Impact of a daily exercise dose on knee joint cartilage – a systematic review and meta-analysis of randomised controlled trials in healthy animals

Abstract

Purpose: To investigate the impact of a daily exercise dose on cartilage composition and thickness, by conducting a systematic review of randomized controlled trials (RCTs) involving healthy animals. Methods: A systematic literature search was performed within the databases MEDLINE, EMBASE, Web of Science and CINAHL for studies published prior to October 2015 with no restriction on publication year or language. We included randomized controlled trials of healthy animals assessing the impact of exercise on cartilage aggrecan content, collagen content and cartilage thickness. The following studies were excluded: non-RCTs, studies which did not investigate any of the outcomes of interest or knee joints, studies with genetically modified animals, comorbidities, non-English full text, and where treatment arms involved other interventions besides exercise. In dog studies, we used the British Kennel Club recommendations for exercise to categorize dogs into low, moderate or high dose of exercise. To classify the dose of exercise programs in rodents as low, moderate or high, we used the American Physiological Society guideline for the design of animal exercise protocols. To determine the dose of exercise in rabbits and sheep, we used the authors’ own classifications, since no existing exercise dosage classifications were found in the literature. A narrative synthesis of the effect of a daily exercise dose on knee cartilage aggrecan, collagen and thickness was performed. We classified the effect of exercise on knee cartilage aggrecan content, collagen content and thickness as: positive (‘+’) or negative (‘-’) when a significant difference (P<0.05) was reported in favor of the exercise (‘+’) or control (‘-’) group; and as similar (‘=’) when no significant difference between the intervention and control groups was reported. A subset of studies reporting sufficient data was combined in meta-analysis using a random-effects model. Meta-regression analyses were performed to investigate the impact of covariates. Results: The literature search resulted in 1,144 publications of which 112 individual studies were identified as potentially eligible and checked in full text. Ultimately, twenty-nine RCTs, involving 64 comparisons, were included. In the low dose exercise group, 21/25 comparisons reported decreased or no effect on cartilage aggrecan, collagen and thickness. In the moderate dose exercise group, all 12 comparisons reported either no or increased effect. In the high dose exercise group, 19/27 comparisons reported decreased effect. A meta-analysis of 14 studies investigating cartilage thickness showed no effect in the low dose exercise group (SMD -0.02; 95%CI -0.42 to 0.38; I2=0.0%), large but non-significant cartilage thickening in the moderate dose exercise group (SMD 0.95; 95%CI -0.33 to 2.23; I2=72.1%) and non-significant cartilage thinning in the high dose exercise group (SMD -0.19; 95%CI -0.49 to 0.12; I2=0.0%). Results were independent of analyzed covariates. Conclusions: Our results suggest that the relationship between daily exercise dose and cartilage composition and thickness is non-linear. While a low daily dose of exercise may have no effects and a high daily dose of exercise may have negative effects, a moderate daily dose of exercise may have positive effects on cartilage composition and thickness in healthy animals.