Case of the Week 01/09/98 (ESMAC conference, Belfast 1998):

What people said...

My first impression was that the main abnormality is Trendelenberg. This
seems to be bilateral, although the pelvic obliquity (right hemipelvis
up) complicates things. I wondered whether this is caused by a scoliosis
(convex left). Looking at the hip moment in the frontal-plane, the right
side is normal, while the left is low, suggesting that only the left hip
abductors are weak. But I confess I'm not sure.

Agreed about the jump knee, and it's interesting that she gets full
extension (even hyperextension on the left) in late stance - suggesting
that the hamstrings are only tight when the hip is flexed. What's
causing the hyperextension on the left? I guess weakness of the left
quadriceps (see knee moment curve). She's using the GRV to stabilise the
knee during loading. Would Botox worsen this hyperextension and result
in recurvatum? I guess it might.

Also agreed about the left ankle, which shows a double bump in it's
power curve, indicating clonus. Both EMGs show abnormal activity at
contact, which is flat-footed. Neither ankle produces a good push-off,
and are in mild equinus.

There are rotational problems, especially on the left, causing
toeing-in, and there are lateral arm thrusts (both arms) to the left in
that leg's swing. I'd be interested in an explanation for these.

What is the primary abnormality, then? I think weakness, especially
left: hip abductors, quads and plantarflexors. The rotational
deformities are secondary to this.

Dr. Chris Kirtley MD PhD
Dept. of Rehabilitation Sciences
The Hong Kong Polytechnic University
Hong Kong
Special Administrative Region of The People's Republic of China

I partly agree with Dr. Kirtley.

In Frontal plane;
    Her pelvis has fixed obliquity throughout the whole gait cycle, but it has its own up and down curve which seems to be exaggerated than normal one. Her upper trunk tilts to left(pelvic obliquity left down) just after right foot off, which is supported by frontal pelvis and hip kinematics and kinetics as Dr. Kirtley mentioned.
    This may be a result from compensated gluteus medius gait left or absolute or relative longer leg right. I think the frontal hip moment may not differentiate these possibilities. Her left knee flexion and ankle dorsiflexion during swing phase shows greater than left leg, so, if she does not have absolute leg length discrepancy, the frontal plane abnormality seems to be a compensated gluteus medius gait which may be due to left gluteus medius weakness.

In Sagittal plane;
    In kinematics, She contacts the ground with her left forefoot(video) and ankle plantar flexed(kinematics) then, the ankle dorsiflexes just a little but never to neutral. During this loading phase, her knee continuously extends from excessively flexed position at initial contact point to hyperextended position in mid and terminal stance phase.
    In kinetics, the ankle exerts an abnormal eccentric plantar flexion moment during loading phase then, in mid stance phase, it reduces plantar flexion moment a little and changes to concentric contraction. During mid stance phase, the knee shows flexion moment with knee extending velocity. It is supported by emg data which shows prolonged hamstring activity with extending knee during mid stance  phase.

    I think almost all of these abnormalities come from the gastrocnemius(GCM) tightness or spasticity. The tight GCM seems to make the knee flexed in terminal swing phase and initial contact point. The forceful knee extension during loading and early single limb stance phase may be due to the tight GCM which acts as a strong knee extensor in closed kinetic chain.

    In this kind of case, I believe that a solid ankle AFO is not a good solution. It forces the ankle to dorsiflex to correct the plantar flexed ankle but it increases the tone of the tight GCM. This will inhibit the knee extension in the terminal swing phase(open kinetic chain) and make the knee to jump(in closed kinetic chain). So, any procedure(Phenol, botox injection or TAL ...) which would release or relax the tight GCM may work for the girl.

    Sorry for long and clumsy words.
    My wrong ideas may be indicated by the members.


Sun G. Chung M.D., Ph.D.
Dept Rehab Med
Seoul National University College of Medicine
Chong Ro Ku YeonGeon Dong Seoul
South Korea
(TEL)82-2-760-2619  (FAX)82-2-743-7473

In response to comments made so far we can clarify that the girl in this case does not have any
scoliosis (fixed or postural) nor does she have a true leg length discrepancy.

