sports medicine Volume 39 Issue 8 August 2009

Sports Med 2009; 39 (8): 607-613 0112-1642/09/0008-0607/$49.95/0

LEADING ARTICLE

ª 2009 Adis Data Information BV. All rights reserved.

The Evolution of Sport Psychiatry, Circa 2009 Ira D. Glick,1 Ronald Kamm2 and Eric Morse3 1 Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Stanford, California, USA 2 Sport Psychiatry Associates, Oakhurst, New Jersey, USA 3 Department of Sports Medicine, North Carolina State University, Raleigh, North Carolina, USA

Abstract

Over the past three decades, the world of both amateur and professional sports has expanded greatly and become more complex. In part related to these changes – and relatively unknown to sports medicine practitioners – the field of sport psychiatry has steadily evolved and grown. This paper focuses on what these changes have been. A sport psychiatrist is a physician-psychiatrist who diagnoses and treats problems, symptoms and/or disorders associated with an athlete, with their family/significant others, with their team, or with their sport, including spectators/fans. The primary aims of the specialty are to (i) optimize health, (ii) improve athletic performance, and (iii) manage psychiatric symptoms or disorders. The training includes medical training to provide knowledge and skills unique to physicians; psychiatric training to provide knowledge and skills inherent in that field, and training and/or experience in sport psychiatry to provide knowledge and skills about psychiatric aspects of sports. The sport psychiatrist first makes an individual, family-systems and phenomenological diagnosis of the clinical situation. Based on this evaluation, he sets goals for not only the athlete, but also for significant others involved. He delivers treatment based on the psychiatric disorder or problem using a combination of medication, psychotherapy or self-help group interventions plus strategies targeted to specific sport performance issues. Evolution of the International Society of Sport Psychiatry as well as the field, including incorporation into school and professional team sports, is described along with a ‘typical day’ for a sport psychiatrist. Case examples, a training curriculum and core literature are included.

Over the past three decades, the world of both amateur and professional sports has greatly expanded. In part related to this phenomenon – and still relatively unknown to sports medicine practitioners – the field of sport psychiatry has steadily evolved and grown.

In 1990, Dan Begel[1] carved out a new subspecialty in psychiatry. He envisaged sport psychiatry as ‘‘the application of the principles of and practice of psychiatry to the world of sports,’’ and identified some of its developmental, occupational, therapeutic and research aspects.

Glick et al.

608

Three observations, in particular, justified the elevation of sport psychiatry to a specific subspecialty. ‘‘First, an athlete’s state of mind has a significant impact on performance; second, participation in sports affects the mood, thinking, personality and health of the participant in specific ways; and third, the psychiatric care of the athlete must be adapted to the athletic context in order to be effective.’’ In the years since the publication of this article, sport psychiatry has continued to evolve. In this current paper, we the authors – three senior members of the International Society for Sport Psychiatry (ISSP) – describe how the field has grown, i.e. what a sport psychiatrist does given the evolution of psychiatric issues associated with amateur and professional sports over the last two decades. We should say, at the onset, that the field suffers from a lack of controlled studies (and data) on incidence, phenomenology or treatment of psychiatric disorders in athletes. There are thus only very limited systematic studies of athletes with psychiatric disorders and no controlled treatment studies that suggest that athletes are to be treated differently to the general clinical population. As such, controlled studies on psychiatry patients who are not primarily athletes are often extrapolated to athletes.

1. What is a Sport Psychiatrist? A sport psychiatrist is a physician-psychiatrist who diagnoses and treats problems, symptoms and/or disorders associated with an athlete, with their family/significant others, with their team, or with their sport, including spectators. The primary aims of the specialty are to (i) optimize physical health, (ii) ethically improve athletic performance including optimizing coping mechanisms and positive psychological strengths, and (iii) manage psychiatric symptoms or disorders. It is different from both general internal medicine and from psychology. The sport psychiatrist can recognize common medical conditions in the same way the internist can recognize common psychiatric conditions like severe mania or depression, but the psychiatrist has the exª 2009 Adis Data Information BV. All rights reserved.

pertise necessary to manage these disabling conditions over the lifetime of the illness with the aim of maintaining athletic performance. Similarly, the sport psychiatrist has the competency to not only prescribe medication, but also provide individual, family or even group therapy (modalities that are at the heart of psychology practice). Psychiatrists do not carry out psychological testing. The training of a sport psychiatrist includes medical training to provide knowledge and skills unique to physicians; psychiatric training to provide knowledge and skills inherent in psychiatry, and training and/or experience in sport psychiatry to provide knowledge and skills about psychiatric aspects of sports (see below).

2. What Does a Sport Psychiatrist Do? 2.1 Diagnosis

The sport psychiatrist initially always makes a detailed diagnosis of a clinical situation and then tries to determine the psychopathology underlying the presenting problem, symptom or disorder (such as anxiety, substance abuse or an eating disorder). The sport psychiatrist must possess special skills in interviewing not only the athlete but also the family or significant others, or the system (i.e. the coach, family, agent, team-mate, owners, league and all the individuals involved in the recreational and business aspect of a particular sport). There is an attempt at understanding the impact of the athlete’s family on the development of the athlete’s mind, and the role that sports play in the family system, as well as specific aspects of sports, such as aggression, anxiety, etc. Obviously, there is a special focus on mental illness as well as on interpersonal problems and disorders. A developing literature[2-5] has helped psychiatrists not familiar with the field see how to approach an athlete who appears to be having a ‘mental problem’, and sport psychiatrists have written articles in non-psychiatric journals that have raised the awareness among sports medicine physicians, and others, to sport psychiatry perspectives on their patients and teams/systems. Sports Med 2009; 39 (8)

The Evolution of Sport Psychiatry, Circa 2009

Table I. Treatments in sports psychiatry Medication management, i.e. pharmacotherapy Psychotherapy: individual, family, group, cognitive behavioural therapy, etc. Performance-enhancing techniques and strategies Substance abuse/dependence management and treatment Mental skills training Self-help groups

2.2 Goals

The goals of treatment should be clearly delineated. These include goals for the athlete and, in some cases, the team or the family. The family, in fact, can be very important in helping to develop goals for the athlete, and to aid in treatment adherence. Derrick Adkins won the 400 metre hurdles at the Summer Olympics in Atlanta in 1996, and was being treated with a selective serotonin reuptake inhibitor (SSRI). He had to taper the SSRI prior to the Olympic trials and the Olympics, in favour of serotonin (5-HT), because of fatigue and slower running times. After the euphoria of winning the Olympics, the patient did not feel it necessary to go back on the SSRI, but his mother noticed him becoming more and more depressed while watching him run on TV in the European tour. The patient was restarted on SSRIs and made a recovery, although his competitive times suffered (table I).[6] 2.3 Treatment

There are a number of ways to categorize the treatments a sport psychiatrist delivers. First, one can describe the evidence-based treatment for a specific psychiatric disorder associated with a particular athlete (e.g. attention deficit hyperactivity disorder [ADHD], eating disorders, substance abuse [including performance-enhancing drugs], etc.). Second, one can subdivide treatment according to the particular modality (or more commonly combination of modalities) prescribed: here we include psychotherapy (individual, family, group, etc.), pharmacotherapy, self-help groups such as Alcoholics Anonymous, etc. Third, one can provide interventions for a specific problem, i.e. suicide, retirement issues, sex or racial ª 2009 Adis Data Information BV. All rights reserved.

609

issues, etc. Finally, one can describe specific strategies such as those found within the rubric of mental skills training, targeted to specific sport performance issues (lack of aggression, lack of confidence, too much arousal) or to unique strategies for a specific sport (e.g. tennis, golf, boxing, football, gymnastics, etc.). Needless to say, the sport psychiatrist has specific skills and techniques to use in his or her work with teams, whether the problem is a high incidence of antisocial behaviour on the team or is related to performance issues or coach-playerteam interpersonal dynamics.[7] There are now professional and team assistance programmes managed by psychiatrists – although their efficacy and effectiveness are still unstudied. The following three cases illustrate some of these issues. 2.3.1 Case Number 1

A boxer in his junior year of high school had been evaluated extensively for ‘abdominal pain’. It soon became apparent that the boxer was suffering from DSM-IV Eating Disorder NOS (not otherwise specified).[8] Pre-occupation with weight had begun when the coach had insisted that he drop down a weight class in order to help his club. Extensive family therapy and psychotherapy helped the patient to recover. Since the patient wanted to compete the following year, the sport psychiatrist contacted the coach regarding the necessity of not putting any pressure regarding attaining a specific weight on the athlete. Followup revealed that the coach disregarded the psychiatrist’s advice, precipitating another episode of anorexia, which resulted in the athlete leaving the sport. 2.3.2 Case Number 2

This case illustrates a sport psychiatrist being contacted via the internet. All of the authors have a website, in addition to the website of the ISSP (www.TheISSP.com). This website has been in existence for almost 10 years and we have handled questions from coaches, parents and athletes, often about youth sport issues. The websites have served as a source of information for the Sports Med 2009; 39 (8)

Glick et al.

610

general public and medical/sport medicine professionals as well. A father contacted a sport psychiatrist because his son, a junior playing lacrosse for a local high school, was suddenly denied significant playing time. Two freshmen had come onto the team and there was the appearance of favouritism, in that one of the freshman’s parents worked at the school, and rumour had it that the coach had promised the parent that the freshman would get significant playing time. The second freshman’s father was the coach of a travelling team on which the patient’s coach’s son was a player. The patient had been a stellar player, and, according to the father, spectators and other coaches were perplexed as to why he would be benched. The benching and the attention surrounding it resulted in a performance decline, and the patient left the team midway through the season. The father contacted us via the website to see whether the boy should switch schools for his senior year, whether he should quit the sport, or whether there was a way to salvage his senior year at the same high school. 2.3.3 Case Number 3

A college football player complained of being picked on by his coach and singled out. He was diagnosed with major depressive disorder and had to leave the team while he was treated with psychotherapy and medication. The depression improved but the decision was made that the player should transfer to another college. A petition was made to the National Collegiate Athletic Association (NCAA) that the patient not lose the year because he had a serious medical disorder. The request was granted and the patient went on to have a successful college career at another university. There have been articles by sport psychiatrists that have helped bring about changes in a sport. Tofler et al.[9] wrote their ‘achievement by proxy’ article in the mid-1990s. It was a significant factor in influencing US gymnastics to raise the age of eligibility for the Olympics to 16 years, as there had been concern in the Sport Medicine and Sport Psychiatry communities that the intense pressure to make the Olympics on 12-, 13- and ª 2009 Adis Data Information BV. All rights reserved.

14-year-old female gymnasts was leading to an increased rate of physical and psychological disorders in this population. 2.4 Evolution of the Field

Over the last two decades, sport psychiatrists (and other mental health professionals) have moved from a peripheral position to becoming incorporated into college and professional team sports.[10] The reasons underlying this are multiple, including not only competitive pressures to win, but also the need to help handle the associated fallout that may follow bizarre behaviour on and off the field, an overemphasis on or inappropriate aggressive tactics or cheating, family problems, the anabolic steroid scandals (baseball, track and cycling most prominently), and finally the abuse of substances used by athletes (in part as a way to manage stress) seen in virtually every sport. Physicians are now heavily involved in the substance-abuse problem, as we have described above, and every professional league has an ‘expert’ – often a psychiatrist – as part of the league structure. They treat Axis I Disorders,[8] e.g. anxiety disorders or Tourette’s disorder. Sport psychiatrists are now being integrated into the treatment team for problems such as gambling, as well as being central to understanding the effects of injuries in certain sports (e.g. brain damage in soccer, football, boxing, etc.) and of career termination issues. In that regard, one of the authors (RK) serves on the board of a boxing organization, Fighters Initiative for Support and Training (FIST). This organization has helped to develop a programme dedicated toward helping fighters transition to another career before brain damage occurs. By way of example, here is a description of a typical day of practice for a pioneering full-time sport psychiatrist (EM): I work with athletes two afternoons a week at my local university’s training room as part of the sports medicine team. Besides providing psychiatric services, I do team substance use prevention work, team building exercises to improve group dynamics, and consultations with Sports Med 2009; 39 (8)

The Evolution of Sport Psychiatry, Circa 2009

coaches. I also see athletes a few hours a week in my private practice. ADHD is probably the most common mental illness I treat. Stimulants are still first-line treatment. Since they can be performanceenhancing, we do extensive testing and clinical interviews by more than one clinician to make a certain diagnosis, rule out other issues, and provide documentation for applying for special accommodations. The proper documentation in the Sports Medicine chart prevents difficulties with positive urine drug tests for stimulants. Any athlete with Olympic aspirations needs to apply for a therapeutic use exemption from USADA (United States Anti-doping Agency). Treating depression or anxiety is probably the next most frequent referral. Athletes’ exercise and nutrition regimen tend to be healthier than non-athletes’. So when they meet DSM-IV criteria for a mood or anxiety disorder, medication management is often required. I usually recommend psychotherapy (sometimes in splittreatment) combined with medication. As a board-certified addiction psychiatrist, evaluating and treating substance use disorders in athletes is my specialized niche. I do the addiction counseling myself. I use motivational interviewing techniques more in athletes than 12-step facilitation or CBT [cognitive behaviour therapy], in comparison with my nonathletes. I find it challenging to convince most athletes to go to a 12-step meeting (AA or NA [narcotics anonymous]). I also work with athletes who suffer with eating disorders. Most fall into the category of ‘Anorexia Athletica’ or in DSM-IV terms, Eating Disorder NOS. We use treatment contracts that will involve input from coaches, parents, athletic trainers, sports medicine physicians, dieticians and other providers. We provide support in some cases to the team or affected teammates. On some teams, we need to do some educational programs and address the team culture. Athletes may have ‘The OverDoers Triad,’ meaning an eating disorder, obsessive-compulsive disorder and exercise dependence. SSRIs can be extremely helpful in ª 2009 Adis Data Information BV. All rights reserved.

611

reducing obsessional thinking and help the athlete do better in therapy. 2.5 Child and Adolescent Sport Psychiatry

Over the past two decades there has been a gradual increase in athletes aged from (mostly) 8–18 years consulting with sports psychiatrists. The reasons for consulting tend to fall in two areas: (i) DSM-IV disorders like anxiety disorder, phobic disorder and even obsessive-compulsive disorder, and (ii) performance-enhancing interventions. For example, an 11-year-old tennis player’s family called for help in ‘focusing’ during a match, while a 15-year-old swimmer asked for help for ‘negative thoughts before races’. Strategies and techniques are designed to (i) learn to concentrate, (ii) set goals, (iii) ‘relax’ and (iv) stay focused. Anecdotally, all have been found to improve performance to varying degrees. In addition, treatment has been found to uncover family conflicts about a child’s (or even spouse’s) athletic choices for performance or more commonly long-standing family conflicts. 2.6 Curriculum

There are now at least two curricula to support sport psychiatry training programmes.[11] They are targeted to psychiatry residents or fellows and are consistent with the requirements set forth by the Residency Review Committee for general psychiatric residency training programmes. The curriculum is used as an elective for psychiatrists, but it can be used as a foundation for developing a fellowship in sport psychiatry as well. The educational objectives include (i) knowledge of sport psychiatry, (ii) specific skills, and (iii) attitudes to be developed during the elective, plus learning experiences actually working with athletes. Learning experiences include both clinical and didactic experiences. 2.7 Organization and Turf Issues

In addition to the psychiatrist, there are many other disciplines involved in the field including trainers, psychologists, counsellors (broadly defined) and self-styled so-called gurus working Sports Med 2009; 39 (8)

Glick et al.

612

with a particular person or field. Unfortunately, in some cases, some of these people are the suppliers of performance-enhancing substances like anabolic steroids, amphetamines or other compounds. On the other hand, useful collaborative relationships are being forged with other professionals – sport psychologists being one example. There is now at least one practice that combines sport psychiatry and sport psychology, as mentioned above. It has an Association for Applied Sport Psychology (AASP)-certified sport psychologist, a masters-level sport psychology consultant, three ISSP sport psychiatrists, another licensed psychologist, and two licensed professional counsellors, all working together (E. Morse, personal communication, January 2008). Since the Begel article was written,[1] the ISSP has grown and now serves, in part, as a resource for the field. The core purpose of the ISSP is ‘‘to facilitate scientific communication about, and understanding of, disorders of the brain and behavior associated with sport, and to advance their prevention and treatment’’. The ISSP has an annual meeting in which the latest techniques and data are presented, has organized and sponsored a well attended symposium at the annual meeting of the American Psychiatric Association (APA) for the last 10 years, and has provided a curriculum for the field. The website is evolving into a forum for the exchange of ideas in the field, the dissemination of a quarterly newsletter, and a membership list to assist in referral to sport psychiatrists in different geographic locations. As mentioned, ISSP members have consulted with all the professional leagues around issues of psychiatric disorders, inappropriate behaviour (including aggression and substance abuse), medical-psychiatric issues like concussions, and other psychiatric issues. One benefit of the symposia has been the destigmatization of emotional problems in sports. At the APA, high-level athletes like Julie Krone (jockey), Derrick Adkins (track), Gerry Cooney (boxing), Pete Harnish (baseball), Wendy Williams (diving), Terry Bradshaw (football) and others have openly discussed their struggle with Axis I Disorders, and Krone and Bradshaw became the first high-profile athletes to acknowlª 2009 Adis Data Information BV. All rights reserved.

edge taking SSRIs while competing. The media has helped raise the public’s awareness of the fact that athletes (long thought to be the epitome of wellness) sometimes receive psychiatric medication too. Ms Krone’s revelation and other athlete’s disclosures have also made it far easier for general psychiatrists to convince their patients of the tolerability of psychiatric medications – after all, if a high-level athlete can take a medication and compete as well or better than before, what does the patient have to lose by trying to add such a medication to their treatment regimen? Members of the ISSP have also hosted media sessions at the annual convention of the APA discussing movies (When They Were Kings,[12] Remember the Titans[13]) with important sport psychiatry themes. Child psychiatrist members of the ISSP have also hosted media sessions at American Association of Child Adolescent Psychiatry annual meetings. Lastly, sport psychiatrists have consulted with media to help the public understand puzzling events in the athletic world as they unfold – a boxer biting another boxer’s ear, a football player making a ‘suicide attempt’, a wrestler killing his family and then himself while using anabolic steroids, and stimulant and anabolic steroid use by baseball players.[14] We envisage that this field will evolve over the next 20 years like other subspecialty areas of psychiatry such as legal psychiatry, child psychiatry, etc., which means implementation of ethical standards, specialized training programmes, certification examinations and boards, etc. Sports psychiatrists will become more involved in amateur as well as professional sports as the stigma of mental illness diminishes. 2.8 Summary and Conclusion

In this paper we have described the development and growth of sport psychiatry. We have described the evolution of what has happened to the field in the last two decades. Although we have described a variety of new roles for the sport psychiatrist, the rapid change in the field and the international growth of sports has led to new challenges and new treatments to improve the Sports Med 2009; 39 (8)

The Evolution of Sport Psychiatry, Circa 2009

health of the athletes (and significant others) with whom we work, as well as their teams. Acknowledgements We are indebted to Dan Begel, MD for helpful comments on an earlier draft. No funding was received for this article, and the authors have no conflicts of interest that are directly related to the content of this article.

References 1. Begel D. An overview of sport psychiatry. Am J Psychiatry 1992; 49: 606-14 2. Begel D, Burton R. editors. Sport psychiatry: theory and practice. New York City: WW Norton & Co., 2000 3. Glick ID, Horsfall JL. Psychiatric conditions in sports: diagnosis, treatment, and quality of life. Physician Sportsmed 2001; 29: 45-55 4. Kamm RL. The sport psychiatry examination. In: Begel D, Burton RW, editors. Sport psychiatry: theory and practice. New York: Norton, 2000: 159-90 5. Stryer HK, Tofler JR, Lapchick R. A developmental overview of child and youth sports in society. Tofler JR, editor. Child Adolesc Psychiatr Clin N Am 1998; 7: 6997-724, vii 6. Adkins D. Exercise, athletic participation, and mental health: when it works and when it doesn’t. Paper presented as part of symposium 26. American Psychiatric Association Annual Convention; 2001 May 7; New Orleans (LA)

ª 2009 Adis Data Information BV. All rights reserved.

613

7. Kamm RL. Group dynamics and athletic success. Presented at the American Group Psychotherapy Association’s Annual Meeting; 2000 Feb 18; Los Angeles (CA) 8. Diagnostic and statistical manual of mental disorders. 4th ed., text revision. Washington, DC: American Psychiatric Association, 2000 9. Tofler IR, Stryer BK, Micheli LJ, et al. Physical and emotional problems of elite female gymnasts. New Engl J Med 1996; 335: 281-3 10. Nicholi A. Psychiatric correlation in professional football. New Engl J Med 1996; 31: 1095-100 11. Bogacki DF, Newmark TS. A psychiatric curriculum directed toward sport psychiatry. Camden (NJ): Cooper University Hospital, 2007 12. Kamm RL, Calhoun J, Tofler I. When we were kings. Media session at the American Psychiatric Association Annual Convention; 1998 Jun 2; Toronto (ON) 13. Kamm RL, Tofler I. Remember the Titans. Media session at the American Psychiatric Association Annual Convention; 2002 May 22; Philadelphia (PA) 14. Mitchell GJ. Report to the Commissioner of baseball of an independent investigation into the illegal use of steroids and other performance enhancing substances by players in major league baseball. New York: Office of the Commissioner of Baseball; 2007 Dec 13

Correspondence: Prof. Ira D. Glick, 401 Quarry Road, Suite 2122, Stanford, CA 94305, USA. E-mail: [email protected]

Sports Med 2009; 39 (8)

Sports Med 2009; 39 (8): 615-642 0112-1642/09/0008-0615/$49.95/0

REVIEW ARTICLE

ª 2009 Adis Data Information BV. All rights reserved.

Aerobic Conditioning for Team Sport Athletes Nicholas M. Stone and Andrew E. Kilding School of Sport and Recreation, AUT University, Auckland, New Zealand

Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Physical Demands of Team Sport Competition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Physiological Adaptations to Aerobic Conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Potential Benefits of Aerobic Fitness for Team Sport Performance . . . . . . . . . . . . . . . . . . . . . . . . . 2. Traditional Aerobic Conditioning for Team Sports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Effectiveness of Traditional Aerobic Conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Traditional Aerobic Conditioning and Soccer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Traditional Aerobic Conditioning and Basketball. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Classic Team Sport Conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Effectiveness of Classic Team Sport Conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Limitations of a Classic Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Sport-Specific Aerobic Conditioning for Team Sports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Examples of Sport-Specific Aerobic Conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Small-Sided Games for Soccer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Small-Sided Games for Other Sports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Traditional versus Sport-Specific Aerobic Conditioning for Team Sports . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Traditional and Sport-Specific Aerobic Conditioning Approaches in Soccer . . . . . . . . . . . . . . . . 5.2 Traditional and Sport-Specific Aerobic Conditioning and Other Sports . . . . . . . . . . . . . . . . . . . . . 5.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Abstract

615 616 617 617 619 619 619 619 621 622 622 622 623 626 627 627 627 627 630 632 632 632 634 636 636 636

Team sport athletes require a high level of aerobic fitness in order to generate and maintain power output during repeated high-intensity efforts and to recover. Research to date suggests that these components can be increased by regularly performing aerobic conditioning. Traditional aerobic conditioning, with minimal changes of direction and no skill component, has been demonstrated to effectively increase aerobic function within a 4- to 10-week period in team sport players. More importantly, traditional aerobic conditioning methods have been shown to increase team sport performance substantially. Many team sports require the upkeep of both aerobic fitness

Stone & Kilding

616

and sport-specific skills during a lengthy competitive season. Classic team sport trainings have been shown to evoke marginal increases/decreases in aerobic fitness. In recent years, aerobic conditioning methods have been designed to allow adequate intensities to be achieved to induce improvements in aerobic fitness whilst incorporating movement-specific and skill-specific tasks, e.g. small-sided games and dribbling circuits. Such ‘sport-specific’ conditioning methods have been demonstrated to promote increases in aerobic fitness, though careful consideration of player skill levels, current fitness, player numbers, field dimensions, game rules and availability of player encouragement is required. Whilst different conditioning methods appear equivalent in their ability to improve fitness, whether sport-specific conditioning is superior to other methods at improving actual game performance statistics requires further research.

Team sport athletes require a wide range of physical and technical abilities. Specifically, a well developed level of aerobic fitness is desirable in order to recover quickly between repeated high-intensity efforts[1-3] that are typically associated with game-defining moments such as scoring or preventing other teams from scoring. Improving an individual player’s and team’s performance using aerobic conditioning practices has become a priority, and methods to do so have been extensively investigated.[4-18] In order to obtain the necessary articles for this review, several health databases were searched, including EBSCO, PubMed and SPORTDiscus. Key search terms used included ‘aerobic’, ‘basketball’, ‘conditioning’, ‘endurance’, ‘high-intensity’, ‘rugby’, ‘soccer’, ‘sport-specific’, ‘team sports’ and ‘training’. This review specifically focuses on the literature relating to aerobic conditioning interventions and their subsequent influence on aerobic fitness, and, when possible, individual/ team sport performance. Based on the articles retrieved, this review differentiates between ‘traditional’, ‘classic’ and ‘modern’ or ‘sportspecific’ aerobic conditioning approaches. Articles were excluded if no measures of aerobic fitness were presented, as this proved difficult to make comparisons with other research. Because of the small number of articles relating to aerobic conditioning interventions and team sport athletes, there was no limit to the search period. ª 2009 Adis Data Information BV. All rights reserved.

1. Physical Demands of Team Sport Competition The acyclic, intermittent nature of team sport competition is made evident through the movement frequencies observed during match play being in excess of 1000, regardless of sport.[19-26] The total distance travelled in a match suggests that a well developed level of aerobic fitness is required, especially with respect to outdoor, field-based team sports such as rugby union[21,27] and soccer.[5,23,25,26,28-31] Average game intensity appears to be relatively constant across a range of team sports, equating to just below lactate threshold (LT), i.e. 80–90% peak heart rate . (HRpeak) or 70–80% peak oxygen uptake (VO2peak).[5,25,30,32-37] However, when defining characteristics of team sport performances are broken down into low-, medium- and highintensity activities, clear differences between sports are seen. Players in rugby union[21] and soccer[26,31] appear to spend the greatest amount of time performing low-intensity movements (80–85% of game time), whereas field hockey[38] and basketball[19,20,39] players spend a considerably lower amount of time involved in lowintensity activity (~50% of game time) and more at medium-intensity (~40%). It should be acknowledged that the notable differences both within and between team sports time-motion analysis studies could be due to several reasons: (i) the method used to classify different exercise intensities Sports Med 2009; 39 (8)

Aerobic Conditioning for Team Sport Athletes

(i.e. classification of ‘zones’) is inconsistent between studies; (ii) the number of games analysed could lead to an over- or underestimation of the durations of working at different intensities; (iii) the fitness level of the players analysed could lead to either more or less time engaged in highintensity activity; and (iv) the importance of the game and the level of the opposition could both increase or decrease the percentage of highintensity activities performed. Nevertheless, the amount of high-intensity exercise carried out appears consistent across most team sports, accounting for about 15–19% of the total distance covered[26,31,40] or 10–15% of game time.[19,20,41] However, depending on player fatigue levels,[25,32,42] the skill level of the opposition, the importance of the match, or whether the team is winning or losing, the amount of high-intensity exercise performed varies from match to match.[43,44] Heart rate[19,20,23,28,33,35,45-47] and blood lactate concentration[5,8,19-21,23,35,48-51] data reinforce these observations, with moderately high heart rates and lactate concentrations measured during team sport competition. The latter has been significantly correlated to the amount of high-intensity activity performed during the 5 minutes before sampling.[19,52] The energy system contribution to team sport performance tends to reflect game intensity. The majority of the adenosine triphosphate (ATP) required to perform is supplied via aerobic pathways,[19,20,22,28,33,39,41,45,52-55] based on the large amounts of game time spent engaged in low- to moderate-intensity activity. Based on the physical demands and characteristics of team sport competition, and the potential importance of aerobic fitness, it is clear that a significant portion of the conditioning programmes of team sport players should focus on improving their ability to repeatedly perform high-intensity exercise and to recover. To a large extent both these abilities can be improved by performing aerobic conditioning. 1.1 Physiological Adaptations to Aerobic Conditioning

Several studies have shown that aerobic conditioning is associated with adaptations in the ª 2009 Adis Data Information BV. All rights reserved.

617

pulmonary,[56-59] cardiovascular,[60-64] neuromuscular[65,66] and metabolic[66-71] systems. However, it is clearly apparent from the literature that the physiological adaptations to training depend upon several factors, including the exercise intensity[72-75] and frequency,[76-79] the duration of training,[69,80] the total length of time of the training programme[76,81,82] and the initial fitness level of the individual.[75,83] These factors interact to determine the overall magnitude of the training response. Depending on the intensity of aerobic conditioning, physiological adaptation may primarily [84] At intensities occur centrally or peripherally. . slightly below LT (~70–80% VO2peak), physiological adaptations occur primarily in the central component.[81,85] Central adaptations include an improvement in the heart’s capacity to pump blood, primarily through increased stroke volume, which occurs because of an increase in end-diastolic volume and an increase in left ventricular mass.[86] Subsequently, these adaptations result in an increased cardiac output, which, according to . the Fick equation, will increase VO2peak.[87] As the training intensity increases above the LT (>80% . VO2peak), significant peripheral adaptation occurs, with substantial changes in muscle capillarization,[88] oxidative enzyme activity,[89] mitochondrial volume and density,[90] and myoglobin,[82] and the preferential use of free fatty acids as an energy substrate.[69] As a consequence of the above central and peripheral adaptations, performancerelated aerobic measures such as whole-body . VO2peak,[91] exercise/work economy,[92] LT[93] and oxygen uptake kinetics[94] are all improved.[95] 1.2 Potential Benefits of Aerobic Fitness for Team Sport Performance

The potential benefits of enhanced aerobic fitness for team sport performance are numerous. Although team sport players appear to spend the majority of their time involved in low-to-moderate intensity activities, defensive/offensive success often depends on the less frequent but higher intensity activities that involve combinations of sprinting, jumping and tackling. These highintensity activities place extreme demands on Sports Med 2009; 39 (8)

618

the anaerobic energy system intermittently throughout the duration of a game. However, overall there is a heavier reliance on the aerobic system, which serves to promote recovery and is engaged during low- to heavy-intensity activities that predominate over 70–90 minutes of wholebody physical activity. Appropriate aerobic conditioning plays a significant part in allowing players to repeatedly perform high-intensity activity. It has been shown . that a high VO2peak is moderately related (r = 0.62 to r = 0.68; p < 0.05) to repeat sprint ability (RSA) in field hockey, rugby union and soccer players, as well as endurance-trained and -untrained populations.[1-3] This suggests that the body’s ability to deliver and use oxygen, both during and between high-intensity sprints, is important.[96] Furthermore, also in soccer, previous studies have demonstrated that players with a higher aerobic power cover greater distance during a . soccer game.[97] Overall, a high VO2peak will likely serve to reduce the metabolic disturbances resulting from anaerobic metabolism. Ultimately, players who are aerobically well trained are likely able to maintain their work rates/power output towards the end of a game compared with those with poorer aerobic fitness. It is possible that short and long RSA could also be influenced by the oxidative potential of the muscle in elite team sport athletes, which may be best reflected by measures of the LT. Indeed, Edwards et al.[98] showed that the LT was a more sensitive indicator to changes in training status in professional soccer players than . VO2peak. Further support for this measure is provided by other studies utilizing fixed blood lactate values. For example, Krustrup et al.[28] demonstrated that the amount of high-intensity running during soccer match play was significantly related to the running speed at 2 mmol/L blood lactate concentration in female players. More recently, Sirotic and Coutts[99] also showed that a significant relationship existed between the running velocity at a blood lactate concentration of 4 mmol/L and prolonged highintensity intermittent running distance (r = 0.77; p < 0.05) in 16 moderately trained women team sport athletes. These studies both demonstrate a ª 2009 Adis Data Information BV. All rights reserved.

Stone & Kilding

clear benefit of having well developed aerobic function. In contrast, however, it should be acknowledged that there are other studies showing that . VO2peak is a poor indicator of the fitness status of team sport athletes,[98] and it does not relate to performance in either short-term[100] or prolonged intermittent exercise tests among professional players (r = 0.18; p > 0.05; n = 8),[40] nor does it determine the total amount of running distance covered during a game.[28] Also, in well trained subjects, Bishop et al.[100] demonstrated that short-term . RSA was not significantly correlated with VO2peak in elite team sport athletes (r = 0.30; p > 0.05). Methodological differences between these conflicting studies such as the standard of player, nature of the games played, different tests used to determine RSA (short and prolonged) and reliability of output measures from time-motion analysis could all account for the inconsistent findings. . In addition to. VO2peak and LT measures, improvements in VO2 kinetics as a result of endurance training have been suggested to increase metabolic efficiency during recovery, which assists in delaying the onset of fatigue.[101] It should be noted that due to recovery time being typically during team of short duration (<30 sec)[20,22,24,102] . sport competition, a faster VO2 response[103,104] would serve to assist in the replenishment of phosphocreatine stores, which would be re-used across multiple high-intensity efforts. Indeed, athletes with a greater muscle oxidative capacity have been reported to boast greater phosphocreatine resynthesis and an increased ability to remove lactate and hydrogen ions (H+) from skeletal muscle,[105-108] which will probably be beneficial for team sport athletes. Given the apparent importance of developing various aerobic fitness measures to enhance various physical output aspects during games in team sports athletes, studies have considered the most effective ways of improving aerobic fitness. Several different approaches can be used to develop the aerobic condition of team sport players. These include a range of traditional, classic and sport-specific (movement-specific) conditioning approaches. Sports Med 2009; 39 (8)

Aerobic Conditioning for Team Sport Athletes

2. Traditional Aerobic Conditioning for Team Sports 2.1 Definition

Traditional aerobic conditioning, defined here as continuous or interval-based straight line running with minimal changes of direction, is used by many athletes and fitness enthusiasts to increase aerobic fitness. Since it has been sug. generally ocgested that VO2peak improvements . cur when a high percentage of VO2peak is elicited during exercise,[109] the general goal of interval conditioning is to accumulate a greater training stimulus at high intensities compared with what can be tolerated in a single bout of continuous exercise. This approach is especially important for trained team sport athletes whose increases in aerobic fitness are limited by cardiac output,[110] specifically stroke volume. A recent study by Zhou et al.[111] found that stroke volume increased continuously with increased workload up . to VO2peak in well trained participants. Consequently, .the increased stroke volume up to the level of VO2peak in trained athletes has been the rationale behind using high-intensity aerobic conditioning interventions.[5,6,9,112] The prescription of interval training is based on three key variables: work interval intensity and duration, recovery interval intensity and duration, and total work duration (work interval number · work duration). These variables can be manipulated to generate a large range of interval training prescriptions designed primarily to stress aerobic and/or anaerobic energy metabolism. Sufficient physiological data are now available to classify different types of aerobic interval training, ranging in intensity from 85% .to 130% of the power or velocity associated with VO2peak.[113,114] 2.2 Effectiveness of Traditional Aerobic Conditioning

Despite the inclusion of traditional aerobic conditioning in the training programmes of many team sports, surprisingly few studies have documented the true effectiveness of traditional aerobic conditioning approaches in relation to improved physiological measures and their ª 2009 Adis Data Information BV. All rights reserved.

619

subsequent influence on team sport performance (table I). To date, there is no evidence suggesting that aerobic conditioning does not result in improved team sport performance. While in the few studies that have reported the effects on individual fitness, individual performance and team sport performance, traditional interval conditioning has elicited favourable changes in aerobic fitness and performance, suggesting that traditional aerobic conditioning is an effective approach.[4,5,7] 2.3 Traditional Aerobic Conditioning and Soccer

In a unique training study, Helgerud et al.[5] were among the first researchers to clearly show that high-intensity ‘traditional’ aerobic interval training significantly influenced a soccer player’s aerobic fitness and, perhaps more importantly, their performance during a soccer match. In this study, traditional interval endurance training was performed twice a week, over an 8-week period, at the beginning of the competitive soccer season. Players performed four 4-minute running intervals at 90–95% HRpeak, interspersed with 3 minutes of active recovery, . jogging at 50–60% HRpeak. Consequently, VO2peak increased by 10.8% over the duration of the study. LT and running economy improved by 16.0% and 6.7%, respectively (p < 0.05), suggesting a marked impact on measures of aerobic function. However, two regular training sessions, involving game play, tactical, technical, strength and sprint training activities, along with one competitive game per week, were performed concurrently with the endurance training. The intensity of these additional training sessions was not reported and, therefore, based on . the control groups’ post-training increases in VO2peak (2%) and number of sprints performed during a match (8%), it is possible that the supplementary training may have contributed towards the improvements in aerobic fitness in the experimental group. Nevertheless, the influence of this improvement on subsequent soccer performance cannot be ignored. The traditional aerobic conditioning resulted in significantly higher exercise Sports Med 2009; 39 (8)

Balabinis et al.[4]

Sport

No. of subjects

Mean (– SD) age (y)

Season

Training intervention duration sessions (wk) per week

620

ª 2009 Adis Data Information BV. All rights reserved.

Table I. Traditional aerobic endurance conditioning used in team sports, and the subsequent influence on aerobic fitness and individual/team performance Study

Findings mode

intensity

Basketball 7 males

22.6 – 0.8

7 SE training

4

100–500 m running intervals every 30–60 sec

85–90% HRpeak

. Pred VO2peak › 13%*

7 males

22.4 – 0.5

7E training

4

100–500 m running intervals every 30–60 sec

85–90% HRpeak

. Pred VO2peak › 7%*

Soccer

22 males

20.2 – 0.7 In

10

1 1

12–15 · 40 m sprints every 30 sec 15 sec work : 15 sec recovery rest · 12–15 sprints

120% MAS

Helgerud et al.[5]

Soccer

19 males

18.1 – 0.8 In

8

2

(4 min work : 3 min recovery) · 4, running intervals

90–95% HRpeak : 50–60% HRpeak

Helgerud et al.[6]

Soccer

21 males

25.0 – 2.9 Pre

8

2

(4 min work : 3 min 90–95% HRpeak : recovery) · 4, running intervals 50–60% HRpeak

MAS › 8%*** Win percentage › 136%

. VO2peak › 11%* LT › 16%* RE fl 7%* No. of sprints › 100%** No. of involvements with ball › 24%* Distance covered › 20%**

. VO2peak › 8%*** RE fl 4%*

E. = endurance; HRpeak = peak heart rate; LT = lactate threshold; MAS = maximal aerobic speed; Pred = predicted; RE = running economy; SE = strength and endurance; VO2peak = peak oxygen uptake; fl indicates decrease; › indicates increase; * p < 0.05, ** p < 0.01, *** p < 0.001.

