http://ajs.sagepub.com/Medicine

The American Journal of Sports

http://ajs.sagepub.com/content/39/5/940The online version of this article can be found at:

 DOI: 10.1177/0363546510393269

2011 39: 940 originally published online January 6, 2011Am J Sports MedRussell J. Coppack, John Etherington and Andrew K. Wills

TrialThe Effects of Exercise for the Prevention of Overuse Anterior Knee Pain: A Randomized Controlled

  

Published by:

http://www.sagepublications.com

On behalf of: 

American Orthopaedic Society for Sports Medicine

can be found at:The American Journal of Sports MedicineAdditional services and information for    

  http://ajs.sagepub.com/cgi/alertsEmail Alerts:

 

http://ajs.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:  

http://www.sagepub.com/journalsPermissions.navPermissions:  

What is This? 

- Jan 6, 2011 OnlineFirst Version of Record 

- May 5, 2011Version of Record >>

at University of Bath - The Library on May 10, 2016ajs.sagepub.comDownloaded from at University of Bath - The Library on May 10, 2016ajs.sagepub.comDownloaded from

The Effects of Exercise for the Preventionof Overuse Anterior Knee Pain

A Randomized Controlled Trial

Russell J. Coppack,*,z BSc (Hons), MSc, John Etherington,* MB ChB, MSc, FFSEM (UK), FRCP,and Andrew K. Wills,y MSc, PhDInvestigation performed at The Centre for Human Performance, Rehabilitation and SportsMedicine, Defence Medical Rehabilitation Centre, Headley Court, Surrey, United Kingdom

Background: Anterior knee pain (AKP) is the most common activity-related injury of the knee. The authors investigated the effectof an exercise intervention on the incidence of AKP in UK army recruits undergoing a 14-week physically arduous trainingprogram.

Hypothesis: Modifying military training to include targeted preventative exercises may reduce the incidence of AKP in a youngrecruit population.

Study Design: Randomized controlled trial; Level of evidence, 1.

Methods: A single-blind cluster randomized controlled trial was performed in 39 male and 11 female training groups (median age:19.7 years; interquartile range, 17-25) undergoing phase 1 of army recruit training. Each group was randomly assigned to either anintervention (n = 759) or control (n = 743) protocol. The intervention consisted of 4 strengthening and 4 stretching exercises com-pleted during supervised physical training lessons (7 per week). The control group followed the existing training syllabus warm-upexercises. The primary outcome was a diagnosis of AKP during the 14-week training program.

Results: Forty-six participants (3.1%; 95% confidence interval [CI], 2.3-4.1) were diagnosed with AKP. There were 36 (4.8%;95% CI, 3.5-6.7) new cases of AKP in the control group and 10 (1.3%; 0.7-2.4) in the intervention group. There was a 75%reduction in AKP risk in the intervention group (unadjusted hazard ratio = 0.25; 95% CI, 0.13-0.52; P \ .001). Three participants(0.4%) from the intervention group were discharged from the military for medical reasons compared to 25 (3.4%) in the controlgroup.

Conclusion: A simple set of lower limb stretching and strengthening exercises resulted in a substantial and safe reduction in theincidence of AKP in a young military population undertaking a physical conditioning program. Such exercises could also be ben-eficial for preventing this common injury among nonmilitary participants in recreational physical activity.

Keywords: anterior knee pain; injury prevention; exercise; military training

Overuse anterior knee pain (AKP) is a diagnosis thatencompasses several conditions, including patellofemoralpain syndrome (PFPS). It is typified by pain at the frontof the knee31 and aggravated by activities causing repeti-tive and high patellofemoral compressive forces such askneeling, squatting, and running.9,11,31 Anterior knee

pain is the most common knee disorder and is a majorproblem for physically active individuals.36 It accountsfor 25% to 40% of all knee problems presenting to sportsmedicine clinics and affects 1 in 4 of active populations.3,15

Military recruits undergoing initial training experienceabrupt increases in physical activity and are thus a high-risk population for developing AKP.4,17,33 An incidence of8.7% has been reported in British Army recruits33 and 5%to 15% in non-UK military recruits.7,21,24 It is the biggestcause of attrition from recruit training17 and has a debilitat-ing effect on sufferers, curtailing or ending participation inphysical activity.5 Prevention of such injuries is thus increas-ingly important in military and civilian populations.26

There is evidence from randomized controlled trials thata multifaceted prevention program can reduce the incidenceof knee and ankle injuries in specific sports1,13,26 and thatstrengthening and stretching exercises can reduce the inci-dence of lower limb overuse injuries in military recruits.4

Evidence also supports the use of targeted exercise for the

zAddress correspondence to Russell J. Coppack, Centre for HumanPerformance, Rehabilitation and Sports Medicine, Defence MedicalRehabilitation Centre, Headley Court, Epsom, Surrey, KT18 6JW, Eng-land, UK (e-mail: russ.coppack916@mod.uk).

*Centre for Human Performance, Rehabilitation and Sports Medicine,Defence Medical Rehabilitation Centre, Headley Court, UK.

yMedical Research Council Unit for Lifelong Health and Ageing, Uni-versity College London, London, UK.

The authors declared that they had no conflicts of interest in theirauthorship and publication of this contribution.

The American Journal of Sports Medicine, Vol. 39, No. 5DOI: 10.1177/0363546510393269� 2011 The Author(s)

940 at University of Bath - The Library on May 10, 2016ajs.sagepub.comDownloaded from

therapeutic treatment of AKP.8,15,18 The rationale includesthe restoration of patellar alignment and function throughquadriceps strengthening, retraining, and stretching.25,34,35

To our knowledge, there are no randomized controlled trialsexamining the effect of these treatment-based protocols spe-cifically on the prevention of AKP.

