Materials and methods

A prospective cohort study of European professional football was carried out over seven consecutive seasons, 2001/2 to 2007/8, and the overall results have recently been published.17 A total of 88 club seasons were included, involving 23 different clubs, five of which participated in all seven seasons. Teams participated in a mean (SD) of 3.8 (2.2) seasons (range 1 to 7). The methodology followed the consensus on injury definitions and data-collection procedures in football studies18 and the Union of European Football Associations (UEFA) model, and has been reported in detail previously.19 Accordingly, three forms were used for data collection; (1) a baseline form with player anthropometrics collected at the beginning of each season, (2) an injury card that was filled in immediately after the injury and sent to the study group each month and (3) an exposure form where individual player exposure to match play and training was registered and sent in on a monthly basis.

Study sample and study period

All players in the first team squads each season were invited to participate. In total, 1065 players agreed to participate and provided signed written consent. Players injured at the start of their first season were included, but their present injuries were not taken into account. Players who left their club before the end of the season were included for as long as they participated. Details of participating clubs and player anthropometrics are available in a separate report.17

Data collection

Baseline forms for anthropometric data and previous medical history were used, as were exposure registration forms and injury cards. A member of each club’s medical team or technical staff registered individual exposure in minutes during club and national team training sessions and matches. The medical team recorded all injuries immediately after the incident and completed the injury cards (one general injury card and one groin-specific injury card, both adjusted to the consensus statement18 before the 2005/6 season). From the 2005/6 season, information about diagnostic investigations conducted was also reported, and confirmation of any clinical examination, x ray, ultrasonography (US) and MRI was requested on the injury card along with space to enter free text. In addition, from the 2006/7 season, it became mandatory to specify on the groin injury card whether the injury was due to trauma or overuse. Injury cards and attendance records were sent to the study group once a month.

Definitions

Definitions are outlined in table 1. Reinjury was defined as an injury of the same type and location as a previous injury that occurred within 2 months of a player’s return to full participation, later referred to as “early recurrence” in the consensus statement.18 A recordable hip/groin injury was defined as an injury located to the hip (hip joint or surrounding soft tissues) or groin (junction between the anteromedial part of the thigh, including the proximal part of the adductor muscle bellies, and the lower abdomen) resulting from playing football and leading to a player being unable to fully participate in future training or match play (ie time-loss injury). Each injury was coded according to a modified version of the Orchard Sports Injury Classification System (OSICS) 2.0,20 which was established when the study started. The diagnoses enthesiopathy, groin tendinosis, femoroacetabular impingement and groin nerve entrapment were added by the study group.

Statistical analysis

All quantitative variables were analysed using a one-way, factorial analysis of variance (ANOVA) and are presented as mean values (SD). The χ² test was used to analyse qualitative variables. Injury incidence is expressed as the number of injuries per 1000 h of exposure and presented with 95% CIs. A Z test was used when analysing incidences between groups. To analyse the influence of age on hip/groin injury incidence, players were divided into four age groups: <21, 21–25, 26–30 and >30 years. The significance level was set at 5% (p<0.05).

Results

The mean age of the players was 26 (4), ranging from 16 to 40 years. The total exposure was 566 000 h (475 000 training hours and 91 000 match hours). A total of 628 hip/groin injuries were recorded, representing between 12 and 16% of all injuries per season (table 2).

The mean number of groin injuries per club was 7.2 (4.3) per season. The total groin injury incidence was 1.1/1000 h (95% CI 1.0 to 1.2) and was significantly higher during match play than training (3.5/1000 (3.1 to 4.0) vs 0.6 (0.5 to 0.7), p<0.001). Ninety-four of the recorded groin injuries were reinjuries, accounting for almost 15% of all groin injuries.

Traumatic injuries constituted 66 (27%) and overuse injuries 175 (73%) of the 241 groin injuries recorded in the 2006/7 season.

