Journal of Strength and Conditioning Research, 2006, 20(2), 345-353© 2006 National Strength & Conditioning Association

THE EFFECTS OF PLYOMETRIC VS. DYNAMICSTABILIZATION AND BALANCE TRAINING ON POWER,BALANCE, AND LANDING FORCE IN FEMALEATHLETES

GREGORY D. MYER/ KEVIN R. FORD,' JENSEN L. BRENT,' AND TIMOTHY E . HEWETT^ ^

'Cincinnati Children's Hospital Research Foundation Sports Medicine Biodynamics Center and HumanPerformance Laboratory, Cincinnati, Ohio 45229; and '^The University of Cincinnati College of Medicine,Departments of Pediatrics, Orthopaedic Surgery and the College of Allied Health Sciences, Department ofRehabilitation Sciences and the Department of Bioengineering, Cincinnati, Ohio 45229.

ABSTRACT. Myer, G.D., K.R, Ford, J.L. Brent, and T.E. Hewett.The effects of plyometric vs. dynamic stabilization and balancetraining on power, balance, and landing force in female athletes,J. Strength Cond. We.s. 20(2): 345-;i53. 2006.—Neuroniasculartraining protocols that include both plyometrics and dynamicbalance exercises can significantly improve biomechanics andneuromuscular performance and reduce anterior cruciate liga-ment injury risk in female athletes. The purpose of this studywas to compare the effects of plyomelrics (PLYO) versus dynam-ic stabilization and balance training (BAL) on power, balance,strengtb. and landing force in female athletes. Either PI-,YO orBAL were included as a component of a dynamic neuromusculartraining regimen that reduced measures related to ACL injuryand increased measures of performance. Nineteen bigh schoolfemale athletes participated in training 3 times a week for 7weeks. The PLYO in = 81 group did not receive any dynamicbalance exercises and the BAL in - 11) group did not receiveany maximum efTort jumps during training. Pretraining vs. post-training measures of impact force and standard deviation of cen-ter of pressure (COP) were recorded during a single leg hop andbold. Subjects were also tested for training effects in strength(isokinetic and isoinertial) and power (vertical jump). Tbe per-cent cbange from pretest to posttest in vertical ground reactionforce was significantly different between tbe BAL and PLYOgroups on tbe dominant side ip < 0.05). Botb groups decreasedtbeir standard deviation of center of pressure (COP) during hoplandings in tbe medial/lateral direction on their dominant side,whicb equalized pretested side to side differences. Both groupsincreased hamstrings strengtb and vertical jump. The results ofthis study suggest tbat both PLYO and BAL training are effec-tive at increasingmeasuresof neuromuscular power and control.A combination of PLYO and BAL training may further maximizetbe effectiveness of preseason training for female athletes.

KEY WORDS, core stability, ACL injury prevention, performancetraining, neuromuscular training, biomechanics, jump training

INTRODUCTION

—^_ nitial interventions aimed at reducing the in-J I cidence of anterior cruciate ligament (ACL) in-T I juries in female athletes were developed based

^ on empirical evidence from coaching and train-ing female athletes and from performance en-

hancement research (11, 15). More current techniques,developed from injury mechanism and objective analysisof training methods, may further reduce traumatic ACLinjuries in females (22, 27). The studies that demonstrate

ACL injury reduction are comprehensive, and includemultiple training components that may induce the neu-romuscular changes required to reduce ACL injury risk.

There is evidence that neuromuscular training notonly decreases ACL injury risk, hut also that it alters bio-mechanical risk factors for ACL injury and improves mea-sures of performance in female athletes (14, 15, 22, 26,27, 29). Effective neuromuscular training protocols haveused plyometric power, hiomechanics and technique,strength, balance, and core stability training to induceneuromuscular changes and potential injury preventioneffects in females (14, 15, 22, 26, 27, 29). However, it isnot known which of these components are most effectiveand efficient or wbether the effects of the separate com-ponents are cumulative. Selective combination of neuro-muscular training components may provide additive ef-fects or a single component may provide tbe majority ofthe biomechanical alterations related to reduction of ACLinjury incidence and improved neuromuscular perfor-mance in female atbletes 114, 15, 22, 26, 27, 29). Addi-tionally, the time investment or difficulty of teachingthese comprehensive protocols may preclude coaches orathletes from undertaking training or may reduce com-pliance witb training (22, 27, 29). To increase the utili-zation of ACL injury prevention training, more effectiveand efficient methods for reducing ACL injury risk andimproving performance in female atbletes sbould be de-veloped. When highly effective training methods are de-veloped, dynamic neuromuscular training can be usedwith high-risk athletes on a widespread basis witb great-er compliance.

Tbe purpose of this study was to compare the effectsof maximum effort plyometric jumping (PLYO) vs. dy-namic stabilization and halance training (BALI on power,balance and landing force in female atbletes. EitherPLYO or BAL were included as a component of a dynamicneuromuscular training regimen that reduced measuresrelated to ACL injury and increased measures of perfor-mance (13, 25, 28). The first hypothesis was that thePLYO study group would demonstrate improvements inpower (e.g., vertical jump) while the BAL group wouldnot. Second, the BAL group would improve postural con-trol measures (e.g., center of pressure [COP| sway), whiletbe PLYO group would not. Tbe tbird hypothesis was tbat

345

346 MYER, FORD, BRENT ET AL.

the BAL group would improve force dissipation measuresand the PLYO group would not. The fourth hypothesiswas that the PLYO groups would increase isokineticstrength. The final hypothesis was that hoth PLYO andBAL would improve isoinertial strength measures.