We conducted a cluster randomized controlled trial ina military recruit population undergoing a 14-week struc-tured and incremental physical conditioning program.Our aim was to investigate the effect of a targeted exerciseintervention on the incidence of AKP during this program.

MATERIALS AND METHODS

Ethical approval was granted by the Ministry of Defence(UK) research ethics committee. All participants gavewritten informed consent. This study is registered athttp://controlled-trials.com, number ISRCTN61493628.

Study Design and Participants

We used a single-blind cluster randomized controlleddesign. All British Army recruits who enlisted at theArmy Training Centre (Pirbright, UK) between July 2006and February 2007 and passed the entry medical examina-tion were invited to participate in the study (n = 1502); nonedeclined. Before enlistment, recruits were divided intosingle-sex troops; these were our units of randomization.Each of the 50 troops (clusters) (average number of recruits[range]: 41 [22-48]) were randomly assigned to 1 of 2 groups:an AKP prevention training program (PTP) or a controlprogram. A simple randomization procedure based on acomputer-generated table of random numbers was used toallocate the intervention, and an external administratorprovided the group assignment. Physicians diagnosingAKP cases were blinded. An attempt was made to blind par-ticipants, but given the physical nature of the intervention,we refrain from calling this a double-blinded study.

Procedures

All participants were initially screened by a general practi-tioner and deemed fit to commence training. Baselineheight, weight, and body mass index (BMI) were recorded,and participants completed a self-administered question-naire to obtain a history of lower limb injury, exposure tostrength training and running before enlistment, and cur-rent and past smoking habits. Baseline fitness was alsoassessed using a 2.4-km timed running test.

All participants undertook the same 14-week phase 1basic military training consisting of 3 to 4 hours of formaldaily training activity. Key elements included endurancemarching with backpacks (15%), military field exercises(24%), running (4%), weapons and foot drill (7%), strengthand conditioning (3%), and theoretical classroom lessons(23%). Because all recruits followed the same common mil-itary syllabus with the same daily schedule and environ-mental conditions, all extrinsic risk factors that may

influence AKP other than the intervention sessions werethe same across the study arms.

The PTP and control sessions were completed duringeach physical training lesson of the common military sylla-bus. All sessions were supervised by qualified army physicaltraining instructors (PTIs). Participant blinding wasattempted through the application of dummy warm-up exer-cises for control group participants. Participants in differentgroups did not attend concurrent physical training sessionsand were instructed not to reveal information about ses-sions to the AKP outcome assessor (physiotherapist).

Intervention and Control Program

Fifteen PTIs instructed and supervised the prevention andcontrol protocols. The instructors were familiarized withthe intervention during a 1-day training workshop. Notraining was required for the control program. Written,standardized program instructions were also provided.

The PTP consisted of 8 exercises (Table 1), delivered insubsets of 4 during the warm-up and warm-down of eachformal physical training lesson (mean = 7 per week).Warm-up exercises comprised closed kinetic chain (CKC)quadriceps and gluteal strengthening exercises. Each exer-cise was performed in sets of 10 to 14 repetitions. Isometrichip abduction repetitions were measured in seconds. Partic-ipants were encouraged to focus on the quality of theirmovement, with emphasis on the position of the hip andknee in relation to the foot. These functional exercisesinvoke a strong eccentric contraction of the hip and kneeextensors.5,8,25,35 During the training period, the exerciseload increased progressively as the number of repetitionsincreased.

Previous research has shown that poor muscle flexibilityis a predisposing factor for the development of AKP.37 Foursoft tissue stretching exercises were undertaken during thewarm-down of the PTP (Table 1). We used the static stretch-ing method, 3 repetitions of a 20-second stretch. The quad-riceps were stretched in the upright standing position. Thestretch was applied by maximal flexing of the knee with thehip slightly extended. The hamstrings were stretched ina supine position by lifting the extended leg while control-ling movement of the pelvis. The gastrocnemius musclewas stretched in a standing position while leaning againsta wall 0.60 m (2 feet) away and keeping the tested leg fullyextended behind the contralateral side. Iliotibial bandstretches were performed in a supine position by crossingthe flexed knee over the contralateral leg and placing thefoot on the floor. A gentle adduction force was applied tothe flexed knee until the lateral thigh was stretched. Thecombined duration of warm-up and warm-down PTP exer-cise was 15 minutes. For practical reasons, body weight–resisted exercises were used (see Appendix, available inthe online version of this article at http://ajs.sagepub.com/supplemental/). Previous research recommends the use oftechnically less demanding exercises to enhance complianceand accommodate the low training experience of recruits.4

The control group undertook existing common militarysyllabus warm-up and warm-down exercises, including

Vol. 39, No. 5, 2011 Effects of Exercise for Preventing Overuse Anterior Knee Pain 941

at University of Bath - The Library on May 10, 2016ajs.sagepub.comDownloaded from

slow running, general upper and lower body stretching,abdominal curls, and pushup drills. These exercises alonewould be unlikely to cause AKP and were the same dura-tion as the PTP. The amount of basic training conductedwas not influenced by the introduction of either program.Although attendance at physical training sessions is com-pulsory (excluding injury, sickness, or administrative rea-sons), compliance was assessed using attendance records.

