- Email: [email protected]
Fingertip touch improves postural stability in patients with peripheral neuropathy
Gait and Posture 14 (2001) 238– 247 www.elsevier.com/locate/gaitpost
Fingertip touch improves postural stability in patients with peripheral neuropathy Ruth Dickstein b,*, Charlotte L. Shupert a, Fay B. Horak a a
Neurological Sciences Institute of Oregon Health Science Uni6ersity, West Campus – Bldg. 1, 505 NW 185 th A6enue, Portland OR 98209, USA b Department of Physical Therapy, Sackler Faculty of Medicine, Tel-A6i6 Uni6ersity, Ramat-A6i6, Tel-A6i6 69978, Israel Accepted 27 November 2000
Abstract The purpose of this work was to determine whether fingertip touch on a stable surface could improve postural stability during stance in subjects with somatosensory loss in the feet from diabetic peripheral neuropathy. The contribution of fingertip touch to postural stability was determined by comparing postural sway in three touch conditions (light, heavy and none) in eight patients and eight healthy control subjects who stood on two surfaces (firm or foam) with eyes open or closed. In the light touch condition, fingertip touch provided only somatosensory information because subjects exerted less than 1 N of force with their fingertip to a force plate, mounted on a vertical support. In the heavy touch condition, mechanical support was available because subjects transmitted as much force to the force plate as they wished. In the no touch condition, subjects held the right forefinger above the force plate. Antero-posterior (AP) and medio-lateral (ML) root mean square (RMS) of center of pressure (CoP) sway and trunk velocity were larger in subjects with somatosensory loss than in control subjects, especially when standing on the foam surface. The effects of light and heavy touch were similar in the somatosensory loss and control groups. Fingertip somatosensory input through light touch attenuated both AP and ML trunk velocity as much as heavy touch. Light touch also reduced CoP sway compared to no touch, although the decrease in CoP sway was less effective than with heavy touch, particularly on the foam surface. The forces that were applied to the touch plate during light touch preceded movements of the CoP, lending support to the suggestion of a feedforward mechanism in which fingertip inputs trigger the activation of postural muscles for controlling body sway. These results have clinical implications for understanding how patients with peripheral neuropathy may benefit from a cane for postural stability in stance. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Light touch; Peripheral neuropathy; Postural stability
1. Introduction The importance of somatosensory information from the feet to the control of upright stance balance is well established [1–4]. Delayed, distorted and/or absent somatosensory information from cutaneous, joint and muscle receptors underlie many of the postural abnormalities in individuals with distal sensory neuropathy [5 – 7]. The prevalence of distal symmetrical neuropathy in patients with long standing diabetes (DM-PN), is about 50% [8], and is related to instability during stance and gait [9], as well as to frequent fall accidents [10,11]. * Corresponding author. Tel.: + 972-3-640-9224; fax: +972-3-6409223. E-mail address: [email protected] (R. Dickstein).
Laboratory studies have demonstrated excessive postural sway in these patients [12], which is further accentuated by darkness or by unreliable visual conditions [13,14]. Postural responses to surface perturbations are also delayed with poor amplitude scaling in patients with peripheral neuropathy [15], but not in patients with profound loss of vestibular function [6], suggesting that somatosensory information is more important for postural responses than vestibular information. Clinicians attempt to improve postural stability of DM-PN patients by providing a walking cane [16]. Recent studies suggest that a cane may be useful in improving balance because it provides an alternative sensory reference and not because it provides an increased mechanical base of support [17,18]. It has been shown that light touch from a fingertip with forces
0966-6362/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 6 6 - 6 3 6 2 ( 0 1 ) 0 0 1 6 1 - 8
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
under 1 N that provide somatosensory cues but not mechanical support, attenuated postural sway similarly to heavy touch that allowed mechanical support in healthy subjects [18– 23]. Furthermore, reduction in sway due to light touch of a fingertip on a stable surface reduced postural sway even more than vision in healthy subjects and in subjects with bilateral loss of vestibular function [22]. Likewise, touch cues via the index finger were shown to be as effective as heavy force contact in providing stabilization of stance in blind subjects [23]. In healthy subjects, light touch from a fingertip also helped to suppress muscle vibration-induced postural sway due to aberrant proprioceptive information from the feet [24], suggesting that fingertip touch may be able to substitute for impaired or altered sensory information from the feet. The studies of Jeka and his colleagues further suggest that somatosensory information from a fingertip is used in a feedforward manner to trigger appropriate postural activity in leg muscles because changes in forces under the fingertip during light touch, but not heavy touch conditions, are phase-advanced over changes in center of pressure (CoP) under the feet [18,21]. The extent to which patients with distal peripheral neuropathy can use a similar strategy to improve postural stability in stance is unknown. Sensory neuropathy secondary to diabetes is discernable in the sensory nerves of the feet prior to the sensory and motor nerves of the hand [8,25]. Thus, it is possible that patients with diabetic neuropathy have enough somatosensory function in the hands to use it for improving balance in stance and gait. In fact, since subjects with somatosensory loss are more unstable than healthy subjects, they may benefit even more than control subjects from finger somatosensory feedback as a sensory substitution for diminished foot CoP information indicating body sway. The goal of the current study was to determine whether DM-PN subjects with profound loss of somatosensory information in the feet would be able to utilize light touch cues from a single fingertip for increasing stance stability. The specific contribution of the somatosensory input versus mechanical support was determined by comparing postural stability in stance conditions in three touch modes: ‘light touch’ (LT), ‘heavy touch’ (HT) and ‘no touch’ (NT). Postural stability during stance was challenged with eyes closed and/or by stance on a compliant, foam surface to determine the environmental conditions in which patients with DM-PN show the most instability and most help from fingertip touch. Forces under the fingertip were measured to determine whether patients with neuropathy could use light-touch feedforward mechanisms for control of posture similar to control subjects.
239
2. Methods
2.1. Subjects and clinical characteristics The subjects were eight adults (five men) suffering from Diabetes Mellitus (DM) with profound sensory peripheral neuropathy (DM-PN), with a mean age of 59.7910.7 years, and eight age-matched healthy volunteers (four men, mean age 60.69 9) who served as controls. All patients were diabetic for at least 4 years, the average duration being 21 years. Only completely independent community dwellers with no functional visual and/or motor deficits were accepted, and none of the patients used a cane in daily life. In all participants, normal function of the vestibular system was verified by sinusoidal horizontal rotation testing of the vestibulo-ocular reflex (VOR). All subjects in the patient and control groups had gain and phase measures of the VOR well within the 95% tile of normal laboratory limits. Specifically, VOR gain was 0.67 for the neuropathy group and 0.5 for the control group compared to B0.45 for the 95th percentile limit of normal gain. Somatosensory sensation of the right index finger was measured by clinical testing of proprioception and vibration sense at the distal interphalangeal (IP) joint, and by static two-point discrimination of the palmar aspect of the distal phalanx using the Mackinnon-Dellon disk criminator [26]. Only two of the DM-PN patients and none of the control subjects had reduced index finger somatosensory sensation. Note that for two-point discrimination, age-adjusted norms were applied; that is, 4 mm distance between the two points was established as upper normal limit [27,28]. The existence of sensory neuropathy in patients’ lower extremities was established by electrodiagnostic testing and by a thorough physical examination of the feet. Patients’ motor nerve (peroneal and posterior tibial nerves) conduction velocities were mostly within normal limits, whereas conduction of their sensory nerves was substantially impaired. Specifically, responses of the two medial and lateral plantar nerves were undetectable in all eight neuropathy subjects and sural nerve responses were absent in six of them. In the two patients with detectable sural nerve responses, the conduction velocity and amplitude of the sural nerve were significantly outside normal limits. The physical examination included routine testing of: ankle joint range of motion and strength of the dorsal and plantar flexor muscles; threshold of vibration sensation at the medial malleoli and the first metacarpophalangeal (MP) joint (128 Hz tuning fork); and testing of kinesthetic sensation at the ankle and the first MP joint. The results of patients’ physical examination are summarized in Table 1. Cutaneous sensation was further assessed by Semmes–Weinstein monofilaments
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
240
[29,30], at three dorsal and two plantar foot zones supplied by five cutaneous branches (The dorsal digital cutaneous, intermediate dorsal, and sural nerve on the dorsal side of the foot, and the medial and lateral plantar nerves on the plantar aspect). The testing protocol followed the interval comparison method [31]. Superficial sensation in each of the five zones was found to be significantly reduced in the DM-PN patients in comparison to the control subjects (P B 0.05; Mann– Whitney U test).
plate for support. In NT trials subjects’ right index finger was held above the touch plate and no contact with the plate was permitted. It was not difficult for the subjects to hold their finger above the touch plate and practice trials were not needed. Each trial was 40 s long. Data collection in each trial started after subjects assumed the required stance with only 5 cm distance between their medial malleoli and facing an art poster 1 m ahead of them. The circumference of each foot was marked on the surface, assuring stance at the same location on the plates in all trials. The same narrow base stance was kept also during foam standing trials. A safety harness was used to minimize fall risks.