The mean absence because of groin injury (all injuries included) was 15 (20) days; 16 (17) days for traumatic injuries and 14 (19) days for overuse injuries (p = 0.45). The mean absence for reinjuries was significantly longer than for index injuries: 23 (27) versus 14 (18) days (p<0.0001).

Eighteen different diagnoses were registered in the hip/groin injury category, the most frequent being adductor injury (64%), followed by hip flexor/iliopsoas injury (8%) (table 3).

There was no significant difference in groin injury incidence between age groups (p = 0.4 to 0.9). It was between 1.0 and 1.2/1000 h in the different age groups.

The preferred kicking leg was reported in 518 cases (players using right and left foot equally were excluded, as were bilaterally injured players), and injury side was analysed with respect to leg dominance. We found that 295 (57%) groin injuries affected the dominant side and 223 (43%) the non-dominant (p = 0.89); thus, leg dominance did not affect injury side.

Figure 1 shows the monthly variance of groin injury incidence. March was the month with the highest incidence of groin injuries throughout the seven seasons, followed by October and November.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Groin injury incidence during training and match play, monthly distribution.

Investigations and correlation to diagnosis

US and MRI were the two most commonly used investigations (table 4). Other investigations used were plain x ray (n = 16), bone scan (n = 3) and electromyography (n = 1). Adductor injuries and hip flexor/iliopsoas injuries were further analysed for differences in diagnostic frequency between different investigative modalities. This indicated that adductor injury was a more common diagnostic entity among injuries diagnosed with US or MRI than among injuries that were diagnosed by clinical examination only (p<0.0001). No such difference was seen for hip flexor/iliopsoas injuries (p = 0.30). Remaining diagnoses were not analysed considering the small number of cases.

Discussion

The main findings in the present study were that hip/groin injuries were a substantial problem in professional football and that more than half of the injuries resulted in the player being sidelined for more than 1 week. Another important finding was that reinjuries to the hip/groin constituted every sixth injury and caused significantly longer absences than index injuries.

Injury incidence

Consistent with other recent studies,3 4 5 6 12 13 14 21 12–16% of all injuries acquired in professional football throughout a season were to the hip/groin. The incidence of injury in this study was consistent between seasons. To the best of our knowledge, this has not been shown before, and it indicates that groin injury is not an increasing problem in professional football.

Injury severity and reinjury

We found that 41% of groin injuries were classified as moderate and 12% as severe, which supports the earlier findings previously published for the first season of this cohort.13 Apparently, hip/groin injuries result in a rather long period of absence, and preventive training would be beneficial. Scientifically validated prevention programmes are lacking, but exercises to increase joint movement22 and strengthen core and groin/gluteal muscles could be important.23 It is also plausible that increasing player and staff awareness (ie, avoid playing with groin pain) could be an important factor.

Previous studies focusing on high-level footballers have reported a relatively high recurrence rate for groin injuries,5 14 ranging from 31 to 50%. In the present study, we found that 15% of all groin injuries were reinjuries. This could partly be due to rather small numbers of analysed groin injuries in the previous studies, as well as increasing knowledge among the medical staff of elite clubs about treating groin injuries cautiously to minimise the risk of reinjury. Better organised medical support in Champions League clubs, as well as larger squads than in teams playing at lower levels, may give the opportunity to thoroughly rehabilitate and rest injured players. In this study, a time limit of 2 months was used to define a recurrent injury, while other studies define these injuries as any other injury in the same location during the same season5 or earlier in the player’s career,14 which could partly explain the discrepancies. Our finding that a reinjury in the groin area caused a significantly longer absence than the index injury further underlines the importance of a correct and quick diagnosis, as well as full rehabilitation before returning to play.