Outcomes

The primary outcome was an incident case of overuse AKPoccurring during the 14-week training period. Anteriorknee pain cases were captured through self-presentationof pain at the army medical center. Participants were ini-tially seen by a medical officer. Five experienced physio-therapists who were blinded to group assignment madea diagnosis of AKP. Before the study, a consultantphysician (J.E.) delivered a specific training session to the5 outcome assessors on the diagnosis of anterior kneepain. The inclusion criteria for an AKP diagnosis were (1)anterior or retropatellar knee pain arising from at least 2of the following: prolonged sitting, stair climbing, squatting,running, kneeling, and hopping/jumping; (2) insidious onsetof symptoms unrelated to a traumatic incident; and (3) pres-ence of pain on palpation of the patellar facets, on step downfrom a 25-cm step, or during a double-legged squat. Theexclusion criteria were signs and symptoms of meniscal orother intra-articular pathologic conditions; ligament laxityor tenderness; tenderness over the patellar tendon, iliotibialband, or pes anserinus tendons; patellar apprehension sign;Osgood-Schlatter or Sinding-Larsen-Johanssen syndrome;evidence of a knee joint effusion or hip or lumbar referredpain; a history of patellar dislocation; or other surgery orstructural damage to the knee. These criteria are sensitiveand specific for diagnosing AKP.8,34

Secondary outcomes were the occupational endpoints ofeach participant: successful completion, medical discharge(MD), discharge as of right (DAOR: a voluntary discharge

at the request of the recruit), unfit for army service(UFAS: recruits incapable of meeting the training stand-ards), backsquadding (recruits held back in training), andother (withdrawal from training for all other reasons).These were obtained using training group attendancerecords and the medical center database. The secondaryoutcomes were also used as a marker of adverse eventsbecause we would expect to see rises in backsquadding ormedical discharges if the intervention protocol were caus-ing other injuries or secondary problems.

Sample Size

The study was designed to detect a 2.5% differencebetween groups in the cumulative incidence of AKP. Thiswas based on reducing the incidence of AKP by 50% inthose cases occurring after day 14 of training (currentknowledge suggests that 10-14 days is the minimum timerequired to stimulate a muscular strength and endurancetraining adaptation).22 The total estimated sample size,assuming equal arms, was 2762 participants (type 1 error =0.05, power = 0.8). Because of military operational commit-ments, the trial was suspended early, and on termination,we had recruited 1502 participants. There were no inter-mediate or data monitoring checks while the trial wasrunning.

Statistical Analyses

We defined a ‘‘time exposed’’ to training for each individualand used the hazard ratio as our effect measure. Exposuretime was defined as the length of time an individual spentin training with his or her original training group free ofAKP. Participants were thus censored at the point theywere removed from training. Reasons for censoring werea period in rehabilitation, sickness absence, back squad-ding, an MD from the army, voluntary discharge(DAOR), or an administrative discharge (UFAS). Partici-pants who successfully completed training with their

TABLE 1Anterior Knee Pain Prevention Training Program (PTP)a

Weeks of Program

Exercise 1-3 4-6 7-9 10-12 13-14

Isometric hip abduction against a wall in standing, secb 10 10 15 15 20Forward lunges—knee over the forward foot, repetitionsb 10 12 12 14 14Single-legged step downs from a 20-cm step, repetitionsb 10 10 12 12 14Single-legged squats to 45� of knee flexion with isometric gluteal

muscle contraction, repetitionsb10 10 12 12 14

Quadriceps (hip and knee) stretches, secc 20 20 20 20 20Iliotibial band (lateral thigh) stretches, secc 20 20 20 20 20Hamstring stretches, secc 20 20 20 20 20Calf (gastrocnemius) stretches, secc 20 20 20 20 20

aAll stretches measured in seconds.bExercises performed during formal physical training session warm-up. All repetitions completed in sets of 3.cStretching exercises performed during formal physical training session warm-down.

942 Coppack et al The American Journal of Sports Medicine

at University of Bath - The Library on May 10, 2016ajs.sagepub.comDownloaded from

original group were censored at the point of exit (14weeks). For participants who developed AKP, exposurewas calculated to the point of diagnosis. The Kaplan-Meiersurvival function was estimated for the control and inter-vention arms and plotted and compared using a log ranktest. There was evidence that the baseline hazard of AKPwas not constant when tested formally in a Poisson model,and thus we used a Cox proportional hazards model to esti-mate the effect of the intervention. Within-cluster (train-ing group) correlation was accounted for using robuststandard errors. There was no evidence that the effect ofthe intervention differed between sexes (P = .18, Waldtest), so we pooled the sexes for our final models. We pres-ent an unadjusted and adjusted effect estimate, as theunadjusted Cox model from a randomized controlled trialcan in certain circumstances produce a biased estimate.For the adjusted model, we controlled for known predictorsof AKP (smoking, gender, previous injury, previous AKP).We tested for differences in the secondary outcomes usinga x2 test of general association. Stata v9.2 was used for allanalyses.

RESULTS

Baseline Characteristics

In total, 1502 recruits agreed to participate in the study;there were no refusals and no exclusions or dropouts beforeconsent. There were 759 participants (556 [73.3%] men;203 [26.7%] women) in the intervention group and 743(536 [72.1%] men; 207 [27.9%] women) in the control arm(Figure 1). The groups were well balanced at baseline interms of height, weight and BMI, social and training his-tory, and previous injuries (Table 2). The total exposuretime was 2034 recruit training months in the interventiongroup and 1824 recruit training months in the controlgroup. The mean individual compliance rate for the 2 pro-grams was 91%, and there was no evidence of a differencebetween groups (P . .05). There were no reports of painfrom the intervention exercises.

