Vol. 19 • Issue 26
• Page 32
sEMG Case Study
Last month, we began a case study of a 46-year-old self-employed contractor referred for an FCE following a motor vehicle accident from which he continued to report persistent symptoms and impairment four years post injury.
The SEMG data presented in last month's column showed the peak amplitudes of each of the eight electrode sites, the median frequency (MF) value (averaged) of each muscle site (by testing both before and after the lifting series using a standardized isometric test), and finally, a representation of the percent workload of each site at the lowest and highest weights to observe changes in motor plans as well as fatigue.
Depending on the body area injured, lifting tasks normally include a squat lift, a level lift and an overhead lift during the test. One of these lifts is normally repeated for additional comparison. Just the schedule of different lift tasks with six weight increments provide invaluable information not otherwise available, addressing numerous questions, a few of which are listed.
SEMG monitoring of lift tasks can provide help in answering several questions:
• Were adjustments in the weights consistent with the different lift positions and the physiological findings of muscle activation seen with SEMG?
• Was there a physiological change when the patient's movement pattern changed or when he reported a change in symptoms?
• If inhibition occurred in a prime mover, did the patient's movement pattern change or did he report any change, e.g., increased difficulty?
• Were patient's complaints regarding fatigue or difficulty continuing consistent with physiological findings of fatigue or delayed recovery from fatigue?
• Was the patient making good self-limiting decisions, i.e., asking to add less weight than protocol when there was good reason for such reduction? If inhibition increased, making the task more difficult to perform, did the patient stop?
These are just a few of the many questions that can be answered with the additional objectivity of SEMG, keeping in mind that no single parameter will provide the answer. The greater the number and complexity of multivariate analysis that is applied to the thousands of data points, the higher the discriminatory ability of the software to separate those individuals with normal movement from those with impairments.
Having used the MF value as a stand-alone value, and because this is one of the spectral frequency values that is often used as the sole determination of pre- and post-test outcomes, we will take the MF value one step further to show how it can break down and be very misleading. For simplicity of the various tasks performed by some of the muscle sites being monitored, we will focus only on the two levels of bilateral placement in the lumbar region.
The examination of MF values last month was limited to values taken immediately as the first test of the FCE. These were followed by the lift task that was shown with peak amplitudes at each weight of the squat lift, as well as a comparison of the percent workload recorded at each site from the lowest weight and highest weight of that lift series (15# and 55#).
Now we will examine the MF values taken before any other FCE tasks were performed and the last MF collection of the day, that was collected five hours later and included a 30-minute rest period to allow for recovery. Some 13 tasks, including three additional sets of lifts, were performed in the interim. While several FCE tasks remained, it has been found that the cables used at the time of this FCE interfered with those tasks. Muscle fatigue was not a concern during the remaining tasks, so the electrodes were removed.
The values representing the initial, rested MF values and those MF values after hours of varied physical tasks and required recovery time suggest to us that the lumbar extensor sites on the left side of the spine are performing much better than the right side. The starting values on the left side were similar and within an acceptable range, and it would appear that recovery of the left upper-lumbar area was excellent while the left-lower lumbar "overshot" its recovery a bit too far. A high-recovery MF value might result from the use of more faster-firing motor units as compared to rested, slower-firing motor units. One could suggest this was an indication of some delayed recovery of the slow-twitch motor units. In comparison, the right upper-lumbar site fails to show expected recovery and the right lower-lumbar starts too low, suggesting some initial inhibition as one likely explanation.
This would be a "neat and tight" explanation of the MF value changes, but let's look a bit deeper into what some of the other parameters tell us about these values. It would seem obvious that the MF values collected from the back extensor sites should reflect MF values from a power spectrum pattern characterizing extensor postural muscles, the use of a high ratio of slow-twitch motor units with a smaller representation of fast-twitch motor units. Rather than inspect the complexity of the entire power spectrum, let us first just compare the average firing rate of the motor units in the sample of each electrode site. We know that as a motor unit fatigues, its rate of firing slows down, and the MF slope, already discussed several months ago, should have a mild negative slope. So represents the average motor unit "zero crossing rate" (ZCR) from the first collection prior to starting the FCE, compared to the ZCR average from the last collection.
