What is an EMG Evaluation?

The EMG evaluation is a standardized series of tests designed to assess the patient's pelvic muscle condition and neurological control, in order to pin-point deficiencies and provide a sound basis for a treatment plan.

---

The Old Kegel Perineometers Were not Accurate

Standardization of testing was a TOP priority in the development of perineometry. The previously published research using the Kegel Perineometers was difficult to understand, because each researcher came up with different ranges of numbers for "before" and "after" scores. The problem was that (1) the Kegel perineometers were NOT calibrated devices, and (2) the process of "eye-balling" of a fluctuating needle was not reliable. Different people used different strategies for interpreting what they saw in the rapidly moving gauge needle. As a result, it was impossible to accurately compare one person's research with another's, to arrive at general conclusions about pelvic muscle dysfunctions and rehabilitation.

Electronic signal processing has enabled us to standardize data collection so that we CAN compare readings between clinicians and begin to make scientific progress.

Contrary to the widely circulated but misleading advertising statements, the initial concerns in EMG evaluation were never simply the measurement of "strength" alone. Several factors are important in the evaluation, as described below.

Standard Evaluation Components

A "resting baseline" was always taken first, to assess the patient's absolute resting level, and to compare with a possibly elevated level after a few contractions. There is no such thing as a resting baseline with manometric devices, which cannot determine if the vaginal orifice is small because it IS small, or because tense muscles make it appear small.

The next two tests, "flicks" and "holds" were taken directly from the Kegel Perineometer research tradition, dating back to 1948.

A "flick" test was performed next, to measure the peak amplitude of contraction of the "phasic" muscle fibers. Usually an average of six short contraction peaks was calculated.

At least one commercially available program miscalculates the "flick" scores as "three second contractions" and takes the average of the entire three second interval. This is quite misleading, since an inevitable delay of up to one or more seconds in making the short contraction reduces the "phasic" measure to a short-term "tonic" measure.

The flick score in and of itself is of no real meaning; it does NOT predict urinary (or fecal) control, for example. It is useful to compare it with the "hold" score.

We also observe the resting levels between and after the "flicks" to see if there was any change, indicating a lack of control.

The Ten-Second Hold

A ten-second "hold" test then performed, consisting of six 10-second rests and 10-second holds; the rests and holds were each averaged together. Then, because an elevated resting score could give an equally elevated contraction score, we elected to subtract the resting level from the contraction level to give a "net strength" score. This is sometimes called a "tonic" muscle score.

It is important to point out that the "hold" score is the AVERAGE of each of the ten seconds of the contraction. It is NOT the maximum score obtained during the 10 second interval. We sometimes refer to this as a ten-second "window of opportunity" score. Because we include all ten seconds, any delay in recruitment of the muscle has the effect of lowering the 10-second average.

Thus the 10-second score is a "combination" measure which reflects BOTH strength and control, it has proved to be a very useful index of BOTH pelvic muscle condition AND neurological control.

Initially, in 1981, that was it. Then I became associated with Hollis Hermann, PT, and helped her with her master's thesis, and she suggested that we include a standard physical therapy measure of endurance -- the number of seconds that a contraction could be held at or above 50% of the maximum (phasic, or short-term) strength.

At least one software program misinterprets the endurance test as the time "above 3 microvolts". There is no scientific tradition or basis for such a test, so it is impossible to compare the the endurance test of any other manufacturer that does follow the industry standards.

The First Office Instrument - 1982

When the first Office Evaluation instrument, the "Clinical Perineometer" was produced by Farrall Instruments, Inc. of Nebraska, we included means for making these measurements semi-automatically. In addition, the C.P. included, quite intentionally, a small 2 inch strip-chart recorder that recorded all the muscle activity as pen deflections. It was intentionally designed to look like a basic EKG, in order to lend scientific credibility to the measurement of pelvic muscles.

The Clinical Perineometer was the same office instrument used by Diane Smith in the famous New Jersey Seniors project, reported in Bagais-Smith et al.

In 1984, after many dozens of Clinical Perineometers were in use, I wrote "Software Standards for Perineometry", in which the use of these standard EMG evaluation tests were released to the public domain and all manufacturers were urged to adopt them for the purpose of scientific sharing. For the next ten years, everyone did adopt them.