Primary Outcome

Of 1502 participants, 46 (3.1%; 95% confidence interval[CI], 2.3-4.1) were diagnosed with AKP. The median timeof diagnosis of AKP was 4.3 weeks (interquartile range[IQR], 3.1-8). Pooling the 2 groups, the incidence of AKPwas 0.012 (95% CI, 0.009-0.016) per recruit-months oftraining. In men, the incidence was 0.012 per recruit-month (95% CI, 0.009-0.017), and in women, the incidencewas 0.011 (95% CI, 0.006-0.020). There was no evidence ofa sex difference in the incidence of AKP (P . .05).

There were 36 new cases of AKP in the control group(cumulative incidence: 4.8%; 95% CI, 3.5-6.7) and 10 inthe intervention group (cumulative incidence: 1.3%; 95%CI, 0.7-2.4). The incidence in the control group was 0.020per recruit-month (95% CI, 0.014-0.027) versus 0.005(95% CI, 0.002-0.009) in the intervention group. Figure 2

shows the cumulative probability of developing AKP overthe course of training in each arm of the study. Therewas very strong evidence for a difference in this survivalfunction between groups (P \ .01, log rank test).

There was no evidence of a gender interaction in theeffect of the intervention (P = .18 from the proportionalhazard model), so we report the pooled analysis. Therewas very strong evidence of a protective effect of theintervention. The hazard (risk) of AKP over trainingwas reduced by 75% (95% CI, 52-87) in the interventiongroup compared to the control group, and this is an esti-mate of the preventable fraction in the control group(Table 3). The absolute risk difference between the con-trol and intervention groups was 20.015 per recruit-month (95% CI, 20.022 to 20.008). The unadjusted andadjusted hazard ratios for the effect of the interventionwere very similar (Table 3). After adjustment for cluster-ing, standard errors were also very similar, suggestingthat the observations within each troop were largelyindependent.

Secondary Outcomes

There was strong evidence for a difference in the secondaryoutcomes between groups (P \ .001). In the control group,25 (3.4%) were medically discharged compared to 3 (0.4%)in the intervention group. Six (0.8%) in the interventiongroup were discharged on the grounds of ‘‘unfit for armyservice’’ compared to 23 (3.1%) in the control group. Themajority of participants in both groups successfully com-pleted training, although a greater proportion from theintervention group had a successful outcome (79% vs68%), and there was no evidence to suggest a differencein voluntary discharge (DAOR) rate between groups (P .

.05) (Table 4).Of the 10 AKP cases in the intervention group, 9 (90.0%)

successfully completed training, and 1 (10.0%) wasUFAS—none were medically discharged. In the controlgroup, 10 (27.8%) participants with AKP were medicallydischarged, 6 (16.7%) were backsquadded, 2 (5.6%) wereUFAS, 2 (5.6%) were voluntarily discharged (DAOR), and16 (44.4%) successfully completed training.

DISCUSSION

Principal Findings

Our study shows that a simple set of targeted exerciseswas effective in producing a large reduction in the risk ofAKP in a young military population undergoing an inten-sive 14-week physical conditioning and training program.We also showed a significant and important improvementin occupational outcomes such as a reduction in dischargesfrom the service in the intervention group. There were noadverse effects among individuals undergoing the inter-vention protocol. As far as we are aware, this is the firstrandomized controlled trial demonstrating the use of a pre-vention protocol for overuse anterior knee pain.

Vol. 39, No. 5, 2011 Effects of Exercise for Preventing Overuse Anterior Knee Pain 943

at University of Bath - The Library on May 10, 2016ajs.sagepub.comDownloaded from

Strengths and Limitations

This study has several strengths. With the exception of thecontrol/intervention sessions, the daily exercise load andexposure to other extrinsic risk factors for AKP of each

individual were similar and monitored and supervised.Because of the military training setting, no individuals werelost to follow-up. The intervention was specifically targetedat anterior knee pain, and the sessions were supervised toensure correct execution and compliance. There was no

TABLE 2Baseline Descriptive Characteristics of Intervention and Control Group Participants byGender (n = 1502): Pretrial Distribution of Knee, Ankle, Acute, and Overuse Injuries

Demographics and Physical Characteristics

Intervention Group (n = 759) Control Group (n = 743)

Variable Men (n = 556) Women (n = 203) Men (n = 536) Women (n = 207)

Age, y, mean 6 SD 19.4 6 2.1 20.2 6 2.9 19.5 6 2.4 19.8 6 2.8Height, m, mean 6 SD 1.76 6 0.06 1.64 6 0.06 1.76 6 0.06 1.65 6 0.06Weight, kg, mean 6 SD 70.7 6 9.6 61.1 6 7.7 69.9 6 9.4 60.9 6 7.8Body mass index, kg/m2, mean 6 SD 22.7 6 2.6 22.6 6 2.3 22.6 6 2.5 22.5 6 2.2Baseline fitness, mean 6 SDa 637.4 6 64.4 754.7 6 64.2 634.2 6 61.1 758.9 6 58.5Smoking habits, No. (%)b

Nonsmokers 282 (50.7) 129 (63.5) 297 (55.4) 121 (58.5)Current smokers 274 (49.3) 74 (36.4) 239 (44.6) 86 (41.5)

1-7 73 (13.1) 22 (10.8) 66 (12.3) 26 (12.5)8-14 121 (21.8) 40 (19.7) 107 (19.9) 44 (21.3).15 80 (14.4) 12 (5.9) 66 (12.3) 16 (7.7)

Running, No. (%)c

None 52 (9.4)g 8 (3.9) 51 (9.5) 7 (3.9)1-5 259 (46.7) 102 (50.2) 236 (44.0) 103 (49.8)6-10 157 (28.3) 68 (33.5) 177 (33.0) 70 (33.9)11-15 43 (7.7) 14 (6.9) 38 (7.1) 17 (8.2).16 44 (7.9) 11 (5.4) 34 (6.3) 10 (4.8)