METHODS

Experimental Approach to the Prohlem

To compare the effects of PLYO versus BAL, a controlledlahoratory cohort experiment was conducted. Female ath-letes were randomized prior to pretesting to one of twotraining groups (PLYO or BAL). The PLYO group did notreceive any dynamic balance exercises and the BAL groupdid not receive any maximum effort jumps during train-ing. To evaluate the effects of each protocol the testingincluded dynamic landing force and COP control, verticaljump, and strength measures.

Suhjects

Nineteen female athletes from a Cincinnati area highschool participated in this study. The age of the partici-pants (mean + SD) was L5.9 ± 0.8 years in the PLYOgroup and 15.6 ± L2 in the BAL group with a range of14 to 17 years. All the subjects listed their primary sportas volleyball, while 12 subjects listed secondary sports in-cluding 5 in basketball, 2 soccer, 2 softball, and 1 swim-ming. Seventy-nine percent of the subjects had at least 6years of experience in their reported primary sport, whilenone had less than 4 years of experience. Eighty-four per-cent of the subjects reported previous participation insome form of off-season training program. This included79'^ in jogging programs, 74*}̂ in weight lifting programs,42'7f in sprinting programs, and lOVr in a jump trainingprogram. Height and mass were assessed at the pretrain-ing test date and the posttraining test date. The initialheight (mean ± SD) of the participants in the PLYOgroup was 169.5± 6.1 cm and mass was 61.4 ± 7.3 kg. Inthe BAL group the initial height and mass (mean ± SD)were 168.1 ± 6.9 cm and 66.4 ± 11.8 kg. Follow-up as-sessment of height and mass at the posttest date revealedno significant change in mean height while mean weightincreased in the PLYO group to 61.8 ± 7.5 kg and in theBAL group to 68.3 ± 11.9 kg (p = 0.01).

Parents or guardians signed informed consent prior toparticipation in the study. Initially, 23 subjects were ran-domized between the 2 groups. Four athletes did not meetthe preestablished compliance criterion that requiredeach subject participate in at least two-thirds (12 of 18)of the training sessions (subject population n = 19). Onesubject made only 11 sessions due to competitive club sea-son travel, thus excluding her from the study analysis.Three subjects did not complete necessary training ses-sions or participate in posttesting, due to injuries duringoutside sport activities (1 acute fibula fracture from aneversion ankle sprain while wearing a rigid ankle ortho-sis, 1 grade 2 and 1 grade 3 inversion ankle sprain). Noacute injuries occurred during the training sessions. The4 suhjects excluded from the original 23 due to noncom-pliance were from the PLYO group. The average numberof completed sessions for both groups was approximately16 (PLYO 16.4 ± 1.5, BAL 15.5 ± 1.4).

All subjects were pretested 1 week prior to tbe initialtraining session. Posttesting was performed approximate-ly 8 weeks after the pretest on all subjects (4 days after

the final training session). The training sessions wereconducted with a minimum 1:3 trainer to athlete ratio,by National Strength and Conditioning Association cer-tified professionals and graduate and undergraduate lev-el student interns to ensure proper institution of protocolguidelines. The subjects were randomized into 2 groups,which allowed for 2 training sessions per day. One groupperformed a protocol that involved maximum effort ply-ometric training (PLYO), while the second group per-formed a protocol that focused on dynamic stabilizationand balance training (BAL). This study was approved bythe Cincinnati Children's Hospital Medical Center Insti-tutional Review Board.

Procedures

Training. The neuromuscular training protocol used inthis study was derived from a training protocol, previ-ously shown to reduce biomechanical measures related toACL injury and increase measures of performance (13,25, 28). The previous protocol was modified to includeonly maximum effort plyometric jumps and cut maneu-vers (PLYO) or to include dynamic lower extremity sta-bilization and balance exercises (BAL). The neuromus-cular training was conducted on Monday, Tuesday, andThursday of each consecutive week. Each training sessionlasted for approximately 90 minutes. Prior to each train-ing session an active warm-up that included 5 agility lad-der runs was performed. Each of the groups participatedin 2 types of training per day; either resistance training,speed interval training, or PLYO or BAL training de-pending on the group. Tahles 1 and 2 depict the protocolsfor the specific PLYO and BAL sessions respectively tak-en from the first session of the week in weeks 1, 3, and6. Tables 3 and 4 depict the corresponding strength andspeed interval training protocols used by both groups.

The focus of tbe PLYO group included an emphasis onperforming jumping movements with maximum effortand power, and also performing cutting techniques withquick reactions and maximal effort. The athletes receivedfrequent feedback regarding the technical performance oftheir jumping and cutting movements. Specifically, theatbletes were instructed to perform maximum effortjumps without knee valgus, with a focus on improvingefficiency and power of the jump. When performing theunanticipated cutting maneuvers, the athletes were againinstructed to decrease knee valgus motion, maintainproper knee and foot alignment and position, while at-tempting to improve the speed and efficiency of the tech-nique. The exercise intensity progressed by adding com-plexity to the movements, and through the addition ofsingle limb maneuvers. The exercises performed by tbePLYO group did not include any form of stabilization,hold or balance.

