* I gratefully acknowledge Bill Buford, Bioengineer formerly at the Paul W. Brand Research Laboratory, Gillis W. Long Hansen’s Disease Center, Carville, Louisiana, for his help in reviewing the monofilament calculations in developing instrument measurements and collaborating on sensibility test design. Also acknowledged is the help of Lillian Brewder, M.Ed., for statistical consultation.The Semmes-Weinstein monofilament form of testing provides information that is not duplicated by any other test, adds insight into the physiology of the peripheral nerve system, and quantifies abnormality. When calibrated correctly, it is one of the few, if not the only, sensibility measurement instruments that approaches requirements for an objective test. This chapter is a summary of what is fact or fiction, what is known and not known regarding monofilament testing, and what is necessary in technique to produce an accurate measurement of sensibility.
The first edition of Rehabilitation of the Hand was intended to record technique for education in hand therapy as correlated with surgery. In this fifth edition, it is rewarding to be able to reference excellent studies that have been done with the monofilaments in various applications that advance our knowledge base for patient treatment.
Myths and Truths
Many fascinating studies of cutaneous sensibility have been executed since the age of Aristotle. † Each offers insight into the many facets of sensory function and what has and has not been attempted in its measurement. All articles list a bibliography of referenced information sources. By reading these references and the studies they in turn reference, one comes into direct contact with the original sources of information. It is often surprising to find that one’s personal interpretation of the original study is quite different from the interpretation of the study by another author. Serious researchers need to obtain original reports that often are more comprehensive and important to understanding than later reviews. A review of only the last 10 years can be misleading.
References .Neurophysiologists are still trying to isolate the components of “normal” sensibility, and much is yet unknown. Quite naturally, the neurophysiologist is interested in “normal” sensory function, and his or her test requirements are adjusted accordingly for normal detection thresholds. The clinician is interested in abnormal thresholds as they compare with normal and needs a test that can clearly define when a subject has dropped off from a normal baseline. * Abnormal thresholds are referenced against normal detection threshold values so that degrees of loss can be determined accurately and monitored for improvement or worsening with treatment.
* References .The monofilament test is simple to perform accurately. It provides an absolute numerical and pictorial record for serial measurements of the same patient, or across patients, for comparison with treatment. The monofilaments can be used in a 10-minute screen at predetermined sites, for limited mapping such as a specific digital nerve area, or for a full mapping to show clearly the relative relationship of an abnormal nerve versus other areas of response. The test’s versatility is one of its strengths.
The monofilament test, described in this chapter as used in the hand and upper extremity, has great value in providing similar screening and mapping all over the body. † Even on the face, where sensitivities far exceed that of the hand, areas of abnormality can be accurately mapped. Only in the plantar area of the foot, which often has degrees of callus, is some allowance needed in the interpretation of higher detection thresholds.
References .Semmes et al and Weinstein used the lightest monofilaments in the test for discrimination of detection areas with normal subjects. For test comparison within normal subjects, different scales of interpretation than those presented here (i.e., between men and women, left and right hands, age groups, and so on) are needed to address finite differences found within normal subjects. However, the intent of the monofilament test in evaluating clinical subjects is to identify abnormal subjects and when patients first fall off from a normal baseline (for both males and females). To identify abnormal subjects from those who fall “within normal limits,” one scale of interpretation has been determined based on known threshold detection values and patient “functional discrimination and recognition.” This interpretation scale, described in this chapter, is used for the hand, arm, shoulder, and rest of the body.
An instrument has to be shown to have instrument reliability before it can be considered valid and used in clinical studies. The repeatability of an instrument is related to its validity. If an instrument is shown not to be repeatable (i.e., in its strength of application), then its validity is nonexistent. One cannot use human performance alone to prove reliability of an instrument that measures in units defined by the National Institute of Standards and Technology (NIST).
The monofilaments clearly have been shown to be accurate and repeatable as an instrument if calibrated correctly, and they have the unique ability to dampen the vibration of the examiner’s hand that occurs with hand-applied devices that do not control for this vibration.
