Is There a Relationship Between Workload and Occurrence of Back Pain and Back Injuries in Athletes?

The back is subjected to a great deal of strain in many sports. Up to 20% of all sports injuries involve an injury to the lower back or neck. Repetitive or high impact loads (e.g., running, gymnastics, skiing) and weight loading (e.g., weightlifting) affect the lower back. Rotation of the torso (e.g., golf, tennis) causes damage to both, the lumbar and thoracic spine. The cervical spine is most commonly injured in contact sports (e.g., boxing, football). One of the factors that increases the odds of injuries in athletes is excessive and rapid increases in training loads. In spite of currently emerging evidence on this issue, little is known about the balance between physiological loading on the spine and athletic performance, versus overloading and back pain and/or injury in athletes. This scoping review aims (i) to map the literature that addresses the association between the training load and the occurrence of back pain and/or injury, especially between the Acute:Chronic Workload Ratio (ACWR) and back problems in athletes of individual and team sports, and (ii) to identify gaps in existing literature and propose future research on this topic. A literature search of six electronic databases (i.e., MEDLINE, PubMed, Web of Science, SCOPUS, SportDiscus, and CINAHL) was conducted. A total of 48 research articles met the inclusion criteria. Findings identified that fatigue of the trunk muscles induced by excessive loading of the spine is one of the sources of back problems in athletes. In particular, high training volume and repetitive motions are responsible for the high prevalence rates. The most influential are biomechanical and physiological variations underlying the spine, though stress-related psychological factors should also be considered. However, limited evidence exists on the relationship between the ACWR and back pain or non-contact back injuries in athletes from individual and team sports. This may be due to insufficiently specified the acute and chronic time window that varies according to sport-specific schedule of competition and training. More research is therefore warranted to elucidate whether ACWR, among other factors, is able to identify workloads that could increase the risk of back problems in athletes.


INTRODUCTION
Back injuries and back pain are among the most common health problems in athletes affecting their performance . Acute injury or pain can be caused by falling, being tackled in team sports, fighting in combat sports, or while lifting heavy weights (Caine and Nassar, 2005;Aasa et al., 2017;Bromley et al., 2018;Plais et al., 2019). However, much more common than acute incidents are chronic back problems (Alzahrani et al., 2019). In many sports, the back and spinal column undergoes elevated stress for a long time (Lawrence et al., 2006). This may result in inflammation around the vertebrae and back muscles, which sometimes causes injuries to the discs but much more often leads to upper or lower back pain . Sciatica, for instance, is back pain that also affects the back of the legs or even the feet . It can occur in cyclists who are in a flexed forward posture or athletes of water and swing sports who perform a great deal of trunk rotation. Particularly in sports with repetitive asymmetric loading, the sacroiliac joint dysfunction is a frequent cause of low back pain in athletes (Peebles and Jonas, 2017). In most of these sports, asymmetric loading causes side-to-side dysbalances that may greatly enhance susceptibility to back pain. A clear example is an association between repeated golf swings and golf-related low back injuries (Cole and Grimshaw, 2016). Trunk rotational power represents the measure that reflects this asymmetric loading during trunk rotation (Zemková et al., 2019a). Its values were found to be significantly higher on the dominant than nondominant side in golfers (∼15%), tennis (∼12%) and hockey players (∼14%) at lower or higher loads, whereas there were no significant between-side differences in a control group of physically active subjects (∼7%). This parameter is also able to reflect the specificity of training programs in the preparatory and competitive periods in canoeists, hockey and tennis players (Poór and Zemková, 2018). Similarly, wheelchair athletes, who suffer from spine curvature disorders, are at greater risk of back problems (Samuelsson et al., 2001). This is related to mainly those participating in sports that involve trunk rotational motions under loading and unloading conditions (Goosey-Tolfrey, 2010). A recent study demonstrated that wheelchair table tennis players exhibit lower mobility in the thoracic and lumbar regions of the spine during trunk flexion as well as lower lumbar inversion and pelvic retroversion than able-bodied athletes (Zemková et al., 2018b). Their limited range of motion (ROM) during trunk rotations and decreased posterior concavity contributes to lower trunk rotational velocity (Zemková et al., 2018b) and it may also increase the risk of spine injury. Velocity of trunk rotational movement is also compromised in older athletes, most likely due to their reduced ROM as a result of increased trunk stiffness with aging (Zemková et al., 2018a).