Strength training, No. (%)d

No 119 (21.4)g 33 (16.7) 128 (23.9) 39 (18.9)g

1 56 (10.1) 20 (9.8) 42 (7.8) 18 (8.7)2-3 260 (46.9) 109 (53.7) 255 (47.6) 102 (49.5).4 120 (21.2) 41 (20.2) 111 (20.7) 47 (22.8)

Injury Type Intervention Group (Pooled Sex) Control Group (Pooled Sex)

Body part, No. (%)Knee 78 (10.3) 74 (9.9)Ankle 108 (14.3) 68 (9.2)

Acute injuries, No. (%)e

Knee ligament injury 10 (1.3) 14 (1.8)Ankle ligament injury 107 (14.1) 68 (9.2)Bone fractures 14 (1.2) 4 (0.5)Meniscus injury 13 (1.7) 7 (1.0)Other 29 (3.9) 28 (3.8)

Overuse injuries, No. (%)f

Anterior knee pain 14 (1.8) 15 (2.0)Patellar tendinitis 1 (0.1) 0 (0.0)Osgood-Schlatter syndrome 12 (1.6) 19 (2.6)

Previous surgery, No. (%)Knee 9 (1.2) 5 (0.7)Ankle 14 (1.8) 6 (0.8)

aBaseline fitness is 2.4 km measured in seconds.bSmoking habits is number of cigarettes per day (%).cRunning is number of miles per week (%).dStrength training is number of sessions per week (%).eValues are pooled (by gender) numbers (%) of participants. Acute injury defined as occurring with a sudden onset associated with a known

trauma.fOveruse injury defined as occurring with a gradual onset without any known trauma.gData missing for one participant.

944 Coppack et al The American Journal of Sports Medicine

at University of Bath - The Library on May 10, 2016ajs.sagepub.comDownloaded from

suggestion that the groups differed at baseline in terms ofrisk factors for AKP or factors that may be related to occupa-tion outcomes (eg, BMI, fitness, smoking, previous injuries).The method of case capture was first hand rather than rely-ing on retrospective information such as medical notes.

In our study, the diagnosis of AKP was generic, with noidentification of specific causes or analysis of predisposingfactors for AKP. Consequently, any views on the effective-ness of our program in combating certain predisposing fac-tors or preventing specific diagnoses responsible for AKPwould be speculative. Unfortunately, because of militaryoperational commitments, data collection was stopped early,and the estimate of the effect lacks precision. However, thestudy was still adequately powered because the effect wesaw was larger than hypothesized. It is possible that therewas some censorship bias because of more participantsfrom the control group being medically discharged; if indi-viduals who were medically discharged were also more likelyto develop AKP, then the effect would have been underesti-mated. Blinding of outcome assessors was not completelysuccessful. In 33% (6 in the intervention group and 9 inthe control group) of all AKP cases, the trained observerresponded that he or she was aware of participant allocation.In procedural intervention trials this can account for anoverestimation of effect by up to 17%.30 Last, there was nofollow-up beyond 14 weeks, and so no conclusions can bedrawn about the long-term benefits of the intervention.

Strengths and Weaknesses in Relation to OtherStudies and Comparison of Findings

The intervention program comprised well-establishedexercises used in the management of AKP.5,8 This regimen

Allocated to PTP intervention group.21 troops (n = 759)Received allocated intervention (n = 759)Did not receive allocated intervention (n = 0)

Allocated to CP control group. 23 troops (n = 743)Received allocated intervention (n = 743)Did not receive allocated intervention (n = 0)

Analyzed (n = 759 )

Lost to follow-up (n = 0 ) Lost to follow-up (n = 0 )

Analyzed (n = 743 )

Assessed for eligibility (n = 1502)

Randomized (44 troops; 1502 participants)

Figure 1. Trial profile: flow of troops and participants through the study. CP, control program; PTP, prevention training program.

Control (CP) group

Intervention (PTP) group

0

1

2

3

4

5

6

7

8

0 2 4 6 8 10 12 14

Log rank test: χ2 = 17·45 , p<0·001

Cu

mu

lati

ve %

wit

h A

KP

Training time (weeks)

Figure 2. Failure plot for anterior knee pain (AKP) in the controland intervention groups over the period of training. The top lineis the cumulative % with AKP in the control group, and the bot-tom line is the cumulative % with AKP in the intervention group.

TABLE 3Effect of Intervention (Hazard Ratio)a

Hazard Ratio 95% Confidence Interval P Value

Unadjusted 0.25 0.13-0.48 \.001Adjustedb 0.25 0.13-0.49 \.001

aEstimates are from a Cox proportional hazards regressionmodel with robust standard errors to account for the clusterdesign.

bAdjusted for gender, age, previous ankle injury, previous epi-sode of anterior knee pain, physical fitness, smoking (cigarettesper day), body mass index, and height.

Vol. 39, No. 5, 2011 Effects of Exercise for Preventing Overuse Anterior Knee Pain 945

at University of Bath - The Library on May 10, 2016ajs.sagepub.comDownloaded from

has produced beneficial effects in randomized controlledtrials.8,9,18 In contrast to many previous injury preventionstudies, which include several lower limb injuries,4,10,13 wechose to focus on the prevention of AKP alone.

