Prevalence and impact of knee osteoarthritis in runners

Abstract

Purpose: The aim of this study was to investigate the prevalence and impact of knee osteoarthritis among recreational runners. Methods: This study is part of a randomized-controlled trial that investigated the effectiveness of an evidence-based online injury prevention program on running injuries, in preparation for a running event (from 5 km up to marathon) in recreational runners (N=2378). After registration, a baseline questionnaire informed on demographics, lifestyle characteristics, running characteristics and injury history. All participants received three follow-up questionnaires during the study period; two weeks before the running event they registered for, one day after the running event and one month after the running event. The follow-up questionnaires informed on running related injuries (RRIs) during follow-up. Participants that reported a new running related knee injury (RRKI) during the study period (N=277)were sent a knee-specific questionnaire after a mean of 16-months (range 11.7-18.6) follow-up. This questionnaire informed on the type of knee complaints, morning stiffness, activity related knee pain and self-reported diagnosis. In addition, the National Institute for Health and Care Excellence (NICE) guideline was followed to diagnose clinical knee osteoarthritis (OA).Participants were asked if they were recovered from their RRKI (yes/no) and time to recovery(weeks) was questioned. Information on medical consumption was obtained and all were asked whether the RRKI restricted their running in terms of running speed, duration and/or frequency. The subscales symptoms and sports of the Knee injury and Osteoarthritis Outcome Score (KOOS/100) and the Anterior Knee Pain Score (AKPS/100) were used to administer OA and patellofemoral specific outcomes at follow-up. T-tests and chi-square tests were used to describe differences between runners with and without knee OA. With cox regression analysis, the association between time-to-recovery and potential prognostic factors was determined. Results: Of the 277 runners that reported a new RRKI (11.6%) during follow-up, a total of 138 (49.8%) responded to the knee-specific follow-up questionnaire and were consequently included in the current study. Responders were on average significantly older (42.3 vs. 39.3 years, p=0.04). No other differences between responders and non-responders were observed. At baseline, study participants (N=138) were on average 42.3 (SD 12.2) years old, had an average BMI of 23.3 (SD 3.0) kg/m2 and the majority was male (59.4%). Participants trained on average 2.2 (SD 0.9) times a week, spend 2.6 (SD 1.5) hours a week on training with an average running speed of 6.0 (SD 0.9) min/km. At baseline, 36.2% of the participants reported an RRKI in the previous 12 months. At 16 month follow-up, 71.0% of the runners were recovered from their knee injury, with a mean recovery time of 10.5 (SD 9.9) weeks. Non-recovered participants had complaints for 54.4 (SD 4.8) weeks up to follow-up. Following self-diagnosis, most participants suffered from knee OA/degenerative meniscus lesions (23.2%)and iliotibial band syndrome (23.2%). Morning stiffness was reported by 26.0% of the participants. Following the NICE guidelines, 13.8% of the runners with incident knee injuries were diagnosed with knee OA at follow-up. Within the group of participants that had clinical knee OA following the NICE guideline, 5.1% was recovered and 35% was not-recovered from their knee injury (p<0.001). The presence of morning stiffness was associated with recovery rate (83.3% versus 36.1%, p<0.001). Runners with clinical knee OA had a significantly lower KOOS symptom score (65.3(20.0) vs 84.8(20.0), p<0.001) and AKPS (80.9(10.7) vs. 93.3(9.6), p<0.001) compared to runners with other types of RRKIs. Runners with a self-reported diagnosis of knee OA more frequently visited a health care professional (78.1% vs. 43.4%, p<0.001) compared to runners with other types of RRKIs, with especially a high percentage of visits to the physiotherapist (80%). The results of the COX regression for time to recovery shows that male sex (HR 1.90; 95%CI 1.17-3.09) was associated with a shorter time to recover from RRKIs, while participants diagnosed with knee OA following the NICE guideline (HR 0.18; 95%CI 0.06-0.53) had a longer time to recovery. Conclusions: Time-to-recovery from RRKIs is relatively long, which emphasizes the need for optimal treatment, education and injury prevention programs in recreational runners, especially in female runners. Furthermore, more knowledge on the impact of running with knee OA is important, given the high number of runners with knee OA symptoms and the longer time-to-recovery of RRKI in runners with knee OA.