In contrast, the BAL group followed a protocol thatemphasized dynamic stabilization and balance. The ath-letes were instructed on methods to dampen landing forcethrough sagittal plane fiexion while avoiding positions ofknee valgus. As this protocol progressed, it also increasedin difficulty by moving from stable ground surfaces, torelatively unstable surfaces such as Airex pads (PerformBetter Inc., Cranston, RI), BOSU trainers (TEAM BOSU,Canton, OH), and Swiss Balls (Perform Better Inc.). Dy-namic balance exercises progressed by altering tbe ath-letes' center of gravity through perturbations or addingexternal weight to the movement or by the addition of

PLYOMF.TRIC VS. DYNAMIC BALANCK TRAINING 347

TABLE 1. Plyometric training at weeks 1, 3, and 6.

Plyometric training Reps

15101010

1015

20

151515151515

15

Week 6

Squat-tuck jumpsBarrier hops flat (front to back)Barrier hops flat (side to side)Crossover hop, hop, hop-athletic

position + stept180'Jump.s (height)Broad jump, jump, jump vertical +

stept3 Barrier hop-reaction (3-way)Forward-backward hops over

barriers + step*'"Box drop-180°-box drop-max

vertical + steptBox drop-ISO" + steptLateral box drop-broad jump +

121212

15

Box drop max vertical-broad jump+ stept

Approach max verticalCrossover step-ski stop max

vertical

61010

Week 1

Athletic positionWall jumpsSquat jumpsTuck jump (with thighs parallel)Line jumps (side to side)Line jump lateral max verticalLunge jump180 Jumps (height)Broad jump vertical + steptBounding in placeForward jumps over barriers + steptForward barrier jumps with middle

box + steptBox drop max vertical + steptBox drop + stept

Week 3

Wall jumpsSquat jumpsTuck jump (with abdominal crunch)Tuck jump (with butt kick)Barrier jumps (front to back)Barrier jumps (side to side)Hop, bop, hop-athletic position + stept180" Jumps (height)Broad jump, jump, jump vertical + steptBounding for distanceLateral barrier hops with staggered

box reactionBack drop-box toucb-max vertical + steptLateral box drop max verticalPower steps

66

61058

10

63

6

8

'''' Time in seconds.t Exercise ends with a quick reaction step.

single limb maneuvers. In addition, this group performedexercises which focused on maintaining dynamic stabili-zation, with movements designed to strengthen the coremusculature of the body. These exercises included torsoflexion, extension, and rotational strength maneuvers.

Testing. A single-leg hop and balance test was per-formed a total of 6 times (3 randomized trials on eachside) using an AccuPower portahle force platform (Ad-

vanced Mechanical Technology, Inc., Watertown, MA).Subjects initiated the movement while balancing on 1foot, and were instructed to hop forward 50 cm and bal-ance for 10 seconds after the landing on the same foot.Standard deviation of center of pressure (COP) in the me-dial-lateral (COPMI.) and anterior-posterior (COP^p) direc-tions, and maximum vertical ground reaction force werecalculated for each trial with Matlab software version 7.0(Mathworks, Natick, MA) Previously, measurements onthe portable platform have been shown to provide reliablemeasures (3).

Knee extensor and knee fiexor strength was assessedisokinetically using a dynamometer (Biodex Medical Sys-tems, NY). Isokinetic testing bas previously been shownto be a reliable measure of quadriceps (r ^ 0.968) andhamstrings ir = 0.848) strengtb in children and adoles-cent populations (12). The patient was secured in a seatedposition on the dynamometer with her trunk perpendic-ular to the floor, hip flexed to 90°, and knee also flexedto 90". Stabilization straps were secured at the waist, dis-tal femur, and distal sbank, just proximal to the medialmalleolus. The test session consisted of 10 knee fiexion-extension repetitions for each leg at 300°- second '. Peakflexion and extension torques were recorded. A warm-upset consisting of 5 knee flexion-extension repetitions foreach leg at 300°- second ' was performed prior to datacollection.

The vertical jump was measured on an MXl verticaljump trainer (MXP Sports, Reading, PA). Prior to thetest, each subject's overhead reach was determined withthe subject reaching directly overhead with both handsup toward the ball, the midline of the basketball wasaligned with the distal interphalangeal joint of the rightand left middle fingers. The subject was told to use a nat-ural over-head reach (no exaggerated superior rotation ofthe shoulder girdle). The digital readout of the systemwas zeroed to subtract reach from jump height and pro-vide actual vertical displacement during the vertical jumptesting (8, 26). Prior authors have demonstrated thatcounter movement vertical jump testing has a test-retestreliability of 0.99 (31).

Prior to isoinertial strength testing the athlete wasinstructed on the proper form for squat, hang clean, andbench press exercises. Tbe athlete was instructed to per-form practice repetitions with the standard barbell. Afterthe exercise orientation, the trainer chose a weight theyestimated the athlete could lift 5 or fewer times. The testwas accepted provided the repetitions completed were 8or less. If the athlete performed more than 8 repetitions,more weight was added and they were retested. The 1repetition maximum (IRM) was predicted with the equa-tion introduced by Wathen et al. I IRM = 100 * rep wt/[48.8 + 53.8 ^•^exp(-.075^'^reps)Ii (21, 34). The squattest-ing required the athlete's thigh to be parallel to the fioorfor each repetition. The hang clean testing required tbeathlete to clean the bar onto her front deltoids and standwith full body extension for each repetition. The benchpress testing required the athlete to touch tbe barbell tober cbest and return to full arm extension for eacb rep-etition. Kravitz and colleagues demonstrated tbat pre-dicted IRM testing provided acceptable levels of accuracy(20).

Statistical AnalysesStatistical means and standard deviations for each vari-able were calculated for each group. All dependent van-

348 MYER, FORD, BRENT ET AL.

TABLE 2. Core stability training at weeks 1, 3, and 6.

