Br J Sports Med 2005;39:196-204
© 2005 BMJ Publishing Group Ltd & British Association of Sport and Exercise Medicine

ORIGINAL ARTICLE

Summary and agreement statement of the 2nd International Conference on Concussion in Sport, Prague 2004

P McCrory8, K Johnston5, W Meeuwisse9, M Aubry1, R Cantu2, J Dvorak3, T Graf-Baumann4, J Kelly6, M Lovell7 and P Schamasch10

1 Chief Medical Officer, International Ice Hockey Federation
2 Department of Surgery, Chief Neurosurgery Service and Director, Sports Medicine Service, Emerson Hospital, Concord, MA, USA; Medical Director, National Center for Catastrophic Sports Injury Research, Chapel Hill, NC, USA
3 Chairman, FIFA Medical Research and Assessment Center (F-MARC); Chairman, Department of Neurology, Schulthess Clinic, Zurich, Switzerland
4 Director, Office for Science Management, Administration and Scientific Director, German Society for Musculo-Skeletal Medicine and Pain Therapy, FIFA Medical Research and Assessment Center (F-MARC), Teningen, Germany
5 Neurosurgeon/Concussion Consultant, Departments of Neurosurgery, Kinesiology & Physical Education, McGill University; Director, Concussion Program, McGill Sport Medicine Centre, Montreal, Canada
6 Professor of Neurosurgery and Rehabilitation Medicine, University of Colorado School of Medicine, Denver, CO, USA
7 Director, Sports Medicine Concussion Program, University of Pittsburgh; Co-director, National Hockey League Neuropsychology Program, Pittsburgh, PA, USA
8 Associate Professor, Center for Health, Exercise and Sports Medicine & The Brain Research Institute, University of Melbourne, Melbourne, Australia
9 Professor and Medical Director, University of Calgary Sport Medicine Centre; Sport Injury Epidemiologist, National Hockey League, Calgary, Alberta, Canada
10 Medical and Scientific Director, International Olympic Committee, Lausanne, Switzerland

Correspondence to:
Associate Professor McCrory
PO Box 93, Shoreham, Victoria 3916, Australia;

Accepted 7 February 2005

/ ABSTRACT

In November 2001, the 1st International Symposium on Concussionin Sport was held in Vienna, Austria to provide recommendationsfor the improvement of safety and health of athletes who sufferconcussive injuries in ice hockey, football (soccer), and othersports. The 2nd International Symposium on Concussion in Sportwas organised by the same group and held in Prague, Czech Republicin November 2004. It resulted in a revision and update of theVienna consensus recommendations, which are presented here.

Keywords: concussion; head injury; brain

This paper is a revision and update of the Vienna consensusrecommendations developed after the 1st International Symposiumon Concussion in Sport.1 The Prague agreement statement is designedto build on the principles outlined in the original Vienna documentand to develop further conceptual understanding of this problem.This document is developed for use by doctors, therapists, healthprofessionals, coaches, and other people involved in the careof injured athletes, whether at the recreational, elite, orprofessional level.

/ BACKGROUND PERSPECTIVE

In November 2001, the 1st International Symposium on Concussionin Sport was held in Vienna, Austria. This meeting was organisedby the International Ice Hockey Federation (IIHF) in partnershipwith the Federation Internationale de Football (FIFA) and theInternational Olympic Committee Medical Commission (IOC). Aspart of the resulting mandate for the future, the need for leadershipand updates was identified. To meet that mandate, the 2nd InternationalSymposium on Concussion in Sport was organised by the same groupand held in Prague, Czech Republic in November 2004.

The original aims of the symposia were to provide recommendationsfor the improvement of safety and health of athletes who sufferconcussive injuries in ice hockey, football (soccer), and othersports. To this end a range of experts were invited to bothmeetings in order to address specific issues of epidemiology,basic and clinical science, injury grading systems, cognitiveassessment, new research methods, protective equipment, management,prevention, and long term outcome. At the conclusion of theinitial conference, a small group of experts were given a mandateby the conference delegates and organising bodies to draft adocument describing the agreement position reached by thosein attendance at that meeting. That document was co-publishedin the British Journal of Sports Medicine, Clinical Journalof Sport Medicine, and Physician and Sportsmedicine.1

The wider interest base resulting from the first meeting anddocument was reflected by the expanded representation. New groupsat the second meeting included trauma surgeons, sport psychologists,and others. This same group has produced the current documentas an update of the original Vienna consensus document and includesa sideline assessment form with a pocket sized summary cardfor use by clinicians.

