Measurement of Socket Discomfort—Part I: Pressure Sensation

Abstract

The purpose of the study was to determine the feasibility of quantifying pressure-related socket discomfort by using psychophysical methods. The first experiment involved the application of known uniform pressure to the residual limbs of three transtibial amputees. Category ratio scales were used to measure sensations of pressure, pain, and perceived discomfort. The second experiment involved altering the geometry at the fibula head and distal tibia of the subjects’ fitted PTB sockets with wafers of varying thickness and use of the same three scales to measure response to the variations in local point-specific pressures. Two of the subjects had more than 32 years of experience with PTB sockets, whereas the third subject had just been fitted with a temporary prosthesis. Results indicated that response functions for global uniform pressure varied among the subjects but were similar for the modifications at specific locations. Both experienced amputees reported constant pressure sensations between 10 and 40 mm Hg of applied uniform global pressure, whereas the inexperienced subject reported a linear increase in pressure sensation over the same range. Pressure and discomfort sensations appeared to be isolated subjectively and assigned different magnitudes among subjects, creating unique response functions for each subject. Possible explanations for unique uniform global pressure response functions could involve years of experience using a PTB socket, residual limb tissue thickness, and morphology or the way in which the amputation surgery was performed. For pressure variations at the fibula head and distal tibia, pressure sensation and perceived discomfort were highly correlated and appeared not to be distinct sensations. Response functions were similar among subjects. Psychophysical measurement could be adapted for use in a clinical setting during the fitting process and offer researchers additional tools for evaluating alternative socket designs and materials. The methods permit the patient to communicate discomfort sensations more precisely and provide the prosthetist with additional insight into the patient’s perception of quality of socket fit.

An extensive methodology exists that can be used to measure the subjective magnitudes of the phenomena the body senses when forces and energies of various types act upon it. Psychophysical measurements of the sensations produced by sound and light form the basis for acoustic and lighting design. ¹ Psychophysical scales also are being used to measure pain in clinic and hospital settings. ² The research question examined in this study is whether the pressure sensations produced in the residual limb by a prosthesis can be measured using psychophysical methods. If so, the data produced by such methods possibly could be used to improve socket design and fitting.

The neural functioning of human visual and auditory systems make it possible to quickly and accurately extract and interpret vast amounts of information from the environment. The skin and muscle tissues, in contrast, are not designed to produce nearly as much information about the environment as the eyes and ears, and the information they do produce plays a much smaller role in cognitive tasks. In comparison with the eyes and ears, the sensory organs of the skin and muscles are distributed at a much lower density, and the data they produce do not result in nearly as rich or detailed a model of the environment. Also, many of the sensory organs of the skin and muscles that generate data on the sensations produced by a socket provide information about the internal state of the body rather than the external environment.

Although various studies have examined the biomechanical properties of tissues and the engineering properties of different liner and socket materials, research on the psychophysics of socket discomfort has been largely neglected. Yet, the patient’s perceptions are the final determiners of comfort. The relevant information is highly subjective in nature. Neurological mechanisms for producing sensations of pain and discomfort have evolved to protect and preserve organisms, and an extensive body of scientific literature exists on the physiology of these somatosensory neural systems.

Two important goals of good socket fit are the avoidance of tissue breakdown and the minimization of discomfort. For sensate residual limbs, these are related goals because tissue breakdown frequently is associated with pain and discomfort, and pain is frequently, but not always, an indication of tissue breakdown. During the fitting process, verbal responses from the patient are often used to identify locations of excess pressure and discomfort in the socket. Typically these responses are fuzzy and lacking any kind of scientific precision, which makes it difficult for the patient to describe accurately sensations being experienced. This, in turn, can make it difficult for the prosthetist to determine how much adjustment to the socket may be necessary to provide significant relief to the patient.

PHYSIOLOGY OF PRESSURE AND DISCOMFORT SENSATION

The basic sensations produced by the neuroreceptors of the limbs include pain, touch (pressure and vibration), temperature, and possibly muscle and tendon tension. Though the term “discomfort” is used frequently to describe socket fit problems, discomfort is not one of the basic sensations. This study proposes that a perception of discomfort involves first an awareness of one or more of the basic sensations, followed by the assignment of a negative valence to the sensation. Factors that could lead to the assignment of a negative valence might include, among others, the extent to which the attentional resources of the individual are controlled by and directed toward the sensation, thereby diverting the attentional resources from desired cognitive tasks, or the expectation that pain will occur if the stimulus causing the sensation is not altered or eliminated. Sensed pain will nearly always be associated with discomfort, but discomfort need not always be associated with pain.

Sources of pain and discomfort include socket related pressure and shear, bone spurs, peripheral neuropathic pain, phenomena related to cortical reorganization after amputation, heat and moisture in the socket, and skin irritation from sources other than forces. ^3,4 Stump pain and discomfort are complex phenomena, and it is not always possible to eliminate them through socket design, suspension method, or quality of fit. Edema occurs frequently for many amputees, and with a swelling of the residual limb, socket interface forces will change, increasing in some areas of the socket.