Stone & Kilding

Sports Med 2009; 39 (8)

Dupont et al.[7]

Aerobic Conditioning for Team Sport Athletes

intensities (3.5%; p < 0.05) when expressed in relation to HRpeak during a soccer match following the training period. This is presumably related to the improved exercise economy and higher LT as a result of the endurance training. It was also clearly shown that the number of sprints and involvements with the ball increased significantly (p < 0.05) from 6.2 to 12.4 (100%) and 47.4 to 58.8 (24%), respectively. Furthermore, a significant increase in the total distance covered (8619 m to 10 335 m, 20%; p < 0.01) during a match was observed. Collectively, these results are very encouraging despite the small sample size (n = 19). Furthermore, it must also be acknowledged that quantifying soccer performance based on a single soccer match has its limitations, due to the tactical/ technical nature of the game. Playing conditions, skill level of the opposition and importance of the game, amongst other factors, may all have contributed to player performance.[43,44] Analysis of several soccer matches is required to confirm these observations. More recently, other studies have reported positive physiological adaptation after traditional training approaches. Helgerud et al.[6] performed another study utilizing the same aerobic interval training approach as previously used[5] with a European Champions League team, and found a similar improvement in . VO2peak (8.1%; p < 0.001). This pre-season improvement was observed despite concurrently training for both maximal strength and aerobic endurance. Furthermore, this increase was exhibited in soccer players of a higher standard compared with those used by Helgerud et al.[5] Running economy improved by 3.7% (p < 0.05) post-training, perhaps augmented by the maximal strength training that was performed, as resistance training has been demonstrated to benefit exercise economy.[115,116] However, it appears that the concurrent training approach may have slightly hindered the improvement in running economy when compared with a previous aerobic training study involving younger soccer players.[5] Other. authors also report significant increases in VO2peak, LT and running economy, in addition to improvements in individual soccer performance, following ª 2009 Adis Data Information BV. All rights reserved.

621

traditional aerobic conditioning programmes[5,6] (see also section 5.1). Using a different approach to the above, Dupont et al.[7] investigated the effect of traditional aerobic conditioning on aerobic fitness in soccer players during the season. A goal of this study was to observe a change in aerobic fitness without any subsequent decrease in match performance. This study demonstrated that one weekly session of short intermittent exercises (12–15 · 15 sec at 120% of maximal aerobic speed) and one weekly session of repeated sprinting exercise (40 m sprints repeated every 30 sec) over 10 weeks induced substantial improvements in aerobic fitness.[7] Maximal aerobic speed improved by 8.1 – 3.1% (p < 0.001). Similar to other studies,[5] the athletes involved in this research also performed two additional team training sessions per week, reportedly involving. ‘light’ exercises. Unfortunately, post-training VO2peak was not reported for this study, making comparisons with other research difficult.[5,6] However, given the type of training performed, the improvements in maximal aerobic speed could be a consequence . of a greater amount of time spent at VO2peak.[117] Greater time spent at a high percentage of . VO2peak during training necessarily induces a positive change in measures of aerobic fitness.[109] With respect to the effects of the conditioning regimen on match performance, the team won 33% of its matches during the 10-week control period (no high-intensity training) and 78% of its matches during the 10-week high-intensity training period. However, this 136% increase must be interpreted with caution, as many factors could have influenced the outcome of games during both the control and the training periods, as described earlier. Overall, the findings of Dupont et al.[5,6] suggest that imet al.[7] and Helgerud . provements in VO2peak can be achieved during the pre-season and/or early in the season without any decrease in match performance. 2.4 Traditional Aerobic Conditioning and Basketball

Traditional aerobic interval . conditioning has also been shown to increase VO2peak in collegiate Sports Med 2009; 39 (8)

Stone & Kilding

622

basketball players who simultaneously trained for both muscular strength and aerobic adaptations.[4] Balabinis et al.[4] examined the effects of endurance, strength and concurrent strength and endurance conditioning on several physiological parameters over a 7-week duration (training phase of the season was not stated). Endurance conditioning involved performing between two and ten repeats of 30–1000 m running intervals every 30–60 sec. Endurance conditioning alone . improved predicted VO2peak by approximately 7%, which was slightly less than that reported by however, greater Helgerud et al.[5] Interestingly, . changes in predicted VO2peak (13%) were observed when concurrent strength and endurance conditioning was performed (table I), despite all experimental groups being equally matched for aerobic fitness (figure 1) at the start of the study. Given that participants in the studies of both et al.[4] displayed Helgerud et al.[5] and Balabinis . similar pre-training VO2peak values (58 and 54–55 mL/kg/min, respectively), the greater increase in aerobic function for the strength and endurance-trained group reported by Balabinis et al.[4] may have been augmented by the periodized strength training programme that the group performed. The strength training undertaken was divided into three phases: (i) maximum strength, 1–4 sets · 3–6 repetitions at 75–95% of one repetition maximum (1RM); (ii) explosive power, 4–5 sets · 5–6 repetitions at 70% 1RM; and (iii) muscular endurance, 3 sets · 30–40 repetitions at 40% 1RM. It is very possible that the latter muscular endurance phase promoted

· VO2peak (mL/kg/min)

65

Pre-test Post-test

additional peripheral oxidative adaptations, which has been previously observed in other studies.[118-122] Although Balabinis et al.[4] did not measure aerobic fitness directly, nor any subsequent influence on game performance following the training period, these results suggest that combining strength and endurance training may be a worthwhile approach in basketball for improving aerobic fitness. 2.5 Summary

In summary, it is clear that traditional aerobic conditioning involving repeated running intervals at intensities ranging between 85% and 95% HRpeak and lasting up to 4 minutes, separated by a maximum of 3 minutes’ active recovery at about 60% HRpeak, appears to promote beneficial changes in aerobic function. Such beneficial changes are noticeable when traditional aerobic conditioning is performed twice per week with 24–48 hours of recovery between sessions over periods ranging from 4 to 10 weeks. Some limited evidence exists to suggest that this improvement in fitness is transferable to the actual game situation and subsequently enhances team sport performance. Furthermore, periodized, concurrent strength and endurance training in team . sport athletes may elevate VO2peak to a greater extent than endurance training alone. The findings of both Dupont et al.[7] and Helgerud et al.[5,6] demonstrate that the aerobic fitness of soccer players can be improved during the competitive season without a decrease in match performance. However, traditional aerobic conditioning methods and their subsequent influence on sport performance is an area requiring further research.

60

3. Classic Team Sport Conditioning 55

3.1 Definition

50 Strength and endurance

Control

Endurance

Strength

. Fig. 1. Mean (–SD) peak oxygen uptake (VO2peak) for concurrent strength and endurance, control, endurance only and strength only training groups’ pre- and post-7-week training intervention.[4]

ª 2009 Adis Data Information BV. All rights reserved.

‘Classic’ team sport conditioning typically integrates strength, power, speed and aerobic conditioning components, directed towards improving the athlete’s overall functional and physical capacities specific to their sport, within a coaching framework.[123] Under such conditions, Sports Med 2009; 39 (8)

Aerobic Conditioning for Team Sport Athletes

some research, but not all,[124-126] suggests that aerobic power can be maintained and even increased, simply from participation in classic team sport training and competition (table II).[127-131] For example, in basketball, it has been demonstrated that the intensity and duration of running during team practice and competition created sufficient stimulus for the maintenance of aerobic power over single[128,132] and consecutive (four) seasons.[129] However, Hunter et al.[129] reported the training objective for most .players was to maintain, not increase, their VO2peak (50 mL/kg/min) over the four seasons, based on . similar VO2peak values reported for other college [133] Subsequently, no improvements in players. . VO2peak were observed. Conversely, a more recent study involving professional basketball players[134] has shown that a conditioning programme involving speed exercises, technical drills, match situations and endurance training resulted . in a marginal increase in VO2peak (6%) from preseason through to the competitive season. 3.2 Effectiveness of Classic Team Sport Conditioning

. Observing an increase in VO2peak during the competitive season is not unique to basketball. Gabbett[131,135] found that amateur rugby league players undertaking a progressively overloaded training programme involving specific skill, speed, muscular power, agility and endurance training exercises, twice per .week, showed an 18–19% increase in predicted VO2peak during the course of a rugby league season. The same author in a later study[130] also demonstrated that junior and senior rugby league players increased aerobic fitness by between 5.1% and 8.6% over a 14-week pre-season training period. Despite having lower training loads, junior rugby league players exhibited greater . adaptations in predicted relative VO2peak than senior rugby league players (8.6% vs 5.1%, respectively). This suggests that junior (~17 years old) and senior (~26 years old) rugby league players adapt differently to an absolute training stimulus and that training programmes should be modified to accommodate differences in training age.[130] Furthermore, across three consecutive pre-season ª 2009 Adis Data Information BV. All rights reserved.

623

. periods, predicted VO2peak values in rugby league players have been shown to improve progressively (2001, 7.7%; 2002, 11.8%; 2003, 15.6%; figure 2), with each pre-season period inducing a significant . increase (p > 0.05) in VO2peak.[127] This improvement was attributed to a periodized conditioning programme consisting of two game-specific training sessions per week of approximately 60–100 minutes’ duration. Interestingly, following the initial season (2001), training loads were decreased through reductions in training duration (2002) and. training intensity (2003), but improvements in VO2peak were still observed. However, it should be noted . that the greater changes in predicted relative VO2peak over the three pre-season periods could possibly be a result of the slightly . lower pre-training VO2peak of players in the 2002 and 2003 pre-season periods.[127] Also, current research has questioned the reliability of the multistage shuttle run test for monitoring changes in . VO2peak.[136] Lamb and Rogers[136] concluded this was due to the large amount of random error associated with various types of predictive equations. In addition, recent research has also failed to show a. strong relationship between laboratory-derived VO2peak and multistage shuttle run test score,[137] particularly at the elite level.[138] For example, Aziz et al.[137] . found only a moderate relationship between VO2peak (absolute and relative) and aerobic endurance performance (multistage shuttle run test score) [r = 0.43 and r = 0.54, respectively] in 37 male soccer players, suggesting that these two tests were measuring different aspects of aerobic fitness. As a provided modified result, Kilding et al.[139] have . equations for predicting VO2peak from the multistage shuttle run test in team sport athletes. Future research should either use more direct methods for detecting changes in aerobic fitness, or utilize more appropriate predictive methods. Recently, an off-season classic conditioning approach has been reported to promote increases in . VO2peak and individual technical skill performance in adolescent basketball players.[140] In this study, players were divided into two training groups: (i) specialized (SP) training, carried out exclusively on the basketball court; and (ii) mixed (MX) training, which was similar to SP training but also included resistance training for muscular strength Sports Med 2009; 39 (8)

624

ª 2009 Adis Data Information BV. All rights reserved.

Table II. Influence of a ‘classic’ approach to team sport conditioning on measures of aerobic fitness Study

Sport

No. of Mean (– SD) males age (y)

Season

Training

Findings

duration

sessions session per week duration (min)

mode

5

Technical, match situations, half and full court 5v5 SSG, and muscular strength and power circuit training

Bogdanis et al.[140]

Basketball

27a

14.7 – 0.5

Off

4 weeks

Hoffman et al.[128]

Basketball

9

18.8 – 0.7

Pre, in

25 weeks

Hunter et al.[129]

Basketball

24

4 competitive seasons

100–120

. VO2peak › 5%* BTS › 17–27%** 1.5 mile run time › 5%

2

1.5–3 mile run

3

6–20 · 100–440 yard runs

. VO2peak › fl

. VO2peak › 6% HRrest fl 18%**

Laplaud et al.[134]

Basketball

8

24.0 – 4.0

Pre, in

4.7 – 0.7 months

19 – 2 hours

Speed, technical, match situations and endurance training

Tavino et al.[126]

Basketball

9

18–22b

Pre, in

6 months

0–5

Weights, aerobic and anaerobic training

36

17.9

Pre

14 weeks

2

60–100

Skill, speed, muscular power, agility and endurance training

41

25.5

Pre

14 weeks

2

60–100

Skill, speed, muscular power, agility and endurance training

. VO2peak fl 5% . Pred VO2peak * › 9% . Pred VO2peak › 5%*

Gabbett

[130]

Rugby League

Rugby League

36

17.9

Off, pre, in

9 months

2

60–100

Skill, speed, muscular power, agility and endurance training

. Pred VO2peak › 19%

Gabbett[131]

Rugby League

52

>18

Off, pre, in

9 months

2

60–100

Skill, speed, muscular power, agility and endurance training

. Pred VO2peak › 18%

Gabbett[127]

Rugby League

79

22.9

2001

10 months

2

60–100

Periodized game-specific training programme

65

19.6

2002

10 months

2

60–100

Periodized game-specific training programme

76

21.5

2003

10 months

2

60–100

Periodized game-specific training programme

a b

. Pred VO2peak * › 8% . Pred VO2peak › 12%* . Pred VO2peak › 16%*

Adolescents.

Range. . BTS = basketball technical skills; HRrest = resting heart rate; Pred = predicted; SSG = small-sided games; VO2peak = peak oxygen uptake; fl indicates decrease; › indicates increase; › fl indicates no change; * p < 0.05, ** p < 0.01.

Stone & Kilding

Sports Med 2009; 39 (8)

Gabbett[135]

Aerobic Conditioning for Team Sport Athletes

625

Pre-season Post-season b 80

4.0

70

Training duration (min)

Training intensity (units)

a 4.5

3.5

3.0

60

50

40

2.5 2001

2003

2002 Year

2001

2002 Year

2003

c 55

· VO2peak (mL/kg/min)

50

45

40

35 2001

2002 Year

2003

. Fig. 2. (a) Overall training intensity, (b) duration, and (c) peak oxygen uptake (VO2peak) of sub-elite rugby league players over three consecutive pre-season preparation periods. Values are mean – 95% CI (reproduced from Gabbett,[127] with permission from BMJ Publishing Group Ltd).

and power (see Bogdanis et al.[140] for more details on training design). Both training programmes included five training sessions per week, each lasting 100–120 minutes, for 4 weeks. No training intensity was prescribed; however, intensity was evaluated by continually monitoring heart rate during the training sessions. Training load in both the SP and MX training groups was similar (low ª 2009 Adis Data Information BV. All rights reserved.

intensity: 58.4 – 3.0 vs 64.2 – 2.7%; moderate intensity: 37.0 – 2.4 vs 32.6 – 1.0%; high intensity: 4.6 – 1.2 vs 3.2 – 1.0% for the SP and MX groups, respectively). A relatively small but substantial improvement in aerobic fitness was observed following both SP and MX training programmes (SP: 4.9 – 1.8%; MX: 4.9 – 1.4%; p < 0.05), despite the training not being specifically designed to Sports Med 2009; 39 (8)

Stone & Kilding

626

[126] resulted in a 4.7% increase in (5 . · per week) VO2peak. However, during the competitive season, only scrimmages (intra-team games) and anaerobic conditioning were continued, in conjunction with approximately two basketball games per week. Consequently, a 5.3% decrease in aerobic fitness over the course of the competitive season was observed. When the conditioning focus of the training programme does not include aerobic conditioning, minimal changes in . VO2peak would be expected, which could be detrimental to individual and team performance. However, it could be argued that the exclusion of aerobic conditioning from in-season professional basketball practice is warranted, given that Gillam[124] observed a negative relationship between points scored per minute of play and the cardiovascular endurance of players (r = -0.66). This is supported indirectly by the more recent findings of Hoffman et al.,[123] who reported a low correlation (r = 0.10) between the total amount of playing time per player and aerobic endurance in basketball athletes, suggesting that those with the highest levels of aerobic fitness had the least amount of playing time. The major determinant of playing time in this study was the coach’s evaluation of the player’s ability.

3.3 Limitations of a Classic Approach

ª 2009 Adis Data Information BV. All rights reserved.

3000 Yo-yo intermittent recovery test performance (m)

* *

*

2500

2000

1500

on as

on En

d

se

ea s ar ts St

ar re p id -p

ep

re p

ar a

tio

n

at io n

1000

Pr

Despite these positive changes in fitness, espe. cially VO2peak, findings from a number of studies refute the notion that classic team sport training can result in substantial changes in aerobic fitness. For example, seasonal changes in aerobic fitness of elite soccer players appear to increase through to the start of the competitive season (25%) and decline, by 5%, when measured at the end of the season, as illustrated in figure 3.[125] A similar trend has been observed in a recent analysis of the aerobic fitness of soccer players over the course of a soccer season.[141] Similarly, college basketball training and competition has been reported to have little effect on aerobic capacity during a season.[124,126] A pre-season training programme consisting of anaerobic conditioning (5 · per week), weight training (3 · per week), scrimmages (2–4 · per week) and aerobic conditioning

M

. improve VO2peak.[140] The small increase could be due to the aerobic stimuli of classic basketball training typically not being adequate to induce more substantial improvements in aerobic fitness,[140] and that insufficient time (5–7%) was spent at training . intensities associated with greater increases in VO2peak.[109] Additionally, aerobic fitness typically declines in team sport athletes during the off-season, which is evident from lower preseason yo-yo intermittent recovery test (IRT) [125] scores than end-season . scores in soccer players. Therefore, a lower VO2peak could have potentially contributed to the increase in aerobic fitness. These results help reinforce the findings of Balabinis et al.[4] in that additional traditional aerobic training in conjunction with resistance training may lead to greater gains in aerobic fitness. Perhaps more relevant to team sport performance, Bogdanis et al.[140] showed an improvement (15–25%; p < 0.01) in four basketball-specific technical skills (speed shot shooting, passing, dribbling and defensive sliding) following the training period for both groups. An improvement in individual technical skill would logically transfer into better individual/team performance during a basketball game, which warrants further research during the competitive season.

Fig. 3. Seasonal changes in yo-yo intermittent recovery test performance for elite soccer players. Values are mean – SEM as well as individual values. * Denotes significant difference (p < 0.05) from pre-preparation period (reproduced from Krustrup et al.,[125] with permission from Lippincott Williams and Wilkins).

Sports Med 2009; 39 (8)

Aerobic Conditioning for Team Sport Athletes

3.4 Summary

Overall, the aerobic capacity of team sport players (basketball, rugby league and soccer) has been shown to increase throughout the preseason and decrease during the competitive season, when using a classic team sport conditioning approach.[125,126] The reduced attention to aerobic conditioning during the competitive season, in some sports, suggests that the importance of aerobic endurance may be underrated. In some instances this may be warranted, if other aspects (technical or physical) are shown to be more important. Accordingly, it appears that coaches, along with strength and conditioning professionals, prioritize training regimens focused on improving anaerobic fitness during the competitive season, most probably because high-intensity activities are associated with important game winning situations, such as scoring points in basketball or a try in rugby union. However, it should be emphasized that a lack of focus on aerobic conditioning is also very likely to influence the ability to repeatedly perform, and recover from, high-intensity activity (sprints), so the absence of aerobic conditioning during the competitive season, regardless of sport, may not represent best practice in terms of optimizing the condition of athletes. Further research is needed to determine the impact of classic team sport conditioning regimens on aerobic fitness and possibly game performance so that the strengths and weaknesses of such approaches can be identified for different team sports. Future training studies can then be developed to further develop strengths and improve on weaknesses. Furthermore, there is a need for strategies to be developed that show coaches of team sports how two components of the game can be worked on simultaneously, such as aerobic endurance and technical skill. 4. Sport-Specific Aerobic Conditioning for Team Sports 4.1 Definition

The departure from traditional aerobic conditioning methods appears to be the result of the ª 2009 Adis Data Information BV. All rights reserved.

627

design and greater use of sport-specific aerobic conditioning sessions. Sport-specific aerobic conditioning generally involves small-sided conditioning games or dribbling tracks/circuits, which incorporate skills and movements specific to the sport into a physical framework. Indeed, such aerobic conditioning methods are being increasingly implemented in professional team sport environments (table III)[142,143] with an increased emphasis on training ‘with the ball’ where possible.[8-14,130] The perceived benefits of performing sport-specific exercise, rather than traditional aerobic conditioning, are that: (i) the training will transfer better into the athletes’ competitive environment; (ii) the greatest training adaptations will occur when the training stimulus simulates the specific movement patterns and physiological demands of the sport;[83] (iii) skill-based conditioning games provide an opportunity to develop decision-making and problem-solving skills, under stressful physical loads;[144] and (iv) it is possible that team sport players may respond better psychologically, in terms of motivation, to sport-specific physical conditioning rather than nonspecific traditional, continuous or interval-based conditioning. In consideration of these factors, researchers have designed and investigated the efficacy of various sport-specific methods to develop aerobic endurance.[9,18] 4.2 Examples of Sport-Specific Aerobic Conditioning

Sport-specific aerobic conditioning can take many forms. One example is a dribbling circuit that incorporates changes of direction, acceleration and deceleration and skills specific to the sport. Such circuits have been utilized to improve aerobic fitness, especially in soccer.[8,9,13] One of the first studies to investigate the effectiveness of this strategy was conducted by Hoff et al.,[9] who designed a soccer-specific dribbling track (figure 4), where accelerations, changes of direction and activities with the ball were used for specific interval training, alongside small-sided games. In Norwegian first division players, Hoff et al.[9] reported that interval training using Sports Med 2009; 39 (8)

Study

No. of subjects

Mean (– SD) age (y)

Season

18 M

14.0 – 0.4

In

Hoff et al.[9]

6M

Kelly and Drust[10]

8M

Chamari et al.[8]

Training intervention duration sessions per week

628

ª 2009 Adis Data Information BV. All rights reserved.

Table III. Soccer-specific aerobic conditioning in team sports and the subsequent influence on aerobic fitness Findings intensity

(4 min : 3 min) · 4, Hoff Track and 4v4 SSG

90–95% HRpeak : 60–70% HRpeak

. VO2peak › 8% RE fl 14%**

22.2 – 3.3

(4 min : 3 min) · 4, Hoff Track (4 min : 3 min) · 4, 5v5 SSG

90–95% HRpeak : 70% HRpeak 90–95% HRpeak : 70% HRpeak

94% HRpeak, 92% . VO2peaka 91% HRpeak, 85% . VO2peaka

18.0 – 1.0

(4 min : 2 min) · 4, 5v5 SSG, 30 · 20 m pitch (4 min : 2 min) · 4, 5v5 SSG, 40 · 30 m pitch (4 min : 2 min) · 4, 5v5 SSG, 50 · 40 m pitch

91% HRpeaka 90% HRpeaka 89% HRpeaka

(2 min : 2 min) · 4, 2v2 SSG, 30 · 20 m pitch (3 min : 1.5 min) · 4, 3v3 SSG, 40 · 30 m pitch (3.5 min : 2 min) · 5, 4v4 SSG, 50 · 30 m pitch (5 min : 1.5 min) · 3, 5v5 SSG, 55 · 30 m pitch (6 min : 1.5 min) · 3, 6v6 SSG, 60 · 40 m pitch (10 min : 1.5 min) · 3, 8v8 SSG, 70 · 45 m pitch (2 min : 2 min) · 5, 5v5 pr SSG, 60 · 35 m pitch (2 min : 2 min) · 5, 6v6 pr SSG, 65 · 30 m pitch

91% HRpeaka 91% HRpeaka 90% HRpeaka 89% HRpeaka 88% HRpeaka 88% HRpeaka 90% HRpeaka 91% HRpeaka

(2 min : 2 min) · 4, 2v2 SSG, 30 · 20 m pitch (3.5 min : 1.5 min) · 4, 3v3 SSG, 43 · 25 m pitch (4 min : 2 min) · 4, 4v4 SSG, 40 · 30 m pitch (6 min : 1.5 min) · 4, 5v5 SSG, 45 · 30 m pitch (8 min : 1.5 min) · 3, 6v6 SSG, 50 · 30 m pitch

89% HRpeaka 91% HRpeaka 90% HRpeaka 89% HRpeaka 88% HRpeaka 88% HRpeaka

Little and Williams[11]

23 (sex not given)

22.8 – 4.5

Little and Williams[12]

28 (sex not given)

24.0 – 5.0

In

8 weeks

2

Continued next page

Stone & Kilding

Sports Med 2009; 39 (8)

mode (work : recovery)

ª 2009 Adis Data Information BV. All rights reserved.

Findings demonstrate the intensities achieved during one bout of the training mode. a

ET . = endurance test; HR = heart rate; HRpeak = heart rate peak; IRT = intermittent recovery test; M = male; pr = pressure half switch; RE = running economy; SSG = small-sided games; VO2peak = peak oxygen uptake; fl indicates decrease; › indicates increase; * p < 0.05, ** p < 0.01, *** p < 0.001.

>83% HRpeak : unknown

Yo-yo IRT › 7%** Yo-yo ET › 44%***

629

(4 min : 3 min) · 3, 3v3 up to 6v6 SSG on varying pitch dimensions 2 8 months Pre, in 20 M Rampinini et al.[14]

24.5 – 4.1

(4 min : 3 min) · 4, Hoff Track 2 10 weeks Pre, in 16.9 – 0.4 11 M McMillan et al.[13]

(8 min : 1.5 min) · 4, 8v8 SSG, 70 · 45 m pitch

No. of subjects Study

Table III. Contd

Mean (– SD) age (y)

Season

Training intervention duration sessions per week

mode (work : recovery)

intensity

90–95% HRpeak : 70% HRpeak

Findings

. VO2peak › 11%*** HR @ 9 km/h fl 5%*

Aerobic Conditioning for Team Sport Athletes

a dribbling track resulted in physical. loads equivalent to 94% HRpeak and 92% VO2peak, which are optimal intensities for developing aerobic fitness.[109] Similarly, it was demonstrated that accompanying interval training sessions using small-sided games (5 vs 5) induced steady-state exercise intensities of 91% .HRpeak, corresponding to approximately 85% VO2peak. The small sample size in this study (n = 6) is an obvious limitation; however, other recent research reinforces this finding where exercise intensities achieved during small-sided games of various conditions were found to range from 87% to 91% HRpeak.[10-12] Together, both training modes provided an optimal training intensity. Interestingly, however, players with the .highest . VO2peak elicited the lowest percentage of VO2peak during small-sided games, suggesting that the playing situation designed for this experiment had a ceiling effect for the achievable intensity, and consequently the development of aerobic endurance. That is, the technical/tactical constraints of the game prevented maximal intensities from being reached for some players. Therefore, for athletes with an already high aerobic capacity, and with a good skill level, the aerobic energy system would not be fully stimulated under these training conditions. It may be preferential to prescribe traditional, interval-based, aerobic conditioning, where high workloads can be achieved for sustained periods. Alternatively, it is possible that with some modification – e.g. reducing the number of players, coach encouragement, or increasing the pitch size – such small-sided games may elicit a more intense/strenuous scenario, which could be physiologically beneficial for athletes with a relatively high initial aerobic fitness. In support, it has been shown that three-a-side small-sided games result in more high-intensity activity, greater overall distance covered, less jogging, less walking, higher heart rates and more tackling, dribbling, goal attempts and passes than five-a-side soccer games.[145] Likewise, some evidence suggests that when player numbers are kept constant, a larger playing area increases the intensity of the activity, while a smaller playing area has the opposite effect.[14] Independent of player numbers and Sports Med 2009; 39 (8)

Stone & Kilding

630

20 cm hurdle cone 30 m

10 m

10 m A B 2m

15 m

10 m Start Fig. 4. Soccer-specific dribbling track – ‘The Hoff Track’ (reproduced from Hoff et al.,[9] with permission from BMJ Publishing Group Ltd).

pitch size, the skill level of the player may influence the achievable exercise intensity. For example, it has been reported that junior players, with less skill, are not able to maintain the skill/ drill/technique at a fast enough pace, or with sufficient consistency, to achieve and maintain the required metabolic stress; as such, training may be counterproductive.[146] Such athletes may not achieve the optimal physiological adaptation during sport-specific aerobic conditioning, which might negatively influence future playing performance. Clearly, examining the characteristics of responders versus non-responders (e.g. age, skill level and fitness level) to modern aerobic conditioning approaches, such as small-sided games and dribbling circuits, presents itself as a future research opportunity. 4.3 Small-Sided Games for Soccer

As mentioned above, the modifiable characteristics/ parameters of small-sided games may be influential in determining physical loads. In a comprehensive ª 2009 Adis Data Information BV. All rights reserved.

study, Rampinini et al.[14] examined the effects of player numbers, field dimensions and coach encouragement on the exercise intensity of smallsided soccer games, designed specifically for aerobic conditioning. Twice per week, over 8 months, 20 amateur soccer players performed a total of 67 three-, four-, five- or six-a-side games as interval training. Games were played on three differentsized pitches with varying dimensions (small: 12–24 · 20–32 m; medium: 15–30 · 25–40 m; large: 18–36 · 30–48 m), with and without coach encouragement. Each small-sided game consisted of three bouts of 4 minutes, with 3 minutes of active recovery separating bouts. A specific training intensity was not prescribed for the work intervals, nor was the type of activity defined for the recovery periods. Although not the primary aim of the study, performance measures following the training period increased, with the group mean for the yo-yo IRT improving by 7.4% (p < 0.01) and for the yo-yo endurance test, by an extraordinary 44.3% (p < 0.001). These increases provide evidence of the benefits of performing sport-specific aerobic conditioning in soccer players. An improvement in yo-yo IRT performance suggests a potential increase in soccer performance, given that the amount of high-intensity running performed during a soccer match has been closely associated with the distance covered during the yo-yo IRT.[125] The factor that had the greatest impact on the physiological response to small-sided games was encouragement, followed by player numbers and field dimensions.[14] Three-a-side games were more intense than four-, five- and six-a-side games, irrespective of field dimensions and coach encouragement. Higher exercise intensities when fewer players are on the pitch might be due to the players having more possession of the ball.[147,148] In the same way, a larger pitch size produced higher exercise intensities than a smaller size (1%; p < 0.017), independent of player numbers and coach encouragement, albeit only marginally. Not surprisingly, in all situations, small-sided games with coach encouragement produced higher heart rate (2.5%) and blood lactate concentration (30%) responses than without. By manipulating such variables, it would be possible to impose a sufficient physiological stress on players already possessing Sports Med 2009; 39 (8)

Aerobic Conditioning for Team Sport Athletes

a high level of aerobic fitness. A factor not considered by Rampinini et al.[14] was the impact of playing rules, or ‘conditions’, on the physiological responses to small-sided games. In contrast to Rampinini et al.,[14] Kelly and Drust[10] showed that when player numbers are kept constant, pitch dimensions do not seem to influence the intensity of small-sided games when expressed as %HRpeak (table III). In addition, no significant difference was observed in the total number of technical actions (passing, receiving, turning, dribbling, interceptions and heading) performed by players when pitch dimensions increased. The similarity in the frequency of technical actions across varying pitch dimensions suggests that pitch size is not a major determinant of the number of technical actions performed.[10] However, important technical actions, such as shots on goal, were significantly greater using a small pitch than with a medium or large pitch (small: 85 – 15; medium: 60 – 18; large: 44 – 9; p < 0.05). Therefore, these findings suggest that altering pitch dimensions should be considered if a combined physical training stimulus and technical work on shooting in soccer is desired. The differing results of Rampinini et al.[14] and Kelly and Drust[10] demonstrate the need for a better understanding of the factors that contribute to overload in small-sided games. Other research into small-sided soccer games has revealed such modes of aerobic conditioning are a reliable aerobic training stimulus.[11,149] The work of Hill-Haas et al.[149] and Little and Williams[11] has demonstrated low variability across a variety of small-sided game formats where player numbers and pitch dimensions have been altered. Such studies have demonstrated low variability in physiological measures (such as %HRpeak[11,149]) and time-motion measures (such as total distance covered and percentage of total time at low velocities[149]) during both continuous and interval-based small-sided games. However, test-retest variability tends to increase for higher velocities,[149] possibly due to global positioning systems only sampling distance at 1 Hz combined with the duration of highintensity efforts being very brief (<2 sec). The reliability of the physiological responses and ª 2009 Adis Data Information BV. All rights reserved.

631

external loads observed during small-sided conditioning games for soccer suggest that such training modes allow for optimized group physical conditioning and therefore represent a viable alternative to traditional running interval training for developing and maintaining aerobic fitness. Similar to the work of Hoff et al.,[9] Chamari et al.[8] reported on an 8-week (twice per week) training study involving 18 young male soccer players. Once per week, players performed four 4-minute bouts on the Hoff track, at 90–95% HRpeak, separated by 3 minutes’ active recovery at 60–70% HRpeak. During the second session, on the following day, players participated in small-sided games (4 vs 4) on a 20 m square pitch, at the same intensity as in session one. The 3-minute active recovery involved two players passing and juggling with the ball. When expressed in mL/kg/min, this . training regimen resulted in a 7.5% increase in VO2peak and a 14% improvement in running economy when running at 7 km/h. Heart rate at 7 km/h also decreased by 9 beats/min, indicating improved stroke volume. Likewise, with 16 young male soccer players, McMillan et al.[13] demonstrated that 10 weeks of aerobic endurance training, using the Hoff track in a similar manner to Chamari et al.,[8] was equally effective in elevating . VO2peak (6.4 mL/kg/min, .or 9%). Given these reported improvements in VO2peak, and considering the findings of previous research,[5] it is reasonable to suggest that a concomitant increase in total distance travelled and average exercise intensity would be observed during a competitive match, following each training period. Unfortunately, weaknesses in both the Chamari et al.[8] and McMillan et al.[13] studies were a low sample size, the lack of a control group, and that no match performance measures were reported following the training intervention periods. Despite these limitations, both studies demonstrate that a specifically designed dribbling track, and small-sided conditioning games, allow young soccer players to perform at high percentages (>85%) of .HRpeak . and VO2peak, resulting in improvements in VO2peak of 7.5–9% over an 8- to 10-week period. Regardless of the recent advances in aerobic conditioning for team sport athletes, elite soccer Sports Med 2009; 39 (8)

Stone & Kilding

632

players who are playing the most matches and are training the most for soccer still have only . modest VO2peak values – typically an average of 60–64 mL/kg/min.[6] Hoff and Helgerud[112] have addressed this, explaining that the stroke volume of the heart needs to be increased by achieving high cardiac outputs for sustained periods during training. This consequently leads towards a training model that is of high aerobic intensity, but also with a duration long enough to create high cardiac output without breaking the muscular-venous pumping action that is fundamental for a high stroke volume.[86] This potentially indicates, from a purely physiological perspective, that small-sided games could have a potential limitation in providing a sufficient stimulus, as they are often more intermittent than traditional methods of training (i.e. interval running), thus setting the muscular-venous pump to zero and consequently not allowing a high stroke volume to be achieved. In consideration of this, the rules/structure of small-sided games could be manipulated to create fewer stops in player movement in an attempt to create a greater cardiovascular load. 4.4 Small-Sided Games for Other Sports

Most previous sport-specific training studies have considered soccer; however, there are also studies in other sports, such as rugby league[18,150] and rugby union.[151] Skill-based conditioning games for rugby league have been designed to develop specific aspects of the game, including scrambling defence and support play, the ability to play the ball at speed, defence line speed, ball control and patience. During such conditioning activities, Gabbett[150] measured similar heart rate (152 vs 155 beats/min) and blood lactate concentrations (5.2 vs 5.2 mmol/L) during competition and training, which suggests that skillbased conditioning games have the capacity to replicate the intensity of rugby league competition. However, simply replicating game intensity is not enough to induce reasonable improvements in aerobic function, which, as outlined previously, depends on factors such as exercise intensity and duration.[109] A more detailed ª 2009 Adis Data Information BV. All rights reserved.

discussion of the effectiveness of skill-based conditioning games, when compared with traditional aerobic conditioning methods for rugby league players, can be found below (section 5). In addition, blood lactate measures have been reported to be a poor indicator of muscle lactate[51] and are directly influenced by the amount of high-intensity activity performed within 5 minutes of the blood sample being taken.[19] Therefore, such observations must be interpreted with caution. Likewise, the primary activities of skill-based conditioning games in rugby league and rugby union tend to be focused largely on technique compared with physical development (although this will happen to some extent at the same time), and this may influence the extent of improvements in specific aerobic parameters such as . VO2peak and exercise economy. 4.5 Summary

There is no doubt that sport-specific aerobic conditioning has the ability to induce positive changes in aerobic fitness (and technique under physical load), as demonstrated by the collective results of the described studies. However, athletes with high fitness levels and young players with limited skill may not benefit from small-sided games if the specifics of the game (number of players involved and pitch size) are not considered. Dribbling circuits may be more beneficial in this respect. Regardless, in most studies, evidence of the impact of sport-specific aerobic conditioning on subsequent game performance is lacking, though the inherent difficulties and challenges in collecting worthwhile match performance data are acknowledged. 5. Traditional versus Sport-Specific Aerobic Conditioning for Team Sports Whilst there has been an increase in the use of sport-specific conditioning approaches for team sports, several researchers have questioned the effectiveness when compared with traditional methods of conditioning (table IV).[16-18,152] Since a number of investigations,[125,126,141] but not all, have shown that aerobic fitness declines Sports Med 2009; 39 (8)

Study

Gabbett[18]

Sport

Rugby League

Impellizzeri et al.[15] Soccer

Reilly and White[16] Soccer

a

Soccer

Mean (– SD) Season Group Training intervention age (y) duration sessions (wk) per week

session mode duration (min) (work : recovery)

Findings

37 M

22.1 – 0.9

In

Spec

9

2

60–100

Skill-based conditioning games

32 M

22.3 – 0.8

In

Trad

9

2

60–100

Speed, power, agility and endurance training

14

Pre and In

Spec

12

2

(4 min : 3 min) · 4, SSG

Unknown : 60–70% HRpeak

15

Pre and In

Trad

12

2

(4 min : 3 min) · 4, running intervals

90–95% HRpeak : 60–70% HRpeak

intensity . Pred VO2peak * › 5% 3 : 1 win-loss ratio . Pred VO2peak * › 5% 3 : 1 win-loss ratio

9

18.2 – 1.4

Spec

6

2

(4 min : 3 min) · 6, 5v5 SSG

Unknown: 50–60% HRpeak

9

18.2 – 1.4

Trad

6

2

(4 min : 3 min) · 6, running intervals

85–90% HRpeak : 50–60% HRpeak

9

Distance covered › 4% HI activity › 26% . VO2peak › 8% . %VO2peak at LT › 4% RE at LT fl 3% Distance covered › 6% HI activity › 23%

. VO2peak › fl Lapeak › fl Lapeak › fl

Spec

4v4 and 8v8 SSG

91% HR-peak

Trad

Running intervals

85% HR-peak

Findings demonstrate the intensities achieved during one bout of the training mode.

HI = high-intensity; HRpeak = heart rate peak; Lapeak = lactate peak; LT . = lactate threshold; M = male; Pred = predicted; RE = running economy; SSG = small-sided games; Spec = sport-specific training group; Trad = traditional training group; VO2peak = peak oxygen uptake; fl indicates decrease; › indicates increase; › fl indicates no change; * p < 0.05.

633

Sports Med 2009; 39 (8)

Sassi et al.[17]

No. of subjects

Aerobic Conditioning for Team Sport Athletes

ª 2009 Adis Data Information BV. All rights reserved.

