Football (Soccer) as a Probable Cause of Long-Term Neurological Impairment and Neurodegeneration: A Narrative Review of the Debate

Football (soccer) is the most widely played sport across the globe. Due to some recent high-profile cases and epidemiological studies suggesting football can lead to neurodegeneration, scientific and public interest has been piqued. This has resulted in research into whether an association between football participation and neurodegeneration or neurological impairment is present. It has been theorised that a combination of repeated sub-concussive and concussive injuries, due to ball-heading and head collisions, may lead to neurodegeneration. However, evidence remains conflicting. Due to the popularity of the sport, and the serious conditions it has been linked to, it is important to determine whether repeated head impacts during football participation can play a causative role in neurodegenerative disease. To answer this question, a review of the current literature was carried out. Epidemiological evidence showed a higher incidence of amyotrophic lateral sclerosis amongst amateur and professional footballers and that footballers in positions that involve less contact and heading, e.g., goalkeepers lead significantly longer lives. Additionally, imaging studies reach a similar conclusion, reporting changes in brain structure, blood flow, and inflammatory markers in footballers when compared to controls. However, studies looking at an association between heading frequency and cognition show a lack of consensus on whether a higher heading exposure results in reduced cognition. Similarly, in neuropathological studies, signs of chronic traumatic encephalopathy (CTE) have been found in some former players, with contrasting studies suggesting low levels of CTE-type pathology are found in the general population, regardless of exposure to head trauma. The majority of studies suggest a link between football and neurodegenerative disease. However, the high prevalence of retrospective cohort and cross-sectional studies, often plagued by recall bias, undermine the conclusions drawn. Therefore, until larger prospective cohort studies are conducted, concrete conclusions cannot be made. However, caution can be exercised to limit head impacts.


Introduction And Background
Football (soccer) is the world's most popular sport, with an estimated 250 million participants [1]. Following the evidence brought to light by epidemiological studies [2], growing interest has been expressed in understanding the association between football and increased rates of neurodegenerative disease in participants.
Neurodegeneration forms an overarching category that includes a range of conditions, such as Alzheimer's disease (AD), Parkinson's disease (PD), and amyotrophic lateral sclerosis (ALS). These diseases are caused by changes in neuronal structure, loss of function, and cell death in a progressive manner. The true nature of these diseases is still poorly understood, with pathological observations not always correlating with clinical symptoms. Chronic traumatic encephalopathy (CTE), for example, may present with cognitive impairment, increased suicidal ideation, or remain asymptomatic even with extensive pathology [3].
Many neurodegenerative conditions share mechanistic features at the cellular level. The common principal causes include protein dysfunction, neuroinflammation, and oxidative stress [4]. Some of the proteins involved may have a role in several diseases; one such example is hyperphosphorylated-tau, which has been implicated in both AD and CTE [5], where hyperphosphorylated-tau is found in neurons and astrocytes at the base of cortical sulci in a perivascular distribution [6].
The risk factors of neurodegenerative disease can vary, with recent studies beginning to unravel the interaction between genetic predisposition and environmental risk factors. One example is traumatic brain injury (TBI), which has been associated with an increased risk of conditions such as AD and PD [7][8][9][10]. Despite these more recent advances in understanding, ageing is still the greatest risk factor for neurodegenerative disease.
Understanding the changes that occur leading to neurodegeneration is often difficult as diagnoses are confirmed on neuropathological analysis. Therefore, other modalities must be employed in the form of cognitive and behavioural testing and imaging studies such as magnetic resonance imaging (MRI). In addition, particularly in the context of contact sports, determining the frequency and severity of a head impact often relies on self-reports, and these frequencies will often be recollected incorrectly. There is also a growing debate surrounding the definition of concussion and its value in describing TBI severity [11]. Despite this, the term is still commonly used and has been included for the purpose of this review.
In recent years, reports of CTE in former American football players [12,13] have prompted research into other contact sports, such as football, and whether they, too, are associated with neurodegenerative changes. Playing football involves a high degree of physical contact and, as a result, both subconcussive and concussive injuries can occur. While concussions usually produce noticeable symptoms, such as headache and dizziness [14], subconcussive injuries do not produce immediate neurological deficits [15]. However, a growing body of literature is linking factors such as repetitive heading with brain damage [16]. These studies provide a possible explanation for the increased rates of long-term neurodegeneration in footballers.
There remains much discourse on the subject, with studies finding evidence both for and against an association between football and long-term neurodegeneration. The topic has also received significant media coverage, partly due to news that Jeff Astle and other high-profile footballers died of CTE, which has led to much public concern [17]. This has even prompted the United Kingdom (UK) Football Association to publish guidelines on limiting heading in training [18].
Our principal aim was to examine the literature to determine if these concerns are justified by considering the evidence for and against there being an association between playing football and long-term neurological impairment and neurodegeneration. We have categorised our findings based on the following study types: epidemiological, cognition, pathological, and imaging.

