*Address correspondence to Karsten Knobloch, PhD, MD, Plastic, Hand and Reconstructive Surgery, Hannover Medical School, Carl-Neuberg-Str. 1, 30625 Hannover, Germany (e-mail: [email protected]).
No potential conflict of interest declared.
The American Journal of Sports Medicine, Vol. 37, No. 7 DOI: 10.1177/0363546509333012
© 2009 The Author(s)
Program for Hamstring Muscle and Patellar
and Achilles Tendon Injuries
An Intervention Study in Premier League Female Soccer
Robert Kraemer, MD, and Karsten Knobloch,* PhD, MD
From Plastic, Hand and Reconstructive Surgery, Hannover Medical School, Hannover, Germany
Background: A soccer-specific balance training has been shown to decrease injury incidence of the anterior cruciate ligament and ankle sprains in randomized controlled trials. However, hamstring injuries and tendinopathy remain significant issues in soccer. Hypothesis: Proprioceptive training can reduce the incidence of hamstring muscle injuries and tendinopathy in elite soccer. There is a dose-effect relationship between balance training duration and injury incidence.
Study Design: Cohort study; Level of evidence, 3.
Methods: Twenty-four elite female soccer players (body mass index, 21.7 ± 1.2; age, 21 ± 4 years) of a German premier league soccer team were prospectively included. Starting in January 2004, an additional soccer-specific proprioceptive multistation training was initiated over 3 years. Injury data/1000 hours of exposure with documentation of all occurred injuries, detailed train-ing, and match exposure data as well as time loss data were 100% complete.
Results: At the end of the 3-year proprioceptive balance training intervention, noncontact hamstring injury rates were reduced from 22.4 to 8.2/1000 hours (P = .021), patellar tendinopathy from 3.0 to 1.0/1000 hours (P = .022), and Achilles tendinopathy from 1.5 to 0.0/1000 hours (P = .035). There was no effect of balance training on contact injuries. Mean time loss of all assessed injuries significantly decreased from 14.4 days during the control period to 1.5 days during intervention periods (P = .003). The more minutes of balance training performed, the lower the rate of overall injuries (r = –0.185, P = .001), hamstring injuries (r = –0.267, P = .003), patellar tendinopathy (r = –0.398, P = .02), and gastrocnemius strains (r = –0.342, P = .002).
Conclusion: Soccer-specific balance training (protective balancing) can reduce noncontact hamstring injuries and patellar and Achilles tendinopathy. A dose-effect relationship between duration of balance training and injury incidence is evident. A proprio-ceptive training program reduced the rehabilitation time in noncontact injuries, which warrants further investigation.
Keywords: balance; proprioception; tendon; injury; prevention; soccer Soccer still enjoys great popularity with about 260 million
soccer players worldwide.6 Because of this growing popu-larity, it is likely that soccer-associated injuries and over-use complaints will gain in importance. Gender seems to play a role among female soccer players compared with male athletes, especially regarding ACL injuries.
Injury incidence rates in female soccer players in the German National League during competitive games are higher than are injury rates during training.8 However,
during the 2004 Olympic tournament in Athens, female soccer was associated with 70 injuries per 1000 playing hours, slightly lower than in male soccer players with 73 per 1000 player hours.13 During the female soccer competi-tion of that tournament, 30 time-loss injuries per 1000 player hours were documented with an overall rate of 79% contact injuries. Although the numbers are varying, female soccer players exhibit a 2-fold to 8-fold elevated risk to rupture their ACLs in contrast to male athletes.22 Hormonal variations with higher estrogen levels have been blamed for the higher injury rate among female athletes; however, this is controversial.22
Concerning ankle sprain injuries, lack of balance abili-ties in athletes is still an issue. Furthermore, loading abil-ity of the lower extremabil-ity during soccer-specific exercises seems to reflect gender differences.20 In the meantime, several preventive programs have been introduced that
modify the injury risk of ACL rupture or ankle sprain inju-ries.19 In one of the first randomized controlled studies, Caraffa et al4 from Perugia demonstrated a significant reduction in ACL rupture from 1.15 to 0.15 per 1000 hours of exposure among male soccer players.
Another randomized controlled study in high school soc-cer and basketball athletes found a significant difference in ankle sprain injuries between a balance training group and a control group. The number of ankle sprains in 765 athletes was 1.13 per 1000 hours of exposure in the bal-ance training group versus 1.87 per 1000 hours of game exposure in the control group.16 The intervention group performed a 5-phase balance training program. Phases 1 through 4 consisted of 5 exercise sessions per week for 4 weeks before the start of the season. In phase 5 (mainte-nance phase), the subjects performed the program 3 times a week for 10 minutes throughout the competitive season. In all phases, each exercise was performed for 30 seconds.
As muscle injuries are a major issue not only in soccer, the influence of an additional balance training interven-tion regarding these injuries still remains to be deter-mined.12 A recent study in male soccer demonstrated a significantly higher rate of muscle injury in players with untreated strength imbalances in comparison with players showing no imbalance in preseason. An isokinetic inter-vention normalizing the isokinetic parameters even reduced the risk factor for injury to that observed in play-ers without imbalances.5 We hypothesize that an addi-tional soccer-specific proprioceptive training intervention will reduce the injury rate of hamstring muscle injuries, a common type of injury in female soccer players, depending on the duration of the balance training intervention.
