Exercise as Medicine During the Course of Hip Osteoarthritis

Inger Mechlenburg; Lisa Cecilie Urup Reimer; Troels Kjeldsen; Thomas Frydendal; Ulrik Dalgas


Exerc Sport Sci Rev. 2021;49(2):77-87. 

In This Article

Exercise and PA as Primary Prevention in Hip OA

Studies investigating risk factors for development of hip OA have evaluated the effects of either different exercise modalities or a person's PA level. In addition, studies including several proxy measures for exercise or PA level such as occupation or muscle strength can be found. A systematic review from 2013 identified publications investigating exercise and risk of hip OA.[28] One study did not find an association between regular exercise and risk of hip OA reporting an odds ratio of 1.2,[15] whereas other studies reported that soccer,[19] racket sports,[20] gymnastics in women,[16] and handball in men[22] increased the risk of developing hip OA. Oppositely, one of the studies did not find associations between hip OA and running, kung fu, and soccer as well as gymnastics in men.[16] Furthermore, one of the included studies found that the odds ratio of having hip OA in persons with high exercise exposure before age 50 was 4.5 compared with low exposure.[20] Another more recent systematic review investigating the relation between exercise and any type of OA found an increased risk of developing OA in people who participated in sport with a relative risk of 1.4 compared with a control group.[21] The risk was highest among soccer players and lowest among runners and tennis players, but no difference in the risk of developing OA was found between elite and nonelite athletes.[21] In addition, the Norwegian HUNT study that enrolled 15,191 women and 14,766 men did not find any associations between exercise intensity levels or exercise volume and an increased risk of having developed hip or knee OA when assessed 11 yr after baseline.[17]

A recent systematic review concluded that running at a recreational level was associated with significantly lower odds of hip or knee OA compared with competitive runners or controls.[14] In addition, the authors stated that these results indicate that long exposure to high-volume or high-intensity running as well as a sedentary lifestyle is associated with a higher risk of hip or knee OA.[14] When the data were stratified by joint, the odds ratio for competitive runners compared with recreational runners was not statistically significant for hip OA. When stratifying data for years of running, that is, above or less than 15 yr, the overall results showed that people running for longer than 15 yr had a greater risk at developing hip or knee OA. When stratifying the data based on either hip or knee OA, the results were no longer statistically significant for hip OA. Of note, most of the included studies were cross-sectional and case-control studies.

Another recent systematic review investigated the association between high-impact exercise and the risk of developing hip OA among elite athletes. The review found that male athletes participating in elite impact sports were at higher risk of developing hip OA compared with matched controls. The odds ratio were 1.8–8.7 depending on the types of sports,[72] with handball being associated with the highest rate of hip OA followed by soccer, hockey, track and field, and basketball. The risk of hip OA in long-distance runners was inconclusive in the aforementioned review,[72] as studies demonstrated lower,[23] higher,[24,26] or similar[24,25] rates of hip OA among male runners when compared with controls. Since the systematic review was published, a case-control study on the prevalence of THA in retired national football athletes was published, reporting a high prevalence of THA of 4.6% among the retired football athletes.[29] Risk factors for THA were age, linemen, and injuries.[29] For comparison, the prevalence of THA in the total U.S. population was 0.83% in 2010.[73] Interestingly, joint injuries have also been investigated in a systematic review. The review identified four case-control and cohort studies showing an odds ratio of 5 for an increased hip OA in persons having had joint injuries.[28] Similarly, Hootman et al.[18] found that hip and knee injuries were a significant independent risk factor for developing hip/knee OA in men.

Ageberg et al.[32] enrolled 27,760 participants and showed that higher leisure time PA offered protection against THA in women. The adjusted relative risk for THA was 0.66 among women with the highest PA, compared with those with the lowest PA after adjusting for age, sex, body mass index (BMI), education, smoking, and marital status.[32] Hootman et al., however, did not find an association between PA and hip OA risk in a longitudinal study of 5283 adults without hip or knee OA. The participants were asked to quantify their PA in leisure time, and a decade later, they were asked whether they had been diagnosed with hip or knee OA.[18] The authors did not find significant associations between hip or knee OA and frequency, pace, or weekly amount of running or walking, after adjusting for sex, age, BMI, previous joint injury or surgery, smoking status, and comorbidity.[18] Similarly, Cheng et al.[27] did not find an association between baseline leisure PA level and physician-diagnosed hip or knee OA 10 yr later in 16,961 participants. However, the authors found that younger men less than 50 yr of age who were running more than 20 miles weekly had a 2.4-fold higher risk of developing OA compared with men who were classified as being sedentary, after controlling for BMI, smoking, and intake of alcohol or caffeine.[27]

Proxy Measures

One proxy measure of exercise/PA is muscle strength. A systematic review from 2013 identified eight cross-sectional studies investigating the association between muscle strength and hip OA.[30] The main findings were that strong evidence showed muscle weakness in patients with hip OA, whereas moderate evidence showed that muscle strength in the affected leg is lower compared with both the opposite leg and healthy controls.[30] Of note, all studies on muscle strength were cross-sectional and thus not allowing conclusions on causality and whether low muscle strength leads to hip OA or whether hip OA leads to low muscle strength. A narrative review from 2016 states that the deep stabilizing muscles of the hip possibly acts as protectors of the joint.[31] This indicates a possible link between muscle strength and development of hip OA, but solid longitudinal studies are lacking in this area. Furthermore, it is challenging to measure hip muscle strength reliably at a level not exceeding recommended limits of agreement of 15% between test sessions.[74]

Another proxy measure of PA is occupation. An umbrella review of systematic reviews from 2019 found that occupational physical tasks related to forces exerted on the hip, such as heavy lifting, kneeling, squatting, and climbing, were associated with an increased risk of hip OA.[33] Particularly, heavy lifting was reported to increase the risk of hip OA in men by 150%.[33] This is somewhat in line with the aforementioned findings linking sports to hip OA, where different exercise modalities seem to have very different impact on hip OA risk.

Altogether, the existing evidence linking exercise/PA and hip OA is inconsistent and seemingly complex. The observed discrepancies may be explained by differences related to study design, outcomes, and populations, but may also reflect a potential nonlinear and modality-dependent dose-response relation as depicted in Table 1.