Several other exercise intervention studies have demon-strated significant reductions in injury rates in theirtrained groups. In these studies, the investigators exam-ined the effect of balance training,13 multifaceted struc-tured warm-ups,26 or plyometric jump training.19 Thehazard ratio of injury found in our study (0.25; 95% CI,0.12-0.51) is of a similar magnitude compared with anotherrandomized controlled trial examining preventative warm-up exercises in adolescents (relative risk, 0.35; 95% CI,0.19-0.63).26 A systematic review of multi-interventiontraining programs incorporating strength and balanceexercises also reported injury risk reductions greaterthan 50% in 5 of 6 of the studies reviewed.1

In contrast to our findings, 2 recent randomized con-trolled trials found no preventative effect of an AKP inter-vention in military recruits4 and elite volleyball players.10

In both studies, compliance rates were poor (�75%), andexercises were performed incorrectly10 or without propersupervision.4 Furthermore, both studies employed a broadapproach that included other knee or lower limb injuries.Another randomized trial reporting low compliance ratesin male soccer players found no preventative effects ofexercise.14 In our study, participants performed all exer-cise sessions under qualified supervision. This ensuredconsistency in the quality of exercise technique and highrates of compliance (�90%). On the basis of these findingsand previous studies,13,26 we suggest that supervision isa potentially key factor in AKP prevention protocols.

Another difference between our study and the onlyother randomized AKP prevention trial with militaryrecruits4 is the type and dosage of exercise. Brushøjet al4 employed 3 sets of 5 exercises, 3 times per week (totalweekly duration = 45 minutes), for 12 weeks. We employed3 sets of 8 exercises, 7 times per week (total weekly dura-tion = 105 minutes), for 14 weeks. A minimum amount ofexercise must be performed before an effect may be

expected.14,22 Depending on the goal of training, the exer-cise dosage in both studies conforms with published recom-mendations.22 However, this number is arbitrary as thereis no evidence on the dose-effect relationship for any exer-cise program to prevent injury.14

Interpretation, Explanation, and Implication

Training programs that address intrinsic risk factors area well-established method of reducing AKP inci-dence.10,32,37 The quadriceps play a key role in the causesof AKP.5,9 Several studies indicate that AKP patients dem-onstrate an imbalance in the neuromotor control betweenthe vastus medialis oblique (VMO) and vastus lateralis(VL).34,36 Reduced motor activity of the VMO relative tothe VL could cause lateral patellar tracking and subluxa-tion.5,8,37 The prevention program that we tested incorpo-rated CKC functional strengthening exercises. Theseexercises have been shown to increase VMO and VL activa-tion patterns6 and improve patellofemoral joint alignmentin healthy adults.20

Previous studies using VMO CKC exercises in AKPpatients have reported positive effects.8,12,25 In our study,a high percentage of AKP cases was sustained in the first5 weeks of training (Figure 2). Previous research in UKmilitary recruits showed a similar temporal pattern.33 Sur-face electromyographic (EMG) techniques reveal thatstrength gains in the early phase of training are associatedwith an increase in the amplitude of EMG activity.16,25

Eccentric contractions appear to involve a different EMGactivation scheme to isometric and concentric contrac-tions,16 and one older study showed relief of AKP afteronly 2 to 4 weeks of eccentric training.2

Because our intervention group focused on quadriceps,pelvis, and hip muscle recruitment, we would expect betterhip and knee motor control during weightbearing activi-ties.25 It is plausible that improved motor control motionhad an important role in the prophylactic effect seen inthe intervention group. Specifically, movement patternswere executed more efficiently, thereby reducing the stresson the patellofemoral joint. The emphasis on controlledeccentric exercise may have increased the muscular capac-ity to absorb high patellofemoral forces during the earlystages of training. However, this view is speculative. Noeffects of this program have been reported in injury pre-vention studies, and a recent prospective randomized trialshowed no alteration in reflex response times of VMO andVL in AKP sufferers using CKC exercises.34 Thus, strongevidence in support of a neuromuscular adaptation in ourprogram is lacking, and further research investigatingthe potential protective effects of exercise on disturbedneuromuscular balance is needed.

Stretching of soft tissue structures has been shown toreduce pain and increase flexibility in AKP patients.5,8,12

This would imply that a stretching program may be bene-ficial in the prevention of AKP. However, scientific evi-dence supporting the protective effects of stretching islacking. Two randomized studies using military recruitsfailed to show any positive effect of stretching on

TABLE 4Training (Occupational) Outcome by Groupa

Group Code

Training OutcomeControl,No. (%)

Intervention,No. (%)

Total,No. (%)

DAOR 71 (9.6) 76 (10.0) 147 (9.8)MD 25 (3.4) 3 (0.4) 28 (1.9)UFAS 23 (3.1) 6 (0.8) 29 (1.9)Successful completion

of training504 (67.8) 605 (79.7) 1109 (73.8)

Otherb 120 (16.2) 69 (9.1) 189 (12.6)Total 743 (100) 759 (100) 1502 (100)

aAbbreviations: DAOR, discharge as of right; MD, medical dis-charge; UFAS, unfit for army service. Pearson x2(4) = 50.2168,Pr = 0.001.

bBacksquadded, rehabilitation, sickness.

946 Coppack et al The American Journal of Sports Medicine

at University of Bath - The Library on May 10, 2016ajs.sagepub.comDownloaded from

individual injury risk,27,28 and it is not possible to deter-mine which elements of our intervention program may beeffective in preventing AKP. Although systematic reviewsreport mixed evidence that stretching exercises for injuryprevention are effective,1 the findings of this trial suggestthat abandoning stretching interventions for AKP preven-tion would be premature.

