Diabetes Care 27:173-178, 2004
© 2004 by the American Diabetes Association, Inc.
Pathophysiology/Complications
Original Article |
Postural Control Mechanisms During Quiet Standing in Patients With Diabetic Sensory Neuropathy
Danik Lafond, PHD1,2,
Hélène Corriveau, PT, PHD3 and
Francois Prince, PHD1,2,4
1 University of Montreal, Department of Kinesiology, Montreal, Quebec, Canada
2 Marie-Enfant Rehabilitation Center, Movement Laboratory, Montreal, Quebec, Canada
3 Sherbrooke Geriatric University Institute, Ageing Research Center of Sherbrooke, Sherbrooke, Quebec, Canada
4 University of Montreal, Department of Surgery, Montreal, Quebec, Canada
Address correspondence and reprint requests to Francois Prince, PhD, University of Montreal, Department of Kinesiology & Surgery, C.P. 6128, Succursale Centre-ville, Montreal, Quebec H3C 3J7, Canada. E-mail: francois.prince{at}umontreal.ca
 |
ABSTRACT
|
|---|
OBJECTIVE—The objective of the present study was to compare postural mechanisms identified by using dual force platform in healthy elderly community-dwelling subjects and diabetic sensory neuropathy (DSN) patients under different visual conditions.
RESEARCH DESIGN AND METHODS—The presence and the severity of the sensory neuropathy was evaluated with a clinical scale. Postural mechanisms and motor strategies of the ankle and hip joints were quantified by testing subjects in quiet stance on a dual force platform under two visual conditions (eyes open and eyes closed). Root mean square (RMS) values of the center of pressure (COP) time-varying signals and normalized cross-correlation function were used to estimate the contribution and the interdependence of postural control mechanisms.
RESULTS—DSN patients show larger RMS values of the COPnet displacement in both anteroposterior and mediolateral (ML) directions. Motor strategies at the ankle joints are altered in DSN patients compared with healthy elderly subjects particularly in the ML direction.
CONCLUSIONS—This experiment is the first to highlight that even with vision, postural mechanisms at the ankle joints are impaired in DSN patients during quiet standing. Our results point out the importance of focusing on postural control instability in ML of DSN patients.
Abbreviations: AP, anteroposterior • COP, center of pressure • COPc, ankle control mechanism • COPv, hip control mechanism • COPr, COP under the right foot • COPl, COP under the left foot • DSN, distal sensory neuropathy • EC, eyes closed • EO, eyes open • ML, mediolateral • RMS, root mean square
|
INTRODUCTION
|
|---|
The postural control system is a complex organization that controls the orientation and the equilibrium of the body during an upright stance (1,2). Past research provides many insights that the postural control system involves sensory afferences integration instead of eliciting reflex responses. Sources of sensory afferences that are viewed to contribute to postural control are 1) vestibular, 2) visual, and 3) proprioceptive. Postural instability in diabetic sensory neuropathy (DSN) patients is usually attributed to the lack of accurate proprioceptive feedback (sensory ataxia) from the lower limbs (3–5). The prevalence of sensory ataxia in diabetic patients ranges from 10 to 90% depending on screening protocol and criteria to define neuropathy (6). Compared with non-neuropathic diabetic patients, DSN patients self-reported 15 times more injuries during gait and reported feeling less safe during unusual situations (7). Peripheral neuropathy has been significantly associated with falling and repetitive falls (8). Evaluation of postural steadiness is usually based on the interpretation of center of pressure (COP) measures using a force platform (9). Only few studies have addressed the problem of postural instability during quiet standing in DSN (3–5,10). High correlations were found between the severity of the neuropathy and the COP measures (5,10,11) but not with diabetes per se (3).