Table IV. Effect of traditional vs sport-specific aerobic conditioning on team sport performance and aerobic fitness

Stone & Kilding

634

throughout the competitive season, a priority for team sport athletes during the competitive season must be to focus on at least maintaining aerobic fitness, while at the same time keeping up the practice of game skills.[16] Sport-specific aerobic conditioning methods may prove useful in achieving such priorities, especially when the time between competitive engagements is short. However, coaches and trainers alike need reassurance that sport-specific methods are as effective as traditional approaches for developing aerobic fitness. It has yet to be clearly identified whether a combined training stimulus of skill-based conditioning games, traditional conditioning activities and strength training would improve physiological capacities to a greater extent than either skill-based conditioning games or traditional conditioning activities alone. Nevertheless, many authors have proposed sport-specific exercises, such as small-sided games, as an alternative mode of aerobic conditioning for team sport athletes.[11,12,16-18,149,152]

5.1 Traditional and Sport-Specific Aerobic Conditioning Approaches in Soccer

Several studies have compared the effectiveness of traditional aerobic interval conditioning with skill-based conditioning games in soccer players.[16,17,152] Reilly and White[16] trained 18 professional, premier league soccer players twice per week for 6 weeks. Sport-specific conditioning involved six 4-minute bouts of 5 versus 5 small-sided games, interspersed with 3 minutes of active recovery (jogging at 50–60% HRpeak). Intensity of work intervals during the small-sided games was not reported. Aerobic interval conditioning involved performing six 4-minute periods of running at 85–90% HRpeak, interspersed with 3 minutes’ active recovery, again jogging at 50–60% HR . peak. After the training intervention, predicted VO2peak increased by 0.2% (57.7 – 3.0 to 57.8 – 3.0 mL/kg/min) for the sport-specific group and by 0.3% (57.8 – 3.2 to 58.0 – 3.2 mL/kg/min) for the aerobic interval group. This negligible improve. ment in VO2peak is somewhat surprising, given that the prescribed intensity of interval work was ª 2009 Adis Data Information BV. All rights reserved.

[5,8,13] where, similar to that previously employed, . despite higher pre-training VO2peak values than reported here, the training intervention resulted in . large (7.5–9%) increases in VO2peak. The lack of improvement in aerobic fitness observed by Reilly and White[16] may be related to the number of players involved in the small-sided games,[145] the rules of the game, and/or the playing area in which [153] Adthe conditioning games were conducted. . ditionally, indirectly measuring VO2peak using the multistage shuttle run test may have influenced the accuracy of such findings.[139] Nevertheless, it was concluded by the authors that small-sided conditioning games were an acceptable substitute for aerobic interval training to maintain fitness during the competitive season. Similarly, Sassi et al.[17] compared the responses of repetitive interval running with small-sided games (4 vs 4 and 8 vs 8) and drills for technical/tactical training in top European league soccer players. Repetitive running consisted of 4 · 1000 m runs, separated by 150 sec of recovery. The authors concluded that small-sided games with the ball could present physiological training stimuli comparable to and sometimes exceeding interval training without the ball. This was demonstrated by the higher heart rates observed during small-sided games (178 – 7 beats/min) than in the repetitive running bouts (167 – 4 beats/min). This finding supports the earlier work of Reilly and Ball,[148] who showed a higher energetic cost of dribbling with a ball compared with normal running. The authors found, irrespective of speed, an added cost of 5.2 kJ/min when a ball was involved and an increase in blood lactate concentration. Collectively, therefore, these findings suggest that small-sided games are adequate alternatives to traditional repetitive running bouts. In a similar study, but of longer duration (14 weeks), 29 soccer players trained twice a week with part of the training session devoted to aerobic interval training.[152] Both the sport-specific (n = 14) and generic training (n = 15) groups completed four bouts of exercise lasting 4 minutes, separated with 3 minutes of active recovery (60–70% HRpeak), as suggested by Helgerud et al.[5] There was, however, no control group for this study. The mode of exercise for the generic

Sports Med 2009; 39 (8)

Aerobic Conditioning for Team Sport Athletes

training group was running around a regular soccer pitch at an intensity corresponding to 90–95% HRpeak. The sport-specific training group played different small-sided games (3 vs 3 with goal keeper, 2–3 ball touches, 25 · 35 m field dimensions; 4 vs 4 with goal keeper, 2 ball touches, 40 · 50 m field; 4 vs 4 and 5 vs 5). The average exercise intensity, expressed as a percentage of HRpeak, during the generic training sessions was not different from that achieved during the sport-specific training sessions (90.7 – 1.2% and 91.3 – 2.2%, respectively), suggesting that both approaches result in sufficient exercise intensities to promote aerobic adaptation. . However, after training, greater increases in VO2peak (7%), LT (10%) and exercise economy at LT (2%) were observed in the generic training group. Despite . similar training intensity and pre-training VO2peak values (56–58[152] and 58 mL/kg/min[16]), these increases are substantially greater than those previously reported by Reilly and White.[16] The differences observed are most probably due to the greater duration of the training intervention (14 vs 6 weeks). Nevertheless, it should be noted that the improvements reported by Impellizzeri et al.[152] are lower than the . corresponding 10%, 16% and 7% increases in VO2peak, LT and exercise economy, respectively, reported by Helgerud et al.[5] after only 8 weeks of interval training. This could be explained by different initial fitness levels and possibly the type of training programme employed by Helgerud et al.[5] prior to the training intervention. More importantly, in addition to the measured increases in aerobic fitness, Impellizzeri et al.[152] observed substantial changes in several measures of match performance, for both training groups, albeit derived from one (post-training) match analysis. Most relevant to soccer performance were the increases in the time spent performing high-intensity activities, 22.8% and 25.5% for the generic and sport-specific training groups, respectively. The amount of high-intensity activity performed is generally accepted to differentiate top-level professional players from those of a lower standard, and therefore it is an important parameter to consider.[25] In addition, ª 2009 Adis Data Information BV. All rights reserved.

635

high-intensity activities are generally associated with critical moments in a soccer match, such as scoring a goal. The total distance covered during match play also increased post-training, by 6.4% and 4.2% for the generic and sport-specific training groups, respectively. However, these increases (594 and 399 m) were lower than the remarkable 1716 m previously reported by Helgerud et al.[5] While soccer players generally do not run around on the pitch without purpose or intent, the total distance travelled during a soccer match is a poor indicator of soccer performance.[30] The differences in the improvements in total distance travelled in these studies could have been influenced by several factors, including: (i) the importance of the match; (ii) the skill level of the opposition; (iii) seasonal variation;[43,44] and (iv) the tactical approach used. Other match performance characteristics evaluated included the time spent performing lowintensity activities, which increased by 18.2% for the generic training group and by 7% for the sport-specific training group. This difference is difficult to explain, given that both groups performed active recovery – jogging at 60–70% HRpeak. Active recovery during training would essentially induce improvements in exercise economy at the intensities associated with recovery, therefore allowing greater ground to be covered during a match situation at lower intensities. Finally, time spent walking decreased in both groups by a similar amount (9.3% and 8.2% for the generic and sport-specific training groups, respectively). This suggests that players were more ‘engaged’ in the game. In summary, the findings of Impellizzeri et al.[152] demonstrate that sport-specific aerobic conditioning has minimal advantages over traditional interval-based aerobic conditioning, with respect to increases in aerobic fitness and, most importantly, match performance characteristics. It should be noted that these findings refer only to soccer. Future studies could examine the influence of combining both sport-specific and traditional aerobic conditioning methods within the same training programme and its subsequent effects on match performance characteristics. Sports Med 2009; 39 (8)

Stone & Kilding

636

5.2 Traditional and Sport-Specific Aerobic Conditioning and Other Sports

In other football codes, interval training using distances and activities specifically related to competition has been reported. For example, in rugby league, activities such as (i) moving up and back over 10 m for periods of 30–90 sec; (ii) repeat tackling efforts on a bag for 5–10 repetitions; and (iii) sprint efforts over distances of 5–60 m have been recommended, with varying exercise-to-rest ratios.[143] Recently, similar skill-based conditioning games have been compared with traditional conditioning activities in rugby league.[18] Skill-based conditioning consisted of games designed to develop passing, catching and ballcarrying technique, tackling technique, scrambling defence and supportive play, play-the-ball speed, defensive line speed and shape, and ball control. Traditional conditioning sessions were not strictly aerobic in nature and consisted of speed, muscular power, agility and aerobic endurance training common to rugby league. Both groups performed twice-weekly training sessions, of approximately 60–100 min duration, over a 9-week in-season training period. Training intensity was estimated using a modified rating of perceived exertion scale:[154] no significant differences were detected between conditioning groups. Gabbett[18] compared the performance of rugby league athletes participating in skill-based conditioning games versus traditional conditioning, . and observed similar increases in predicted VO2peak (4.7% and 5.2%, respectively). In terms of performance, both groups won 75% of their games during the training period; however, the teams adopting the skill-based conditioning games approach scored significantly more points per game (61%; p < 0.05) and conceded fewer points per game compared with traditional training methods. As previously mentioned for soccer,[7] these differences could be influenced by several factors, including, but not limited to, ground and environmental conditions, injuries and the quality of the opposition.[18] However, skill-based conditioning games have been recommended for team sport athletes as a method of developing skills under pressure and ª 2009 Adis Data Information BV. All rights reserved.

fatigue.[155,156] As such, the sport-specific training regimen employed by Gabbett[18] appeared to transfer better into the competitive rugby league environment, enhancing decision-making while under pressure from opposition and ultimately resulting in greater points scored and fewer points conceded. However, many factors can influence the points differential during rugby league, making this method of assessing the impact of sport-specific aerobic conditioning methods on team performance somewhat unreliable. Therefore, future research could examine the effect of traditional and sport-specific aerobic conditioning methods on standardized running protocols that reflect the competition demands of rugby league. 5.3 Summary

It appears that sport-specific and traditional aerobic conditioning approaches are equivocal in soccer and rugby league, in terms of developing aerobic fitness and match performance. As expected, the magnitude of response in most instances is dependent upon the intensity, frequency and duration of training, as well as the total duration of the training programme and the initial fitness level of the athletes involved. Sportspecific conditioning games may be slightly more strenuous than traditional training approaches, as demonstrated by elevated heart rate responses, which may potentially evoke greater improvements in cardiovascular function and subsequently aerobic fitness. These higher responses can be attributed to the additional physical demands imposed on players during activities such as small-sided games,[157] where the use of a ball increases the metabolic cost of performing any given activity, and possibly increased motivation and enthusiasm of players when completing sport-specific conditioning games/drills. Unfortunately, many studies are hampered by a small sample size, lack of a control group, and indirect measures of aerobic fitness. 6. Conclusion It has been well established (especially in basketball, rugby union and soccer) that team Sports Med 2009; 39 (8)

Aerobic Conditioning for Team Sport Athletes

sport athletes require a well developed level of aerobic fitness, in order to maintain repeated high-intensity efforts and to recover adequately between such activities throughout a typical game (40–90 min). Research to date suggests that these adaptations can be achieved by regularly performing aerobic conditioning. Traditional aerobic conditioning, with minimal changes of direction and no skill component, has been demonstrated to effectively increase aerobic function within a 4- to 10-week period in team sport players. More importantly, traditional aerobic conditioning methods have been shown to increase sport performance substantially, with increases in total distance covered and the number of sprints performed during a match. Many professional team sports require the upkeep of both aerobic fitness and sport-specific skills during the competitive season. With classic team sport trainings being shown to evoke marginal increases/decreases in aerobic fitness, sportspecific aerobic conditioning methods have been designed to allow adequate intensities. to be achieved to induce improvements in VO2peak. Such activities have incorporated movement- and skill-specific tasks, such as small-sided games and dribbling circuits. Sport-specific conditioning methods have . been demonstrated to promote increases in VO2peak; however, little research to date has addressed the subsequent effects on game performance. The effectiveness of sport-specific conditioning appears to be influenced by the skill level of the athlete, where those with a lower skill level may not be able to maintain the skill or drill at a suitable intensity to promote the desired aerobic adaptations. Current fitness must also be considered. Players with already high levels of fitness may easily achieve the desired physical load during small-sided games and thus not achieve a training effect. Skill- and fitnessrelated issues can be overcome by manipulating conditions such as player numbers, field dimensions, game rules and coach encouragement: smaller playing numbers, larger playing areas and coach encouragement tend to increase the metabolic loading of small-sided games. When traditional and sport-specific conditioning approaches are compared, results are equivocal. Both ª 2009 Adis Data Information BV. All rights reserved.

637

approaches promote similar increases in aerobic fitness and sport performance when training intensity and volume are constant. Definitely the most important benefit of performing sportspecific conditioning is that it allows for both aerobic fitness and game skills to be developed simultaneously. Acknowledgements The authors are grateful to AUT University for allowing time to prepare this paper. The authors have no conflicts of interest that are directly relevant to the content of this review.

References 1. Bishop D, Edge J, Goodman C. Muscle buffer capacity and aerobic fitness are associated with repeated-sprint ability in women. Eur J Appl Physiol 2004; 92 (4-5): 540-7 2. Bishop D, Spencer M. Determinants of repeated-sprint ability in well-trained team-sport athletes and endurancetrained athletes. J Sports Med Phys Fitness 2004; 44 (1): 1-7 3. McMahon S, Wenger HA. The relationship between aerobic fitness and both power output and subsequent recovery during maximal intermittent exercise. J Sci Med Sport 1998; 1 (4): 219-27 4. Balabinis CP, Psarakis CH, Moukas M, et al. Early phase changes by concurrent endurance and strength training. J Strength Cond Res 2003; 17 (2): 393-401 5. Helgerud J, Engen LC, Wisloff U, et al. Aerobic endurance training improves soccer performance. Med Sci Sports Exerc 2001; 33 (11): 1925-31 6. Helgerud J, Kemi OJ, Hoff J. Pre-season concurrent strength and endurance development in elite soccer players. In: Hoff J, Helgerud J, editors. Football (soccer): new developments in physical training research. Trondheim: Norwegian University of Science and Technology, 2003: 55-66 7. Dupont G, Akakpo K, Berthoin S. The effects of in-season, high-intensity interval training in soccer players. J Strength Cond Res 2004; 18 (3): 584-9 8. Chamari K, Hachana Y, Kaouech F, et al. Endurance training and testing with the ball in young elite soccer players. Br J Sports Med 2005; 39: 24-8 9. Hoff J, Wisloff U, Engen LC, et al. Soccer specific aerobic endurance training. Br J Sports Med 2002; 36: 218-21 10. Kelly DM, Drust B. The effect of pitch dimensions on heart rate responses and technical demands of small-sided soccer games in elite players. J Sci Med Sport 2009;12 (4): 475-9 11. Little T, Williams AG. Suitability of soccer training drills for endurance training. J Strength Cond Res 2006; 20 (2): 316-9

Sports Med 2009; 39 (8)

638

12. Little T, Williams AG. Measures of exercise intensity during soccer training drills with professional soccer players. J Strength Cond Res 2007; 21 (2): 367-71 13. McMillan K, Helgerud J, Macdonald R, et al. Physiological adaptations to soccer specific endurance training in professional youth soccer players. Br J Sports Med 2005; 39: 273-7 14. Rampinini E, Impellizzeri FM, Castagna C, et al. Factors influencing physiological responses to small-sided soccer games. J Sports Sci 2007; 25 (6): 659-66 15. Impellizzeri FM, Marcora SM, Castagna C, et al. Physiological and performance effects of generic versus specific aerobic training in soccer players. Int J Sports Med 2006; 27 (7): 483-92 16. Reilly T, White C. Small-sided games as an alternative to interval training for soccer players [abstract]. J Sports Sci 2004; 22 (6): 559 17. Sassi R, Reilly T, Impellizzeri F. A comparison of smallsided games and interval training in elite professional soccer players [abstract]. J Sports Sci 2004; 22 (6): 562 18. Gabbett TJ. Skill-based conditioning games as an alternative to traditional conditioning for rugby league players. J Strength Cond Res 2006; 20 (2): 309-15 19. Abdelkrim NB, El Fazaa S, El Ati J. Time-motion analysis and physiological data of elite under-19-year-old basketball players during competition. Br J Sports Med 2007; 41 (2): 69-75 20. McInnes SE, Carlson JS, Jones CJ, et al. The physiological load imposed on basketball players during competition. J Sports Sci 1995; 13 (5): 387-97 21. Deutsch MU, Maw GJ, Jenkins D, et al. Heart rate, blood lactate and kinematic data of elite colts (under-19) rugby union players during competition. J Sports Sci 1998; 16: 561-70 22. Duthie G, Pyne D, Hooper S. Time motion analysis of 2001 and 2002 super 12 rugby. J Sports Sci 2005; 23 (5): 523-30 23. Bangsbo J, Norregaard L, Thorso F. Activity profile of competition soccer. Can J Sport Sci 1991; 16 (2): 110-6 24. Bloomfield J, Polman R, O’Donoghue P. Physical demands of different positions in FA premier league soccer. J Sports Sci Med 2007; 6: 63-70 25. Mohr M, Krustrup P, Bangsbo J. Match performance of high-standard soccer players with special reference to development of fatigue. J Sports Sci 2003; 21 (7): 519-28 26. Rienzi E, Drust B, Reilly T, et al. Investigation of anthropometric and work-rate profiles of elite South American international soccer players. J Strength Cond Res 2000; 40 (2): 162-9 27. Morton AR. Applying physiological principles to rugby training. Sports Coach 1978; 2: 4-9 28. Krustrup P, Mohr M, Ellingsgaard H, et al. Physical demands during an elite female soccer game: importance of training status. Med Sci Sports Exerc 2005; 37 (7): 1242-8 29. Mayhew SR, Wenger HA. Time-motion analysis of professional soccer. J Hum Move Studies 1985; 11: 49-52 30. van Gool D, van Gerven D, Boutmans J. The physiological load imposed on soccer players during real match-play. In: Reilly T, Lees A, Davids K, et al., editors. Science and football. London: E & F N Spon, 1988: 51-9

ª 2009 Adis Data Information BV. All rights reserved.

Stone & Kilding

31. Withers RT, Maricic Z, Wasilewski S, et al. Match analysis of Australian professional soccer players. J Hum Move Studies 1982; 8: 159-76 32. Bangsbo J, Mohr M, Krustrup P. Physical and metabolic demands of training and match-play in the elite football player. J Sports Sci 2006; 24 (7): 665-74 33. Reilly T. Energetics of high-intensity exercise (soccer) with particular reference to fatigue. J Sports Sci 1997; 15 (3): 257-63 34. Reilly T, Bangsbo J, Franks A. Anthropometric and physiological predispositions for elite soccer. J Sports Sci 2000; 18: 669-83 35. Coutts A, Reaburn P, Abt G. Heart rate, blood lactate concentration and estimated energy expenditure in a semiprofessional rugby league team during a match: a case study. J Sports Sci 2003; 21 (2): 97-103 36. Boyle PM, Mahoney CA, Wallace FM. The competitive demands of elite male field hockey. J Sports Med Phys Fitness 1994; 34: 235-41 37. Florida-James G, Reilly T. The physiological demands of Gaelic football. Br J Sports Med 1995; 29: 41-5 38. Spencer M, Rechichi C, Lawrence S, et al. Time-motion analysis of elite field hockey during several games in succession: a tournament scenario. J Sci Med Sport 2005; 8 (4): 382-91 39. Bishop DC, Wright C. A time-motion analysis of professional basketball to determine the relationship between three activity profiles: high, medium and low intensity and the length of the time spent on court. Int J Perf Analysis Sport 2006; 6 (1): 130-8 40. Bangsbo J, Lindquist F. Comparison of various exercise tests with endurance performance during soccer in professional players. Int J Sports Med 1992; 13 (2): 125-32 41. Docherty D, Sporer B. A proposed model for examining the interference phenomenon between concurrent aerobic and strength training. Sports Med 2000; 30 (6): 385-94 42. Mohr M, Krustrup P, Bangsbo J. Fatigue in soccer: a brief review. J Sports Sci 2005; 23 (6): 593-9 43. Rampinini E, Coutts AJ, Castagna C, et al. Variation in top level soccer match performance. Int J Sports Med 2007; 28: 1018-24 44. Rampinini E, Impellizzeri FM, Castagna C, et al. Technical performance during soccer matches of the Italian Serie A league: effect of fatigue and competitive level. J Sci Med Sport 2009; 12: 227-33 45. Ali A, Farrally M. Recording soccer players’ heart rates during matches. J Sports Sci 1991; 9: 183-9 46. Capranica L, Tessitore A, Guidetti L, et al. Heart rate and match analysis in pre-pubescent soccer players. J Sports Sci 2001; 19 (6): 379-84 47. Mohr M, Krustrup P, Nybo L, et al. Muscle temperature and sprint performance during soccer matches-beneficial effect of re-warm-up at half time. Scand J Med Sci Sports 2004; 14: 156-62 48. Atkins SJ. Performance of the yo-yo intermittent recovery test by elite professional and semiprofessional rugby league players. J Strength Cond Res 2006; 20 (1): 222-5 49. McLean DA. Analysis of the physical demands of international rugby union. J Sports Sci 1992; 10 (3): 285-96

Sports Med 2009; 39 (8)

Aerobic Conditioning for Team Sport Athletes

50. Al-Hazzaa HM, Almuzaini KS, Al-Refaee SA, et al. Aerobic and anaerobic power characteristics of Saudi elite soccer players. J Sports Med Phys Fitness 2001; 41 (1): 54-61 51. Krustrup P, Mohr M, Steensberg A, et al. Muscle and blood metabolites during a soccer game: implications for sprint performance. Med Sci Sports Exerc 2006; 38 (6): 1165-74 52. Bangsbo J. Energy demands in competitive soccer. J Sports Sci 1994; 12: S5-12 53. Hoffman JR, Epstein S, Einbinder M, et al. The influence of aerobic capacity on anaerobic performance and recovery indices in basketball players. J Strength Cond Res 1999; 13 (4): 407-11 54. Treadwell PJ. Computer-aided match analysis of selected ball games (soccer and rugby union). In: Reilly T, Lees A, Davids K, et al., editors. Science and football. London: E & FN Spon, 1988: 282-7 55. Ekblom B. Applied physiology of soccer. Sports Med 1986; 3 (1): 50-60 56. Acevedo EO, Goldfarb AH. Increased training intensity effects on plasma lactate, ventilatory threshold, and endurance. Med Sci Sports Exerc 1989; 21 (5): 563-8 57. Casaburi R, Storer TW, Ben-Dov I, et al. Effect of . endurance training on possible determinants of VO2 during heavy exercise. J Appl Physiol 1987; 62 (1): 199-207 58. Hill NS, Jacoby C, Farber HW. Effect of an endurance triathlon on pulmonary function. Med Sci Sports Exerc 1991; 23 (11): 1260-4 59. Tzankoff SP, Robinson S, Pyke FS, et al. Physiological adjustments to work in older men as affected by physical training. J Appl Physiol 1972; 33 (3): 346-50 60. Andrew GM, Guzman CA, Becklake MR. Effect of athletic training on exercise cardiac output. J Appl Physiol 1966; 21 (2): 603-8 61. Coyle EF, Hemmert MK, Coggan AR. Effects of detraining on cardiovascular responses to exercise: role of blood volume. J Appl Physiol 1986; 60 (1): 95-9 62. Green HJ, Sutton JR, Coates G, et al. Response of red cell and plasma volume to prolonged training in humans. J Appl Physiol 1991; 70 (4): 1810-5 63. Wilmore JH, Stanforth PR, Gagnon J, et al. Cardiac output and stroke volume changes with endurance training: the Heritage family study. Med Sci Sports Exerc 2001; 33 (1): 99-106 64. Wilmore JH, Stanforth PR, Gagnon J, et al. Heart rate and blood pressure changes with endurance training: the Heritage family study. Med Sci Sports Exerc 2001; 33 (1): 107-16 65. Desaulniers P, Lavoie PA, Gardiner PF. Endurance training increases acetylcholine receptor quantity at neuromuscular junctions of adult rat skeletal muscle. Neuroreport 1998; 9 (16): 3549-52 66. Lucia A, Hoyos J, Pardo J, et al. Metabolic and neuromuscular adaptations to endurance training in professional cyclists: a longitudinal study. Jpn J Physiol 2000; 50 (3): 381-8 67. Green HJ, Jones S, Ball-Burnett M, et al. Adaptations in muscle metabolism to prolonged voluntary exercise and training. J Appl Physiol 1995; 78 (1): 138-45

ª 2009 Adis Data Information BV. All rights reserved.

639

68. Green HJ, Jones S, Ball-Burnett ME, et al. Early muscular and metabolic adaptations to prolonged exercise training in humans. J Appl Physiol 1991; 70 (5): 2032-8 69. Holloszy JO, Coyle EF. Adaptations of skeletal muscle to endurance exercise and their metabolic consequences. J Appl Physiol 1984; 56 (4): 831-8 70. LeBlanc PJ, Peters SJ, Tunstall RJ, et al. Effects of aerobic training on pyruvate dehydrogenase and pyruvate dehydrogenase kinase in human skeletal muscle. J Physiol 2004; 557 (2): 559-70 71. Wibom R, Hultman E, Johansson M, et al. Adaptation of mitochondrial ATP production in human skeletal muscle to endurance training and detraining. J Appl Physiol 1992; 73 (5): 2004-10 . 72. Burke EJ, Franks BD. Changes in VO2max resulting from bicycle training at different intensities holding total mechanical work constant. Res Q 1975; 46 (1): 31-7 73. Faria IE. Cardiovascular response to exercise as influenced by training of various intensities. Res Q 1970; 41 (1): 44-50 74. Gaesser GA, Rich RG. Effects of high- and low-intensity exercise training on aerobic capacity and blood lipids. Med Sci Sports Exerc 1984; 16 (3): 269-74 75. Wenger HA, MacNab RB. Endurance training: the effects of intensity, total work, duration and initial fitness. J Sports Med Phys Fitness 1975; 15 (3): 199-211 76. Fox EL, Bartels RL, Billings CE, et al. Frequency and duration of interval training programs and changes in aerobic power. J Appl Physiol 1975; 38 (3): 481-4 77. Hickson RC, Overland SM, Dougherty KA. Reduced training frequency effects on aerobic power and muscle adaptations in rats. J Appl Physiol 1984; 57 (6): 1834-41 78. Pollock ML, Miller HS, Linnerud AC, et al. Frequency of training as a determinant for improvement in cardiovascular function and body composition of middle-aged men. Arch Phys Med Rehabil 1975; 56 (4): 141-5 79. Raven PB, Gettman LR, Pollock ML, et al. A physiological evaluation of professional soccer players. Br J Sports Med 1976; 10 (4): 209-16 80. Terjung RL. Muscle fiber involvement during training of different intensities and durations. Am J Physiol 1976; 230 (4): 946-50 81. Cunningham DA, McCrimmon D, Vlach LF. Cardiovascular response to interval and continuous training in women. Eur J Appl Physiol Occup Physiol 1979; 41 (3): 187-97 82. Hickson RC, Rosenkoetter MA. Reduced training frequencies and maintenance of increased aerobic power. Med Sci Sports Exerc 1981; 13 (1): 13-6 83. McArdle WD, Katch FI, Katch VL. Exercise physiology: energy, nutrition, and human performance. 4th ed. Baltimore (MD): Williams & Wilkins, 1996 84. Saltin B, Nazar K, Costill DL, et al. The nature of the training response; peripheral and central adaptations of one-legged exercise. Acta Physiol Scand 1976; 96 (3): 289-305 85. Sale D, MacDougall D. Specificity in strength training: a review for the coach and athlete. Can J Appl Sport Sci 1981 Jun; 6 (2): 87-92 86. Astrand PO, Rodahl K. Textbook of work physiology. 3rd ed. New York: McGraw-Hill, 1987

Sports Med 2009; 39 (8)

640

87. Laffite LP, Mille-Hamard L, Koralsztein JP, et al. The effects of interval training on oxygen pulse and performance in supra-threshold runs. Arch Physiol Biochem 2003; 111 (3): 202-10 88. Gute D, Fraga C, Laughlin MH, et al. Regional changes in capillary supply in skeletal muscle of highintensity endurance-trained rats. J Appl Physiol 1996; 81 (2): 619-26 89. Gollnick PD, Armstrong RB, Saltin B, et al. Effect of training on enzyme activity and fiber composition of human skeletal muscle. J Appl Physiol 1973; 34 (1): 107-11 90. Harms SJ, Hickson RC. Skeletal muscle mitochondria and myoglobin, endurance, and intensity of training. J Appl Physiol 1983; 54 (3): 798-802 91. Billat. LV, Flechet B, Petit B, et al. Interval training at VO2max: effects on aerobic performance and overtraining markers. Med Sci Sports Exerc 1999; 31 (1): 156-63 92. Jones AM. A five year physiological case study of an Olympic runner. Br J Sports Med 1998; 32: 39-43 93. Davis JA, Frank MH, Whipp BJ, et al. Anaerobic threshold alterations caused by endurance training in middleaged men. J Appl Physiol 1979; 46 (6): 1039-46 94. Carter H, Jones AM, Barstow TJ, et al. Effect of endurance training on oxygen uptake kinetics during treadmill running. J Appl Physiol 2000; 89: 1744-52 95. Jones AM, Carter H. The effect of endurance training on parameters of aerobic fitness. Sports Med 2000; 29 (6): 373-86 96. Bogdanis GC, Nevill ME, Boobis LH, et al. Contribution of phosphocreatine and aerobic metabolism to energy supply during repeated sprint exercise. J Appl Physiol 1996; 80 (3): 876-84 97. Bangsbo J. Physiological demands of soccer. In: Ekblom B, editor. Football (soccer). London: Blackwell Scientific, 1994: 43-59 98. Edwards AM, Clark N, Macfayden AM. Lactate and ventilatory thresholds reflect the training status of professional soccer players where maximum aerobic power is unchanged. J Sports Sci Med 2003; 2: 23-9 99. Sirotic AC, Coutts AJ. Physiological and performance test correlates of prolonged, high-intensity, intermittent running performance in moderately trained women team sport athletes. J Strength Cond Res 2007; 21 (1): 138-44 100. Bishop D, Lawrence S, Spencer M. Predictors of repeatedsprint ability in elite female hockey players. J Sci Med Sport 2003; 6 (2): 199-209 101. Tomlin DL, Wenger HA. The relationship between aerobic fitness and recovery from high intensity intermittent exercise. Sports Med 2001; 31 (1): 1-11 102. Spencer M, Lawrence S, Rechichi C, et al. Timemotion analysis of elite field hockey, with special reference to repeated-sprint activity. J Sports Sci 2004; 22 (9): 843-50 103. Phillips SM, Green HJ, MacDonald MJ, . et al. Progressive effect of endurance training on VO2 kinetics at the onset of submaximal exercise. J Appl Physiol 1995; 79 (6): 1914-20

ª 2009 Adis Data Information BV. All rights reserved.

Stone & Kilding

104. Dupont G, Millet GP, Guinhouya C, et al. Relationship between oxygen uptake kinetics and performance in repeated running sprints. Eur J Appl Physiol 2005; 95: 27-34 105. McCully KK, Kakihira H, Vandenborne K. Noninvasive measurements of activity-induced changes in muscle metabolism. J Biomech 1991; 21 Suppl. 1: 153-61 106. McMahon S, Jenkins D. Factors affecting the rate of phosphocreatine resynthesis following intense exercise. Sports Med 2002; 32: 761-84 107. Neya M, Ogawa Y, Matsugaki N, et al. The influence of acute hypoxia on the prediction of maximal oxygen uptake using multi-stage shuttle run test. J Sports Med Phys Fitness 2002; 42 (2): 158-64 108. Takahashi H, Inaki M, Fujimoto K. Control of the rate of phosphocreatine resynthesis after exercise in trained and untrained human quadriceps muscles. Eur J Appl Physiol 1995; 71: 396-404 109. Wenger HA, Bell GJ. The interaction of intensity, frequency and duration of exercise training in altering cardiorespiratory fitness. Sports Med 1986; 3 (5): 346-56 . 110. Richardson RS. What governs skeletal muscle VO2max? New evidence. Med Sci Sports Exerc 2000; 32 (1): 100-7 111. Zhou B, Conlee RK, Jensen R, et al. Stroke volume does not plateau during graded exercise in elite male distance runners. Med Sci Sports Exerc 2001; 33 (11): 1849-54 112. Hoff J, Helgerud J. Endurance and strength training for soccer players: physiological considerations. Sports Med 2004; 34 (3): 165-80 113. Billat LV. Interval training for performance: a scientific and empirical practice. Special recommendations for middle- and long-distance running – part I, aerobic interval training. Sports Med 2001; 31 (1): 13-31 114. Billat LV. Interval training for performance: a scientific and empirical practice. Special recommendations for middle- and long-distance running – part II, anaerobic interval training. Sports Med 2001; 31 (2): 75-90 115. Millet GP, Jaouen B, Borrani F, et al. Effects of concurrent endurance and strength training on running . economy and VO2 kinetics. Med Sci Sports Exerc 2002; 34 (8): 1351-9 116. Paton CD, Hopkins WG. Effects of high-intensity training on performance and physiology of endurance athletes [online]. Available from URL: http://www.sportsci.org/ jour/04/cdp.pdf [Accessed 2004 Sep 15] 117. Billat VL, Slawinski J, Bocquet V, et al. Intermittent runs at the velocity associated with maximal oxygen uptake enables subjects to remain at maximal oxygen uptake for a longer time than intense but submaximal runs. Eur J Appl Physiol 2000; 81: 188-96 118. Hickson RC, Rosenkoetter MA, Brown MM. Strength training effects on aerobic power and short-term endurance. Med Sci Sports Exerc 1980; 12 (5): 336-9 119. Hoff J, Gran A, Helgerud J. Maximal strength training improves aerobic endurance performance. Scand J Med Sci Sports 2002; 12 (5): 288-95 120. McCall GE, Byrnes WC, Dickinson A, et al. Muscle fiber hypertrophy, hyperplasia, and capillary density in college

Sports Med 2009; 39 (8)

Aerobic Conditioning for Team Sport Athletes

121.

122.

123.

124.

125.

126.

127.

128.

129.

130.

131.

132.

133.

134.

135.

136.

men after resistance training. J Appl Physiol 1996; 81 (5): 2004-12 Osteras H, Helgerud J, Hoff J. Maximal strength-training effects on force-velocity and force-power relationships explain increases in aerobic performance in humans. Eur J Appl Physiol 2002; 88: 255-63 Paavolainen L, Hakkinen K, Hamalainen I, et al. Explosive-strength training improves 5-km running time by improving running economy and muscle power. J Appl Physiol 1999; 86 (5): 1527-33 Hoffman JR, Tenenbaum G, Maresh CM, et al. Relationship between athletic performance tests and playing time in elite college basketball players. J Strength Cond Res 1996; 10 (2): 67-71 Gillam GM. Identification of anthropometric and physiological characteristics relative to participation in college basketball. Nat Strength Cond Assoc J 1985; 7 (3): 34-6 Krustrup P, Mohr M, Amstrup T, et al. The yo-yo intermittent recovery test: physiological response, reliability, and validity. Med Sci Sports Exerc 2003; 35 (4): 697-705 Tavino LP, Bowers CJ, Archer CB. Effects of basketball on aerobic capacity, anaerobic capacity, and body composition of male college players. J Strength Cond Res 1995; 9 (2): 75-7 Gabbett TJ. Reductions in pre-season training loads reduce training injury rates in rugby league players. Br J Sports Med 2004; 38: 743-9 Hoffman JR, Fry AC, Howard R, et al. Strength, speed and endurance changes during the course of a division I basketball season. J Appl Sport Sci Res 1991; 5 (3): 144-9 Hunter GR, Hilyer J, Forster MA. Changes in fitness during 4 years of intercollegiate basketball. J Strength Cond Res 1993; 7 (1): 26-9 Gabbett TJ. Performance changes following a field conditioning program in junior and senior rugby league players. J Strength Cond Res 2006; 20 (1): 215-21 Gabbett TJ. Changes in physiological and anthropometric characteristics of rugby league players during a competitive season. J Strength Cond Res 2005; 19 (2): 400-8 Maresh CM, Wang BC, Goetz KL. Plasma vasopressin, renin activity, and aldosterone response to maximal exercise in active college females. Eur J Appl Physiol 1985; 54: 398-403 Bolonchuk W, Lukaski H, Siders W. The structural, functional, and nutritional adaptation of college basketball players over a season. J Sports Med Phys Fitness 1991; 31 (2): 165-72 Laplaud D, Hug F, Menier R. Training-induced changes in aerobic aptitudes of professional basketball players. Int J Sports Med 2004; 25 (2): 103-8 Gabbett TJ. Physiological and anthropometric characteristics of junior rugby league players over a competitive season. J Strength Cond Res 2005; 19 (4): 764-71 Lamb KL, Rogers L. A re-appraisal of the reliability of the 20 m multi-stage shuttle run test. Eur J Appl Physiol 2007; 100: 287-92

ª 2009 Adis Data Information BV. All rights reserved.

641

137. Aziz AR, Mukherjee S, Chia MYH, et al. Relationship between measured maximal oxygen uptake and aerobic performance with running repeated sprint ability in young elite soccer players. J Sports Med Phys Fitness 2007; 47 (4): 401-7 138. O’Gorman D, Hunter A, McDonnacha C, et al. Validity of field tests for evaluating endurance capacity in competitive and international-level sports participants. J Strength Cond Res 2000; 14 (1): 62-7 139. Kilding AE, Aziz AR, Teh KC. Measuring and predicting maximal aerobic power in international-level intermittent sport athletes. J Sports Med Phys Fitness 2006; 46 (3): 366-72 140. Bogdanis GC, Ziagos V, Anastasiadis M, et al. Effects of two different short-term training programs on the physical and technical abilities of adolescent basketball players. J Sci Med Sport 2007; 10 (2): 79-88 141. Krustrup P, Mohr M, Nybo L, et al. The yo-yo ir2 test: physiological response, reliability, and application to elite soccer. Med Sci Sports Exerc 2006; 38 (9): 1666-73 142. Lawson E. Incorporating sport-specific drills into conditioning. In: Foran B, editor. High-performance sports conditioning. Champaign (IL): Human Kinetics, 2001: 215-66 143. Meir R, Newton R, Curtis E, et al. Physical fitness qualities of professional rugby league football players: determination of positional differences. J Strength Cond Res 2001; 15 (4): 450-8 144. Gabbett T. Increasing training intensity in country rugby league players. Rugby League Coaching Magazine 2001; 20: 30-1 145. Platt D, Maxwell A, Horn R, et al. Physiological and technical analysis of 3 v 3 and 5 v 5 youth football matches. Insight FA Coaches Assoc J 2001; 4 (4): 23-4 . 146. Castagna C, Belardinelli R, Abt G. The VO2 and HR response to training with the ball in youth soccer players. In: Reilly T, Cabri J, Araujo D, editors. Science and football V: the proceedings of the fifth world congress on science and football. New York: Routledge, 2005: 462-4 147. Balsom P. Precision football. Kempele: Polar Electro Oy, 1999 148. Reilly T, Ball D. The net physiological cost of dribbling a soccer ball. Res Q Exerc Sport 1984; 55: 267-71 149. Hill-Haas S, Coutts A, Rowsell G, et al. Variability of acute physiological responses and performance profiles of youth soccer players in small-sided games. J Sci Med Sport 2008; 11 (5): 487-90 150. Gabbett T. Do skill-based conditioning games simulate the physiological demands of competition? Rugby League Coaching. Manuals 2004; 32: 27-31 151. Gamble P. A skill-based conditioning games approach to metabolic conditioning for elite rugby football players. J Strength Cond Res 2004; 18 (3): 491-7 152. Impellizzeri FM, Rampinini E, Marcora SM. Physiological assessment of aerobic training in soccer. J Sports Sci 2005; 23 (6): 583-92 153. Bangsbo J. Fitness training in soccer: a scientific approach. Spring City (PA): Reedswain Publishing, 2003

Sports Med 2009; 39 (8)

642

154. Foster C, Florhaug JA, Franklin J, et al. A new approach to monitoring exercise training. J Strength Cond Res 2001; 15 (1): 109-15 155. Mallo Sainz J, Navarro E. Analysis of the load imposed on under-19 soccer players during typical football training [abstract]. J Sports Sci 2004; 22: 510 156. Gabbett TJ. Training injuries in rugby league: an evaluation of skill-based conditioning games. J Strength Cond Res 2002; 16 (2): 236-41

ª 2009 Adis Data Information BV. All rights reserved.