Our results showed that a reduction in AKP wasreflected in significantly lower MD rates in the interven-tion group (Table 4). Beneficial results were also observedfor training days lost and completion of training. A previ-ous study with UK military recruits showed that AKPcases accounted for 25% of all medical discharges andtook longer to train.33 One non-UK study found thata 46.6% reduction in all injuries resulted in a 40.0% reduc-tion in medical discharge rates in male recruits who fol-lowed a modified exercise program.29 A further studyshowed a 14.2% reduction in attrition in recruits undergo-ing a prevention conditioning protocol.23 These studieswere nonrandomized and suffer from design limitations.However, combined with our data, they do demonstratethat reduced injury incidence can significantly improveoccupation outcomes.

Generalizability of Results

Although the age group (17-30) in our study is representa-tive of the typical case of AKP in the general population,we do not know if our results can be generalized to otherage groups or to groups with different activity levels. Con-sequently, our findings are specific to a young military pop-ulation and cannot be considered a panacea for all AKPsufferers. If the goal of training is to develop eccentricmotor control patterns that are resistant to injury, it maybe beneficial to introduce our program with younger ado-lescents who have not yet established their motor pat-terns.26 Introduction of preventative exercise programsmay also lead to fewer injuries and training hours lost inathletic groups and lower medical costs among the generalpopulation. However, achieving high compliance ratesamong civilian populations who are not subject to manda-tory training programs could be a problem. Furtherresearch is needed among populations other than militaryrecruits focusing on specific types of injuries.1

CONCLUSION

A targeted exercise protocol significantly reduced the inci-dence of AKP and improved secondary occupation out-comes in military recruits undergoing a 14-week trainingprogram. We recommend that preventative trainingshould be introduced as a core component of phase 1army training. It is possible that close supervision of cor-rect exercise technique by professionally qualified staffmay be a crucial factor in AKP prevention. Future researchshould focus on the long-term benefits of preventativeexercise programs and the relative effectiveness of this reg-imen among nonmilitary and recreational populations.

ACKNOWLEDGMENT

We acknowledge and offer grateful thanks to the medicalcenter, physiotherapy department, and physical trainingwing staff at the ATC Pirbright. Particular thanks areextended to the sword company clerk, Mrs Lorraine Con-nelly, for her invaluable help and cooperation duringdata collection. We also thank Gethin Owen and RichardHorsley (DMRC), as well as James Bilzon and Rachel Izard(ARTD Occ Med), for their support in instigating thisstudy. We also thank Luisa Zuccolo (University of Bristol)for her useful comments in reviewing this manuscript.This work was sponsored by the Army Recruitment andTraining Division (ARTD), UK.

REFERENCES

1. Aaltonen S, Karjalainen H, Heinonen A, Parkkari J, Kujala UM. Pre-

vention of sports injuries: systematic review of randomized controlled

trials. Arch Int Med. 2007;167(15):1585-1592.

2. Bennett JG, Stauber WT. Evaluation and treatment of anterior knee

pain using eccentric exercise. Med Sci Sports Exerc.

1986;18(5):526-530.

3. Bizzini M, Childs JD, Piva SR, Delitto A. Systematic review of the

quality of randomised controlled trials for patellofemoral pain syn-

drome. J Orthop Sports Phys Ther. 2003;33:4-20.

4. Brushøj C, Larsen K, Albrecht-Beste E, Nielsen MB, Løye F, Holmich

P. Prevention of overuse injuries by a concurrent exercise program in

subjects exposed to an increase in training load: a randomised con-

trolled trial of 1020 army recruits. Am J Sports Med. 2008;36(4):

663-670.

5. Clark DI, Downing N, Mitchell J, Coulson L, Syzpryt EP, Doherty M.

Physiotherapy for anterior knee pain: a randomised controlled trial.

Ann Rheum Dis. 2000;59:700-704.

6. Coqueiro K, Bevilaqua-Grossa D, Berzin F, Soares AD, Candolo C,

Monteiro-Pedro V. Analyses on the activation of the VMO and VLL

muscles during semi-squat exercises with and without hip adduction

in individuals with patellofemoral pain syndrome. J Electromyogr

Kinesiol. 2005;15(6):596-603.

7. Cowan DN, Jones BH, Frykman PN, et al. Lower limb morphology

and risk of overuse injury among male infantry trainees. Med Sci

Sports Exerc. 1996;28(8):945-952.

8. Crossley K, Bennell K, Green S, Cowan S, McConnell J. Physical

therapy for patellofemoral pain: a randomized, double-blinded,

placebo-controlled trial. Am J Sports Med. 2002;30(6):857-865.

9. Crossley K, Bennell K, Green S, McConnell J. A systematic review of

physical interventions for patellofemoral pain syndrome. Clin J Sport

Med. 2001;11(2):103-110.

10. Cumps E, Verhagen EA, Duerinck, Deville A, Duchene L, Meeusen R.

Effect of a preventative programme on the prevalence of anterior

knee pain in volleyball players. Eur J Sports Sci. 2008;8(4):183-192.

11. Dorotka R, Jiminez-Boj E, Kypta A, Koller B. The patellofemoral pain

syndrome in recruits undergoing military training: a prospective 2-

year follow-up study. Mil Med. 2003;168(4):337-340.

12. Douciette SA, Goble EM. The effect of exercise on patellar tracking in

lateral patellar compression syndrome. Am J Sports Med.

1992;20(4):434-440.

13. Emery CA, Cassidy D, Klassen TP, Rosychuk RJ, Rowe BH. Effec-

tiveness of a home-based balance-training program in reducing

sports-related injuries among healthy adolescents: a cluster random-

ized controlled trial. CMAJ. 2005;172(6):749-754.