Interestingly, neuromuscular gender-specific differences seem to exist regarding the biomechanics of the gastrocne-mius muscle in soccer players. Female players demonstrate a greater lateral gastrocnemius activity and a mediolat-eral gastrocnemii imbalance that are not present among male players during soccer-specific exercises.15
Given this observation and the relevant number of over-use injuries of the Achilles and patellar tendons,9 we also decided to determine the effect of proprioceptive training on patellar and Achilles tendinopathy. We hypothesized that an additional soccer-specific balance training inter-vention would affect overuse injury rates of tendon inju-ries. To prove the aforementioned hypotheses, we chose a prospective crossover cohort study design over 3 years with the German premier league female soccer team of the FC Bayern Muenchen.
METHODS
Injury Definitions
An injury is defined according to the recent consensus article for injury definitions as any physical complaint sustained by a soccer player that results from a soccer match or soccer training, irrespective of the need for medi-cal attention or time loss from soccer playing activities.3,11
An injury that results in a soccer player receiving medical attention is referred to as a medical attention injury. An injury that results in a soccer player being unable to take full part in future soccer training or competition is defined as a time-loss injury.
Contact injuries are defined either as injuries resulting from accidental opponent contact or as injuries resulting from foul play by the injured player herself or foul play by an opponent player.
On the other hand, noncontact injuries are defined as injuries that occurred without any opponent soccer player movement in correlation with the injury incidence. Furthermore, noncontact injuries include overuse injuries that occur over a certain time period.
Assessment of Data
Injury data of 24 premier league soccer playing women were assessed prospectively from the beginning of season 2003/2004 until the end of season 2005/2006. The first half of season 2003/2004 was defined as the control period with no additional balance training intervention performed. During all seasons, duration of training and competitive games were prospectively documented in minutes for every participating soccer player. Furthermore, detailed data were assessed regarding injury occurrence with the clinical diagnosis and duration of time loss by the medical team (physical therapists and/or team physician).
The intervention period with a soccer-specific balance training began with the second half of season 2003/2004 for 2.5 years. The duration of the balance training inter-vention was documented in minutes in addition to the regular soccer training. The control period in 2003 started with 24 female soccer players aged 21 ± 4 years with a mean body mass index of 21.7 ± 1.2 (height, 171 ± 4 cm; weight, 63 ± 5 kg) and a body fat of 26.9% ± 1.2%.
All injuries corresponding to the above-mentioned defi-nitions were recorded using a standard form containing the following information:
1. date of injury;
2. when the injury occurred:
(a) during competition or official training (training and warm-up performed as part of the official competition format)
(b) during training
(c) during other forms of training (strength, stamina, or other activity);
3. if the injury was new or a recurrence;
4. if the athlete continued the match or training after the incident or not;
5. the injury mechanism;
6. injury type (concussion, contusion, strain, disloca-tion, fracture, skin abrasion/wound, other); and 7. anatomical location and side.
As mentioned above, duration of training, match, and bal-ance training time were documented for every half-season during control and intervention periods.
Balance Training
From the second half-season 2003/2004, a balance and coordination training intervention was implemented in addition to the regular soccer training activities of the premier league soccer players. The proprioceptive training was scheduled and performed on a regular weekly basis, and duration in minutes was documented. However, because of compliance or other circumstances, the sched-uled balance training intervention was not equal through-out the 2.5 years of observation. Therefore, the detailed amount of balance training in each half-season was docu-mented in minutes. The basic balance training principles were soccer-specific multistation exercises according to the “protective balancing” principles. Progression in level of difficulty from easy to complex was a main feature of the exercises. The duration of each exercise was between 15 and 30 seconds. Exercises were performed with nonfa-tigued muscles after a short regular warm-up of about 10 to 15 minutes. As an essential part of the protective bal-ancing program, the soccer players had to wear multistud-ded soccer shoes on the different balance grounds to simulate the soccer-specific environment during the prop-rioceptive training as much as possible.
The following exercises were implemented: (1) foot stand on right and left foot; (2) jump forward in single-foot stand with flexed knee at landing and balancing; (3) jump backward in single-foot stand and balancing; (4) row jumping single foot; (5) row jumping bipedal; (6) obstacle course forward and backward; (7) obstacle course sideways; (8) bipedal jumping on forefoot; (9) side-ways jumping in single-legged stand; (10) sitting on a wobble board with balancing torso; (11) jumping forward over a line, landing with flexed knees, and balancing; and (12) standing on both hands and feet with diagonal balanc-ing. All exercises were performed with no additional weight; that is, players had to bear only their own body weight on 1 or 2 legs or all extremities depending on the exercise. In addition, balance training was implemented in soccer-specific match play training on balance boards.
Training and Intervention Duration
Duration of training and match times during the control period in 2003 was 2010 hours. Duration of training and match times during the first intervention period in 2004 was 2120 hours with 990 minutes of balance training. During each season of 2004/2005, 2004 and 1434 hours of training and match were performed, with 1080 and 450 minutes of balance training during each season period, respectively. During season 2005/2006, 2421 and 2100 hours of training and match were performed, respectively, with 720 minutes of balance training in 2005 and 840 minutes in 2006. All soccer players performed the same interventional training.
