Case of The Week 4/8/97 (Iliac Crest Bone Harvest Surgery): What people said...

Markedly slow speed, caused by short stride length, with long stance phase bilaterally (but especially on the left) and much reduced cadence. Step length is a centimetre longer on the left. Trunk and pelvis rotated to the left, causing the operated side to lead the healthy side through the cycle, as Andreas mentions. Left hemipelvis lower than right, and reduced knee flexion (both stance and swing) on the right. In the video the right leg seems externally rotated, but the kinematics show, in fact, that this is an illusion caused (presumably) by the pelvic rotation. If anything, the left is a little bit internally rotated. I get the impression of a little circumduction on the left (which would fit with the pelvic drop on that side), but I may be imagining it.

What would one expect of an operation on the iliac crest? I guess it might disrupt iliacus, gluteus medius and/or the abdominals. Of the three, it seems that the major effect is on the gluteus medius. The frontal hip moment curves are almost normal, but power generation in that plane is grossly abnormal, with normally concentric activity becoming eccentric, especially on the operated side. In other words, she has a Trendelenberg gait, in which the pelvis drops on the left during stance due to weakened hip abductors on the right.

So, in summary, the main effect of the surgery is to compromise the hip abductors (not surprising since the gluteus medius is cut through in the operation). General gait measures show that the gait is very slow (was this her natural walking speed?) due to a combination of increased stance duration and reduced stride length. It's difficult to be certain why - primary or compensatory. I'd say the latter.

What could we do for her? Strengthening of the weak gluteus medius might be expected to be fairly successful, since the moment curve seems to indicate that there's no paralysis of the muscle - just inadequate power generation.

Younger and Chapman (1) reviewed 239 patients who had 243 autogenous bone grafts and found an 8.6% incidence of major complications and 20.6% incidence of minor complications. I guess this patient would be classified as "minor", but a walking velocity of 0.66 m/s is quite a disability. Synthetic bone grafts are unlikely to replace autogenous bone grafts so it looks like we'll be seeing the gait complications of the operation for some time yet!

1. Younger EM, Chapman MW: Morbidity at bone graft donor sites. J Orthop Trauma 1989;3:192-5.

Chris -- Dr. Chris Kirtley (Kwok Kei Chi) MD PhD Assistant Professor Department of Rehabilitation Sciences The Hong Kong Polytechnic University Hong Kong

Well, I'm a pediatric PT, but I'll have a bash!

Observationally, I notice a lack of reciprocal arm swing and a sense of this lady being "careful" and somewhat tentative, especially when loading the right. I note that she is 63 and wonder if I have to consider primary balance and other aging issues as well as the surgery, which was 2 years ago. She also has some adipose tissue around her tummy and hips, which makes me wonder about marker jiggle and tends to make me expect a larger measurement error due to marker placement.

Observationally, she certainly look to have a classic Trendelenberg gait: the clue is the trunk movement (could it be the dreaded gluteus medius lurch??) Her velocity is decreased but to the naked eye, her stride doesn't look too bad, so I suspect decreased cadence as the primary reason for decreased velocity. This makes me wonder again about primary balance issues. All 3 rockers are present at both ankles and she doesn't seem to have clearance issues. The left knee looks OK, the right a little stiff, perhaps slightly decreased peak knee flexion. The right hip looks a little stiffer than the left and the pelvis definitely has some major asymmetries, the most notable being in the coronal plane. The video is a bit oblique, so I have some trouble seeing transverse pelvic issues.

Kinematics confirm that she leans way to the right while swinging through the left leg (direct link to right abductor weakness). The pelvis up and rotated forward on the right throughout the cycle. There is decreased dynamic ROM in the coronal plane at the pelvis: the right is held up, but the pelvis doesn't activeley move in this plane - she uses her trunk and the pelvis moves with the trunk.

The right hip actually abducts a little in swing: this is a compensation for both the decreased right peak knee flexion and the increased right plantar flexion we see at initial contact, but an interesting one considering we are assuming right abductor insufficiency. The right hip is a little laterally rotated to maintain a normal FPA with the forward rotated right hemipelvis.

Kinetics show decreased power bursts at hip and ankle, which we expect because of decreased velocity. The moment patterns are not too far from normal (which is based on healthy young adults, anyway...they may be normal for the elderly). What interests me most in the kinetics is the inverted power graph at the right hip in the coronal plane, again suggesting problems with the right abductor muscle group.

As a PT, I would indeed target specific right hip abductor strengthening for this lady in order to try to improve the Trendelenberg. I am also interested in further testing balance and finding out about her overall strength/fitness. My gut instinct makes me wonder if there is another process at work here other than the previous bone graft. General balance and fitness training may be part of the answer that will help improve velocity and bring cadence up to closer to normal.

Jenni Dabelstein

I'm a PhD student at the department of Biophysics and Orthopedics of the Nijmegen University Hospital. The aim of my study is the gait recovery in patients treated for benign and malign bone and soft tissue tumors of the lower limb. Especially i'm interested in the patients treated with limbsaving procedurs. On the gait analysis I'm an absolute beginner and i have learned a lot from the "cases of the week"

What I see in my patients with pelvic tumors is rather the same as shown in this case. There is an assymetry in the gait between affected and non- affected leg (stance fase, swing fase and kineticts), a more or less classic Trendelburg gait, a slow (prefferred) walking speed and reduced stride time. Som of these parameters cannot be declared by the fact that their is a weakness of the muscles. After two years you will expect that during walking in daily life the muscle strength is returned. Even it is a good idea to start a rehabilitation for the muscle weakness.