Core stability training Time* RepsWeek 1

Deep hold positionBox butt touchLine jump (forward)-deep holdLine jump (lateral)-deep holdBox drop-deep holdSingle leg squat-deep holdBOSU(F)t deep holdBOSU(F)T drop squatsBOSU(R)t jump stick landing-deep holdBOSU(R)t hoth knees deep holdB0SU(R)1; crunchesBOSU(R)$ swivel crunch (feet planted)BOSU(R)$ single leg pelvic bridgesBOSU{R)$ supermans

Week 3

BOSU(F)t drop stick-deep holdBOSU(F)T deep hold partner perturbationsBox drop (lateral)-deep holdSingle leg line hop (front-back)-deep holdSingle leg line hop (side-side)-deep holdSingle leg squat-heel touchesSwiss ball both knees deep holdBOSU(R):i: single leg step-stick deep holdDouble crunchTable double crunchTable double swivel crunchTable reverse hyperextensionsB0SU(R)1: lateral crunchB0SU(R)1: swimmers

Week 6

Double BOSUlFit deep hold (partner perturbations)BOSU(F)t drop single leg airex stick deep holdBOSU(R)$ single leg deep hold partner ball tossSwiss ball both knees deep hold (partner perturbations)BOSU(R)t single leg (4 + way] hop stick deep holdBOSU(F)t single leg ball pick upAirex walking lungesBOSU(F)t single leg squatsBOSU(F)t single leg deep hold (partner perturbations)Straight leg lifts with toe punchStraight leg lateral double crunchBOSU(R)$ double crunchBOSU(R)$ opposite swivel crunch (feet up)Swiss ball reverse hack hyperextensions

5

20

20

20

10

2520

30

5884

10688

10

35401212

10

488

10

625158

121010

55

1512151212

* Time in seconds.t BOSU flat side up.t BOSU round side up.

ables (vertical ground reaction force, SD of COP, isoki-netic and isoinertial strength and vertical jump) were cal-culated for percent change and analyzed with a 1-wayanalysis of variance (ANOVA) to determine differencesbetween groups. A mixed-design repeated measures AN-OVA {2 X 2) was also used to test for the effect of trainingand training group on vertical jump height and IRMstrength variahles. A mixed-design repeated measuresANOVA ( 2 X 2 X 2 ) was used to test for the effect oftraining, training group and side on vertical ground re-action force, COPMT., COP^J, and isokinetic strength. Sta-tistical analyses were conducted in SPSS for Windows,Release 12.0 (SPSS Inc., Chicago, Illinois). Comparison ofprior biomeehanical data during dynamic tasks foundthat the ratio of within group to between group differ-

ences was 0.73 (23). Based on these data, a power anal-ysis revealed that to achieve 90% statistical power in thecurrent study, with an exploratory alpha level of 0.05, aminimum of 8 subjects per group (PLYO and BAL) wererequired. Therefore, we recruited a team with at least 16athletes that could he randomized into 2 groups.

RESULTSThe BAL and PLYO groups demonstrated differing effectsof training on force dissipation during single leg landings.Figure 1 is a representative graph of landing ground re-action force profiles in a BAL trained athlete pretrainingvs. posttraining. The percent change from pretest to post-test in vertical ground reaction force was significantly dif-ferent between the BAL and PLYO groups on the domi-

PLYOMETRIC VS. DYNAMIC BALANCE TRAINING 349

TABLE 3. Strength training at weeks 1, 3, and 6. TABLE 4. Speed training at weeks 1, 3, and 6.

Strength training Sets Reps

21222111

T-H

112

121012121015101510151020

Week 1Dumbbell hang snatchBench butt touchBarbell squatBench pressLying leg curlLat pulldownsBall squat dumbbell floor touchesDumbbell shoulder pressRussian hamstring curlSeated cable rowHip abduction-adduction at 60° and 120° secDouble crunch

Week 3Hang cleanLeg pressDumbbell incline pressPVont lunges + pressInverted lying pull-upsStretch dumhbell dead lift3-Way dumbbell shoulder circuitBench reverse hyperextensionaKnee flexion-extension at 120° and 300''-secBand good morningsBack extensionsAnkle circuit

Week 6

Dumbbell hang snatchBarbell squatSingle leg band assisted squatBench pressLying leg curlLat pulldownsDumbbell shoulder pressBand shoulder pressRussian hamstring curlStanding cable rowHip ahduction-adductiun at 60° and 120°-secDouble crunch

Speed training Grade {%) Time*#ofSets

108101010121012102012

5585888815

30

nant side (p < 0.05). Figure 2 shows the BAL group re-duced their impact force by 7.0% while the PLYO grouphad an increase of 7.67/. The non-dominant side showeda similar trend, although the results were not statisticallysignificant (BAL -5.49'^, PLYO -hO.3%, p = 0.33). Thepercent increase from pretest to posttest was not differentbetween groups for any other measured variables {p >0.05).

Figure 3 shows the effects of the two different trainingregimens on dynamic force stabilization during single-leglanding. Both the PLYO and BAL groups decreasedCOPML on their dominant side (side X training interac-tion; p < 0.05) during landing onto a force plate from asingle-leg hop, which equalized pretested side to side(dominant to nondominant) differences. Neither BAL norPLYO training affected COP^p during single-leg landingon the portable force platform {p > 0.05).

