Dynamic Knee Stability and Perturbation Training

Dynamic Knee Stability

• Rudolph et al, KSSTA 2001• 31 active subjects

– 10 uninjured– 11 copers– 10 non-copers

• Screening evaluation• Quadriceps strength testing• Knee joint laxity testing

Testing

• EMG testing– Normalized to maximum EMG– Variables

• Muscle onset (threshold 2.5x ave rest EMG)• Termination of activity• Magnitude (Integration over weight acceptance

interval)• Co-contraction of VL-LH, VL-MG

Testing

• 3D motion analysis– Self-selected walking, jogging speeds– Joint motions, moments– Support moments

WalkingCoper Non-coper ControlInvolved Uninvolved Involved Uninvolved Involved Uninvolved

Vert GRF at loading (F=8.499, P=0.017)*

1.25%BW (±0.030)

1.23%BW (±0.022)

1.22%BW (±0.031)

1.26%BW (±0.033)

1.31%BW* (±0.031)

1.29%BW* (±0.033)

Peak knee flexion angle (negative=flexion) (F=8.499, P=0.017)**

-22.8° (±1.9)

-24.5° (±1.8)

-21.9°** (±1.9)

-25.9° (±1.9)

-26.5° (±1.9)

-26.5° (±1.9)

Knee moment at PKF (F=6.212, P=0.034)**

0.368 (±0.07)

0.437 (±0.68)

0.314** (±0.071)

0.542 (±0.071)

0.558 (±0.074)

0.601 (±0.071)

Soleus integral over wt acceptance (t=2.894, P=0.020)**

8.655 (±1.292)

7.489 (±0.679)

9.811** (±1.362)

6.626 (±0.716)

8.302 (±1.292)

7.612 (±0.679)

*Control group different from copers and non-copers (P<0.05)**Non-copers' involved side different from all others (P<0.05)

Walking

Distribution of support moments on the involved side during weight acceptance, walking. Non-copers* had lower knee moments (F=5.402, P=0.045) and higher hip moments (F=3.979, P=0.056) than copers or uninjured subjects

Walking

Knee flexion angle External knee flexion moment

Quadriceps index

Copers Copers

r=0.029, P=0.932 r=0.135, P=0.693

Non-copers Non-copers

r=0.933, P=0.000* r=0.716, P=0.030*

Lateral hamstrings Onset-to-peak EMG

Copers Copers

r=0.672, P=0.030* r=0.765, P=0.010*

Non-copers Non-copers

r=0.095, P=0.824 r=0.408, P=0.316

Non-coper Coper

Quadriceps strength (t=4.033, P=0.001)* 75.3% (±11%) 97.1% (±12.7%)

Walking

• Regression analyses – 79.5% of the variability in the knee moment at peak

knee flexion accounted for by the variability in the onset-to-peak of the hamstrings and quadriceps muscles (F=6.009, P=0.030) in the copers only.

Jogging

Copers Non-copers Controls

Involved Un-involved Involved Un-involved Involved Un-involved

Velocity m/s per LL (F=4.00, P=0.03)*

4.041 (±0.23) 4.00 (±0.19) 4.137

(±0.24)4.236 (±0.21)

4.745* (±0.23)

4.885* (±0.19)

Stride length m/LL (F=4.30, P=0.029)*

3.089 (±0.13)

3.034 (±0.11)

3.194 (±0.18)

3.297 (±0.15)

3.575* (±0.15)

3.592* (±0.13)

Vertical ground reaction force (F=2.849, P=0.075)

2.172 (±0.07)

2.204 (±0.07)

2.084 (±0.07)

2.156 (±0.07)

2.322 (±0.07)

2.357 (±0.07)

JoggingKnee flexion Knee moment

Jogging

Non-copers had significantly greater hip (F=3.3994, P=0.030) and less knee (F=4.727, P=0.017) extensor moments on the involved sides

Jogging

Non-copers had significantly greater co-contraction between vastus lateralis and medial gastrocnemius in the involved limb (*F=3.609, P=0.041)

Peak knee flexion angle

Knee moment at peak knee flexion

Passive laxity

CopersNS

r=0.203, P=1.000

Non-copersNS

r=-0.866, P=0.015*

Quadriceps index

CopersNS

r=-0.133, P=1.000

Non-copersNS

r=-0.798, P=0.060**

VL-LH co-contraction

CopersNS

r=-0.417, P=0.231

Non-copersNS

r=-0.670, P=0.048*

Jogging

• Regression analyses – 83.5% of the variability in the knee moment at peak

knee flexion accounted for by the variability in the amount of VL-LH and VL-MG co-contraction (F=15.231, P=0.004) in the non-copers only.

