View Full Version : responses to EMG fatigue

08-12-1998, 03:44 PM
Thanks to all who responded to my question on measuring EMG fatigue
measures. The information was helpful in making the decision to do both 1)
an absolute submaximal test based on the first evaluation which will stay
constant over the subject's recovery to full strength and 2) a functional
submaximal test based on the subject's strength on that day.

Regards, Janice Eng

Here are the original questions and the responses:

We are interested in examining EMG fatigue measures (median power
frequency) during an isometric contraction of the quadricep muscles of
people who have severe muscle atrophy secondary to immobilization from a
fracture. We will be following these subjects from the point of cast
removal to full strength recovery. After surveying the literature we felt
that a submaximal contraction kept at the same absolute force level would
be best since the maximal force output will vary (likely increase) over
their recovery.

However, we are wondering:
QUESTION 1) By not using a maximal contraction, could this lead to
misinterpretation of the data since different motor units may be required
to reach the maximal force?


1) If you want to standardize the contractions you can define a functional
maximum contraction, for each subject, and use a submaximal level of that
contraction (i.e. 70% of functional max). That way all your subjects will
be at the same relative contraction level, even though they may differ in
absolute terms

2) The use of submaximal forces is normal. A maximal contraction can only be
sustained for a few seconds. You will have to think about the level of the
contraction - obviously the lower %Max the longer the contraction can be
maintained. There is a Rohmert scale which indicates the expected periods
for which contractions of various levels can be maintained.

3) I think you had better keep track of both absolute and relative
measures. I don't believe that both have been reported in concert and
would be very valuable. I would use an absolute load and record its
%max at each test along the way. The big problem as you mention is the
%max as strength increases --- if the load starts out as a modest
contaction, say 30% max, it will really drop off as strength increases.
Maybe the way around it is to take one relative and one absolute
measure. That way you could reasonably track a similar proportion of
motor unit types (using relative force) and monitor the shift to smaller
MU (as strength increases) with the absolute force (and expect a
decrease in MPF).

4) In this case I believe it may be useful to carry out different
'fatigue' level trials with different static loads, and try to achieve some
correlation between fatigue and isometric load. I suspect that the
relationship would be non-linear. Typical trials could be carried out for
loads between 35 - 85% of maximum isometric contraction. EMG normalised
response versus time and load should provide some interesting results.
Trials will have to be conducted over several days to get a true reading of
fatigue level per load. Also, the subject will get conditioned over time,
so this has to be taken into account.

A more efficient method may be to perform fatigue level tests (isometric)
with a sample of normal healthy subjects at various loads and attempt to
achieve a mathematical correlation between fatigue and load over time. Then
use these curves to compare fatigue levels of the injured subjects.

5)I did a masters thesis (Allen Akers) comparing the fatigue of the
quadriceps in
two different types of athletes and a control group. It was not a
longitudinal study, but having researched the topic and learned quite
a bit from the results I gathered, I would say that you would best be
served by using a maximal contraction. I would assume that you are
only concerned with changes within an individual, so if you utilize a
statistical design in which you can use the maximal force output as a
covariate, I think you will be introducing less error than if you
make assumptions about the fibers performing the contraction. I am
currently working on my PhD in advanced statistics and research
design and I'm quite confident that this is the best route to go. My
masters was in exercise physiology and I did three years of research
on EMG analysis (unfortunately, all on my own) so I'm also pretty
sure that you will be introducing a lot of error by using the
submaximal effort due to the changes in fiber type utilization during
the strengthening process.

6) Whether you do a maximal or submaximal contraction, in your patient
population you cannot be sure that they are recruiting the same motor
units each time. If you have them perform a standardized protocol, at
least you can claim that you are assessing changes in response to that
protocol over time.
In addition, you have to consider that each time you apply your
electrodes, they will not be in EXACTLY the same position, and therefore
your myoelectric signal parameters will be effected, and particularly so
if the limb geometry is changing (due to recovery from atrophy).
You should probably use a standardized electrode configuration and
protocol, but still normalize your data in terms of slope parameters,

QUESTION 2) Also in our case, by keeping the absolute force value of the
contraction the same over time, the relative percent of the fatiguing
contraction from
maximum (submaximal percentage) would actually be decreasing over the
recovery. Could this be a confounding variable when it comes to
interpreting the fatigue measures?

Over the period of sustained submaximal contraction, the
relative contraction level will actually increase since you will experience
a decrease in functional reserve. This is not a problem. You can terminate
the isometric contraction when the subject can no longer sustain the
original contraction level. At this point you know that the subjects have
reached their new (diminished) maximal capacity. This would be a good
standard way to terminate the contraction session. The recovery process
should not hinder the interpretation of data collected during the fatiguing
contractions (IMHO).

Yes, this will certainly be a problem. I would suggest keeping your
relative force level constant, but that would require determining a maximum
prior to each trial. I suspect this is something you would prefer not to do.

Thanks to Linda McLean , Ari V Sirin ,
Vaughan Kippers v.kippers@mailbox.uq.edu.au, Peter Keir ,
Rene Ferdinands , and Allen Akers for
their helpful responses.

__________________________________________________ ____
Janice Eng, PhD, PT
Assistant Professor
School of Rehabilitation Sciences, Faculty of Medicine
University of British Columbia
T325-2211 Wesbrook Mall, Vancouver, BC, Canada, V6T 2B5
Tel: (604)714-4105 Fax: (604)822-7624

To unsubscribe send UNSUBSCRIBE BIOMCH-L to LISTSERV@nic.surfnet.nl
For information and archives: http://www.bme.ccf.org/isb/biomch-l