2.2. Experimental protocol Subjects were tested while standing barefoot on a dual force platform surface, near a circular (11 cm diameter) touch plate mounted on a small AMTI force transducer. The touch plate was positioned at the subjects’ right side (all subjects were right-handed), 90 cm height from the floor, immediately anterior to the right foot. Each participant was tested in 12 sensory conditions that were randomly presented twice in two separate blocks of 12 trials. The conditions consisted of three touch modes, ‘heavy’ (HT), ‘light’ (LT) and ‘no’ (NT) touch, in each of two visual (eye open or closed) and two surface (firm and foam) conditions. The foam consisted of 7.5 cm, medium density Temper foam. During HT trials, subjects were allowed to use the touch plate for mechanical support by transmitting onto it as much force as they chose with their index finger. During LT trials, the touch plate was contacted by the right index finger pulp with a pressure not exceeding 1 N. An audio-feedback device attached to the touch plate informed subjects if the limit had been exceeded, thus preventing them from using the touch
2.3. Dependent 6ariables The data was sampled at 240 Hz and included the three dimensional forces measured under each foot, trunk antero-posterior (AP) and medio-lateral (ML) angular velocity and the three dimensional forces under the right index finger. Trunk velocity was measured by a Watson angular rate sensor, which was attached to the sternum. The root mean square (RMS) of the trunk velocity for each plane was used as a measure of trunk stability. Whole body AP and ML movements of the center of pressure (CoP) were derived from the dual force platform data. The RMS of CoP sway was calculated for the 40-s duration of each trial. The mean magnitude of forces transmitted via the index finger onto the touch-plate in the antero-posterior, medio-lateral and vertical axis were determined after the raw data were low passed filtered at 15 Hz, based on a residual analysis [32].
Table 1 Clinical data Patients Normal
Controls Reduced
Absent
Normal
Reduced
Absent
Ankle strength
16
–
–
16
–
–
Ankle ROM
16
–
–
16
–
–
Joint position Toe Ankle Rt index finger
– 10 6
15 6 2
1 – –
16 16 –
– – –
– – –
Vibration Toe Ankle Rt index finger
– 1 5
10 13 3
6 2 –
9 16 8
7 – –
– – –
Superficial sensation Dorsal foot Plantar foot
4 2
12 14
– –
16 16
– –
– –
Numbers are counts from both legs. For superficial sensation, reduced sensation was assigned to subjects whose Semmes–Weinstein mean feet scores fell beyond the upper 95% confidence interval of the corresponding mean of the control subjects.
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
241
Fig. 1. The CoP in a representative subject with peripheral neuropathy due to diabetes mellitus during stance under three fingertip touch conditions: no touch, light touch, and heavy touch. The subject stood on foam with eyes closed and showed decreased CoP sway with fingertip touch; heavy touch reduced sway more than light touch in this condition. A-P, antero-posterior; M-L, medio-lateral.
2.4. Analysis Descriptive methods and multiple analysis of variance (MANOVA) with one fixed (group) and three repeated measures factors (surface * vision * touch mode) were employed. Post-hoc testing compared the effects of LT with HT and of the effects of LT to NT. Temporal synchronization between AP and ML movements of the CoP and between corresponding fingertip forces, was determined by calculation of the cross-correlation between them. The correlation values were normalized by Fisher’s Z-transform method [33] and then subjected to a 2*2*2*2 (Group* Surface* Vision* Touch mode) MANOVA analysis. In addition, similar to the analysis used by Jeka et al. [21,22], the time point of the highest correlation value between each pair of CoP and finger force data was established within a time window of 9625 ms.
3. Results
3.1. CoP postural sway Both light touch and heavy touch of a fingertip significantly attenuated postural sway in both the DM-
PN subjects and the control subjects under all sensory conditions (F= 87.9, P B 0.0001 for AP CoP; F =27.4, PB 0.0001 for ML CoP). Fig. 1 shows examples of CoP sway from a representative neuropathy subject during with NT, LT, and HT in the condition resulting in the most sway for both groups, e.g. eyes closed standing on foam. Heavy touch had a larger effect than LT on reducing CoP sway in both the AP and ML directions (F= 65, PB 0.0001 for AP CoP, F=7.2, P= 0.01 for ML CoP). The difference between heavy touch and light touch was much less apparent, however, when subjects stood on the firm surface. The amount of reduction in both AP and ML CoP sway due to light and heavy touch was similar in the two groups. The effect of light touch was comparable to the effect of vision on reducing the CoP sway in both groups. The effect of touch on average AP and ML CoP RMS values under each condition in the two groups are presented in Fig. 2A and B. The effect of touch in the two DM-PN patients with reduced sensation in their fingertip was not different from the other six subjects. In general, ML but not AP CoP sway of the DM-PN patients was larger than sway of the healthy subjects (F= 4.6, P= 0.05 for ML CoP but P = 0.06 for AP CoP). Stance on a compliant surface or absence of
242
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
vision significantly increased CoP sway (F = 85.2 PB 0.0001 and F=32.3, PB 0.0001). A significant interaction between these effects (F =24.6, P B 0.0001 for AP CoP and F= 15.0, P = 0.001 for ML CoP) indicated that CoP sway was largest when subjects stood on the compliant surface with their eyes closed. The effect of touch on attenuating CoP sway was larger on a compliant than a firm surface (F =24.8, P B 0.0001 for AP CoP and F= 16.5, P B0.0001 for ML CoP) and was larger without vision than with vision (F = 2.7, P =0.08 for AP CoP and F = 7.9, P =0.001 for ML CoP).
3.2. Trunk 6elocity Like CoP, trunk velocity was significantly reduced by light touch and heavy touch in both patients with DM-PN and control subjects under all sensory conditions (F=11.6, PB 0.0001 for AP trunk velocity and F =9.8, P =0.0006 for ML trunk velocity). Fig. 3 illustrates the trunk velocity of one patient and one control subject in the most challenging condition (eyes closed, standing on foam). Unlike CoP sway, light touch attenuated AP and ML trunk RMS angular velocity at a similar magnitude to the attenuation caused by heavy touch for all visual and surface condi-
tions in both patients and control subjects (P \ 0.05 for differences between conditions of LT and HT). The effects of touch on the average AP and ML trunk velocity RMS for each condition in the two groups is summarized in Fig. 4. Like CoP sway, the amount of reduction of trunk velocity due to touch was similar in the control subjects and DM-PN subjects. Furthermore, like CoP sway, the effect of light touch in reducing trunk velocity was not statistically different from the effect of vision for both groups. During stance on a firm surface, trunk angular RMS velocity was comparable in the two groups. Stance on the foam was associated with substantially a higher trunk RMS velocity in the DM-PN patients than in the control subjects with this difference larger in the ML direction (F= 3.8, P= 0.06 and F= 4.7, P= 0.04 for the AP and ML directions respectively). Mean AP trunk RMS velocity was larger than ML trunk velocity (range between 0.57 and 1.36 deg/s for AP and between 0.37 and 1.14 deg/s for ML trunk RMS velocity). Closing the eyes and standing on foam significantly increased trunk velocity RMS for both the AP and ML directions (F=41.06 PB 0.0001, F= 26.7 PB 0.0001 for AP velocity and F= 63.4 PB 0.0001, F=24.3 PB 0.0001 for ML velocity).
Fig. 2. Effects of light touch and heavy touch on mean ( 9 S.E.) antero-posterior (A) and medio-lateral (B) CoP RMS. Subjects stood with feet close together under four sensory conditions: eyes open or closed and on a firm or foam surface. Black bars indicate DM-PN group data and white bars indicate control group data. N, no touch; L, light touch; and H, heavy touch.
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
243
Fig. 3. Medio-lateral (A) and antero-posterior (B) angular velocity of a DM-PN patient and of a control subject (C and D respectively), during stance on foam surface with eyes closed. (presented are 5 – 6 s in the middle of a 40-s long trial). X-axis tick marks denote 1 s.