In spite of the fact that low back pain in athletes is among the most prevalent musculoskeletal condition with incidence rates of 1-30%, it is often neglected in research studies (Graw and Wiesel, 2008). The prevalence of low back pain is 1-94%, (the highest in rowing and cross-country skiing), and its point prevalence is 18-65% (the highest in rowing and the lowest in basketball) . Prevalence rates vary during an athlete's season with the highest rates observed during the peak season . The evidence on risk factors is mostly restricted to lumbar spine and hip flexibility, back muscle strength, trunk extensor/flexor endurance, and some anthropometric characteristics (Moradi et al., 2015;Trompeter et al., 2017). However, less attention has been paid to sport related factors. Among those, high training volume and repetitive motions contribute to the high prevalence rates . Athletes involved in impact sports are at risk for certain spinal pathologies that are associated with repetitive loading of the spine (Lawrence et al., 2006). For instance, higher incidence rates of spondylolysis and degenerative disk disease have been reported in elite athletes who participate in more intense training over a longer period of time than those who do not (Lawrence et al., 2006). Though muscle fatigue is often a symptom of back pain in athletes, factors like the number of training sessions per week only rarely have been investigated . The concept that compares the amount of acute workload performed in a 1week relative to a 4-week chronic workload can provide useful information for designing the optimal workload that would improve athlete performance, and at the same time, decrease the likelihood of back pain and related injuries.
The Acute:Chronic Workload Ratio (ACWR) is a simplified version of the fitness-fatigue model that was first introduced by Banister et al. (1975). It is definied as the ratio between training loads in recent periods (∼5-10 days) and over longer periods (∼4-6 weeks) (Foster et al., 1999;Hulin et al., 2016b). The ACWR is calculated by using the rolling or exponentially weighted moving average model (Hunter, 1986). Though both models are used in practice, the first one is unable to deal with the decaying nature of fatigue and fitness effects over time and the non-linearity of workload and the occurrence of injury; thus it may less precisely reflect variations of accumulated loads (Williams et al., 2017;Menaspa, 2017). The second one, which attributes a decreasing weighting in order to compensate for the load latency effects (Williams et al., 2017), is more sensitive to detect increases in injury risk at higher ACWR ranges during the preseason and in-season periods (Griffin et al., 2020).
These models have been used to better control and understand the training process. Several studies have documented the association between the actual and modeled performance in a variety of individual sports (e.g., Taha and Thomas, 2003;Jobson et al., 2009) or injuries in team sports (e.g., Hulin et al., 2016a;Bowen et al., 2017;Murray et al., 2017). Contrary to this, less attention has been paid to the investigation and use of these models in association with the occurrence of injuries or pain that especially involve the spine and back muscles. So far the training volume, internal and external loads in different types of training, sport-specific demands on the spine in relation to back pain and/or injury have been primarily investigated (e.g., Maselli et al., 2015;Newlands et al., 2015;van Hilst et al., 2015). There is however, no sufficient evidence to determine the relationship between ACWR and the occurrence of back problems in athletes. A question also remains as to whether ACWR can be predictive of an athlete's risk of back pain and/or injury.
To this end, two main questions were addressed in this scoping review: (1) Is the training load associated with the occurrence of back problems in athletes of individual and team sports? (2) Is there a relationship between the ACWR and back pain and/or injury in these athletes? Accordingly, this scoping review aimed at (i) mapping the literature that addresses the association between the training load and the occurrence of back pain and/or injury, especially between the ACWR and back problems in athletes of individual and team sports, and (ii) identifying gaps in existing literature and proposing future research on this topic.