It is true that some closeness in force of application is produced by the monofilaments in the long kit and occasions may occur when forces overlap. This simply means that not all the monofilaments available for testing need to be used. For the sake of time, it is more expedient to use those monofilaments determined most critical for identification of “normal versus abnormal” and changes in “functional discrimination and recognition.” This is why certain monofilaments are specifically selected for the minikit, and in most cases, these are all that is needed for screening or full mapping of sensibility. The forces of application of the monofilaments in the minikit never overlap; thus the minikit is actually more sensitive as an instrument than the long kit. The long kit does include additional monofilaments needed for testing within normal subjects. In normative studies, inclusion of the lighter above-threshold monofilaments is critical for accurate measurement.
That the monofilaments have instrument reliability if calibrated correctly does not mean there is no room for improvement. The Weinstein Enhanced Sensory Test (WEST) was developed as an improved instrument. Following review and discussion regarding the minikit filaments and a prototype set that placed the minikit monofilaments in one holder, Weinstein, one of the original authors of the Semmes-Weinstein monofilament test, added nonslip coating and a rounded tip, less fragile handle and certified the monofilament force of application. Because the changes in the WEST monofilaments are potentially different enough in stimulus to affect threshold detection, exact comparisons cannot yet be drawn between the WEST and the original Semmes-Weinstein. Although the instruments are very close in stimulus, additional clinical testing in patients will direct whether interpretation scales need be adjusted for the WEST.
Other versions of the Semmes-Weinstein monofilaments have been designed. Most of these keep the original 90-degree angle of the filament to the rod and 38-mm length of the monofilament and include monofilaments of standard diameter. In particular, instrument versions in other handles have been made available for overseas testing in countries where cost is often a deciding factor in the availability of instruments and equipment. The application technique has been found to be important in ensuring accuracy of any form of the Semmes-Weinstein monofilament test.
When calibrated and applied correctly, the monofilaments are a valid test for testing sensibility detection thresholds. Studies have clearly demonstrated their ability to accurately detect the clinical condition intended. * Used in standard protocols, the monofilament test is being used to compare patient data in multicenter studies and is providing information regarding peripheral nerve changes with treatment not previously available with less sensitive and uncontrolled instruments.
* References .Normative studies were originally conducted by Semmes and Weinstein. Weinstein has published extensively on the monofilaments over a 50-year period. More recent normative studies have been made and are available for reference, but these only confirm the original normative thresholds defined by Weinstein with current instruments. Newer instruments need review and clinical trials as well as comparison with results from currently available touch-detection threshold instruments.
The force of application the monofilaments produce is related to their diameter size and length. Monofilaments, sized according to their diameter, if of the same length, will be repeatable in force of application within a very small range, usually within milligrams ( Fig. 13-1 ). By comparison, peripheral nerves that are clinically acute and changing increase or decrease in their threshold detection by several grams, and changes in nerve function and the improvement or worsening direction of change are detected and documented quite easily. Very small changes in force of application would not be expected to significantly affect threshold of detection. However, testing should be as controlled and repeatable as possible if results are to be considered reliable. Clinicians either need to request calibration information on test kits they use or measure the calibration, particularly for reports on patient studies.
Unfortunately, the readily available top-loading balances can be inaccurate for measurement of monofilament force of application because the scales depend on a spring mechanism for measurement. Because the monofilament is elastic (one of its desired physical properties), it is opposed by the spring mechanism, and the balance readout is variable. If a top-loading instrument such as a Metlar is all that is available for measurement, some estimate of its force of application can be made. However, these should not be used to adjust the application force of a given filament because the result may be to make changes that make the instrument out of standard with other instruments used for patient comparison. If measured on a top-loading balance-type scale, a monofilament should not be hand applied, but rather should be mounted in a test tube holder and brought down to where it bends in contact with the scale for a reading. The scale should be allowed to accommodate for the elasticity of the monofilament. This can be reasonably accomplished by waiting a few seconds and then taking the measurement at the same point in time for each measurement.