This protocol represents a work in progress, and, as with allother recommendations or proposals, it must be updated as newinformation is added to the current state of the literatureand understanding of this injury.

/ BACKGROUND ISSUES

Definition of concussion
Over 35 years ago, the Committee on Head Injury Nomenclatureof the Congress of Neurological Surgeons proposed a "consensus"definition of concussion.2,3 This definition was recognisedas having a number of limitations in accounting for the commonsymptoms of concussion. In the Vienna document, a revised consensusdefinition was proposed as follows: "Sports concussion is definedas a complex pathophysiological process affecting the brain,induced by traumatic biomechanical forces". Several common featuresthat incorporate clinical, pathological, and biomechanical injuryconstructs that may be used in defining the nature of a concussivehead injury include the following.

  1. Concussion may be caused by a direct blow to the head, face,neck, or elsewhere on the body with an "impulsive" force transmittedto the head.
  2. Concussion typically results in the rapid onsetof short livedimpairment of neurological function that resolvesspontaneously.
  3. Concussion may result in neuropathologicalchanges, but theacute clinical symptoms largely reflect a functionaldisturbancerather than structural injury.
  4. Concussion resultsin a graded set of clinical syndromes thatmay or may not involveloss of consciousness. Resolution ofthe clinical and cognitivesymptoms typically follows a sequentialcourse.
  5. Concussionis typically associated with grossly normal structuralneuroimagingstudies.

No changes were made to the definition by the Prague Group beyondnoting that in some cases post-concussive symptoms may be prolongedor persistent.

Pathophysiological basis of concussion
At this time, there is no existing animal or other experimentalmodel that accurately reflects a sporting concussive injury.It is noted that, in experimental models, of more severe injurya complex cascade of biochemical, metabolic, and gene expressionchanges occur.4 Whether similar metabolic changes occur in sportsconcussion, however, remains speculative at this time.5

Concussion grading scales
The Vienna recommendation that injury grading scales be abandonedin favour of combined measures of recovery in order to determineinjury severity (and/or prognosis) and hence individually guidereturn to play decisions received continued support.

It was also noted that concussion severity can only be determinedin retrospect after all concussion symptoms have cleared, theneurological examination is normal, and cognitive function hasreturned to baseline.6 There is limited published evidence thatconcussion injury severity correlates with the number and durationof acute concussion signs and symptoms and/or degree of impairmenton neuropsychological testing.7,8,9,10,11,12 The developmentof validated injury severity scales continues in the publishedliterature.13

Subtypes of concussion
One of the issues speculated on at the Vienna conference waswhether concussion represents a unitary phenomenon with a linearspectrum of injury severity or whether different concussionsubtypes exist. These subtypes may represent differences inclinical manifestations (confusion, memory problems, loss ofconsciousness), anatomical localisation (such as cerebral versusbrainstem), biomechanical impact (rotational versus linear force),genetic phenotype (apolipoprotein epsilon 4 (ApoE4) positiveversus ApoE4 negative), neuropathological change (structuralinjury versus no structural injury), or an as yet undefineddifference. These factors may operate independently or interactwith each other. It is clear that the variations in clinicaloutcome with the same impact force require a more sophisticatedapproach to the understanding of this phenomenon than currentlyavailable.14

Significance of loss of consciousness
The traditional approach to severe traumatic brain injury usingloss of consciousness as the primary measure of injury severityhas acknowledged limitations in assessing the severity of sportingconcussive injury. Findings in this field describe associationof loss of consciousness with specific early deficits but doesnot necessarily imply severity.13,15 As such the presence ofloss of consciousness as a symptom would not necessarily classifythe concussion as complex (see below).

Significance of amnesia
There is renewed interest in the role of post-traumatic amnesiaand its role as a surrogate measure of injury severity.13,16Published evidence suggests that the nature, burden, and durationof the clinical post-concussive symptoms may be more importantthan the presence or duration of amnesia alone.8,15,17 Furtherit must be noted that retrograde amnesia varies with the timeof measurement after the injury and hence is poorly reflectiveof injury severity.18,19

Paediatric concussive injury
The general recommendations outlined in the Vienna documentwere originally designed for the management of adult sportingconcussion. Agreement was reached, however, that identifiedthose recommendations as relevant and useful to management ofchildren as well. In broad terms it was felt that the recommendationsshould be applicable to children (defined as 5–18 yearsof age) whereby children should not be allowed to return toplay or training until clinically completely symptom free. Inaddition, the concept of "cognitive rest" was introduced withspecial reference to a child’s need to limit exertionwith activities of daily living and to limit scholastic activitieswhile symptomatic. There was also a recognition by the groupthat additional research is needed to better clarify the potentialdifferences between adults and children with regard to recoveryfrom injury and to develop cognitive assessment tools that betterevaluate the younger athlete.