METHODS
The article was designed as a scoping review (Armstrong et al., 2011). In order to answer the above questions and to identify a gap in existing research in the field, a literature search was provided. Electronic literature was searched on MEDLINE, PubMed, Web of Science, SCOPUS, SportDiscus, and CINAHL databases. Further searches were conducted on Elsevier, SpringerLink, EBSCOhost, and Google Scholar. Besides articles in peer-reviewed journals, also conference proceedings were analyzed. The search was confined to studies closely associated with the major topic of this review, i.e., investigating the relationship between training loads and the occurrence of back problems in athletes, particularly between the ACWR and back pain and/or injury in athletes of individual and team sports. Our primary focus was concentrated on outcome measures of the ACWR and data related to back problems in athletes of individual and team sports. Using this approach, however, led to the identification of a limited number of studies that were able to meet the eligibility criteria for this review. Therefore the search was later widened to include all relevant studies that investigated the relationship between the training load and back pain and/or injury in athletes. Together these help us to identify gaps in the current literature regarding the training load, especially the ACWR and their associations with back pain and/or injury in athletes of individual and team sports, and suggest a proposal for future research.
The target population was athletes of individual (cricket, cycling, dancing, diving, golf, gymnastics, horseback riding, jiu jitsu, judo, powerlifting, rowing, running, skiing, swimming, triathlon, weightlifting, wrestling) and team sports (basketball, florball, handball, hockey, soccer, tennis, volleyball). The most frequent terms "Acute:Chronic Workload Ratio, " "back injuries, " "back pain, " "back problems, " "competition, " "external training load, " "fitness level, " "injury risk, " "internal training load, " "low back pain, " "neck/cervical pain, " "non-contact injury, " "thoracic pain, " and "training" were combined with particular individual and team sports. Additional searches were performed by using the words from subheadings, such as factors contributing to back pain and/or injury in athletes with respect to their localization, intensity, frequency, and duration. The key inclusion criterion was that studies involved athletes of individual and team sports with back pain and/or injury, a specified training program, and objective or subjective measures relevant to this review. Studies were excluded if they were incomplete (abstracts, etc.), not peerreviewed, did not contain original research, and were not written in the English. Studies that failed to meet these criteria for this review were excluded. Figure 1 represents particular phases of the search process.

Overview of Studies Dealing With Back Pain and/or Injury in Athletes of Individual Sports
The key data elements that were sought for each study were categorized as follows: (1) study design, (2) study population (number of athletes, sex, age, type of sport, and performance Frontiers in Physiology | www.frontiersin.org level), (3) training characteristics, (4) low back pain information, (5) diagnostic methods used for identification of lumbar spine, low back pain and lower back injuries, and (6) study outcomes.
Out of the 31 research articles (Table 1), 21 studies (68%) included both sexes, six studies (19%) dealt with male athletes, and four studies (13%) focused on female athletes. A total of 2999 athletes of mean age of 26.1 years was evaluated, out of which 1819 were males and 1180 were females. Fifteen studies were conducted with elite athletes, nine studies assessed non-elite athletes, and six studies were conducted in both elite and non-elite athletes. For the purpose of this review, elite athletes were defined as those competing professionally at national level, international level or at the Olympic Games. Athletes competing at district and regional competition levels, as well as recreational and amateur athletes were defined as non-elite. Also athletes competing at high schools, colleges, or universities were considered as non-elite.
Lumbar spine abnormalities and alterations, low back injuries and pain were diagnosed in many different ways: using imaging systems such as magnetic resonance imaging (MRI), radiographs, CT, Spinal Mouse, motion analysis, ultrasonic measurements, through clinical examination, physical examination, various questionnaires, surveys, body charts, interviews, pain scales, and combinations of the above.