3.3. Fingertip forces Patients with somatosensory loss due to DM-PN exerted similar forces from their index finger onto the touch plate as the control subjects. Mean vertical and lateral forces across all LT and HT conditions are given in Table 2. The only significant difference between groups was in the laterally directed forces: DM-PN patients employed a larger rightward-directed force (toward the touch plate) than the controls (Table 2). For all subjects, standing on foam was associated with significantly larger vertical fingertip force than standing on a firm surface (F =12.9, P =0.003). The horizontal forces during HT conditions were directed forward and rightward. AP forces were neither affected by group nor by vision or by surface condition. However, ML forces during HT conditions and foam stance were associated with a greater amount of force directed toward the right side as compared with stance on a firm surface. The forces applied onto the touch bar during LT conditions by the two patients who had reduced finger sensation (as evident from the two-point discrimination test), were not larger than the forces applied by the other patients.
3.4. Correlation between AP CoP and AP fingertip forces The correlation between the AP forces exerted onto the touch plate and AP CoP movements under subjects’ feet indicated that a forward movement of the CoP was associated with a backward directed force onto the touch plate and vice-versa. Mean correlation range was |= 0.6–0.8 and 0.3–0.6 (for controls and patients, respectively). Correlations between touch plate forces and CoP were significantly larger during LT than HT (F= 10.6, P= 0.005), and during foam stance than on a firm surface (F= 4.6, P= 0.04). The highest correlation values were obtained when changes in the CoP lagged behind the changes in fingertip force exerted onto the touch plate. Mean time lag between AP forces on the touch plate and AP CoP ranged from 95 to 342 ms. The time lag was significantly longer during LT than during HT trials (F= 9.8, P= 0.006). Correlation values between ML CoP movements and ML forces exerted onto the touch plate were low during stance on the firm surface and became significantly higher during stance on the foam (F= 29.4, PB
244
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
0.0001). This surface-dependent increase in correlation between ML CoP and fingertip force was larger in the patients than control subjects (F = 6.1, P = 0.02). Correlation values during stance on the foam ranged between 0.4 and 0.6, with a positive association between leftward movement of the CoP and a rightward force from the fingertip on the plate. Time lags between the ML movements of the CoP and ML forces exerted onto the plate were very small during stance on a firm surface and became substantially larger when standing on the foam. Nevertheless, in both surface conditions during LT trials, changes in the CoP lagged behind changes in the ML force on the touch plate (time lags values between 68 and 293 ms), while during HT these time lags were negligible.
4. Discussion Trunk and CoP sway of the DM-PN patients, as well as the healthy subjects, decreased with light and heavy fingertip touch. The effects of touch became more apparent in conditions of deficient surface and/or visual information in both subject groups. The effectiveness of touch to decrease sway was larger in subjects with neuropathy than in control subjects, especially in the most difficult conditions. When standing on foam, subjects with neuropathy had larger than normal CoP and
trunk sway with no touch, but similar sway as controls with light and heavy touch. The larger effect of touch in the neuropathy than control group might be expected because neuropathy subjects may need to substitute fingertip touch for deficient foot somatosensory information in order to reduce their larger than normal postural instability. Thus, the results support a compensatory increase in sensitivity or ‘weighting’ of fingertip somatosensory information for controlling excessive postural sway in subjects with chronic somatosensory loss in the feet. In fact, the influence of light fingertip touch on reducing postural sway becomes larger when either control or neuropathy subjects are more unstable in conditions of standing on foam and without vision. Despite the likelihood of decreased somatosensory information in the fingertip of patients with peripheral neuropathy, they were able to take advantage of additional feedback to reduce postural sway. It is not possible to determine whether subjects are using cutaneous light pressure information or joint and muscle proprioceptive information to attenuate postural sway with light touch [18], nor to determine accurately which of these sensory pathways are adequate in the patients. Nevertheless, these results show that subjects with neuropathy in their feet and legs sway more than normal and functionally benefit significantly from light fingertip touch.
Fig. 4. Group means ( 9S.E.) of antero-posterior (A) and medio-lateral (B) RMS angular trunk velocity under the same sensory conditions as Fig. 2.
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
245
Table 2 Mean vertical and lateral forces (N) applied by the index finger to the touch plate. Vertical (positive is downward) Light touch Patients Controls Group effect Surface effect
Lateral (positive is rightward) Heavy touch
0.27 9 0.02 12.3 912.9 0.269 0.03 6.5 9 3.5 NS Vertical forces are larger during foam stance F= 12.9, P =0.003
In contrast to former findings [18,19] light touch in the current study was not as effective as heavy touch in decreasing CoP sway, but had an intermediate influence between the effects of ‘heavy’ and ‘no touch’. These results stem from the challenging stance conditions imposed on our subjects since the reduced effect of light touch compared to heavy touch was especially obvious during stance on the foam surface. The instability in subjects with somatosensory loss was further increased when they stood on the foam surface. This increased instability suggests either that the foam surface removed some surface somatosensory information that neuropathy patients were using on firm surface or that the foam produced additional mechanical instability. Standing on foam also increased instability in the control subjects such that the CoP and trunk velocity was at least as large, or larger, than for patients with neuropathy on the firm surface. The effectiveness of fingertip touch as a sensory substitution for somatosensory loss in the feet can be seen by the similar level of postural sway (both CoP and trunk velocity) between the conditions of light, and especially heavy fingertip touch when standing on foam, and the condition of no touch when standing on the firm surface. Regardless of the amount of excessive sway in neuropathy patients, fingertip touch reduced sway so it was similar to normal subjects using touch. These results suggest that touch doesn’t add a consistent amount or consistent percent of stability but rather touch allows a specific level of stability in both neuropathy patients and control subjects. While prior reports have documented an increased CoP sway in diabetic patients with peripheral neuropathy [7,34,35], the AP CoP sway amplitude of the current patients was comparable to that of the healthy subjects, although ML sway on a firm surface was somewhat larger. It is possible that the intactness of the motor nerves supplying the leg muscles contributed to the small loss CoP sway amplitude in the patients despite their significant loss of somatosensory information. Alternatively, the absence of excessive CoP sway of the patients could result from an a-priori avoidance from relying on sensory information from their feet for balance tasks. Perhaps, their prolonged inability to rely
Light touch
Heavy touch
0.03 0.98 0.008 0.06 Patients exert larger rightward forces F=5.5, P =0.03 Rightward forces are larger during foam stance F =9.8, P =0.0007
on surface inputs through their feet habituated them to depend more on vestibular, visual and/or proximal somatosensory inputs, and to execute postural corrections and adaptations at higher body segments without excessively moving the COP under the feet [6,36]. Reliance on vestibular information could be the preferred strategy for substitution of the deficient surface information. This suggestion is supported by the observations that patients with profound sensory neuropathy can preserve their balance on a CoP sway-referenced surfaces as successfully as healthy subjects, provided that their vestibular system is intact [37]. On the other hand, their postural stability is adversely affected by standing on a surface that is sway-referenced to the center of mass (CoM), suggesting, for these subjects, CoM information is more critical for stance stability than the CoP [37]. The fact that ML CoP sway during foam stance was accentuated in the patients with neuropathy is explainable by the distorted foot sensory information, combined with the relatively narrow stance base in this study. Other studies have reported increased lateral sway in the elderly compared to the young, especially in elderly subjects with a history of falls. This increased ML sway suggests either that those elderly subjects may, in fact, have posturally significant somatosensory loss or that ML sway tends to increase prior to AP sway in many types of increased instability. In addition, the fact that the touch plate was positioned laterally, may have encouraged the DM-PN subjects in our study to adjust their balance in this dimension. The fact that patients transferred more force onto the plate than the healthy subjects during the ‘heavy touch’ conditions, is indicative of their greater sense of instability compared with the healthy controls, and of a subsequent greater reliance on the touch plate, when this option was available. Trunk velocity was also higher in the patients than in the healthy controls, mainly when standing on foam. Thus, the additive influence of disruptive surface somatosensory information [38] and the deficient foot sensory information due to neuropathy led patients to use excessive trunk movements in both planes for balancing. This excessive trunk velocity probably reflects
246
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
increased reliance on a hip strategy for postural control in patients with somatosensory loss as reported previously. The effect of light touch on trunk velocity was impressive, being comparable to that of heavy touch for both the AP and ML directions. In contrast to its low potential to decrease CoP excursion, light touch was as effective as heavy touch in reducing trunk velocity. Because normal regulation of the CoM position is dependent on continuous monitoring of the CoP position [39], which is impaired in the DM-PN patients, the contribution of external touch cues via the upper extremity appears to be of great importance. Specifically, somatosensory information via the arm is useful as an orientation reference for perceiving the alignment of the body with respect to the environment, facilitating responses for the control of an upright trunk [18–20]. Neuropathy subjects appear capable of using an ankle, instead of hip, strategy to control postural sway when using fingertip touch. The arm in contact with the touch plate is in a better biomechanical position than the foot for sensing postural instability, thus enabling faster responses to correct trunk sway [19,40]. Our results support the application of a feedforward ‘sensory-motor strategy’ that uses somatosensory information from the finger to result in changes in forces under the feet to reduce sway. This feedforward strategy reduces body sway both in the frontal and sagittal planes for both control and neuropathy subjects. Time lags in the order of 300 ms were measured between fingertip forces during light touch and subsequent CoP movements for both subject groups similar to previous studies with healthy subjects and subjects with vestibular or visual loss [20,41]. Thus, any slowing of nerve conduction in the limbs of neuropathy subjects was not sufficient to disable this fast, feedforward mechanism. In conclusion, for both DM-PN patients and healthy subjects, fingertip somatosensory inputs through light touch were as effective as heavy touch in attenuating trunk velocity but less effective in decreasing CoP sway. Since patients with peripheral neuropathy showed larger trunk velocity than healthy subjects, the findings suggest an important functional role for LT in controlling stance balance in this subject group. Additionally, the application of a feedforward mechanism, in which fingertip inputs trigger the activation of postural muscles for controlling body sway, have clinical implications for understanding how patients with peripheral neuropathy may benefit from a cane for postural stability in stance.
