Major international events, such as the Olympic Games, Fédération Internationale de Football Association (FIFA) Football World Cup and the International Rugby Board (IRB) Rugby World Cup (RWC), provide worldwide showcases for sport. The intensity of media attention afforded to these events, however, means that all aspects of the sport are invariably placed under close scrutiny. In particular, the nature and causes of injuries sustained during these competitions attract a large amount of public interest because the loss of key athletes can have marked effects on a country’s performance. Because injuries during major competitions raise awareness of the risks associated with injury, it is essential that governing bodies collect sufficient and valid epidemiological data to provide evidence-based responses to stakeholder questions about these and other medical issues.

Although detailed information about the nature and causes of rugby injuries is available at the amateur1 2 and professional3^–5 levels, information at the international level is more limited. Jakoet and Noakes6 reported an incidence of 32 match injuries/1000 player-match hours during RWC 1995, and Best et al7 reported an incidence of 98 match injuries/1000 player-match hours for RWC 2003. Some studies have reported results for individual countries competing in the RWC; Wekesa et al8 reported an incidence of 196 match injuries/1000 player-match hours during six RWC 1995 regional qualifying matches in Kenya, and Brooks et al9 reported 230 match injuries/1000 player-match hours for England during RWC 2003. Bathgate et al10 reported an incidence of 69 match injuries/1000 player-match hours for Australia over the period 1994 to 2000, which included results obtained for Australia during RWC 1995 and 1999. Differences in the definitions and procedures used in these and other studies identified the need to develop a consistent approach for injury surveillance studies in rugby so that meaningful comparisons of results may be made across studies. Consequently, an international meeting was facilitated by the IRB in 2006 to address this issue, and the recommendations from this meeting were published as an international consensus statement on definitions and procedures that should be implemented for injury surveillance studies in rugby.11

The aim of the current study was to record, analyse and report the incidence, severity, nature and cause of match and training injuries sustained by players competing in the IRB RWC 2007.

METHODS

The RWC 2007 injury surveillance study covered a 7-week period during September and October 2007. Over this period, the IRB and the host rugby union (France) implemented a number of arrangements to provide medical support for the 20 countries competing in the tournament and to record information about the injuries sustained by players, non-playing support staff and IRB tournament officials. The study protocol was approved by the University of Nottingham Medical Ethics Committee.

Match day and non-match-day medical support

The tournament medical officer (TMO), who had overall responsibility for the provision of medical support services, was supported by a city medical officer (CMO) at each host city. The TMO was responsible for establishing a network of medical specialists that would be available to teams (players and non-playing support staff) and IRB officials on match and non-match-days; this support was additional to the arrangements normally made by each country for their own players. Specialists available included a sports medicine doctor; physiotherapist, radiologist, anaesthetist, ophthalmologist, resuscitator, and orthopaedic, maxillofacial, plastic, neural and spinal surgeons. At each stadium, there were fully equipped treatment areas available for the care of injured players and officials. CMOs submitted medical reports containing cases of suturing, specialist pitch-side medical treatment and diagnostic imaging provided for players during each game, and also submitted records of medical treatments and examinations provided for players, non-playing team staff and IRB officials at other times during the tournament.

Injury surveillance study

Definitions and procedures used throughout the study were compliant with the IRB approved international consensus statement.11 An injury surveillance manual, which included the aims of the study, definitions, procedures and all necessary report forms required to implement the study, was provided to each country’s medical team before the start of the tournament. Players’ baseline information (normal playing position, age in years, free-standing stature in cm, body mass in kg, dominant leg and arm) and consent were obtained, and baseline data were reported as means (SD). Match exposures were calculated on the basis of 15 players (8 forwards, 7 backs) per team exposed for 80 minutes (first half 0–20, 20–40+, second half 40–60, 60–80+ minutes). No allowances were made for players temporarily or permanently dismissed during a match (yellow or red cards). No matches required extra time to be played. Training exposures were calculated on the basis of the number of players attending (forwards, backs) and the number, length (minutes) and structure (skills: full contact, semi-contact, no contact; conditioning: weights, non-weights) of training sessions undertaken by each team.

The primary injury definition used in the study11 was “Any physical complaint sustained by a player during a RWC match or training session that prevented the player from taking a full part in all training activities or match play for more than 1 day following the day of injury, irrespective of whether match or training sessions were actually scheduled”. When required, injuries were followed up after the tournament in order to obtain return-to-play/training dates. To allow results from this study to be compared with data collected during RWC 1995 and 2003, injuries were also recorded using a secondary definition that encompassed injuries resulting in a player missing ⩾1 matches. Unless specifically stated, however, all data reported in this study refer to results obtained using the primary “time-loss” injury definition. Injuries were reported as recurrences on the basis of clinical judgement by the player’s medical team using the definition:11 “An injury of the same type and at the same site as an index injury and which occurred after the player’s return to full participation from the index injury”. Absences due to illnesses and non-sport related medical conditions were not included. Team doctors were responsible for recording injury details (location, type, Orchard Code,12 number of days and matches missed, use of diagnostic tests and invasive procedures) and risk factors (playing position, time, activity at time of injury) for every match and training injury.