Past reports on postural steadiness of DSN patients only quantify the net COP changes in anteroposterior (AP) and mediolateral (ML) directions. During feet side-side configuration, postural mechanisms have been reported (12) to be under ankle mechanism in AP direction, whereas hip abductors/adductors motor activities are associated with ML control. The objective of the present study is to compare postural mechanisms by using dual force platform in healthy elderly subjects and DSN patients. Ankle and hip postural strategies in both AP and ML directions will be investigated and compared.
|
RESEARCH DESIGN AND METHODS
|
|---|
Eleven elderly patients with type 2 diabetes with DSN and 20 healthy elderly subjects were studied (Table 1). Subjects were age matched because postural instability has been related to age (5,13). The DSN patients were recruited from a diabetes clinic. The DSN patients were not selected based on their complaints of instability. The severity of polyneuropathy was quantified using a scoring system developed by Valk et al. (14). Briefly, a total score of 0 is graded as no polyneuropathy, 1–9 as mild, 10–18 as moderate, and 18–33 as severe polyneuropathy. Valk et al. (15) have shown that this clinical evaluation has a good intraexaminer reliability and a good sensibility and specificity. Five DSN subjects were classified as mild, four as moderate, and two as having severe polyneuropathy. The eligibility criteria for a healthy subject consisted of being  60 years of age, living independently in the community, and having no neurological or musculoskeletal impairments. Healthy subjects were excluded if they reported visual, somatosensory impairments, vestibular dysfunction, or if they reported at least one fall in the past 6 months. Finally, the healthy subject had to have a fasting blood glucose level <7.8 mmol/l and HbAlc <11.1 mmol/l as well as a score lower than 2 on the Valk scale to be eligible for the study. The clinical evaluation has been described in details elsewhere (10). None of the subjects were using tricyclic antidepressants or antipsychotic drugs. Furthermore, none of the subjects had a distal acuity score below 20/16 on the Snellen card (16).
COP analysis and postural control mechanisms definition
When a subject stands erect in double stance on one force platform, the COP is calculated as follows:
 | (Eq. 1) |
in which M is the moment, F is the reaction force, x, y, and z are, respectively, the ML, AP, and vertical direction, and X0, Y0, and Z0 are the offsets from the geometric center of the force platform.
However, in a double stance on two adjacent force platforms, the COPnet is determined in AP and ML directions in respect to Eq. 2. The COPnet(t) is, therefore, the weighted sum of the time-varying position of the COP from each force platform.
 | (Eq. 2) |
in which COPl and COPr are the COP coordinates under the left limb and the right limb, respectively. The following ratio is the time-varying signal representing the relative vertical load (V) under each foot.
 | (Eq. 3) |
This ratio is 0.5 when the body weight is equally distributed under each foot.
If the two limbs do not have exactly the same loading on each force plate, Eq. 2 will be modified to reflect the average loading of each limb during the evaluation period.
 | (Eq. 4) |
The ratios  and  represent the average vertical loads under each foot. Because the = 1 - in quiet standing, the COPc(t) is viewed as the motor contribution of ankle plantarflexors/dorsiflexors and ankle evertors/invertors in the AP and ML directions, respectively (12). Now if the COPc(t) is removed from the COPnet(t), the contribution of the load/unload mechanism, COPv(t), which is reflected by the fluctuations of V, can be estimated as follows:
 | (Eq. 5) |
Biomechanical analysis showed that the COPv(t) is attributed to the motor contribution of the hip adductors/abductors (12). Therefore, the COPnet is the summation of two mechanisms, one relative to the control of ankle musculature (COPc) and the other attributed to the control of the hip musculature (COPv).
Data collection
Subjects were instructed to stand as still as possible in an upright posture on two adjacent force platforms. Data acquisition was made in a double leg stance with feet at pelvic width. Tracings were done of foot placement, and subjects were required to remain within these tracings for all trials to ensure that feet position remained constant. Under the "eyes open" (EO) condition, subjects were instructed to look straight ahead with their head straight and their arms hanging at their sides in a comfortable position. Four successive trials with eyes open followed by four successive trials with eyes closed (EC) lasting 120 s were collected with a resting period of  5 min between trials. Ground reaction forces and moments were acquired by two AMTI force platforms (Model OR6–5; Advance Mechanical Technology, Waterton, MA). Analog signals were sampled at a frequency of 20 Hz with an A/D converter and recorded on a Pentium computer. Force platforms always allowed the temperature to stabilize for at least 45 min before any data collection to minimize any electronic drifts. The data collected were thereafter transformed (in N and N x m) by multiplying the data array by full calibration matrix to ensure true values of reaction forces and moments. The force platform signals were filtered with a zero lag sixth order Butterworth low pass filter at 10 Hz. The accuracy of the COP measured in the present study is 0.2 mm (10).