Conclusions

COPERS NON-COPERS

• Normal knee motions and moments

• Less co-activation• Muscle activation-

important factor in stability

• Compensation related to quadriceps strength, passive knee laxity

• Reduced knee flexor moment

• Reduced knee motion• Transfer control to hip• Possible delayed force

production?

Perturbation training

• Fitzgerald et al, PT 2000• 26 subjects completed training

– 14 subjects in standard group– 12 subjects in perturbation group

• Screening exam– Pass “rehab candidate” criteria

Training programs

STANDARD PROGRAMPERTURBATION

TRAINING

• Resistance training to quads and hams

• Cardiovascular endurance training

• Agility training• Sport-specific skill

training

• AP, ML on Balance Master

• AP, ML rotary on tiltboard

• Rollerboard/Platform• Multi-directional on

rollerboard

Treatment outcomes

• Unsuccessful rehab– Episode of knee giving way– Status reduction from rehab candidate to high risk for

reinjury on retesting• Outcome measures

– MVIC quadriceps– Single-limb hop tests– Knee joint laxity– KOS-ADLs– KOS-Sports– Global Rating Scale

Results

Greater number of subjects in the standard group had unsuccessful rehabilitation (χ2=5.27, critical value=3.84, P<.05)

Positive likelihood ratio was 4.88 ([11/18]/1–[7/8])

Results

KOS-ADLS interaction. P<.05

KOS-Sports interaction. P=.12

Results

GRS interaction. P<.05

X-over Hop interaction. P<.05

Hop TestingPerturbation (Immed after)

Perturbation (F/u)

Standard (Immed after)

Standard (F/u)

Single Hop 101%(±14%)* 68%(±48%)

X-over Hop 105%(±13%) 104%(±16%) 100%(±15%) 64%(±55%)

Triple Hop 99%(±12%)* 59%(±51%)

* P<.05 at follow-up

Conclusions

More subjects in standard group (50%) had unsuccessful rehab compared to pert group (92%)

~5x more likely to successfully return to high-level activities if receive perturbation training

Pre to post trainingADLs, GRS, X-over hop improved in both groups

Post training to follow-upMaintained in pert group, in standard group

Development of Dynamic Stability

• Chmielewski et al, J Electromyo & Kinesiology 2002• 9 subjects

– Passed screen and wanted to attempt to return to activity

• Quadriceps strength testing• Screening exam

EMG testing

• VL, LH, MG, SOL• Muscle timing onset• Termination of activity• Muscle activity duration• Time to peak amplitude• Peak amplitude• Integral of muscle activity during loading response

Training

• Perturbation training• Agility training• Resistance strength training for quads, hams, and gastrocs

Screening examination

Pre training Post training

Quadriceps index 90.9%(±12.5%) 91.3%(±5.8%)

Timed Hop 97.3%(±7.95%) 96.0%(±4.9%)

KOS-ADLs 91.8%(±6.6%) 97.3%(±2.3%)*

GRS 83.7%(±13.6%) 94.3%(±4.3%)*

P<.05

EMG testing

* VL integral of activity during walking is significantly increased after perturbation training (p<0.05)

VL activity integrated from 100 ms before initial contact to peak knee flexion is less before training (a) compared to after training (b).

EMG testing

Timing of muscle activity during walking before and after perturbation training

Conclusions

• Quadriceps activity integral after training• Relationship of quadriceps activity, peak magnitude, time-

to-peak activity with hamstrings and soleus activation– Influenced by training

• Self-reports sports , functional test