Statistics
As recommended by the injury consensus statement,11 we calculated the injury incidence per 1000 hours of exposure
to training and match for each half-season. An indepen-dent-samples t test was applied for parametric, ordinal variables of nonconnected samples regarding mean time loss. Chi-square and the Fisher exact test were applied on nominal data regarding injury incidence. Correlation analysis with Pearson correlation coefficient was per-formed between the duration of balance training (minutes) compared with the overall noncontact injury rate, the non-contact hamstring injury rate, and the patellar and the Achilles tendon injury rate. A P < .05 was considered sta-tistically significant. The SPSS statistical software pack-age 16.0 for Windows (SPSS Inc, Chicago, Illinois) was used for statistical analysis.
RESULTS
Overall Injury Rates
Overall injury rate (Figure 1 and Table 1) during the control period before the intervention program was 47.7/1000 hours of training and match exposure in the year 2003. In detail, 13.4/1000 hours (28%) injuries were caused by foul play and 34.3/1000 hours (72%) occurred without foul play. After ini-tiation of the intervention program, noncontact overall injuries decreased by 29% (NS [not significant]) and contact injuries only by 8% (NS) within the first 6 months of balance training. One year after training implementation, noncon-tact injuries had decreased significantly by 65% from 34.4/1000 hours to 12.0/1000 hours (P = .021) and remained constant (12.8/1000 hours; –63% vs baseline; P = .025). Overall, the mean injury rate of all noncontact injuries dur-ing all intervention periods significantly decreased by 42% (P = .045) versus the control period, whereas contact injury rates remained equal to the control period in 2003.
Figure 1. Noncontact injury rate per 1000 hours of exposure to soccer activities with and without time-loss injuries during the control period and the 5 balance training periods com-pared with the duration of balance training during the inter-vention periods. Injury rate decreases were significantly reciprocal to duration of balance training (correlation coeffi-cient Pearson r = –0.185, P = .001).
Muscle Injuries
During the control period in 2003, 23.4 muscle injuries per 1000 hours of exposure occurred, without any muscle injury after foul play. Overall muscle injuries during the intervention period were significantly reduced from 23.4/1000 hours to 12.7/1000 hours of exposure (P = .032).
Hamstring injuries (Figure 2) occurred most often, with an injury rate of 12.4/1000 hours of training and match
TABLE 1
Overall Injury Rate per 1000 Hours of Exposure to Soccer Activities With and Without Time Loss Injuriesa
Injuries Overall: No Foul vs Foul/ 1000 Hours
of Exposure FoulNo Foul FoulNo Foul FoulNo Foul FoulNo Foul FoulNo Foul FoulNo Foul FoulNo Foul Hamstring strain 11.9 b 0 9.0b 0 5.0 0 9.5 0 6.6 0 4.8b 0 7.0 ± 2.0c 0 Hamstring rupture 0.5 0 0 0 0 0 0 0 0.8 0 0.5 0 0.3 ± 0.3 0 Groin pain 1.0 0 0 0 0 0 0 0 0 0 0 0 0 0 Back muscle strain 3.5 b 0 3.3 0.5 1.5d 0 3.5d 0.5 0.8b 0 1.0 0 2.0 ± 1.2c 0.2 ± 0.2 Back muscle rupture 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Contusion 0.5 10.0 0.5 5.2 0 12.0 0.5 5.5 0.8 10.3 0.5 9.5 0.5 ± 0.3 8.5 ± 2.7 Patellar tendon 3.0d 0 1.9 0.5 1.0d 0 2.0 0.5 1.7 0 1.0d 0 1.5 ± 0.4c 0.2 ± 0.2 Shin splint 0 0 0 0 1.0 0 0.5 0 0 0 0 0 0.3 ± 0.4 0 Achilles tendon 1.5b 0 1.9 0 1.5 0.5 1.5 0 0b 0 1.0 0 1.2 ± 0.7 0.1 ± 0.2 Knee strain 2.0b 1.5d 0b 0.5d 0.5 0 0 0.5 0 0.4 0b 0.5 0.1 ± 0.2 0.4 ± 0.2 Knee ligament rupture 0.5 0.5 0 0 0.5 0 0 0 0 0 0 0 0.1 ± 0.2 0 Knee luxation 0 0 0 0 0 0.5 0 0 0 0 0 0 0 0.1 ± 0.2 Ankle sprain 0 0.5b 0.5 4.2b 0 1.5 0.5 4.5 0 0c 0.5 1.4 0.3 ± 0.2 2.3 ± 1.7 Ankle ligament rupture 0 0.5 0 0 0 0.5 0 0 0.4 0.4 0 0 0.1 ± 0.2 0.2 ± 0.2 Shoulder instability 0.5 0 0 0 0 0.5 0 0 0 0.4 0 0 0 0.2 ± 0.2 Knee/ankle instability 2.5 b 0 0.5b 0.9 0 0 0.5 1.0 0 0 0.5b 0 0.3 ± 0.2 0.4 ± 0.5 Gastrocnemius muscle strain 6.0 b 0 6.6 0.5 0.5b 0 7.0b 0.5 1.2 0 1.9b 0 3.4 ± 2.8c 0.2 ± 0.2 Gastrocnemius muscle rupture 0.5 0 0 0 0 0 0 0 0 0 0 0 0 0 Fracture/ osteochondral lesion 0.5 0 0.5 0 0 0 0.5 0 0 0 1.0 0 0.4 ± 0.4 0 Plantar fasciitis 0 0 0 0 0.5 0 0 0 0.4 0 1.0 0 0.4 ± 0.4 0 Injuries overall/1000 hours of exposure 34.3d 13.4 24.5 12.3 12.0b,d 15.5 36.3b 18.1 12.8d 11.6 13.3 11.9 19.8 ± 9.4c 13.9 ± 2.5 Control Period 990 min 450 min 720 min 840 min
Mean Injury Rate ± SD During All Intervention Periods Season 2003/2004
1080 min
Season 2004/2005 Season 2005/2006
aData separated into injuries due to foul play and without foul play during the control period without the balance training intervention and
during the balance training periods. Time in minutes reflects intervention period with proprioceptive training duration.