Formal cognitive assessment is currently problematic until lateteen years because of the continuing cognitive maturation thatoccurs during this period, which, in turn, makes the utilityof comparison with either the person’s own baseline performanceor population norms limited.20

Because of the different physiological response during childhoodto head trauma, a conservative return to play approach is recommended.It may be appropriate to extend the amount of time of asymptomaticrest and/or the length of the graded exertion in children andadolescents. Future research is needed in this area.

/ A NEW CLASSIFICATION OF CONCUSSION IN SPORT

Historically, concussions have been classified with a numberof different grading systems. In the Vienna Statement, thisapproach was abandoned. One of the key developments by the PragueGroup is the understanding that concussion may be categorisedfor management purposes as either simple or complex.

Simple concussion
In simple concussion, an athlete suffers an injury that progressivelyresolves without complication over 7–10 days. In suchcases, apart from limiting playing or training while symptomatic,no further intervention is required during the period of recovery,and the athlete typically resumes sport without further problem.Formal neuropsychological screening does not play a role inthese circumstances, although mental status screening shouldbe a part of the assessment of all concussed athletes. Simpleconcussion represents the most common form of this injury andcan be appropriately managed by primary care physicians or bycertified athletic trainers working under medical supervision.21The cornerstone of management is rest until all symptoms resolveand then a graded programme of exertion before return to sport.All concussions mandate evaluation by a medical doctor.

Complex concussion
Complex concussion encompasses cases where athletes suffer persistentsymptoms (including persistent symptom recurrence with exertion),specific sequelae (such as concussive convulsions), prolongedloss of consciousness (more than one minute), or prolonged cognitiveimpairment after the injury. This group may also include athleteswho suffer multiple concussions over time or where repeatedconcussions occur with progressively less impact force. In thisgroup, there may be additional management considerations beyondsimple return to play advice. Formal neuropsychological testingand other investigations should be considered in complex concussions.It is envisaged that such athletes would be managed in a multidisciplinarymanner by doctors with specific expertise in the managementof concussive injury such as a sport medicine doctor with experiencein concussion, sports neurologist, or neurosurgeon.

/ CLINICAL ISSUES

Pre-participation physical examination
Recognising the importance of concussion history, and appreciatingthe fact that many athletes will not recognise all the concussionsthey may have suffered in the past, a detailed concussion historyis of value.22–25 Such a history may identify athletesthat fit into the "complex" category outlined above and providesan opportunity for the doctor to educate the athlete about thesignificance of concussive injury.

A structured concussion history should include specific questionsas to previous symptoms of a concussion, not just perceivednumber of past concussions. It is also worth noting that dependenceon the recall of concussive injuries by team mates or coacheshas been shown to be unreliable.22 The clinical history shouldalso include information about all previous head, face, or neckinjuries, as these may have clinical relevance to the presentinjury. It is worth emphasising that, with maxillofacial andneck injuries, co-existent concussive injuries may be missedunless specifically assessed. Specific questions pertainingto disproportionate impact versus symptom severity matchingmay alert the clinician to a progressively increasing vulnerabilityto injury.

As part of the clinical history, it is advised that detailson protective equipment used at the time of injury be sought,both for recent and remote injuries. The benefit of this approachallows modification and optimisation of protective behaviourand an opportunity for education.

It is specifically recommended that:

  1. both a baseline cognitive assessment (such as the Prague SCATtest in the absence of computerised neuropsychological testing)and symptom score is performed as part of the pre-participationevaluation;
  2. although formal baseline neuropsychological screeningmay bebeyond the resources of many sports or indivifdual athletes,it is recommended that, in organised high risk sports, considerationbe given to having cognitive evaluation regardless of the ageor level of performance.

Signs and symptoms of acute concussion
The suspected diagnosis of sports concussion made on the sidelineis applicable to both medical and non-medical personnel andcan include clinical symptoms, physical signs, cognitive impairment,and/or loss of consciousness.

If any one of the following symptoms or problems is present,a head injury should be suspected and appropriate managementinstituted. These will be summarised on the sideline concussionassessment tool (SCAT) that accompanies this document (fig 1).