Incidence of injury is reported here as the number of injuries/1000 player-hours of exposure (95% CI) and severity of injury is reported as mean (95% CI) and median (95% CI) values and grouped within the categories:11 minimal (2–3 days), mild (4–7 days), moderate (8–28 days) and severe (>28 days).

Data analysis

The t test13 was used to identify significant differences (p<0.05) in anthropometric data between groups. Incidence and severity of injury are reported with 95% CI and p values (z-test) for comparisons.13 The χ² test was used to identify significant differences (p<0.05) in the numbers of injuries among groups.13

RESULTS

In total, 626 players (342 forwards, 284 backs) took part in the study, of whom 26 were IRB-approved replacements for injured players (16 forwards, 10 backs). All countries provided players’ baseline information and informed consent, and submitted appropriate exposure and injury report forms. Table 1 shows the mean age, stature and body mass of the cohort as a function of playing position.

There were 48 (40 group stage, 8 knockout stage) matches in the tournament, which equated to 1920 (1024 forwards, 896 backs) player-match hours. These matches resulted in 161 time-loss injuries (86 forwards, 75 backs), of which 140 were acute (74 forwards, 66 backs), 18 gradual onset (11 forwards, 7 backs), 9 were recurrent (6 forwards, 3 backs) and 83 (44 forwards, 39 backs) resulted in the injured player missing a subsequent match. This number of match injuries equates to an average of 1.7 injuries/team-game (range 0 to 5). During the 7 week tournament period, 17046 (9188 forwards, 7858 backs) player-training hours were recorded: this training exposure resulted in 60 time-loss injuries (32 forwards, 28 backs), of which 41 were acute (22 forwards, 19 backs), 18 gradual onset (9 forwards, 9 backs), 16 were recurrent (8 forwards, 8 backs) and 45 (22 forwards, 23 backs) resulted in the injured player missing a subsequent match. Ten injuries resulted in >10 days of absence from training, but the players were not recorded as having missed a match because the players sustained the injuries during their country’s last game and they returned to their country in the off-season. No player sustained a catastrophic or career-ending injury during the tournament, but 26 of the original 600 declared squad players were replaced because of a competition-related injury and one player retired from the tournament as a consequence of a cardiovascular condition experienced during a non-rugby related activity. Overall, 122 (55.2%) injured players (match: 46 forwards, 43 backs; training: 16 forwards, 17 backs) had imaging techniques used in the assessment of their injuries, 8 (3.6%) players (match: 3 forwards, 2 backs; training: 1 forward, 2 backs) required surgery and 6 (2.7%) players (match: 2 forwards, 3 backs; training: 1 forward, 0 backs) received corticosteroid injections. In addition, 51 (42 forwards; 9 backs) players temporarily left the field of play during matches to receive medical attention from their team doctor of which 28 (21 forwards, 7 backs) had lacerations sutured by the IRB match doctor. Tournament medical staff provided a range of treatment and/or injury investigation services to 116 players and non-players on match days and to 63 players and non-players at other times (table 2).

Incidence, severity and nature of injury

There were 3434 player days lost from matches and training as a consequence of injury (2369 match injuries, 1065 training injuries). The incidences and mean and median severities of match and training injuries for forwards and backs are presented in table 3, whilst the incidence of injury as a function of grouped severity is shown in table 4.

The incidences of match and training injuries resulting in players missing at least one match and the incidence of match injuries requiring players to leave the pitch temporarily for medical treatment are shown in table 5.

The incidence and mean severity of injuries are presented as functions of injury location in tables 6 (match injuries) and 7 (training injuries) and as a function of injury type in tables 8 (match injuries) and 9 (training injuries). Cross-tabulations of the major groupings for injury type and location of match and training injuries are presented in table 10 and the injuries causing the greatest total loss of time for all players during the tournament are shown in table 11.

After match injuries (n = 161), 58 players (29 forwards, 29 backs) were immediately removed from play, and 39 (23 forwards, 16 backs) were removed from play later in the game. During matches, lower limb injuries sustained by backs were significantly more likely to occur to their non-dominant leg (11 dominant, 34 non-dominant, p<0.001), whereas there was no significant difference for forwards (20 dominant, 21 non-dominant, p = 0.876). Similarly, upper limb match injuries sustained by backs were significantly more likely to be to their non-dominant arm (3 dominant, 12 non-dominant, p = 0.020), whereas there was again no significant difference for forwards (7 dominant, 9 non-dominant, p = 0.617). There were no significant differences in the numbers of training injuries to the dominant and non-dominant leg for backs (13 dominant, 14 non-dominant, p = 0.847) or forwards (12 dominant, 12 non-dominant, p = 1.000). There were too few upper limb injuries sustained during training (n = 3) to compare the results.

Injury causation

The incidence of match injuries is shown as a function of the period of play in table 12 and the match activity at the time of injury in table 13. The incidence of training injuries as a function of the training activity at the time of injury is shown in table 14.