In the clinic, trauma caused by pressure can be classified by observation into one of five stages, which imply that damage begins at the surface of the skin and proceeds inward. ⁵ However, several studies have concluded that pressure-caused trauma may begin in the muscle close to bone and proceed outward. ^6,7 Muscle appears to be more susceptible to pressure than skin and subcutaneous tissue because of its greater metabolic activity and need for temperature control. ⁸ Studies also have shown that the occurrence of blisters is related to the amount of pressure applied, the coefficient of friction, the thickness of the stratum corneum, and the number of rubbing cycles. ^9–11 Average capillary blood pressure ranges from 12 mm Hg on the venous side to 32 mm Hg on the arterial side of the limb. ¹² When blood flow is diminished through the capillaries, nourishment decreases and the byproducts of metabolism accumulate. Pressure-duration relationships have been developed for pressure sores, but most involve continuous pressure. ^13,14 There are fewer studies of repeated pressure loading of the type encountered inside a socket during gait. One study found two distinct responses to pressure; under conditions of 30 mm Hg and 10 to 15 minutes of loading, normal subjects exhibited complete recovery via rapid reactive hyperemia, but debilitated subjects revealed an impaired recovery control mechanism. ¹⁵

The residual limb typically is covered with hairy, nonglabrous skin. However, the mechanoreceptors of hairy skin have received much less study than those of the glabrous skin of the palms and finger tips, and their function is not as well understood. In glabrous skin, the distinctions among the four types of receptors (FAI, FAII, SAI, and SAII) is relatively clear. ¹⁶ In hairy skin, it appears that three channels exist; the Ph (Pacinian hairy—responds to 45- to 500-Hz vibrations), NPh low (non-Pacinian hairy—responds to 0.4- to 4-Hz vibration), and NPh mid (non-Pacinian hairy—responds to 4- to 45-Hz vibration for large areas of stimulation and 4- to 150-Hz vibration for small areas). ¹⁷ Pacinian corpuscles inhabit the deep tissue around major blood vessels and muscles and at lower densities than in glabrous skin. The NPh low receptors appear to lie deep in the skin. They produce sensations of pressure, respond to maintained pressure and to stretch, are affected by static indentation and produce resting discharges. They have large receptive fields and correspond to SAII receptors of glabrous skin. The NPh mid receptors appear to lie more superficially, and their stimulation is perceived as flutter. Hair follicle receptors, which sense hair movement, appear to involve FA fibers.

Many pressure-sensitive receptors fire rapidly as force is applied and then slow down or cease to fire when the pressure is maintained at a constant level. These mechanoreceptors convey information only when the skin is undergoing indentation. This characteristic of mechanoreceptors helps prevent overloading the cognitive functions of the brain. Thus, individuals are able to sit for long periods without being continually made aware of the pressure on their gluteal muscles, and clothes can be worn without a person being conscious of their weight.

Spatial acuity is a measure of the ability of the central nervous system (CNS) to discriminate among patterns of stimuli applied to the skin. Lateral inhibition amplifies patterns of sensation. On the extremities, spatial acuity increases from proximal to distal. Both Stevens and Choo ¹⁸ and Weinstein ¹⁹ found that the finger tips have a spatial acuity approximately 20 times greater than the lateral skin of the upper forearm, and the dorsal surface of the hallux has a spatial acuity roughly 8 times greater than the calf and 4 times greater than the anterior thigh. However, spatial acuity declines with age. Decreases between ages 23 and 74 average 50% on the fingertip and approximately 75% on the hallux. The decline is hypothesized to be due to a thinning of mechanoreceptors in the skin, though CNS changes and changes in the physical properties of the skin have not been ruled out as factors. Stevens and Choo conjecture that loss of spatial acuity in the foot may lead to balance problems in the elderly and in diabetics who experience neuropathy.

Studies by Teuber and Weinstein found that spatial acuity of the distal residual limb was greater than that of the sound contralateral limb at the same location and approached that of the missing extremity. ^20,21 It was hypothesized that regions of the somatosensory cortex that were innervated by the missing limb became responsive to neighboring neurons, which in the case of the lower limb, represented portions of the residual limb adjacent to the amputation. Numerous researchers have studied the possible neural mechanisms that would enable this learning to occur. ^22–27

Nociceptors respond in a distinctive manner to noxious stimuli. ²⁸ Many nociceptors have thresholds toward the upper end of the harmless range of stimuli, but most encode only in the noxious range. Mechanical nociceptors consisting of large myelinated fibers (A-fibers) and having large receptive fields respond to overtly noxious pressure. On the hairy skin of the limbs, each receptive field consists of many distinct points. High-threshold mechanoreceptors (HTM) respond to strong mechanical stimulation. C-fibers, which are unmyelinated, typically respond to heat. Receptive fields for C-fibers have clear boundaries and consist of single zones, unlike the multi-point A-fibers. Innervation densities are high. Polymodal nociceptors, which fall into the C-fiber category, respond to strong pressure, heating, and some irritant chemicals. Irritant chemicals include histamine and bradykinin, which are released when tissues become damaged.