In 1985 the "Computerized Perineometer" program for Commodore C-64 computers was released. It was able to produce a one-page report of resting, flick, hold, and endurance scores for the standard evaluation. In 1987 the IBM version of my software was released, following the same standard. In 1988-89, Hollister/InCare came out with their PRS8900 Office System. In 1991 SRS came out with the "Orion-Perry Teacher" cartridge for the popular Orion 8600 computer system.

In 1993 Frost and Sullivan, a market research company, reported that PerryMeter-SRS had 69% of the biofeedback-for-incontinence market, with Hollister a distant second at 28%. All other companies shared the remaining 3% of the market. In other words, 97% of the products people used conformed to the software standards for EMG evaluations.

Historic Interest:

The Original "Handbook of Perineometry", which was the instruction manual for the Perry products of the 1980s and early 1990s, is available on the website in several chapters. The Table of Contents, with annotations, is at http://www.incontinet.com/hoptoc.htm. From there you can download individual chapters, as explained.

The Original Report is generated automatically by the Commodore and Orion versions of our software; it is a separate BASIC language program on the IBM-J&J-I-330 version. After conducting the standard EMG evaluation, the following is printed: (I suggest you PRINT this page now, so you can see it while reading the discussion that follows.) =====================================================================

PerryMeter Pelvic Muscle Diagnostic Assessment Report
Version 1.1 - Copyright 1987-92 by John D. Perry
NAME OF CLINIC OR HOSPITAL GOES HERE
Name of Responsible Professional Goes Here
Street Address, City and State Here

======================================================================

Resting Tension at start of evaluation was 4.4 microvolts
Resting Level after Short Contractions was 5.8
Change in Resting due to First Test was 1.4

TEST 1: MAXIMUM PHASIC MUSCLE STRENGTH (Short Contractions Test)

TEST 2: AVERAGE TONIC MUSCLE STRENGTH (10-Second Hold Test)

The Net Tonic Strength, 11.3, is the single most important score.

TEST 3: TIMED ENDURANCE TEST (Time Above 50% of Maximum Strength)

Examiner's Comments:

Evaluation Performed by: ____________________________

====================================================================

Comments: While the evaluation report allows computer-typed comment lines as shown above, I myself prefer to use hand-written comments, circling relevant scores and scribbling their interpretation in the margin. For example, an improvement in tonic strength might be hand-marked "100% improvement from last week -- That's Great!" as a form of encouragement to the patient.

Immediately after the evaluation, usually three photocopies are made; one for the referring physician, one for the insurance company, and one for the patient to take home and tape to the refrigerator to get family support and encouragement. (Also to point to and say "Mom has to do her exercises, kids, so YOU have to do the dishes!)

The report format was intended to make the relevant information instantly obvious to busy third parties, such as insurance claims reviewers who do not have even minutes to spend trying to decipher a tradtional biofeedback instrument data matrix output page. The latter often require that the user manually average together all the trials which start with "sht" (abreviation for "short") for example to derive an average flick score. The format was, in fact, developed out of extreme frustration in trying to make sense out of such "standardized, general purpose" reports as produced by the general-purpose program supplied by the instrument's manufacturer.

Another benefit of an easy-to-read and interpret report is more subtle. Yesterday my wife had a mammogram done, and she was able to bring with her all her past mammogram negatives. In just a few minutes, the radiologist was able to give a preliminary report: No Change, based on the present and past records. Now that EMG evaluations are approaching 15 years old, many patients (who fail to keep up with their exercises) will become RETURN patients, probably in another city far removed from their earlier home. The existence of a paper record of previous evaluations enables the clinician to do a far better job of interpreting present readings. I have already encountered patients who said they "did biofeedback" several years ago, but had no idea of the therapist's name, the instruments used, or the readings they obtained.

Chapter Two in The Handbook of Perineometry on this website -- <http://www.incontinet.com/hopchap2.htm> provides an excellent detailed discussion of the use of EMG perineometry and the information that can be gleaned from it. Since the Handbook was revised seven times in seven years, it is fairly well written. The following is a updated synopsis of that much longer discussion.

Diagnostic Interpretation. Pelvic Muscles, like all voluntary muscles in the body, should exhibit three characteristics: (1) They should contract on command, (2) they should relax when not contracted, and (3) they should respond quickly to commands. These three facts form the basis for interpretation of EMG findings.