Overview of Studies Dealing With Back Pain and/or Injury in Athletes of Team Sports
Out of the 17 research articles ( Table 2), seven studies (41%) included both sexes, six (35%) were focused on male and four (24%) on female athletes. The number of athletes in respective studies ranged from 29 to 1110 depending on the research methods used (retrospective vs. prospective). Ten out of these studies (56%) dealt with senior players, seven (39%) with junior players and one study (5%) with both age categories. Participants were top athletes (4 studies), elite players (4), nonelite players of different performance levels (5), and players of more than one performance level (5). Top athletes were considered those competing at the international level and taking part in top international competitions (Olympic Games, World Tours, and World Championships). Elite athletes were defined as the members of national teams not competing at the highest level or young athletes who were members of academies for talented youth. Non-elite athletes were from sport clubs, high schools, and colleges.
Regarding the team sports category, the following groups of athletes were included: soccer players (7 studies), volleyball players (5), beach volleyball and handball players (3), field hockey, floorball, handball and basketball players (2), and the players of different types of rugby (1). Studies included were from the following countries: Norway (4 studies), Germany (3), United Kingdom, Sweden, and Finland (2), Switzerland, United States, Australia, Iran, and Portugal (1). Players included were from the same countries except of three studies which were conducted at major international events such as World Championships or World Tours.
This review includes two main types of studies: prospective cohort studies and retrospective studies. Low back injuries and pain or lumbar spine abnormalities were diagnosed in the following ways: using imaging systems (MRI), through physical examination by medical staff or physiotherapists, questionnaires, interviews, and their various combinations.

The Association of Workload With the Occurrence of Back Problems in Athletes of Individual and Team Sports
There is no conclusive evidence to date to demonstrate that the cause of low back pain in athletes of individual sports lies in undue abnormalities of the spine. Only one study (Koyama et al., 2013) confirmed lumbar disc degeneration as a statistically significant variable accounting for low back pain. However, no direct association between the training load and back injuries or abnormalities in the spine was confirmed. Nonetheless, there were trends toward an increased risk of injuries with increasing levels of performance and high physical loads.
The highest prevalence of low back pain was identified in field hockey, floorball, rugby, and beach volleyball (Nosco et al., (2) Pain deteriorated with training (10) Pain improved with training (2) -MRI examination -Neurological examination of lower extremities and the spine -Back pain questionnaire -MRI abnormalities were observed in 3 divers at baseline -MRI abnormalities were observed in 4/7 divers trained more than 8 h/week (at follow-up) -MRI abnormalities were also found in 8/11 divers trained less than 8 h/week -High prevalence of back pain and MRI changes was observed; their causal relationship was confirmed -Young age was associated with high frequency of back pain -Growth spurt was associated with high risk of occurrence of degenerative changes of the thoracolumbar spine -Mild structural scoliosis in the thoracic spine in 1 diver -Tenderness in the thoracic in 8 divers -Tenderness in the lumbar spine in 5 divers -5 divers presented MRI abnormalities and tenderness in the thoracolumbar spine -MRI abnormalities but no tenderness in 5 divers -5 divers presented tenderness but no MRI abnormalities in the spine -MRI abnormalities or tenderness in 3 divers -Abnormalities in the thoracolumbar spine at baseline MRI examination in 12 out of 18 divers -Abnormalities in the thoracolumbar spine at follow-up MRI examination in 12 out of 17 divers -20 new abnormalities in the spine were observed at follow-up -LBP group cycled significantly more in a week than NLBP group -Training volume of 160 km/week or more was associated with 3.6 times higher incidence of LBP than in cyclists with training volume less than 160 km/week -Training volume less than 160 km/week reduced the risk of new occurrence of LBP in recreational cyclists Muyor et al., 2011 -60  Running related injury was defined as any pain of musculoskeletal origin attributed to running and associated with absence at least one training unit (Bovens et al., 1989;Macera et al., 1989;van Middelkoop et al., 2007van Middelkoop et al., , 2008 -10-point pain numerical rating scale -Survey -Running related injuries in lumbar spine in 12 (14%) runners, duration of LBP: 2.4 ± 0.8 weeks, pain intensity: 5.2 ± 2.5