The excitation of only small numbers of nociceptors, and possibly just a single nociceptor, can produce pain sensation in normal subjects. The quality, or character, of the pain may also vary among fiber types (A versus C) and by skin type. C-fibers are associated with burning pain or itch on hairy skin, and A-fibers are associated with pricking pain. Temporal summation may be involved in pain perception. With repetitive and long-duration pressure stimuli, pain increases although nociceptor activity does not. CNS mechanisms may thus enhance or prolong pain perception. Tissue inflammation typically results in primary hyperalgesia or an increase in pain sensitivity. This sensitization appears to be mediated by chemicals released after tissue injury, including histamine and bradykinin, among others. After nerve injury, nociceptors may show increased sensitivity to catecholamines, and many regenerating nociceptors become sensitive to catecholamines. Tissue inflammation also may lead to catecholamine sensitivity.

Mechanoreceptor and nociceptor nerve impulses are generated and transmitted to higher levels in the peripheral nervous system (PNS) and then to the CNS, where the impulses are perceived as sensations. There are three pathways. The first major pathway is the dorsal column pathway, which carries primarily touch and proprioceptive impulses, and conducts potentials along large myelinated nerves rapidly up the spinal cord to the brain stem, where the nerve fibers cross to the other side of the brain and thence to the thalamus and somatosensory cortex. ²⁹ The second major pathway is the spinothalamic, which primarily carries sensations related to pain and noxious stimulation, and some information about temperature and touch. It consists of two distinct pathways. For both the dorsal column and spinothalamic pathways, processing of impulses begins in the dorsal horns of the spinal cord. The spinothalamic pathway conducts impulses more slowly than the dorsal column pathway and is comprised of smaller diameter fibers. It ascends the spinal cord on the opposite side from the peripheral receptors and divides into two branches at the brain stem where the evolutionary older paleospinothalamic pathway, which appears to be specialized for signaling dull or burning pain, receives input from small, unmyelinated, slow-conducting nerve fibers and projects to the thalamus and thence to the somatosensory cortex. The second branch, the more recent neospinothalamic pathway, appears specialized for sharp or pricking pain and carries impulses from small, myelinated, slow-conducting nerve fibers that terminate in the skin. It also projects to the thalamus, and thence to the somatosensory cortex and the association cortex; in the former pain is felt as sensation, and in the latter pain is processed cognitively.

Both of these spinothalamic pathways assist the individual to determine the nature of a traumatic event, its location, its severity, and its duration. ³⁰ However, the spinothalamic pathway performs less well for detecting trauma in deep tissues and visceral structures. Cutaneous pain tends to be perceived as sharp, pricking, or burning and is well localized; the quality of deep tissue pain tends to be more diffuse and dull, and may be described by some as “sickening.”

The third major pathway is the spinoreticular, which appears to carry out the processing of affective or emotional aspects of pain. The neurons comprising this pathway enter the reticular formation and then project to both the thalamus and the dorsal and ventral noradrenergic bundles (DNB and VNB). The DNB plays a very significant role in the activation and regulation of attentional resources, which in turn can generate states of vigilance and facilitate responses to threats such as tissue damage. It is this pathway that increases alertness to threat and draws information processing resources away from activities that may be competing for cognitive processing. The VNB innervates the hypothalamus and plays a role in stress response, which may contribute to the emergence of pathological pain, as well as helping to ameliorate pain of limited duration. Glucocorticoids released during stress response may reduce inflammation and block the sensitization of nociceptors. From the DNB and VNB projections enter the limbic frontal cortex, where an emotional dimension is added to perception of pain. Because the limbic system is responsible for memory and emotion, pain stimuli become enmeshed in emotional response and develop an emotional component.

Although pain serves a protective function, the neural systems involved in the perception of pain function crudely. Performance is better for damage to cutaneous tissue than deep tissue. The location of deep tissue damage is often misperceived, and the magnitude of perceived pain may bear little relationship to the severity of tissue trauma. Further, nociceptors adjust their threshold level as a function of their chemical environment. From a clinical perspective, the emotional intensity (affective magnitude) of pain is of great importance because it represents the extent to which a patient consciously perceives a tissue-traumatizing event to be occurring and it determines distress.

Aging has profound effects on the skin, which can affect awareness of tissue traumatization. As mentioned previously, spatial acuity decreases. The presence of diabetes or renal disease, among other pathologies, also may influence skin aging. In general, the organizational characteristics of young skin become more disordered with age, producing a wide range of potential problems for the prosthetist. Cell reproduction and repair slow down. As the skin ages, the dermal papillae of the papillary layer diminish in height until their surface is nearly flat. This reduces the area of contact between the dermis and epidermis and makes the boundary layer more susceptible to separation under shearing loads, and the peeling back of the epidermis occurs under lower levels of shear. ³¹ The dermis also thins with increasing age, offering less protection from trauma for underlying tissue. There is less collagen per unit area and the collagen is less dense. Collagen provides tensile strength and prevents the skin from being damaged by overstretching. It also provides the structural framework to support the vessels and nerves embedded in the dermis. With aging, the organization of the collagen fibers becomes increasingly chaotic, with the result that the skin becomes leathery and tough. It is believed that additional cross-linking occurs and the fibrous network becomes more random and disarrayed. The collagen becomes less water soluble and stiffer with age. Though the stiffening of the collagen should make skin more resistant to stretching, just the opposite occurs—the skin becomes loose and wrinkled. However, this may be due to changes in the elastin fibers, which also exhibit disorganization, so the skin does not recoil as quickly when stretched. The elastin fibers become thicker and more branched. The ground substance may also change with age, providing less lubrication for the fibers when they move.