Category 1: PC Muscle Weakness. Muscle atrophy or weakness is the most common finding, especially in urinary and fecal incontinence patients. The "net tonic strength" of the AVERAGE woman was found to be 9 microvolts [Ten-second long average, narrow bandpass, RMS]. The range has been 0.2 microvolts [a 72 year old woman with Parkinson's Disease and stress incontinence] to 45 microvolts [a 23 year old woman]. A healthy woman in the 30-50 age group should be able to register 12-15 microvolts; by age 75 a more likely healthy score would be 5-10 microvolts.

Computerized displays make it easy to assess tonic muscle strength. The ideal display should be a "square wave", that rises immediately on command, stays consistently high for another nine seconds, and drops rapidly on release. Among SI patients, however, a "saw tooth" display, with three or four separate peaks during the ten-second test, is a common observation. These represent separate "phasic" contractions, rather than a single "tonic" contraction.

Category 2: PC Muscle Tension. The original Kegel Perineometer operated on the basis of physical displacement, and was unable to differentiate between a "small" vaginal orifice and a "tense" one. EMG allows us to determine if the resting level is within normal limits. Normal resting levels are in the range of 1-2 microvolts (narrow, RMS). Persons who are skilled at systematic relaxation (e.g., yoga or TM) often register below 1.0 mV. Readings of 3-4 mV or more are generally associated with reports of PAIN as the presenting symptom. Pelvic tension often masquerades as frequent Urinary Tract Infections (without a culturable agent), or may actually cause frequent UTIs directly (by interference with the body's self-repair mechanisms.)

Type A: Situational Pelvic Tension. Some persons develop pelvic tension only in response to specific situations. Vaginismus (bracing in fear of hurt caused by penetration) is a most obvious example. One patient (age 75) tested perfectly normal -- except when thinking about her 95 year old mother, which caused her resting to increase to over 5 mV. Since "mother" lived in the patient's house, she was a constant source of irritation at home. One herpes patient got extremely tense when ever she was asked to visualize the face of the person who gave her the herpes infection. One patient presented with complaints of rectal pain whenever she had to walk more than a few minutes. It was finally discovered that she was contracting her anal sphincter every time tried to lift her legs to walk -- a kind of neurological short-circuit! She was perfectly "normal" during the standard evaluation. Therapists need to be especially creative sometimes to see if there are mental or physical "triggers" that will evoke tension. It is easy for inexperienced therapists to get "false negatives" because they haven't explored the tension-producing stimuli.

Type B: Chronic Pelvic Tension. Chronic tension often develops as a response to the irritation of scar tissue from prior surgeries in the pelvic region, from lower back and disk problems, and from postural imbalances effecting the pelvis itself. In an unknown percentage of cases it results from dysfunctional behavioral habits, such as holding back urination for excessive amounts of times. Nurses, for example, often complain that they "don't have time to pee" while on duty. Chronic pelvic tension is also associated with "sensory urge incontinence"; there is so much "noise" coming from the bladder area that the patient does not recognize the need to urinate until the signals get extremely urgent, and then there is little time to get to the toilet.

Category 3: Neuromuscular Control. We assume that pelvic muscles should exhibit the same level of control as other muscles, but that is often not the case, especially in older patients suffering stress incontinence. Many of them require several seconds (3-4) to effectively recruit a maximal contraction. If they get one second warning of a sneeze, and take 3 seconds to contract, leakage is likely to occur before the sphincter is tightly closed. Some of the newer pelvic muscle programs (such as the Orion and Orion/PC) actually record and print the "latency" to contract and the "delay" in letting go; any number greater than 1 second should be considered suspect. When the patient takes a long time (>1 sec) to recruit the PC, it is sometimes helpful to ask them to rapidly open and close their hand, to illustrate how quickly their other muscles respond. In resistant cases, an EMG sensor on the forearm can be used to display comparative data. (Ask the patient to make a fist every time she contracts the PC, and show the different recruitment times as two signals on the screen.)

Dysmenorrhea -- a special case. Painful menstrual cramping is just that -- cramping of the pelvic muscles in response to the minor pain associated with the monthly degeneration of the uterine lining. Patients who are evaluated with EMG during their worse cramps quickly see the cause of their pain (acute muscle tension), and can quickly learn how to eliminate at least 95% of their pain by learning to do pelvic muscle contractions during cramps.

