The effects of running gait retraining on biomechanics, performance, pain and injury: a systematic review and meta-analysis

INTRODUCTION
Running gait retraining is an emerging area of interest in the prevention [1] and treatment [2,3] of running-related injuries. There is increasing evidence that runners can successfully modify their running biomechanics using various gait retraining strategies [4,5]. This review aimed to synthesise evidence from randomised controlled trials (RCTs) that investigated the effectiveness of gait retraining on running biomechanics, performance, pain and injury in distance runners.
METHODS
Seven electronic databases (MEDLINE, EMBASE, CINAHL, Web of Science, SPORTDiscus, PEDro, and CENTRAL) were searched from inception to March 2021 to identify RCTs evaluating running gait retraining compared to no intervention, usual training, placebo, or standard care, and reporting biomechanical, performance, or pain and injury outcomes. Only studies with longitudinal follow-up were included. Gait retraining interventions were categorised into step-rate, non-rearfoot footstrike (NRFS), impact, and multi-parameter subgroups. Risk of bias was assessed using the Cochrane Risk of Bias tool [6]. We conducted meta-analysis with standardised mean differences (SMD) for outcomes with homogeneous data from at least two studies and classified evidence quality as very low, low, moderate or high according to Grading of Recommendations Assessment, Development and Evaluation (GRADE) criteria. SMDs were described as trivial (<0.20), small (0.20–0.59), moderate (0.60–1.19), and large (>1.19).
RESULTS AND DISCUSSION
A total of 19 studies from 16 independent cohorts involving 673 participants met the inclusion criteria in our review. There was moderate evidence that gait retraining targeting an increase in step rate does moderately increase step rate (SMD 1.03 [95% Confidence Interval (CI): 0.63, 1.44]) and cause a small decrease in average vertical loading rate (AVLR) (SMD -0.57 [95% CI: -1.05, -0.09]) (Figure 1). There was low-quality evidence that NRFS retraining moderately increases knee flexion at initial contact (SMD 0.74 [95% CI: 0.11, 1.37]) and NRFS and multi-parameter gait retraining does not significantly alter measures of running performance (SMD ≤0.32, p ≥0.192). There was a lack of studies that report on pain and injury outcomes with conflicting findings. Individual study findings demonstrated reduced injury incidence (hazard ratio 0.38 [95% CI: 0.25, 0.59]; rate ratio 0.52 [95% CI: 0.31, 0.86]) following impact and multi-parameter gait retraining and conflicting findings regarding knee pain changes following step-rate and NRFS retraining.
CONCLUSIONSThe findings from our review indicate step rate interventions were effective at increasing step rate and reducing AVLR, and NRFS interventions cause an increase in knee angle at initial contact. The retraining interventions in this review did not have any negative or positive effects on measures of running performance. Notably, there is preliminary evidence that gait retraining targeting a reduction in landing impacts may be an effective injury prevention strategy for healthy runners, but further RCT evidence is required before recommending gait retraining for injury treatment. Clinicians and researchers can selectively use the gait retraining approaches described in this review to target specific running biomechanics, but not enhance running performance.The GRADE evidence quality was generally rated as low to moderate due to imprecision related to the small size of most studies and risk of bias concerns that were common to several included studies. There was also large heterogeneity between gait retraining interventions employed in studies that limited the pooling of data. Further research involving injured runners is required because they are the population considered to be in most need of retraining.
REFERENCES1. Chan ZYS et al. Am J Sports Med 46: 388-95, 2018.2. Esculier JF et al. Br J Sports Med 52: 659-66, 2018.3. Davis IS et al. Curr Rev Musculoskelet Med 13: 103-4, 2020.4. Crowell HP et al. J Orthop Sports Phys Ther 40: 206-13, 2010.5. Roper JL et al. Clin Biomech (Bristol, Avon) 35: 14-22, 2016.6. Higgins JP et al. BMJ 343:d5928, 2011