Data analysis
A measurement of the average contribution of the COPc(t) and the COPv(t) to the COPnet(t) was calculated for each trial from the root mean square (RMS) as follows:
 | (Eq. 6) |
in which N is the number of sample data.
The dependence of between the COPl(t) and COPr(t) as well as the dependence of the two separates motor mechanisms [COPc(t) and the COPv(t)] on the COPnet(t) was estimated by a normalized cross-correlation at zero lag to obtain results lying between +1 and -1. The "cross-correlation" measures of the similarity between two different datasets at different time lags. In the present study, we calculated the cross-correlation at zero lag because we are interested in postural mechanisms contributions at the same point in time. If the signals are identical, then the correlation coefficient is 1; if they are totally different, the correlation coefficient is 0; and if they are identical except that trends are moving in the opposite directions (phase is shifted by exactly 180°), then the correlation coefficient is -1. A strong correlation indicates that the information contained in the two COP time series overlap substantially and indicates a dependence of COP time series in this study. All COP time series were calculated and analyzed with MATLAB 5.1 (Mathworks, Natick, MA). The ANOVA and the Student’s t test were used to compare healthy subjects with DSN subjects. Statistical significance was assumed at P < 0.05 (two-tailed).
|
RESULTS
|
|---|
Comparisons of subject characteristics for the two groups are presented in Table 1. There were significant differences between the two groups in all sensory tests and the Tinetti mobility scale. The COP time-varying signals of one representative trial of 10 s are presented (Fig. 1). Dual force plates first allow quantifying the relative load under each foot. The left and right vertical ground reaction forces are perfectly out of phase as depicted in Fig. 1A. When one limb is unloading, the contralateral limb is loaded proportionally. In AP direction, the COPnet trajectory is approximately at the mid-distance between COPl and COPr trajectories (Fig. 1B). However, in the ML direction, the COPnet trajectory is not in relation to either COPl or COPr (Fig. 1C). The COPnet oscillations are considerably larger then either COPl or COPr. In ML direction, the COPv trajectory is in good agreement to the COPnet, whereas the contribution of the COPc is negligible (Fig. 1D). In contrast, the COPc trajectory is close to the COPnet and the amplitude of the COPv, contributing very little to the COPnet in the AP direction (Fig. 1E).
View larger version (31K):
[in this window]
[in a new window]
|
Figure 1— A: Right (black line) and left (gray line) vertical forces expressed in fraction of body weight. B: Right COP (black line), left COP (gray line), and COPnet (dashed line) time series in AP direction (in millimeters). C: Right COP (black line), left COP (gray line), and COPnet (dashed line) time series in ML direction (in millimeters). D: COPc (black line), COPv (gray line), and COPnet (dashed line) time series in AP direction (in millimeters). E: COPc (black line), COPv (gray line), and COPnet (dashed line) time series in ML direction (in millimeters).
|
|
The contribution of the COPc and COPv to the COPnet displacement was quantified using the RMS values of each of these time-varying signals (Table 2). The RMS of the COPnet is significantly greater for the DSN compared with the healthy elderly subjects in both AP and ML directions. The mean RMS values during EO conditions were 1.97 and 3.58 mm for the healthy group, whereas the DSN groups present a higher value with a mean of 2.77 and 4.91 mm in ML and AP, respectively. For the healthy group, the EC condition significantly increases the COPc, COPv, and COPnet in AP direction. In ML, the RMS values of the COPv and the COPnet were increased significantly (P < 0.05) during the EC compared with the EO conditions but not the COPc. It should be noted, however, that the COPc in ML and the COPv in AP have limited contribution to the COPnet. DSN patients show a significant increase of RMS values of the COPv and the COPnet during EC condition in the AP direction only. When expressed in percentage of COPnet (%COPnet), there is no difference between healthy subjects and DSN patients for the COPc and COPv in both AP and ML directions. There is an effect of vision where the EC condition reduces the relative contribution of the COPc in ML and the COPv in AP for the healthy group.