Looking At The Numbers The first section of the printout (above) provides information about the initial resting level, the change after the short contractions, and (together with the "REST" column of the tonic test #2), information about any change during the "holds". In this case we see pathologically high initial resting that got worse by 1.4 mV with short contractions and stayed worse during the holds. However, notice the gradual decline from 6.7 before the first hold to 4.7 after the fifth trial. This improvement can be used to explain to the patient how 10-second holds may be helpful in reducing her pelvic tension. It is almost certain that this patient is experiencing pelvic pain, and should be asked about UTIs and retention as well.

Phasic muscle strength is reported in Test 1. In this case, it is extremely high and relative consistent. Often this is a sign that the patient has been doing Kegel exercises improperly; i.e., only making short contractions, with no attempt to hold for increasing times, as Kegel taught. Phasic strength has NOT been shown to be especially significant, except to help interpret the next test.

Tonic muscle strength is shown in Test 2. "11.3 microvolts" is the NET score, but in this case we see that it is reduced by an average resting level of 5.5 during the test. It seems likely that tonic strength is NOT a problem for this patient.

That is confirmed in part by the results of Test 3, where the patient was able to hold a single contraction above 50% of maximum strength for more than "30 seconds". The crucial range is greater or less than 10 seconds; usually stress incontinence and some urge patients will start at only 1 or 2 seconds duration; when they get past 10 seconds, they quickly reach 20 or 30 after just a few more days of at-home EMG practice.

The information provided by the computer printout is extremely precise and provides a snapshot view of the patient's pelvic muscle problems. It is so clear that Lesley and I often do "blind" interpretations at our workshops to show how much clinical information it yields. We ask a seminar participant to volunteer to be tested without providing us any background or history. After the EMG eval we tell them what we think is wrong. I won't say we have perfect predictability (especially with borderline results), but over the years we probably have been right more than nine times out of ten.

Other Examples Of EMG Evaluation Reports The PerryMeter (IBM) evaluation report (1987) was based on the very similar Commodore version (1985) and has been effectively duplicated in the Orion/Perry Teacher cartridge from SRS (1991) and the new Orion/PC Perry Protocol (1996). I understand (but have not been able to confirm) that the Hollister/InCare system provides a similar report [If anyone has such a report, please FAX it to me at 305-294-5115 so I can update that statement!]

On the other hand, a number of manufacturers of general purpose biofeedback instruments have attempted to cash in on the biofeedback for incontinence market by offering "incontinence" programs that are simply repackaged general biofeedback programs. Examples of this are J&J/Physiodata, shown below, and The Prometheus Group.

See if you can make sense out of this Physiodata printout. Calculate the Resting Levels, flicks, holds, and endurance scores, and make a note of how long those calculations took. Then ask yourself if the patient, or the patient's insurance reviewer, will be able to come up with the same results. [Alternatively, how long will it take to compose a letter conveying the necessary information, and how will you be reimbursed for your time in doing so?]

=============================================================== 
PELVIC FLO0R APPLICATI0N REP0RT 
Date: 09/1O/1993, Fri                  Protocol: Sitting Only 
Client/Patient: XXXXXXXXXXXXXXXXXX     Patient Position:Sitting 
Length of Session: 15 mm.              Single Session 
Row mode,All periods                   09/10/1993,Fri 10:18    
current timing: Seconds/Average 1.000 Averages/Trial- 20 