The number of active sweat glands decreases, and the skin is less able to remove excess body heat. As a result, the skin feels drier, but the elderly individual is more at risk of heat stroke during periods of extremely hot weather. Interestingly, sebaceous oil glands found throughout the skin increase in size but decrease oil output. Although aged skin can repair as well as young skin, repair may take significantly longer. The number of vessels and capillaries declines with aging, and the capillary loops disappear. As a result, less nourishment becomes available to sustain the cells of the dermis and epidermis and to repair damage. One consequence of this is that blistering may occur rarely or may take much longer to occur in aged skin. Because extracellular fluids cannot move as quickly to the site of shear-related trauma in aged skin, blisters may not form. Instead, skin damage may progress directly to loss of the epidermis. This may be accelerated by the decline in skin elasticity mentioned above. Because of the reduced vascularity, absorption of substances through the skin may be reduced. Although permeability of the skin does not appear to change, clearance time in the dermis may increase.

Also, aged skin is more likely to feel cool, and the elderly are more likely to feel cold, or even experience reduced pain perception. Meissner’s corpuscles decrease continuously with age. The number of dermal cells decreases, and the remaining fibroblasts have a reduced cytoplasm volume and decreased rate of metabolism. This may be accompanied by the loss of subcutaneous fat and an increased risk of trauma, as well as increased susceptibility to pressure-related tissue damage.

With aging, acute inflammatory reactions are muted and reduced. Thus, reaction to noxious stimuli may be delayed, and a long latency period may develop, followed by tissue collapse and ulceration. Cell immunity also declines with age. Although aged skin appears to show decreased hypersensitivity to substances, this may be linked to reduced immunity to diseases such as skin cancer. Dermatitis tends to become chronic and widespread and may not respond quickly to treatment. Healing may be delayed and unpredictable. Skin problems in diabetics also are related to reduced vascularity, and diabetics are prone to the problems discussed above.

PSYCHOPHYSICAL MEAUREMENT OF PRESSURE AND DISCOMFORT

The two-stage theory of sensory scaling states that in the first stage, a stimulus creates a sensation according to a psychophysical law based on neurophysiological processes, and in the second stage the sensation produces a subjective response according to a sensory response law. ³² In the context of socket fit, when pressure is applied to a residual limb, pressure-sensing neural cells transmit information to higher processing centers in the CNS (the first stage, or psychophysical response). The higher processing centers make the individual aware of the sensation and assign to it a subjective magnitude (the second stage, or sensory response). During the typical psychophysics laboratory experiment, only the stimulus and response can be measured. The intermediate steps, where the CNS processes the information received from the pressure-sensing cells, are not observable, though state-of-the-art imaging techniques are increasing the feasibility of viewing CNS functioning.

The least powerful scale used in psychophysical modeling is the categorical scale, which consists simply of categories, which may or may not lie along an underlying psychological continuum. If there is an underlying continuum, then responses can be arranged in order of magnitude, and an ordinal scale developed. The ordinal scale has no true origin (zero point) and the psychological distances between categories on the continuum cannot be established. Thus, the mathematical manipulation of category ratings is limited.

The next level of scale is the interval scale, which has equal intervals. Though it lacks a true origin, the distances between responses along the scale can be obtained by addition and subtraction, and the intervals of different scales can be compared. With an interval scale, the use of standard parametric statistics can be rigorously defended. The highest level of scale is the ratio, which has all the features of the interval scale but in addition has a true origin or zero point. With the existence of a zero point on the scale, ratios of scale values can be calculated, and different scales can be compared by means of a ratio. The responses of independent stimuli also can be added together. Order, interval, and origin relationships are preserved. Because of this, virtually any appropriate mathematical model can be used to establish the relationship between the magnitude of a physical stimulus and the magnitude of the subjective perception of that stimulus. Since Stevens ³³ first proposed the power law, many studies have found that the relationship between the magnitude of an applied stimulus and the subjective magnitude of the sensation it causes can be described best by a power function. The values of the exponent have been found to vary with the sensory modality and the stimulus conditions.

