of funding.
Role of the funding source
This project was partly funded by (1) Special Research Fund Ghent-University and (2) Research Foundation – Flanders. The funding sources had no involvement in the study design, collection, analysis and interpretation of the data; in the writing of the manuscript; and in the decision to submit the manuscript for publication.
Conflicts of interest
No conflicts of interest were declared.
Acknowledgements
The authors gratefully acknowledge the management of RC Racing Ghent, track and field club, and the Radiology Department of General Hospital Jan Palfijn Ghent, for logistic support in organizing this study. Additionally, the authors would like to thank the runners and controls for their willingness to participate and especially the runners for their motivation to keep up the good (running) work during the Start To Run program.
Risk of osteoarthritis associated with long-term weight-bearing sports. A radiologic survey of the hips and knees in female ex-athletes and population controls.
Positive effects of moderate exercise on glycosaminoglycan content in knee cartilage. A four-month, randomized, controlled trial in patients at risk of osteoarthritis.
00338-9&id=gr1.jpg){kind=link}
00338-9&id=gr2.jpg){kind=link}