Evidence for and against a correlation by study type
Research studies conducted across a range of academic disciplines have found evidence that both corroborate and dispute an association between repeated head impacts (RHIs) in football and long-term neurodegeneration. We explore the current evidence across these study types.

Epidemiological Studies
Epidemiological studies allow us to investigate the potential link between football as an exposure and neurodegenerative disease as an outcome over time. One particularly important research effort has been that of the Football's Influence on Lifelong Health and Dementia Risk (FIELD) study [19][20][21]2], a collaborative effort investigating physical and mental health outcomes in former football players based in Scotland. Additionally, several other cohort studies across Europe have also examined this correlation.
Chiò et al. found a statistically significant link between a career lifespan of more than five years and ALS when studying 7,325 male Italian professional football players [22]. A follow-up study identified three further cases, particularly amongst midfielders, strengthening this correlation [23]. No ALS cases were found in a group of basketball players and road cyclists, suggesting that football is a specific risk factor for ALS. RHIs from heading and other risk factors, such as genetic predisposition, for example, mutations in the SOD1 and TDP-43 genes [24], and chemicals such as herbicides [25] are some explanations proposed by the authors. There are studies that corroborate the link between RHIs and increased neurodegenerative risk. Tierney and Higgins quantified the frequency of headers during 7,147 football games in six European leagues for three seasons by reviewing match footage and found that defenders consistently performed the greatest number of headers [26]. Interpreted with the results from a retrospective cohort study on former male professional Scottish footballers by Russell et al., which revealed that defenders had the highest risk of neurodegenerative disease [19], it could be argued that RHIs are a potentially significant risk factor. This would have to be further investigated with robust longitudinal studies before making definitive conclusions. An additional piece of supportive evidence came from the study performed by Russell et al., who found that the risk of neurodegenerative disease was lowest for goalkeepers [19]. Comparably, Śmigielski et al. observed that goalkeepers had a five-to-eight-year longer lifespan amongst international football players born before 1923 who played in the first three World Cups or in the 1946/1947 season of the main European leagues [27]. In general, it appears that the risk is high in footballers of all playing positions, as supported by Russell et al. [19] and Mackay et al. [2], whose study revealed a higher mortality rate from neurodegenerative disease amongst 7,676 former Scottish professional football players as part of the FIELD study; a higher frequency of prescribed dementia-related medications was also noted in this population.
In addition to a potentially higher risk of ALS amongst footballers, results from some studies also suggest an earlier onset. Pupillo et al. discovered 34 ALS cases, nearly double the expected number, amongst 23,586 male professional Italian players who were followed up from the age of 15 until either 2018 or their death [28]. The mean age at diagnosis was 20.2 years younger than that of the general population. Moreover, Gamez and Carmona found the mean age of ALS symptom onset to be 23.7 years younger [29]. However, these studies had several limitations, such as basing the results on incomplete medical records and having reduced statistical power from limited case numbers.
Ultimately, strong evidence can be obtained through cohort studies as temporal associations can be observed. Despite this, further longitudinal studies, including prospective matched-cohort studies, are needed before drawing definitive conclusions on the causality between football and neurodegenerative disease risk.
A summary of epidemiological studies investigating the association between participation in football and neurodegenerative disease is given in Table 1