bP
< .01.
cP
< .05 versus control period.
dP
< .05.
exposure before implementation of the training interven-tion in 2003. There were no contact-related hamstring injuries during all seasons from 2003 to 2006. During the first season of the training intervention, hamstring strain significantly dropped from 11.9/1000 hours to 9.0/1000 hours (P = .009) and decreased significantly during season 2005/2006 to 4.8/1000 hours of exposure (P = .001 vs base-line). The overall mean injury rate of hamstring strains during all intervention periods significantly dropped by
43% versus the control period in 2003 (P = .048). Both time-loss and no time-time-loss mean injury rates of all intervention periods significantly decreased by 83% (P = .003 vs control period) and 32% (P = .047 vs control period), respectively.
Muscle injuries of the lower leg had an incidence rate of 6.5/1000 hours of exposure in 2003, mainly with the diag-nosis of a gastrocnemius muscle strain. After the first year of balance training, incidence of noncontact gastrocnemius muscle strain decreased significantly to 0.5/1000 hours (P = .001). Gastrocnemius muscle strain remained constant at 1.2 to 1.9 injuries per 1000 hours during season 2005/2006 (P = .009 vs baseline) after an intermittent
incidence peak at the end of season 2004/2005 of 7.0/1000 hours (P = .001 vs both seasons 2004/2005 and 2005/2006) when balance training duration was the lowest over the intervention period. Mean injury rate of gastrocnemius muscle strain during all intervention periods was signifi-cantly reduced by 38% versus the control period (P = .048). Mean injury rate of non–time-loss gastrocnemius muscle strain significantly dropped 38% (P = .045). Gastrocnemius muscle strain causing time loss remained equal during all periods.
Regarding muscle injuries to the back, only noncontact back muscle strains were apparent during all seasons. Back muscle strain decreased significantly between the control period from 3.5 to 1.5/1000 hours of exposure (P = .001) after 1 year of balance training. Overall back muscle strain rate as well as back muscle strain rate without time loss during all intervention periods decreased significantly by 43% (P = .045) versus the control period. There were no time-loss injuries to the back during all seasons.
Tendon Injuries and Balance Training
During the control period 2003, the patellar tendon was injured 3.0 times per 1000 hours as a noncontact injury without any contact injuries. Noncontact injury rate of the patellar tendon decreased 1 year after the implementation of the training program and was statistically significant (3.0/1000 hours vs 1.0/1000 hours; P = .022). Noncontact injury rate of the patellar tendon remained low (1.0/1000 hours) until the end of season 2005/2006 versus 2003 (P = .022). Regarding the mean injury rate of the patellar ten-don during all intervention periods, there was a significant reduction by 50% (P = .038). Noncontact time-loss injuries (Figure 3) to the patellar tendon were reduced significantly during the intervention periods by up to 100% (P = .046).
During the control period 2003, there were only noncon-tact injuries with an incidence rate of 1.5/1000 hours of training and game exposure. Noncontact Achilles tendon injuries were prevented during the balance training imple-mentation (1.5/1000 hours vs 0/1000 hours; P = .035). Mean injury rate of the Achilles tendon causing time loss during all intervention periods was significantly reduced versus the control period in 2003 by 90% (P = .046).
Ligament Injuries
The main injury to the knee was a noncontact ligament strain, which significantly decreased from 2.0/1000 hours to 0.0/1000 hours of exposure (P = .001) after implementa-tion of the addiimplementa-tional balance training intervenimplementa-tion. There were also significantly fewer contact injuries to the knee ligaments after the first 3 months of the balance training intervention (1.5 vs 0.5/1000 hours; P = .011). During season 2005/2006, knee strain injuries did not occur again (P = .001 vs control period). Mean knee strain injury rate during all intervention periods dropped significantly by 95% (P = .004) versus the control period. During the bal-ance training intervention periods, time-loss knee strain injuries significantly decreased by 100% (P = .021). Figure 2. Hamstring injury rate per 1000 hours of exposure
to soccer activities with and without time-loss injuries during the control period and the 5 balance training periods com-pared to the duration of balance training during the interven-tion periods. Hamstring injury rate decreases were significantly reciprocal to duration of balance training (correlation coeffi-cient Pearson r = –0.267, P = .003).
Figure 3. Patellar tendon injury rate per 1000 hours of expo-sure to soccer activities with and without time-loss injuries during the control period and the 5 balance training periods compared with the duration of balance training during the intervention periods. Patellar tendinopathy rate decreases were significantly inverse to duration of balance training (cor-relation coefficient r = –0.398, P = .020).