Stone & Kilding

157. Reilly T, Robinson G, Minors DS. Some circulatory responses to exercise at different times of day. Med Sci Sports Exerc 1984; 16 (5): 477-85

Correspondence: Dr Andrew E. Kilding, School of Sport and Recreation, AUT University, Private Bag 92006, Auckland 1020, New Zealand. E-mail: [email protected]

Sports Med 2009; 39 (8)

Sports Med 2009; 39 (8): 643-662 0112-1642/09/0008-0643/$49.95/0

REVIEW ARTICLE

ª 2009 Adis Data Information BV. All rights reserved.

The Exercise-Induced Stress Response of Skeletal Muscle, with Specific Emphasis on Humans James P. Morton,1 Anna C. Kayani,2 Anne McArdle2 and Barry Drust1 1 Research Institute for Sport and Exercise Sciences, Liverpool John Moores University, Liverpool, UK 2 School of Clinical Sciences, University of Liverpool, Liverpool, UK

Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. General Function of Heat Shock Proteins (HSPs) and Regulation of HSP Expression . . . . . . . . . . . . . . . 1.1 Function of HSPs in the Unstressed Cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Regulation of HSP Expression. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 Function of HSPs in the Stressed Cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. HSP Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Ubiquitin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 aB-Crystallin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 HSP27 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 HSP60 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 HSP70 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6 HSP90 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. The Effect of Acute Exercise on HSP Content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Endurance Exercise: ‘Non-Damaging’ Exercise Protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Resistance Exercise and Down-Hill Running: ‘Damaging’ Exercise Protocols. . . . . . . . . . . . . . . . . 3.3 Time-Course and Muscle Fibre Specificity of the Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Individual Variation of the Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.1 Training Status. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.2 Recent Activity Levels/Thermal History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.3 Energy Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.4 Sex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.5 Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. The Effects of Exercise Training on HSP Content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Baseline HSP Content. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Magnitude of the Stress Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Possible Physiological Signals Initiating the Exercise-Induced Stress Response . . . . . . . . . . . . . . . . . . . 5.1 Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Oxidative Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Metabolic Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4 Cytokine Production and Inflammatory Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Biological Significance of Exercise-Induced Increases in Muscle HSP Content . . . . . . . . . . . . . . . . . . . 7. Conclusions and Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

644 645 645 645 646 646 646 646 647 647 647 647 648 648 649 650 650 650 650 651 651 651 651 651 652 652 653 654 655 655 656 657

Morton et al.

644

Abstract

Skeletal muscle adapts to the stress of contractile activity via changes in gene expression to yield an increased content of a family of highly conserved cytoprotective proteins known as heat shock proteins (HSPs). These proteins function to maintain homeostasis, facilitate repair from injury and provide protection against future insults. The study of the exercise-induced production of HSPs in skeletal muscle is important for the exercise scientist as it may provide a valuable insight into the molecular mechanisms by which regular exercise can provide increased protection against related and non-related stressors. As molecular chaperones, HSPs are also fundamental in facilitating the cellular remodelling processes inherent to the training response. Whilst the exercise-induced stress response of rodent skeletal muscle is relatively well characterized, data from humans are more infrequent and less insightful. Data indicate that acute endurance- and resistance-type exercise protocols increase the muscle content of ubiquitin, aB-crystallin, HSP27, HSP60, HSC70 and HSP70. Although increased HSP transcription occurs during exercise, immediately post-exercise or several hours following exercise, time-course studies using western blotting techniques have typically demonstrated a significant increase in protein content is only detectable within 1–2 days following the exercise stress. However, comparison amongst studies is complicated by variations in exercise protocol (mode, intensity, duration, damaging, non-damaging), muscle group examined, predominant HSP measured and, perhaps most importantly, differences in subject characteristics both within and between studies (training status, recent activity levels, nutritional status, age, sex, etc.). Following ‘non-damaging’ endurancetype activities (exercise that induces no overt structural and functional damage to the muscle), the stress response is thought to be mediated by redox signalling (transient and reversible oxidation of muscle proteins) as opposed to increases in contracting muscle temperature per se. Following ‘damaging’ forms of exercise (exercise that induces overt structural and functional damage to the muscle), the stress response is likely initiated by mechanical damage to protein structure and further augmented by the secondary damage associated with inflammatory processes occurring several days following the initial insult. Exercise training induces an increase in baseline HSP levels, which is dependent on a sustained and currently unknown dose of training and also on the individual’s initial training status. Furthermore, trained subjects display an attenuated or abolished stress response to customary exercise challenges, likely due to adaptations of baseline HSP levels and the antioxidant system. Whilst further fundamental work is needed to accurately characterize the exercise-induced stress response in specific populations following varying exercise protocols, exercise scientists should also focus their efforts on elucidating the precise biological significance of the exercise-induced induction of HSPs. In addition to their potential cytoprotective properties, the role of HSPs in modulating cell signalling pathways related to both exercise adaptation and health and disease also needs further investigation. As a nonpharmacological intervention, exercise and the associated up-regulation of HSPs and the possible correction of maladapted pathways may therefore prove effective in providing protection against protein misfolding diseases and in preserving muscle function during aging.

ª 2009 Adis Data Information BV. All rights reserved.

Sports Med 2009; 39 (8)

Exercise-Induced Stress Response of Skeletal Muscle

Cells respond to stress via changes in gene expression to yield a family of highly conserved proteins known as heat shock proteins (HSPs).[1] As their name suggests, HSPs were originally found to be up-regulated after exposure to elevated temperatures.[2,3] Since these initial findings, the cellular content of HSPs has also been shown to increase following other stressors such as ischaemia,[4] protein degradation,[5] hypoxia,[6] acidosis,[7] oxidative stress,[8] increased intracellular calcium[9] and energy depletion.[10] Accordingly, the terms ‘stress proteins’ and ‘cellular stress response’ were introduced to reflect the array of stressors known to initiate HSP expression.[11] During the last two decades, the stress of exercise has also been shown to up-regulate HSP content in tissues of various animal species.[12-17] These findings have been extended by data from human studies demonstrating that several HSPs are also up-regulated in the skeletal muscle of humans following various types of exercise protocols. These data are still in their infancy, however, and as yet the exercise-induced stress response of human skeletal muscle is far from well understood. The ability of exercise to initiate the heat shock response provides an important research area for the exercise scientist, as cells demonstrating elevated levels of HSPs following non-damaging episodes of stress are subsequently capable of withstanding normally lethal or damaging stresses.[18-22] The study of HSPs, in relation to exercise, may therefore increase our understanding of the cellular and molecular mechanisms underpinning the increased protection to contraction-induced damage associated with regular exercise. The possibility of using exercise (as a non-pharmacological approach) as a means of harnessing the endogenous protective systems of a cell to provide ‘cross-tolerance’ to other stressors is also an exciting area with obvious health implications.[23] The present review provides an up-to-date and critical review of the effects of acute and chronic exercise on HSP expression of skeletal muscle, where specific emphasis is given to human studies. The reader is firstly introduced to the generic function of HSPs, the regulation of heat ª 2009 Adis Data Information BV. All rights reserved.

645

shock gene transcription, and the major HSP families found in skeletal muscle. The effects of acute and chronic exercise on HSP expression are then discussed, as are the possible methodological factors influencing the response and the potential physiological stressors signalling the response. We conclude by highlighting the possible biological significance of exercise-induced HSP expression and by offering some future research directions. Whilst our intended focus is on data derived from human skeletal muscle, we also review findings (where appropriate) from other species/cell types that may be applicable to the exercising human. 1. General Function of Heat Shock Proteins (HSPs) and Regulation of HSP Expression 1.1 Function of HSPs in the Unstressed Cell

In unstressed cells, constitutively expressed HSPs function as molecular chaperones necessary for facilitating the correct folding of newly synthesized proteins, preventing the aggregation of aberrantly folded proteins, facilitating the refolding of denatured proteins and safely transporting proteins to their correct cellular compartment.[24,25] As a chaperone, HSPs are therefore considered to perform important ‘housekeeping’ functions.[26,27] 1.2 Regulation of HSP Expression

Expression of HSP genes is dependent on the presence of short nucleotide sequences known as heat shock elements (HSEs) located upstream in the promoter region of heat shock responsive genes. The presence of one or more functional HSE is the identifying feature of an HSP gene.[11] The HSE is the binding site for a constitutively expressed protein known as the heat shock factor (HSF). HSF1 is the major stress responsive transcription factor in mammalian cells and is thought to be constitutively expressed in the cytoplasm of unstressed cells.[28] In these situations, HSF1 is thought to exist in an inactive monomer state that co-precipitates with HSP70 and HSP90.[28] During stressful insults, the freely Sports Med 2009; 39 (8)

Morton et al.

646

available HSPs that are present in unstressed situations are subsequently depleted by interactions with unfolded proteins. Those HSPs that are bound with HSF1 also have a higher affinity for binding for unfolded proteins than HSF1. The accumulation of damaged and/or malfolded proteins therefore displaces HSP70 and HSP90 from the HSF1 complex, thereby freeing HSF1 for activation.[28-31] This sequestration of HSP70 and HSP90 releases HSF1 from its inactive monomeric state, allowing translocation to the nucleus, trimerization, hyperphosphorylation and binding to the HSE of the HSP genes. Data also demonstrate that HSF1 is able to directly sense stress (e.g. heat and oxidative stress), form homotrimers and bind to the HSE.[32] In a negative feedback loop mechanism, the increased expression of HSPs will eventually accumulate to a level at which ‘free’ HSPs can re-form the complex with HSF1, thus reverting the latter back to its inactive monomeric form.[33] The exercise-induced activation of HSF1 was first noted in rat hearts[34] and recently observed in human skeletal muscle by data from our laboratory.[35] 1.3 Function of HSPs in the Stressed Cell

An increased cellular content of HSPs following stress functions to restore cellular homeostasis and to provide cytoprotection against further insults.[24] An increased content of HSPs is thought to promote cellular recovery by binding with misfolded and unfolded proteins and facilitating the refolding of these proteins when cellular conditions become more favourable.[24] The stress-induced expression of HSPs therefore acts in a manner analogous to their chaperone function, reflecting the operation of a feedback system that responds to increases in misfolded proteins by elevating the synthesis of the chaperones that help them refold.[36] Direct evidence for protective effects of increased HSP content is provided in those studies employing transgenic approaches, whereby cells and tissues that overexpress HSPs show considerable protection to normally damaging stresses.[37] An increased HSP content following non-damaging stresses (i.e. pre-conditioning) also provides cytoprotection ª 2009 Adis Data Information BV. All rights reserved.

against subsequent periods of normally lethal or damaging stresses.[18-22] It should be noted, however, that although evidence for protective effects of increased muscle HSP content is well documented in animal tissue, comparable findings in humans remain elusive. 2. HSP Families HSPs can be classified into a number of ‘families’ based on their molecular mass. Several HSPs are expressed in skeletal muscle, the most prominent of which include the small HSPs (range 8–27 kDa in size), HSP60 (60 kDa), HSP70 (70 kDa) and HSP90 (90 kDa). A brief overview of the location and specific functions (where known) of these proteins follows. 2.1 Ubiquitin

Ubiquitin is the smallest HSP (8 kDa) that is expressed in human skeletal muscle.[38] This protein is a highly conserved protein that is constitutively expressed in the cytosol. Ubiquitin is considered an HSP because it contains a HSE in the promoter region.[39] Ubiquitin’s primary role is in chromatin structure and in protein degradation.[24] Intracellular proteins that are to be degraded are first covalently modified by the addition of ubiquitin and subsequently targeted for degradation via the ubiquitin-proteasome pathway.[40] The observed increases in ubiquitin levels following damaging contractions of the elbow flexors[38] was therefore likely to facilitate the targeting and removal of damaged and denatured proteins. 2.2 aB-Crystallin

aB-crystallin (22 kDa) belongs to a family of crystallins found in vertebrate lenses. It is also expressed in cardiac and type I and IIa skeletal muscle fibres where it tends to co-localize with HSP27 at the I-band and M-line.[41] aB-crystallin functions as a molecular chaperone to prevent aggregation of denatured proteins or to facilitate refolding upon the removal of stress.[42] Similar to HSP27, aB-crystallin is also involved in stabilization of actin filaments following stressful Sports Med 2009; 39 (8)

Exercise-Induced Stress Response of Skeletal Muscle

insults to the cytoskeleton.[43] It has also been proposed to be involved in regulation of desmin intermediate filaments where it may help stabilize the Z-line.[44] An increased content of desmin and aB-crystallin in the vastus lateralis of humans at 14 days following downhill running suggests these proteins may be involved in remodelling of the Z-disk structures so as to increase resistance to mechanical stresses.[45] Together with HSP27, aB-crystallin is thought to play an important physiological role that functions to protect the cytoskeleton and contractile machinery during stress.[45] 2.3 HSP27

HSP27 is localized in the cytosol in unstressed conditions and is translocated to within or around the nucleus following stress.[46] Data from various cell types have shown that HSP27 is involved in microfilament stabilization,[47] signal transduction,[48] growth,[24] differentiation and transformation processes,[46] and in providing protection against thermal[11] and oxidative stress.[49] HSP27 has been proposed to play a direct role in protecting skeletal muscle from contraction-induced damage via interactions with cytoskeletal elements and in regulation of the glutathione system.[50] HSP27 is increased approximately 3-fold 48 hours following damaging contractions of the elbow flexors.[51-53] Both HSP27 and aB-crystallin translocate to cytoskeletal/myofibrillar proteins immediately following damaging contractions of human vastus lateralis muscle[54] – an effect consistent with observations in mice.[55] This translocation of HSP27 following contraction-induced damage is likely an attempt to limit cytoskeletal disruption and to aid in repair of injured structures. 2.4 HSP60

HSP60 is constitutively expressed in the mitochondrial matrix during normal conditions, where it facilitates the correct folding and assembly of proteins as they enter the mitochondria and in aiding protein transport across intracellular membranes.[56-58] HSP60 is increased in the vastus lateralis of man[59,60] following acute non-damaging ª 2009 Adis Data Information BV. All rights reserved.

647

exercise. Trained subjects also display elevated resting levels of muscle HSP60 content,[61] presumably to protect and/or repair any denatured proteins and also to facilitate increased mitochondrial protein import and assist in mitochondrial biogenesis. 2.5 HSP70

The most highly conserved of HSPs and the most widely studied is the HSP70 family. Four major isoforms of the HSP70 family exist in mammalian cells, the most prominent of which include a cognate isoform (referred to as HSC70 or HSP73) and an inducible isoform (referred to as HSP70 or HSP72). These proteins are involved in interaction with cell signalling pathways,[48] messenger RNA (mRNA) stabilization and degradation,[62] assisting in protein degradation,[63] and as regulators of cell death.[64] The most well documented role of the HSP70 family is as a molecular chaperone and in cytoprotection.[65] The cytoprotective mechanisms of HSP70 are suggested to be similar to its chaperone role where it maintains correct protein folding and translocation, refolds misfolded proteins, prevents protein aggregation and assists in the degradation of unstable proteins.[66] Elevated levels of HSP70 are associated with acquired thermotolerance[66] and with providing cross-tolerance to non-related stressors. This cytoprotective effect is documented in animal studies where muscle HSP70 content has been up-regulated using pre-conditioning stresses[18,19] or transgenic modification.[37] HSP70 is also upregulated in male vastus lateralis muscle following both acute [59,60] and chronic exercise.[67,68] Based on findings from animal tissue, it is therefore thought that increased HSP70 in tissues of an exercising human may permit increased tolerance to the biochemical stresses that accompany exercise.[11] 2.6 HSP90

HSP90 acts in a chaperone-like manner that is involved in the folding and activation of substrate proteins including protein kinases, transcription factors and steroid hormone receptors.[24] The stress-induced expression of HSP90 functions to Sports Med 2009; 39 (8)

Morton et al.

648

prevent aggregation of partially unfolded proteins by maintaining them in a folding competent state for refolding.[69] Similar to HSP70, HSP90 may self-regulate its stress-induced synthesis through its interaction with HSF1.[28] Inhibition of HSP90 function delays and impairs recovery from heat shock, suggesting that a multi-component chaperone complex involving several HSPs is necessary for optimal protection.[70] To the authors’ knowledge, no data are available examining the exercise-induced production of HSP90 in human skeletal muscle. 3. The Effect of Acute Exercise on HSP Content During the 1990s, studies employing treadmill-running rodents provided consistent evidence that acute exercise up-regulates HSP70 content in the heart,[15,16] brain,[17] liver[15] and skeletal muscle.[12,13,15-16] Subsequent studies demonstrated that acute contractile activity also up-regulates aB-crystallin,[41] HSP25,[55] HSP60,[14] HSC70[14] and HSP90[12] in skeletal muscle of various animal species. In the last decade, accumulating data have demonstrated that exercise also up-regulates HSP content of human skeletal muscle. Comparison among studies, however, is difficult due to disparate exercise protocols, differing timing of biopsy sampling, muscle examined, differing subject characteristics both within and between studies, and predominant HSP measured (most notably HSP70). The following sections provide a critical review of exerciserelated HSP data, outlining some of the methodological factors that may influence the response. 3.1 Endurance Exercise: ‘Non-Damaging’ Exercise Protocols

Several investigators have utilized endurance exercise protocols to study the exercise-induced stress response of human skeletal muscle. These protocols can typically be referred to as ‘nondamaging’ in that they induce no overt structural (as indicated by circulating creatine kinase levels) or functional (as indicated by maximal voluntary muscle force) damage to the muscle. It is acª 2009 Adis Data Information BV. All rights reserved.

knowledged, however, that such exercise stresses may result in a transient and reversible oxidation of muscle proteins that may possibly be acting as the exercise ‘signal’ that initiates the stress response (this concept is discussed further in section 5.2). Data regarding the exercise-induced stress response of human muscle were first provided by Puntschart et al.[71] These authors demonstrated that although HSP70 mRNA increased 4-fold in the vastus lateralis muscle immediately after 30 minutes of treadmill running at the anaerobic threshold, HSP70 protein levels did not change within 3 hours after cessation of exercise. The authors therefore suggested there may be a time delay between transcription and translation and that 3 hours post-exercise is not sufficient for newly synthesized protein to be detected. Similarly, Walsh et al.[72] observed 6-fold significant increases in HSP72 mRNA at 2 hours following 1 hour of running, yet HSP72 protein content did not significantly change at 2, 8 or 24 hours postexercise. Consistent with these observations, cycling protocols also induced increases in HSP70 mRNA immediately following exercise[73] or 2 hours post-exercise[74] with no accompanying increase in protein. Evidence from rodent muscle suggests that the exercise protocols used in the aforementioned studies may not have been sufficient in terms of exercise intensity to initiate an overall increase in HSP content.[75] However, the above protocols were sufficient to induce an increase in transcription, suggesting the appropriate signalling pathway was activated. Alternatively, it is possible that the timing of biopsy sampling may not have been appropriate to detect any newly synthesized HSPs. In order to address this hypothesis, Khassaf et al.[59] performed a time-course study where biopsies were obtained from the vastus lateralis for up to 6 days following onelegged cycling. Both HSP60 and HSP70 content of sedentary subjects tended to increase 24 hours post-exercise, although these increases only became significant at 3 and 6 days into the recovery period, respectively. The authors also observed that this response displays individual variation in terms of both time-course and magnitude, which Sports Med 2009; 39 (8)

Exercise-Induced Stress Response of Skeletal Muscle

appeared related to baseline HSP levels. Typically, individuals with lower levels of HSPs responded to exercise with a faster (i.e. 24–48 hours post-exercise) and larger induction, whereas subjects with higher baseline levels exhibited a slower (i.e. 72 hours to 6 days post-exercise) and smaller response. Using a running exercise protocol, Morton et al.[60] subsequently confirmed this timecourse where peak responses of HSP70 in the vastus lateralis of recreationally active subjects typically occurred at 48 hours post-exercise. Consistent with the findings from Khassaf and colleagues, we also observed similar individual variability, which again appeared related to baseline HSP levels. Interestingly, the running exercise protocol did not induce up-regulation of aB-crystallin or HSP27, suggesting that specific HSPs are differentially expressed according to characteristics of the exercise stress. 3.2 Resistance Exercise and Down-Hill Running: ‘Damaging’ Exercise Protocols

In contrast to ‘non-damaging’ protocols, some researchers have also used resistance exercise and down-hill running protocols to study the stress response of human muscle. These exercise protocols can be defined as ‘damaging’ in that they induce overt structural (as indicated by significantly increased circulating creatine kinase levels) and functional (as indicated by significantly reduced maximal voluntary force) damage to the muscle. Thompson et al.[51-53] investigated the HSP response to acute bouts of resistance exercise that are damaging in nature. In the first of these studies, HSP27 and HSP70 increased 2- and 10-fold, respectively, in the biceps brachii of untrained subjects 48 hours after lengthening contractions. The authors subsequently examined the role of HSPs in the ‘repeated bout effect’ in which the previous resistance exercise protocol was performed on two occasions, separated by 4 weeks.[53] The relative magnitude of the increase in HSP27 and HSP70 response was the same after both bouts, although the baseline levels of both proteins in the previously exercised muscle were lower before bout 2. Although the rates of ª 2009 Adis Data Information BV. All rights reserved.

649

synthesis of HSP27 and HSP70 were similar after both bouts, the absolute increase in protein levels was lower after bout 2. These data may represent an adaptive mechanism of the stress response whereby baseline levels of stress proteins may be down-regulated following repeated bouts of exercise yet the magnitude of HSP expression following acute bouts of exercise is maintained or increased. Similar findings have been observed in skeletal muscle of trained rodents[76] and leukocytes of trained humans[77] following acute exercise and an in vitro heat stress, respectively. Data are also available examining the stress response following downhill running protocols,[45,52] which are also damaging in nature due to their bias towards lengthening contractions. HSP27 and aB-crystallin increased approximately 3-fold in the vastus lateralis muscle at 24 hours post-exercise and remained elevated for 14 days.[45] In contrast, no increases in HSP27 or HSP70 have been observed 48 hours following downhill running.[52] It is difficult to explain the discrepancies between studies, although the subjects in the latter report were predominantly female and thus may provide evidence of a sexspecific response to exercise (see section 3.4.4. for further discussion). Interpretation of data from damaging exercise protocols is complicated by the inflammatory response that occurs several days following contractions. It is therefore possible that the reported increases in HSP content in these studies are also due to the increased presence of phagocytic cells, given that such cells contain relatively high levels of HSPs.[78] The invasion of phagocytic cells also results in an oxidative burst, thus contributing to secondary muscle injury that may further augment the intramuscular expression of HSPs. Analysis of homogenates from muscles that have undergone damaging protocols therefore makes it difficult to specifically quantify those HSPs induced by ‘muscle’ as opposed to those due to changes in phagocytic cell content and secondary processes.[79] Non-damaging exercise protocols such as those described in section 3.1 are therefore suggested to provide a ‘cleaner’ methodological approach to study the exercise-induced regulation of HSPs.[79] Sports Med 2009; 39 (8)

Morton et al.

650

3.3 Time-Course and Muscle Fibre Specificity of the Response

Whereas HSP mRNA are increased during,[73] immediately after[71] or several hours post-exercise,[72] studies employing serial biopsies and western blotting techniques have consistently demonstrated increases in protein levels are only detectable at 1–14 days post-exercise.[45,51-53,59-60,72,80-81] Recent data, however, have provided new evidence on the time-course of the response. Using an isometric protocol and immunohistochemistry techniques, Tupling et al.[81] observed an 80% increase in HSP70 content of the vastus lateralis immediately post-exercise that was specific to type I fibres only. In addition to a novel time-course, this study thus also provided evidence for a muscle fibre-specific expression of HSPs. However, the time-course and fibre specificity of the stress response may be unique to particular HSPs, to the characteristics of the chosen exercise protocol (e.g. intensity, duration, mode, damaging/non-damaging, fibre recruitment) and to the population under investigation. Clearly, further work is needed using a variety of analytical techniques, experimental designs and subject populations before the time-course of synthesis and degradation of HSPs following exercise can be accurately defined. 3.4 Individual Variation of the Response

The stress response of human muscle displays high individual variation (in terms of both magnitude and time-course), which appears to be due, in part, to individual differences in baseline HSP content.[59,60,81] In considering factors that determine baseline HSP levels and/or the extent of the heat shock response to various stresses, data from various cell types have indicated that training status,[76] recent activity levels,[82] thermal history,[66] energy availability,[83] sex[84] and age[85] are all possible determinants. The roles of the above factors in influencing the magnitude of the exercise-induced stress response are discussed in the following sections. An understanding of these factors has important implications both for research design and for prescribing exercise protocols that are intended to up-regulate HSP ª 2009 Adis Data Information BV. All rights reserved.

content with the potential aim of providing cytoprotection. 3.4.1 Training Status

Skeletal muscle from trained rodents[86] and humans[60,87] exhibits a diminished stress response to customary exercise, likely because adaptations occurring during training function to reduce the degree of homeostatic imbalance that occurs during an acute stress. Alternatively, data from trained rodent muscle[76] and leukocytes from trained humans[77] suggest that trained individuals may respond to an acute exercise stress with a faster and larger production of HSPs, thereby helping the cell to recover more quickly. Although uncertainty in this area remains (largely due to variations in cell type studied and intensity of exercise protocol), it is evident that subtle differences in training status between subjects may contribute to the individual variation observed previously, both within and between studies. The impact of training status on baseline HSP levels and also on the magnitude of the stress response to acute exercise is discussed in more detail in section 4. 3.4.2 Recent Activity Levels/Thermal History

The cellular content of HSPs at any given time may simply be a reflection of the previous exposure to stress, e.g. exercise or hyperthermia. Indeed, in human skeletal muscle, HSPs can remain elevated for 6,[59,81] 7[60] or 14 days following acute exercise.[45] Heat-acclimatized humans also display higher baseline levels of HSPs in skin fibroblasts.[88] Cells demonstrating elevated levels of HSPs following recent stresses may therefore not need to mount a further stress response as they may already be preconditioned to such challenges. Variations in the baseline levels of HSPs and in the magnitude of the exercise-induced production of HSPs may therefore be related to the timing of previous exposures to stress. Non-heat-acclimatized and fully ‘rested’ individuals should therefore be utilized in all investigations. However, precisely how long is required to alleviate the acute expression of HSPs is not well defined. Sports Med 2009; 39 (8)

Exercise-Induced Stress Response of Skeletal Muscle

3.4.3 Energy Availability

Reduced glucose availability activates the stress response in vitro,[10] and a reduced carbohydrate availability has been suggested to be a contributing factor to the exercise-induced production of HSPs in human skeletal muscle.[73,83] Although it is presently unclear how carbohydrate availability may regulate HSP expression, it is nevertheless apparent that intersubject variations in endogenous and exogenous substrate availability may contribute to variable stress responses. Further discussion on the role of carbohydrate in activating the exercise-induced stress response is provided in section 5.3. 3.4.4 Sex

Female rodents[89] and female humans[90] appear to experience exercise-induced muscle damage to a lesser degree compared with male counterparts. It may therefore be hypothesized that females should display a decreased intracellular accumulation of denatured proteins compared with males; it thus follows that females should exhibit an attenuated HSP response to exercise.[91] In keeping with this hypothesis, an attenuated HSP70 response of rodent heart, lung, liver and skeletal muscle was observed in females compared with males.[91] The same workers demonstrated that removal of ovaries from female rodents resulted in muscle HSP70 induction similar to that of male rodents, and further showed that oestrogen treatment given to ovariectomized animals reversed this effect.[84] A sex-specific response of human skeletal muscle to exercise may therefore explain some of the discrepancies between studies reviewed previously, although this remains to be formally tested following an acute exercise stress. Nevertheless, preliminary data from our group demonstrate that human female muscle shows a lack of HSP adaptation to short-term exercise training (Morton et al.[92]). 3.4.5 Age

Tissues from aged animals[93] and lymphocytes[94] and monocytes[95] from elderly humans show a reduced production of HSPs following heat stress. Following exercise, the myocardium[96] and skeletal muscle[85] from aged ª 2009 Adis Data Information BV. All rights reserved.

651

rodents also exhibit a diminished HSP response compared with younger animals. The precise mechanisms underlying the attenuated stress response of aged skeletal muscle following contractile activity is an active area of research,[79] and may be related to transcriptional defects. Similar to sex, the influence of aging on the exercise-induced stress response of human skeletal muscle has not yet been investigated. 4. The Effects of Exercise Training on HSP Content Whereas increased production of HSPs following acute exercise is thought to regain cellular homeostasis and repair damage to denatured proteins,[97-101] increases in HSP content occurring with repetitive exercise are suggested to represent adaptations that serve to maintain homeostatic balance during a given stress.[102] Up-regulation of baseline HSP content during chronic exercise may therefore be a crucial component of the cellular and molecular mechanisms by which regular exercise confers protection against related and nonrelated stressors.[11] As molecular chaperones, increased baseline HSP levels in trained muscle are also important in modulating cell signalling pathways and facilitating cellular remodelling processes that are inherent to the training response, such as mitochondrial biogenesis,[103] protein turnover,[104] muscle hypertrophy/fibre transition[100,105,106] and regulation of atrophic[107,108] and apoptotic pathways.[109] A wealth of data indicate that chronic contractile activity induces an increase in baseline HSP levels in cardiac[110-114] and skeletal muscle[115-118] of rodents. In contrast to animal tissue, data from human muscle are limited concerning the influence of training status on both baseline HSP levels and the ability to produce HSPs following an acute stress exposure. 4.1 Baseline HSP Content

Liu et al.[67] observed that the largest traininginduced increase in HSP70 content of the vastus lateralis of elite male rowers occurred at the end of a distinct training phase where the maximum amount of exercise had been completed. Sports Med 2009; 39 (8)

652

With the exercise volume decreasing during further training periods, HSP70 levels declined towards resting values. These data suggest that the HSP response to training is related to the total amount of training undertaken, although at this stage the authors could not comment on whether such a response was dependent more on intensity or volume of exercise. In a subsequent study also using elite male rowers, the researchers demonstrated that the highest levels of HSP70 also occurred following training periods where the highest intensities of training had been undertaken.[68] This finding is physiologically reasonable given that high-intensity exercise is associated with greater thermal, metabolic and oxidative stress, all of which are known activators of the stress response. Collectively, these data suggest that training-induced increases in HSP levels are only maintained if the training stimulus is held constant or is increased. When the stimulus is reduced, HSP levels appear to return to pretraining levels. These data are significant in that HSP-mediated protection is also likely to return to pre-training values. It is therefore possible that a down-regulation of HSP content upon the cessation of an exercise training programme (or in a period of de-training) may be one of multiple mechanisms by which the repeated bout effect can be lost. The precise dose of exercise that is needed to consistently maintain elevated muscle HSP content throughout training is currently not known. Although the above studies provide insightful data regarding HSP responses to training, the subjects used in these investigations were already of elite nature and thus it is possible that the shortterm training interventions were insufficient to induce further ‘gross’ changes in muscle HSP levels. As such, the pre-training intervention levels of HSPs in these subjects may have already been appropriate to counteract the stress of regular training. It is therefore possible that the HSP levels of these subjects are somewhat higher than those of less conditioned or sedentary populations. In order to address this hypothesis, Morton et al.[61] employed a cross-sectional design and obtained muscle biopsies from the vastus lateralis of trained and untrained male subjects. A significant increase in aB-crystallin, HSP60 and MnSOD (manganese ª 2009 Adis Data Information BV. All rights reserved.

Morton et al.

superoxide dismutase) content and a tendency for increased levels of HSP70 and HSC70 was evident in trained muscle compared with untrained subjects, thus demonstrating that exercise training is associated with an up-regulation of baseline HSP and antioxidant defences. 4.2 Magnitude of the Stress Response

Trained rodents display a blunted HSP70 production in soleus muscle compared with sedentary rodents following acute customary exercise,[86] a finding attributed to a training-induced increase in baseline antioxidant systems. It was therefore suggested that HSPs may have a secondary effect on modulation of antioxidant defences, providing additional protection when the primary system is overwhelmed.[86] A preliminary study from humans also demonstrated that trained rowers did not show any increase in muscle HSP70 gene expression within 6 hours following high-intensity resistance exercise or low-intensity endurance exercise.[87] Unfortunately, a less conditioned control group was not studied. More recently, we have shown that the exercise-induced induction of HSP60, HSC70 and HSP70 previously observed in vastus lateralis muscle of recreationally active subjects[60] is abolished when trained subjects undertake the experimental protocol.[61] Collectively, these findings suggest that trained subjects display an attenuated or abolished stress response to customary exercise, likely due to an increased baseline level of endogenous defences such as HSPs and antioxidant defence proteins. We therefore suggest that the exercise protocol must present a novel homeostatic disruption and exceed a ‘critical threshold’ exercise intensity needed to overwhelm baseline defence systems in order for a stress response to occur (see figure 1). 5. Possible Physiological Signals Initiating the Exercise-Induced Stress Response The stress response is activated both in vitro and in vivo following various stressors, including thermal, oxidative, mechanical, metabolic and cytokine production. These stressors are similar to the homeostatic perturbations occurring in Sports Med 2009; 39 (8)

Exercise-Induced Stress Response of Skeletal Muscle

Individual characteristics 1. Training status (i.e. baseline HSP levels/ antioxidants, resting glycogen content/ utilization during exercise, etc.) 2. Age 3. Sex

Necessary fibre recruitment

Absolute workload

Novelty of exercise stress

Critical threshold

↑ HSPs Fig. 1. The critical threshold hypothesis and possible factors contributing to the attainment of a ‘critical threshold’ that is necessary for induction of heat shock proteins (HSPs) following an acute exercise stress. The model suggests that the individuals’ characteristics (in particular, that of training status and the accompanying baseline levels of HSPs and antioxidant defences) will determine the absolute workload that is necessary to attain the critical threshold. This workload should be sufficient to activate a significant proportion of total muscle fibres (so as to present a ‘whole’ muscle stress) and must also be relatively novel so as to overwhelm baseline defence systems. › indicates increased.

contracting skeletal muscle; thus, it is difficult to isolate the precise signal(s) that is responsible for initiating the exercise-induced stress response (see figure 2). In the present section, data concerning the role of the above factors in contributing to the exercise-induced stress response are discussed. 5.1 Temperature

Supporting evidence for the role of temperature in activating the exercise-induced stress response is provided from a number of species/ cell types and experimental approaches. For example, both in vitro[21] and in vivo[119,120] heating protocols that increased cell temperature to 42C induced increases in HSP70 content of ª 2009 Adis Data Information BV. All rights reserved.

653

C2C12 skeletal muscle myotubes and rodent muscle, respectively. The exercise-induced increase of HSP70 in rodent myocardium and soleus muscle following exhaustive treadmill running is significantly enhanced when exercise is performed under elevated ambient temperatures.[121] Elevations in cellular temperature per se induce increases in reactive oxygen species[122] – also thought to be a potent activator of the exercise-induced stress response. In contrast, numerous data suggest that thermal stress is not the sole signal responsible for the exercise-induced expression of HSPs. Skidmore et al.[16] observed increases in HSP70 content in rodent skeletal and cardiac muscle 30 minutes after 1 hour of treadmill running that were independent of an increase in core temperature. Electrical stimulation of mouse[14] and rabbit[117] muscle induced increases in muscle HSP content in the absence of an increased muscle temperature. The magnitude of HSP increases observed in animal studies following treadmill running[12,13,15-16] or electrical stimulation protocols[14,117] is also somewhat larger than that observed following in vivo heating protocols,[119,120] thus suggesting that additional signals arising during exercise are also contributing to the heat shock response. In order to address this hypothesis in human tissue, Morton et al.[123] passively heated the vastus lateralis muscle to comparable temperatures to that observed during exercise and observed no significant increases in HSP70 content of the vastus lateralis muscle at 2 or 7 days postheating (figure 3). Furthermore, Palermo and colleagues[35] subsequently demonstrated that the exercise-induced activation of HSF1 is not enhanced when the exercise is performed under conditions of elevated ambient temperature, despite significantly increased muscle and core temperatures in this condition. Although findings from both animal and human studies demonstrate that heat stress per se is not the primary activator of the exercise-induced stress response, it may be slightly premature to conclusively dismiss the role of temperature. For example, it is possible that a smaller increase in cell temperature than that in the above cited studies may have a compounded effect on another exercise Sports Med 2009; 39 (8)

Morton et al.

654

Response

Time-course

HSF1 activation

During

Exercise

↑ Tmus?

↓ Glycogen?

↑ ROS?

↑ Mechanical damage?

HSP mRNA

Hours

HSPs

Days

HSPs

1–2 weeks

Inflammatory processes?

Fig. 2. The potential stressors arising during exercise that may be signalling the stress response and a summary of the time-course of the response in human muscle. Depending on the nature of the exercise protocol (i.e. intensity, duration, damaging/non-damaging), an increase in muscle temperature (Tmus), increased reactive oxygen species (ROS), glycogen depletion, and mechanical damage to muscle proteins may signal the response independently or in combination with each other. Heat shock factor-1 (HSF1) is activated during exercise, thereby causing increased heat shock protein (HSP) transcription where elevated levels of transcript persist for up to several hours post-exercise. Increased HSP content is typically detected in the days following exercise. If the protocol has been damaging in nature, this increase may be exacerbated by an inflammatory response occurring several days following the exercise stress and is likely to be prolonged due to the damaging and regeneration processes. › indicates increased; fl indicates decreased.

stressor(s), depending on the nature of activation and transcription of HSF1. 5.2 Oxidative Stress

In contrast to the traditionally held view that reactive oxygen species (ROS) are ‘injurious molecules’, exercise-induced increases in ROS are now accepted as important cellular messengers in signal transduction involved in activation of transcription factors such as nuclear factor (NF)-kB AP1 (activator protein 1) and HSF1.[124] McArdle et al.[14] demonstrated that 15 minutes of non-damaging isometric contractions of the hindlimbs of mice results in increased release of superoxide anions from within the muscle ª 2009 Adis Data Information BV. All rights reserved.

into the extracellular space and an accompanying transient and reversible oxidation of muscle protein thiols. The contraction protocol subsequently resulted in increased HSP70 content of the soleus and extensor digitorum longus muscles. These findings appear consistent with data from other cell types,[125,126] in that oxidation of muscle protein thiols may be part of a signalling mechanism leading to increased HSP production. These data suggest that subtle moments of redox imbalance may act as a signal for HSP adaptation either by direct interaction with HSF1[32] or by causing some minor oxidative damage to proteins subsequently leading to an up-regulation of HSPs. Data from McArdle and colleagues also demonstrate that the increased HSP70 content Sports Med 2009; 39 (8)

Exercise-Induced Stress Response of Skeletal Muscle

observed in human vastus lateralis muscle following one-legged cycling[59] was abolished following vitamin C,[78] vitamin E[80] and b-carotene supplementation.[80] It should be noted, however, that antioxidant supplementation in these studies induced a significant increase in baseline muscle HSP70 content, suggesting that an attenuation of the stress response may be related to increased baseline protective systems. Alternatively, it may be that an elevation of tissue antioxidant capacity directly scavenges exercise-generated ROS, thereby abolishing transcriptional activity of the HSP genes. Experimental support for this hypothesis was provided by Fischer et al.,[127] where the combination of vitamin C and E supplementation inhibited increases in muscle HSP72 mRNA expression immediately following 3 hours of knee extensor exercise. Despite the supporting evidence for a role of ROS in activating the stress response, further work is needed to clarify the precise source of ROS production and site of action. 5.3 Metabolic Stress

In cell culture studies, reduced glucose availability,[10] adenosine triphosphate (ATP) depletion,[128] acidosis[7] and increased intracellular calcium levels[9] all induce increased cellular con250

Passive heating Exercise

HSP70 (% of resting content)

* 200

*

150

100

50

0 Rest

2 days

7 days

Time Fig. 3. The effects of passive heating and exercise on heat shock protein-70 (HSP70) content of the vastus lateralis muscle at 2 and 7 days following each intervention. * denotes significant difference from resting values (p < 0.05). Figure compiled from data obtained in Morton et al.[60,123]

ª 2009 Adis Data Information BV. All rights reserved.