Cognition Studies
Studies looking at the impact of football on cognition allow us to determine if football impacts an individual's functionality. Research primarily looks at this in two populations, either those who are young and actively playing football or those who are older and retired from the sport. Although cognition does not always reflect the extent of neuropathology present, it can serve as a marker of neurodegeneration, particularly when there is a progressive effect on cognition.
Stewart et al. [30] and Levitch et al. [31] showed that active players with greater repetitive heading in the two weeks prior to cognitive testing being carried out had decreased psychomotor speed, memory, and attention. These same deficits are also seen acutely following concussion [30]. This suggests that, although sub-concussive, heading can induce similar cognitive deficits to concussion.
Interestingly, Levitch et al. also found that moderate levels of short-term heading (defined as occurring within the previous two weeks) improved attention levels [31]. Strauss et al. echoed this, finding footballers with no or low exposure to repetitive heading exhibited better attention, processing speed and memory compared to non-athletes [32]. However, once players pass a 'threshold value' in heading frequency, these benefits are no longer seen, with athletes with high heading frequency showing decreased cognition compared to other footballers [31].
Furthermore, studies have shown that active players with a greater history of long-term frequent heading (lasting more than 12 months) had lower cognitive testing scores in sections regarding attention, planning, verbal learning, and memory [31,[33][34][35]. Levitch et al. theorised that this long-term reduction is due to the accumulation of damage induced by short-term heading, thus suggesting heading contributes to long-term neurodegeneration in footballers [31]. This conclusion is further supported by studies showing that amateur footballers performed worse in cognitive testing than controls [36,37].
Of note, participants in these studies were relatively young and all but one study used amateur footballers as opposed to professionals. This means that the individuals may not have had sufficient exposure to footballrelated head impacts to induce the levels of cognitive deficit and neurodegeneration that may predispose one to diseases such as CTE.
Koerte et al. [38] and Bruno et al. [39] addressed this by carrying out cognitive testing on retired professional footballers and found that these cognitive deficits persist following retirement. They found that retired footballers with a greater history of heading had reduced cognitive performance [39] and that footballers had worse memory than age-matched controls [38].
A limitation of many of these studies is that they depended on participants to report their own heading frequencies. This may have led to the collection of inaccurate data, especially considering that heading has been associated with worse memory. Future studies should use video recordings of games and training to corroborate data, thus reducing self-reporting bias. Additionally, confounding factors such as the force with which a ball was headed, rotational versus linear force experienced, participation in other contact sports, and level of football exposure during childhood and post-retirement were often not accounted for.
Cognition studies have been useful in identifying factors that may contribute to long-term neurodegeneration in footballers. They have also shown that the cognitive decline in retired professional footballers is more advanced than in the general population [38]. However, due to the cross-sectional nature of these studies, it is difficult to determine whether the cognitive decline seen is truly due to football and whether participation contributes to the development of long-term neurodegenerative diseases such as CTE.
While there is evidence pointing to a dose-dependent relationship between heading frequency and neurodegeneration [30,39], studies that are longitudinal with larger sample sizes should be carried out. These should follow players throughout their careers and into retirement, helping establish the timeline and natural history of cognitive changes that occur following exposure to football. These should be followed by post-mortem pathological examination of participants.
Conversely, many studies investigating the cognitive effects of football have failed to observe any long-term impairments among players. Both Straume-Naesheim et al. [40] and Guskiewicz et al. [41] demonstrate that heading exposure or the number of unintentional RHIs, such as concussions, does not correlate with a worse neuropsychological performance on testing. The authors, therefore, conclude that playing football does not lead to cognitive impairment during a player's career, meaning that the risk of long-term neurodegenerative outcomes is low. However, a limitation of both these studies was that there was no player follow-up and, once again, large cohort longitudinal studies are needed to confirm these findings. This was addressed by Kemp et al., who, in a longitudinal prospective study, showed that professional football players after a fiveyear follow-up displayed no changes in neurology, psychology, or structural brain imaging [42]. However, it should be noted that the sample population was relatively young and research into an older population is needed. Additionally, even though older studies had found an association between a history of unintentional RHIs and reduced cognition, more recent studies, such as those by Stewart et al. [30], Lipton et al. [33], and Bruno et al. [39], have not found this. These differing conclusions may be due to a change in the management of unintentional RHIs over time.
Furthermore, Vann Jones et al. demonstrate that retired football players, overall, have no greater risk of cognitive decline and dementia than the general population [43]. They also found that, in contradiction to other studies [19,27,29,38], player position and career length were not associated with an increased risk. As a result, the study suggests that the link between football and long-term neurodegenerative outcomes is lacking, and any cognitive impairment that is observed in players may be reversible by the time they have retired. However, the relevance of these findings is limited as the study had a small sample size and no control group.