Ankle sprain due to foul play increased during the first period of the training intervention in 2003/2004 from 0.5 to 4.2/1000 hours of exposure (P = .000) and dropped to 0.0/1000 hours during season 2005/2006 (P = .000). Mean ankle sprain contact injury rate during all intervention periods significantly increased 5-fold versus the control period in 2003 (P = .008). Noncontact ankle sprain remained stable at an injury incidence rate of 0.0 to 0.5/1000 hours during all seasons.
Mean Time Loss
Mean time loss of all noncontact injuries (Figure 4 and Table 2) during all intervention periods significantly decreased from 14.4 days to 1.5 days (P = .003). In detail, the mean time loss of hamstring strain decreased signifi-cantly by 81% (P = .001). Time loss due to Achilles tendi-nopathy (12.0 vs 0 days, P = .009) and ligament strain of the knee (10.2 vs 0 days, P = .031) did not occur during the intervention period. Apparent changes in mean time loss of injuries due to foul play were not significant.
Correlation of Balance Training and Injury Rates
We found a significant correlation between the duration of the additional training intervention and the overall inci-dence of injuries during all seasons (correlation coefficient Pearson r = –0.185, P = .001). Overall injury rate increased from 12.0 to 36.3/1000 hours of exposure during a 58% reduction in balance training duration during season 2004/2005. After intensifying the balance training, injury rates again dropped to 12.8 and 13.3/1000 hours during season 2005/2006.
The more minutes of balance training performed, the lower the overall injury rate (r = –0.185, P = .001), ham-string injuries (r = –0.267, P = .003), patellar tendinopathy (r = –0.398, P = .02), and gastrocnemius strains (r = –0.342,
P = .002) (Table 3). Achilles tendinopathy was not
corre-lated with the duration of balance training (r = –0.012,
P = .957), nor were ankle sprains with the amount of
balance training (r = 0.279, P = .055).
DISCUSSION
The main findings of this interventional study were as fol-lows: soccer-specific balance training (protective balanc-ing) can reduce noncontact hamstring and back muscle injuries in parallel with knee ligament injuries. In addi-tion, overuse injuries of the patellar and Achilles tendon can be decreased by an additional proprioceptive training intervention depending on the balance training duration.
Current scientific data report a reduction of injuries of the ACL and the ankle joint after a balance training pro-gram. A prospective controlled trial including 80 male soc-cer players found an additional proprioceptive training as an effective strategy for reduction of ankle sprain injuries.17,19 Furthermore, the same trial compared 3 dif-ferent prevention strategies regarding reduction of ankle sprain injuries. Effects of the proprioceptive training inter-vention were superior to strengthening of the peroneal muscles and the use of orthoses. Injury incidence of non-contact ankle sprains in our study showed a low injury rate over 3 years without relevant changes. It is likely that our initial low injury rate of ankle sprain resulted from a lower risk factor profile among our premier league female soccer team.8 Current literature propagates proprioceptive training as a beneficial intervention regarding knee strain reduction.4 According to that, our proprioceptive training intervention showed a significant decrease in knee strain injuries during the interventional periods as well.
Interestingly, we found a significant decrease of hamstring time-loss injuries as both minor and severe injuries, that is, time-loss injuries, during the intervention periods. Injury rate was reciprocal to the duration of the performed balance training. In other words, the more balance training per-formed, the fewer hamstring injuries occurred. This is of particular interest because hamstring injuries account for a significant number of injuries in soccer.6 Recently, it was found that the rate of hamstring muscle injury among pro-fessional soccer players was significantly increased in sub-jects with untreated strength imbalances in comparison with players showing no imbalance in preseason.5 The risk of injury remained significantly higher in players with strength imbalances who had subsequent compensating training but no final isokinetic control test than in players without imbal-ances. Conversely, normalizing the isokinetic parameters within players suffering from isokinetic imbalance even reduced the injury incidence compared with that observed in players without imbalances. Eccentric training has been reported to reduce the number of hamstring injuries in soc-cer players, whereas flexibility training alone was not able to reduce this type of injury in a recent cohort study.1
We found that a comprehensive balance training approach using soccer-specific proprioceptive tasks in female soccer players could significantly reduce the number of noncontact Figure 4. Time-loss hamstring injury rate per 1000 hours of
exposure to soccer activities of time-loss injuries during the control period and the 5 balance training periods compared with the duration of balance training during the intervention periods.
hamstring and back muscle injuries depending on the balance training load. Furthermore, a close relationship to the amount of balance training minutes performed could be demonstrated. Thus, proprioceptive training might be of additional benefit in the prevention of hamstring and back muscle injuries in soccer players. Further studies are warranted to elucidate the role of a potential combination of eccentric training and a soccer-specific proprioceptive training intervention in soccer players in a randomized
controlled trial. Furthermore, the lower threshold of balance training has to be established to fulfill a preven-tive effect.