Accurate measurement of the dynamic force of the monofilaments requires instrumentation that can measure their force dynamics exactly. Special instrumentation was created for this purpose in earlier measurements. Bell and Buford developed measurement instrumentation sensitive enough to measure the dynamic force of application of the monofilaments ( Fig. 13-2 ). The signature of a monofilament repeatedly applied can be measured and examined for spikes in force, vibration of the tester’s hand, or subthreshold application. In more recent sensory instrument testing, researchers have developed an instrument measuring system using commercially available force transducers and a custom program to measure the force output (developed in Labview, National Instruments, Austin, Texas). Newer measurement systems include a direct reading into a computer program for analysis.
It is not uncommon to find that in the area of sensibility testing, authors and clinicians champion one or more methods of sensibility testing. Often, this is based on tradition, on what seems to be most popular, or on opinion that has been raised about a particular test. Sensory function is complicated, and depending on the question, one may need more than one test to obtain an adequate picture of abnormality.
One real criticism of most instruments for measuring sensibility is their lack of control on applied stimulus, which the monofilaments do have by virtue of their elastic property and ability to accommodate some of the normal and variable vibration of the examiner’s hand (which often exceeds the resolution for normal touch detection threshold). The elasticity of the monofilament and its bending at a specific force means the force it can apply is limited. Without this elasticity, any application by any probe instrument far exceeds the resolution of normal touch detection threshold and widely varies from a few milligrams in one application to over several grams or several hundred grams in another. This is because there is no limit on the force and no control on variables such as vibration.
Several computerized instruments that can reproduce part of the monofilament touch-threshold stimulus available range have been developed, and some of these are becoming commercially available. It should be recognized that computerized instruments have the same limitations as handheld instruments if their stimulus is hand applied. The best of the computerized instruments is designed to automatically apply and control the stimulus and has a built-in limit on how much force is applied. In instruments that still depend on a hand application of the stimulus, some improved control is possible with the accompanying direct feedback of measured force (or pressure) output, but in measurements in our laboratory, it was impossible for any instrument to have sufficient control on force of application if the testing probe was hand applied.
In addition, force measurement systems must be carefully examined for sensitivity. It is rare that one is accurate to less than 1 g. In fact, testing of strain gauges reported in literature to be accurate for less than 1 g found they often were not. The measurement system needs to be shown not to just average or electronically filter and smooth signals because averaging can hide peaks of higher force and the hand-applied vibration to which the patient would most certainly respond. It is common for computerized instruments to filter and average signals; some filtering of external “noise” such as the frequency produced by lights in a measurement area is even required. Instruments reported to be accurate need to have their forces and range of forces measured, and these measurements need to be available along with instrument specifications for any instrument sold for clinical testing of patients.
One of the biggest criticisms of the computerized instruments for threshold detection is that they have yet to produce, in one stimulus probe of a single diameter, the full range of application forces available in the Semmes-Weinstein monofilament set. Interestingly, proponents of computerized instruments argue that the computerized stimulus is better because it does not change for area. The area would change if the stimulus was designed to measure greater levels of sensory loss as needed to define “loss of protective sensation.” Computerized instruments are primarily designed to assess light touch thresholds and do not yet offer a test to determine “protective sensation” threshold. Certainly, when it comes to sensory abnormality and function discrimination or recognition, whether or not protective sensation is present is a defining factor needed for patient treatment because it determines if the patient is still relatively safe with sharp or hot objects, or conversely, in danger of wounds and amputations from complications of burns and injuries that could be incurred through use of everyday objects.
Undoubtedly, computerized instruments with sufficient control and range eventually will be available and add to our understanding of sensibility and abnormality. They also may help us define and adjust what is needed for an optimal touch-threshold test and have more application in research laboratories than in clinical settings. Once optimal stimulus and control are defined, the real challenge is to make this a readily available practical instrument at a reasonable cost so that it can be widely used for patient testing. To this end, the monofilaments have so far remained a valuable test with relative control and practicality and may eventually be found to be as accurate in needed sensitivity and accuracy and more desirable for practicality and portability than a more expensive computerized battery-dependent instrument.
Because of the large differences in sensibility testing instruments and techniques, practitioners wanting to develop an objective sensibility testing technique cannot escape the need to critique references and instruments for their merits and limitations. An instrument is not accurate or inaccurate just because it is said to be so; rigorous studies and measurements have to be conducted and published. Clinicians should treat as fact only what can be clearly proven as fact; all else should be treated as conjecture and opinion. In this way, advances in sensibility testing can be made that will help put to rest some of the myths perpetuated in testing and prevent a very good test like the Semmes-Weinstein monofilament test from not being used just because it is not recommended by some experienced examiners who champion another form of testing.