655

tent of HSPs. Exercise-related metabolic stress has therefore been suggested to be involved in initiating the exercise-induced stress response.[129] In myogenic cultured cells, ATP depletion to 30% of control levels and a fall in pH from 7.3 to 6.9 was sufficient to induce HSF1 activation.[128] Reductions in pH to 6.7 in the face of maintained ATP levels failed to activate HSF1,[128] suggesting that ATP depletion, independent of acidosis, may be an important metabolic pathway for the expression of HSPs during exercise. However, it is difficult to ascertain the contribution of ATP depletion per se during whole body exercise in humans because ATP levels are not depleted to such levels. Availability of carbohydrate for the contracting musculature may also be an important contributing stressor to the exercise-induced expression of HSPs. The increase in HSP72 gene expression occurring during exhaustive cycling coincided with the time-point at which muscle glycogen levels were reduced to low levels.[73] Using a two-legged exhaustive knee extensor protocol, Febbraio et al.[83] also observed a 2-fold increase in HSP72 gene expression immediately post-exercise only in the leg that performed a glycogen-depleting protocol 24 hours prior to the exercise bout. Furthermore, Tupling et al.[81] also observed that peak HSP70 induction following isometric contraction was highest and the time-course more rapid in those muscle fibres that exhibited largest glycogen depletion during exercise. Whilst these data indicate a potential role of carbohydrate availability as a contributing factor to the exercise-induced expression of HSPs, it is presently difficult to offer precise mechanisms underpinning these findings. Furthermore, we have recently demonstrated that the training-induced increase in HSP70, HSP60 and aB-crystallin in the vastus lateralis and gastrocnemius muscles occurs independently of both endogenous and exogenous carbohydrate availability.[130] 5.4 Cytokine Production and Inflammatory Stress

It is well documented that interleukin (IL)-6 is produced locally in contracting skeletal muscle.[131] In addition to its hormonal role regulating substrate metabolism, it is also suggested Sports Med 2009; 39 (8)

Morton et al.

656

that muscle-derived IL-6 may act as a signalling molecule involved in the regulation of the cellular stress response.[132] Febbraio et al.[132] demonstrated that the infusion of IL-6 into resting human skeletal muscle induced a significant increase in HSP72 gene expression in vivo. These findings are based, however, on resting skeletal muscle and hence are not representative of contracting skeletal muscle where a further array of alterations in cellular homeostasis is occurring. Furthermore, experimental evidence against a role of IL-6 was provided by an antioxidant supplementation study in which the increase in HSP72 gene expression in the vastus lateralis immediately following 3 hours of two-legged knee extensor exercise was abolished following vitamin C and E supplementation, whilst having no effect on IL-6 gene expression.[133] Cytokine production, at least that of IL-6, is therefore unlikely to be the dominant signal for activation of the exercise-induced stress response, particularly during non-damaging aerobic type activities. However, the invasion of monocytes into the muscle and the associated production of cytokines may be responsible, in conjunction with mechanical damage to proteins, for the exerciseinduced expression of HSPs observed following damaging exercise protocols.[51-53] In these circumstances, the controlled and deliberate generation of ROS by phagocytic cells may also contribute to the increased intramuscular expression of HSPs. 6. Biological Significance of Exercise-Induced Increases in Muscle HSP Content The cytoprotective properties of increased HSP content were initially observed in studies demonstrating that cells exposed to sub-lethal heat shock were subsequently protected against normally lethal levels of heat stress.[134-136] This concept of acquired thermotolerance led to the suggestion of cross-tolerance in which cells that display elevated levels of HSPs following stressor A (which in most cases has been heat stress) are subsequently protected against a non-related stressor B.[137] Cross-tolerance has important ª 2009 Adis Data Information BV. All rights reserved.

implications for exercise whereby the exerciseinduced production of HSPs may be an important mechanism by which exercise can provide protection to cells, tissues and organs against a variety of stressors and diseases.[23] Unfortunately, conclusive evidence for protective effects of increased muscle HSP content is only available in animal tissue, and comparable findings in humans remain elusive. Nevertheless, relevant findings from animal tissue that have translational potential for human research deserve a brief review. Prior heat stress and the associated increased muscle content of HSP70 provided significant protection to rodent skeletal muscle against necrosis induced by ischaemia-reperfusion.[18,19] Maglara et al.[21] also demonstrated that damage to C2C12 mouse skeletal muscle myotubes induced by either the calcium inophore A23187 or the mitochondrial uncoupler 2,4-dinitrophenol was significantly reduced by a prior period of hyperthermia, which induced significant increases in HSP25, HSC70 and HSP70 content. It was therefore suggested that an increased cellular content of HSPs may provide protection against the muscle damage that occurs by a pathological increase in intracellular calcium or uncoupling of the mitochondrial respiratory chain (i.e. exercise). Further data from our group also demonstrated that skeletal muscle of adult and aged transgenic mice overexpressing HSP70 displays enhanced recovery from damage induced by lengthening contractions compared with wild type animals.[37] Subsequent data revealed that HSP70 overexpression is associated with reduced accumulation of oxidation products in muscle of aged mice and the maintenance of NF-kB activation following non-damaging contractions.[138] These findings suggest that interventions that lead to lifelong elevations of muscle HSP content may help preserve muscle function and the ability to adapt and re-generate following stressful episodes occurring during aging. Unfortunately, Kayani et al.[139] could not re-create the protective effects of increased muscle HSP70 content when achieved via the physiological stress of lifelong exercise, likely due to differences in absolute muscle content of HSP70 in trained versus transgenic animals (2-fold increases vs Sports Med 2009; 39 (8)

Exercise-Induced Stress Response of Skeletal Muscle

10-fold increase). However, the severity of the contraction protocol may have masked any protective effects of training-induced HSP levels; it is also not representative of the challenges that face elderly humans. Clearly, translational research in adult and aged humans is needed and a cause-and-effect relationship established by absolute muscle content of HSPs and any protection offered against physiologically relevant stresses. HSP72 gene[140,141] and protein expression[142] are reduced in patients with type 2 diabetes mellitus and this is inversely correlated with insulin sensitivity. Studies in mice also demonstrated that elevated HSP72 content of skeletal muscle (achieved via heat shock, transgenic approaches or pharmacological interventions) provides significant protection against diet- or obesityinduced hyperglycaemia, hyperinsulinaemia, glucose intolerance and insulin resistance.[142] This protection was associated with HSP72-mediated blocking of inflammatory pathways. Similar to the ‘aging’ studies cited above, translational research in humans is therefore needed in order to investigate whether exercise-induced increases in muscle HSP content have similar regulatory effects. However, regardless of the clinical setting, a clear exercise intensity and dose-response relationship needs to be established in order to administer more precise exercise interventions needed to induce optimal levels of muscle HSP content. 7. Conclusions and Future Directions The addition of exercise to the list of known inducers of HSPs opened a novel field of investigation that remains an active area of research. Available data now demonstrate that various forms of acute and chronic exercise up-regulate several HSPs in human skeletal muscle. Despite these recent advances, the exercise-induced stress response is far from clearly characterized and understood. This is due to variations in timing of biopsy sampling between studies, restrictions of analysis to one or two HSPs, differing analytical techniques and differing subject characteristics both within and between studies. Future reª 2009 Adis Data Information BV. All rights reserved.

657

searchers should focus their efforts on obtaining tissue from carefully chosen time points (with respect to their chosen exercise protocol), examining the response of several HSP families (both magnitude of increase and subcellular location), and utilizing subjects from relatively homogeneous and diverse populations for both within- and between-study comparisons, respectively. Exercise training up-regulates baseline HSP levels in a complex manner that is dependent on a sustained and currently unknown dose of exercise and also on the individual’s initial training status. The magnitude of the stress response following acute exercise in trained muscle is also not well defined, although initial data suggest that trained muscle exhibits a blunted production of HSPs to customary exercise, likely due to simultaneous increases of baseline HSP levels and antioxidant defences. Given the array of ‘cross-talk’ that occurs between signalling pathways in skeletal muscle, it is difficult to conclude that exercise induces HSP expression through a unique ‘stress’. The extent to which each potential stressor contributes to the exercise-induced induction of HSPs is also likely dependent on the mode, intensity, duration and contractile nature of the exercise stimulus. Following resistance or damaging exercise, HSP expression is likely initially initiated by mechanically induced damage to muscle proteins and further augmented by the inflammatory response occurring several days into the recovery period. Following endurance and aerobic type activities, expression of HSPs is likely mediated by redox signalling where heatinduced radicals may also contribute to overall cellular oxidant activity. Whilst further fundamental work is needed to accurately characterize the stress response of specific populations following varying exercise protocols, exercise scientists should also focus their efforts on elucidating the precise biological significance of the exercise-induced induction of HSPs. In addition to their potential cytoprotective properties, the role of HSPs in modulating cell signalling pathways related to both exercise adaptation and health and disease also needs further investigation. As a non-pharmacological Sports Med 2009; 39 (8)

Morton et al.

658

intervention, exercise and the associated upregulation of HSPs (and hence the possible correction of maladapted signalling pathways) may therefore prove effective in providing protection against protein misfolding diseases and in preserving muscle function during aging. Acknowledgements No sources of funding were used to assist in the preparation of this review. The authors have no conflicts of interest that are directly relevant to the content of this review.

References 1. Kilgore JL, Musch TI, Ross CR. Physical activity, muscle and the HSP70 response. Can J Appl Physiol 1998; 23: 245-60 2. Ritossa F. A new puffing pattern induced by temperature shock and DNP in Drosophila. Experimenta 1962; 13: 571-3 3. Tissieres A, Mitchell HK, Tracy UM. Protein synthesis in salivary glands of Drosophila melanogaster: relation to chromosome puffs. J Mol Biol 1974; 84: 389-98 4. Marber MS, Meistrel R, Chi SR, et al. Overexpression of the rat inducible 70 k-Da heat stress protein in a transgenic mouse increases the resistance of the heart to ischemic injury. J Clin Invest 1995; 95: 1446-56 5. Chiang HL, Terlecky SR, Plant CP, et al. A role for a 70kilodalton heat shock protein in lyosomal degradation of intracellular proteins. Science 1989; 246: 382-5 6. Guttman SD, Glover CVC, Allis CD, et al. Heat shock, deciliation and release from anoxia induce the synthesis of the same polypeptides in starved T. pyriformis. Cell 1980; 22: 299-307 7. Weitzel G, Pilatus U, Rensing L. Similar dose response of heat shock protein synthesis and intracellular pH change in yeast. Exp Cell Res 1985; 159: 252-6 8. Adrie C, Ritcher C, Bachelet M. Contrasting effects of NO and peroxynitrites on HSP70 expression and apoptosis in human monocyctes. Am J Physiol 2000; 279: C452-60 9. Welch WJ, Garrells JI, Thomas GP, et al. Biochemical characterization of the mammalian stress proteins and identification of two stress proteins as glucose and Ca2+ ionophore regulated proteins. J Biol Chem 1983; 258: 7102-11 10. Sciandra JJ, Subjeck JR. The effect of glucose on protein synthesis and thermosensitivity in Chinese hamster ovary cells. J Biol Chem 1983; 258: 12091-3 11. Locke M. The cellular stress response to exercise: role of stress proteins. Exerc Sports Sci Rev 1997; 25: 105-36 12. Locke M, Noble EG, Atkinson BG. Exercising mammals synthesise stress proteins. Am J Physiol 1990; 258: C723-9 13. Hernando R, Manso R. Muscle fibre stress in response to exercise: synthesis, accumulation and isoform transactions of 70 kDa heat shock proteins. Eur J Biochem 1997; 243: 460-7

ª 2009 Adis Data Information BV. All rights reserved.

14. McArdle A, Pattwell D, Vasilaki A, et al. Contractile activity-induced oxidative stress: cellular origin and adaptive responses. Am J Physiol 2001; 280: C621-7 15. Salo DC, Donovan CM, Davies KJA. HSP70 and other possible heat shock or oxidative stress proteins are induced in skeletal muscle, heart and liver during exercise. Free Radic Biol Med 1991; 11: 239-46 16. Skidmore R, Gutierrez JA, Guerrio V, et al. HSP70 induction during exercise and heat stress in rats: role of internal temperature. Am J Physiol 1995; 268: R92-7 17. Walters G, Ryan KL, Tehrany MR, et al. HSP70 expression in the CNS in response to exercise and heat stress in rats. J Appl Physiol 1998; 84: 1269-77 18. Garramone RR, Winters RM, Das DK, et al. Reduction of skeletal muscle injury through stress conditioning using the heat shock response. Plast Reconstr Surg 1994; 93: 1242-9 19. Lepore DA, Hurley JV, Stewart AG, et al. Prior heat stress improves survival of ischemic-reperfused skeletal muscle in vivo. Muscle Nerve 2000; 23: 1847-54 20. McArdle F, Spiers S, Aldemir H, et al. Preconditioning of skeletal muscle against contraction-induced damage: the role of adaptations to oxidants in mice. J Physiol 2004; 561: 233-44 21. Maglara AA, Vasilaki A, Jackson MJ, et al. Damage to developing mouse skeletal muscle myotubes in culture: protective role of heat shock proteins. J Physiol 2003; 548: 837-46 22. Suzuki K, Smolenski RT, Jayakumar J, et al. Heat shock treatment enhances graft cell survival in skeletal myoblast transplantation to the heart. Circulation 2000; 102 Suppl. 3: III216-21 23. Locke M. Overview of the stress response. In: Locke M, Noble EG, editors. Exercise and stress response: the role of stress proteins. Boca Raton (FL): CRC Press LLC, 2002: 1-12 24. Welch WJ. Mammalian stress response: cell physiology, structure/function of stress proteins, and implications for medicine and disease. Physiol Rev 1992; 72: 1063-81 25. Lancaster GI, Febbraio MA. Mechanisms of stressinduced cellular HSP72 release: implications for exerciseinduced increases in extracellular HSP72. Exerc Immunol Rev 2005; 11: 46-52 26. Fehrenbach E, Niess AM. Role of heat shock proteins in the exercise response. Exerc Immunol Rev 1999; 5: 57-77 27. Fehrenbach E, Northoff H. Free radicals, exercise, apoptosis and heat shock proteins. Exerc Immunol Rev 2001; 7: 66-89 28. Zuo J, Guo Y, Guettouche T, et al. Repression of heat shock transcription factor HSF1 activation by HSP90 (HSP90 complex) that forms a stress-sensitive complex with HSF1. Cell 1998; 94: 471-8 29. Baler R, Welch WJ, Vollemy R. Heat shock gene regulation by nascent polypeptides and denatured proteins: HSP70 as a potential autoregulatory factor. J Cell Biol 1992; 117: 1151-9 30. Baler R, Zou J, Vollemy R. Evidence for a role of HSP70 in the regulation of the heat shock response. Cell Stress Chaperone 1996; 1: 33-9 31. Morimoto RI. Cells in stress: transcriptional activation of heat shock genes. Science 1993; 259: 1409-10

Sports Med 2009; 39 (8)

Exercise-Induced Stress Response of Skeletal Muscle

32. Ahn SG, Thiel D. Redox regulation of mammalian heat shock factor 1 is essential for Hsp gene activation and protection from stress. Gene Dev 2003; 17: 516-28 33. Ho JSL, Westwood JT. Transcriptional regulation of the mammalian heat shock genes. In: Locke M, Noble EG, editors. Exercise and stress response: the role of stress proteins. Boca Raton (FL): CRC Press LLC, 2002: 12-78 34. Locke M, Noble EG, Tanguay R, et al. Activation of heat shock transcription factor in rat heart after heat shock and exercise. Am J Physiol 1995; 268: C1387-94 35. Palermo J, Broome CS, Rasmussen P, et al. Heat shock factor activation in human muscles following a demanding intermittent exercise protocol is attenuated with hyperthermia. Acta Physiol 2008; 193: 79-88 36. Dobson CM. Protein folding and misfolding. Nature 2003; 426: 884-90 37. McArdle A, Dillmann WH, Meistril R, et al. Overexpression of HSP70 in mouse skeletal muscle protects against muscle damage and age related muscle dysfunction. FASEB J 2004; 18: 335-57 38. Thompson HS, Scordillis SP. Ubiquitin changes in human biceps muscle following exercise-induced muscle damage. Biochem Biophys Res Commun 1994; 204: 1193-8 39. Bond U, Schlesinger MJ. The chicken ubiquitin gene contains a heat shock promoter and expresses an unstable mRNA in heat shocked cells. Mol Cell Biol 1986; 6: 4602-10 40. Glickman MH, Ciechanover A. The ubiquitin-proteasome proteolytic pathway: destruction for the sake of construction. Physiol Rev 2001; 82: 373-428 41. Neufer PD, Ordway GA, Williams RS. Transient regulation of c-fos, alpha B-crystallin and HSP70 in muscle during recovery from contractile activity. Am J Physiol 1998; 274: C341-6 42. Jakob U, Gaestel M, Engel K, et al. Small heat shock proteins are molecular chaperones. J Biol Chem 1993; 268: 1517-20 43. Mounier N, Arrigo A-P. Actin cytoskeleton and small heat shock proteins: how do they interact. Cell Stress Chaperone 2002; 7: 167-76 44. Atomi Y, Yamada S, Strohman R, et al. aB-crystallin in skeletal muscle: purification and localisation. J Biochem 1991; 110: 812-22 45. Feasson L, Stockholm D, Freyssenet D, et al. Molecular adaptations of neuromuscular disease-associated proteins in response to eccentric exercise in human skeletal muscle. J Physiol 2002; 543: 297-306 46. Arrigo AP, Landry J. Expression and function of the low molecular weight heat shock proteins. In: Morimoto RI, Tissieries A, Georgopoulos C, editors. The biology of heat shock proteins and molecular chaperones. Cold Spring Harbour (NY): Cold Spring Harbour Laboratory Press, 1994: 335-73 47. Lavoie J, Gingras-Bretan G, Tanguay RM, et al. Induction of Chinese hamster HSP27 gene expression in mouse cells confers tolerance to heat shock: HSP27 stabilization of the microfilament organization. J Biol Chem 1993; 268: 3420-9 48. Gabai VL, Sherman MY. Interplay between molecular chaperones and signalling pathways in survival of heat shock. J Appl Physiol 2002; 92: 1743-8

ª 2009 Adis Data Information BV. All rights reserved.

659

49. Escobedo J, Pucci AM, Koh TJ. HSP25 protects skeletal muscle cells against oxidative stress. Free Radic Biol Med 2004; 37: 1455-62 50. Koh TJ. Do small heat shock proteins protect skeletal muscle from injury? Exerc Sport Sci Rev 2002; 30: 117-21 51. Thompson HS, Scordillis SP, Clarkson PM, et al. A single bout of eccentric exercise in increases HSP27 and HSC/HSP70 in human skeletal muscle. Acta Physiol Scand 2001; 171: 187-93 52. Thompson HS, Maynard EB, Morales ER, et al. Exerciseinduced HSP27, HSP70 and MAPK responses in human skeletal muscle. Acta Physiol Scand 2003; 178: 61-72 53. Thompson HS, Clarkson PM, Scordillis SP. The repeated bout effect and heat shock proteins: intramuscular HSP27 and HSP70 expression following two bouts of eccentric exercise in humans. Acta Physiol Scand 2002; 174: 47-56 54. Paulsen G, Vissing K, Kalhovde JM, et al. Maximal eccentric exercise induces a rapid accumulation of small heat shock proteins on myofibrils and a delayed HSP70 response in humans. Am J Physiol 2007; 293: R844-53 55. Koh TJ, Escobedo J. Cytoskeletal disruption and small heat shock protein translocation immediately after lengthening contractions. Am J Physiol 2004; 286: C713-22 56. Hood DA. Contractile activity induced mitochondrial biogenesis in skeletal muscle. J Appl Physiol 2001; 90: 1031-5 57. Hood DA, Takahashi M, Connor MK, et al. Assembly of the cellular powerhouse: current issues in mitochondrial biogenesis. Exerc Sport Sci Rev 2000; 28: 68-73 58. Martinus RM, Ryan MT, Naylor DJ, et al. Role of chaperones in the biogenesis and maintenance of mitochondrion. FASEB J 1995; 9: 371-8 59. Khassaf M, Child RB, McArdle A, et al. Time course of responses of human skeletal muscle to oxidative stress induced by non-damaging exercise. J Appl Physiol 2001; 90: 1031-5 60. Morton JP, MacLaren DPM, Cable NT, et al. Time-course and differential expression of the major heat shock protein families in human skeletal muscle following acute non-damaging treadmill exercise. J Appl Physiol 2006; 101: 176-82 61. Morton JP, MacLaren DPM, Cable NT, et al. Trained men display increased basal heat shock protein content of skeletal muscle. Med Sci Sports Exerc 2008; 40: 1255-62 62. Laroia G, Cuesta R, Brewer G, et al. Control of mRNA decay by heat shock ubiquitin proteasome pathway. Science 1999; 284: 499-503 63. Goldberg AL. Protein degradation and protection against misfolded or damaged proteins. Nature 2003; 426: 895-9 64. Samali A, Orrenius S. Heat shock proteins: regulators of stress response and apoptosis. Cell Stress Chaperone 1998; 3: 228-36 65. Kiang JG, Tsokos GC. Heat shock protein 70kDa: molecular biology, biochemistry and physiology. Pharmacol Ther 1998; 80: 183-201 66. Kregel K. Heat shock proteins: modifying factors in physiological responses and acquired thermotolerance. J Appl Physiol 2002; 92: 2177-86

Sports Med 2009; 39 (8)

660

67. Liu Y, Mayr S, Optiz-Gress A, et al. Human skeletal muscle HSP70 response to training in highly trained rowers. J Appl Physiol 1999; 86: 101-4 68. Liu Y, Lormes W, Baur C, et al. Human skeletal muscle HSP70 response to physical training depends on exercise intensity. Int J Sports Med 2000; 21: 351-5 69. Freeman BC, Morimoto RI. The human cytosolic molecular chaperones HSP90, HSP70 (HSC70) and HDJ-1 have distinct roles in recognition of non-native protein and protein refolding. EMBO J 1996; 15: 2969-75 70. Duncan RF. Inhibition of HSP90 function delays and impairs recovery from heat shock. FEBS J 2005; 272: 5244-56 71. Puntschart A, Vogt M, Widmer HR, et al. HSP70 expression in human skeletal muscle after exercise. Acta Physiol Scand 1996; 157: 411-7 72. Walsh RC, Koukoulas I, Garnham A, et al. Exercise increases serum HSP72 in humans. Cell Stress Chaperone 2001; 6: 386-93 73. Febbraio MA, Koukoulas I. HSP72 gene expression progressively increases in human skeletal muscle during prolonged exhaustive exercise. J Appl Physiol 2000; 89: 1055-60 74. Febbraio MA, Mesa JL, Chung J, et al. Glucose ingestion attenuates the exercise-induced increase in circulating heat shock protein 72 and heat shock protein 60 in humans. Cell Stress Chaperone 2004; 9: 390-6 75. Milne KJ, Noble EG. Exercise-induced elevation of HSP70 is intensity dependent. J Appl Physiol 2002; 92: 561-8 76. Gonzalez B, Hernando R, Manso R. Stress proteins of 70 kDa in chronically exercised skeletal muscle. Eur J Physiol 2000; 440: 42-9 77. Fehrenbach E, Niess AM, Schlotz E, et al. Transcriptional and translational regulation of heat shock proteins in leukocytes of endurance runners. J Appl Physiol 2000; 89: 704-10 78. Khassaf M, McArdle A, Esanu C, et al. Effect of vitamin C supplementation on antioxidant defence and stress proteins in human lymphocytes and skeletal muscle. J Physiol 2003; 549: 645-52 79. Vasilaki A, McArdle F, Iwanejiko L, et al. Adaptive responses of mouse skeletal muscle to contractile activity: the effect of age. Mech Ageing Dev 2006; 127: 830-9 80. Jackson MJ, Khassaf M, Vasilaki A, et al. Vitamin E and the oxidative stress of exercise. Ann N Y Acad Sci 2004; 1031: 158-68 81. Tupling AR, Bombardier E, Stewart RD, et al. Muscle fibre specific response of HSP70 expression in human quadriceps following acute isometric exercise. J Appl Physiol 2007; 103: 2105-11 82. Campisi J, Leem TH, Greenwood BN, et al. Habitual physical activity facilitates stress-induced HSP72 induction in brain, peripheral and immune tissues. Am J Physiol 2003; 284: R520-30 83. Febbraio MA, Steensberg A, Walsh R, et al. Reduced glycogen availability is associated with an elevation in HSP72 in contracting human skeletal muscle. J Physiol 2002; 538: 911-7 84. Paroo Z, Dipchand ES, Noble EG. Estrogen attenuates post-exercise HSP70 expression in skeletal muscle. Am J Physiol 2002; 282: 245-51

ª 2009 Adis Data Information BV. All rights reserved.

Morton et al.

85. Vasilaki A, Jackson MJ, McArdle A. Attenuated HSP70 response in skeletal muscle of aged rats following contractile activity. Muscle Nerve 2002; 25: 902-5 86. Smolka MB, Zoppi CC, Alves AA, et al. HSP72 as a complementary protection against oxidative stress induced by exercise in the soleus muscle of rats. Am J Physiol 2000; 279: R1539-45 87. Nething K, Wang L, Liu Y, et al. Blunted HSP70 response to acute exercise in well trained skeletal muscle [abstract]. Med Sci Sports Exerc 2004; 36: S318 88. Lyashko VN, Chernicov VG, Ivanov VI, et al. Comparison of the heat shock response in ethnically and ecologically different human populations. Proc Nat Acad Sci U S A 1994; 91: 2492-5 89. Amelink GJ, Bar PR. Exercise-induced muscle protein leakage in the rat: effects of hormonal manipulation. J Neuro Sci 1986; 76: 61-8 90. Shutamate JB, Brooke MH, Carroll JE, et al. Increased serum creatine kinase after exercise: a sex linked phenomenon. Neurology 1979; 29: 902-4 91. Paroo Z, Tidus PM, Noble EG. Estrogen attenuates HSP72 expression in acutely exercised male rodents. Eur J Appl Physiol 1999; 80: 180-4 92. Morton JP, Holloway K, Woods P, et al. Exercise traininginduced gender specific heat shock protein adaptations in human skeletal muscle. Muscle Nerve 2009; 39: 230-3 93. Kregel KC, Moseley PL. Differential effects of exercise and heat stress on liver HSP70 accumulation with aging. J Appl Physiol 1996; 80: 262-77 94. Jurivich DA, Qiu L, Welk JF. Attenuated stress responses in young and old human lymphocytes. Mech Ageing Dev 1997; 94: 233-49 95. Njemini R, Lambert M, Demanet C, et al. The induction of heat shock protein 70 in peripheral mononuclear blood cells in elderly patients: a role for inflammatory markers. Hum Immunol 2003; 64: 575-85 96. Demirel HA, Hamilton KL, Shanley RA, et al. Age and attenuation of exercise-induced myocardial HSP72 accumulation. Am J Physiol 2003; 285: H1609-15 97. Ritz MF, Masmoudi A, Matter N, et al. Heat stressing stimulates nuclear protein kinase C raising diacylglycerol levels: nuclear protein kinase C activation precedes HSP70mRNA expression. Receptor 1993; 3: 311-24 98. Jatella M. Overexpression of major heat shock protein HSP70 inhibits tumour necrosis factor-induced activation of phospholipase A2. J Immunol 1993; 151: 4286-94 99. Zietara MS, Skorkowski EF. Thermostability of lactate dehydrogenase LDH-A4 isoenzyme: effect of heat shock protein DnaK on the enzyme activity. Int J Biochem Cell Biol 1995; 27: 1169-74 100. Locke M, Atkinson BG, Tanquay RM, et al. Shifts in type I fiber proportion in rat hindlimb muscle are accompanied by changes in HSP72 content. Am J Physiol 1994; 266: C1240-46 101. Tupling AR, Gramolini AO, Duhamel TA, et al. HSP70 binds to the fast twitch skeletal muscle sarco(endo)plasmic reticulum Ca2+ ATPase (SERCA1a) and prevents thermal inactivation. J Biol Chem 2004; 279: 52382-9

Sports Med 2009; 39 (8)

Exercise-Induced Stress Response of Skeletal Muscle

102. Noble EG. Heat shock proteins and their induction with exercise. In: Locke M, Noble EG, editors. Exercise and stress response: the role of stress proteins. Boca Raton (FL): CRC Press LLC, 2002: 43-78 103. Takahashi M, Chesley A, Freyssenet D, et al. Contractile activity-induced adaptations in the mitochondrial protein import system. Am J Physiol 1998; 274: C1380-7 104. Reid MB. Response of the ubiquitin-proteasome pathway to changes in muscle activity. Am J Physiol 2005; 288: R1423-34 105. Kilgore JL, Timson BF, Saunders DK, et al. Stress protein induction in skeletal muscle: comparison of laboratory models to naturally occurring hypertrophy. J Appl Physiol 1994; 76: 598-601 106. O’Neill DET, Aubrey FK, Zeldin DA, et al. Slower skeletal muscle phenotypes are critical for constitutive expression of HSP70 in overloaded rat plantaris muscle. J Appl Physiol 2006; 100: 981-7 107. Naito H, Powers SK, Demirel HA, et al. Heat stress attenuates skeletal muscle atrophy in hindlimb-unweighted rats. J Appl Physiol 2000; 88: 359-63 108. Selsby JT, Dodd SL. Heat treatment reduces oxidative stress and protects muscle mass during immobilization. Am J Physiol 2005; 289: R134-9 109. Siu PM, Bryner RW, Martyn JK, et al. Apoptotic adaptations from exercise training in skeletal and cardiac muscles. FASEB J 2004; 18: 1150-2 110. Demirel HA, Powers S, Caillaud C, et al. Exercise training reduces myocardial lipid peroxidation following shortterm ischemia-reperfusion. Med Sci Sports Exerc 1998; 30: 1211-6 111. Powers SK, Demirel HA, Vincent HK, et al. Exercise training improves myocardial tolerance to in vivo ischemia-reperfusion in the rat. Am J Physiol 1998; 275: R1468-77 112. Powers SK, Locke M, Demirel HA. Exercise, heat shock proteins, and myocardial protection from I-R injury. Med Sci Sports Exerc 2001; 33: 386-92 113. Noble EG, Moraska A, Mazzeo RS, et al. Differential expression of stress proteins in rat myocardium after free wheel or treadmill run training. J Appl Physiol 1999; 86: 1696-701 114. Harris MB, Starnes JW. Effects of body temperature during exercise on myocardial adaptations. Am J Physiol 2001; 280: H2271-80 115. Atalay M, Oksala NKJ, Laaksonen DE, et al. Exercise training modulates heat shock protein response in diabetic rats. J Appl Physiol 2004; 97: 605-11 116. Kelly DA, Tidus PM, Houston ME, et al. Effect of vitamin E deprivation and exercise training on induction of HSP70. J Appl Physiol 1996; 81: 2379-85 117. Neufer PD, Ordway GA, Hand GA, et al. Continuous contractile activity induces fiber type specific expression of HSP70 in skeletal muscle. Am J Physiol 1996; 271: C1828-37 118. Ecochard L, Lhenry F, Sempore B, et al. Skeletal muscle HSP72 levels during endurance training: influence of peripheral arterial insufficiency. Eur J Physiol 2000; 440: 918-24 119. Oishi Y, Taniguchi K, Matsumoto H, et al. Muscle typespecific response of HSP60, HSP72 and HSC73 during

ª 2009 Adis Data Information BV. All rights reserved.

661

120.

121.

122.

123.

124.

125.

126.

127.

128.

129.

130.

131.

132.

133.

134.

135.

recovery after elevation of muscle temperature. J Appl Physiol 2002; 92: 1097-103 Oishi Y, Taniguchi K, Matsumoto H, et al. Differential responses of HSPs to heat stress in slow and fast regions of rat gastrocnemius muscle. Muscle Nerve 2003; 28: 587-94 Kim K-B, Kim M-H, Lee D-J. The effect of exercise in cool, control and hot environments on cardioprotective HSP70 induction. J Physiol Anthropol Appl Hum Sci 2004; 23: 225-30 Zuo LI, Christofi FL, Wright VP, et al. Intra- and extracellular measurement of reactive oxygen species produced during heat stress in diaphragm muscle. Am J Physiol 2000; 279: C1058-66 Morton JP, MacLaren DPM, Cable NT, et al. Elevated core and muscle temperature to levels comparable to exercise do not increase heat shock protein content of physically active men. Acta Physiol 2007; 190: 319-27 Pattwell DM, Jackson MJ. Contraction-induced oxidants as mediators of adaptation and damage in skeletal muscle. Exerc Sport Sci Rev 2004; 32: 14-8 Freeman ML, Borrelli MJ, Syed K, et al. Characterization of a signal generated by oxidation of protein thiols that activates the heat shock transcription factor. J Cell Physiol 1995; 164: 356-66 McDuffee AT, Senisterra G, Huntley S, et al. Proteins containing non-native disulfide bonds generated by oxidative stress can act as signals for the induction of the heat shock response. J Cell Physiol 1997; 171: 143-51 Fischer CP, Hiscock NJ, Basu S, et al. Vitamin E isoform specific inhibition of the exercise-induced heat shock protein 72 expression in humans. J Appl Physiol 2006; 100: 1679-87 Benjamin IJ, Horie S, Greenberg ML, et al. Induction of stress proteins in cultured myogenic cells: molecular signals for the activation of heat shock transcription factor during ischemia. J Clin Invest 1992; 89: 1685-9 Liu Y, Steinacker JM. Changes in skeletal muscle heat shock proteins: pathological significance. Front Biosci 2001; 6: 12-5 Morton JP, Croft L, Bartlett J, et al. Reduced carbohydrate availability does not modulate training-induced heat shock protein adaptations but does up-regulate oxidative enzyme activity in human skeletal muscle. J Appl Physiol 2009; 106: 1513-21 Febbraio MA, Pedersen BK. Muscle derived interleukin-6: mechanisms for activation and biological role. FASEB J 2002; 16: 1335-47 Febbraio MA, Steensberg A, Fischer CP. IL-6 activates HSP72 gene expression in human skeletal muscle. Biochem Biophys Res Commun 2002; 296: 1264-6 Fischer CP, Hiscock NJ, Febbraio MA, et al. Effects of antioxidant treatment on HSP72 and IL-6 gene expression in human contracting skeletal muscle [abstract]. J Physiol 2002; 539P: S120 Landry J, Bernier D, Chreiten P, et al. Synthesis and degradation of heat shock proteins during development and decay of thermotolerance. Cancer Res 1982; 42: 2457-61 Li GC, Meyer JL, Mak JY, et al. Heat-induced protection of mice against thermal death. Cancer Res 1983; 43: 5758-60

Sports Med 2009; 39 (8)

662

136. Mizzen LA, Welch WJ. Characterization of the thermotolerant cell: I. Effects on protein synthesis activity and the regulation of heat-shock protein 70 expression. J Cell Biol 1988; 106: 1105-16 137. Li GC. Induction of thermotolerance and enhanced heat shock protein synthesis in Chinese hamster fibroblasts by sodium arsenite and by ethanol. J Cell Physiol 1983; 115: 116-22 138. Broome CS, Kayani AC, Palomero J, et al. Effect of lifelong overexpression of HSP70 in skeletal muscle on age-related oxidative stress and adaptation after nondamaging contractile activity. FASEB J 2006; 20: 1549-51 139. Kayani AC, Close GL, Jackson MJ, et al. Prolonged treadmill running increases HSP70 in skeletal muscle but does not affect age-related functional deficits. Am J Physiol 2008; 294: R568-76 140. Bruce CR, Carey AL, Hawley JA, et al. Intramuscular heat shock protein 72 and heme-oxygenase mRNA are reduced

ª 2009 Adis Data Information BV. All rights reserved.

Morton et al.

in patients with type 2 diabetes: evidence that insulin resistance is associated with a disturbed antioxidant defence mechanism. Diabetes 2003; 52: 2338-45 141. Kurucz I, Morva A, Vaag A, et al. Decreased expression of heat shock protein 72 in skeletal muscle of patients with type 2 diabetes correlates with insulin resistance. Diabetes 2002; 51: 1102-9 142. Chung J, Nguyen A-K, Henstridge DC, et al. HSP72 protects against obesity-induced insulin resistance. PNAS 2008; 105: 1739-44

Correspondence: Dr James P. Morton, Research Institute for Sport and Exercise Sciences, Liverpool John Moores University, 15-21 Webster Street, Liverpool, L3 2ET, UK. E-mail: [email protected]

Sports Med 2009; 39 (8)

REVIEW ARTICLE

Sports Med 2009; 39 (8): 663-685 0112-1642/09/0008-0663/$49.95/0

ª 2009 Adis Data Information BV. All rights reserved.

