PARIS – Vigorous, but not moderate, physical exercise was associated with a significantly greater risk of knee osteoarthritis progression over 1 year in a longitudinal analysis of 99 patients.
Engaging in vigorous activity was associated with a 1.38-times increased risk for progression (95% confidence intervals, 1.04-1.83; P = .025), defined as an increase in either cartilage or meniscus defect scores at 1 year.
© Stockbyte/Thinkstock
Vigorous exercise was associated with a significantly greater risk of knee osteoarthritis progression over one year.
The odds ratios were not significantly increased for moderate activity (OR, 0.78; 95% CI, 0.47-1.28; P = .322) but were close to significance for walking activity (OR, 1.50; 95% CI, 1-2.25; P = .052), lead author Deepak Kumar, Ph.D., said at the World Congress on Osteoarthritis.
The analyses adjusted for age, body mass index, sex, pain, Kellgren-Lawrence (KL) score, and modified Whole Organ Magnetic Resonance Scores for cartilage and meniscus in the first block, and International Physical Activity Questionnaire (IPAQ) scores in the second block.
"We did not see that moderate activity was associated with progression, and this indicates that we need to further investigate the optimal dosage of physical activity for someone with knee osteoarthritis," said Dr. Kumar, a postdoctoral scholar, radiology department, University of California, San Francisco.
Although greater physical activity has been shown to reduce pain and improve function, the results support more recent work suggesting that certain types of activity may be associated with disease progression. Walking 10,000 steps or more per day was found to increase the risk of meniscus and cartilage lesions on MRI in people with knee OA (Ann. Rheum. Dis. 2013;72:1170-5), while high and very low levels of self-reported activity were associated with greater cartilage relaxation times on 2T MRI at 4 years in data from the Osteoarthritis Initiative (Osteoarthritis Cartilage 2013;12:1558-66).
For the current analysis, 99 participants in the ongoing, longitudinal Cartilage Loading and Unloading (CLOC) knee OA study underwent 3T MRI of the knee using a sagittal, high-resolution 3D fast spin-echo Cube sequence at baseline, which was repeated 1 year later. Cartilage and meniscus defects were graded by experienced radiologists. Participants with OA had a baseline radiographic KL score of more than 1 and were symptomatic; controls had a KL of 0 or 1 and no knee symptoms.
By 1 year, 35 participants showed progression (14 with OA and 21 controls) and 64 did not. Surprisingly, there was greater progression in the lateral compartment (11 lateral menisci, 12 lateral tibiae) and, more expectedly, in the patella in 12 persons, Dr. Kumar said at the meeting, sponsored by the Osteoarthritis Research Society International.
No significant baseline differences existed between nonprogressors and progressors with respect to age (53.2 years vs. 50.8 years), body mass index (24.4 kg/m2 vs. 24.5 kg/m2), and sex (62.5% vs. 48.5% female), he said.
Knee injury and Osteoarthritis Outcome Score subscale scores were also similar for pain (85.7 vs. 82.4) and symptoms (84.5 vs. 83.4).
Progressors had engaged, however, in significantly more metabolic equivalent-minutes per week of vigorous exercise than did nonprogressors (2,410.6 vs. 1,413.1; P = .046), Dr. Kumar said. Metabolic equivalent-minutes per week were similar for moderate activity (1,094.1 vs. 858.7; P = .396) and walking (1,646.2 vs. 1,245.1; P = .151).
During the discussion following the formal presentation, an attendee said the study is potentially very valuable because it suggests that something all clinicians want to do is "dangerous" but questioned whether some of the MRI data were "overread" given the almost 40% rate of progression in a relatively fit cohort. A Danish attendee also remarked that her group has experienced so many problems with patients filling out the IPAQ questionnaire that they no longer trust the data.
Dr. Kumar responded that the MRI readings were performed by expert radiologists and were reliable, and that more stringent definitions of progression are being explored. His group is also working on a new questionnaire to better define physical activity levels.
Data are also being analyzed from the rest of the cohort (160 participants) and over a longer, 3-year follow-up, he said in an interview.
"We are also identifying quantitative MRI and biomechanical metrics that may be more sensitive to disease progression in knee OA. These will help us understand the factors that are related to worsening of knee OA and develop therapies."
The National Institutes of Health–National Institute of Arthritis and Musculoskeletal and Skin Diseases funded the work. Dr. Kumar and his coauthors reported no conflicting interests.