Pathological Studies
Analysing the brain's pathology post-mortem offers a unique opportunity to identify any insults that may have occurred. To understand more about the potentially pathological relationship between football and neurodegenerative disorders, former footballers' brains have been examined.
Several initial case reports, such as by Hales et al. [44] and Grinberg et al. [45], highlighted that AD may be incorrectly diagnosed in the place of CTE in patients with a significant football history. The authors suggested a different explanation for the findings was needed, with the chronic injurious effects of playing football as a potential theory. However, there were no comparison groups, and, due to the possible coexistence of AD and CTE pathology, the extent to which these both contribute to clinical symptoms required investigation. The coexistence of multiple pathologies was also found in a FIELD study conducted by Lee et al., in which seven former footballers with dementia were examined neuropathologically postmortem [46]. A mixed presence of CTE and other pathologies was seen in five of the footballers, with two participants deemed to have sufficiently extensive CTE to explain their dementia. One limitation common to the previous pathology studies was the lack of prospective data collection. This, however, was undertaken by Ling et al., where clinical data and information about the participants' playing careers, including rates of concussion, were collected [47]. Out of the 14 footballers analysed, six displayed abnormalities of the septum pellucidum, consistent with a history of chronic RHIs and four had pathologically confirmed CTE.
These studies allude to the potential for neurodegeneration due to football participation. The crosssectional nature of neuropathological studies makes it difficult to identify the relationship between potential causes and effects. Yet again, this highlights the need for prospective cohort studies followed by post-mortem analyses.
On the other hand, a pathological study by Iverson et al. points towards the lack of an association [48]. In a post-mortem analysis of individuals without a known history of RHIs, they found that six out of eight individuals were positive for hyperphosphorylated-tau in neurons, astrocytes, and cell processes in the cortical sulci. Of the six, five met the diagnosis of CTE. Findings from this study are important in demonstrating that RHIs are not essential for the pathogenesis of CTE, and we can, therefore, speculate that the higher incidence of CTE in football players may be due to factors other than head trauma. A drawback of the study, however, is that it did not set a threshold of pathology for diagnosing CTE. It has been established that small levels of hyperphosphorylated-tau are common amongst the general population [49], therefore, unless a commonly agreed diagnostic threshold is used across studies, some individuals with small levels of pathology may have been misdiagnosed with CTE. This is further reflected by the presence of ageing-related tau astrogliopathy (ARTAG), which shares features seen in CTE, and the two pathologies may well coexist [50].
A summary of the pathological studies describing the incidence of neurodegenerative disease in professional football players is given in Table 2.

Imaging Studies
One of the few ways of studying the brain and its structure in vivo is through imaging. Modalities such as MRI have been most useful in identifying changes in brain structure and volume. These have yielded interesting results concerning the health of current footballers' brains. Therefore, to further investigate the potential neurodegenerative effects of football, imaging studies have been carried out looking at a range of factors which may predict neurodegeneration.
Four separate studies [32,33,38,51], including one large longitudinal study conducted by Strauss et al. [32], used MRI to determine the impact of RHIs on white matter tracts, cortical thickness, and their relationship to cognition. They revealed that a high heading frequency was associated with poor white matter integrity and a reduction in bilateral parieto-occipital thickness, and this correlated with worse memory performance when compared to activity-matched controls. Another study [52] using magnetic resonance spectroscopy scanned 11 footballers with no history of head trauma and 14 age-and activity-matched non-contact sports controls. This study demonstrated a significant increase in neuroinflammatory markers, such as choline, myoinositol, and glutathione, in footballers, the latter two markers correlating with estimated lifetime headers. In addition to the neuroinflammatory role which may lead to the neurodegenerative effects seen in footballers, impaired cerebral blood flow regulation may also be implicated, as was demonstrated by a study conducted by Marley et al. [36]. The 16 amateur football players showed an equal baseline cerebral blood flow to age-and activity-matched controls. However, when challenged with hypo-and hypercapnia, the control group responded more effectively, as determined with the use of transcranial doppler ultrasound. To add to this, the footballers scored more poorly on cognitive testing, thus providing evidence that the sport contributes to neurodegeneration. Robust imaging studies have been conducted in athletes from different sports, for example in rugby players [53], and efforts should be made to adapt these methods for studying neurological outcomes in footballer players.
A summary of the cognition and imaging studies investigating the structural and neuropsychological correlates of football participation is given in Table 3.