DISCUSSION

The medical support services provided by the host rugby union were used extensively by teams (9.4 treatments/investigations per team) to support the diagnosis and treatment of injured players. The services used most frequently were MRI and radiography examinations and match-day suturing of lacerations: only two countries requested diagnostic ultrasound examinations, which probably reflects the availability of this imaging technique within most teams’ medical facilities. Although a cost-benefit analysis was not undertaken as part of this study, the demand for these medical facilities throughout the tournament justifies the necessity for and the continued provision of these services at future events.

The potential for making comparisons between RWC 2007 and RWC 1995 and 2003 results is limited for several reasons. Previous studies did not report players’ anthropometric data and training injuries or differentiate between results for forwards and backs. RWC 2003 results7 also related to professional rugby but the data for RWC 1995 referred to the amateur era of rugby.6 Finally, but importantly, the definition of injury used in the RWC 1995 and 2003 studies grouped injuries that caused players to miss matches on an equal basis with minor lacerations that only required suturing to allow players to return to the field of play. Best et al highlighted that the injury definition adopted during the RWC 2003 study had limitations because the large number of minor lacerations recorded could distort the conclusions:7 these authors therefore suggested that the definition used in future RWC studies should be revised. In order to assess the results obtained in the present study, the data were also compared with information reported for professional rugby in England,3 4 as a similar injury definition was used and results were presented for forwards and backs and for match and training activities.

Forwards and backs in this study were on average older, taller and heavier than previously reported for rugby union players.3 9 14 15 These differences may simply reflect the requirements of modern international rugby16 and/or the continued temporal changes in height and weight that have been reported for rugby union players.17 Although direct comparisons could not be made, injuries that resulted in players missing matches in the present study, together with players leaving the pitch temporarily for medical treatment, approximated to the injury definitions used in the RWC 1995 and 2003 studies. On this basis, the incidence of injury calculated for the present study (69.8 injuries/1000 player-hours) was higher than that for RWC 19956 (32) but lower than that reported for RWC 2003.7 The incidence of missed-match injuries obtained for RWC 2007 does, however, agree closely with the incidence of missed-match injuries reported for English professional rugby3 (40 for all players, 38 forwards, 41 backs). The time-lapse between an individual team's matches at RWC 2003 and 2007 varied from 4 to 10 days, so there was a considerable variation in the threshold severity value for recording a missed-match injury. Additionally, at the end of the tournament, some players returned to their countries during the off-season; in these cases, injuries sustained by players in their team’s final games at RWC 2007 were not recorded as missed-match injuries, irrespective of the time the players were absent from training. These factors contribute to the variability seen in incidence, severity, nature and causation of injuries observed in studies using the missed match injury definition, and they make comparisons of results with studies using the time-loss injury definition questionable. The incidence of time-loss match injuries in the present study (83.9 injuries/1000 player-hours) was similar to the incidence of time-loss injuries (91) reported for professional players in England.3 The incidence of time-loss training injuries during RWC 2007 (3.5 injuries/1000 player-hours) was, however, higher than that reported (2.0) for professional players in England.4 The average number of days lost per injury was not recorded in previous RWC studies; however, the results obtained in this study were comparable with those reported for professional players in England3 4 (match: 18 forwards, 18 backs; training: 24 forwards, 23 backs).

In the present study, fewer injuries were sustained to the head and fewer classified as lacerations than was the case in either RWC 1995 or RWC 2003.6 7 these differences can largely be explained by the inclusion of match-day laceration/blood injuries in the two earlier studies. Only five players were reported to have sustained a concussion at RWC 2007, of whom two missed a match, which is a similar situation to that reported for RWC 2003.7 Results from this study therefore support the view expressed by Best et al that concussions were under-reported at the RWC and is an issue that requires further investigation.7 The knee was the most commonly injured structure during both matches and training, followed by the posterior thigh and shoulder during matches and the posterior thigh during training. Knee ligament and posterior thigh muscle injuries also caused the greatest loss of time for players for both match and training injuries.

As reported previously,3 6 7 the tackle was the activity/event responsible for the highest incidence of injury; however, in the present study, a much higher proportion of tackle injuries were the result of being tackled rather than from tackling. This difference may indicate that from a personal injury perspective, players’ tackling techniques have improved over the last 10 years. After tackles, collisions, rucks and scrums were responsible for the highest incidence of match injuries but importantly, the incidence of collision injuries was almost four times higher than that reported for English professional rugby.18 This may reflect a different style of play or a different interpretation of the laws of rugby at the international level; in either case, this warrants further investigation. In training, full-contact skills activities, not surprisingly, were the most likely to cause injury.

CONCLUSION

This study shows that the international consensus statement on definitions and procedures11 provides an appropriate and achievable methodology for injury surveillance studies in rugby union. The results presented for the incidence, severity, nature and causation of match and training injuries sustained at Rugby World Cup 2007 provide a benchmark for future injury surveillance studies at the Rugby World Cup.