Figure 4 demonstrates the relative effect of the 2training regimens on strength, as measured on an isoki-netic dynamometer. Both groups increased isokinetichamstrings peak torque {p < 0.01), and hamstrings toquadriceps ratio (p < 0.01). Both training protocols alsosignificantly improved vertical jump (p < 0.001, Figure

Week

RunRunRunRunRunRunRunRun

(6 mph)(8-10 mph)(6-9 mph)(6-8 mph)(6-8 mph)(8-10 mph)(10 mph)(10-12 mph)

Week 3

Run (10-14 mph)Hold (10-12 mph)Run (10 mph)Run (9 mph)Run (8 mph)Run (8 mph)Run (10-12 mph)Retrograde (3-5 mph)

Week 6

Run (12-14 mph)Run (12-14 mph)Run (10-13 mph)Run (12-14 mph)Run (12-16 mph)Run (14-20 mph IRetrograde (5~9 mph)Retrograde (6-12 mph)

Time in seconds.

0010152015100

10152025303505

1015202010050

102020202010108

1281210862015

101010666106

PRE-POST

Time (ms]

FIGURE 1. Mean pretest (dotted linel and posttest (solid line)for the ground reaction forces measured during the single leglanding from a representative subject in the dynamicstabilization and balance training group.

5) and predicted IRM measures of bench press (p <0.001), hang clean {p < 0.001) and parallel squat (p <0.001; Table 5).

DISCUSSION

Comprehensive neuromuscular training can lead to si-multaneous improvements in athletic performance andmovement hiomechanics in female athletes (13, 25, 28).The training protocol in the present study used similar

350 MYP:U, FORD, BRENT ET AL.

HBAL• PLYO

FIGURE 2. Vertical impact force change from pre- to posttestwas calculated as a percent change (negative represented areduction in force) and analyzed. '''Percent reduction of peakimpact three was significantly different between groups p <0.05. BAL = dynamic stabilization and balance training; PLYO= maximum effort plyometric jumping.

SAL Oomnarl BAL Nondomlnant PLYO Oomiranl PLYO Nondominant

FIGURE 3. SD of center of pressure in the medial-lateraldirection p < 0.05. Pre = pretest; Post = posttest; BAL =dynamic stabilization and balance training; PLYO = maximumeffort plyometric jumping.

strength and speed training components as a previouslystudied comprehensive protocol (26). Hov^'ever, the pro-tocol was modified by excluding either dynamic balanceor maximum effort plyometric exercises. The 2 differentprotocols demonstrated similar effects on neuromuscularpower, body sway and dynamic strength. However, thedecreased impact force demonstrated hy the BAL groupmay suggest that balance and dynamic stabilizationtraining are important to improve force attenuation strat-egies when landing from a single-leg hop.

BAL Dominant BAL Nondominant PLYO Dommant PLYO Nondi

FIGURE 4. •''̂ Peak hamstrings torque normalized to bodyweight p < 0.01. PRE = pretest; POST = posttest; BAL =dynamic stabilization and balance training; PLYO - maximumeffort plyometric jumping.

FIGURE 5. '̂ '''"•'Vertical jump in the dynamic stabilization andbalance training I BAL) and maximum effort plyometricjumping (PLYOl groups p < 0.001. PRE = pretest; POST =posttest.

The findings supported the hypothesis that therewould be a difference between the 2 groups in decreasinglanding force when landing from a single-leg hop. As ex-pected, the BAL group significantly reduced impact land-ing forces while the PLYO group (without balance/holdexercises) did not. This result may be related to the in-clusion in the BAL group protocol of force dissipationtechniques, which used both double and single-leg deepknee flexion exercises that emphasized coronal planeknee control. It was demonstrated previously that ply-ometric training that incorporates dynamic stabilizationexercises can reduce impact landing forces (151. The re-

TABLE 5.

Bench pressHang cleanSquat(kg)

Predicted

(kg)(kg)

1 repetition •maximum calculations."

Pre

33.0 ±30.2 ±43.9 ±

4.24.210.9

BAL

40.42.82.

Post

5 ±.4 ±.1 ±

3.2.8.

744

Pre

30.9 ±27.5 ±44.0 ±

5.83.55.8

PLYO

36.40.81.

Post

3 ±3 ±2 ±

659

.7

.5

.0

PPP

p value

< 0.001< 0.001< 0.001

Pre = pretest; Post = posttest; BAL = dynamic stabilization and balance training; PLYO - maximum effort plyometric jumping.

PLYOMETRIC VS. DYNAMIC BALANCE TRAINING 351

suits of the current study may indicate that the improvedforce dissipation strategies at landing are more related tothe dynamic stabilization component of the training,rather than the plyometrics component. The inability ofthe PLYO group, without held landings and balancedrills, to decrease impact force may indicate that the in-corporation of exercises that focus on both force dissipa-tion and dynamic balance may be necessary to obtain posttraining force reduction during single limb landing.

Both groups significantly improved their COP^i ontheir dominant side following training, which equalizedpretested side to side (dominant to nondominant) differ-ences. Protocols that use balance training only, or whenused in combination with plyometric training, can de-crease side to side differences in strength measures (10,15). Females often demonstrate leg dominance, which isan imbalance between muscular strength and joint ki-nematics between contralateral lower extremity mea-sures during dynamic tasks (7, 13). Female athletes maygenerate lower hamstrings torques in the non-dominantleg than in the dominant leg (15). Ford et al. showed thatadolescent female athletes had significant side-to-side dif-ferences in maximum knee valgus angle compared tomales during landing (7).