Shoulder Muscle Activity and Function in Common Shoulder Rehabilitation Exercises Rafael F. Escamilla,1,2,3 Kyle Yamashiro,3 Lonnie Paulos1 and James R. Andrews1,4 1 2 3 4

Andrews-Paulos Research and Education Institute, Gulf Breeze, Florida, USA Department of Physical Therapy, California State University, Sacramento, California, USA Results Physical Therapy and Training Center, Sacramento, California, USA American Sports Medicine Institute, Birmingham, Alabama, USA

Contents Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Rotator Cuff Biomechanics and Function in Rehabilitation Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Supraspinatus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Infraspinatus and Teres Minor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 Subscapularis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Deltoid Biomechanics and Function in Rehabilitation Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Scapular Muscle Function in Rehabilitation Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Serratus Anterior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Trapezius . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Rhomboids and Levator Scapulae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Abstract

663 667 667 671 676 678 681 681 682 682 683

The rotator cuff performs multiple functions during shoulder exercises, including glenohumeral abduction, external rotation (ER) and internal rotation (IR). The rotator cuff also stabilizes the glenohumeral joint and controls humeral head translations. The infraspinatus and subscapularis have significant roles in scapular plane abduction (scaption), generating forces that are two to three times greater than supraspinatus force. However, the supraspinatus still remains a more effective shoulder abductor because of its more effective moment arm. Both the deltoids and rotator cuff provide significant abduction torque, with an estimated contribution up to 35–65% by the middle deltoid, 30% by the subscapularis, 25% by the supraspinatus, 10% by the infraspinatus and 2% by the anterior deltoid. During abduction, middle deltoid force has been estimated to be 434 N, followed by 323 N from the anterior deltoid, 283 N from the subscapularis, 205 N from the infraspinatus, and 117 N from the supraspinatus. These forces are generated not only to abduct the shoulder but also to stabilize the joint and neutralize the antagonistic effects of undesirable actions. Relatively high force from the rotator cuff not only helps abduct the shoulder but also neutralizes the superior directed force generated by the deltoids at lower abduction angles. Even though anterior deltoid force is

Escamilla et al.

664

relatively high, its ability to abduct the shoulder is low due to a very small moment arm, especially at low abduction angles. The deltoids are more effective abductors at higher abduction angles while the rotator cuff muscles are more effective abductors at lower abduction angles. During maximum humeral elevation the scapula normally upwardly rotates 45–55, posterior tilts 20–40 and externally rotates 15–35. The scapular muscles are important during humeral elevation because they cause these motions, especially the serratus anterior, which contributes to scapular upward rotation, posterior tilt and ER. The serratus anterior also helps stabilize the medial border and inferior angle of the scapular, preventing scapular IR (winging) and anterior tilt. If normal scapular movements are disrupted by abnormal scapular muscle firing patterns, weakness, fatigue, or injury, the shoulder complex functions less efficiency and injury risk increases. Scapula position and humeral rotation can affect injury risk during humeral elevation. Compared with scapular protraction, scapular retraction has been shown to both increase subacromial space width and enhance supraspinatus force production during humeral elevation. Moreover, scapular IR and scapular anterior tilt, both of which decrease subacromial space width and increase impingement risk, are greater when performing scaption with IR (‘empty can’) compared with scaption with ER (‘full can’). There are several exercises in the literature that exhibit high to very high activity from the rotator cuff, deltoids and scapular muscles, such as prone horizontal abduction at 100 abduction with ER, flexion and abduction with ER, ‘full can’ and ‘empty can’, D1 and D2 diagonal pattern flexion and extension, ER and IR at 0 and 90 abduction, standing extension from 90–0, a variety of weight-bearing upper extremity exercises, such as the push-up, standing scapular dynamic hug, forward scapular punch, and rowing type exercises. Supraspinatus activity is similar between ‘empty can’ and ‘full can’ exercises, although the ‘full can’ results in less risk of subacromial impingement. Infraspinatus and subscapularis activity have generally been reported to be higher in the ‘full can’ compared with the ‘empty can’, while posterior deltoid activity has been reported to be higher in the ‘empty can’ than the ‘full can’.

This review focuses on the scientific rationale behind choosing and progressing exercises during shoulder rehabilitation and training. Specifically, shoulder biomechanics and muscle function are presented for common open and closed chain shoulder rehabilitation exercises. Although weightbearing closed chain positions do occur in sport, such as a wrestler in a quadriceps position with hands fixed to the ground, it is more common in sport for the hand to move freely in space against varying external loads, such as in throwing a football, discus or shot put, passing a basketball, pitching a baseball, swinging a tennis racket, baseball bat or golf club, or lifting a weight overª 2009 Adis Data Information BV. All rights reserved.

head. The movements performed in these latter activities are similar to the movements that occur in open chain exercises. Nevertheless, weightbearing exercises are still used in shoulder rehabilitation, such as facilitation of proprioceptive feedback mechanisms, muscle co-contraction, and dynamic joint stability.[1] A summary of glenohumeral and scapular muscle activity (normalized by a maximum voluntary isometric contraction [MVIC]) during numerous open and closed chain shoulder exercises commonly used in rehabilitation, with varying intensities and resistive devices, are shown in tables I–IX. Several exercises presented Sports Med 2009; 39 (8)

Significantly less EMG amplitude compared with standing forward scapular punch (p < 0.002). e

IR = internal rotation. * There were no significant differences (p = 0.122) in tubing force among exercises; - there were significant differences (p < 0.001) in EMG amplitude among exercises.

Significantly less EMG amplitude compared with D2 diagonal pattern extension, horizontal abduction, internal rotation (p < 0.002). d

Significantly less EMG amplitude compared with standing scapular dynamic hug (p < 0.002).

ª 2009 Adis Data Information BV. All rights reserved.

c Significantly less EMG amplitude compared with standing internal rotation at 0 abduction (p < 0.002).

b

99 – 36 46 – 29 122 – 22 300 – 90

Significantly less EMG amplitude compared with push-up plus (p < 0.002).

104 – 54

94 – 27

47 – 26

49 – 25

665

a

<20a,d <20a,d <20a,d <20a,d <20a,d 21 – 12a <20a <20a <20a <20a <20a <20a 25 – 12a,b,c,d <20a,b,c,d 39 – 22a,d 51 – 24a,d 46 – 24a,d 76 – 32 28 – 12a <20a <20a <20a <20a <20a 46 – 24a 40 – 23a 33 – 25a,b <20a,b,d,e 62 – 31a 54 – 35a 33 – 28a 58 – 38a 53 – 40a 50 – 23a 58 – 32a 60 – 34a 260 – 50 270 – 30 260 – 40 270 – 40 260 – 50 270 – 30

Standing forward scapular punch Standing IR at 90 abduction Standing IR at 45 abduction Standing IR at 0 abduction Standing scapular dynamic hug D2 diagonal pattern extension, horizontal adduction, IR (throwing acceleration) Push-up plus

<20a,b,c,d <20a,b,c,d 26 – 19 40 – 27 38 – 20 39 – 26

Teres major Latissimus EMG (%MVIC)- dorsi EMG (%MVIC)Upper Lower Supraspinatus Infraspinatus Pectoralis subscapularis subscapularis EMG (%MVIC)- EMG (%MVIC)- major EMG EMG (%MVIC)- EMG (%MVIC)(%MVIC)Tubing force (N)* Exercise

Table I. Mean (– SD) tubing force and glenohumeral electromyograph (EMG), normalized by a maximum voluntary isometric contraction (MVIC), during shoulder exercises using elastic tubing and bodyweight resistance, with intensity for each exercise normalized by a ten-repetition maximum. Data for muscles with EMG amplitude >45% of a MVIC are set in bold italic type, and these exercises are considered to be an effective challenge for that muscle (adapted from Decker et al.,[10] with permission)

Shoulder Muscle Activity and Function

in tables I–IX that demonstrated effective glenohumeral and scapular muscle recruitment and muscle activity are illustrated in figures 1–10. To help generalize comparisons in muscle activity from tables I–IX, 0–20% MVIC was considered low muscle activity, 21–40% MVIC was considered moderate muscle activity, 41–60% MVIC was considered high muscle activity, and >60% MVIC was considered very high muscle activity.[2] Because many papers that analyse muscle activity during shoulder exercises involve the use of electromyography (EMG), such as exercises shown in tables I–IX, it is important that clinicians understand what information EMG can and cannot provide. Although EMG-driven mathematical knee models have been successfully developed to estimate both knee muscle and joint force and stress,[3,4] clinically applicable mathematical shoulder models have not yet been developed to estimate individual shoulder muscle and joint forces and stress during exercise. Therefore, the clinician should be careful not to equate EMG with muscle or joint force. However, a somewhat linear relationship between muscle EMG and force has been demonstrated during near isometric and constant velocity contractions.[5-7] However, this relationship may be highly nonlinear during rapid or fatiguing muscle contractions.[8] During muscle fatigue, EMG may increase, decrease or stay the same while muscle force decreases.[9] In addition, EMG amplitude has been shown to be similar or less in maximum eccentric contractions compared with maximum concentric contractions, even though peak force is greater with maximum eccentric contractions. Therefore, caution should be taken in interpreting the EMG signal during exercise. Nevertheless, shoulder EMG during exercise can still provide valuable information to the clinician that can be applied to shoulder rehabilitation and training. EMG provides information on when, how much and how often a muscle is active throughout an exercise range of motion (ROM). For example, early after rotator cuff surgery the recovering patient may want to avoid exercises that generate high rotator cuff activity so as not to stress the healing tissue, but exercises that Sports Med 2009; 39 (8)

Escamilla et al.

666

Table II. Mean (– SD) rotator cuff and deltoid electromyograph (EMG), normalized by a maximum voluntary isometric contraction (MVIC), during shoulder external rotation exercises using dumbbell resistance with intensity for each exercise normalized by a ten-repetition maximum. Data for muscles with EMG amplitude >45% of an MVIC are set in bold italic type, and these exercises are considered to be an effective challenge for that muscle (adapted from Reinold et al.,[12] with permission) Exercise

Infraspinatus EMG (%MVIC)*

Teres minor EMG (%MVIC)*

Supraspinatus EMG (%MVIC)*

Middle deltoid EMG (%MVIC)*

Posterior deltoid EMG (%MVIC)*

Side-lying external rotation at 0 abduction

62 – 13

67 – 34

51 – 47e

36 – 23e

52 – 42e

e

43 – 30e

Standing ER in scapular plane at 45 abduction and 30 horizontal adduction

53 – 25

55 – 30

Prone ER at 90 abduction

50 – 23

48 – 27

Standing ER at 90 abduction

50 – 25

39 – 13

32 – 24

a

c,e

38 – 19

68 – 33

49 – 15e

79 – 31

57 – 32

55 – 23e

59 – 33e

c,e

11 – 6

c,d,e

31 – 27a,c,d,e

Standing ER at approximately 15 abduction with towel roll

50 – 14

46 – 41

41 – 37

Standing ER at 0 abduction without towel roll

40 – 14a

34 – 13a

41 – 38c,e

11 – 7c,d,e

27 – 27a,c,d,e

Prone horizontal abduction at 100 abduction with ER (thumb up)

39 – 17a

44 – 25

82 – 37

82 – 32

88 – 33

a

Significantly less EMG amplitude compared with side-lying external rotation at 0 abduction (p < 0.05).

b

Significantly less EMG amplitude compared with standing external rotation in scapular plane at 45 abduction and 30 horizontal adduction (p < 0.05).

c

Significantly less EMG amplitude compared with prone external rotation at 90 abduction (p < 0.05).

d

Significantly less EMG amplitude compared with standing external rotation at 90 abduction (p < 0.05).

e

Significantly less EMG amplitude compared with prone horizontal abduction at 100 abduction with external rotation (thumb up; p < 0.05).

ER = external rotation. * There were significant differences (p < 0.01) in EMG amplitude among exercises.

activate scapular muscles with minimal cuff activity may be appropriate during this phase of rehabilitation. During more advanced phases of rotator cuff rehabilitation, employing exercises that produce moderate to higher levels of rotator cuff activity may be appropriate. In the scientific literature there is a wide array of methods used during EMG studies involving shoulder exercises, so the clinician should interpret EMG data cautiously. A practical application of EMG is to compare the EMG signal of one muscle across different exercises of relative intensity, and express the EMG signal relative to some common reference, such as percentage of a MVIC (tables I–IX). For example, in table I supraspinatus activity was significantly greater in the standing scapular dynamic hug (62 – 31% MVIC) compared with the standing internal rotation (IR) at 45 abduction (33 – 25% MVIC), with intensity for both of these exercises expressed by a ten-repetition maximum (10 RM). It is more difficult to compare muscle activity between studies when exercise intensity is different between exercises. For example, in one study ª 2009 Adis Data Information BV. All rights reserved.

exercise intensity may be 30% 1 RM, while another study examining the same exercises and muscles may involve an exercise intensity of 80% 1 RM. It is obvious that the normalized EMG would be much higher in the study that used the 80% 1 RM intensity. Comparing muscle activity between studies is also difficult for other reasons, such as differences in MVIC determination, the use of different normalization techniques, EMG differences in isometric versus dynamic contractions, fatigued versus nonfatigued muscle, surface versus indwelling electrodes, electrode size and placement, and varying signal processing techniques. Another difficulty in interpreting EMG data is that some studies perform statistical analyses (tables I–IV)[1,10-12] while other studies do not (tables V–IX).[13-16] Without statistics, it may be more difficult to compare and interpret muscle activity among exercises. For example, for one exercise a muscle may have a normalized mean activity of 50% and a standard deviation of 50%, and for another exercise this same muscle may only have a normalized mean activity of 20% Sports Med 2009; 39 (8)

Shoulder Muscle Activity and Function

667

and standard deviation of 40%. From the mean activity it may appear that the exercise with 50% activity is more effective than the exercise with 20% activity. However, the high standard deviations implies there was high variability in muscle activity among subjects, and statistically there may be no significant difference in muscle activity between these two exercises. 1. Rotator Cuff Biomechanics and Function in Rehabilitation Exercises Rotator cuff muscles have been shown to be a stabilizer of the glenohumeral joint in multiple shoulder positions.[17] Appropriate rehabilitation progression and strengthening of the rotator cuff

is important in order to provide appropriate force to help elevate and move the arm, compress and centre the humeral head within the glenoid fossa during shoulder movements, and resist humeral head superior translation due to deltoid activity.[18-22] This latter function is important in early humeral elevation when the resultant force vector from the deltoids is directed in a more superior direction. This section presents rotator cuff biomechanics and function during a large array of shoulder exercises. 1.1 Supraspinatus

The supraspinatus compresses, abducts and provides a small external rotation (ER) torque to

Table III. Mean (– SD) trapezius and serratus anterior muscle activity (electromyograph [EMG] normalized by a maximum voluntary isometric contraction [MVIC]) during shoulder exercises using dumbbell or similar resistance with intensity for each exercise normalized by a fiverepetition maximum. Data for muscles with EMG amplitude >50% of a MVIC are set in bold italic type, and these exercises are considered to be an effective challenge for that muscle (adapted from Ekstrom et al.,[11] with permission) Exercise

Upper trapezius EMG (%MVIC)*

Shoulder shrug

119 – 23

Middle trapezius EMG (%MVIC)* 53 – 25b,c,d

Lower trapezius EMG (%MVIC)*

Serratus anterior EMG (%MVIC)*

21 – 10b,c,d,f,g,h

27 – 17c,e,f,g,h,i,j

63 – 17

a

79 – 23

45 – 17

Prone horizontal abduction at 135 abduction with ER (thumb up)

79 – 18

a

101 – 32

97 – 16

Prone horizontal abduction at 90 abduction with ER (thumb up)

66 – 18 a

87 – 20

74 – 21c

Prone external rotation at 90 abduction

20 – 18a,b,c,d,e,f,g

45 – 36b,c,d

79 – 21

Prone rowing

a

D1 diagonal pattern flexion, horizontal adduction and ER

66 – 10

Scaption above 120 with ER (thumb up) ‘full can’

79 – 19 a

49 – 16b,c,d

a

b,c,d

Scaption below 80 with ER (thumb up) ‘full can’

72 – 19

47 – 16

a,b,c,d,e,f,g,h

Supine scapular protraction with shoulders horizontally flexed 45 and elbows flexed 45

7–5

Supine upward scapular punch

7 – 3a,b,c,d,e,f,g,h

a

21 – 9

a,b,c,d,f,g,h

7–3

a,b,c,d,f,g,h

12 – 10b,c,d

39 – 15

c,d,h

14 – 6c,e,f,g,h,i,j 43 – 17e,f 9 – 3c,e,f,g,h,i,j 57 – 22e,f

b,c,d,f,g,h

100 – 24

61 – 19c

96 – 24

50 – 21c,h

62 – 18e,f

b,c,d,f,g,h

53 – 28e,f

11 – 5b,c,d,f,g,h

62 – 19e,f

5–2

Significantly less EMG amplitude compared with shoulder shrug (p < 0.05).

b

Significantly less EMG amplitude compared with prone rowing (p < 0.05).

c

Significantly less EMG amplitude compared with prone horizontal abduction at 135 abduction with external rotation (p < 0.05).

d

Significantly less EMG amplitude compared with prone horizontal abduction at 90 abduction with external rotation (p < 0.05).

e

Significantly less EMG amplitude compared with D1 diagonal pattern flexion, horizontal adduction and external rotation (p < 0.05).

f

Significantly less EMG amplitude compared with scaption above 120 with ER (thumb up) [p < 0.05].

g

Significantly less EMG amplitude compared with scaption below 80 with ER (thumb up) [p < 0.05].

h

Significantly less EMG amplitude compared with prone external rotation at 90 abduction (p < 0.05).

i

Significantly less EMG amplitude compared with supine scapular protraction with shoulders horizontally flexed 45 and elbows flexed 45 (p < 0.05).

j

Significantly less EMG amplitude compared with supine upward scapular punch (p < 0.05).

ER = external rotation. * There were significant differences (p < 0.05) in EMG amplitude among exercises.

ª 2009 Adis Data Information BV. All rights reserved.

Sports Med 2009; 39 (8)

Escamilla et al.

668

Table IV. Mean (– SD) ground reaction force on hand (normalized by bodyweight [BW]) and glenohumeral electromyograph (EMG), normalized by a maximum voluntary isometric contraction (MVIC), during low-to-high demand weight-bearing shoulder exercises. Data for muscles with EMG amplitude >40% of a MVIC are set in bold italic type, and these exercises are considered to be an effective challenge for that muscle (adapted from Uhl et al.,[1] with permission) Exercise

Prayer Quadruped Tripod Bipod (alternating arm and leg)

Ground reaction force on hand (%BW)*

Supraspinatus EMG (%MVIC)-

6 – 3a,b,c,d,e,f 19 – 2

a,b,c,d,e

32 – 3

a

34 – 4

a a

2 – 2a,b,c,d 6 – 10

a,b

10 – 11

a

12 – 13

a

14 – 14

a

Infraspinatus EMG (%MVIC)-

Anterior deltoid EMG (%MVIC)-

Posterior deltoid EMG (%MVIC)-

Pectoralis major EMG (%MVIC)-

4 – 3a,b,c,d,e

2 – 4a,b,c

4 – 3a,d,e

7 – 4a,b,c

a,b,c,d,e

a,b,c

a,d,e

10 – 4a,b,c

11 – 8

37 – 26

a

42 – 33

a

44 – 31

a

6–6

12 – 10

a,b

18 – 10

a

6–4

27 – 16

a

16 – 8a,b

28 – 16

a

22 – 10a

a

33 – 20

Push-up

34 – 3

31 – 16

18 – 12

Push-up feet elevated

39 – 5a

18 – 16a

52 – 32a

37 – 15

23 – 14a

42 – 28

One-arm push-up

60 – 6

29 – 20

86 – 56

46 – 20

74 – 43

44 – 45

a

Significantly less compared with the one-arm push-up (p < 0.002).

b

Significantly less compared with the push-up feet elevated (p < 0.002).

c

Significantly less compared with the push-up (p < 0.002).

d

Significantly less compared with the pointer (p < 0.002).

e

Significantly less compared with the tripod (p < 0.002).

f

Significantly less compared with the quadruped (p < 0.002).

* There were significant differences (p < 0.001) in ground reaction force among exercises; - there were significant differences (p < 0.001) in EMG amplitude among exercises.

the glenohumeral joint. From three-dimensional (3-D) biomechanical shoulder models, predicted supraspinatus force during maximum effort isometric scapular plane abduction (scaption) at the 90 position was 117 N.[18] In addition, supraspinatus activity increases as resistance increases during scaption movements, peaking at 30–60 for any given resistance (table IX). At lower scaption angles, supraspinatus activity increases to provide additional humeral head compression within the glenoid fossa to counter the humeral head superior translation from the deltoids (table IX).[13] Due to a decreasing moment arm with abduction, the supraspinatus is more effective during scaption at smaller abduction angles, but it still generates abductor torque (a function of both moment arm and muscle force) at larger abduction angles.[18-20] The abduction moment arm for the supraspinatus peaks at approximately 3 cm near 30 abduction, but maintains an abduction moment arm of greater than 2 cm throughout shoulder abduction ROM.[19,20] Its ability to generate abduction torque during scaption appears to be greatest with the shoulder in neutral rotation or in slight IR or ER.[19,20] ª 2009 Adis Data Information BV. All rights reserved.

This is consistent with both EMG and magnetic resonance imaging (MRI) data while performing scaption with IR (‘empty can’)[23,24] and scaption with IR (‘full can’),[25] with both exercises producing similar amounts of supraspinatus activity.[16,25,26] Even though supraspinatus activity is similar between ‘empty can’ and ‘full can’ exercises, there are several reasons why the ‘full can’ may be preferred over the ‘empty can’ during rehabilitation and supraspinatus testing. Firstly, the internally rotated humerus in the ‘empty can’ does not allow the greater tuberosity to clear from under the acromion during humeral elevation, which may increase subacromial impingement risk because of decreased subacromial space width.[27,28] Secondly, abducting in extreme IR progressively decreases the abduction moment arm of the supraspinatus from 0 to 90 abduction.[19] A diminished mechanical advantage may cause the supraspinatus to have to work harder, thus increasing its tensile stress (which may be problematic in a healing tendon). Thirdly, scapular kinematics are different between ‘empty can’ and ‘full can’ exercises. Scapular IR (transverse Sports Med 2009; 39 (8)

Anterior deltoid EMG (%MVIC)

Middle deltoid EMG (%MVIC)

Posterior deltoid EMG (%MVIC)

Supraspinatus Subscapularis Infraspinatus EMG EMG EMG (%MVIC) (%MVIC) (%MVIC)

Teres minor EMG (%MVIC)

Pectoralis major EMG (%MVIC)

Latissimus dorsi EMG (%MVIC)

Flexion above 120 with ER (thumb up)

69 – 24

73 – 16

£50

67 – 14

52 – 42

66 – 15

£50

£50

£50

Abduction above 120 with ER (thumb up)

62 – 28

64 – 13

£50

£50

50 – 44

74 – 23

£50

£50

£50

Scaption above 120 with IR (thumb down) ‘empty can’

72 – 23

83 – 13

£50

74 – 33

62 – 33

£50

£50

£50

£50

Scaption above 120 with ER (thumb up) ‘full can’

71 – 39

72 – 13

£50

64 – 28

£50

60 – 21

£50

£50

£50

Military press

62 – 26

72 – 24

£50

80 – 48

56 – 48

£50

£50

£50

£50

Prone horizontal abduction at 90 abduction with IR (thumb down)

£50

80 – 23

93 – 45

£50

£50

74 – 32

68 – 36

£50

£50

Prone horizontal abduction at 90 abduction with ER (thumb up)

£50

79 – 20

92 – 49

£50

£50

88 – 25

74 – 28

£50

£50

Press-up

£50

£50

£50

£50

£50

£50

£50

84 – 42

55 – 27

Prone rowing

£50

92 – 20

88 – 40

£50

£50

£50

£50

£50

£50

Side-lying ER at 0 abduction

£50

£50

64 – 62

£50

£50

85 – 26

80 – 14

£50

£50

Side-lying eccentric control of 0–135 horizontal adduction (throwing deceleration)

£50

58 – 20

63 – 28

£50

£50

57 – 17

£50

£50

£50

ER = external rotation; IR = internal rotation.

669

Sports Med 2009; 39 (8)

Exercise

Shoulder Muscle Activity and Function

ª 2009 Adis Data Information BV. All rights reserved.

Table V. Peak (– SD) glenohumeral electromyograph (EMG), normalized by a maximum voluntary isometric contraction (MVIC), over 30 arc of movement during shoulder exercises using dumbbells. Data for muscles with EMG amplitude >50% of a MVIC over at least three consecutive 30 arcs of motion are set in bold italic type, and these exercises are considered to be an effective challenge for that muscle (adapted from Townsend et al.,[16] with permission)

Escamilla et al.

55 – 34 £50 £50 Push-up with hands separated

ª 2009 Adis Data Information BV. All rights reserved.

ER = external rotation; IR = internal rotation.

58 – 45 73 – 3

69 – 31 57 – 36

80 – 38 £50

£50 £50

£50 £50 £50 Push-up plus

£50

77 – 49 £50 Prone extension at 90 flexion

£50

£50

£50 £50 £50 £50 81 – 76

89 – 62

59 – 51 112 – 84 Prone rowing

£50

£50 £50 56 – 46 114 – 69

£50

£50 £50 Press-up

67 – 50

£50 £50 £50 £50

£50

96 – 73 75 – 27 Prone horizontal abduction at 90 abduction with ER (thumb up)

£50

£50 £50 £50 87 – 66

£50

108 – 63 62 – 53 Prone horizontal abduction at 90 abduction with IR (thumb down)

63 – 41

£50 £50 66 – 38 96 – 57

£50

£50 64 – 26

56 – 24

60 – 42 82 – 36 £50 £50

£50

Military press

£50

84 – 20 91 – 52 65 – 79 69 – 46 £50 54 – 16 Scaption above 120 with ER (thumb up) ‘full can’

60 – 22

£50 72 – 46

74 – 65 96 – 53

96 – 45 £50

64 – 53 £50

£50

£50 52 – 30 Abduction above 120 with ER (thumb up)

68 – 53

£50 £50 Flexion above 120 with ER (thumb up)

60 – 18

Pectoralis minor EMG (%MVIC) Lower serratus anterior EMG (%MVIC) Middle serratus anterior EMG (%MVIC) Rhomboids EMG (%MVIC) Levator scapulae EMG (%MVIC) Lower trapezius EMG (%MVIC) Middle trapezius EMG (%MVIC) Upper trapezius EMG (%MVIC) Exercise

Table VI. Peak (– SD) scapular electromyograph (EMG), normalized by a maximum voluntary isometric contraction (MVIC), over 30 arc of movement during shoulder exercises using dumbbells with intensity normalized for each exercise by a ten-repetition maximum. Data for muscles with EMG amplitude >50% of a MVIC over at least three consecutive 30 arcs of motion are set in bold italic type, and these exercises are considered to be an effective challenge for that muscle (adapted from Moseley et al.,[15] with permission)

670

plane movement with medial border moving posterior, resulting in ‘winging’) and anterior tilt (sagittal plane movement with the inferior angle moving posterior), both of which decrease subacromial space width, are greater in the ‘empty can’ compared with the ‘full can’.[29] This occurs in part because humeral IR in the ‘empty can’ tensions both the posteroinferior capsule and rotator cuff (infraspinatus primarily), which originate from the posterior glenoid and infraspinous fossa. Tension in these structures contributes to an anterior tilted and internally rotated scapula, which protracts the scapula. This is clinically important, as Smith et al.[30] reported that relative to a neutral scapular position, scapular protraction significantly reduced glenohumeral IR and ER strength by 13–24% and 20%, respectively. Moreover, scapular protraction has been shown to decrease subacromial space width, increasing impingement risk.[31] In contrast, scapular retraction has been shown to both increase subacromial space width[31] and enhance supraspinatus force production during humeral elevation compared with a protracted position.[32] This emphasizes the importance of strengthening the scapular retractors and maintaining good posture. Fourthly, although both the ‘empty can’ and ‘full can’ test positions have been shown to be equally accurate in detecting a torn supraspinatus tendon, the use of the ‘full can’ test position may be desirable in the clinical setting because there is less pain provocation,[33] and it has been shown to be a more optimal position for supraspinatus isolation.[25] The supraspinatus is active in numerous shoulder exercises other than the ‘empty can’ and ‘full can’. High to very high supraspinatus activity has been quantified in several common rotator cuff exercises, such as prone horizontal abduction at 100 abduction with ER, prone ER at 90 abduction, standing ER at 90 abduction, flexion above 120 with ER, military press (trunk vertical), side lying abduction, proprioceptive neuromuscular facilitation (PNF) scapular clock, and PNF D2 diagonal pattern flexion and extension (tables I, II, V and VII).[10,12,14,16,34-39] When these shoulder exercises are compared with each other, mixed results have been reported. Some Sports Med 2009; 39 (8)

35 – 17 69 – 40 46 – 31 69 – 47 36 – 24 45 – 36 19 – 11 Standing forward scapular punch

19 – 13 12 – 8

ª 2009 Adis Data Information BV. All rights reserved.

EMG data support prone horizontal abduction at 100 abduction with ER over the ‘empty can’ in supraspinatus activity,[35,38] while other EMG data show no difference in supraspinatus activity between these two exercises.[37] In contrast, MRI data support both ‘empty can’ and ‘full can’ over prone horizontal abduction at 100 abduction with ER in activating the supraspinatus.[26] Interestingly, high to very high supraspinatus activity has also been reported in several exercises that are not commonly thought of as rotator cuff exercises, such as standing forward scapular punch, rowing exercises, push-up exercises, and two-hand overhead medicine ball throws (tables I, IV and VII).[1,10,40,41] The supraspinatus also provides weak rotational torques due to small rotational moment arms.[20] From 3-D biomechanical shoulder models, predicted supraspinatus force during maximum effort ER in 90 abduction was 175 N.[18] The anterior portion, which is considered the strongest,[42] has been shown to be a weak internal rotator at 0 abduction (0.2 cm moment arm), no rotational ability at 30 abduction, and a weak external rotator at 60 abduction (approximately 0.2 cm moment arm).[20] In contrast, the posterior portion of the supraspinatus has been shown to provide an ER torque throughout shoulder abduction, with an ER moment arm that progressively decreases as abduction increases (approximately 0.7 cm at 0 abduction and 0.4 cm at 60 abduction).[20] When anterior and posterior portions of the supraspinatus are viewed as a whole, this muscle provides weak ER regardless of abduction angle, although it appears to be a more effective external rotator at smaller abduction angles.[20] ER = external rotation; IR = internal rotation.

27 – 17

29 – 16 109 – 58 46 – 38 69 – 50

31 – 15

Standing low scapular rows at 45 flexion

34 – 23

98 – 74 40 – 26 81 – 65 18 – 10 15 – 11 Standing mid scapular rows at 90 flexion

26 – 16

101 – 47

112 – 62 42 – 22

42 – 28 74 – 53

99 – 38 32 – 14 61 – 41

31 – 25

26 – 12

15 – 11

Flexion above 120 with ER (thumb up)

34 – 17

47 – 34

671

Standing high scapular rows at 135 flexion

24 – 21

50 – 57 96 – 50

63 – 38 41 – 30

30 – 21 97 – 55

71 – 43 41 – 21

27 – 16

28 – 18

19 – 15

16 – 11

21 – 11 Standing extension from 90–0

6–6 Standing IR at 0 abduction

Standing IR at 90 abduction

51 – 30

32 – 51 93 – 41 10 – 6 74 – 47

46 – 20

16 – 8

4–3

89 – 47

84 – 39 20 – 13

50 – 21 57 – 50

72 – 55 8–7

50 – 22

6–6

22 – 12

13 – 7

12 – 8

Standing ER at 0 abduction

Standing ER at 90 abduction

45 – 21 90 – 50 64 – 33 69 – 48 30 – 17 13 – 8

44 – 16

33 – 22 89 – 57 36 – 32 94 – 51 27 – 20 30 – 11

D2 diagonal pattern extension, horizontal adduction, IR (throwing acceleration) Eccentric arm control portion of D2 diagonal pattern flexion, abduction, ER (throwing deceleration)

22 – 12

Infraspinatus EMG (%MVIC) Teres minor EMG (%MVIC) Supraspinatus EMG (%MVIC) Subscapularis EMG (%MVIC) Middle deltoid EMG (%MVIC) Anterior deltoid EMG (%MVIC) Tubing force (N) Exercise

Table VII. Mean (– SD) tubing force and rotator cuff and deltoid electromyograph (EMG), normalized by a maximum voluntary isometric contraction (MVIC), during shoulder exercises using elastic tubing. Data for muscles with EMG amplitude >45% of an MVIC are set in bold italic type, and these exercises are considered to be an effective challenge for that muscle (adapted from Meyers et al.,[14] with permission)

Shoulder Muscle Activity and Function

1.2 Infraspinatus and Teres Minor

The infraspinatus and teres minor comprise the posterior cuff, which provides glenohumeral compression, ER and abduction, and resists superior and anterior humeral head translation by exerting an posteroinferior force to the humeral head.[22] The ER provided from the posterior cuff helps clear the greater tuberosity from under the coracoacromial arch during overhead movements, minimizing subacromial impingement. Sports Med 2009; 39 (8)

Escamilla et al.

672

Table VIII. Mean (– SD) tubing force and glenohumeral and scapular electromyograph (EMG), normalized by a maximum voluntary isometric contraction (MVIC), during shoulder exercises using elastic tubing. Data for muscles with EMG amplitude >45% of an MVIC are set in bold italic type, and these exercises are considered to be an effective challenge for that muscle (adapted from Meyers et al.,[14] with permission) Exercise

Tubing force (N)

Pectoralis major EMG (%MVIC)

Latissimus dorsi EMG (%MVIC)

D2 diagonal pattern extension, horizontal adduction, IR (throwing acceleration)

30 – 11

36 – 30

26 – 37

Eccentric arm control portion of D2 diagonal pattern flexion, abduction, ER (throwing deceleration)

13 – 8

22 – 28

35 – 48

Triceps brachii EMG (%MVIC)

Lower trapezius EMG (%MVIC)

Rhomboids EMG (%MVIC)

Serratus anterior EMG (%MVIC)

6–4

32 – 15

54 – 46

82 – 82

56 – 36

11 – 7

22 – 16

63 – 42

86 – 49

48 – 32

Biceps brachii EMG (%MVIC)

Standing ER at 0 abduction

13 – 7

10 – 9

33 – 39

7–4

22 – 17

48 – 25

66 – 49

18 – 19

Standing ER at 90 abduction

12 – 8

34 – 65

19 – 16

10 – 8

15 – 11

88 – 51

77 – 53

66 – 39

Standing IR at 0 abduction

16 – 8

36 – 31

34 – 34

11 – 7

21 – 19

44 – 31

41 – 34

21 – 14

Standing IR at 90 abduction

16 – 11

18 – 23

22 – 48

9–6

13 – 12

54 – 39

65 – 59

54 – 32

Standing extension from 90–0

21 – 11

22 – 37

64 – 53

10 – 27

67 – 45

53 – 40

66 – 48

30 – 21

Flexion above 120 with ER (thumb up)

26 – 12

19 – 13

33 – 34

22 – 15

22 – 12

49 – 35

52 – 54

67 – 37

Standing high scapular rows at 135 flexion

15 – 11

29 – 56

36 – 36

7–4

19 – 8

51 – 34

59 – 40

38 – 26

Standing mid scapular rows at 90 flexion

15 – 11

18 – 34

40 – 42

17 – 32

21 – 22

39 – 27

59 – 44

24 – 20

Standing low scapular rows at 45 flexion

12 – 8

17 – 32

35 – 26

21 – 50

21 – 13

44 – 32

57 – 38

22 – 14

Standing forward scapular punch

19 – 11

19 – 33

32 – 35

12 – 9

27 – 28

39 – 32

52 – 43

67 – 45

ER = external rotation; IR = internal rotation.

From 3-D biomechanical shoulder models, the maximum predicted isometric infraspinatus force was 723 N for ER at 90 abduction and 909 N for ER at 0 abduction.[18] The maximum predicted teres minor force was much less than the infraspinatus during maximum ER at both 90 abduction (111 N) and 0 abduction (159 N).[18] The effectiveness of the posterior cuff to laterally rotate depends on glenohumeral position. For the infraspinatus, its superior, middle and inferior heads all generate its largest ER torque at 0 abduction, primarily because its moment arm is greatest at 0 abduction (approximately 2.2 cm).[20] As the abduction angle increases, the moment arms of the inferior and middle heads decrease slightly but stay relatively constant, while the moment arm of the superior head progressively decreases until it is about 1.3 cm at 60 abduction.[20] These data imply that the infraspinatus is a more effective external rotator at ª 2009 Adis Data Information BV. All rights reserved.

lower abduction angles compared with higher abduction angles. Although infraspinatus activity during ER has been shown to be similar at 0, 45 and 90 abduction (table II),[12,14,25] ER at 0 abduction has been shown to be the optimal position to isolate the infraspinatus muscle,[25] and there is a trend towards greater infraspinatus activity during ER at lower abduction angles compared with higher abduction angles.[12,43] The teres minor generates a relatively constant ER torque (relatively constant moment arm of approximately 2.1 cm) throughout arm abduction movement, which implies that the abduction angle does not affect the effectiveness of the teres minor to generate ER torque.[20] Teres minor activity during ER is similar at 0, 45 and 90 abduction (table II).[12,14] In addition, both infraspinatus and teres minor activities are similar during external rotation movements regardless of abduction positions.[12,16,34] Sports Med 2009; 39 (8)

Shoulder Muscle Activity and Function

673

What is not readily apparent is the significant role of the infraspinatus as a shoulder abductor in the scapular plane.[18-20] From 3-D biomechanical shoulder models, predicted infraspinatus force during maximum isometric effort scaption (90 position) is 205 N, nearly twice the predicted force from supraspinatus in this position.[18] Liu et al.[19] reported that in scaption with neutral rotation the infraspinatus had an abductor moment

arm that was small at 0 abduction but increased to 1 cm at 15 abduction and remained fairly constant throughout increasing abduction angles. Moreover, infraspinatus activity increases as resistance increases, peaking at 30–60 for any given resistance (table IX).[13] As resistance increases, infraspinatus activity increases to help generate a higher torque in scaption, and at lower scaption angles infraspinatus activity increases to resist superior

Table IX. Mean (– SD) glenohumeral electromyograph (EMG), normalized by a maximum voluntary isometric contraction (MVIC), during scaption with neutral rotation and increasing load using dumbbells (adapted from Alpert et al.,[13] with permission) Anterior deltoid EMG (%MVIC)

Middle deltoid EMG (%MVIC)

Posterior deltoid EMG (%MVIC)

Supraspinatus EMG (%MVIC)

Infraspinatus EMG (%MVIC)

Teres minor EMG (%MVIC)

Subscapularis EMG (%MVIC)

0% NMWa 0–30

22 – 10

30 – 18

2–2

36 – 21

16 – 7

9–9

6–7

30–60

53 – 22

60 – 27

2–3

49 – 25

34 – 14

11 – 10

14 – 13

60–90

68 – 24

69 – 29

2–3

47 – 19

37 – 15

15 – 14

18 – 15

90–120

78 – 27

74 – 33

2–3

42 – 14

39 – 20

19 – 17

21 – 19

120–150

90 – 31

77 – 35

4–4

40 – 20

39 – 29

25 – 25

23 – 18

25% NMWa 0–30

42 – 14

55 – 28

5 – 11

64 – 37

39 – 16

17 – 16

14 – 10

30–60

82 – 20

81 – 21

6–8

79 – 29

64 – 23

24 – 23

32 – 15

60–90

97 – 33

87 – 26

4–4

65 – 21

60 – 24

23 – 21

34 – 18

90–120

96 – 30

85 – 28

4–4

53 – 18

49 – 24

21 – 17

28 – 18

120–150

71 – 39

70 – 36

10 – 6

41 – 23

43 – 30

32 – 26

18 – 19

50% NMWa 0–30

68 – 21

79 – 30

12 – 18

89 – 45

69 – 27

36 – 28

31 – 14

30–60

113 – 33

96 – 24

11 – 14

98 – 35

93 – 27

45 – 33

54 – 24

60–90

113 – 41

91 – 26

10 – 11

82 – 27

80 – 30

40 – 27

50 – 31

90–120

90 – 34

79 – 28

9 – 10

53 – 17

56 – 28

27 – 22

28 – 22

120–150

47 – 38

44 – 35

14 – 15

29 – 8

40 – 28

30 – 22

16 – 18

75% NMWa 0–30

81 – 18

88 – 30

14 – 19

99 – 45

85 – 30

48 – 34

40 – 20

30–60

127 – 44

104 – 33

13 – 14

109 – 37

108 – 33

61 – 37

61 – 32

60–90

121 – 45

97 – 27

14 – 13

91 – 25

96 – 35

54 – 30

50 – 31

90–120

88 – 35

79 – 28

15 – 16

56 – 17

63 – 28

39 – 27

27 – 22

120–150

38 – 33

35 – 26

20 – 22

28 – 12

32 – 18

36 – 23

18 – 15

90% NMWa 0–30

96 – 33

108 – 43

14 – 14

120 – 49

93 – 16

41 – 28

54 – 19

30–60

129 – 47

115 – 45

15 – 9

122 – 37

104 – 24

56 – 27

78 – 41

60–90

135 – 53

102 – 36

13 – 11

104 – 33

86 – 20

54 – 22

67 – 40

90–120

97 – 41

78 – 30

12 – 6

67 – 31

47 – 12

32 – 20

41 – 29

120–150

26 – 14

19 – 14

16 – 9

22 – 19

26 – 15

23 – 12

26 – 17

a

NMW = normalized maximum weight lifted in pounds, where 100% of NMW was calculated pounds by the peak torque value (in footpounds) that was generated from a 5-second maximum isometric contraction in 20 scaption divided by each subject’s arm length (in feet). Mean (– SD) NMW was 21 – 8 pounds (approximately 93 – 36 N).