(Continued)
Frontiers in Physiology | www.frontiersin.org   -Similarity in incidence of LBP in the FL players and controls (56% in FL players, 55% in controls) -Onset age of LBP for controls and the FH was 16.45 ± 2.12 and 16.23 ± 1.80 years -Duration of symptoms was less than 3 weeks in controls (85%) and FH group (82%) -Episode of LBP of 1 month or longer was 15% in controls and 14% in FH -Similarities in pain distal to the buttock in both groups -Non-significant differences in the incidence of LBP and pain characteristics between the groups -In both groups a high incidence of LBP (>50%) with an onset of LBP at a mean age of approximately 16 years

(Continued)
Frontiers in Physiology | www.frontiersin.org at the beginning of spring season -The highest incidence in the knee, the lower leg, and the back (39/18/16%; 0.6, 0.3, and 0.2/1000 h of soccer, respectively -Back the third of most common body location of acute and overuse injuries -National team players and players in the three highest ranked teams although exposed to more playing hours during the year not different in injury incidence as non-national players -Injured players higher practice/game ratio than non-injured players -High amount of re-injuries (>50%) Age of elite/semi-elite/non-athletic controls 23.3 ± 4 years, 23.8 ± 4 years, and 23.9 ± 4.5 years --Almost linear trend of increasing LBP severity from non-athletes, to the semi-elite and elite athletes -Levels of the sensory, affective and total pain score significantly higher in elite athletes -Elite athletes approximately two times more experienced discomforting or greater LBP, and less likely no LBP as other groups; semi-elite athletes less likely experienced discomforting or greater LBP; and non-athletes more likely not experienced LBP (χ 2 = 18.67, p < 0.001) -The age of first episode of LBPbetween groups non-significant differences -Elite athletes 2-4 times attributed sporting activity as a cause of current LBP -Elite players had significantly higher levels and more frequent LBP -LBP in elite players was more attributed to sporting activity when compared to semi-elites and non-athletes -The difference of onset of the first time LBP non-significant between groups -Reported etiology between groups is different Frontiers in Physiology | www.frontiersin.org 1999; Pasanen et al., 2008;Hoskins et al., 2009;Haydt et al., 2012;Külling et al., 2014;Fett et al., 2019). Among pain caused by overuse, low back pain was the most often reported pain with an incidence of 20-86% depending on the age category, sex, performance level, and the time of occurrence. In soccer, volleyball, and handball, low back pain was the third most common pain caused by overuse (Augustsson et al., 2006;Anza et al., 2013;Bere et al., 2015;Tunås et al., 2015;Bowen et al., 2017;Aasheim et al., 2018). In soccer, pain caused by overuse of groin and hip with a rate of 38-57% was the most common, followed by low back pain caused by overuse, whereas in volleyball and handball overuse injuries of shoulder and knee at a rate of 46-57% and 12-59%, respectively, were the most common, followed by low back pain. The risk factors also include training and match load, which is characterized by the specific movement pattern of sport games, such as unilateral load, repetitive static flexion, hyperextension combined with rotation of the spine, jumps, impacts etc. (Snellman et al., 2001;Pasanen et al., 2008;Bere et al., 2015;Fett et al., 2019). However, the evidence is not unambiguous. The findings are mostly based on comparisons of athletes and non-athletes, and players with other athletes, top, elite, and nonelite players (Bahr and Reeser, 2003;Jacobson and Tegner, 2007;Hoskins et al., 2009;Anza et al., 2013;Bere et al., 2015;Haag et al., 2016;Aasheim et al., 2018).
In recent years, more attention has been directed toward objective quantification of training and match load, especially using the ACWR, and the examination of its relationship with injury risk (Griffin et al., 2020). Although 15 studies dealt with this issue in team sport games, only 2 of them can be included in this review (Bacon and Mauger, 2017;Bowen et al., 2017). In other studies, there was no distinction between the type and site of injuries.