Side-to-side imbalances in muscular strength, flexihil-ity and coordination have been shown to he importantpredictors of increased injury risk (2, 13, 16). Knapnik etal. demonstrated that side-to-side balance in strength andflexibility is important for the prevention of injuries andwhen imbalances are present, the athlete has an in-creased risk of injury. Baumhauer et al. also found thatindividuals with muscle strength imbalances exhibited ahigher incidence of injury (2). Hewett et al developed amodel to predict ACL injury risk with high sensitivity andspecificity. Nearly half of the parameters in the predictivemodel were side-to-side differences in lower extremity ki-nematics and kinetics (13). Side-to-side imbalances mayincrease risk for both limbs. Over-refiance on the domi-nant limb can put greater stress and torques on the dom-inant knee, while the weaker knee may be at risk due toan inability of the musculature on that side to effectivelyabsorb the high forces associated with sporting activities.Therefore, training programs should target side-to-sideimbalances in females. However, the results of the studymay suggest that the mode of exercise used may not beas critical as the focus on single limb exercises (24). Bothprotocols in the current study progressed to single leg ac-tivities to increase exercise difficulty and intensity, whichmay be related to the improved symmetry in COP ,̂, aftertraining for both groups.

Both the PLYO and BAL significantly improved theCOP measures along the medial/lateral axis as opposedto no change in the COP ,̂,. The improvements in forcedissipation and postural control were expected in the BALgroup; however, it was hypothesized that the PLYO groupwould not improve postural sway with training. The me-dial/lateral COP improvements by both groups contrastthe findings of Paterno et al. who used a combined dy-namic balance and plyometric protocol (28) and foundthat improvements in body sway measures occurred inthe anterior/posterior plane. The difference in resultsmay be related to the difference in measurements of dy-namic balance. The current study measured COP controlof the athlete in relation to a stable surface, while Pater-no et al. measured body sway on an unstable platform

(28). The study results may be indicative of a more sport-specific and injury relevant training adaptation, since theathletes in the current study performed a more sportsrelated task that required a dynamic ground based land-ing with immediate control of the ground reaction forcesto maintain halance and COP control. The role of COPcontrol and improved balance in reducing lower extremityinjury risk (4, 32, 33) further illustrates the positive na-ture of this improved measure.

Both the BAL and PLYO groups showed large increas-es in open chain isokinetic hamstring measures (Figure5). Complimentary to the increased hamstring strength,both groups significantly improved their hamstringstrength relative to their quadriceps strength (H/Q ratio).These results were expected as previous authors havedemonstrated that isolated plyometric and balance train-ing can improve H/Q ratio (10, 36). Additionally, bothgroups received some form of isolated hamstrings exerciseper training bout. Fry and Powell hypothesized that gen-eralized strength training may not improve H/Q ratio un-less hamstring isolated exercises are incorporated intothe protocol (9). Electromyographic analysis demon-strates that during the leg press hamstring co-activationis significantly reduced compared to the squat exercise(35). Therefore, the observed improvements in the H/Qratio may be a combinatory effect between training vari-ables (PLYO and BAL) and the isolated hamstringstrengthening and squat exercises incorporated into bothgroups' protocols. The increased H/Q ratio from bothPLYO and BAL groups should he considered a positiveadaptation to injury prevention training as low H/Q ratioshave been related to increased risk of traumatic lowerextremity injury in female athletes (30).

Both the PLYO and BAL group improved theirstrength measures (squat and hang clean) and power(vertical jump) measures. Fatouros and colleagues alsofound combinatory effects of plyometrics and resistancetraining to not only increase jump performance but legstrength (5). The PLYO increases were also similar to theresults demonstrated by others who found the effects ofplyometrics appear to be combinatory with resistancetraining. Adams et al. reported that a combined plyome-tric and squat training program significantly increasedvertical jump versus training with squats or plyometricsalone (1). We hypothesized that the PLYO group woulddemonstrate increased strength and power similar to pre-vious reports (15, 26). However, improvements in verticaljump with a protocol that focused on dynamic stabiliza-tion and balance was unexpected. The measured improve-ments in vertical jump may be more related to the resis-tance training included in the protocol, which may be animportant component in all injury prevention protocols(18, 19). Training protocols that focus on resistance train-ing alone may not reduce ACL injury rates. However,there is inferential evidence that resistance training mayreduce injury risk due to the beneficial adaptations thatoccur in hones, ligaments and tendons following training(6, 17). Protocols that supplement plyometric and tech-nique training with strength training may significantlyreduce ACL injuries in female athletes (15). Thus, it ap-pears that resistance training may be effective at reduc-ing knee injuries when combined with other training com-ponents. However, the efficacy of a single-faceted resis-tance training protocol on ACL injury prevention has yetto be demonstrated in the literature.

352 MvER, FORD, BiiRNT KT AL.

PRACTICAL APPLICATIONS

The goals of the study to evaluate the effects of includingmaximum effort plyometric jumps or dynamic lower ex-tremity stabilization exercises in a dynamic neuromus-cular training program shown to reduce measures relatedto ACL injury and increase measures of performancewere met. The results of this study indicate that bothPLYO and BAL training are effective at increasing mea-sures of lower extremity neuromuscular power and con-trol as well as decreasing leg dominance. However, theBAL component may be more important to improve singlelimb force attenuation strategies. A combination of PLYOand BAL training may further maximize the effectivenessof preseason training for female athletes.