The dependence or the independence between COP time-varying signals was quantified by the normalized cross-correlation function (Table 3). In general, there is a very strong and significant correlation between COPc and COPnet and between COPv and COPnet in AP and ML, respectively. A relatively high and significant correlation between COPc and COPv (mean ± SD, 0.64 ± 0.19) and between COPc and COPnet (0.77 ± 0.14) is found for the healthy group in ML. This high dependence is significantly decreased in DSN patients between COPc and COPv (-0.11 ± 0.39) and between COPc and COPnet (0.53 ± 0.22) in ML. In AP, there is a strong and significant dependence between COPl and COPr for the healthy subjects (0.77 ± 0.13) and the DSN patients (0.81 ± 0.06). The relation between COPv and COPnet is high for the healthy subjects (-0.60 ± 0.52) and DSN patients (-0.54 ± 0.63) during EC condition. This relation is significantly decreased in DSN patients (-0.17 ± 0.62) when compared with healthy elderly individuals (-0.62 ± 0.56) during EC. For the healthy group, the EC condition decreases significantly the dependence between COPc and COPv and between COPc and COPnet in the ML direction. However, in the AP direction, only the relation between COPl and COPr is affected by the vision conditions. In DSN patients, the dependence between COPc and COPv was affected by vision in the ML direction. In AP, the dependence between COPl and COPr is significantly increased, and the relation between the COPv and COPnet is affected significantly in AP.
|
CONCLUSIONS
|
|---|
Several studies have demonstrated, using summary COP measures, that DSN patients have a poor postural control during quiet standing compared with healthy elderly individuals. DSN showed larger sway area (3,17,18), larger speed of sway (3,17,18), larger COP range (3,5), higher RMS values of the COP-COM variable (10), and an increase in the power of medium-high frequency band (4,17) of a power spectral analysis. Our results confirm the findings of previous research regarding postural instability of DSN patients in quiet standing. The present study showed RMS values of COPnet in DSN patient during EO that are comparable with the EC results of age-matched healthy elderly subjects as found previously (3). These results highlighted that DSN patients, not selected for any complaint of instability a priori, have a poor postural control that may lead them to a higher risk of falling.
The COPnet is equivalent of the COP measured from one force platform. However, the contributions of the individual limbs were not possible until two platforms were used to identify independent postural mechanisms. The COPc and the COPv represent the postural control at the ankle and hip, respectively. Therefore, the main objective of this study was to compare DSN patients with healthy elderly subjects using postural control mechanisms developed and measured with dual force platforms. The ankle and hip postural mechanisms that have different contributions in AP and ML directions have been calculated. In ML, the COPc has only a negligible effect as shown by low RMS values in the two groups. However, the COPc contributes almost totally to the COPnet in AP. This dominance in AP has been attributed by biomechanical inverse dynamics to the activation of the ankle plantar flexors/dorsiflexors motor control (12). It is not surprising if one considers the axes of rotation of the ankle joints and their symmetrical alignment during quiet standing. Furthermore, the COPr and COPl trajectories vary synergistically in AP. This observation is supported by high cross-correlation coefficients between COPr and COPl in both healthy subjects and DSN patients. In addition, the COPnet is perfectly in phase and dependant of the COPc in AP. In contrast, the COPv in ML contributes to over 90% of the COPnet, whereas in AP, its contribution is only 15–20%. This loading/unloading mechanism dominant in ML has been attributed to the control of the hip adductors/abductors (12). Evidence of hip adductors/abductors motor control in quiet standing in ML direction is supported by the dependence of the COPr and COPl trajectories and by the almost perfect correlation between COPv and COPnet. A closer look on the relation between COPr and COPl in healthy elderly subjects reported out of phase changes with an average cross-correlation coefficient of -0.55. This result is in agreement with those reported in young healthy adults of approximately -0.68 (12). This means that the COPr tends to cancel out the COPl; when the COPr trajectory changes medially by right ankle eversion motor activity, there is a similar left ankle eversion motor activity tending to change medially the COPl trajectory. However, in DSN patients, this relation is decreased, and low cross-correlation coefficients have been attempted with higher variability. In ML, the relation between COPc and COPv is positively correlated with very low variability in healthy elderly subjects. DSN patients have obtained a significantly lower correlation between COPc and COPv lying around zero with a relatively high standard deviation. Furthermore, the relation between COPc and COPnet is significantly lower in DSN patient compared with healthy elderly individuals. Our results showed that left and right evertor/invertor motor activities are not as well matched in DSN patients as seen in age-matched healthy subjects. This study is the first to demonstrate that ankle motor activities are affected in DSN patients during quiet standing.