One problem associated with pure ratio scaling is that it provides little insight into the subjective significance of any given magnitude. If one is interested in learning how much discomfort is associated with a given magnitude, and further wishes to make interindividual comparisons among observers, then verbal descriptors need to be attached to the scale. This has led to the development of a combined category ratio scale, which attaches categorical qualities to a ratio scale. The upper endpoint of the scale is anchored by the word “maximal.” With this scale, it is possible to compare two individuals who may have different levels of tolerance to the stimuli. Borg, who has developed a category ratio scale used in ergonomics and pain studies, claims that the scale is applicable to all sensory modes because the psychological continuum between minimal and maximal sensation covers the same range for all modalities across all individuals, even though the objective range of the stimuli for different modalities may be quite different and the absolute value of the maximum tolerable stimulus may vary from individual to individual. ³⁴

Variants of the category ratio scale have been developed to assess pain intensity in a clinical setting. ^35–37 The features of all the scales that qualify them as a variant of Borg’s category ratio scale are the descriptors used to anchor the end points. Their origins are labeled with “no pain,” and the upper ends are labeled with “pain as bad as it could be” or alternative wording that indicates the maximum pain imaginable. However, none of the scales contain intermediate verbal descriptors. Although use of the VAS is widespread in hospital and clinic settings, there is some evidence that it may not be as sensitive as the verbal descriptors for separating pain intensity from pain unpleasantness. ³⁸

EXPERIMENTAL METHODS

Three subjects with unilateral transtibial amputations who had been fitted with PTB sockets and Pelite liners were selected from the patient clientele at the residency site and asked to participate in the experiment. Psychophysical scales were used to measure response to global pressure on the residual limb in terms of perceived pressure, perceived discomfort, and perceived pain. The same scales then were used to measure the ability of the subjects to detect and evaluate variations in socket geometry and pressure. All subjects were volunteers, and each read and signed a letter of informed consent that had been approved along with the human subject protocol by the Biomedical Sciences Committee of the Institutional Review Board of the University of Nevada, Las Vegas. Two of the patients had sustained amputations as the result of trauma and had been prosthesis users for 57 and 32 years, respectively. The third subject had experienced an amputation approximately 7 months earlier and was being fitted for a temporary prosthesis for the first time; data were collected before the actual delivery of the prosthesis. Additional information on the subjects is presented in Table 1.

Table 1: Subject characteristics.

Touch thresholds were measured on both the residual limb and contralateral limb by using calibrated Semmes-Weinstein monofilaments and the rapid threshold procedure (RPT). ^39,40 Monofilament strengths were 0.02, 0.04, 0.08, 0.16, 0.32, 0.64, 1.2, 2.4, 4.8, 9.6, 20, 40, 80, and 160 g. Spatial acuity was measured on both the residual limb and contralateral limb so that comparisons could be made. Measurements were taken using the same type of testing equipment and procedures as Stevens and Choo ¹⁸ in their landmark study of spatial acuity of the body surface. Stimulators featuring a gap on one side and a solid edge on the other side were applied to the skin, and subjects asked to indicate whether the solid edge or the edge with the gap had been applied. Gap widths were 5, 7.5, 10, 12.5, 15, 20, 30, 40, 50, 60, 70, 80, 90, and 100 mm. All subjects exhibited normal touch thresholds and spatial acuity on their residual limbs. Data were collected over two sessions lasting approximately 1.5 hours each.

To examine psychophysical response functions, the residual limb of each subject was enveloped in an Iceross Icecast inflatable pressure bladder to a level just proximal to the patella. The bladder was inflated in a random sequence to pressures of 10, 20, 30, and 40 mm Hg during each of three sequential runs so that each pressure was applied three times. Each inflation pressure was held for 15 seconds before requesting judgements of perceived pressure, discomfort, and pain. After responses were provided, air was released from the bladder, which was allowed to remain in a deflated state for 15 seconds to permit blood to perfuse back into the tissues before reinflation. Subjective perceptions of pressure, discomfort, and pain were elicited using the category-ratio scale developed by Borg. The scales were printed in a large font on 8 1/2- by 11-inch sheets of paper and plasticized so that they could be held by the subjects. Subjects were asked to study the scale before the experiment began and to provide numbers that corresponded to the amount of stimulus being experienced. The scales are shown in Figure 1.

Figure 1: Pressure, discomfort, and pain category ratio scales.

To measure sensations related to changes in socket geometry at specific locations, a check socket was fabricated by pouring plaster into the Pelite liners of the two experienced subjects’ current fitted prostheses and bubble forming Duraplex over the positive model, thus duplicating the geometry of the sockets for the prostheses that the subjects were wearing. The third subject was in the process of being fitted for a first, temporary prosthesis, and thus had no prior experience with use of a prosthesis. For this subject, the check socket being used to fabricate the laminated socket and representing the end point of initial fitting was duplicated. The number of ply of sock that the subject was currently wearing were pulled over the residual limb before donning the socket, to reproduce as nearly as possible the geometries and pressures being experienced by the subject inside the socket of the prosthesis. The two subjects who had been long-time wearers were asked to verify that the fit was similar to their definitive prostheses, and minor adjustments were made if needed.