REFERENCES

1. ADAMS, K,, J.P. O'SHEA, K.L. O'SHEA, AND M. CLIMSTEIN. Theeffect of six weeks of squat, piyoinetric and squat-plyometrictraining on power production. J. Strength Cond. Res. 6:36-41.1992.

2. BAUMHAUER. J. , D. ALOSA, A. RENSTWJM, S. TKKVINO, AND B.BEYNNON. A prospective study of ankle injury risk factors. Am.J. Sport Med. 23:564-570. 1995.

3. BKENT. J.L., K.R. FoKD, G.D. MYER. A.D. HAKRISON, AND T.E.HKWE'IT. Reliability of single leg landings on a portable forceplatform. Med. Set. Spurts Exerc. 37(Suppll:S400.

4. CARAFFA, A., G. CERULLI, M. PROJETTI, G. AISA, AND A. Ry/.zo.Prevention of anterior cruciate ligament injuries in soccer, Aprospective controlled study of proprioceptive training. KneeSurg. Sports Traumatol. Arthro.sc. 4:19-21. 1996.

5. FATUUROS, I.G., A.Z. JAMUKTAB, D. LEONTSINI, T. KYRIAKOS,N. AGGELOUSI.S, N. KOH'I'OPOULOS, AND P. BUCKENMEYEK. Eval-uation of plyometric exercise training, weight training, andtheir combination on vertical jumping performance and legstrength. J. Strength Cond. Res. 14:470-476. 2000.

6. FLKCK, S.J., AND J.E. FAI.KEL. Value of resistance training forthe reduction of sports injuries. Sports Med. 3:61-68. 1986.

7. FORD, K.R., G.D. MYER, AND T.E. HEWETI-. Valgus knee motionduring landing in high school female and male basketball play-ers. Med. Sci. Sports Exerc. 35:1745-1750. 2003.

8. FORD, K.R., G.D. Mvi-Ui, R.L. SMITH, R.N. BYRNES, S.E. Doi'i-RAK, AND T.E. HEWET'I'. Effects ol' an overhead goal on dropvertical jump maneuvers. J. Strength Cimd. Res. 19. 2005.

9. FRY, A.C, AND D.R. POWELL. Hamstring/quadricep parity withthree different weight training methods. J. Sports Med. Phys.Fitness 27:362-367. 1987.

10. HEITKAMF, H.C., T. HDRSTMANN, F, MAYER, J. WF.I.I.ER, ANDH.H. DiCKHUTH. Gain in strength and muscular balance afterbalance training. Int. J. Sports Med. 22:285-90. 2001.

11. HEWETT, T.E., T.N. LINDENFEI.D, J.V. RICCX>RENE, AND F.R.NOYES. The efl'ect of neuromuscular training on the incidenceof knee injury in female athletes. A prospective study. Am. J.Sports Med. 27:699-706. 1999.

12. HEWE'IT, T.E., G.D. MYER. AND K.R. FORD. Decrease in neu-romuscular control about the knee with maturation in femaleathletes. J. Bone Joint Surg. Am. 86-A:1601-1608. 2004.

13. HEWETT, T.E., G.D. MYER, K.R. FORD, R.S. HEIDT, JR., A.J.CoLOsiMO, S.G. MCLEAN, A.J. VAN DEN BOGERT, M.V. PATER-NO, AND p. Succop. Biomechanical measures of neuromuscularcontrol and valgus loading of the knee predict anterior cruciateligament injury risk in female athletes. Am. J. Sports Med. 33:492-501. 2005.

14. HEWKT-1', T.E., M.V. PATKRNO, AN[> K,R. NOYES, Differences insingle leg balance on an unstable platform between female andmale normal, ACL-deficient and ACL-reconstructed knees. In:Proprioception and Neuromuscular Control in Joint Stability. S.Lephardt and F.H. Fu, eds. Champaign, IL: Human Kinetics,1999. pp. 77-88.

15. HEWETT, T.E., A.L. STROUPE, T.A. NANCE, AND F.R. NOYES.Plyometric training in female athletes. Decreased impact forc-es and increased hamstring torques. Am. J. Sports Med. 24:765-773. 1996.

16. KNAPIK, J.J., C.L. BAUMAN, B.H. JONES. J.M. HARRIS, AND L.VAUCHAN. Preseason strength and flexibility imhalances as-sociated witb athletic injuries in female collegiate athletes.Am. J Sports Med. 19:76-81. 1991.

17. KRAEMER, W.J., N.D. DUNCAN, AND J.S. VOI.EK. Resistancetraining and elite athletes: Adaptations and program consid-erations. J. Orthop Sports Phys. Ther 28:110-119. 1998.

18. KHAEMER, W.J., K. HAKKINEN, N,T. TRIPi.E'rr-McBRIDE, A.C.FRY, L .P . KOZIRIS, N.A. RATAMESS, J.E. BAUER, J.S. VOLEK, T.McCoNNELL, R.U. NEWTON, S.E. GORDON, D. CUMMINGS, J.HAUTH, F . PuLi.o, J.M. LYNCH, S.A. MAZZETTI, AND H.G.KNUTTGEN. Physiological changes with periodized resistancetraining in women tennis players. Med. Sci. Sports Exerc. 35:157-168. 2003.

19. KRAEMER, W.J., S.A. MAZZETTI, B.C. NINDI., L.A. GOTSIIAI.K,J.S. VOLEK, J.A. BUSH, J.O. MAJtx, K. DOHI, A.L. GOMEZ, M.MILES, S.J. FLECK, R.U. NEWTON, AND K, HAKKINEN. Effect ofresistance training on women's strengtb/power and occupation-al performances. Med. ScL Sports Exere. 33:1011-1025. 2001.