14. Engebretsen AH, Mykleburst G, Holme I, Engebretsen L, Bahr R. Pre-

vention of injuries among male soccer players: a prospective, ran-

domized intervention study targeting players with previous injuries

or reduced function. Am J Sports Med. 2008;36(6):1052-1060.

Vol. 39, No. 5, 2011 Effects of Exercise for Preventing Overuse Anterior Knee Pain 947

at University of Bath - The Library on May 10, 2016ajs.sagepub.comDownloaded from

15. Fagan V, Delahunt E. Patellofemoral pain syndrome: a review on the

associated neuromuscular deficits and current treatment options. Br

J Sports Med. 2008;42:789-795.

16. Gabriel DA, Kamen G, Frost G. Neural adaptations to resistive exer-

cise: mechanisms and recommendations for training practices.

Sports Med. 2006;36:133-149.

17. Gemmell IM. Injuries among female army recruits: a conflict of legis-

lation. J R Soc Med. 2002;95(1):23-27.

18. Heintjes E, Berger MY, Bierma-Zeinstra SM, Bernsen RM, Verhaar

JA, Koes BW. Exercise therapy for patellofemoral pain syndrome.

Cochrane Database Syst Rev. 2003;(4):CD003472.

19. Hewett TE, Lindenfeld TN, Riccobene JV, Noyes FR. The effect of

neuromuscular training on the incidence of knee injury in female ath-

letes: a prospective study. Am J Sports Med. 1999;27(2):699-706.

20. Ingersoll CD, Knight KL. Patellar location changes following EMG

biofeedback or progressive resistive exercises. Med Sci Sport Exerc.

1991;23:1122-1127.

21. Jones BH, Cowan DN, Tomlinson JP, Robinson JR, Polly DW, Fryk-

man PN. Epidemiology of injuries associated with physical training

among young men in the army. Med Sci Sports Exerc.

1993;25(2):197-203.

22. Kraemer WJ, Adams K, Cafarelli E, et al. American College of Sports

Medicine position stand: progression models in resistance training

for healthy adults. Med Sci Sports Exerc. 2002;34(2):364-380.

23. Lee L, Kumar S, Kok WL, Lim CL. Effects of a pre-training condition-

ing programme on basic military training attrition rates. Ann Acad

Med Singapore. 1997;26:3-7.

24. Milgrom C, Finestone A, Eldad A, Shlamkovitch N. Patellofemoral

pain caused by overactivity: a prospective study of risk factors in

infantry recruits. J Bone Joint Surg Am. 1991;73(7):1041-1043.

25. Nakagawa TH, Muniz TB, Baldon RM, Maciel CD, Reiff RB, Serrao

FV. The effect of additional strengthening of hip abductor and lateral

rotator muscles in patellofemoral pain syndrome: Clin Rehab.

2008;22(5):1051-1060.

26. Olsen OE, Mykleburst G, Engebretsen L, Holme I, Bahr R. Exercises

to prevent lower limb injuries in youth sports: cluster randomised

controlled trial. BMJ. 2005;330(7489):449.

27. Pope R, Herbert R, Kirwan J. Effects of ankle dorsiflexion and pre-

exercise calf muscle stretching on injury risk in army recruits. Aust

J Physiother. 1998;44(3):165-172.

28. Pope R, Herbert RD, Kirwan JD, Graham BJ. A randomized trial of

pre-exercise stretching for prevention of lower-limb injury. Med Sci

Sports Exerc. 2000;32(2):271-277.

29. Rudzki SJ, Cunningham MJ. The effect of a modified physical train-

ing program in reducing injury and medical discharge rates in Austra-

lian army recruits. Mil Med. 1998;164:648-653.

30. Schultz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of

bias: dimensions of methodological quality associated with estimates

of treatment effects in controlled trials. JAMA. 1995;273(5):

408-412.

31. Tumia N, Maffulli N. Patellofemoral pain in female athletes. Sports

Med Arthrosc Rev. 2002;10(1):69-75.

32. Van Tiggelen D, Witvrouw E, Roget P, Cambier D, Danneels L, Ver-

donk R. Effect of bracing on the prevention of anterior knee pain:

a prospective randomized study. Knee Surg Sports Traumatol

Arthrosc. 2004;12:434-439.

33. Wills AK, Ramasamy A, Ewins DJ, Etherington J. The incidence and

occupational outcome of overuse anterior knee pain during army

recruit training. J R Army Med Corps. 2004;150:264-269.

34. Witvrouw E, Cambier D, Danneels L, et al. The effect of exercise reg-

imens on reflex response time of vasti muscles in patients with ante-

rior knee pain: a prospective randomized intervention study. Scand J

Med Sci Sports. 2003;13:251-258.

35. Witvrouw E, Danneels L, Van Tiggelen D, Willems TM, Cambier D.

Open versus closed kinetic chain exercises in patellofemoral pain:

a 5-year prospective randomized study. Am J Sports Med.

2004;32(5):1122-1130.

36. Witvrouw E, Lysens R, Bellemans J, et al. Which factors predict out-

come in the treatment program of anterior knee pain? Scand J Med

Sci Sports. 2002;12:40-46.

37. Witvrouw E, Lysens R, Bellemans J, Cambier D, Vanderstraeten G.

Intrinsic risk factors for development of anterior knee pain in an ath-

letic population: a two-year prospective study. Am J Sports Med.

2000;28(4):480-498.

For reprints and permission queries, please visit SAGE’s Web site at http://www.sagepub.com/journalsPermissions.nav

948 Coppack et al The American Journal of Sports Medicine

at University of Bath - The Library on May 10, 2016ajs.sagepub.comDownloaded from