Overall, the literature supports the association of training load with the occurrence of back problems in athletes of individual and team sports. A review by Moradi et al. (2015) identified many potential general and sport-specific risk factors for back pain in athletes, however, the evidence exists only for previous low back pain, decreased lumbar extension or flexion, and high body weight. Differences in the association between potential risk factors and back pain may be ascribed to the type of sport, level of competition, and training characteristics (Moradi et al., 2015). Evidently, properly designed training programs might not affect postural and core stability or increase the risk of injuries. For instance, lower lumbar lordosis and thoracic kyphosis, as well as anterior pelvic tilt while standing were identified in Latin American style professional dancers (Muyor et al., 2017). Though specific postures and movements in dance modify their spinal curvatures, it does not alter the spinal morphology in standing when compared to non-dancers. Further, Belavý et al. (2017) demonstrated that chronic running is associated with better intervertebral discs (IVD) composition and IVD hypertrophy. Generally, pain perception and processing is different in athletes as compared to normally active controls (Tesarz et al., 2012). This suggests a compensatory response of the endogenous antinociceptive system to the noxious input during exhaustive endurance training (Tesarz et al., 2013). More specifically, an overstressing of the endogenous pain inhibitory pathways may lead to exhaustion over time (Tesarz et al., 2013). This may result in disinhibition of pain processing when the acute pain is transmitted in the chronic pain and spatial pain spreading (Tesarz et al., 2013). On the contrary, the endogenous pain inhibitory pathways may be protected from chronic overstressing over time by a shift in the activation threshold, which may increase the efficiency of pain inhibition (Tesarz et al., 2013). Specific changes in pain thresholds and higher tolernce to pain at rest in athletes (Tesarz et al., 2012) has to be also taken into account when interpreting self-reported back pain and its relationship with physiological loading of the spine in athletes.
Actually, training is a protective factor against injury. As shown, high chronic workloads decrease the injury risk (Gabbett and Domrow, 2005). A lower risk of sustaining a subsequent injury was reported in athletes of team sports when they trained over 18 weeks before their initial injury (Gabbett and Domrow, 2005). A decreased risk of injury is also associated with welldeveloped physical qualities (Quarrie et al., 2001;Gabbett and Domrow, 2005;Gastin et al., 2015). Thus, progressively increased high training loads not only improve physical fitness of players but may also protect against injury (Gabbett, 2016). For instance, heavier rugby players with faster speed generate greater impact forces which may increase the recurrent rates of contact injuries . However, greater intermittent running speed at high intensity as well as muscle strength and power may decrease the risk of injuries in these players . This indicates that not only overtraining but also undertraining may increase the injury risk (Lyman et al., 2001;Orchard et al., 2009;Cross et al., 2016).
So far, the relationship between the ACWR and back pain/or injury in athletes remains unclear. This is mainly due to the limited evidence and varying methodological quality of particular studies. Nonetheless, it is most likely that sudden spikes in acute training loads may have detrimental effects, whereas high chronic loads and adequate strength and endurance of core muscles could provide protective effects on increasing the occurrence of back pain and/or injury. A structured back-strengthening program has been recommended for reduction of back pain experienced by athletes (Trainor and Wiesel, 2002). In practice, however, functional strength training for back pain prevention can be similar to programs for back pain rehabilitation (Wirth et al., 2017). Most exercises have not been tested for their effectiveness and compared with the load used for strength training. Neither are core strength exercises more effective than traditional resistance exercises (Smith et al., 2014) in individuals with back pain (Macedo et al., 2016;Saragiotto et al., 2016). Adaptations in voluntary activation of trunk muscles to core stability exercises have provided a basis for exercise guidelines (Wirth et al., 2017). However, adaptations of morphological structures that are essential for the core stability have not been sufficiently addressed in research studies. Therefore, the guidelines used for rehabilitation of back pain are insufficient for strength training in athletes (Wirth et al., 2017). Currently, there is no evidence-based exercise program for athletes with back pain and/or injury. Most studies investigated only a part with a narrow group of athletes who usually performed different types of exercise (Daniels et al., 2020). Recent evidence suggests that poor load management is a main injury risk factor (Soligard et al., 2016). Recently, the IOC provided the consensus statement that include guidelines for monitoring of training, competition and psychological load, prescription of training and competition load, athlete well-being and injury (Soligard et al., 2016).