Flexible copolymer wafers approximately 1 mm thick and 23 mm in diameter were taped on the inside of the check socket at the distal tibia and the apex of the fibula head to vary socket geometry and pressure against the residual limb. The experiment consisted of taping 0, 1, or 2 wafers at each location randomly such that each of the three possible states (0, 1, or 2 wafer thicknesses) was presented three times, for a total of nine combinations of socket alteration (3 variations at the fibula head × 3 variations at the distal tibia). A low-cost socket pressure sensor was utilized to measure relative pressure between the wafers and residual limb. The pressure-sensing element was placed on top of the wafers, or in the cases where there was no wafer, directly on the socket. [During the experiment, it was found that the pressure measurements taken inside the socket with the low cost sensor were not reliable. Curving the surface of the sensors to fit the socket created a pressure reading independent of the force applied by the residual limb. These data will not be reported.] The sensing element was approximately 0.5 mm thick and 18 mm in diameter, and the tape was approximately 0.1 mm thick. Thus, the three conditions at each site represented added thicknesses of 0.6, 1.6, and 2.6 mm to the inside of the socket. It was expected that the 0.6-mm thickness of the pressure sensor would be virtually undetectable by the subjects.

For each of the nine socket test conditions, the three scales described previously to measure perceived pressure, discomfort, and pain were utilized to measure perceptions as influenced by socket geometry variations at the sites of the wafers on the distal tibia and fibula head. The wording of the tasks was changed to measure perceptions “at this location.” The subjects stood with a height-adjustable platform supporting their residual limb and check socket. The platform height was adjusted to equalize hip height, as would be done for a routine check socket fitting. Subjects were asked to rock back and forth in their check sockets. Each subject then was asked to evaluate the distal tibia and fibula head by using the three category ratio scales. The socket was removed, the number of wafers changed, and the socket redonned until all nine combinations had been presented.

RESULTS AND DISCUSSION

Touch thresholds for each of the subjects are shown in Table 1. It was found that all subjects had touch thresholds below 10-g force for all points assessed on the exposed residual limb and contralateral limb; thus, none could be considered to have a limb-threatening neuropathy. Thresholds varied substantially between subjects and among assessment locations, however. In general, subject 2, who was the youngest, had the lowest thresholds, 0.32 g of force. Subject 3, who was the oldest, had the highest thresholds on the fibula head and distal tibia. All subjects had higher touch thresholds on the fibula head than on the distal tibia for both the residual limb and contralateral limb. This may have reflected morphologic differences in touch sensitivity between the respective dermatomes. Spatial acuity measurements also are shown in Table 1. Although there were variations between the residual limb and contralateral limb for each subject, no conclusions regarding the improvement of spatial acuity after amputation could be drawn.

Because it was assumed that the subjective scales had ratio properties, least-squares curves were fitted to the data. Although psychophysical research suggested that nonlinear relationships could exist, the range of stimuli and responses was considerably smaller than the range of the scales. Given the small range of pressures applied, it was assumed that a linear approximation would be adequate for analysis purposes. Objectives were to examine and compare the R-square values and the slopes of the regression lines. Given a sample size of n = 12 (the number of points used to fit the least squares lines involving global pressure), any R-square greater than + 0.33 was significantly different from zero at the 0.05 level, and any R-square greater than + 0.49 was significantly different from zero at the 0.01 level. Given a sample size of n = 9 (the number of points used to develop the least-squares curves involving check socket geometry), any R-square value greater than + 0.42 was significantly different from zero at the 0.05 level, and any R-square value greater than + 0.64 was significantly different from zero at the 0.01 level. All reported results were significant at the 0.01 level.

Application of the pressure bladder simulated the pressure distribution that would be expected in a socket produced by a pressure casting technique. This could result in an elongation of tissues and a uniform pressure distribution over the surface of the limb. As a result of the uniform pressure distribution, tissues of low stiffness and a relatively great thickness, such as adipose tissue, would deflect more than tissues of high stiffness and relatively small thickness, such as fascia and tendon. High global pressure also could result in the compression of tissue via the expulsion of fluids and possibly deformation of capillaries, veins, and arteries.