References [1] Kavounoudias A, Roll R, Roll JP. The plantar sole is a ‘dynamometric map’ for human balance control. Neuroreport 1999;9:3247 – 52.
[2] Horak FB, Macpherson JM. Postural orientation and equilibrium. In: Smith JL, editor. Handbook of Physiology: Section 12: Exercise Regulation and Integration of Multiple Systems. New York: Oxford University Press, 1996:255 – 92. [3] Wu G, Chiang JH. The significance of somatosensory stimulations to the human foot in the control of postural reflexes. Exp Brain Res 1997;114:163 – 9. [4] Mauritz KH, Dietz V, Haller M. Balancing as a clinical test in the differential diagnosis of sensory-motor disorders. J Neurol Neurosurg Psychiatry 1980;43:407 – 12. [5] Thoumie P, Do MC. Changes in motor activity and biomechanics during balance recovery following cutaneous and muscular deafferentation. Exp Brain Res 1996;110:289 – 98. [6] Horak FB, Nashner LM, Diener HC. Postural strategies associated with somatosensory and vestibular loss. Exp Brain Res 1990;82:167 – 77. [7] Semoneau GG, Ulbrecht JS, Derr JA, Becker MB, Cavanagh PR. Postural instability in patients with diabetic sensory neuropathy. Diabetes Care 1994;12:1411 – 21. [8] Green DA, Sima AAF, Albers JW, Pfeifer MA. Diabetic neuropathy. In: Rifkin H, Porte D, editors. Diabetes Mellitus. Theory and Practice. New York: Elsevier, 1997:710 – 53. [9] Courtemanche R, Teasdale N, Boucher P, Fleury M, Lajoie Y, Bard C. Gait problems in diabetic neuropathy patients. Arch Phys Med Rehab 1996;7:849 – 55. [10] Richardson JK, Ashton-Miller JA. Peripheral neuropathy: an often-overlooked cause of falls in elderly. Postgrad Med 1996;6:161 – 72. [11] Cavanagh PR, Derr JA, Ulbrecht JS, Maser RE, Orchard TJ. Problems with gait and posture in neuropathic patients with insulin-dependent diabetes mellitus. Diabetic Med 1992;9:469 – 74. [12] Uccioli L, Giacomini PG, Monticone G, Magrini A, Durola L, Bruno E, Parisi L, Menzinger G. Body sway in diabetic neuropathy. Diabetes Care 1995;18:339 – 44. [13] Girolamo S, Menzinger G. Body sway in diabetic neuropathy. Diabetes Care 1995;18:339 – 44. [14] Ledin T, Odkvist LM, Vrethem M, Moller C. Dynamic posturography in assessment of polyneuropathic disease. J Vestibular Res 1991;1:123 – 8. [15] Inglis JT, Horak FB, Shupert CL, Rycewicz-Jones C. The importance of somatosensory information in triggering and scaling automatic postural responses in humans. Exp Brain Res 1994;101:159 – 64. [16] Ashton-Miller JA, Yeh MW, Richardson JK, Galloway T. A cane reduces loss of balance in patients with peripheral neuropathy: results from a challenging unipedal balance test. Arch Phys Med Rehab 1996;77:446 – 52. [17] Jeka JJ. Light touch contact as a balance aid. Phys Ther 1997;5:476 – 87. [18] Jeka JJ, Lackner JR. Fingertip contact influences human postural control. Exp Brain Res 1994;100:495 – 502. [19] Holden M, Ventura J, Lackner JR. Stabilization of posture by precision contact of the index finger. J Vestibular Res 1994;4:285 – 301. [20] Jeka JJ, Lackner JR. The role of haptic cues from rough and slippery surfaces in human postural control. Exp Brain Res 1995;103:267 – 76. [21] Jeka J, Oie K, Schoner G, Dijkstra T, Henson E. Position and velocity coupling of postural sway to somatosensory drive. J Neurophys 1998;79:1661 – 74. [22] Lackner JR, Dizio P, Jeka J, Horak F, Krebs D, Rabin E. Precision contact of the fingertip produces postural sway of individuals with bilateral vestibular los. Exp Brain Res 1999;126:459 – 66. [23] Jeka JJ, Easton RD, Bentzen BL, Lackner JR. Haptic cues for orientation and postural control in sighted and blind individuals. Percept Psychophys 1996;58:409 – 23.
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247 [24] DiZio P, Lackner JR, Rabin E. Sensorimotor and environmental conditions constraining use of the hand as a guide for posture. In: Proceedings of the 13th International Symposium: Multisensory Control of Posture & Gait, Paris, June, 1997. [25] Porte D, Sherwin RS. Diabetes Mellitus. Stanford CT: Appelton & Lange, 1997. [26] Mackinnon SE, Dellon AL. Two point discrimination tester. J Hand Surg Am 1985;1:906 –7. [27] Desroisiers J, Hebert R, Bravo G, Dutil E. Hand sensibility of healthy older people. JAGS 1996;44:974 –8. [28] Louis DS, Green TL, Jacobson KE, Rasmussen C, Kolowich P, Goldstein SA. Evaluation of normal values for stationary and moving two point discrimination in the hand. J Hand Surg 1984;9A:552 – 5. [29] Bell JA. Semmes – Weinstein monofilaments testing for determining cutaneous light touch/deep pressure sensation. The Star/ Carville, November – December, (from ‘ Directions for using Semmes– Weinstein filaments, 1984. [30] Mueller MJ. Identifying patients with diabetes mellitus who are at risk for lower-extremity complications. Use of Semmes – Weinstein Monofilaments. Phys Ther 1996;76:68 – 71. [31] Holewski JJ, Stess RM, Graf PM, Grunfeld C. Aesthesiometry: quantification of cutaneous pressure sensation in diabetic peripheral neuropathy. J Rehab Res Dev 1988;25:1 –10. [32] Winter DA. Biomechanics and Motor Control of Human Movement. New York: Wiley, 1990:36 –45.
247
[33] Snedecor GW, Cochran WG. Statistical Methods. Iowa State University, 1987:186. [34] Wu G, Zhao W. The role of mechanoreceptive information in the stability of human upright posture. A theoretical consideration. Motor Control 1997;1:3 – 19. [35] Boucher P, Teasdale N, Courtemanche R, Chantal B, Fleury M. Postural stability in diabetic polyneuropathy. Diabetes Care 1995;18:638 – 44. [36] Nutt JG, Horak FB. Gait and balance disorders. In: Watts R, Koller W, editors. Movement Disorders. Neurologic Principles and Practice. New York: McGraw-Hill, 1995:51. [37] Shupert CL, Dickstein R, Horak FB, Peterka R, Jones C. Surface sway-referencing and peripheral neuropathy disrupt somatosensory information for postural stability in stance. Gait Posture, 9 (Suppl. 1):S51. [38] Chiang JH, Wu G. The influence of foam surface on somatosensory receptors in the human foot — a rotational platform study. Gait Posture 1996;4:122 – 9. [39] Winter DA. Biomechanics and Motor Control of Human Movement. New York: Wiley, 1990:93 – 6. [40] Jeka JJ, Schoner G, Dijkstra T, Ribeiro P, Lackner JR. Coupling of fingertip somatosensory information to head and body sway. Exp Brain Res 1997;113:475 – 83. [41] Clapp SC, Wing AM. Light touch contribution to balance in normal bipedal stance. Exp Brain Res 1999;125:521 – 4.