Gaps in Current Studies Investigating the Workload in Relation to Back Problems in Athletes and Proposals for Future Research
One of the former studies reported that the incidence of injury is increased when the duration, intensity, and load of training sessions and matches in rugby league is increased (Gabbett, 2004a;Gabbett and Domrow, 2007). There is a U-shaped relationship for 4-week cumulative loads with an increase in injury risk with higher loads (Cross et al., 2016). The risk of injury in professional rugby union players increases if they have high 1-week cumulative loads or large week-to-week changes in the training load (Cross et al., 2016). Further, higher risk of injury is also associated with a large increase in acute workload in elite cricket fast bowlers (Hulin et al., 2014). A sharp increases in running workload, the ACWR > 2.0 for either total distance or a high-speed distance, increases the likelihood of injury in elite footballers in both a given and the subsequent week (Murray et al., 2017). Controlling for training load in a given week may decrease the odds of injury in the subsequent week. Windt et al. (2017) demonstrated that increasing participation in preseason training may decrease the risk of in-season injury in elite rugby league players. Though increased amounts of high-velocity running are associated with higher risk of soft-tissue injury of the lower body, distances covered at moderate and low speeds are protective against this injury . Decreasing the training load during an early-competition phase can reduce the odds of injury without compromising the agility performance in athletes of collision sports (Gabbett and Domrow, 2007). Likewise, reductions in pre-season training loads decrease training injury rates and at the same time increase the maximal aerobic power in rugby league players (Gabbett, 2004b). It seems that high chronic training loads may decrease the risk of injuries in athletes. For instance, elite rugby league players are more resistant to injury when train at a high than a low chronic workload with ACWRs 0.85-1.35, whereas they are less resistant to injury when ACWRs is ∼1.5 and they are subjected to"spikes" in acute workload (Hulin et al., 2016b). Very high and high chronic workloads have protective effects against match injury following shorter recovery periods between matches (Hulin et al., 2016a). Alternatively, a high aerobic capacity and playing experience protect injury in elite Gaelic football players against rapid changes in workload and high ACWRs > 2.0 (Malone et al., 2017a). Players better tolerated increased distances and exposures to maximal velocity when train at higher than at lower chronic loads (Malone et al., 2017b). Over-and under-exposure them to maximal velocity increased the injury risk (Malone et al., 2017b). Accordingly, high chronic workloads, combined with reductions in acute workloads before competition, decrease the risk of injury but might also improve sporting performance (Gabbett, 2016). In general, ACWRs ≥ 1.5 (i.e., greater acute than chronic training load) are associated with a higher injury risk than ACWRs in the range from 0.8 to 1.3. While most studies showed the association of Acute:Chronic markers with risk of injuries, its application in predictions of injuries is questioned (Fanchini et al., 2018;McCall et al., 2018). Rather, it can identify workloads that athletes should use to make a decision when risk of injury is increased or decreased (Hulin and Gabbett, 2019). Therefore, the ACWR should not be viewed in isolation.
Contrary to these studies, the suggested association between training loads and back pain and/or injury is based on case studies, expert opinions, and unpublished data, whereas some other findings are controversial. Analysis of the literature identified several limitations in relation to workloads and back problems in athletes. First, the studies that use the same methods are limited. Second, injury identification is often based only on players' self-assessment. Third, the exposure hours are calculated as average values for training and playing, and in most cases are based on data reported by individuals. Additionally, prevalence of back pain and/or injury in some sports can be affected by sample size. Finally, numbers of injury definitions and time occurrence categories make a comparison among studies more difficult. Further studies using a wider methodological approach are necessary to deepen the knowledge and understanding the association between the training load, especially ACWR and back pain and/or injury in athletes of individual and team sports.