Figure 2 indicates the relationship between pressure produced by the bladder and subjects’ perception of global or deep pressure on the residual limb. Both experienced subjects tended to report constant pressure sensations over a range of applied pressures typically associated with blood pressure in the capillaries of the arteriolar limb. Subject 1, who was experienced, gave identical responses for the same pressure settings during each of the three trials. Pressure perception remained consistently at “moderate” up through 30 mm Hg and then became “somewhat strong” at 40 m m Hg. Subject 3, who also was an experienced wearer, consistently reported 20 and 30 mm Hg of pressure as having the same level of perceived intensity, followed by an upturn in perceived intensity at 40 mm Hg. Subject 2, who was inexperienced, exhibited a more consistent linear response, increasing from “just noticeable” at 10 mm Hg to “very strong” at 40 mm Hg. Although the plot for subject 2 suggests scattering of the data, the scattering reflects the fact that perceived strength of the pressure at 20, 30, and 40 mm Hg increased as the trials were repeated, possibly indicating an increasing sensitivity to global pressure as exposure increased. Similar to Subject 2, the pressure perceptions of subject 3 tended to increase as the trials were repeated. [The fact that both experienced subjects exhibited plateaus of constant reported pressure sensation, whereas the inexperienced subject did not, raises several questions for further research. The residual limbs of the experienced subjects contained little or no adipose tissue, and the muscles had undergone substantial atrophy. The inexperienced subject had both adipose tissue and an initial tissue volume that diminished rapidly after prosthesis use. Possible future research questions are: 1) Is there a relationship between muscle atrophy and global pressure sensation? (Does the number of active mechanoreceptors remain constant as muscle atrophies and the density of receptors per unit volume of tissue increases? If not, does the nervous system make adjustments to preserve the accuracy of pressure sensation?); 2) Is there a relationship between the volume of adipose tissue and muscle and pressure sensation? (Does less stiff tissue have a greater sensitivity to pressure gradients than stiffer tissue? Is the observed phenomenon related to the first stage of the two-stage theory?); 3) Does a prosthesis wearer adapt over the years to the pressure sensations of a fitted PTB socket with its nonuniform pressures such that it becomes difficult to detect global pressure changes? (Are global pressure judgments relative to the sensations of the PTB type of socket? Is the observed phenomenon related to the second stage of the two-stage theory?); and 4) Did the traumatic nature of the amputations and the resulting surgeries influence the sensations?]

Figure 2: Applied global pressure versus perception of pressure.

All three subjects indicated a shift in sensation from “moderate” to “somewhat strong” or higher at pressures between 30 and 40 mm Hg. It is possible that this shift may reflect the range of pressures within which blood flow into the capillaries is reduced to the point that nociceptors or deep pressure receptors begin to alert the individual to a dangerously diminished blood flow, and a concomitant decrease in the ability of the cardiovascular system to supply oxygen and nourishment to the tissues of the residual limb and remove the byproducts of metabolism.

Figure 3 plots the relationship between perceived pressure and perceived discomfort for subjects 1 and 2. Subject 3 reported no discomfort (“0”) for all of the pressure levels on each trial, and these data were not plotted. Subject 1, who was an experienced wearer, exhibited a response function in which no discomfort was experienced until perceived pressure became “moderate” but discomfort then increased rapidly to “moderate” as pressure sensation approached “somewhat strong.” The response function for subject 2, who was inexperienced, behaved in fairly classic manner, increasing linearly throughout the range of deep pressure sensation. Comparison of the regression coefficients indicated that the slope of the least-squares line for subject 1 was over 5 times as great as the slope of the line for subject 2 for the range of pressures in the experiment. The results indicate differences in sensory response functions between the two subjects; subject 1 reported no pressure sensation less than “moderate” or greater than “somewhat strong,” whereas subject 2 reported pressure sensations as low as “just noticeable” and as high as “very strong.” R-square values for both subjects were high, indicating a strong monotonic and linear relationship between the scales for deep pressure perception and discomfort for the range of experimental values. However, both subjects first reported discomfort when pressure was perceived as moderate. Thus, both subjects appeared to have the ability to isolate the magnitude of the sensation of pressure from the magnitude of perceived discomfort, suggesting that the scales were tapping two distinct subjective phenomena.

Figure 3: Perception of global pressure versus perceived discomfort.

When subjects reported discomfort, they were asked where it was most noticeable. Subject 1 reported the fibula head as the location of discomfort whenever pressure was reported as “somewhat strong.” Subject 2 reported the distal tibia as the location of discomfort when global pressures above “moderate” were reported; the location corresponded to a suture line where the subject previously had an infection and still experienced sensitivity and occasional pain. Subjects were asked if they were experiencing any pain for each pressure level during each trial, and all stated that no pain was being experienced.

When the wafers were inserted into the check sockets, pressures were increased at specific locations. A PTB socket, sometimes referred to as “bony contact socket,” places unequal pressures on the tissues, which results in pressure gradients within the residual limb. The distal tibia and fibula head are areas where the skin and underlying tissues are thin and where relief is provided in a PTB socket to minimize pressure. The introduction of wafers at these locations increased the pressure at points where small changes in socket geometry could be expected to produce large deformations of tissue relative to tissue thickness and increase pressure gradients, pressure sensation, and discomfort. In contrast to the uniform pressure relationships, the relationships obtained between pressure sensation and discomfort at both the fibula head and distal tibia were strong and similar, suggesting common underlying psychophysical response and sensory response mechanisms. Subject 1, who was experienced, reported pressure sensations for all combinations of wafer thickness, including the 0.6-mm sensor, as “moderate” and discomfort as “mild.” Because of the lack of variation in these data, they are not plotted. However, subject 1 also stated that all of the resulting pressures would produce intolerable discomfort if the socket were worn for a day. Figure 4 plots the relationships between pressure sensation and discomfort for the other two subjects. R-square values were high. The regression coefficients were numerically close for all four curves (0.96 and 1.3 for subject 2; 0.99 and 1.1 for subject 3), suggesting that the psychophysical relationship between pressure sensation and perceived discomfort may have been fairly similar for the two subjects at the fibula head and distal tibia. The magnitude of the regression coefficients indicates that a one-unit increase in pressure sensation produced approximately a one-unit increase in discomfort. Because the magnitudes of all four regression coefficients were close to 1, and the intercepts of all four regression curves were close to zero, the two scales appear to have been measuring highly similar subjective phenomena—the magnitudes of the pressure sensation and the discomfort perception were the same.