STATISTICS at trial level 
Al EMG Pelvic signal 
&/BASE    Mean Stan.Dev. Maximum Minimum Data P 
Trial# 1  0.56   0.01      0.58    0.55    5 
Trial# 2  0.79   0.12      0.92    0.61    5 
Trial# 3  0.75   0.18      0.96    0.54    5 
&/2SEC    Mean Stan.Dev. Maximum Minimum Data P 
Trial# 1  0.86   0.44      1.59    0.50    5 
Trial# 2  0.89   0.67      2.23    0.52    5 
Trial# 3  0.90   0.66      2.23    0.55    S 
Trial# 4  0.85   0.59      2.03    0.54    5 
Trial# 5  0.93   0.55      2.02    0.63    5 
Trial# 6  0.90   0.32      1.51    0.62    5 
Trial# 7  0.80   0.33      1.46    0.59    5 
Trial# 8  0.98   0.35      1.60    0.62    5 
Trial# 9  0.88   0.30      1.47    0.67    5 
Trial#10  0.99   0.23      1.32    0.71    5 
&/H1      Mean Stan.Dev. Maximum Minimum Data P 
Trial# 1  0.67   0.03      0.72    0.63    5 
Trial# 2  0.80   0.12      0.94    0.60    5 
Trial# 3  2.54   0.30      3.00    2.21    5 
Trial# 4  2.60   0.22      2.79    2.26    5 
Trial# 5  0.67   0.06      0.78    0.60    5 
&/H2      Mean Stan.Dev. Maximum Minimum Data P 
Trial# 1  0.65   0.05      0.74    0.59    5 
Trial# 2  0.74   0.10      0.83    0.57    5 
Trial# 3  2.64   0.31      3.18    2.24    5 
Trial# 4  2.16   0.40      2.51    1.38    5 
Trial# 5  0.61   0.03      0.67    0.59    5 
&/H3      Mean Stan.Dev. Maximum Minimum Data P 
Trial# 1  0.76   0.12      0.90    0.57    5 
Trial# 2  0.71   0.07      0.83    0.64    5 
Trial# 3  2.46   0.10      2.62    2.32    5 
Trial# 4  2.28   0.46      2.67    1.47    5 
Tria1# 5  0.65   0.05      0.75    0.58    5 
&/H4      Mean Stan.Dev. Maximum Minimum Data P 
Trial# 1  0.87   0.05      0.93    0.82    5 
Trial# 2  0.92   0.14      1.13    0.70    5 
Trial# 3  2.22   0.25      2.66    1.90    5 
Trial# 4  2.11   0.27      2.52    1.72    5 
Trial# 5  0.95   0.51      1.96    0.65    5 
&/H5      Mean Stan.Dev. Maximum Minimum Data P 
Trial# 1  0.62   0.03      0.66    0.60    5 
Trial# 2  0.78   0.06      0.87    0.70    5 
Trial# 3  2.47   0.33      2.77    1.88    5 
Trial# 4  2.19   0.29      2.55    1.80    5 
Trial# 5  0.69   0.08      0.82    0.61    5 
&/H6      Mean Stan.Dev. Maximum Minimum Data Pts 
Tr1al# 1  0.82   0.07      0.91    0.70    5 
Tria1# 2  0.82   0.08      0.91    0.69    5 
Trial# 3  2.43   0.66      3.39    1.50    5 
Trial# 4  1.49   0.37      2.19    1.12    5 
Trial# 5  0.69   0.05      0.77    0.62    5 
&/ENDUR   Mean Stan.Dev. Maximum Minimum Data Pts 
Trial# 1  2.86   0.41      3.49    2.25    5 
======================================================== 

Three final observations on EMG eval reports:

First, although some instruments default to two channels for measuring EMG levels in the abdominal muscles, there is no point in reporting such measurements in evaluation reports since there is no way to interpret the data. That's because EMG signals are attenuated by fatty deposits, and without measuring the fat content of the abdominal site, there is no way to know if the signal is really "high" or "low" compared with other people's abdominal muscles. Even if we could estimate the signal loss due to fat, there are no published norms for abdominals.

Second, and for the same reason, it is difficult to know what to make of readings taken through externally applied surface patch electrodes on the perineum or around the anus. Patients differ significantly in the amount of fatty tissue between the buttocks skin and the underlying muscles, and this makes such measurements highly questionable. [The only exception to this is in evaluating for fecal incontinence, where the external anal sphincter CAN be reasonable accessed with surface electodes near it.]

Third, a few programs have built-in options for making evaluation measurements in more than one posture. It is difficult to know why one would want to do that, since the information is highly redundant. Moreover, not a single published study has ever shown or even hinted that such redundant measurement provides any additional information of clinical significance.

Typically, the use of varying postures is reserved to middle and later stages of THERAPY, as part of teaching the patient to generalize to real-world situations. The standard posture for evaluation is "seated in a comfortable reclining chair". In any case, repeating the evaluation in three different postures adds up to a extra half-hour of time which could better be spent teaching correct exercise patterns. Finally, it should be obvious that EVERY biofeedback program COULD repeat the evaluation in multiple postures -- if anyone could think of a good reason for doing so. It's a gimmick.

--------------------

Copyright 1997 by John D. Perry

We Want to Know What You Think!