Summary
Evidence from some pathological studies appears to suggest that football may be inflicting injurious effects on the brain, which may be leading to the development of neurodegenerative conditions, particularly CTE. The findings by Hales et al. [44] and Grinberg et al. [45] highlight the potential for an AD misdiagnosis in former footballers displaying possible symptoms of CTE. Lee et al. [46] and Ling et al. [47] also found the presence of CTE mixed with other neurodegenerative pathologies, such as AD. Given that CTE is largely associated with RHIs, these findings raise the question of whether physical trauma encountered during football, including headers and head-to-player impacts, plays a significant role in the manifestation of clinically diagnosed neurodegeneration. However, not all pathological studies corroborate this. Iverson et al.
found individuals without a known history of RHIs with pathological findings associated with CTE, with most meeting the diagnostic criteria for CTE [48]. This suggests that the development of the pathological signs of CTE could be due to reasons other than RHIs, challenging football as a potential risk factor. However, small levels of hyperphosphorylated-tau are common amongst the general population [49] in the form of ARTAG [54]. Therefore, the results may have been skewed by diagnosing CTE based on a baseline threshold that is not clinically significant. Nevertheless, pathological studies suggesting a potential link between football and CTE have several limitations, such as their cross-sectional nature and small sample sizes, which means that a definitive conclusion cannot be confidently drawn and further post-mortem analyses following prospective cohort studies are needed. The combination of a confident assessment of RHI burden during life; accounting for lifestyle and genetic factors; and detailed neuropathological analyses.
Studies looking at the cognitive effects of football have found that heading is associated with reduced cognition among players. Studies by Koerte et al. [38] and Bruno et al. [39] have especially highlighted that reduced cognition due to football can persist until retirement, which may be indicative of long-term neurodegenerative changes. However, Levitch et al. [31] and Strauss et al. [32] found that low to moderate levels of heading may lead to improved cognitive abilities. Additionally, some studies have refuted the association between heading and cognitive decline entirely. In a cohort of retired football players, Vann Jones et al. demonstrated that long-term football participation did not cause an increased risk of cognitive decline or dementia; although, as previously mentioned, the relevance of these findings is limited because of the small sample size and lack of a control group [43]. Overall, whilst there is little evidence to refute an association between heading in football and reduced cognition, the evidence in favour of an association is insufficient. This is primarily due to the limitations of the study methodologies. Regarding unintentional RHIs, recent studies by Stewart et al. [30], Lipton et al. [33], and Bruno et al. [39] suggest that an association with cognitive decline does not exist.
Most imaging studies agree that playing football can lead to neurodegenerative changes in some individuals. This includes changes such as poor white matter integrity, reduced cortical thickness, increased neuroinflammatory markers, and impaired cerebral blood flow. From the studies included in this review, only Kemp et al. failed to detect any neurodegenerative changes through imaging, but this may be attributed to the sample population being young [42]. It is also impossible to deny the beneficial health effects of playing football, with evidence showing that the sport improves overall fitness [55] and, through moderate heading, cognitive ability [31].

Conclusions
Overall, the literature suggests that a career in professional football is a possible risk factor for neurodegenerative disease. However, there may be other factors at play, such as genetic predisposition and environmental exposures. Nevertheless, there appears to be a paucity of studies providing counter evidence. Until more concrete evidence is found, we recommend that, rather than banning heading altogether, caution should be taken by minimising the amount of heading outside of matches. For example, introducing limitations of heading during training could provide a significant reduction in the RHI burden. We also recommend that leagues educate players about the possible neurological consequences of playing football. This will allow players to make informed decisions about the level of risk to which they expose themselves.
While most studies argue in favour of a correlation between football and long-term neurodegenerative disease, the evidence is insufficient to establish a causal relationship. This is mainly due to the methodological limitations of current studies. Additionally, we aimed to investigate the correlation between football and long-term neurodegenerative diseases, such as AD, PD, ALS, and CTE. However, most of the literature was focused on the link between football and ALS or CTE. This suggests a disproportionate focus on these disorders and highlights a need for further studies investigating a link with other neurodegenerative conditions, such as PD. Future work should focus on forming prospective, populationmatched studies incorporating serum or imaging biomarkers for TBI and RHI. In addition, considerable effort should be directed towards the untangling of the clinicopathological presentations of CTE and other neurodegenerative conditions. This would greatly improve current research and provide considerable insight when developing future RHI management guidelines.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.