But in our opnion the reason why these kind of patients walks so badly is that there is not a complete level of re-automatisation of walking. The difference between preferred walking velocity and maximal walking velocity is small. This means they still depend on systems like, the visual system and cognitive function, the perfrom the walking 'task'. We have measured the infleuence of these system on thier gait parameters like walking velocity stride- and double supporttime. And we found that under walking with constraints they walks more badly. And we have concluded that the gait -automatisation in these patients is not completed so it is difficult for them to walk in the daily situation with lots of hinderances and distrators.

I hope that this is a usefull contribution to the discussion.

Rico de Visser Rico de Visser University of Nijmegen The Netherlands


I must say, Rico, that I haven't heard of such a method of differentiating between automated and subconsious walking being described previously - great idea! Rather like the Stroup test for distracting subjects in a motor control task.

This brings us back to the old question of what determines an individual's walking speed, cadence and stride length. I had a discussion about this recently with Josef Kollmittzer in Vienna (perhaps you'd like to comment, Sepp?). Basically, I now view cadence and stride length something like heart rate and stroke volume. Cardiac output (walking velocity) can be controlled by altering either parameter. Of the two, stride length seems to be the most sensitive to pathology, and may be compensated for by increasing the cadence, as increasing the pulse compensates for circulatory disturbances. In his Physical Therapy of 1990 (70:340-347), Winter noted that elderly people being screened for fitness had normal cadence, but reduced stride length.

Actually, I just looked at the review article in a recent Gait & Posture (1997) Gait in the Elderly (Prince et al 5: 128-135). I was quite shocked to see that there seems to be no real consensus of what happens to the basic gait measures in the normal elderly! In the discussion on pp 130-131, all the studies seem to be on a few subjects and the results are quite contradictory. Is the situation really as bad as that? I know that Francoise is a subscriber to CGA, so I'm hoping he'll take this point up.

So, for this woman, who is 63, I find I'm now totally confused about what her expected parameters shoud be! One paper in the review (Judge J Geront. Med. Sci. 1996: 51A: M303-312) would put her normal velocity at 1.03 m/s, cadence 116 st/min (really???!) and stride length (by deduction) 1.065 m/s. So her percentage reductions come out to 36% for velocity, 26% for cadence and 14% for stride length. That would seem to contradict what I've just been saying about stride length being the most sensitive to pathology!

Now I'm really!


-- Dr. Chris Kirtley (Kwok Kei Chi) MD PhD Assistant Professor Department of Rehabilitation Sciences The Hong Kong Polytechnic University Hong Kong

PS: Just noticed that Jenni Dabelstein wrote...

>Her velocity is decreased but to the naked eye, her stride doesn't look too bad, so I suspect decreased cadence as the primary reason for decreased velocity. This makes me wonder again about primary balance issues.

So maybe her cadence is the problem, then. Why would that be? How would "primary balance issues" influence it?

Chris --

I guess I commented on cadence as function of balance because of my background in CP. Unlike normal children, children with CP frequently have difficulty altering stride length, most notably in the spastic population. Instead of lengthening stride, then cadence to go faster, they tend to instead use an increased cadence alone to increase speed. The use of increased cadence is a gait compensation to normalise their speed.

In a CP patient who had a reasonably functional stride length but a decreased cadence, I (as a PT) start to wonder about what is preventing them from compensating in this expected way. Most often it is a combination of factors, but often centred around "balance" activities.

In my head, "primary" balance issues in gait functionally centre around the child's ability to move themselves rapidly through space, righting their body as their COG moves around wildly (as it often does in CP). Multiple sensory systems are involved.

However, to use "normal" balance systems, normal range of motion and strength are also required. This is a basic premise of PT. In a patient whose neurological balance systems are relatively intact, strength or ROM defecits can result in very poor "balance"...I term this a "secondary" balance issue.

In normal patients, I would also add, it is difficult to generate a normal stride length when cadence is decreased, unless considerable voluntary effort is expended.

So, back to the lady at hand. The case history reported no ROM defecits. There probably was an abductor weakness related directly to the surgery, which needs to be addressed. But, was the weakness enough to explain the decreased velocity, which was more affected by decreased cadence than decreased stride?? I guess I thought not, which is a clinical judgement, of course.

The theory of "re-automisation of walking" seems to fit in with my thoughts. The lady in question has had a surgery and an insult to her body, with some resultant weakness. Her automatic balance systems have not fully adjusted to her body's new status....why can't she compensate for the weakness? Perhaps she needs some help to re-educate balance strategies? Perhaps the elderly need more help in movement re-education than a younger person whose systems are perhaps more open to change and can compensate better???

As a PT, if I didn't think that weakness or pain were enough to explain the clinical picture, I would go digging. I don't think the gait analysis gave me enough information...I want more!

I also want to find out about "normal" elderly values and am confused by the literature...does anyone out there have clinical experience to share??

My final question: do we know that the lady in the case study presented a "normal" picture pre-op?

Jenni Dabelstein

Thanks for your reply, which as you'll notice, I've forwarded to the list (mailto:[n/a]).

I like your reasoning, and there is some evidence to support it: her right and left stance durations are increased over the normal (62%) at 67 and 70% respectively, but her base of support (4.2 cm) is not widened - in fact its rather narrow. So I'm not sure about the balance hypothesis!

I'll ask her physician, Andy Kopf, to comment on her pre-op status.

Chris -- Dr. Chris Kirtley (Kwok Kei Chi) MD PhD Assistant Professor Department of Rehabilitation Sciences The Hong Kong Polytechnic University Hong Kong

Back to Case 4/8/97