Fingertip touch improves postural stability in patients with peripheral neuropathy Ruth Dickstein b,*, Charlotte L. Shupert a, Fay B. Horak a a
Neurological Sciences Institute of Oregon Health Science Uni6ersity, West Campus – Bldg. 1, 505 NW 185 th A6enue, Portland OR 98209, USA b Department of Physical Therapy, Sackler Faculty of Medicine, Tel-A6i6 Uni6ersity, Ramat-A6i6, Tel-A6i6 69978, Israel Accepted 27 November 2000
Abstract The purpose of this work was to determine whether fingertip touch on a stable surface could improve postural stability during stance in subjects with somatosensory loss in the feet from diabetic peripheral neuropathy. The contribution of fingertip touch to postural stability was determined by comparing postural sway in three touch conditions (light, heavy and none) in eight patients and eight healthy control subjects who stood on two surfaces (firm or foam) with eyes open or closed. In the light touch condition, fingertip touch provided only somatosensory information because subjects exerted less than 1 N of force with their fingertip to a force plate, mounted on a vertical support. In the heavy touch condition, mechanical support was available because subjects transmitted as much force to the force plate as they wished. In the no touch condition, subjects held the right forefinger above the force plate. Antero-posterior (AP) and medio-lateral (ML) root mean square (RMS) of center of pressure (CoP) sway and trunk velocity were larger in subjects with somatosensory loss than in control subjects, especially when standing on the foam surface. The effects of light and heavy touch were similar in the somatosensory loss and control groups. Fingertip somatosensory input through light touch attenuated both AP and ML trunk velocity as much as heavy touch. Light touch also reduced CoP sway compared to no touch, although the decrease in CoP sway was less effective than with heavy touch, particularly on the foam surface. The forces that were applied to the touch plate during light touch preceded movements of the CoP, lending support to the suggestion of a feedforward mechanism in which fingertip inputs trigger the activation of postural muscles for controlling body sway. These results have clinical implications for understanding how patients with peripheral neuropathy may benefit from a cane for postural stability in stance. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Light touch; Peripheral neuropathy; Postural stability
1. Introduction The importance of somatosensory information from the feet to the control of upright stance balance is well established [1–4]. Delayed, distorted and/or absent somatosensory information from cutaneous, joint and muscle receptors underlie many of the postural abnormalities in individuals with distal sensory neuropathy [5 – 7]. The prevalence of distal symmetrical neuropathy in patients with long standing diabetes (DM-PN), is about 50% [8], and is related to instability during stance and gait [9], as well as to frequent fall accidents [10,11]. * Corresponding author. Tel.: + 972-3-640-9224; fax: +972-3-6409223. E-mail address: [email protected] (R. Dickstein).
Laboratory studies have demonstrated excessive postural sway in these patients [12], which is further accentuated by darkness or by unreliable visual conditions [13,14]. Postural responses to surface perturbations are also delayed with poor amplitude scaling in patients with peripheral neuropathy [15], but not in patients with profound loss of vestibular function [6], suggesting that somatosensory information is more important for postural responses than vestibular information. Clinicians attempt to improve postural stability of DM-PN patients by providing a walking cane [16]. Recent studies suggest that a cane may be useful in improving balance because it provides an alternative sensory reference and not because it provides an increased mechanical base of support [17,18]. It has been shown that light touch from a fingertip with forces
0966-6362/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 6 6 - 6 3 6 2 ( 0 1 ) 0 0 1 6 1 - 8
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
under 1 N that provide somatosensory cues but not mechanical support, attenuated postural sway similarly to heavy touch that allowed mechanical support in healthy subjects [18– 23]. Furthermore, reduction in sway due to light touch of a fingertip on a stable surface reduced postural sway even more than vision in healthy subjects and in subjects with bilateral loss of vestibular function [22]. Likewise, touch cues via the index finger were shown to be as effective as heavy force contact in providing stabilization of stance in blind subjects [23]. In healthy subjects, light touch from a fingertip also helped to suppress muscle vibration-induced postural sway due to aberrant proprioceptive information from the feet [24], suggesting that fingertip touch may be able to substitute for impaired or altered sensory information from the feet. The studies of Jeka and his colleagues further suggest that somatosensory information from a fingertip is used in a feedforward manner to trigger appropriate postural activity in leg muscles because changes in forces under the fingertip during light touch, but not heavy touch conditions, are phase-advanced over changes in center of pressure (CoP) under the feet [18,21]. The extent to which patients with distal peripheral neuropathy can use a similar strategy to improve postural stability in stance is unknown. Sensory neuropathy secondary to diabetes is discernable in the sensory nerves of the feet prior to the sensory and motor nerves of the hand [8,25]. Thus, it is possible that patients with diabetic neuropathy have enough somatosensory function in the hands to use it for improving balance in stance and gait. In fact, since subjects with somatosensory loss are more unstable than healthy subjects, they may benefit even more than control subjects from finger somatosensory feedback as a sensory substitution for diminished foot CoP information indicating body sway. The goal of the current study was to determine whether DM-PN subjects with profound loss of somatosensory information in the feet would be able to utilize light touch cues from a single fingertip for increasing stance stability. The specific contribution of the somatosensory input versus mechanical support was determined by comparing postural stability in stance conditions in three touch modes: ‘light touch’ (LT), ‘heavy touch’ (HT) and ‘no touch’ (NT). Postural stability during stance was challenged with eyes closed and/or by stance on a compliant, foam surface to determine the environmental conditions in which patients with DM-PN show the most instability and most help from fingertip touch. Forces under the fingertip were measured to determine whether patients with neuropathy could use light-touch feedforward mechanisms for control of posture similar to control subjects.
239
2. Methods
2.1. Subjects and clinical characteristics The subjects were eight adults (five men) suffering from Diabetes Mellitus (DM) with profound sensory peripheral neuropathy (DM-PN), with a mean age of 59.7910.7 years, and eight age-matched healthy volunteers (four men, mean age 60.69 9) who served as controls. All patients were diabetic for at least 4 years, the average duration being 21 years. Only completely independent community dwellers with no functional visual and/or motor deficits were accepted, and none of the patients used a cane in daily life. In all participants, normal function of the vestibular system was verified by sinusoidal horizontal rotation testing of the vestibulo-ocular reflex (VOR). All subjects in the patient and control groups had gain and phase measures of the VOR well within the 95% tile of normal laboratory limits. Specifically, VOR gain was 0.67 for the neuropathy group and 0.5 for the control group compared to B0.45 for the 95th percentile limit of normal gain. Somatosensory sensation of the right index finger was measured by clinical testing of proprioception and vibration sense at the distal interphalangeal (IP) joint, and by static two-point discrimination of the palmar aspect of the distal phalanx using the Mackinnon-Dellon disk criminator [26]. Only two of the DM-PN patients and none of the control subjects had reduced index finger somatosensory sensation. Note that for two-point discrimination, age-adjusted norms were applied; that is, 4 mm distance between the two points was established as upper normal limit [27,28]. The existence of sensory neuropathy in patients’ lower extremities was established by electrodiagnostic testing and by a thorough physical examination of the feet. Patients’ motor nerve (peroneal and posterior tibial nerves) conduction velocities were mostly within normal limits, whereas conduction of their sensory nerves was substantially impaired. Specifically, responses of the two medial and lateral plantar nerves were undetectable in all eight neuropathy subjects and sural nerve responses were absent in six of them. In the two patients with detectable sural nerve responses, the conduction velocity and amplitude of the sural nerve were significantly outside normal limits. The physical examination included routine testing of: ankle joint range of motion and strength of the dorsal and plantar flexor muscles; threshold of vibration sensation at the medial malleoli and the first metacarpophalangeal (MP) joint (128 Hz tuning fork); and testing of kinesthetic sensation at the ankle and the first MP joint. The results of patients’ physical examination are summarized in Table 1. Cutaneous sensation was further assessed by Semmes–Weinstein monofilaments
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
240
[29,30], at three dorsal and two plantar foot zones supplied by five cutaneous branches (The dorsal digital cutaneous, intermediate dorsal, and sural nerve on the dorsal side of the foot, and the medial and lateral plantar nerves on the plantar aspect). The testing protocol followed the interval comparison method [31]. Superficial sensation in each of the five zones was found to be significantly reduced in the DM-PN patients in comparison to the control subjects (P B 0.05; Mann– Whitney U test).
plate for support. In NT trials subjects’ right index finger was held above the touch plate and no contact with the plate was permitted. It was not difficult for the subjects to hold their finger above the touch plate and practice trials were not needed. Each trial was 40 s long. Data collection in each trial started after subjects assumed the required stance with only 5 cm distance between their medial malleoli and facing an art poster 1 m ahead of them. The circumference of each foot was marked on the surface, assuring stance at the same location on the plates in all trials. The same narrow base stance was kept also during foam standing trials. A safety harness was used to minimize fall risks.