Most research that criticizes monofilaments does so without actual direct comparison to the monofilament test in terms of results. Studies that do compare another instrument to the monofilament testing are not always done in a protocol that would stand up to requirements for a fair comparison and also ignore unique aspects of the monofilament test. One must be wary of instrument studies that use only two examiners, particularly those that use the study to support their conclusion for another form of testing. It is easy to find agreement between two examiners, whereas the addition of a third examiner can adversely change the level of interrater reliability. To determine the relative control and validity of another instrument with the monofilaments, a comparison study requires a valid protocol with direct comparison instrument stimulus. Confirmation of results by independent examiners is preferable before judgments are made regarding the relative validity or usefulness of one test versus another.
Attempts have been made to divide clinical testing instruments into those that test slowly versus quickly adapting end organs, referencing the monofilaments as only testing slowly adapting fibers. However, in any given skin area of 10 mm 2 , there are close to 3000 sensory end organs. Given the stimulus variability in application force of most instruments used for testing sensibility it would seem impossible to selectively stimulate slowly versus quickly adapting fibers.
Although in a laboratory situation some end organs can be shown to be rapidly adapting in response to a stimulus and other end organs to be slowly adapting, such laboratory testing requires highly sophisticated instrumentation and measures end organ response under extreme control and artificial circumstances. Any clinical sensibility testing instrument would have to be much more controlled before it could possibly duplicate the testing of a controlled laboratory setting. In addition, neurophysiologists tell us that still other nerve fibers respond in between the quickly or slowly adapting and possibly act as moderators of the other two types. Still to be determined is their role in response to a stimulus and how all the nerve fibers are interpreted at the central cortex level; certainly instruments with variable application force cannot accomplish this.
The application force frequency of the monofilaments has been examined in a laboratory setting with and without strict control. The findings suggest that they cannot, even with the best of control, stimulate only one fiber population (slowly or rapidly adapting fibers). The monofilaments were applied to a strain gauge that could measure both the frequency of their application signal and their force of application ( Fig. 13-3 ). Frequency signals were detected throughout the available frequency spectrum at both low and high frequency. End organs are known to respond at certain defined levels of low or high frequencies. All instruments tested produced low and high frequency signals of sufficient strength to stimulate all end organs. It makes sense clinically that both rapidly and slowly adapting end organs are stimulated in the monofilament test, in that a patient will respond quickly with initial application of the stimulus. A slow enough monofilament application that does not appear to excite the rapidly adapting end organs is harder for a patient to perceive.
Thus arguments that the monofilament test should not be used because it stimulates only the slowly adapting end organs are without merit. However, regardless of whether the monofilament test measures slowly or rapidly adapting end organs, what is most important is that it is clinically reliable as an instrument and valid in identifying the abnormal condition it is intended to measure.
Force versus Pressure
Monofilaments have been reported in “force” values to understand relative differences in their weight (applied to the skin during application). Force is easier to understand in ordinal rank than pressure. Pressure is force divided by area. Because pressure values can change by changing the area measured, they can be misleading. Both pressure and force are units of measurement and do not change the stimulus applied, only the way it is reported. However, at least two areas of variability exist in accurate reporting of pressure measurements of the monofilament as applied to the skin:
When a monofilament is bent against the skin, the full area of the monofilament tip does not always come into full contact with the skin, but rather with a crescent-shaped edge. For an accurate calculation of the pressure applied with a filament of a particular force, one would have to calculate the area of the crescent-shaped edge that comes in contact with the skin and divide this into the force. This is not practical if even possible.
The area of application of a monofilament applied to the skin can and does change with clinical situations. Soft pliable skin may fully accommodate the whole area of the filament tip, whereas hard, inelastic skin that occurs with callus and scar may not. The area of a given sized monofilament is constant. It is no more accurate to report the monofilament application in pressure than it is to report it in force, and it is less accurate to report the stimulus in pressure because of the possibilities for inaccuracy induced by variable contact. The monofilaments are reported in force along with their area for each size monofilament. Their relative pressure can be calculated simply by dividing the force by the area.