20. KRAVITZ, L., C. AKAI^N, K. NOWICKI, AND S.J. KINZEY. Predic-tion of 1 repetition maximum in higb-school power lifters. J.Strength Cond. Res. 17:167-172. 2003.

21. LESUER, D.A., J.H. MCCORMICK, J.L. MAYHEW, R.L. WAS.SER-STEiN, AND M.D. ARNOLD. The accuracy of prediction equationsfor estimating 1-RM performance in the bench press, squat,and deadlift. J Strength Cond. Res. 11:211-213. 1997.

22. MANDELBAUM, B.R., H.J. SILVERS, D. WATANABE, J. KNARR, S.THOMAS, L. GRIFFIN, D.T. KIRKENDALL, AND W.J. GARRETT. Ef-fectiveness of a neuromuscular and proprioceptive trainingprogram in preventing the incidence of A(3L injuries in femaleatbletes: Two-year follow up. Am. J. Sports Med. 2005.

23. MCLEAN, S.G., R.J. NEAL, P.T. MYERS, AND MR. WALTERS.Knee joint kinematics during the sidestep cutting maneuver:Potential for injury in women. Med. Sci. Sports Exerc. 31:959-968. 1999.

24. MYER, G.D., K.R. FORD, AND T.E. HEWEIT. Rationale and clin-ical techniques for anterior cruciate ligament injury preventionin female athletes. J. Athl. Train. 39:352-364. 2004.

25. MYER. G.D., K.R. FORD, AND T.E. HEWETT. The effects of gen-der on quadriceps muscle activation strategies during a ma-neuver that mimics a high ACL injury risk position. J. Electro-myogi: Kinesiol. 15:181-189. 2005.

26. MYER, G.D., K.R. FORD, J.P. PALUMBO, AND T.E. HEWETT. Neu-romuscular training improves performance and lower-extrem-ity biomechanics in female athletes. J. Strength Cond. Res. 19:51-60. 2005.

27. MYKLEBUST, G., L. ENCEBRETSEN, LH. BRAEKKEN, A. SK-JOL-BERG, O.E. OLSEN, AND R. BAHR. Prevention of anterior cruci-ate ligament injuries in female team handhalt players: A pro-spective intervention study over three seasons. CUn. J. SportMed. 13:71-78. 2003.

28. PATERNO, M.V., G.D. MYER. K.R. FORD, AND T.E. HEWETI',Neuromuscular training improves single-limb stability inyoung female athletes. J. Orthop. Sports Phys. Ther. 34:305-317. 2004.

29. PETERSEN, W., C. BRAUN, W. BOCK, K. SCHMIDT, A. WEIMANN,W. DRESCHER, E. EiLiNG, R. STANGE, T. FUCHS, J . HEDDERICH,AND T. ZANTOP. A controlled prospective case control study ofa prevention training program in female team handball play-ers: The German experience. Arch. Orthop Trauma Surg. 2005.

30. SoDERMAN, K., H. ALEREDSON, T. PIETILA, AND S. WERNER.Risk factors for leg injuries in female soccer players: A pro-spective investigation during one out-door season. Knee Surg.Sports Traumatol Arthrosc. 9:313-321. 2001.

31. STOCKBRUCJGER, B.A., AND R.G. HAENNEL. Validity and reli-ability of a medicine ball explosive power test. J. StrengthCond. Res. 15:431-438. 2001.

PLYOMETRIC VS. DYNAMIC BALANCE TRAINING 353

32. TROPP, H., J . EKSTRAND, AND J. GILLQUIST. Stabilometry infunctional instability of the ankle and its value in predictinginjury. Med. Sci. Sports Exerc. 16:64-66. 1984.

33. TROFF, H., AND P. ODENRICK. Postural control in single-limbstance. J. Orthop. Res. 6:833-839. 1988.

34. WATHEN, D. Load assignment. In: Essentials of strength train-ing and conditioning. T.R. Baechle, ed. Champaign, IL: HumanKinetics, 1994. pp. 435^39.

35. WiLK, K.E., R.F. EscAMiLLA, G.S. FLEISIG, S.W. BARRENTINE,J.R. ANDREWS, AND M.L. BOYD. A comparison of tibiofemoraljoint forces and electromyographic activity during open andclosed kinetic chain exercises. Am. J. Sports Med. 24:518-527.1996.

36. WiLKERSON, G.B., M.A. COLSTON, N.L SHORT, K.L. NEAL, P.E.HoEWiscHER, AND J.J. PiXLEY. Neuromuscular changes in fe-male collegiate athletes resulting from a plyometric jump-training program. J. Athl. Train. 39:17-23. 2004.

Acknowledgments

The authors would like to acknowledge funding support fromNational Institutes of Health Grant R01-AR049735-01A1. Tbeauthors would also like to thank Coach Kerry Buttkovich andtbe Division 1 2004 Ohio State Semifinalist and 2005 StateChampion Seton High School Volleyball team for theirparticipation in this study. Tbe authors would also like toacknowledge Hanni Cowley, Elizabetb Brougber, Dr. Jon Divine,Adrick Harrison, Rachel Heyl, Rachel Mees, Monica Naltner,Carmen Quatman, Nick Palumbo, Mark Paterno, AnnieSchmolt, and Merry Jo Ford for their assistance with trainingand testing of athletes and for their input to the manuscriptedits.

Address correspondence to Gregory D. Myer, greg.myer@cchm c. org