2.2. Experimental protocol Subjects were tested while standing barefoot on a dual force platform surface, near a circular (11 cm diameter) touch plate mounted on a small AMTI force transducer. The touch plate was positioned at the subjects’ right side (all subjects were right-handed), 90 cm height from the floor, immediately anterior to the right foot. Each participant was tested in 12 sensory conditions that were randomly presented twice in two separate blocks of 12 trials. The conditions consisted of three touch modes, ‘heavy’ (HT), ‘light’ (LT) and ‘no’ (NT) touch, in each of two visual (eye open or closed) and two surface (firm and foam) conditions. The foam consisted of 7.5 cm, medium density Temper foam. During HT trials, subjects were allowed to use the touch plate for mechanical support by transmitting onto it as much force as they chose with their index finger. During LT trials, the touch plate was contacted by the right index finger pulp with a pressure not exceeding 1 N. An audio-feedback device attached to the touch plate informed subjects if the limit had been exceeded, thus preventing them from using the touch
2.3. Dependent 6ariables The data was sampled at 240 Hz and included the three dimensional forces measured under each foot, trunk antero-posterior (AP) and medio-lateral (ML) angular velocity and the three dimensional forces under the right index finger. Trunk velocity was measured by a Watson angular rate sensor, which was attached to the sternum. The root mean square (RMS) of the trunk velocity for each plane was used as a measure of trunk stability. Whole body AP and ML movements of the center of pressure (CoP) were derived from the dual force platform data. The RMS of CoP sway was calculated for the 40-s duration of each trial. The mean magnitude of forces transmitted via the index finger onto the touch-plate in the antero-posterior, medio-lateral and vertical axis were determined after the raw data were low passed filtered at 15 Hz, based on a residual analysis [32].
Table 1 Clinical data Patients Normal
Controls Reduced
Absent
Normal
Reduced
Absent
Ankle strength
16
–
–
16
–
–
Ankle ROM
16
–
–
16
–
–
Joint position Toe Ankle Rt index finger
– 10 6
15 6 2
1 – –
16 16 –
– – –
– – –
Vibration Toe Ankle Rt index finger
– 1 5
10 13 3
6 2 –
9 16 8
7 – –
– – –
Superficial sensation Dorsal foot Plantar foot
4 2
12 14
– –
16 16
– –
– –
Numbers are counts from both legs. For superficial sensation, reduced sensation was assigned to subjects whose Semmes–Weinstein mean feet scores fell beyond the upper 95% confidence interval of the corresponding mean of the control subjects.
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
241
Fig. 1. The CoP in a representative subject with peripheral neuropathy due to diabetes mellitus during stance under three fingertip touch conditions: no touch, light touch, and heavy touch. The subject stood on foam with eyes closed and showed decreased CoP sway with fingertip touch; heavy touch reduced sway more than light touch in this condition. A-P, antero-posterior; M-L, medio-lateral.
2.4. Analysis Descriptive methods and multiple analysis of variance (MANOVA) with one fixed (group) and three repeated measures factors (surface * vision * touch mode) were employed. Post-hoc testing compared the effects of LT with HT and of the effects of LT to NT. Temporal synchronization between AP and ML movements of the CoP and between corresponding fingertip forces, was determined by calculation of the cross-correlation between them. The correlation values were normalized by Fisher’s Z-transform method [33] and then subjected to a 2*2*2*2 (Group* Surface* Vision* Touch mode) MANOVA analysis. In addition, similar to the analysis used by Jeka et al. [21,22], the time point of the highest correlation value between each pair of CoP and finger force data was established within a time window of 9625 ms.
3. Results
3.1. CoP postural sway Both light touch and heavy touch of a fingertip significantly attenuated postural sway in both the DM-
PN subjects and the control subjects under all sensory conditions (F= 87.9, P B 0.0001 for AP CoP; F =27.4, PB 0.0001 for ML CoP). Fig. 1 shows examples of CoP sway from a representative neuropathy subject during with NT, LT, and HT in the condition resulting in the most sway for both groups, e.g. eyes closed standing on foam. Heavy touch had a larger effect than LT on reducing CoP sway in both the AP and ML directions (F= 65, PB 0.0001 for AP CoP, F=7.2, P= 0.01 for ML CoP). The difference between heavy touch and light touch was much less apparent, however, when subjects stood on the firm surface. The amount of reduction in both AP and ML CoP sway due to light and heavy touch was similar in the two groups. The effect of light touch was comparable to the effect of vision on reducing the CoP sway in both groups. The effect of touch on average AP and ML CoP RMS values under each condition in the two groups are presented in Fig. 2A and B. The effect of touch in the two DM-PN patients with reduced sensation in their fingertip was not different from the other six subjects. In general, ML but not AP CoP sway of the DM-PN patients was larger than sway of the healthy subjects (F= 4.6, P= 0.05 for ML CoP but P = 0.06 for AP CoP). Stance on a compliant surface or absence of
242
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
vision significantly increased CoP sway (F = 85.2 PB 0.0001 and F=32.3, PB 0.0001). A significant interaction between these effects (F =24.6, P B 0.0001 for AP CoP and F= 15.0, P = 0.001 for ML CoP) indicated that CoP sway was largest when subjects stood on the compliant surface with their eyes closed. The effect of touch on attenuating CoP sway was larger on a compliant than a firm surface (F =24.8, P B 0.0001 for AP CoP and F= 16.5, P B0.0001 for ML CoP) and was larger without vision than with vision (F = 2.7, P =0.08 for AP CoP and F = 7.9, P =0.001 for ML CoP).
3.2. Trunk 6elocity Like CoP, trunk velocity was significantly reduced by light touch and heavy touch in both patients with DM-PN and control subjects under all sensory conditions (F=11.6, PB 0.0001 for AP trunk velocity and F =9.8, P =0.0006 for ML trunk velocity). Fig. 3 illustrates the trunk velocity of one patient and one control subject in the most challenging condition (eyes closed, standing on foam). Unlike CoP sway, light touch attenuated AP and ML trunk RMS angular velocity at a similar magnitude to the attenuation caused by heavy touch for all visual and surface condi-
tions in both patients and control subjects (P \ 0.05 for differences between conditions of LT and HT). The effects of touch on the average AP and ML trunk velocity RMS for each condition in the two groups is summarized in Fig. 4. Like CoP sway, the amount of reduction of trunk velocity due to touch was similar in the control subjects and DM-PN subjects. Furthermore, like CoP sway, the effect of light touch in reducing trunk velocity was not statistically different from the effect of vision for both groups. During stance on a firm surface, trunk angular RMS velocity was comparable in the two groups. Stance on the foam was associated with substantially a higher trunk RMS velocity in the DM-PN patients than in the control subjects with this difference larger in the ML direction (F= 3.8, P= 0.06 and F= 4.7, P= 0.04 for the AP and ML directions respectively). Mean AP trunk RMS velocity was larger than ML trunk velocity (range between 0.57 and 1.36 deg/s for AP and between 0.37 and 1.14 deg/s for ML trunk RMS velocity). Closing the eyes and standing on foam significantly increased trunk velocity RMS for both the AP and ML directions (F=41.06 PB 0.0001, F= 26.7 PB 0.0001 for AP velocity and F= 63.4 PB 0.0001, F=24.3 PB 0.0001 for ML velocity).
Fig. 2. Effects of light touch and heavy touch on mean ( 9 S.E.) antero-posterior (A) and medio-lateral (B) CoP RMS. Subjects stood with feet close together under four sensory conditions: eyes open or closed and on a firm or foam surface. Black bars indicate DM-PN group data and white bars indicate control group data. N, no touch; L, light touch; and H, heavy touch.
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
243
Fig. 3. Medio-lateral (A) and antero-posterior (B) angular velocity of a DM-PN patient and of a control subject (C and D respectively), during stance on foam surface with eyes closed. (presented are 5 – 6 s in the middle of a 40-s long trial). X-axis tick marks denote 1 s.