Because questionnaires are mainly used for identification of back pain, more objective methods evaluating spine curvature and flexibility (Muyor et al., 2017), strength, power and endurance of core muscles (Zemková et al., 2016a(Zemková et al., , 2019aZemková and Jeleň, 2019) and hamstrings (Zemková et al., 2015), as well as postural and core stability after perturbations (Zemková et al., 2016b) should be applied. In addition to sportspecific testing methods, basic physical fitness tests can be used (Zemková and Hamar, 2018). Athletes undergoing the high-level of training demonstrated similar deconditioning of the lumbar extensor muscles and used similar strategies to maintain spinal stability after unexpected perturbations to ensure the spine from pain and damage as compared to non-athletes with low back pain (Catalá et al., 2018). Assessment of training interventions for strengthening the core muscles is also necessary, not only in adult athletes with a predisposition to back pain and/or injury but also in adolescents (Zemková and Hamar, 2018). There is increasing evidence that an experience of back pain in 14year-old individuals may be associated with pain in adulthood (Coenen et al., 2017). Repetitive flexion and extension loadings of young Functional Spinal Units causes MRI and histological changes in the growth zones and endplates that are most likely first signs of fatigue and an explanation for the spine injuries in adolescent athletes (Thoreson et al., 2017a). In spite of the fact that back pain and/or injury is one of the most prevalent diagnosis in athletes, the mechanism of back problems is poorly understood so far (Ball et al., 2019). Research studies conducted on athletes with back pain highlight physiological and biomechanical mechanisms as influential, whereas psychological factors are often neglected. Heidari et al. (2017) emphasizes the importance of mainly stress-related factors in prevention and rehabilitation of back pain. Therefore, more longitudinal studies are required in comparison with frequently conducted shortterm experiments in order to investigate long-term changes in strength of core muscles and neuromuscular control of spine stability in athletes.
Collectivelly, the association between the ACWR and injury in a variety of sports has been well established. However, our review highlights the paucity of research studies directly evaluating the association between the ACWR and back pain and/or injury across a larger range of sports. The inconsistency in the methodology of calculating weekly training loads and missing specification of injuries are main limiting factors that restricted comparison between findings. To reduce the variability between studies, researchers need to clearly define the ACWR methodology and specify back problems. The acute and chronic time periods used for assessment of the ACWR may depend on the specific nature of particular sports. Future investigations would benefit from more information on the fundamentals of competition and training in particular sports, physical fitness attributes of athletes, and other back pain and/or injury risk factors.

CONCLUSION
Although the relationship exists between the training load and the occurrence of back problems in athletes of individual and team sports, this is not specifically related to ACWR. In comparison with biomechanical, physiological or stress-related psychological factors, sport-specific patterns underlying high prevalence rates are not fully elucidated. The fatigue of trunk muscles induced by excessive loading of the spine is one of the sources of back problems in athletes. Practitioners can address this issue by load management via the use of the ACWR. However, they should be aware that the acute and chronic time window may vary according to specific structure of individual and team sports and their training and competition schedules. Though this may provide important feedback on athlete's workload, the questionable is the sensitivity and specificity of the ACWR to predict back problems. This is mainly due to the limited evidence on applications of this systems-model approach for identification of risk of back pain and back injuries. Future research should focus on investigation of ACWR validity in combination with multifaceted monitoring systems in prevention of back problems in athletes of both individual and team sports. Nonetheless, the ACWR can be used as an effective tool for monitoring the athlete's responses to training thereby enable practitioners to design workloads that decrease the occurrence of back problems. It seems that high chronic workloads and adequate muscle strength and endurance have a protective effect on increasing the occurrence of sport-related injuries.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation, to any qualified researcher.