Osteoarthritis and multimorbidity in male retired professional footballers compared to men in the general population

Background: Football is one of the most popular sports world-wide. Professional footballers are at higher risk of injuries of lower limb joints and the head due to the nature of the game. This can lead to short- and long-term health consequences such as osteoarthritis (OA) and cognitive disorders. However, whether footballers are at higher risk of developing knee pain, multimorbidity and OA of the foot/ankle after retirement remains unknown. Objectives: [1] to systematically review published observational studies on multimorbidity and long-term conditions in professional footballers only or compared to the general population; [2] to examine the incidence of knee pain in male retired professional footballers (referred to as footballers) compared to men in the general population (referred to as controls) and associated risk factors in each group; [3] to examine the association between knee pain at baseline and incidence of multimorbidity at follow-up in footballers and controls; [4] to examine the incidence of multimorbidity in footballers compared to controls and associated risk factors in each group; and [5] to examine the association between radiographic knee OA and foot and/or ankle OA in footballers and controls. Method: A systematic literature review of observational studies for multimorbidity in professional footballers was undertaken in 3 databases in addition to the Google Scholar until 31st December 2021. The Newcastle-Ottawa Scale was used to assess the quality of the observational studies. The prevalence, incidence and associated risk factors of multimorbidity, or specific long-term conditions were extracted or calculated for footballers compared to the general population. The pooled prevalence of multimorbidity and each long- term condition and 95% confidence interval (CI) were calculated. An observational cohort study was undertaken in retired male professional footballers and aged matched general population controlled men aged ≥ 40 years using a postal questionnaire to examine the incidence of knee pain and multimorbidity over the 6-year period, and radiographic examination in a sample from each group to examine the association between radiographic knee and foot/ankle OA. Knee pain was defined as anterior knee pain experienced for most days of the previous month using body pain mannequin. Multimorbidity was defined as having at least 2 long-term conditions in an individual from 45 conditions included. Risk factors for incident knee pain and multimorbidity were examined per group. Radiographic knee OA was defined using Kellgren and Lawrence grade ≥ 2 in either tibiofemoral or patellofemoral compartments irrespective of the side. Foot and/or ankle OA was defined using the La Trobe Atlas (i.e., the presence of at least ≥2 in osteophytes (OP) plus joint space narrowing (JSN) at any joint of foot/ankle). Incidence (%) of knee pain and multimorbidity was estimated for both groups and adjusted odd ratio (aOR) and 95% CI was calculated using logistic regression. Risk factors for knee pain and multimorbidity were examined using multivariable logistic regression. OR and 95% CI was calculated using logistic regression for association between radiographic OA at the knee and foot/ankle joints. Correlation coefficient β was calculated using linear regression for summated radiographic OP and JSN scores at both sites. Results 56 studies (42 cross-sectional studies, 4 case-control studies, 8 cohort studies, and 2 mixed design studies (cross-sectional and case-control)) were eligible for the systematic review from 18 different countries. There were 24 studies for footballers versus controls, and 32 studies for footballers only. The age ranged from 18 to 93 years old. Quality of included studies was regarded as medium quality. Most of the retrieved studies focused on a single long-term condition, and some studies reported on more than one long-term condition but did not report on multimorbidity. OA was found to be the most common condition being investigated, followed by neurodegenerative diseases, hip or knee pain, psychological and cardiovascular diseases. Methods used to examine long-term conditions or their risk factors were highly heterogenous. Of the 14 studies provided prevalence of knee OA in footballers, the pooled prevalence was 31% in male retired footballers (95% CI 20 – 42%). Of the 6 studies provided prevalence of knee OA, the pooled prevalence was 26% in male controls (95% CI 7 – 44%). Of the 4 studies provided prevalence of knee pain in footballers, the pooled prevalence was 53% in male retired footballers (95% CI 0 – 100%). Of the two studies provided prevalence of knee pain, the pooled prevalence was 27% in male controls (95% CI 26 – 28%). Of the 4 studies provided prevalence of ankle OA, the pooled prevalence was 9% (95% CI 7 - 10%) in male retired footballers. Of the 7 studies provided prevalence of hip OA in footballers, the pooled prevalence was 7% (95% CI 4 - 11%) in male retired footballers. Of the two studies provided prevalence of hip OA in controls, the pooled prevalence was 2% (95% CI 1 - 4%) in controls. Joint injury was the major risk factor associated with the increased risk of OA in footballers. For neurodegenerative disease, only one comparative study was identified from the literature, the incidence of neurodegenerative disease was higher in footballers (5.0%) compared to the general population controls (1.6%) over 18 years follow-up (p<0.001). Playing position (midfielder and defender), head injury and career duration might increase risk for neurodegenerative disease. For the following conditions, only one comparative study was identified from the literature, the prevalence of self-reported acute myocardial infarction was 3.08% in male retired footballers and 9.46% in controls (p<0.001), hypertension was 22.98% in footballers and 34% in controls (p<0.001), and diabetes was 3.77% in footballers and 11.47% in controls (p<0.001). Of 486 footballers and 610 controls with data collected at two waves, 222 footballers and 402 controls were at risk for knee pain. There was no significant difference in the incidence of knee pain between footballers and controls (24.0% and 28.8% respectively, p=0.18) (aOR 0.77, 95% CI 0.52 – 1.16). Whilst body mass index (BMI) (aOR 1.17, 95% CI 1.04 – 1.31) and football-related knee injury (aOR 2.42, 95% CI 1.22 – 4.79) were risk factors for incident knee pain in footballers, BMI (aOR 1.08, 95% CI 1.01 – 1.14) and higher socioeconomic status (aOR 0.88, 95% CI 0.81 – 0.95) were associated with incident knee pain in controls. No association was observed between knee pain at baseline and incidence of multimorbidity at follow-up in footballers and controls. A total of 317 footballers and 362 controls were at risk for multimorbidity. The incidence of multimorbidity was lower in footballers (37.5%) than controls (52.0%) (p<0.001) (aOR 0.64, 95% CI 0.45 - 0.89). Whilst age (aOR 1.04, 95% CI 1.01 – 1.07), the presence of one condition at baseline (aOR 2.98, 95% CI 1.78 – 4.99) and analgesic use (aOR 2.88, 95% CI 1.53 – 5.40) were risk factors for incidence of multimorbidity in footballers, age (aOR 1.05, 95% CI 1.03 – 1.08), the presence of one condition at baseline (aOR 2.73, 95% CI 1.73 – 4.31) and higher socioeconomic status (aOR 0.88, 95% CI 0.81 – 0.95) were associated with the incidence of multimorbidity in controls. A total of 72 footballers and 147 controls underwent radiographs for both knee and foot/ankle joints. The association between knee OA and foot/ankle OA was not statistically significant in footballers (aOR 4.13, 95% CI 0.72 – 23.66; p=0.111), but was significant in controls (aOR 2.57, 95% CI 1.01 – 6.53; p=0.046). Similar finding were observed using the linear regression analysis for summated radiographic scores of OP and JSN. Conclusions Although professional footballers may be more likely to develop OA and neurodegenerative disease, they may be less likely to develop other long-term conditions, such as diabetes and cardiovascular diseases, and multimorbidity. In addition to the constitutional risk factors such as BMI, football-related head and joint injuries may explain the additional risks of OA and neurodegenerative disease. However, the general health and physical fitness may also explain the lower risk of other conditions and multimorbidity in footballers compared to controls. The results of this PhD project suggest the preventative strategy should be tailored to the population specifications to mitigate the specific risks and improve the general health status.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.