Compliance and Proprioceptive Training
As with every intervention, the compliance needed to achieve the expected results is of utmost importance. As far as prop-rioceptive training is concerned, each balance training TABLE 2
Mean Time Loss per Injury During the Control Period Without the Balance Training Intervention and the Balance Training Periodsa
Injuries Overall: No Foul vs Foul/1000 Hours of
Exposure, d FoulNo Foul FoulNo Foul FoulNo Foul FoulNo Foul FoulNo Foul FoulNo Foul FoulNo Foul Hamstring strain 10.5 0 2.0 0 0 0 2.0 0 3.8 0 2.0 0 2.0 ± 1.3 0.0 Hamstring rupture 70.0 0 0 0 0 0 0 0 21.0 0 0 35.0 4.2 ± 9.4 7.0 ± 15.7 Groin pain 28.0 0 0 0 0 0 0 0 0 0 0 0 0 0.0 Back muscle strain 0 0 0 0 0 0 0 0 0 0 0 0 0 0.0 Back muscle rupture 0 0 0 0 0 0 0 0 0 0 0 0 0 0.0 Contusion 2.7 2.7 0 3.0 0 1.8 0 3.0 0 23.9 0 0 0 6.3 ± 9.9 Patellar tendon 21.0 0 0 0 0 0 0 0 28.5 0 0 0 5.7 ± 12.7 0.0 Shin splint 0 0 0 0 0 0 0 0 0 0 0 0 0 0.0 Achilles tendon 12.0 0 0 0 0 2.0 0 0 0 0 0 0 0b 0.4 ± 0.9 Knee strain 10.2 10.2 0 56.0 0 0 0 56.0 0 7.0 0 42.0 0c 32.2 ± 26.9 Knee ligament rupture 83.7 83.7 0 0 0 83.5 0 0 0 0 0 0 0 16.7 ± 37.3 Knee luxation 0 0 0 0 0 0 0 0 0 0 0 0 0 0.0 Ankle sprain 0 7.0 0 6.3 0 0 0 6.3 0 0 0 0 0 2.5 ± 3.5 Ankle ligament rupture 0 21.0 0 0 0 56.0 0 0 0 35.0 0 0 0 18.2 ± 26.0 Shoulder instability 21.0 0 0 0 0 7.0 0 0 0 0 0 0 0 1.4 ± 3.1 Knee/ankle instability 42.0 0 0 35.0 0 0 0 35.0 0 0 0 0 0 14.0 ± 19.2 Gastrocnemius muscle strain 1.0 0 0 7.0 0 0 7.0 0 0 0 0 0 1.4 ± 3.1 1.4 ± 3.1 Gastrocnemius muscle rupture 21.0 0 0 0 0 0 0 0 0 0 0 0 0 0.0 Fracture/ osteochondral lesion 0 0 18.0 0 0 0 18.0 0 0 0 70.0 0 21.2 ± 28.7 0.0 Plantar fasciitis 0 0 0 0 0 0 0 0 0 0 0 0 0 0.0 Injuries overall/1000 hours of exposure 14.4c 5.9 1.0c 4.7 0.0 6.5 1.3 4.4 2.3 2.9 3.1c 3.3 1.5 ± 1.2 4.4 ± 1.4
Control Period 990 min 450 min 720 min 840 min
Mean Injury Rate ± SD During All Intervention Periods Season 2003/2004
1080 min
Season 2004/2005 Season 2005/2006
aTime in minutes reflects intervention period with proprioceptive training duration. bP < .05 versus control period.
minute has to be justified with the overall general soccer training with the trainer. A recent randomized intervention trial among 508 soccer players revealed a poor compliance of only 20% to 30% for proprioceptive training.7 A large, cluster randomized trial studied the effect of a set of exercises (“the 11”) to prevent injuries in female youth soccer.21 The inter-vention group was taught the 11 exercises for core stability, lower extremity strength, neuromuscular control, and agility, to be used as a 15-minute warm-up program for soccer train-ing over an 8-month season. No difference was observed in the overall injury rate between the intervention (3.6 injuries/1000 hours, confidence interval, 3.2-4.1) and control group (3.7 injuries/1000 hours, confidence interval, 3.2-4.1; relative risk = 1.0, confidence interval, 0.8-1.2; P = .94) or in the incidence for any type of injury. During the first 4 months of the season, the training program was used during 60% of the training sessions, but only 14 of 58 intervention teams completed more than 20 prevention training sessions. The authors concluded that they could not observe any effect of the injury prevention program on the injury rate, most likely because the compliance with the program was low. Other studies have not yet reported in detail on the training load of the proprioceptive intervention. We demonstrated a close relationship between the balance training load and the injury rate. In seasons with a high balance training load, a low over-all number of noncontact injuries was encountered, whereas in those months with less balance training, a substantially higher number of injuries was encountered. This holds true for hamstring and back muscle injuries, patellar tendinopa-thy, as well as gastrocnemius muscle strains. As our interven-tion program was made a part of the regular soccer training of the female soccer team, the compliance of our intervention was 100% with the documented mean seasonal training duration.