Injury pattern and investigations

Adductor-related injury was by far the most common diagnosis, representing almost three-quarters of all cases together with iliopsoas-related injuries. This is in concordance with earlier studies.3 11 US and/or MRI was used in 68% of the cases in the adductor-related injury group, while the same investigations were used in less than 50% of cases in the iliopsoas-related injury group in the 2005/6 to 2007/8 seasons. Diagnosis of almost every third adductor-related and every second iliopsoas-related injury was based solely on clinical examination. Adductor-related injuries were diagnosed more frequently with US (79%) or MRI (69%) than by clinical examination only (50%), indicating that adductor-related injury is not overestimated when based on clinical examination only in the studied cohort. However, we did not have access to the MRI investigations or the statements of the radiologists responsible, therefore some investigations may have been normal and the diagnosis based on clinical symptoms anyway. Thus, overestimation of some diagnoses cannot be ruled out.

The third most common entity was unspecified groin pain, representing 5% of the total number of hip/groin injuries. We interpret this as meaning that groin injuries are a troublesome area in clinical assessments and that there is a lack of documented guidelines for examining and diagnosing groin injury. Sportsman’s hernia could constitute a part of this category, argued by some authors as a common but underdiagnosed source of unclear groin pain.8 In the present study, sportsman’s hernia was diagnosed in 4% of the cases. Moreover, only 16 plain x rays were conducted, a number that arguably could be too low considering the increasing knowledge of hip joint anomalies causing chronic groin pain. In recent years, femoroacetabular impingement and labral tears have gained interest as differential diagnoses for groin pain. We found a relatively small number of such cases, but it is possible that the increased awareness and knowledge about this diagnostic entity along with improvements of MRI and arthroscopic techniques will result in an increased number of reported cases in the future.

Availability and user knowledge as well as local traditions influence the use of the different investigative modalities. The medical staffs in several of the investigated teams have access to US on a daily basis and are using it both to diagnose injuries and to monitor the rehabilitation process. Arguably, MRI is less user-dependent and has recently been shown to have an excellent level of specificity for groin injuries,24 but no study has yet shown that MRI is superior to US in diagnosing hip/groin injury.

Seasonal variation

March was found to be the month with the highest incidence of groin injuries, followed by October and November. This finding correlates with exposure to match play. A squad rotation policy, low-intensity training in these periods and thorough planning of the match schedule could be of value. May had the lowest incidence of all months during the competitive season, corresponding to the end of the season with lower match-play intensity.

Study strengths and weaknesses

The obvious strength of this study is that it was conducted with a prospective design, avoiding the risk of recall bias. Being conducted over several seasons in a homogenous population, to our knowledge our study comprises one of the largest samples of prospectively recorded hip and groin injuries in sports to date. The method used to record and classify data followed the consensus statement18 and has successfully been implemented in previous studies, making relevant comparisons between studies possible. Another strength of this study is that the hip/groin injuries were diagnosed by experienced club medical practitioners and that many injuries were examined radiologically.

Still, a limitation of this study is the lack of common definitions and diagnostic criteria for groin injuries. To minimise the risk of inter-rater differences, we supplied each medical team with a thorough manual with definitions and study examples. Nevertheless, diagnostic procedures rely on personal, club, local and national traditions, and we had no specific criteria or time frame for referral for radiological examination. This should be considered when interpreting the results. Furthermore, since we did not have access to the radiological statements, we do not know to what extent injuries had normal findings or multiple pathology on radiological examination. To be able to validate and standardise the categorisation of injuries in future studies, the MRI statements should be collected, or preferably the images could be reviewed by an independent radiologist.

Finally, we also recognise the risk of underestimating some cases of groin pain, since only injuries resulting in absence from training or match play were registered. Due to the overuse nature of many groin injuries, it is often possible to play for some period with hip/groin pain without time loss, and these cases were not registered under the current injury definition.

Conclusion

Groin injuries are common in professional football, the incidence studied over several seasons being consistent. More than 50% of groin injuries cause absences of more than 1 week, and reinjuries cause significantly longer absences than index injuries, emphasising the importance of quick and correct diagnosis as well as thorough rehabilitation before return to play. Diagnosis still relies to a rather large extent on clinical examination, and the value of radiological examinations and the use of independent radiologists to further sharpen diagnostics should be investigated further.