We considered her to be a mildly affected diplegic and that weakness was probably not a major
factor, certainly in terms of a primary cause of her gait abnormality.

We were thinking along the lines that this was a problem of dynamic hamstring tightness
combined with weak and overactive gastrocnemei. Popliteal angles are low yet tightness is seen
when this two joint muscle is on the stretch proximally and distally at terminal  swing /
initial contact. However at terminal stance the hamstrings are slackened proximally and no
longer restrict knee extension.

Do people think the gastrocnemius and / or hamstrings should be targeted with treatment and if
so how? What would be the likely out come of this treatment?
Would the left thigh rotation and trunk / pelvic movement limit the effectiveness of this

Oxford Gait Laboratory Team
Nuffield Orthopaedic Centre
Oxford                                                  tel:     01865 227609
OX3 7LD                                         fax:    01865 744277

Some comments on ESMAC case 2 (case KA)

1 - In relation to the pelvis and hips, while we see some degree of
pelvic obliquity we are confused by the hip Abd/Add kinematics. Where
there would appear to be adduction of the left hip when weight bearing
this is not borne out by the kinematics. We find the hip kinematic
graph confusing here.

2 - With regard to tripping, this may be due to a weak Tibialis anterior
muscle as a drop foot is displayed during swing. An alternative
explanation would be a spastic Tibialis posterior muscle which has been
masked by casting. EMGs of the Tibialis anterior/Tibialis posterior
muscles would be essential here.

3 - If there are skeletal rotational problems present these should be
confirmed by ultrasound scan or CT scan.

Yours sincerely,

Prof. Tim o' Brien,
Professor of Orthopaedic Surgery,
Central Remedial Clinic,
Dublin 3,

Dear Chris,

The concept that muscles can produce accelerations in a direction
opposite to their moments (e.g., gastrocnemius producing a knee extension
acceleration) is reviewed in an article that was written by Felix Zajac
and Mike Gordon.  (Zajac FE , Gordon ME: Determining muscle's force and
action in multi-articular movement. Exercise and Sport Sciences Reviews 17:
187-230, 1989).  We have performed dynamic simulations of gait and found
similar results (e.g., Piazza, S. J., Delp S. L.: Influence of muscles
on knee flexion during the swing phase of normal gait. Journal of
Biomechanics vol. 29, pp. 723-733, 1996.).  I hope this information is helpful to

I think there are about 50 research centers that use SIMM these days.
Hopefully there will be more when the PC version comes out later this


Scott L. Delp
Assoc. Professor                      Rehab. Inst. of Chicago (1406)
Biomedical Engineering            345 East Superior Street
and Rehabilitation Medicine      Chicago, IL  60611
Northwestern University          fax: 312-908-2208
email:               voice: 312-908-8860

On the subject of gastrocs action, I've looked up the papers that Scott Delp mentioned. Here are two salient paragraphs:

"Our finding that the gastrocnemius (G) acts to extend the knee in swing was unexpected because the G passes posterior to the knee... G is a biarticular muscle that produces a knee flexion moment and an ankle plantarflexion moment. These moments produce opposing accelerations of the knee joint: knee flexion moment accelerates the knee in flexion and ankle plantarflexion moment accelerates the knee in extension. If the latter acceleration is larger than the former, a net knee extension acceleration results". (Piazza & Delp 1996, J. Biomech. 29:733).

"Biarticular muscles can have a multitude of actions. For example, G, which exerts both ankle-extensor and knee flexor torques, can act to (a) extend the ankle & flex the knee, (b) flex the ankle & flex the knee, or (c) extend the ankle & flex the knee. Since a muscle's action depends on the position of the body and on the muscle's interaction with external objects (such as the ground), it can vary among motor tasks, and even during a single motor task". (Zajac & Gordon 1989, Ex. & Sports Sci. Rev. 17:187).

I think clinicians should be aware of these findings when making intuitive judgements about where to inject their Botulinum! Muscle action is a much more complicated business than meets the eye.

Dr. Chris Kirtley MD PhD
Dept. of Rehabilitation Sciences
The Hong Kong Polytechnic University
Hong Kong
Special Administrative Region of The People's Republic of China

Back to Case 01-09-98