Balance Training and Tendon Injuries
Eccentric training and stretching in Achilles tendinopathy as well as eccentric training on a decline board in patellar tendinopathy have been shown to significantly decrease chronic tendon pain.10,18 Recently, a randomized controlled trial was published studying the effect of an eccentric training and stretching program to reduce Achilles and
patellar tendinopathy in Danish soccer players.9 The authors found that a prophylactic eccentric training and stretching program reduced the risk of developing ultrasonographic abnormalities in the patellar tendons but had no positive effects on the risk of injury. Currently, there is only 1 trial focusing on the proprioceptive capacity in tendinopathy. In radial epicondylitis, 1 clinical trial studied elbow propriocep-tion in patients with a tennis elbow compared with subjects with a healthy elbow.14 Both absolute error and variable error of threshold to detection of a passive movement were greater in the radial epicondylitis-diagnosed elbows than in the control elbows, and there was a tendency toward a greater absolute error of joint position sense compared with the control elbows. We found a significant reduction of both patellar and Achilles tendon injury rate after implementa-tion of an addiimplementa-tional soccer-specific proprioceptive training. Regarding the pathomechanism underlying this observa-tion, one has to speculate. Given a similar lack of propriocep-tion at the patellar and/or the Achilles level in susceptible persons, a proprioceptive training intervention might be beneficial to restore a more normal proprioceptive capacity for ligaments as well as for tendons. Sensory input might be higher in proprioceptive-trained individuals, and this might enhance muscle-tendon unit function and integration.
No Preventive Effect in Contact Injuries
Current scientific data describe an injury incidence reduc-tion of the lower extremities with proprioceptive training.16 But oftentimes, current studies do not discriminate between reduction of contact and noncontact injury incidence. Our study demonstrated a relevant reduction of certain noncon-tact injuries but could not demonstrate a decrease of overall injury rate due to foul play, as well as loss and no time-loss contact injuries. Regarding this, contact injuries do not seem to be influenced by a proprioceptive training interven-tion. Player-to-player contact is more frequent among com-petition injuries, whereas noncontact mechanisms are more common among practice injuries.23 In contrast to our train-ing intervention, Fédération Internationale de Football Association's (FIFA’s) “the 11” also features an instruction point regarding “fair play,” which indeed seems to be an important factor of possible further injury rate reduction and should always be a part of preventive training pro-grams.12 An alternative for a prevention program aiming at injury reduction of both contact and noncontact injuries was developed by an Icelandic sports medicine group. In a ran-domized trial, video scenes of several common sports injury mechanisms were selected and presented to soccer players. The soccer players analyzed the video sequences to develop and implement preventive strategies. However, during the next season, no difference was observed in injury incidence between the intervention group with video tape analysis and control groups without video analysis.2
Balance Training Duration and Injury Rates
In our study, we demonstrated a correlation between dura-tion of the addidura-tional training intervendura-tion and incidence of injuries. We encountered a 2-fold increase of hamstring, TABLE 3
Correlation Coefficient According to Pearson Between the Amount of Balance Training and the Incidence
of Hamstring Strains, Patellar Tendinopathy, Achilles Tendinopathy, and the Overall Noncontact
Injury Rate in Female Soccer Players
Correlation
Clinical Diagnosis Coefficient (Pearson) P Hamstring injury –0.267 .003 Patellar tendinopathy –0.398 .02 Gastrocnemius muscle strain –0.342 .002
Ankle sprain 0.279 .055
Achilles tendinopathy –0.012 .957 Overall injury rate –0.185 .001
a 15-fold increase of gastrocnemius muscle, a 3-fold increase of ankle, and a 3-fold increase of patellar tendon injuries during the intervention period 2004/2005 after a 58% drop in balance training intervention duration from 1080 minutes to 450 minutes. Consequently, elevated injury rates again significantly decreased after intensify-ing the balance trainintensify-ing intervention durintensify-ing the next sea-son. So far, no study has shown in detail such a correlation between the duration of an additional proprioceptive train-ing within a season and the direct effect on injury inci-dence. For the first time, our study demonstrates a reciprocal correlation between the duration of the training intervention and the noncontact injury incidence for over-all injuries, hamstring injuries, patellar tendinopathy, and gastrocnemius muscle strains.
Balance Training and Rehabilitation Time
Furthermore, there is a lack of scientific data regarding effects of eccentric and balance training on tertiary pre-vention, that is, reduction of injury reccurrence and time loss after primary injury incidence. In our data, we showed a statistical reduction of injury incidence and mean time loss after the proprioceptive training implementation. Consequently, we can assume a positive effect not only on injury incidence but also on duration of rehabilitation time after primary sports injuries. In elite sports, shortening of rehabilitation time cannot be underestimated regarding the economic influence on an elite sports club. We found an overall reduction of rehabilitation time of a given injury during the proprioceptive intervention.
Limitations
This study was designed as a prospective crossover study without a control group. A reasonable further step to prove our results with a higher level of evidence would be a prospec-tive randomized controlled trial over a perennial time. The crossover study setting was chosen because the authors of the study were asked by the soccer club for the immediate imple-mentation of a practical training program to reduce the rel-evant muscle injury rate during the last seasons in the soccer club. As our study was designed as a prospective crossover study, we did not document injury occurrence within the soc-cer team retrospectively, although injury surveillance over a longer period without proprioceptive training as the control period would have been more reliable. However, even if we encountered a higher than normal control injury rate, we clearly found a dose-response relationship between the amount of balance training (protective balancing) and the injury rate. Experience shows that compliance in interven-tional training should not be underestimated for injury out-come. Our results in minor soccer division teams must be critically scrutinized due to frequent suboptimal training conditions in minor soccer leagues, with less training time a week and without a detailed injury surveillance system by a physical therapist or doctor.