ª 2009 Adis Data Information BV. All rights reserved.

Sports Med 2009; 39 (8)

674

Escamilla et al.

activity during ‘full can’ compared with ‘empty can’ exercise.[16] Moreover, MRI data demonstrate similar infraspinatus activity during abduction with IR and abduction with ER.[26] The infraspinatus is active in numerous shoulder exercises other than ‘empty can’, ‘full can’, abduction and ER exercises (tables V and VII). High to very high infraspinatus activity has been quantified in prone horizontal abduction at 100 abduction with ER and IR, flexion, side-lying abduction, standing extension from 90 to 0, and D1 and D2 diagonal pattern flexion (tables V

Fig. 1. D2 diagonal pattern extension, horizontal adduction and internal rotation (see tables I, VII and VIII for muscle activity during this exercise).

humeral head translation due to the increased activity from the deltoids.[22] Another finding is that the abductor moment arm of the infraspinatus generally increased as abduction with IR increased,[19] such as performing the ‘empty can’ exercise. In contrast, the abductor moment arm of the infraspinatus generally decreased as abduction with ER increased,[19] similar to performing the ‘full can’ exercise. Otis et al.[20] reported similar findings: the abductor moment arms for the three heads of the infraspinatus (greatest in superior head and least in inferior head) were approximately 0.3–1.0 cm at 45 of ER, 0.5–1.7 cm at neutral rotation and 0.8–2.4 cm at 45 of IR. These data imply that the infraspinatus may be more effective in generating abduction torque during the ‘empty can’ compared with the ‘full can’. However, EMG data demonstrate greater infraspinatus ª 2009 Adis Data Information BV. All rights reserved.

Fig. 2. Press-up (see tables V and VI for muscle activity during this exercise).

Sports Med 2009; 39 (8)

Shoulder Muscle Activity and Function

Fig. 3. Prone external rotation at 90 abduction (see tables II and III for muscle activity during this exercise).

and VII).[12,14,16,26,35-37,43,44] When these shoulder exercises are compared with each other, mixed results have been reported. Some EMG data support prone horizontal abduction at 100 abduction with ER over the ‘empty can’ and ‘full can’ in infraspinatus activity,[35] while other EMG data and MRI data show no difference in infraspinatus activity between these exercises.[26,37] High to very high infraspinatus activity has been reported in several closed chain weight-bearing exercises, such as a variety of push-up exercises and when assuming a bipod (alternating arm and leg) position (table IV).[1,10] In contrast to the infraspinatus, the teres minor generates a weak shoulder adductor torque due to its lower attachments to the scapula and humerus.[18-20] A 3-D biomechanical model of the shoulder reveals that the teres minor does not generate scapular plane abduction torque when it contracts, but rather generates an adduction torque and 94 N of force during maximum effort scapular plane adduction.[18] In addition, Otis et al.[20] reported the adductor moment arm of the teres minor was approximately 0.2 cm at 45 IR and approximately 0.1 cm at 45 ER. These data imply that the teres minor is a weak adductor of the humerus regardless of the rotational position of the humerus. In addition, because of its posterior position at the shoulder, it also helps generate a weak horizontal abduction torque. Therefore, although its activity is similar to the infraspinatus during ª 2009 Adis Data Information BV. All rights reserved.

675

ER, it is hypothesized that the teres minor would not be as active as the infraspinatus during scaption, abduction and flexion movements, but would show activity similar to the infraspinatus during horizontal abduction type movements. This hypothesis is supported by EMG and MRI data, which show that teres minor activity during flexion, abduction and scaption is considerably less than infraspinatus activity (tables V and IX).[13,16,26,34,35,37] Even though the teres minor generates an adduction torque, it is active during humeral elevation movements as it contracts to enhance joint stability by resisting superior humeral head translation and providing humeral head compression within the glenoid fossa.[22] This is especially true at lower abduction angles and when abduction and scaption movements encounter greater resistance (table IX).[13] In contrast to arm abduction, scaption and flexion, teres minor activity is much higher during prone horizontal abduction at 100 abduction with ER, exhibiting activity similar to the infraspinatus (tables II and V).[12,16,26,35,37] Teres minor activity is also high to very high during standing high, mid and low scapular rows and standing forward scapular punch, and even

Fig. 4. Prone horizontal abduction at 90–135 abduction with external rotation (see tables II and III for muscle activity during this exercise).

Sports Med 2009; 39 (8)

Escamilla et al.

676

Fig. 5. Rowing (see tables III, V and VI for muscle activity during this exercise).

during internal rotation exercises to help stabilize the glenohumeral joint.[14] 1.3 Subscapularis

The subscapularis provides glenohumeral compression, stability, IR and abduction. From 3-D biomechanical shoulder models, predicted subscapularis force during maximum effort IR was 1725 N at 90 abduction and 1297 N at 0 abduction.[18] Its superior, middle and inferior heads generate its largest IR torque at 0 abduction, with a peak moment arm of approximately 2.5 cm.[20] As the abduction angle increases, the moment arms of the inferior and middle heads stay relatively constant, while the moment arm of the superior head progressively decreases until it is about 1.3 cm at 60 abduction.[20] These data imply that the upper portion of the subscapularis muscle (innervated by the upper subscapularis nerve) is a more effective internal rotator at lower abduction angles compared with higher abduction angles. However, there is no significant difference in upper subscapularis activity among IR exercises at 0, 45 or 90 abduction (table I).[10,45] Abduction angle does ª 2009 Adis Data Information BV. All rights reserved.

not appear to affect the ability of the lower subscapularis (innervated by the lower subscapularis nerve) to generate IR torque.[20] However, lower subscapularis muscle activity is affected by abduction angle. Decker et al.[10] reported significantly greater lower subscapularis activity with IR at 0 abduction compared with IR at 90 abduction (table I), while Kadaba et al.[45] reported greater lower subscapularis activity with IR at 90 abduction compared with IR at 0 abduction. Performing IR at 0 abduction produces similar amounts of upper and lower subscapularis activity.[10,45,46] The movement most optimal for isolation and activation of the subscapularis muscle is the Gerber lift-off against resistance,[25,46,47] which is performed by ‘lifting’ the dorsum of the hand off the mid-lumbar spine (against resistance) by simultaneously extending and internally rotating the shoulder.[48] Although this was originally developed as a test (using no resistance) for subscapularis tendon ruptures,[48] it can be used as an exercise since: (a) it tends to isolate the subscapularis muscle by minimizing pectoralis major, teres major, latissimus dorsi, supraspinatus and infraspinatus activity when performed with no resistance;[25,46] (b) it generates as much or more subscapularis activity compared with resisted IR at 0 or 90 abduction;[25,46,47] and (c) it avoids the subacromial impingement position associated with IR at 90 abduction.[25] It is important to begin the Gerber lift-off test/exercise with the hand in the mid-lumbar spine, as lower

Fig. 6. Push-up plus (see tables I, IV and VI for muscle activity during this exercise).

Sports Med 2009; 39 (8)

Shoulder Muscle Activity and Function

677

tor moment arm of the subscapularis generally increased as abduction with ER increased, such as performing the ‘full can’. This implies that the ‘full can’ may be more effective in generating subscapularis activity compared with the ‘empty can’. While most studies that have examined the ‘empty can’ exercise have reported low subscapularis activity,[26,36,37] Townsend et al.[16] reported high to very high subscapularis activity during the ‘empty can’ and low subscapularis activity during the ‘full can’ (table V). In contrast, scaption with neutral rotation as well as flexion and abduction above 120 with ER generated high to very high subscapularis EMG amplitude (tables V, VII and IX).[13,14,16] The subscapularis is active in numerous shoulder exercises other than flexion, abduction, scaption and IR exercises. High to very high subscapularis activity has been quantified in side-lying abduction, standing extension from

Fig. 7. Scaption with external rotation (full can) [see tables III, V and VI for muscle activity during this exercise].

and upper subscapularis activity decreases approximately 30% when the exercise begins at the buttocks level.[46] Performing the Gerber lift-off test produces similar amounts of upper and lower subscapularis activity.[46] The subscapularis generates significant abduction torque during humeral elevation.[19,20] From 3-D biomechanical shoulder models, predicted subscapularis force during maximum effort scaption (90 isometric position) was 283 N, approximately 2.5 times the predicted force from supraspinatus in this position.[18] Liu et al.[19] reported that in scapular plane abduction with neutral rotation the subscapularis generated a peak abductor moment arm of 1 cm at 0 abduction and then slowly decreased to 0 cm at 60 abduction. Moreover, the abductor moment arm of the subscapularis generally decreased as abduction with IR increased,[19] such as performing the ‘empty can’. In contrast, the abducª 2009 Adis Data Information BV. All rights reserved.

Fig. 8. Flexion (see tables V, VI, VII and VIII for muscle activity during this exercise).

Sports Med 2009; 39 (8)

678

Escamilla et al.

humeral head to resist superior humeral head translation,[22] but these muscles also neutralize the IR and ER torques they generate, further enhancing joint stability. 2. Deltoid Biomechanics and Function in Rehabilitation Exercises The abductor moment arms during scaption at 0 abduction with neutral rotation are approximately 0 cm for the anterior deltoid and 1.4 cm for

Fig. 9. Side-lying external rotation at 0 abduction (see tables II and V for muscle activity during this exercise).

90–0, military press, D2 diagonal pattern flexion and extension, and PNF scapular clock, depression, elevation, protraction and retraction movements (tables I, V and VII).[10,14,16,36,39,43] Even ER exercises have generated high to very high subscapularis activity (table VII) to help stabilize the glenohumeral joint.[14] Although prone horizontal abduction at 100 abduction with ER was an effective exercise for the supraspinatus, infraspinatus and teres minor, it is not an effective exercise for the subscapularis (table V).[16,37] High to very high subscapularis activity has been reported in the push-up, standing scapular dynamic hug, standing forward scapular punch, standing high, mid and low scapular rows, and two-hand overhead medicine ball throw (tables I and VII).[10,14,40,41] Otis et al.[20] reported that the superior, middle and inferior heads of the subscapularis all had abductor moment arms (greatest in the superior head and least in the inferior head) that vary as a function of humeral rotation. These moment arm lengths for the three muscle heads are approximately 0.4–2.2 cm at 45 of ER, 0.4–1.4 cm at neutral rotation and 0.4–0.5 cm at 45 of IR. This implies that the subscapularis is more effective with scaption with ER compared with scaption with IR. Moreover, the simultaneous activation of the subscapularis and infraspinatus during humeral elevation not only generate both abductor moments and inferior directed force to the ª 2009 Adis Data Information BV. All rights reserved.

Fig. 10. Standing scapular dynamic hug-forward scapular punch (see tables I, VII and VIII for muscle activity during this exercise).

Sports Med 2009; 39 (8)

Shoulder Muscle Activity and Function

the middle deltoid, and progressively increase with increasing abduction.[19,20] By 60 abduction the moment arms increase to approximately 1.5–2 cm for the anterior deltoid and 2.7–3.2 cm for the middle deltoid. From 0 to 40 abduction the moment arms for the anterior and middle deltoids are less than the moment arms for the supraspinatus, subscapularis and infraspinatus.[19,20] This implies that the anterior and middle deltoids are not effective abductors at low abduction angles (especially the anterior deltoids), while the supraspinatus, infraspinatus and subscapularis are more effective abductors at low abduction angles. These biomechanical data are supported by EMG data, in which anterior and middle deltoid activity generally peaks between 60 and 90 of scaption, while supraspinatus, infraspinatus and subscapularis activity generally peaks between 30 and 60 of scaption (table IX).[13] The abductor moment arm for the anterior deltoid changes considerably with humeral rotation, increasing with ER and decreasing with IR.[19] At 60 ER and 0 abduction, a position similar to the beginning of the ‘full can’, the anterior deltoid moment arm was 1.5 cm (compared with 0 cm in neutral rotation), which makes the anterior deltoid an effective abductor even at small abduction angles.[19] By 60 abduction with ER, its moment arm increased to approximately 2.5 cm (compared with approximately 1.5–2 cm in neutral rotation).[19] In contrast, at 60 IR at 0 abduction, a position similar to the beginning of the ‘empty can’ exercise, its moment arm was 0 cm, which implies that the anterior deltoid is not an effective abductor with humeral IR.[19] By 60 abduction and IR, its moment arm increased to only about 0.5 cm.[19] Although the abductor moment arms for the middle and posterior deltoids did change significantly with humeral rotation, the magnitude of these changes was too small to be clinically relevant. From EMG and MRI data, both the anterior and middle deltoids exhibit similar activity between the ‘empty can’ and ‘full can’ (table V).[16,26] Additional exercises that have exhibited high to very high anterior and middle deltoid activity are shown in tables IV, V and VII;[1,14,16,40,43,44,49-52] examples are D1 and D2 diagonal pattern ª 2009 Adis Data Information BV. All rights reserved.

679

flexion, flexion, push-up exercises, bench press, dumbbell fly, military press, two-hand overhead medicine ball throws, press-up, dynamic hug and standing forward scapular punch. Comparing exercises, anterior and middle deltoid activity was significantly greater performing a free weight bench press compared with a machine bench press.[52] There was no difference in mean anterior deltoid activity among the dumbbell fly and the barbell and dumbbell bench press, but both the anterior deltoid and pectoralis major were activated for longer periods in the barbell and dumbbell bench press compared with the dumbbell fly.[50] Bench press and military press technique variations also affect deltoid activity. Anterior deltoid increased as the trunk became more vertical, such as performing the incline press and military press,[51] but was less in the bench press and least in the decline press.[51] Hand grip also affects shoulder biomechanics and deltoid activity during the bench press. Compared with a narrow hand grip, employing a wider hand grip resulted in slightly greater anterior deltoid activity during the incline press and military press.[51] In contrast, compared with a wide hand grip, employing a narrow hand grip resulted in greater anterior deltoid and clavicular pectoralis activity during the decline press and bench press.[51] This is consistent with biomechanical data during the bench press, in which a greater shoulder extension torque is generated by the load lifted with a narrower (95% biacromial breadth) hand grip (peak torque of approximately 290 N m when bar was near chest) compared with a wider (270% biacromial breadth) hand grip (peak torque of approximately 210 N m when bar was near chest), which must be countered by a shoulder flexor torque generated by the shoulder flexors (primarily the anterior deltoid and clavicular pectoralis major).[53] This greater shoulder flexor torque occurred because throughout the bench press movement the load is further away from the shoulder axis with a narrower hand grip (moment arm of approximately 7 cm at starting and ending positions and approximately 21 cm when bar was near chest) compared with a wider hand grip







Sports Med 2009; 39 (8)

Escamilla et al.

680

(moment arm of approximately 4 cm at starting and ending positions and approximately 15 cm when bar was near chest).[53] Push-up technique variations also affect deltoid activity.[1,54] Anterior deltoid activity was least in a standard push-up, greater in a push-up with feet elevated and greatest in a one-arm push-up. Moreover, anterior deltoid activity was 60–70% of a MVIC during a plyometric push-up (clapping) and one-arm push-up, but only 40–50% of a MVIC during the standard push-up, push-up with hands staggered (left or right hand forward relative to other hand) and push-ups with one on both hands on a basketball.[54] These data illustrate how these exercises can be progressed in terms of increasing muscle activity. However, the plyometric and onearm push-up resulted in approximately double lumbar spinal compressive loads compared with performing standard, ball or staggered hand pushups, which may be problematic for individuals with lumbar spinal problems.[54] Moreover, these higher intensity push-up exercises result in greater loading of the glenohumeral joint resulting from greater muscle activity and greater ground reaction forces transmitted from the floor to the shoulder (plyometric push-up). The posterior deltoid does not effectively contribute to scapular plane abductor from 0–90, but more effectively functions as a scapular plane adductor due to an adductor moment arm.[19,20] Because its adductor moment arm decreases as abduction increases, this muscle becomes less effective as a scapular plane adductor at higher abduction angles, and may change to a scapular plane abductor beyond 110 abduction.[19,20] These biomechanical data are consistent with EMG and MRI data, in which posterior deltoid activity is low not only during scaption but also during flexion and abduction (tables V and IX).[13,16,26] However, high to very high posterior deltoid activity has been reported in the ‘empty can’ exercise when compared with the ‘full can’ exercise, which implies that IR during scaption increases posterior deltoid activity.[26,37] During rowing exercises and prone horizontal abduction at 100 abduction with ER and IR, both the posterior and middle deltoids produced high to very high activity (tables II ª 2009 Adis Data Information BV. All rights reserved.

and V), but low anterior deltoid activity.[12,16,37] Posterior and middle deltoid activity remain similar between IR and ER positions while performing prone horizontal abduction at 100 abduction (table V).[16] Other exercises that have exhibited high to very high posterior and middle deltoid activity include D1 diagonal pattern extension and D2 diagonal pattern flexion, pushup exercises, shoulder extension and side-lying ER at 0 abduction (tables IV and V).[1,14,16,43,44] Peak isometric abduction torque has been reported to be 25 N m at 0 abduction and neutral rotation.[19] Up to 35–65% of this torque is from the middle deltoid, up to 30% from the subscapularis, up to 25% from the supraspinatus, up to 10% from the infraspinatus, up to 2% from the anterior deltoid and 0% from the posterior deltoid.[19] This implies that both the deltoids and rotator cuff provide significant abduction torque. The ineffectiveness of the anterior and posterior deltoids to generate abduction torque may appear surprising,[19,20] but the low abduction torque for the anterior deltoid does not mean this muscle is only minimally active. In fact, because the anterior deltoid has an abductor moment arm near 0 cm at 0 abduction, the muscle could be very active and generating very high force but very little torque because of the small moment arm. At 0 abduction deltoid force attempts to translate the humeral head superiorly, which is resisted largely by the rotator cuff. Therefore, highly active deltoids may also result in a highly active rotator cuff, especially at low abduction angles during humeral elevation. The aforementioned torque data are complemented and supported by muscle force data from Hughes and An,[18] who predicted forces from the deltoids and rotator cuff during maximum effort abduction with the arm 90 abducted and in neutral rotation. Posterior deltoid and teres minor forces were only 2 N and 0 N, respectively, which further demonstrates the ineffectiveness of these muscles as shoulder abductors. In contrast, middle deltoid force was the highest, at 434 N, which supports the high activity in this muscle during abduction exercises (tables II, V, VII and IX). The anterior deltoid generated the second highest force of 323 N, which may appear




Sports Med 2009; 39 (8)

Shoulder Muscle Activity and Function

surprising given the low abductor torque for this muscle at 0 abduction. However, force and torque are not the same, and in this study by Hughes and An[18] the shoulder was positioned at 90 abduction (a position in which the deltoids are effective abductors), while in the study by Liu et al.[19] the shoulder was positioned at 0 abduction (a position in which the deltoids are not effective abductors). As previously mentioned, the moment arm of the anterior deltoid progressively increases as abduction increases. It is also important to remember that muscle force is generated not only to generate joint torque, but also to provide joint stabilization. During maximum effort abduction, Hughes and An[18] also predicted 608 N of force from the subscapularis (283 N), infraspinatus (205 N) and supraspinatus (117 N). These large forces are generated not only to abduct the shoulder but also to stabilize the glenohumeral joint and neutralize the superior directed force generated by the deltoids, especially at lower abduction angles. 3. Scapular Muscle Function in Rehabilitation Exercises Appropriate scapular muscle strength and balance is important because the scapula and humerus move together as a unit during humeral elevation, referred to as scapulohumeral rhythm. Near 30–40 of humeral elevation the scapula begins to upwardly rotate in the frontal plane, rotating approximately 1 for every 2 of humeral elevation until 120 humeral elevation, and thereafter rotating approximately 1 for every 1 humeral elevation until maximal humeral elevation, for a total of approximately 45–55 of upward rotation.[55,56] Interestingly, scapulohumeral rhythm is affected by humeral rotation. For example, it has been demonstrated that from 0–90 scapular plane abduction the scapula rotates upwardly 28–30 with neutral humeral rotation, 36–38 with humeral ER and 40–43 with IR.[57] Moreover, from 0–90 of scaption, scapular IR (winging) and anterior tilt are greater with humeral IR (‘empty can’) compared with humeral ER (‘full can’); scapular IR and anterior tilt are associated with a smaller subacromial space width, increasing impingement risk. ª 2009 Adis Data Information BV. All rights reserved.

681

During humeral elevation, in addition to scapular upward rotation, the scapula normally posterior tilts (inferior angle moving anterior in sagittal plane) approximately 20–40 and externally rotates (lateral border moves posterior in transverse plane) approximately 15–35.[55,56] If these 3-D sequences of normal scapular movements are disrupted by abnormal scapular muscle firing patterns, weakness, fatigue or injury, the shoulder complex functions less efficiently and injury risk is increased. The primary muscles that cause and control scapular movements include the trapezius, serratus anterior, levator scapulae, rhomboids and pectoralis minor. The function of these muscles during shoulder exercises is discussed below. 3.1 Serratus Anterior

The serratus anterior works with the pectoralis minor to abduct (protract) the scapula and with the upper and lower trapezius to upwardly rotate the scapula. The serratus anterior is an important muscle because it contributes to all components of normal 3-D scapular movements during humeral elevation, which includes upward rotation, posterior tilt and external rotation.[55,56] The serratus anterior also helps stabilize the medial border and inferior angle of the scapula, preventing scapular IR (winging) and anterior tilt. Tables III, VI and VIII show several exercises that elicit high to very high serratus anterior activity, such as D1 and D2 diagonal pattern flexion, D2 diagonal pattern extension, supine scapular protraction, supine upward scapular punch, military press, IR and ER at 90 abduction, flexion, abduction, scaption above 120 with ER, and push-up plus.[11,14,15,41,49] Serratus anterior activity tends to increase in a somewhat linear fashion with humeral elevation (tables III and VI).[11,15,56,58,59] However, increasing humeral elevation increases subacromial impingement risk,[27,28] and humeral elevation at lower abduction angles also generates high to very high serratus anterior activity (table III).[11] It is interesting that performing IR and ER at 90 abduction generates high to very high serratus anterior activity (tables III and VIII), because these Sports Med 2009; 39 (8)

Escamilla et al.

682

exercises are usually thought to primarily work rotator cuff muscles.[11,14] Not surprising is the high activity generated during the push-up. When performing the standard push-up, push-up on knees and wall push-up, serratus activity is greater when full scapular protraction occurs after the elbows fully extend (push-up plus).[60] Moreover, serratus anterior activity was lowest in the wall push-up plus, exhibited moderate activity during the push-up plus on knees, and high to very high activity during the standard push-up plus and push-up plus with the feet elevated (greatest activity with feet elevated)[49,60,61] – which illustrates how these exercises can be progressed. Additional exercises that have been shown to be effective in activating the serratus anterior is the standing scapular dynamic hug,[49] PNF scapular depression and protraction movements,[39] ‘empty can’,[37] and the wall slide.[59] The wall slide begins by slightly leaning against the wall with the ulnar border of forearms in contact with wall, elbows flexed 90 and shoulders abducted 90 in the scapular plane. From this position the arms slide up the wall in the scapular plane while leaning into the wall. The wall slide produces similar serratus anterior activity compared with scaption above 120 with no resistance. One advantage of the wall slide compared with scaption is that, anecdotally, patients report that the wall slide is less painful to perform.[59] This may be because during the wall slide the upper extremities are supported against the wall in a closed chain position, making it easier to perform. 3.2 Trapezius

General functions of the trapezius include scapular upward rotation and elevation for the upper trapezius, retraction for the middle trapezius, and upward rotation and depression for the lower trapezius. In addition, the inferomedialdirected fibres of the lower trapezius may also contribute to posterior tilt and external rotation of the scapula during humeral elevation,[56] which decreases subacromial impingement risk.[61,62] Tables III, VI and VIII show several exercises that elicit high to very high trapezius activity, such as shoulder shrug, prone rowing, prone ª 2009 Adis Data Information BV. All rights reserved.

horizontal abduction at 90 and 135 abduction with ER and IR, D1 diagonal pattern flexion, standing scapular dynamic hug, PNF scapular clock, military press, two-hand overhead medicine ball throw, and scaption and abduction below 80, at 90 and above 120 with ER.[11,15,39,40,49] During scaption, upper trapezius activity progressively increases from 0 to 60, remains relatively constant from 60 to 120 and continues to progressively increase from 120 to 180.[58] High to very high middle trapezius activity occurs in the shoulder shrug, prone rowing and prone horizontal abduction at 90 and 135 abduction with ER and IR.[11,15] Some studies have reported high to very high middle trapezius activity during scaption at 90 and above 120,[11,49,58] while other EMG data show low middle trapezius activity during this exercise.[15] High to very high lower trapezius activity occurs in the prone rowing, prone horizontal abduction at 90 and 135, abduction with ER and IR, prone and standing ER at 90 abduction, D2 diagonal pattern flexion and extension, PNF scapular clock, standing high scapular rows, and scaption, flexion and abduction below 80 and above 120 with ER.[11,14,15,39] Lower trapezius activity tends to be low at <90 of scaption, abduction and flexion, and then increases exponentially from 90 to 180.[11,15,39,58,59,63] Significantly greater lower trapezius activity has been reported in prone ER at 90 abduction compared with the ‘empty can’ exercise.[34] 3.3 Rhomboids and Levator Scapulae

Both the rhomboids and levator scapulae function as scapular adductors (retractors), downward rotators and elevators. High to very high rhomboid activity has been reported during D2 diagonal pattern flexion and extension, standing ER at 0 and 90 abduction, standing IR at 90 abduction, standing extension from 90 to 0, prone horizontal abduction at 90 abduction with IR, scaption, abduction and flexion above 120 with ER, prone rowing, and standing high, mid and low scapular rows (tables VI and VIII).[14,15] High to very high levator scapulae activity has been reported in Sports Med 2009; 39 (8)

Shoulder Muscle Activity and Function

scaption above 120 with ER, prone horizontal abduction at 90 abduction with ER and IR, prone rowing, and prone extension at 90 flexion (table VI).[15] 4. Conclusions During shoulder exercises the rotator cuff abducts, externally rotates and internally rotates, and stabilizes the glenohumeral joint. Although the infraspinatus and subscapularis generate muscle forces two to three times greater than the supraspinatus force, the supraspinatus still remains a more effective shoulder abductor because of its more effective moment arm. Both the deltoids and rotator cuff provide significant abduction torque, with an estimated contribution up to 35–65% by the middle deltoid, 30% by the subscapularis, 25% by the supraspinatus, 10% by the infraspinatus and 2% by the anterior deltoid. During abduction, middle deltoid force has been estimated to be 434 N, followed by 323 N from the anterior deltoid, 283 N from the subscapularis, 205 N from the infraspinatus and 117 N from the supraspinatus. These forces are generated not only to abduct the shoulder but also to stabilize the joint and neutralize the antagonistic effects of undesirable actions. Relatively high force from the rotator cuff not only helps abduct the shoulder but also neutralizes the superior directed force generated by the deltoids at lower abduction angles. Even though anterior deltoid force is relatively high, its ability to abduct the shoulder is low due to a very small moment arm, especially at low abduction angles. The deltoids are more effective abductors at higher abduction angles, while the rotator cuff muscles are more effective abductors at lower abduction angles. During maximum humeral elevation the scapula normally upwardly rotates 45–55, posterior tilts 20–40 and externally rotates 15–35. The scapular muscles are important during humeral elevation because they cause these motions, especially the serratus anterior, which contributes to scapular upward rotation, posterior tilt and ER. The serratus anterior also helps stabilize the medial border and inferior angle of the scapula, ª 2009 Adis Data Information BV. All rights reserved.

683

preventing scapular IR (winging) and anterior tilt. If normal scapular movements are disrupted by abnormal scapular muscle firing patterns, weakness, fatigue or injury, the shoulder complex functions less efficiently and injury risk increases. Scapula position and humeral rotation can affect injury risk during humeral elevation. Compared with scapular protraction, scapular retraction has been shown to both increase subacromial space width and enhance supraspinatus force production during humeral elevation. Moreover, scapular IR (winging) and anterior tilt, both of which decrease subacromial space width and increase impingement risk, are greater when performing the ‘empty can’ compared with the ‘full can’. There are several exercises in the literature that exhibit high to very high activity from the rotator cuff, deltoids and scapular muscles, such as prone horizontal abduction at 100 abduction with ER, flexion, abduction and scaption with ER, D1 and D2 diagonal pattern flexion and extension, ER and IR at 0 and 90 abduction, standing extension from 90 to 0, a variety of weight-bearing upper extremity exercises (such as the push-up), standing scapular dynamic hug, forward scapular punch and rowing exercises. Supraspinatus activity in the ‘empty can’ and ‘full can’ is similar, although the ‘full can’ results in less risk of subacromial impingement. Infraspinatus and subscapularis activity have generally been reported to be higher in the ‘full can’ compared with the ‘empty can’, while posterior deltoid activity has been reported to be higher in the ‘empty can’ than the ‘full can’. Acknowledgements No sources of funding were used to assist in the preparation of this review. The authors have no conflicts of interest that are directly relevant to the content of this review.

References 1. Uhl TL, Carver TJ, Mattacola CG, et al. Shoulder musculature activation during upper extremity weight-bearing exercise. J Orthop Sports Phys Ther 2003; 33 (3): 109-17 2. DiGiovine NM, Jobe FW, Pink M, et al. Electromyography of upper extremity in pitching. J Shoulder Elbow Surg 1992; 1: 15-25

Sports Med 2009; 39 (8)

684

3. Escamilla RF, Fleisig GS, Zheng N, et al. Biomechanics of the knee during closed kinetic chain and open kinetic chain exercises. Med Sci Sports Exerc 1998; 30 (4): 556-69 4. Zheng N, Fleisig GS, Escamilla RF, et al. An analytical model of the knee for estimation of internal forces during exercise. J Biomech 1998; 31 (10): 963-7 5. Lawrence JH, De Luca CJ. Myoelectric signal versus force relationship in different human muscles. J Appl Physiol 1983; 54 (6): 1653-9 6. Solomonow M, Baratta R, Zhou BH, et al. Historical update and new developments on the EMG-force relationships of skeletal muscles. Orthopedics 1986; 9 (11): 1541-3 7. Perry J, Bekey GA. EMG-force relationships in skeletal muscle. Crit Rev Biomed Eng 1981; 7 (1): 1-22 8. Bigland-Ritchie B. EMG/force relations and fatigue of human voluntary contractions. Exerc Sport Sci Rev 1981; 9: 75-117 9. Thomas CK, Johansson RS, Bigland-Ritchie B. EMG changes in human thenar motor units with force potentiation and fatigue. J Neurophysiol 2006; 95 (3): 1518-26 10. Decker MJ, Tokish JM, Ellis HB, et al. Subscapularis muscle activity during selected rehabilitation exercises. Am J Sports Med 2003; 31 (1): 126-34 11. Ekstrom RA, Donatelli RA, Soderberg GL. Surface electromyographic analysis of exercises for the trapezius and serratus anterior muscles. J Orthop Sports Phys Ther 2003; 33 (5): 247-58 12. Reinold MM, Wilk KE, Fleisig GS, et al. Electromyographic analysis of the rotator cuff and deltoid musculature during common shoulder external rotation exercises. J Orthop Sports Phys Ther 2004; 34 (7): 385-94 13. Alpert SW, Pink MM, Jobe FW, et al. Electromyographic analysis of deltoid and rotator cuff function under varying loads and speeds. J Shoulder Elbow Surg 2000; 9 (1): 47-58 14. Meyers JB, Pasquale MR, Laudner KG, et al. On-the-field resistance-tubing exercises for throwers: an electromyographic analysis. J Athletic Training 2005; 40 (1): 15-22 15. Moseley Jr JB, Jobe FW, Pink M, et al. EMG analysis of the scapular muscles during a shoulder rehabilitation program. Am J Sports Med 1992; 20 (2): 128-34 16. Townsend H, Jobe FW, Pink M, et al. Electromyographic analysis of the glenohumeral muscles during a baseball rehabilitation program. Am J Sports Med 1991; 19 (3): 264-72 17. Lee SB, Kim KJ, O’Driscoll SW, et al. Dynamic glenohumeral stability provided by the rotator cuff muscles in the mid-range and end-range of motion: a study in cadavera. J Bone Joint Surg Am 2000; 82 (6): 849-57 18. Hughes RE, An KN. Force analysis of rotator cuff muscles. Clin Orthop Relat Res 1996; (330): 75-83 19. Liu J, Hughes RE, Smutz WP, et al. Roles of deltoid and rotator cuff muscles in shoulder elevation. Clin Biomech (Bristol, Avon) 1997; 12 (1): 32-8 20. Otis JC, Jiang CC, Wickiewicz TL, et al. Changes in the moment arms of the rotator cuff and deltoid muscles with abduction and rotation. J Bone Joint Surg Am 1994; 76 (5): 667-76

ª 2009 Adis Data Information BV. All rights reserved.

Escamilla et al.

21. Burke WS, Vangsness CT, Powers CM. Strengthening the supraspinatus: a clinical and biomechanical review. Clin Orthop Relat Res 2002; (402): 292-8 22. Sharkey NA, Marder RA. The rotator cuff opposes superior translation of the humeral head. Am J Sports Med 1995; 23 (3): 270-5 23. Jobe FW, Moynes DR. Delineation of diagnostic criteria and a rehabilitation program for rotator cuff injuries. Am J Sports Med 1982; 10 (6): 336-9 24. Rowlands LK, Wertsch JJ, Primack SJ, et al. Kinesiology of the empty can test. Am J Phys Med Rehabil 1995; 74 (4): 302-4 25. Kelly BT, Kadrmas WR, Speer KP. The manual muscle examination for rotator cuff strength: an electromyographic investigation. Am J Sports Med 1996; 24 (5): 581-8 26. Takeda Y, Kashiwaguchi S, Endo K, et al. The most effective exercise for strengthening the supraspinatus muscle: evaluation by magnetic resonance imaging. Am J Sports Med 2002; 30 (3): 374-81 27. De Wilde L, Plasschaert F, Berghs B, et al. Quantified measurement of subacromial impingement. J Shoulder Elbow Surg 2003; 12 (4): 346-9 28. Roberts CS, Davila JN, Hushek SG, et al. Magnetic resonance imaging analysis of the subacromial space in the impingement sign positions. J Shoulder Elbow Surg 2002; 11 (6): 595-9 29. Thigpen CA, Padua DA, Morgan N, et al. Scapular kinematics during supraspinatus rehabilitation exercise: a comparison of full-can versus empty-can techniques. Am J Sports Med 2006; 34 (4): 644-52 30. Smith J, Dietrich CT, Kotajarvi BR, et al. The effect of scapular protraction on isometric shoulder rotation strength in normal subjects. J Shoulder Elbow Surg 2006; 15 (3): 339-43 31. Solem-Bertoft E, Thuomas KA, Westerberg CE. The influence of scapular retraction and protraction on the width of the subacromial space: an MRI study. Clin Orthop Relat Res 1993; (296): 99-103 32. Kibler WB, Sciascia A, Dome D. Evaluation of apparent and absolute supraspinatus strength in patients with shoulder injury using the scapular retraction test. Am J Sports Med 2006; 34 (10): 1643-7 33. Itoi E, Kido T, Sano A, et al. Which is more useful, the ‘‘full can test’’ or the ‘‘empty can test,’’ in detecting the torn supraspinatus tendon? Am J Sports Med 1999; 27 (1): 65-8 34. Ballantyne BT, O’Hare SJ, Paschall JL, et al. Electromyographic activity of selected shoulder muscles in commonly used therapeutic exercises. Phys Ther 1993; 73 (10): 668-77; discussion 677-82 35. Blackburn TA, McLeod WD, White B, et al. EMG analysis of posterior rotator cuff exercises. Athletic Training 1990; 25 (1): 40-5 36. Horrigan JM, Shellock FG, Mink JH, et al. Magnetic resonance imaging evaluation of muscle usage associated with three exercises for rotator cuff rehabilitation. Med Sci Sports Exerc 1999; 31 (10): 1361-6 37. Malanga GA, Jenp YN, Growney ES, et al. EMG analysis of shoulder positioning in testing and strengthening the supraspinatus. Med Sci Sports Exerc 1996; 28 (6): 661-4

Sports Med 2009; 39 (8)

Shoulder Muscle Activity and Function

38. Worrell TW, Corey BJ, York SL, et al. An analysis of supraspinatus EMG activity and shoulder isometric force development. Med Sci Sports Exerc 1992; 24 (7): 744-8 39. Smith J, Dahm DL, Kaufman KR, et al. Electromyographic activity in the immobilized shoulder girdle musculature during scapulothoracic exercises. Arch Phys Med Rehabil 2006; 87 (7): 923-7 40. Cordasco FA, Wolfe IN, Wootten ME, et al. An electromyographic analysis of the shoulder during a medicine ball rehabilitation program. Am J Sports Med 1996; 24 (3): 386-92 41. Hintermeister RA, Lange GW, Schultheis JM, et al. Electromyographic activity and applied load during shoulder rehabilitation exercises using elastic resistance. Am J Sports Med 1998; 26 (2): 210-20 42. Itoi E, Berglund LJ, Grabowski JJ, et al. Tensile properties of the supraspinatus tendon. J Orthop Res 1995; 13 (4): 578-84 43. Kronberg M, Nemeth G, Brostrom LA. Muscle activity and coordination in the normal shoulder: an electromyographic study. Clin Orthop Relat Res 1990; (257): 76-85 44. Ekholm J, Arborelius UP, Hillered L, et al. Shoulder muscle EMG and resisting moment during diagonal exercise movements resisted by weight-and-pulley-circuit. Scand J Rehabil Med 1978; 10 (4): 179-85 45. Kadaba MP, Cole A, Wootten ME, et al. Intramuscular wire electromyography of the subscapularis. J Orthop Res 1992; 10 (3): 394-7 46. Greis PE, Kuhn JE, Schultheis J, et al. Validation of the liftoff test and analysis of subscapularis activity during maximal internal rotation. Am J Sports Med 1996; 24 (5): 589-93 47. Suenaga N, Minami A, Fujisawa H. Electromyographic analysis of internal rotational motion of the shoulder in various arm positions. J Shoulder Elbow Surg 2003; 12 (5): 501-5 48. Gerber C, Krushell RJ. Isolated rupture of the tendon of the subscapularis muscle: clinical features in 16 cases. J Bone Joint Surg Br 1991; 73 (3): 389-94 49. Decker MJ, Hintermeister RA, Faber KJ, et al. Serratus anterior muscle activity during selected rehabilitation exercises. Am J Sports Med 1999; 27 (6): 784-91 50. Welsch EA, Bird M, Mayhew JL. Electromyographic activity of the pectoralis major and anterior deltoid muscles during three upper-body lifts. J Strength Cond Res 2005; 19 (2): 449-52 51. Barnett C, Kippers V, Turner P. Effects of variations of the bench press exercise on the EMG activity of five shoulder muscles. J Strength Cond Res 1995; 9 (4): 222-7

ª 2009 Adis Data Information BV. All rights reserved.

