Eduard Alentorn-Geli, MD, MSc, PhD
Study Design
Systematic review and meta-analysis.
Background
Running is a healthy and popular activity worldwide, but data regarding its association with osteoarthritis (OA) are conflicting.
Objectives
To evaluate the association of hip and knee OA with running and to explore the influence of running intensity on this association.
Methods
PubMed, EMBASE and Cochrane Library databases were used to identify studies investigating the occurrence of OA of the hip and/or knee among runners. Studies comparing this occurrence between runners and controls (sedentary, non-running individuals) were meta-analyzed. Runners were regarded as ‘competitive’ if they were reported as professionals/elite athletes, or participated in International competitions. Recreational runners were individuals running in a non-professional (amateur) manner. The prevalence and odds ratio (95% CI) for OA between runners (at competitive and recreational level) and controls were calculated. Subgroup analyses were conducted for OA location (hip or knee), gender and years of exposure to running (less or more than 15 years).
Results
Twenty-four studies (n=123,173 individuals) were included and 16 (n=112,192 individuals) were meta-analyzed. The overall prevalence (95% CI) of hip and knee OA was 13.3% (11.62-15.2) in competitive runners, 3.5% (3.38-3.63) in recreational runners and 10.23% (9.89-10.58) in controls. The odds ratio (95% CI) for hip and/or knee OA between recreational runners and controls was 0.66 (0.57-0.76). The odds ratio (95% CI) for hip and/or knee OA in competitive runners was higher than that in recreational runners (OR (95% CI) 1.34 (0.97-1.86) and 0.66 (0.57-0.76) respectively (controls as reference group); p=0.0001).
Conclusions
Recreational runners had a lower occurrence of OA compared with competitive runners and controls. These results indicated that a more sedentary lifestyle or long exposure to high-volume and/or high-intensity running are both associated with hip and/or knee OA. However, it was not possible to determine whether these associations are causative or confounded by other risk factors, such as previous injury.
Level of Evidence
Etiology/Harm, Level 2a. J Orthop Sports Phys Ther, Epub 13 May 2017. doi:10.2519/jospt.2017.7137