3.3. Fingertip forces Patients with somatosensory loss due to DM-PN exerted similar forces from their index finger onto the touch plate as the control subjects. Mean vertical and lateral forces across all LT and HT conditions are given in Table 2. The only significant difference between groups was in the laterally directed forces: DM-PN patients employed a larger rightward-directed force (toward the touch plate) than the controls (Table 2). For all subjects, standing on foam was associated with significantly larger vertical fingertip force than standing on a firm surface (F =12.9, P =0.003). The horizontal forces during HT conditions were directed forward and rightward. AP forces were neither affected by group nor by vision or by surface condition. However, ML forces during HT conditions and foam stance were associated with a greater amount of force directed toward the right side as compared with stance on a firm surface. The forces applied onto the touch bar during LT conditions by the two patients who had reduced finger sensation (as evident from the two-point discrimination test), were not larger than the forces applied by the other patients.
3.4. Correlation between AP CoP and AP fingertip forces The correlation between the AP forces exerted onto the touch plate and AP CoP movements under subjects’ feet indicated that a forward movement of the CoP was associated with a backward directed force onto the touch plate and vice-versa. Mean correlation range was |= 0.6–0.8 and 0.3–0.6 (for controls and patients, respectively). Correlations between touch plate forces and CoP were significantly larger during LT than HT (F= 10.6, P= 0.005), and during foam stance than on a firm surface (F= 4.6, P= 0.04). The highest correlation values were obtained when changes in the CoP lagged behind the changes in fingertip force exerted onto the touch plate. Mean time lag between AP forces on the touch plate and AP CoP ranged from 95 to 342 ms. The time lag was significantly longer during LT than during HT trials (F= 9.8, P= 0.006). Correlation values between ML CoP movements and ML forces exerted onto the touch plate were low during stance on the firm surface and became significantly higher during stance on the foam (F= 29.4, PB
244
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
0.0001). This surface-dependent increase in correlation between ML CoP and fingertip force was larger in the patients than control subjects (F = 6.1, P = 0.02). Correlation values during stance on the foam ranged between 0.4 and 0.6, with a positive association between leftward movement of the CoP and a rightward force from the fingertip on the plate. Time lags between the ML movements of the CoP and ML forces exerted onto the plate were very small during stance on a firm surface and became substantially larger when standing on the foam. Nevertheless, in both surface conditions during LT trials, changes in the CoP lagged behind changes in the ML force on the touch plate (time lags values between 68 and 293 ms), while during HT these time lags were negligible.
4. Discussion Trunk and CoP sway of the DM-PN patients, as well as the healthy subjects, decreased with light and heavy fingertip touch. The effects of touch became more apparent in conditions of deficient surface and/or visual information in both subject groups. The effectiveness of touch to decrease sway was larger in subjects with neuropathy than in control subjects, especially in the most difficult conditions. When standing on foam, subjects with neuropathy had larger than normal CoP and
trunk sway with no touch, but similar sway as controls with light and heavy touch. The larger effect of touch in the neuropathy than control group might be expected because neuropathy subjects may need to substitute fingertip touch for deficient foot somatosensory information in order to reduce their larger than normal postural instability. Thus, the results support a compensatory increase in sensitivity or ‘weighting’ of fingertip somatosensory information for controlling excessive postural sway in subjects with chronic somatosensory loss in the feet. In fact, the influence of light fingertip touch on reducing postural sway becomes larger when either control or neuropathy subjects are more unstable in conditions of standing on foam and without vision. Despite the likelihood of decreased somatosensory information in the fingertip of patients with peripheral neuropathy, they were able to take advantage of additional feedback to reduce postural sway. It is not possible to determine whether subjects are using cutaneous light pressure information or joint and muscle proprioceptive information to attenuate postural sway with light touch [18], nor to determine accurately which of these sensory pathways are adequate in the patients. Nevertheless, these results show that subjects with neuropathy in their feet and legs sway more than normal and functionally benefit significantly from light fingertip touch.
Fig. 4. Group means ( 9S.E.) of antero-posterior (A) and medio-lateral (B) RMS angular trunk velocity under the same sensory conditions as Fig. 2.
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
245
Table 2 Mean vertical and lateral forces (N) applied by the index finger to the touch plate. Vertical (positive is downward) Light touch Patients Controls Group effect Surface effect
Lateral (positive is rightward) Heavy touch
0.27 9 0.02 12.3 912.9 0.269 0.03 6.5 9 3.5 NS Vertical forces are larger during foam stance F= 12.9, P =0.003
In contrast to former findings [18,19] light touch in the current study was not as effective as heavy touch in decreasing CoP sway, but had an intermediate influence between the effects of ‘heavy’ and ‘no touch’. These results stem from the challenging stance conditions imposed on our subjects since the reduced effect of light touch compared to heavy touch was especially obvious during stance on the foam surface. The instability in subjects with somatosensory loss was further increased when they stood on the foam surface. This increased instability suggests either that the foam surface removed some surface somatosensory information that neuropathy patients were using on firm surface or that the foam produced additional mechanical instability. Standing on foam also increased instability in the control subjects such that the CoP and trunk velocity was at least as large, or larger, than for patients with neuropathy on the firm surface. The effectiveness of fingertip touch as a sensory substitution for somatosensory loss in the feet can be seen by the similar level of postural sway (both CoP and trunk velocity) between the conditions of light, and especially heavy fingertip touch when standing on foam, and the condition of no touch when standing on the firm surface. Regardless of the amount of excessive sway in neuropathy patients, fingertip touch reduced sway so it was similar to normal subjects using touch. These results suggest that touch doesn’t add a consistent amount or consistent percent of stability but rather touch allows a specific level of stability in both neuropathy patients and control subjects. While prior reports have documented an increased CoP sway in diabetic patients with peripheral neuropathy [7,34,35], the AP CoP sway amplitude of the current patients was comparable to that of the healthy subjects, although ML sway on a firm surface was somewhat larger. It is possible that the intactness of the motor nerves supplying the leg muscles contributed to the small loss CoP sway amplitude in the patients despite their significant loss of somatosensory information. Alternatively, the absence of excessive CoP sway of the patients could result from an a-priori avoidance from relying on sensory information from their feet for balance tasks. Perhaps, their prolonged inability to rely
Light touch
Heavy touch
0.03 0.98 0.008 0.06 Patients exert larger rightward forces F=5.5, P =0.03 Rightward forces are larger during foam stance F =9.8, P =0.0007
on surface inputs through their feet habituated them to depend more on vestibular, visual and/or proximal somatosensory inputs, and to execute postural corrections and adaptations at higher body segments without excessively moving the COP under the feet [6,36]. Reliance on vestibular information could be the preferred strategy for substitution of the deficient surface information. This suggestion is supported by the observations that patients with profound sensory neuropathy can preserve their balance on a CoP sway-referenced surfaces as successfully as healthy subjects, provided that their vestibular system is intact [37]. On the other hand, their postural stability is adversely affected by standing on a surface that is sway-referenced to the center of mass (CoM), suggesting, for these subjects, CoM information is more critical for stance stability than the CoP [37]. The fact that ML CoP sway during foam stance was accentuated in the patients with neuropathy is explainable by the distorted foot sensory information, combined with the relatively narrow stance base in this study. Other studies have reported increased lateral sway in the elderly compared to the young, especially in elderly subjects with a history of falls. This increased ML sway suggests either that those elderly subjects may, in fact, have posturally significant somatosensory loss or that ML sway tends to increase prior to AP sway in many types of increased instability. In addition, the fact that the touch plate was positioned laterally, may have encouraged the DM-PN subjects in our study to adjust their balance in this dimension. The fact that patients transferred more force onto the plate than the healthy subjects during the ‘heavy touch’ conditions, is indicative of their greater sense of instability compared with the healthy controls, and of a subsequent greater reliance on the touch plate, when this option was available. Trunk velocity was also higher in the patients than in the healthy controls, mainly when standing on foam. Thus, the additive influence of disruptive surface somatosensory information [38] and the deficient foot sensory information due to neuropathy led patients to use excessive trunk movements in both planes for balancing. This excessive trunk velocity probably reflects
246
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247
increased reliance on a hip strategy for postural control in patients with somatosensory loss as reported previously. The effect of light touch on trunk velocity was impressive, being comparable to that of heavy touch for both the AP and ML directions. In contrast to its low potential to decrease CoP excursion, light touch was as effective as heavy touch in reducing trunk velocity. Because normal regulation of the CoM position is dependent on continuous monitoring of the CoP position [39], which is impaired in the DM-PN patients, the contribution of external touch cues via the upper extremity appears to be of great importance. Specifically, somatosensory information via the arm is useful as an orientation reference for perceiving the alignment of the body with respect to the environment, facilitating responses for the control of an upright trunk [18–20]. Neuropathy subjects appear capable of using an ankle, instead of hip, strategy to control postural sway when using fingertip touch. The arm in contact with the touch plate is in a better biomechanical position than the foot for sensing postural instability, thus enabling faster responses to correct trunk sway [19,40]. Our results support the application of a feedforward ‘sensory-motor strategy’ that uses somatosensory information from the finger to result in changes in forces under the feet to reduce sway. This feedforward strategy reduces body sway both in the frontal and sagittal planes for both control and neuropathy subjects. Time lags in the order of 300 ms were measured between fingertip forces during light touch and subsequent CoP movements for both subject groups similar to previous studies with healthy subjects and subjects with vestibular or visual loss [20,41]. Thus, any slowing of nerve conduction in the limbs of neuropathy subjects was not sufficient to disable this fast, feedforward mechanism. In conclusion, for both DM-PN patients and healthy subjects, fingertip somatosensory inputs through light touch were as effective as heavy touch in attenuating trunk velocity but less effective in decreasing CoP sway. Since patients with peripheral neuropathy showed larger trunk velocity than healthy subjects, the findings suggest an important functional role for LT in controlling stance balance in this subject group. Additionally, the application of a feedforward mechanism, in which fingertip inputs trigger the activation of postural muscles for controlling body sway, have clinical implications for understanding how patients with peripheral neuropathy may benefit from a cane for postural stability in stance.