CONCLUSION
A soccer-specific balance training intervention can help to reduce hamstring and back muscle injuries. The more bal-ance training units performed, the lower the injury rates for hamstring injuries, patellar tendinopathy, gastrocne-mius strains, and overall injuries. Overuse injuries of the patellar and Achilles tendon can be reduced by an addi-tional balance intervention. Finally, we found a decrease in rehabilitation time in our observed cohort, which has not been published yet and warrants further investigation. REFERENCES
1. Arnason A, Andersen TE, Holme I, Engebretsen L, Bahr R. Prevention of hamstring strains in elite soccer: an intervention study. Scand J
Med Sci Sports. 2008;18(1):40-48.
2. Arnason A, Engebretsen L, Bahr R. No effect of a video-based aware-ness program on the rate of soccer injuries. Am J Sports Med. 2005;33(1):77-84.
3. Brooks JH, Fuller CW. The influence of methodological issues on the results and conclusions from epidemiological studies of sports inju-ries: illustrative examples. Sports Med. 2006;36(6):459-472. 4. Caraffa A, Cerulli G, Projetti M, Aisa G, Rizzo A. Prevention of anterior
cruciate ligament injuries in soccer: a prospective controlled study of proprioceptive training. Knee Surg Sports Traumatol Arthrosc. 1996;4(1):19-21.
5. Croisier JL, Ganteaume S, Binet J, Genty M, Ferret JM. Strength imbal-ances and prevention of hamstring injury in professional soccer players: a prospective study. Am J Sports Med. 2008;36(8):1469-1475. 6. Dvorak J, Junge A, Graf-Baumann T, Peterson L. Football is the most
popular sport worldwide. Am J Sports Med. 2004;32(1 suppl):3S-4S. 7. Engebretsen AH, Myklebust G, Holme I, Engebretsen L, Bahr R.
Prevention of injuries among male soccer players: a prospective, randomized intervention study targeting players with previous injuries or reduced function. Am J Sports Med. 2008;36(6):1052-1060. 8. Faude O, Junge A, Kindermann W, Dvorak J. Injuries in female soccer
players: a prospective study in the German National League. Am J
Sports Med. 2005;33(11):1694-1700.
9. Fredberg U, Bolvig L, Andersen NT. Prophylactic training in asymp-tomatic soccer players with ultrasonographic abnormalities in Achilles and patellar tendons: the Danish Super League Study. Am J Sports
Med. 2008;36(3):451-460.
10. Frohm A, Saartok T, Halvorsen K, Renström P. Eccentric treatment for patellar tendinopathy: a prospective randomised short-term pilot study of two rehabilitation protocols. Br J Sports Med. 2007; 41(7):e7.
11. Fuller CW, Ekstrand J, Junge A, et al. Consensus statement on injury definitions and data collection procedures in studies of football (soc-cer) injuries. Br J Sports Med. 2006;40:193-201.
12. Junge A, Dvorak J. Injuries in female football players in top-level international tournaments. Br J Sports Med. 2007;41(suppl 1):i3-7. 13. Junge A, Langevoort G, Pipe A, et al. Injuries in team sport
tourna-ments during the 2004 Olympic games. Am J Sports Med. 2006;34(4):565-576.
14. Juul-Kristensen B, Lund H, Hansen K, Christensen H, Danneskiold-Samsoe B, Bliddal H. Poorer elbow proprioception in patients with lateral epicondylitis than in healthy controls: a cross-sectional study.
J Shoulder Elbow Surg. 2008;17(1 suppl):72S-81S.
15. Landry SC, McKean KA, Hubley-Kozey CL, Stanish WD, Deluzio KJ. Neuromuscular and lower limb biomechanical differences exist between male and female elite adolescent soccer players during an unanticipated run and crosscut maneuver. Am J Sports Med. 2007;35(11):1901-1911.
16. McGuine TA, Keene JS. The effect of a balance training program on the risk of ankle sprains in high school athletes. Am J Sports Med. 2006;34(7):1103-1111.
17. Mohammadi F. Comparison of 3 preventive methods to reduce the recurrence of ankle inversion strains in male soccer players. Am J
Sports Med. 2007;35(6):922-926.
18. Petersen W, Welp R, Rosenbaum D. Chronic Achilles tendinopathy: a prospective randomized study comparing the therapeutic effect of eccentric training, the AirHeel brace, and a combination of both. Am
J Sports Med. 2007;35(10):1659-1667.
19. Silvers HJ, Mandelbaum BR. Prevention of anterior cruciate ligament injury in the female athlete. Br J Sports Med. 2007;41(suppl 1):i52-59.
20. Sims EL, Hardaker WM, Queen RM. Gender differences in plantar loading during three soccer-specific tasks. Br J Sports Med. 2008;42(4):272-277.
21. Steffen K, Myklebust G, Olsen OE, Holme I, Bahr R. Preventing inju-ries in female youth football: a cluster-randomized controlled trial.
Scand J Med Sci Sports. 2008 Jan 14 [Epub ahead of print].
22. Toth AP, Cordasco FA. Anterior cruciate ligament injuries in the female athlete. J Gend Specif Med. 2001;4(4):25-34.
23. Yard EE, Schroeder MJ, Fields SK, Collins CL, Comstock RD. The epidemiology of United States high school soccer injuries, 2005-2007. Am J Sports Med. 2008;36(10):1930-1937.