685

52. McCaw ST, Friday JJ. A comparison of muscle activity between a free weight and machine bench press. J Strength Cond Res 1994; 8 (4): 259-64 53. Wagner LL, Evans SA, Weir JP, et al. The effect of grip width on bench press performance. Int J Sport Biomech 1992; 8: 1-10 54. Freeman S, Karpowicz A, Gray J, et al. Quantifying muscle patterns and spine load during various forms of the pushup. Med Sci Sports Exerc 2006; 38 (3): 570-7 55. McClure PW, Michener LA, Sennett BJ, et al. Direct 3dimensional measurement of scapular kinematics during dynamic movements in vivo. J Shoulder Elbow Surg 2001; 10 (3): 269-77 56. Ludewig PM, Cook TM, Nawoczenski DA. Three-dimensional scapular orientation and muscle activity at selected positions of humeral elevation. J Orthop Sports Phys Ther 1996; 24 (2): 57-65 57. Sagano M, Magee D, Katayose M. The effect of glenohumeral rotation on scapular upward rotation in different positions of scapular-plane elevation. J Sport Rehab 2006; 15: 144-55 58. Bagg SD, Forrest WJ. Electromyographic study of the scapular rotators during arm abduction in the scapular plane. Am J Phys Med 1986; 65 (3): 111-24 59. Hardwick DH, Beebe JA, McDonnell MK, et al. A comparison of serratus anterior muscle activation during a wall slide exercise and other traditional exercises. J Orthop Sports Phys Ther 2006; 36 (12): 903-10 60. Ludewig PM, Hoff MS, Osowski EE, et al. Relative balance of serratus anterior and upper trapezius muscle activity during push-up exercises. Am J Sports Med 2004; 32 (2): 484-93 61. Graichen H, Bonel H, Stammberger T, et al. Three-dimensional analysis of the width of the subacromial space in healthy subjects and patients with impingement syndrome. AJR Am J Roentgenol 1999; 172 (4): 1081-6 62. Ludewig PM, Cook TM. Alterations in shoulder kinematics and associated muscle activity in people with symptoms of shoulder impingement. Phys Ther 2000; 80 (3): 276-91 63. Wiedenbauer MM, Mortensen OA. An electromyographic study of the trapezius muscle. Am J Phys Med 1952; 31 (5): 363-72

Correspondence: Prof. Rafael Escamilla, Department of Physical Therapy, California State University, Sacramento, 6000 J Street, Sacramento, CA 95819-6020, USA. E-mail: [email protected]

Sports Med 2009; 39 (8)

Sports Med 2009; 39 (8): 687-695 0112-1642/09/0008-0687/$49.95/0

RESEARCH REVIEW

ª 2009 Adis Data Information BV. All rights reserved.

Endurance Athletes’ Stroke Volume Response to Progressive Exercise A Critical Review Thomas Rowland Baystate Medical Center, Springfield, Massachusetts, USA

Abstract

Untrained subjects typically demonstrate a plateau in stroke volume beyond the early stages of an acute bout of progressive exercise. Some studies have indicated that the stroke volume pattern in highly trained endurance athletes may differ, continuing to rise progressively to the point of maximal exercise. This suggests that the mechanism for generating cardiac output in these athletes may be influenced by cardiac functional factors, particularly augmented diastolic filling. This review of studies assessing stroke volume changes with exercise in athletes reveals a wide disparity of reported patterns, ranging from those indicating a decline, to stable values (a true plateau), to a progressive increase to exhaustion of over 40%. Differences in testing methodology might help to explain these variable results, but which pattern truly reflects that expected in highly trained endurance athletes remains uncertain. Ancillary data suggest that if a non-plateau of stroke volume is typical of athletes, augmented diastolic filling might be the responsible mechanism. However, it is not clear whether this might reflect superior upstream (atrial pressure/volume) or downstream (ventricular diastolic function, compliance) factors.

During an acute bout of upright progressive exercise in a healthy, untrained person, stroke volume typically rises by 20–30% at low work intensities, then changes little (or ‘plateaus’) to the point of subject exhaustion.[1-3] The initial increase presumably reflects mobilization of blood sequestered by gravity in the lower extremities upon assuming the upright position,[4] as evidenced by failure to observe such a rise when exercise conditions are independent of gravity (supine cycling,[5,6] arm exercise,[7] weightlessness in space,[8] swimming[9] and aquatic exercise[10]). Other than during this initial phase of ventricular refilling, then, it can be as-

sumed that stroke volume does not contribute to the rise in cardiac output during progressive upright exercise. (Some have contended that stroke volume may actually decrease in the last seconds of an exhaustive test.[11,12]) As work intensity increases, left ventricular enddiastolic dimension (LVED) in healthy subjects remains stable or slightly declines.[13,14] This more or less constant ventricular filling volume (preload) during progressive exercise is achieved as heart rate rises to match increases in systemic venous return (Bainbridge reflex).[15] Systolic and diastolic function improve in parallel as exercise intensity increases.[16] These changes in inotropic and

Rowland

688

lusitropic function serve to maintain a constant stroke volume and ventricular filling volume, respectively, as time periods of systolic ejection and diastolic filling shorten with increasing heart rate. This model of cardiovascular response, derived from studies initially performed in the 1950s,[17] has been supported by more contemporary techniques.[2] It indicates a primary role of peripheral, non-cardiac factors (particularly fall in peripheral vascular resistance from arteriolar dilatation triggered by local metabolic regulators) in facilitating and controlling blood circulation with exercise. In this schema, the heart is relegated to a secondary role as a force-feed pump, its dynamics dictated by the magnitude of systemic venous return. Highly trained endurance athletes are characterized by their superior cardiac functional capacity – higher maximal stroke volume and cardiac output – compared with non-athletes. Athletes have been found in some studies to exhibit the same pattern of stroke volume response to progressive exercise as non-athletes (a plateau after an early rise), their greater maximal values being accounted for by a shift upwards of the work intensity-stroke volume curve.[18,19] By this interpretation, functional characteristics of the heart are no different in endurance athletes than in non-athletes. The higher maximal stroke volume in athletes is accounted for by a greater resting stroke volume, which in turn reflects a larger resting LVED, part of an overall expansion of the cardiovascular system. By this schema, factors influencing resting LVED (such as plasma volume and bradycardia) are responsible for the higher maximal stroke volume . (and, by extension, maximal cardiac output [ Q] and aerobic power . [VO2]) of trained individuals. In this model, then, variations in cardiac responses to exercise between trained and untrained persons are quantitative rather than qualitative, related to differences in heart dimensions rather than ventricular function. Findings in other studies have challenged this concept, indicating a continuous rise in stroke volume, without plateau, during progressive upright exercise in highly trained endurance athletes.[20-27] This information suggests that the cardiovascular dynamics during exercise may differ between such athletes and non-athletes, and that the mechanics of generating a superior ª 2009 Adis Data Information BV. All rights reserved.

stroke volume and cardiac output can be altered by the training state. Specifically, such data imply that functional rather than simply dimensional features of the heart (specifically, superior diastolic and/or systolic function) during exercise may serve to differentiate cardiovascular capacity in endurance athletes from that of non-athletes. Such a conclusion would require re-thinking the relative roles of peripheral versus central (i.e. cardiac) determinants of cardiovascular responses to exercise in endurance athletes. In untrained individuals, the research data accumulated over the last 100 years indicates that ‘‘the heart has relatively little effect on the normal regulation of cardiac output’’.[17] However, confirming functional changes during exercise as important in generating increases in stroke volume in endurance athletes would indicate that circulatory responses to the metabolic demands of exercise in trained individuals are controlled by central (i.e. cardiac) as well as peripheral factors. An accurate understanding of the pattern of stroke volume response to progressive exercise in trained individuals is obviously critical to such insights. Unfortunately, such a differentiation of stroke volume dynamics in endurance athletes and non-athletes is far from clear, as the extant research literature is replete with conflicting information. It is the purpose of this article to review these investigations, with the goals of (i) providing an overview of current research data; (ii) attempting to delineate factors that might be responsible for differences in findings of stroke volume plateau versus non-plateau in testing of endurance athletes; and (iii) considering candidate mechanisms that might be responsible for possible stroke volume non-plateau in highly trained endurance athletes. The reader is forewarned that, based on the current literature, a conclusion regarding the veracity of a non-plateau pattern of stroke volume response as characteristic of endurance athletes will unfortunately not be forthcoming. 1. Methods Studies describing stroke volume responses to progressive exercise in highly trained endurance Sports Med 2009; 39 (8)

Stroke Volume in Trained Subjects

athletes were culled from multiple sources, including a computer literature search, personal literature files, and reference lists from published articles. The following were inclusive criteria for this literature evaluation: (i) endurance .athletes, with . an average group maximal VO2 (VO2max) value >60 mL/kg/min (adult men) or >55 mL/kg/min (adult women and young adolescents), or, if this value was not provided, a group in which such values would be expected; (ii) young athletes (<35 years of age); (iii) upright cycle or treadmill exercise; (iv) stroke volume estimated during (not after) maximal or near maximal exercise (average peak heart rate >175 beats/min); and (v) resting as well as at least two submaximal values of stroke volume. Studies in which stroke volume was estimated by non-conventional methodology were not included. The variability in study design and presentation of findings in these studies precluded any statistical analysis by which the patterns of stroke volume responses could be differentiated. As expected, early refilling rise in stroke volume with upright exercise was observed equally in athletes and nonathletes. For the purposes of this review, then, change in stroke volume for each study was considered as the percent change from a point determined visually from early-mid exercise intensities (approximately heart rate 120 beats/min or 30–50% . VO2max) to maximal (or near-maximal) exercise. In an effort to discern factors that might be responsible for the variability in reported change in stroke volume, studies were divided into two groups – those indicating a stroke volume plateau, defined arbitrarily as <20% increase (equivalent to a change in stroke index of approximately 10–15 mL/m2), and those demonstrating >20% rise in mean stroke volume value (the non-plateau group). 2. Results and Discussion A total of 21 studies were identified that met the inclusion criteria. Of these, 13 reported a change in stroke volume <20%, while eight described an increase >20% (table I). The overall average change in stroke volume among these reports was +14%. Although the magnitude of stroke volume change was dichotomized for this ª 2009 Adis Data Information BV. All rights reserved.

689

review, it is apparent from table I that there exists a continuum of stroke volume responses in these reports, ranging from two in which stroke volume declined, seven in which little or no change was observed (a true plateau), to two describing a rise exceeding +40%. The present body of research information thus precludes any confident conclusion regarding characteristic stroke volume responses in endurance athletes versus those of non-athletes. In assessing factors that might explain the marked variability in these results, the information in table I provides no obvious differentiating influence of level of aerobic fitness, maximal heart rate, sport, sex, age or date of publication. However, certain observations regarding the potential influences of methodology and testing protocol may be pertinent. 2.1 Methodology and Protocols

Six of the eight studies indicating a nonplateau (>20% rise) in stroke volume utilized the acetylene rebreathing technique, and five of these were performed in the same laboratory using a nonconventional supramaximal testing protocol.[21-24,26] In this method, following a standard progressive test to volitional exhaustion, subjects were allowed a minute of rest before pedaling again at a supramaximal work load, during which measurement of cardiac output, heart rate and . VO2 were defined as ‘maximum’. The extent to which this procedure might influence cardiac dynamics and stroke volume calculation at peak exercise (cardiac output/heart rate) is not clear. For example, Huonker et al.[37] suggested that the rise in stroke volume in the terminal stages of progressive exercise in athletes might reflect recruitment of additional muscle groups (as would be expected during supramaximal exercise), with augmented skeletal muscle pump function, levels of systemic venous return and cardiac filling. It may be argued that this approach ensures a true maximal stress. However, two observations suggest that the nonconventional nature of this protocol might be problematic. First, the final stroke volume point in these studies created an upward deviation of the trend observed during Sports Med 2009; 39 (8)

Rowland

690

Table I. Features of reports of pattern of stroke volume response during progressive exercise in trained endurance athletes (see text) . Study No. Age (y) sex Sport Testing HRmax %DSV VO2max Studies showing DSV <20% Ginzton et al.[28]

15

19 M F

DR

C Echo

187

-26

Ogawa et al.[29]

15

28 M

ET

T Acet

64

178

-9

Rivera et al.[30]

11

32 M

DR

T Acet

70

185

0

Rowland et al.[31]

7

12 M

Cyc

C Dopp

60

184

0

Rowland and Roti[19]

8

31 M

Cyc

C Dopp

74

183

0 0

Nottin et al.[18]

10

11 M

Cyc

C Dopp

59

195

Roti and Rowland (U)

8

34 F

Cyc

C Dopp

60

178

0

Rowland et al.[32]

8

14 M

Cyc

C Dopp

198

+1

Zhou et al.[27]

10

26 M

DR

T Acet

73

187

+3

Vanfraechem[33]a

17

21 M

Socc

C Imp

65

175

+10

Ekblom and Hermansen[34]

70

13

29 M

ET

T Dye

192

+10

Ahmad and Dubiel[35]

9

24 M

DR

C Rad

182

+16

Rowland et al.[36]

8

11 M

DR

C Dopp

190

+17

Studies showing DSV >20% Gledhill et al.[22]

7

23 M

Cyc

C Acet

69

190

+23

Warburton et al.[25]a

10

26 M

Cyc

C Rad

68

184

+24

Krip et al.[23]b

12

24 M

Cyc

C Acet

64

187

+30

Crawford et al.[20]b

12

6M6F

DR

C Echo

177

+30

Wiebe et al.[26]a

6

25 F

ET

C Acet

64

192

+35

Zhou et al.[27]

5

30 M

DR

T Acet

84

180

+39

Warburton et al.[24]a

9

22 M

Cyc

C Acet

69

194

+44

Ferguson et al.[21]b

7

23 F

ET

C Acet

64

200

+44

a

No untrained controls.

b

DSV same as controls.

%DSV = average percent change in stroke volume from low-mid intensity to maximal exercise; Acet = acetylene rebreathing; C = cycle ergometer; Cyc = cyclists; DR = distance runners; Dopp = Doppler ultrasound; Dye = dye dilution; Echo = two-dimensional ultrasound; Imp = thoracic bioimpedance; M = male; Rad = radioET = endurance trained; F = female; HRmax = average maximal heart rate (beats/min); . nuclide angiography; Socc = soccer players; T = treadmill; U = unpublished study; VO2max = average maximal aerobic power (mL/kg/min).

submaximal exercise, and in fact, in three of these reports the final ‘maximal’ value established the stroke volume pattern as a ‘non-plateau’ (figure 1). Secondly, in two of the three studies using this protocol in which untrained control subjects were included, a similar non-plateau of stroke volume response was observed in both trained and untrained subjects.[21,23] Five of the six studies employing Doppler ultrasound have indicated a strict plateau in stroke volume in athletes. Stroke volume values during exercise by this technique are estimated as the product of aortic flow velocity and resting aortic outflow cross-sectional area. It assumes a constant aortic outflow tract diameter with increasing ª 2009 Adis Data Information BV. All rights reserved.

work intensity, which is true in non-athletes[38] but has not been examined in highly trained individuals. Kasikcioglu et al.[39] reported that aortic distensibility, measured at rest 3 cm above the aortic valve, was greater in athletes than in non-athletes, raising the possibility that aortic outflow diameter might increase with exercise in this group. If so, traditional Doppler techniques would underestimate stroke volume at high exercise intensities in athletes. Also, three of these reports involved young adolescent athletes who, although highly trained, had experienced a limited duration of sports participation. The extent that differences in testing protocol and measurement technique are responsible for Sports Med 2009; 39 (8)

Stroke Volume in Trained Subjects

691

the variation observed in stroke volume pattern during progressive exercise in athletes must be considered. However, as these factors appear to be the only ones that might differentiate a plateau from a non-plateau in these investigations, future studies should be conducted with attention to these possible influences. 2.2 Indirect Evidence

A unique rise of stroke volume in endurance athletes compared with non-athletes would need to be achieved by a progressive rise in ventricular preload and/or augmented myocardial contractility (in response to an exaggerated fall in systemic vascular resistance and/or increased sympathetic stimulation) during exercise. It is instructive, then, to examine findings in studies that have assessed systolic and diastolic function as well as changes in ventricular dimensions during progressive exercise in endurance athletes. Any differences in these factors between trained and untrained individuals would (i) support the existence of unique stroke volume response patterns in athletes, and (ii) provide insights into mechanisms responsible for a progressive rise in stroke volume in trained subjects. 2.2.1 Systolic Function

Reported measures of ventricular systolic function at rest have generally been comparable 200

A

Stroke volume (mL)

180 B

160 140

C

120 100 80 60 40 20 60

80

100 120 140 160 Heart rate (beats/min)

180

200

Fig. 1. Pattern of stroke volume response to progressive exercise in three studies utilizing a supra-maximal protocol (see text). A = Gledhill et al.,[22] B = Krip et al.,[23] C = Warburton et al.[25]

ª 2009 Adis Data Information BV. All rights reserved.

in trained and untrained subjects.[40,41] Also, with few exceptions,[42,43] exercise studies have consistently demonstrated equal rises in ventricular ejection fraction or shortening fraction (gross markers of systolic contractile function) in athletes compared with non-athletes.[28,32,35,44-46] Similarly, endurance athletes and non-athletes have typically demonstrated similar declines in systemic vascular resistance (which influences ventricular after-load, wall stress and contractile function) during progressive exercise.[29,43,47-49] These data, then, imply no significant differences in systolic function at rest or during exercise in athletes compared with non-athletes. 2.2.2 Ventricular Diastolic Filling

Studies that have assessed LVED during exercise by radionuclide angiography or twodimensional echocardiography in endurance athletes and non-athletes have provided conflicting results. Some have indicated no group differences,[28,32,48,49] while others have indicated a rise in LVED in athletes as work load increases compared with the flat profile typical of nonathletes.[20,25,35,45,49] This research information would suggest, therefore, if a progressive rise in stroke volume is typical of highly trained endurance athletes, that such a pattern reflects a progressive increase in ventricular diastolic filling volume (pre-load), with increases of stroke volume created by a Frank-Starling mechanism. Exaggerated increases in filling volume would indicate a progressively greater rise in the atrialventricular pressure gradient during exercise in athletes versus non-athletes. Such a gradient is an expression of the balance between upstream factors (atrial pressure, volume and contractile force) and downstream factors (rate of ventricular relaxation, myocardial compliance and ventricular suction). In non-trained individuals, the atrial-ventricular gradient rises approximately 4-fold during a maximal progressive test and is generated by downstream factors, particularly improvements in diastolic function (rate of myocardial relaxation).[16,50,51] The contribution of upstream factors is precluded by the heart rate response matching systemic venous return, Sports Med 2009; 39 (8)

Rowland

692

which prevents a significant rise in left atrial pressure or volume. A number of recent studies have indicated greater transmitral flow rates at supine rest in athletes, based on higher early (E wave) and lower late (A wave) diastolic velocities, as well as a greater E/A ratio.[52-58] Others have reported no differences in these measures between trained and untrained individuals.[59-61] Attributing greater mitral gradients to superior diastolic function in athletes is clouded by the observation that these mitral flow velocities are markedly influenced by heart rate. Gilliam et al.[62] reported with cardiac pacing at 85 and then at 60 beats/min, average peak E increased from 0.30 to 0.48 m/s, peak A fell from 0.79 to 0.68 m/s, and peak E/A rose from 0.37 to 0.70 cm/s . Harrison et al.[63] found that when pacing rate was increased from 70 to 88 beats/min, peak A increased from 0.37 to 0.50 m/s and E/A fell from 1.63 to 1.24 m/s. Peak E, however, was not altered. Since athletes typically exhibit resting bradycardia compared with non-athletes, separating the relative contributions of upstream increases in left atrial volume (due to increased filling time) and superior myocardial relaxation properties becomes problematic. Assessment of ventricular function by myocardial velocity in the longitudinal plane (tissue Doppler imaging [TDI]) may provide more specific information, since early and late diastolic velocities (Em and Am, respectively) are less influenced by heart rate and ventricular loading than mitral blood flow velocities. The caveat with this technique is that myocardial velocities are directly related to heart size.[64] Thus, when Tumuklu et al.[65] and others[18,58,66] describe higher relaxation myocardial velocities (Em) at rest in trained individuals, it is not clear whether simply the larger ventricular size typical of endurance athletes might be responsible. Others have described similar TDI measures in resting athletes and non-athletes.[67,68] More pertinent to the present question, however, is characterization of diastolic function during exercise. Unfortunately, data providing comparisons of mitral flow velocities or TDI between athletes and non-athletes are limited, ª 2009 Adis Data Information BV. All rights reserved.

restricted largely to low-grade supine exercise[37,55,69] or post-exercise.[58] DiBello et al.[70] studied the effects of semisupine maximal cycle exercise on mitral E velocity in ten elite adult male distance runners and ten matched sedentary controls. Increases in stroke volume were greater in the athletes. Mean E values at rest were similar at rest (0.82 and 0.80 cm/s in the athletes and controls, respectively) and at 50% maximal exercise (1.28 and 1.29 cm/s). At peak exercise, heart rates were similar in the two groups (167 and 164 beats/min). Average peak E velocity at end exercise was higher in the athletes (1.54 vs 1.42 cm/s), but their mean endurance time was also greater (5.05 – 0.9 vs 4.28 – 0.7 min). Thus, it was not possible to decipher whether improved endurance fitness was the result or the cause of the enhanced E velocity. 2.3 Upstream Factors

Greater mitral flow velocities in athletes (and, hence, higher diastolic filling volumes) could also reflect upstream factors, particularly atrial pressure/volume relationships. In untrained subjects, tachycardia during progressive exercise is tightly linked to rises in atrial volume and pressure as systemic venous return increases, thereby serving to ‘defend’ a stable LVED.[15] Changes in atrial compliance that might accompany increases in blood volume and vagal tone during athletic training could alter this reflex. Dampening of the heart rate response to atrial expansion as venous return increases would permit greater ventricular filling volumes. This, in turn, would create a rise in stroke volume during high-intensity exercise via a Frank-Starling mechanism, which is not observed in non-athletic subjects. In untrained individuals, the constancy of atrial and ventricular dimensions during progressive exercise might reflect pericardial constraints on chamber size. Hoit et al.[71] demonstrated that pericardiectomy in dogs resulted in increases in resting LVED and stroke volume, accompanied by a rise in the early diastolic mitral pressure gradient (peak E wave velocity). It is conceivable, then, that changes in pericardial compliance in Sports Med 2009; 39 (8)

Stroke Volume in Trained Subjects

trained athletes might permit similar ventricular enlargement and augment stroke volume during exercise. However, no experimental evidence exists to examine such an effect.

693

peripheral mechanisms in controlling circulatory flow during exercise in highly trained individuals. Acknowledgements

2.4 Other Factors

Additional variables might also influence the patterns of stroke volume response to progressive exercise in endurance athletes and untrained individuals. For example: (i) heart rate does not increase linearly with work load, often demonstrating a tapering at high exercise intensities;[72] (ii) decrease in systolic time intervals with tachycardia is associated with augmentation of ventricular contractility;[73] and (iii) b1-receptor sensitivity and/or density may influence cardiac function during exercise.[74] How such factors might serve to influence and differentiate stroke volume responses between endurance athletes and non-athletes remains to be clarified. 3. Conclusions Whether or not (i) endurance athletes demonstrate a progressive rise in stroke volume during a progressive exercise test, and (ii) the mechanisms surrounding their generation of superior cardiac output differ from those of untrained individuals cannot be satisfactorily answered based on the current body of research data. From this review, it is apparent that equivalent amounts of evidence can be mustered to support both sides of the argument. The available information suggests the possibility that testing methodology and protocol might be responsible for the marked variation in stroke volume observed in these reports. If a non-plateau in stroke volume is typical of elitelevel endurance athletes, enhanced diastolic filling would appear to be the most likely mechanism, with a reliance on a Frank-Starling mechanism to augment stroke volume as work intensity rises. In this case, the relative contributions of upstream and downstream factors that would augment transmitral flow velocity would need to be delineated. Continued efforts to address these questions are important in seeking insights into the relative roles of central versus ª 2009 Adis Data Information BV. All rights reserved.

The author received no funding for the preparation of this review and has no conflicts of interest directly relevant to its contents.

References 1. Higginbotham M, Morris KG, Williams RS. Regulation of stroke volume during submaximal and maximal upright exercise in normal man. Circ Res 1986; 58: 281-91 2. Rowland TW. Circulatory responses to exercise: are we misreading Fick? Chest 2005; 127: 1023-30 3. Rowland T, Potts J, Potts T, et al. Cardiac responses to progressive exercise in normal children: a synthesis. Med Sci Sports Exerc 2000; 32: 253-9 4. Notarius CF, Magder S. Central venous pressure during exercise: role of the muscle pump. Can J Physiol Pharmacol 1996; 74: 647-51 5. Bevegard S, Holmgran A, Jonsson B. The effect of body position on the circulation at rest and during exercise, with special reference to the influence on the stroke volume. Acta Physiol Scand 1960; 49: 279-98 6. Rowland T, Garrison A, DeIulio A. Circulatory responses to progressive exercise: insights from positional differences. Int J Sports Med 2003; 24: 512-7 7. Magel JR, McArdle WD, Toner M. Metabolic and cardiovascular adjustment to arm training. J Appl Physiol 1978; 45: 75-9 8. Atkov OY, Bednenko VS, Formina GA. Ultrasound techniques in space medicine. Aviat Space Environ Med 1987; 58 Suppl. 9: A69-73 9. Rowland T, Bougault V, Walther G, et al. Cardiac responses to swim bench exercise in age group swimmers and nonathletic children. J Sci Med Sport 2009; 12: 266-72 10. Christie JL, Sheldahl LMN, Tristahi FE. Cardiovascular regulation during head out water immersion. J Appl Physiol 1990; 69: 657-64 11. Gonzalez-Alonso J. Point: stroke volume does decline during exercise at maximal effort in healthy individuals. J Appl Physiol 2008; 104: 275-80 12. Warburton DER, Gledhill N. Counterpoint: stroke volume does not decline during exercise at maximal effort in healthy individuals. J Appl Physiol 2008; 104: 276-8 13. Pokan R, Von Duvillard SP, Hofman P, et al. Change in left atrial and ventricular dimensions during and immediately after exercise. Med Sci Sports Exerc 2000; 32: 1713-8 14. Rowland T, Blum JW. Cardiac dynamics during upright exercise in boys. Am J Hum Biol 2000; 12: 749-57 15. Linden RJ. The size of the heart. Cardioscience 1994; 5: 225-33 16. Rowland T, Heffernan K, Jae SY, et al. Tissue Doppler assessment of ventricular function during cycling in

Sports Med 2009; 39 (8)

Rowland

694

17. 18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34. 35.

7- to 12-year old boys. Med Sci Sports Exerc 2006; 38: 1216-22 Guyton AC. Regulation of cardiac output. N Engl J Med 1967; 277: 805-12 Nottin S, Vinet A, Stecken F, et al. Central and peripheral cardiovascular adaptations to exercise in endurance trained children. Acta Physiol Scand 2002; 34: 637-42 Rowland T, Roti MW. Cardiac responses to progressive upright exercise in adult male cyclists. J Sports Med Phys Fitness 2004; 44: 179-85 Crawford MH, Petru MA, Rabinotwitz C. Effect of isotonic training on left ventricular volume during upright exercise. Circulation 1985; 72: 1237-43 Ferguson S, Gledhill N, Jamnik VK, et al. Cardiac performance in endurance-trained and moderately active young women. Med Sci Sports Exerc 2001; 33: 1114-9 Gledhill N, Cox D, Jamnik R. Endurance athletes’ stroke volume does not plateau: major advantage is diastolic function. Med Sci Sports Exerc 1994; 26: 1116-21 Krip B, Gledhill N, Jamnik V, et al. Effects of alterations in blood volume on cardiac function during maximal exercise. Med Sci Sports Exerc. 1997; 29: 1469-76 Warburton DER, Gledhill N, Jamnik VK, et al. Induced hypervolemia, cardiac function, VO2max, and performance of elite cyclists. Med Sci Sports Exerc 1999; 31: 800-8 Warburton DER, Haykowsky MJ, Quinney HA, et al. Myocardial response to incremental exercise in endurancetrained athletes: influence of heart rate, contractility, and the Frank-Starling effect. Exp Physiol 2002; 87: 613-22 Wiebe CG, Gledhill N, Jamnik VK, et al. Exercise cardiac function in young through elderly endurance trained women. Med Sci Sports Exerc 1999; 31: 684-91 Zhou B, Conlee RK, Jensen R, et al. Stroke volume does not plateau during graded exercise in elite male distance runners. Med Sci Sports Exerc 2001; 33: 1849-54 Ginzton LE, Conant R, Brizendine M, et al. Effect of long-term high intensity aerobic training on left ventricular volume during maximal upright exercise. J Am Coll Cardiol 1989; 14: 364-71 Ogawa T, Spina RJ, Martin WH, et al. Effects of aging sex, and physical training on cardiovascular responses to exercise. Circulation 1992; 86: 494-503 Rivera AM, Pels AE, Sady SP, et al. Physiological factors associated with the lower maximal oxygen consumption of master runners. J Appl Physiol 1989; 66: 949-54 Rowland T, Wehnert M, Miller K. Cardiac responses to exercise in competitive child cyclists. Med Sci Sports Exerc 2000; 32: 747-52 Rowland T, Unnithan V, Fernhall B, et al. Left ventricular response to dynamic exercise in young cyclists. Med Sci Sports Exerc 2002; 34: 637-42 Vanfraechem JHP. Stroke volume and systolic time interval adjustments during bicycle exercise. J Appl Physiol 1979; 46: 588-92 Ekblom B, Hermansen L. Cardiac output in athletes. J Appl Physiol 1968; 25: 619-25 Ahmad M, Dubiel JP. Left ventricular response to exercise in regular runners and controls: a radionuclide evaluation. Clin Nucl Med 1990; 15: 630-5

ª 2009 Adis Data Information BV. All rights reserved.

36. Rowland T, Goff D, Popowski B, et al. Cardiac responses to exercise in child distance runners. Int J Sports Med 1998; 19: 385-90 37. Huonker M, Ko¨nig, Keul J. Assessment of left ventricular dimensions and functions in athletes and sedentary subjects at rest and during exercise using echocardiographic, Doppler sonography and radionuclide ventriculography. Int J Sports Med 1996; 17 Suppl. 3: S173-9 38. Rowland T, Obert P. Doppler echocardiography for the estimation of cardiac output with exercise. Sports Med 2002; 32: 973-86 39. Kasikcioglu E, Kayseriglioglu A, Oflaz H, et al. Aortic distensibility and left ventricular diastolic functions in endurance athletes. Int J Sports Med 2005; 26: 165-70 40. Fagard RH. Athlete’s heart: a meta-analysis of the echocardiographic experience. Int J Sports Med 1996; 17 Suppl. 3: S149-8 41. Colan SD, Sanders SP, Borow KM. Physiologic hypertrophy: effects on left ventricular systolic mechanics in athletes. J Am Coll Cardiol 1987; 9: 776-83 42. Anholm JD, Foster C, Carpenter J, et al. Effect of habitual exercise on left ventricular response to exercise. J Appl Physiol 1982; 52: 1648-51 43. Boutcher SH, McLaren PF, Cotton Y, et al. Stroke volume response to incremental submaximal exercise in aerobically trained, active, and sedentary men. Can J Appl Physiol 2003; 28: 12-26 44. Brandao MUP, Wajngarten M, Rondon E, et al. Left ventricular function during dynamic exercise in untrained and moderately trained subjects. J Appl Physiol 1993; 75: 1989-95 45. Rubal BJ, Moody JM, Damore S, et al. Left ventricular performance of the athletic heart during upright exercise: a heart rate-controlled study. Med Sci Sports Exerc 1986; 18: 134-40 46. Schairer JR, Stein PD, Keteyian S, et al. Left ventricular response to submaximal exercise in endurance-trained athletes and sedentary adults. Am J Cardiol 1992; 70: 930-3 47. Hanson JS, Tabakin BS. Comparison of the circulatory response to upright exercise in 25 ‘‘normal’’ men and 9 distance runners. Br Heart J 1965; 27: 211-9 48. Nottin S, Nguyen L-D, Terbah M, et al. Left ventricular function in endurance-trained children by tissue Doppler imaging. Med Sci Sports Exerc 2004; 36: 1507-13 49. Tomai F, Ciavolella M, Gaspardone A, et al. Peak exercise left ventricular performance in normal subjects and in athletes assessed by first-pass radionuclide angiography. Am J Cardiol 1992; 70: 531-5 50. Cheng CP, Igarashi Y, Little WC. Mechanism of augmented rate of ventricular filling during exercise. Circ Res 1992; 70: 9-19 51. Rowland T, Mannie E, Gawle L. Dynamics of left ventricular diastolic filling during exercise. Chest 2001; 120: 145-50 52. Casa P, D’Andrea A, Galderisi M, et al. Pulsed Doppler tissue imaging in endurance athletes: relation between left ventricular preload and myocardial regional diastolic function. Am J Cardiol 2000; 85: 1131-6 53. Hoogsteen J, Hoogeveen A, Schaffers H, et al. Left atrial and ventricular dimensions in highly trained cyclists. Int J Cardiovasc Imag 2003; 19: 211-7

Sports Med 2009; 39 (8)

Stroke Volume in Trained Subjects

54. Kasikcioglu E, Oflaz H, Akhan H, et al. Right ventricular myocardial performance index and exercise capacity in athletes. Heart Vessels 2005; 20: 147-52 55. McFarlane N, Northridge DB, Wright AR, et al. A comparative study of left ventricular structure and function in elite athletes. Br J Sports Med 1991; 25: 45-8 56. Neilan TG, Yoerger DM, Douglas PS, et al. Persistent and reversible cardiac dysfunction among amateur marathon runners. Eur Heart J 2006; 27: 1079-84 57. Palazzuoli A, Puccetti L, Pastorelli M, et al. Transmitral and pulmonary venous flow study in elite male runners and young adults. Int J Cardiol 2002; 84: 47-51 58. Vinereanu D, Florescu N, Sculthorpe N, et al. Left ventricular long-axis dysfunction is augmented in the hearts of endurance-trained compared to strength-trained athletes. Clin Sci 2002; 103: 249-57 59. Fagard R, Van den Broeke C, et al. Noninvasive assessment of systolic and diastolic left ventricular function in female runners. Eur Heart J 1987; 8: 1305-11 60. George KP, Gates PE, Birch KM, et al. Left ventricular morphology and function in endurance-trained female athletes. J Sports Sci 1999; 17: 633-42 61. Gregoire JM, Vandenbossche JL, Messin R, et al. Diastolic function and modifications of left ventricular architecture in ultraendurance athletes. Acta Clin Belg 1989; 44: 388-95 62. Gilliam LD, Homma S, Novick SS, et al. The influence of heart rate on Doppler mitral inflow patterns [abstract]. Circulation 1987; 76 Suppl. IV: IV-123 63. Harrison MR, Clifton GD, Pennell AT, et al. Effect of heart rate on left ventricular diastolic transmitral flow velocity patterns assessed by Doppler echocardiography in normal subjects. Am J Cardol 1991; 67: 622-7 64. Krieg A, Scharhag J, Kindermann W, et al. Cardaic tissue Doppler imaging in sports medicine. Sports Med 2007; 37: 15-30 65. Tumuklu MM, Ildizli M, Ceyhan K, et al. Alterations in left ventricular structure and diastolic function in professional football players: assessment by tissue Doppler

ª 2009 Adis Data Information BV. All rights reserved.

695

66.

67.

68.

69.

70.

71.

72.

73.

74.

imaging and left ventricular flow propagation velocity. Echocardiography 2007; 24: 140-8 Koc M, Bozkurt A, Akpinar O, et al. Right and left ventricular adaptation to training determined by conventional echocardiography and tissue Doppler imaging in young endurance athletes. Acta Cardiol 2007; 62: 13-8 Baldi JC, McFarlane K, Oxenham HC. Left ventricular diastolic filling and systolic function of young and older trained and untrained men. J Appl Physiol 2003; 95: 2570-5 Schmidt-Trucka¨ss A, Schmid A, Ha¨ussler G, et al. Left ventricular wall motion during diastolic filling in endurance-trained athletes. Med Sci Sports Exerc 2001; 33: 189-95 Nixon JV, Wright AR, Porter TR, et al. Effects of exercise on left ventricular diastolic performance in trained athletes. Am J Cardiol 1991; 68: 945-9 DiBello V, Santoro G, Talarico L, et al. Left ventricular function during exercise in athletes and sedentary men. Med Sci Sports Exerc 1996; 28: 190-6 Hoit BD, Dalton N, Bhargava V, et al. Pericardial influences on right and left ventricular filling dynamics. Circ Res 1991; 68: 197-208 Hofmann P, Pokan R, Seibert F-J, et al. The heart rate performance curve during incremental cycle ergometer exercise in healthy young male subjects. Med Sci Sports Exerc 1997; 29: 762-8 Hofmann P, Pokan R, Preidler K, et al. Relationship between heart rate threshold, lactate turn point and myocardial function. Int J Sports Med 1994; 15: 232-7 Hofmann P, Wonisch M, Pokan R, et al. b1-Adrenoceptor mediated origin of the heart rate performance curve deflection. Med Sci Sports Exerc 2005; 37: 1704-9

Correspondence: Dr Thomas Rowland, Department of Pediatrics, Baystate Medical Center, 759 Chestnut St, Springfield, MA, USA. E-mail: [email protected]

Sports Med 2009; 39 (8)