References [1] Kavounoudias A, Roll R, Roll JP. The plantar sole is a ‘dynamometric map’ for human balance control. Neuroreport 1999;9:3247 – 52.
[2] Horak FB, Macpherson JM. Postural orientation and equilibrium. In: Smith JL, editor. Handbook of Physiology: Section 12: Exercise Regulation and Integration of Multiple Systems. New York: Oxford University Press, 1996:255 – 92. [3] Wu G, Chiang JH. The significance of somatosensory stimulations to the human foot in the control of postural reflexes. Exp Brain Res 1997;114:163 – 9. [4] Mauritz KH, Dietz V, Haller M. Balancing as a clinical test in the differential diagnosis of sensory-motor disorders. J Neurol Neurosurg Psychiatry 1980;43:407 – 12. [5] Thoumie P, Do MC. Changes in motor activity and biomechanics during balance recovery following cutaneous and muscular deafferentation. Exp Brain Res 1996;110:289 – 98. [6] Horak FB, Nashner LM, Diener HC. Postural strategies associated with somatosensory and vestibular loss. Exp Brain Res 1990;82:167 – 77. [7] Semoneau GG, Ulbrecht JS, Derr JA, Becker MB, Cavanagh PR. Postural instability in patients with diabetic sensory neuropathy. Diabetes Care 1994;12:1411 – 21. [8] Green DA, Sima AAF, Albers JW, Pfeifer MA. Diabetic neuropathy. In: Rifkin H, Porte D, editors. Diabetes Mellitus. Theory and Practice. New York: Elsevier, 1997:710 – 53. [9] Courtemanche R, Teasdale N, Boucher P, Fleury M, Lajoie Y, Bard C. Gait problems in diabetic neuropathy patients. Arch Phys Med Rehab 1996;7:849 – 55. [10] Richardson JK, Ashton-Miller JA. Peripheral neuropathy: an often-overlooked cause of falls in elderly. Postgrad Med 1996;6:161 – 72. [11] Cavanagh PR, Derr JA, Ulbrecht JS, Maser RE, Orchard TJ. Problems with gait and posture in neuropathic patients with insulin-dependent diabetes mellitus. Diabetic Med 1992;9:469 – 74. [12] Uccioli L, Giacomini PG, Monticone G, Magrini A, Durola L, Bruno E, Parisi L, Menzinger G. Body sway in diabetic neuropathy. Diabetes Care 1995;18:339 – 44. [13] Girolamo S, Menzinger G. Body sway in diabetic neuropathy. Diabetes Care 1995;18:339 – 44. [14] Ledin T, Odkvist LM, Vrethem M, Moller C. Dynamic posturography in assessment of polyneuropathic disease. J Vestibular Res 1991;1:123 – 8. [15] Inglis JT, Horak FB, Shupert CL, Rycewicz-Jones C. The importance of somatosensory information in triggering and scaling automatic postural responses in humans. Exp Brain Res 1994;101:159 – 64. [16] Ashton-Miller JA, Yeh MW, Richardson JK, Galloway T. A cane reduces loss of balance in patients with peripheral neuropathy: results from a challenging unipedal balance test. Arch Phys Med Rehab 1996;77:446 – 52. [17] Jeka JJ. Light touch contact as a balance aid. Phys Ther 1997;5:476 – 87. [18] Jeka JJ, Lackner JR. Fingertip contact influences human postural control. Exp Brain Res 1994;100:495 – 502. [19] Holden M, Ventura J, Lackner JR. Stabilization of posture by precision contact of the index finger. J Vestibular Res 1994;4:285 – 301. [20] Jeka JJ, Lackner JR. The role of haptic cues from rough and slippery surfaces in human postural control. Exp Brain Res 1995;103:267 – 76. [21] Jeka J, Oie K, Schoner G, Dijkstra T, Henson E. Position and velocity coupling of postural sway to somatosensory drive. J Neurophys 1998;79:1661 – 74. [22] Lackner JR, Dizio P, Jeka J, Horak F, Krebs D, Rabin E. Precision contact of the fingertip produces postural sway of individuals with bilateral vestibular los. Exp Brain Res 1999;126:459 – 66. [23] Jeka JJ, Easton RD, Bentzen BL, Lackner JR. Haptic cues for orientation and postural control in sighted and blind individuals. Percept Psychophys 1996;58:409 – 23.
R. Dickstein et al. / Gait and Posture 14 (2001) 238–247 [24] DiZio P, Lackner JR, Rabin E. Sensorimotor and environmental conditions constraining use of the hand as a guide for posture. In: Proceedings of the 13th International Symposium: Multisensory Control of Posture & Gait, Paris, June, 1997. [25] Porte D, Sherwin RS. Diabetes Mellitus. Stanford CT: Appelton & Lange, 1997. [26] Mackinnon SE, Dellon AL. Two point discrimination tester. J Hand Surg Am 1985;1:906 –7. [27] Desroisiers J, Hebert R, Bravo G, Dutil E. Hand sensibility of healthy older people. JAGS 1996;44:974 –8. [28] Louis DS, Green TL, Jacobson KE, Rasmussen C, Kolowich P, Goldstein SA. Evaluation of normal values for stationary and moving two point discrimination in the hand. J Hand Surg 1984;9A:552 – 5. [29] Bell JA. Semmes – Weinstein monofilaments testing for determining cutaneous light touch/deep pressure sensation. The Star/ Carville, November – December, (from ‘ Directions for using Semmes– Weinstein filaments, 1984. [30] Mueller MJ. Identifying patients with diabetes mellitus who are at risk for lower-extremity complications. Use of Semmes – Weinstein Monofilaments. Phys Ther 1996;76:68 – 71. [31] Holewski JJ, Stess RM, Graf PM, Grunfeld C. Aesthesiometry: quantification of cutaneous pressure sensation in diabetic peripheral neuropathy. J Rehab Res Dev 1988;25:1 –10. [32] Winter DA. Biomechanics and Motor Control of Human Movement. New York: Wiley, 1990:36 –45.
247
[33] Snedecor GW, Cochran WG. Statistical Methods. Iowa State University, 1987:186. [34] Wu G, Zhao W. The role of mechanoreceptive information in the stability of human upright posture. A theoretical consideration. Motor Control 1997;1:3 – 19. [35] Boucher P, Teasdale N, Courtemanche R, Chantal B, Fleury M. Postural stability in diabetic polyneuropathy. Diabetes Care 1995;18:638 – 44. [36] Nutt JG, Horak FB. Gait and balance disorders. In: Watts R, Koller W, editors. Movement Disorders. Neurologic Principles and Practice. New York: McGraw-Hill, 1995:51. [37] Shupert CL, Dickstein R, Horak FB, Peterka R, Jones C. Surface sway-referencing and peripheral neuropathy disrupt somatosensory information for postural stability in stance. Gait Posture, 9 (Suppl. 1):S51. [38] Chiang JH, Wu G. The influence of foam surface on somatosensory receptors in the human foot — a rotational platform study. Gait Posture 1996;4:122 – 9. [39] Winter DA. Biomechanics and Motor Control of Human Movement. New York: Wiley, 1990:93 – 6. [40] Jeka JJ, Schoner G, Dijkstra T, Ribeiro P, Lackner JR. Coupling of fingertip somatosensory information to head and body sway. Exp Brain Res 1997;113:475 – 83. [41] Clapp SC, Wing AM. Light touch contribution to balance in normal bipedal stance. Exp Brain Res 1999;125:521 – 4.