The data results provided for us challenged the "neat and tidy" assumptions we made from the MF values. First, each muscle has an expected profile of the relative ratio of slow- and fast-twitch fibers that comprise each muscle. These relative ratios are affected by high levels of highly specific exercise training; the more highly a muscle is trained, the greater the percent of slow-twitch fibers are expected to be present. This increase represents a pattern of "adaptation of use."
When a muscle such as the quadriceps group is confined by casting and bed rest for only a few days, the opposite pattern has already begun, where "adaptation of disuse" is degrading slow-twitch motor units to fast-twitch units. Thus, the relatively high ZCR values prior to the FCE tasks suggests that the latter adaptation has already had an effect on the lumbar extensor group as a whole. The only value that, as a stand-alone parameter, might look more positive is the right lower lumbar, but we just saw that the MF value of this same site was highly suggestive of significant inhibition.
The left lumbar sites, to which we felt some reflection of less injury (at least prior to any physical demands) now show high starting values for postural muscles. They do not have the expected negative slopes that would occur consistent with a healthy pattern of fatigue resulting from an expected profile of slow-twitch motor units. This leaves significant doubt as to whether any of these lumbar muscle sites are performing as expected.
Another way to examine muscle performance is not the average speed of motor-unit activity, but a parameter that gives us information about how much of the muscle site sampled by the electrode is actively being recruited for the task. Each motor unit within the sample being collected might represent collective motor-unit activity that provides information to the electrodes and its speed (the average of which we have already examined). Or we could consider each motor unit a penny and stack piles of pennies by virtue of the recorded speed when the signal reaches the electrodes.
The more motor units that are active, the higher our stack of pennies becomes, and we get the power spectrum, representing both its speed (frequency in Hertz or cycles per second) and location. By the pennies' location along the spectrum, we would have greater knowledge of how many muscle fibers might be active. If the total power was much lower than expected values, we would then have strong reason to assume the muscle was experiencing a high level of inhibition, and that the extent of inhibition would decrease as the total power increased. Inhibition is a continuum, not an "all-or-nothing" event.
When the total power of the four lumbar sites was extracted and once again compared from pre-FCE tasks-and after the required 30 minutes of relaxation after the completion of the first 13 tasks-the results negated all the positive assumptions we had started with when looking only at the MF data provided.
Prior to the FCE, the site that looked "most healthy" was the left lower-lumbar site, yet this site is totally eliminated due to inhibition that was just short of 100 percent. While the other sites had slightly higher levels of total power pre-FCE, by the end of the test the only site that was not at the edge of full inhibition was the right upper-lumbar site. This site was the one where the MF slope was so steep and negative that it could not possibly be considered functional for any length of time necessary for his job. These conclusions are very different from those that we would have gotten without any SEMG analysis, yet vary widely from those so commonly used without further checks and balances.
We have come full circle, starting from a position where we thought the solid evidence of the MF data would support the fatigue resiliency of at least some of the lumbar extensor group. But the more we examined other parameters that analyzed the same action a bit differently, we were left with the loss of support for our initial assumption.
In fact, these final conclusions (that were supported by numerous other parameters included in the full analysis) supported that the patient was not able to perform his job as he was at that time. The full FCE analysis was able to make specific recommendations that would allow the patient to enter a short rehabilitation program that would specifically address the physiological needs of muscle fatigue, failure to recover, inhibition and motor control adaptation involving secondary stress to other locations.
The patient's motivation made the success of such a program very high, even when he understood that the work would continue long after therapy was formally terminated. n
Barbara J. Headley has a clinical practice in Centennial, CO, and is active in lecturing after years of research with sEMG and studying the relationship between pain and movement dysfunction. Learn more about her products and courses by calling 303-993-8296 or visiting her Website at www.barbaraheadley.com