The monofilament test is capable of measuring a range of responses, for example, from “within normal limits” to “diminished light touch,” to “diminished protective sensation” to “loss of protective sensation” and detection of any residual sensibility that may still be responsive to treatment.
Although the marking numbers of the monofilaments are related to a log scale of the force, they are used primarily as index numbers for ordering and for identification of filaments included in sets. Conversion tables specify the relative force for interpretation ( Fig. 13-4 , Table 13-1 ).
|Filament Markings *||Calculated Force (g)|
|Blue||Diminished light touch||3.22– 3.61||0.166–0.408|
|Purple||Diminished protective sensation||3.84– 4.31||0.697–2.06|
|Red||Loss of protective sensation||4.56–6.65||3.63–447|
* Minikit monofilaments are in bold. Descriptive levels based on other scales of interpretation and collapse of data from 200 patient tests.
Monofilaments are relatively inexpensive and can be made available to national and international treatment programs for standard measurement and comparison. However, one should keep in mind that realities of the business side of medicine may cause financial considerations to drive opinion more than the accuracy and practical value of the instrument. So long as clinicians continue to buy inaccurate instruments, they will continue to be sold and their accuracy potentially exaggerated. As the demand for monofilaments grows, more and more companies will make them available. Clinicians need to critically evaluate instruments they use, ask for measurement specifications, and return inaccurate instruments for replacement.
Nerve Conduction Velocity
Tests of nerve conduction velocity are recommended along with the Semmes-Weinstein monofilament test where available. Nerve conduction tests depend greatly on examiner technique and other variables. The test can vary according to the time of day, temperature of the extremity, size of the electrodes, placement of the electrodes, and individual testing instrument. Used with other tests of sensibility, the test can be of great help in determining the location of nerve injury and in numerically quantifying and documenting nerve problems. It is important to realize that the test in no way can determine what a patient does and does not feel. Attempts to directly correlate nerve conduction with diminished functional cutaneous sensation have been largely unsuccessful. Both nerve conduction tests and cutaneous sensibility tests appear to be needed to adequately clinically classify and monitor peripheral nerve function. Usually, the tests will correlate for evidence of involvement or noninvolvement of a nerve. However, exceptions are found. I have found cases in which nerve conduction was “absent” when some Semmes-Weinstein monofilaments could still be felt. In less frequent instances, cases have been seen in which the monofilaments have been “within normal limits” and a nerve shows a “slowed” nerve conduction response.
Advantages of Monofilament Testing
Advantages of monofilament testing include that the filaments bend when the peak-force threshold has been achieved and that a relatively consistent force is continued by the filaments until they are either removed from the skin contact or are severely curved. When they are severely curved, the force on the skin is less than the desired threshold. In addition to controlling the force of application, the filament design attempts to control the velocity of application. If applied too quickly, the filament force will exceed the desired threshold. Otherwise, the bending of the filament minimizes the vibration of the examiner’s hand.
A minikit set containing a normal threshold filament and the heaviest filament for each functional level was developed in 1977. This kit greatly reduces testing time. The instrument can be used for screening at selected sites or full mapping.
Color coding of the filament force produces a mapping that provides the examiner with differential thresholds of touch in areas of normal or relatively normal sensibility and areas of diminution. If the application technique is consistent, the mappings produced can be serially compared for changes in neural status. The mappings can be predictors of the rate of neural return or diminution. They can be predictors as well of the quality of neural return, or severity of diminution ( Figs. 13-5 to 13-7 and Plates 24 to 26 ). Attempts to correlate increasing or decreasing touch thresholds with levels of patient function appear superior to that associated with many other forms of testing. Used in combination with other clinical tests of sensory function, particularly a test of sensory nerve conduction, the monofilament test can lead to the resolution of patient problems not resolved by other forms of testing and can clarify other test results. The monofilaments are increasingly used by neurophysiologists in studies to determine end-organ response. Like the other tests of sensibility, they could be made more objective through careful consideration of their physical properties.