View larger version (73K):
[in this window]
[in a new window]
/ Figure 1Sport concussion assessment tool (SCAT).
  1. Cognitive features (see below)
  2. Unaware of period, opposition,score of game
  3. Confusion
  4. Amnesia
  5. Loss of consciousness
  1. Typical symptoms (see SCAT (fig 1) for standard symptomscale);other symptoms such as a subjective feeling of slownessandfatigue after an impact may indicate that a concussion hasoccurredor has not fully resolved.26
  2. Headache or pressurein the head
  3. Balance problems or dizziness
  4. Nausea
  5. Feeling"dinged","foggy", stunned, or "dazed"
  6. Visual problems—forexample,seeing stars or flashinglights, double vision
  7. Hearingproblems—forexample, ringing in the ears
  8. Irritabilityor emotional changes
  1. Physical signs
  2. Loss of consciousness/impairedconscious state
  3. Poor coordination or balance
  4. Concussiveconvulsion/impactseizure
  5. Gait unsteadiness/loss of balance
  6. Slow to answerquestions or follow directions
  7. Easily distracted,poor concentration
  8. Displaying inappropriate emotions—forexample, laughing,crying
  9. Vomiting
  10. Vacant stare/glassy eyed
  11. Slurred speech
  12. Personality changes
  13. Inappropriate playingbehaviour—forexample, running inthe wrong direction
  14. Significantly decreasedplaying ability

Sideline evaluation of cognitive function is an essential componentin the assessment of this injury. Brief neuropsychological testbatteries that assess attention and memory function have beenshown to be practical and effective. Such tests include theMaddocks questions27 and the Standardised assessment of concussion.28It is worth noting that standard orientation questions—forexample, time, place, person—have been shown to be unreliablein the sporting situation when compared with memory assessment.27,29

It is recognised, however, that abbreviated testing paradigmsare designed for rapid concussion evaluation on the sidelinesand are not meant to replace comprehensive neuropsychologicaltesting, which is sensitive enough to detect subtle deficitsthat may exist beyond the acute episode, nor should they beused as a stand alone tool for the ongoing management of sportsconcussions. It should also be recognised that the appearanceof symptoms may be delayed several hours after a concussiveepisode.

Convulsive and motor phenomena
A variety of acute motor phenomena—for example, tonicposturing—or convulsive movements may accompany a concussion.30,31Although dramatic, these clinical features are generally benignand require no specific management beyond the standard treatmentfor the underlying concussive injury.

Development of the sport concussion assessment tool (SCAT)
Figure 1 outlines the SCAT. The intent was to create a standardisedtool that could be used for patient education as well as forphysician assessment of sports concussion. The SCAT was developedby combining the following existing tools into a new standardisedtool:

  1. Sideline evaluation for concussion.28,29
  2. Management of concussionsports palm card; AmericanAcademyof Neurology and the BrainInjury Association.32
  3. Standardised assessment of concussion.33
  4. Sideline concussion check; UPMC, Thinksafe, Sports MedicineNew Zealand Inc and the Brain Injury Association.
  5. McGill abbreviatedconcussion evaluation (ACE) (unpublished).
  6. National HockeyLeague physician evaluation form (unpublished).
  7. The UK JockeyClub assessment of concussion.34
  8. Maddocks questions.27

The authors gave input through a process of collaboration anditerative review. The SCAT was evaluated for face and contentvalidity on the basis of scientific literature35 and clinicalexperience of the authors. The memory questions, specifically,were modified from the validated Maddocks questions to makethese questions less football-specific.27

/ INVESTIGATIONAL ISSUES

Neuropsychological assessment after concussion
The application of neuropsychological testing in concussionhas been shown to be of value and continues to contribute significantinformation in concussion evaluation.10,11,36,37 It has beenshown that cognitive recovery may precede or follow clinicalsymptom resolution, suggesting that the assessment of cognitivefunction should be an important component in any return to playprotocol.12 It must be emphasised, however, that neuropsychologicalassessment should not be the sole basis of a return to playdecision but rather be seen as an aid to the clinical decisionmaking. Although neuropsychological screening may be performedor interpreted by other healthcare professionals, the finalreturn to play decision should remain a medical one in whicha multidisciplinary approach has been taken.

Neuropsychological testing should not be performed while theathlete is symptomatic because it adds nothing to return toplay decisions, and it may contaminate the testing process byallowing practice effects to confound the results. In certaincases, however, serial follow up after the injury is valuable,both as a means to encourage athlete compliance and for comparisonpurposes.