Figure 4: Perception of pressure at specific locations versus perceived discomfort.

Only subject 2 provided experimental data complete enough to facilitate a comparison of regression coefficients between global deep pressures created by the pressure bladder and location specific pressures created by altering socket geometry. The coefficients were roughly 2 and 2.5 times greater for locally applied pressures at the distal tibia and fibula head, respectively, indicating a much more rapid increase in discomfort with perceived pressure at these sites. This result suggests that the relationship between the magnitude of pressure sensation and magnitude of discomfort may be dependent on the particular site on a limb and whether pressure is applied globally or at a point. More research would be needed to evaluate this hypothesis.

Subject 3 reported “very mild” and “moderate” pain once each at the distal tibia for 2.6-mm inserts. Subject 3 also reported “moderate” pain once and “somewhat strong” pain twice when 2.6-mm inserts were placed at the fibula head. “Just noticeable” pain was reported at the fibula head once each for 0.6- and 1.6-mm inserts, and “very mild” pain was reported once for the 0.6-mm insert. Figure 5 shows for subject 3 the relationships between perceived pressure and pain and between pain and discomfort for the fibula head. Pain was not reported until a “moderate” pressure was perceived. For levels of pressure perception above “moderate,” the scale magnitude assigned to pain was one scale unit less than the magnitude assigned to pressure half the time, and equal to the magnitude assigned to pressure the other half of the time, suggesting that subject 3 might have been able to subjectively isolate the two sensations of pain and pressure some of the time. Subject 3 also appeared able to distinguish between the magnitudes of pain and discomfort, nearly always assigning discomfort a higher magnitude than pain at levels of pain below “somewhat strong.” Based on R-square values, the relationship between pain and discomfort was strongest.

Figure 5: Subject 3’s perception of pressure versus pain, and pain versus discomfort at the fibula head.

Conclusions

Results indicate that psychophysical methods for measuring sensations of pressure, discomfort, and pain are feasible and may hold promise as tools to help with the fitting of sockets as well as for research conducted on liner and socket designs and materials. The methods provide patients with a means for expressing and communicating the sensations and discomfort problems associated with socket fit and allow the prosthetist to quantify quality of fit. Although the research was carried out on patients using PTB sockets that feature nonuniform pressures, an experiment featuring uniform pressure also was carried out. The data suggested that psychophysical response functions for pressure sensation may change with experience and exposure to the forces encountered inside a socket. In particular, the data of the experiment indicated distinctive pressure sensation responses among the three subjects, which could reflect different amounts of experience with the fitting and wearing of a prosthesis, different residual limb tissue characteristics, or different response functions. The inexperienced patient might present a fitting challenge different from the experienced user, who may be able to more quickly identify problem areas. Further research involving a larger number of subjects would be needed to verify this.

The methods could provide helpful information for the fitting of many other socket types including total contact and ischial containment sockets. For example, if a prosthetist wanted to ensure an even distribution of pressure when fitting a total contact socket, geometry could be varied until the patient reported similar levels of pressure sensation throughout the socket, and no locations with discomfort greater than some very small amount, such as “just noticeable.” For sockets that load specific anatomical locations, such as the standard PTB socket, pressure sensations greater than “moderate” might be indicators of potential problems. To establish relationships between verbal descriptors and potential fitting problems, research involving a larger number of subjects and longitudinal data would be needed. Also, more locations inside a socket would need to be researched, such as the popliteal shelf and hamstring tendon areas, gastrocnemius region, and patella tendon bar. For patients who are experiencing continuing fitting problems, the methods could be used to maintain daily diaries in which the patients would record levels of discomfort and related data such as activity, medication, or ply of sock. This information could assist the prosthetist in evaluating fitting problems.

The scales can be applied quickly and easily. None of the experimental subjects expressed any difficulty using the scales, but it is recognized that some elderly and senile individuals might not be able to understand or carry out the task instructions. Also, the scales must be worded in a language with which the patient has a high level of fluency. Although an established category ratio scale was employed, alternative scaling methods could be explored; for example, scales featuring gradations of color densities or color changes, accompanied by verbal descriptors, might be developed. None of the subjects had impaired sensations, but it would be desirable to study subjects who are experiencing neuropathy in their residual limbs. Psychophysical methods could be used to increase knowledge on the sensations of deep pressure that are experienced by such individuals, a subject for which virtually no information exists.

Acknowledgments

The author wishes to thank Ossur-Flexfoot for lending an Iceross Icecast pressure bladder. The author is indebted to Jerry Fullerton, CPO, and owner of Superior Limb and Brace, Las Vegas, for helping secure subjects and for providing support and encouragement during the project.

Note: This research was self-financed by the author to fulfill his Prosthetics Residency research project requirement.