The effect of vision was significant in the RMS values of all COP time-varying signals in both AP and ML directions for the healthy group. However, only RMS values of COPc and COPnet in AP were affected by vision deprivation in the DSN group. The dependence or independence of the COP time-varying signals were not differently affected by vision compared with healthy elderly subjects in ML. These results supported the findings previously discussed that postural control mechanisms are affected in ML even with vision in DSN subjects. Our results point out the importance of focusing on postural control instability in ML of DSN patients.
|
Footnotes
|
|---|
A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.
Received for publication March 18, 2003.
Accepted for publication September 17, 2003.
|
References
|
|---|
- Macpherson JM, Fung J, Jacobs R: Postural orientation, equilibrium, and the spinal cord. Adv Neurol 72:227–232, 1997[Medline]
- Horak FB, Macpherson JM: Postural orientation and equilibrium. In Handbook of Physiology. Exercise: Regulation and integration of Multiple Systems. Rowell LG, Shepherd JT, Eds. New York, Oxford University Press, 1996.
- Boucher P, Teasdale N, Courtemanche R, Bard C, Fleury, M: Postural stability in diabetic polyneuropathy. Diabetes Care 18:638–645, 1995[Abstract]
- Oppenheim U, Kohen-Raz R, Alex D, Kohen-Raz A, Azarya M: Postural characteristics of diabetic neuropathy. Diabetes Care 22:328–333, 1999[Abstract/Free Full Text]
- Simoneau GG, Ulbrecht JS, Derr JA, Becker MB, Cavanagh PR: Postural instability in patients with diabetic sensory neuropathy. Diabetes Care 17:1411–1421, 1994[Abstract]
- Vinik AI, Park TS, Stansberry KB, Pittenger GL: Diabetic neuropathies. Diabetologia 43:957–973, 2000[Medline]
- Cavanagh PR, Derr JA, Ulbrecht JS, Maser RE, Orchard TJ: Problems with gait and posture in neuropathic patients with insulin-dependent diabetes mellitus. Diabet Med 9:469–474, 1992[Medline]
- Richardson JK, Ching C, Hurvitz EA: The relationship between electromyographically documented peripheral neuropathy and falls. J Am Geriatr Soc 40:1008–1012, 1992[Medline]
- Prieto TE, Myklebust JB, Myklebust BM: Characterization and modeling of postural steadiness in the elderly: a review. IEEE Trans Rehabil Eng 1:26–34, 1993
- Corriveau H, Prince F, Hebert R, Raîche M, Tessier D, Maheux P, Ardilouze JL: Evaluation of postural stability in elderly with diabetic neuropathy. Diabetes Care 23:1187–1191, 2000[Abstract/Free Full Text]
- Yamamoto R, Kinoshita T, Momoki T, Arai T, Okamura A, Hirao K, Sekihara H: Postural sway and diabetic peripheral neuropathy. Diabetes Res Clin Pract 52:213–221, 2001[Medline]
- Winter DA, Prince F, Stergiou P, Powell C: Medial-lateral and anterior-posterior motor responses associated with center of pressure changes in quiet standing. Neurosci Res Commun 12:141–148, 1993
- Sheldon JH: The effect of age on the control of sway. Gerontol Clin 5:129–138, 1963
- Valk GD, Nauto JJP, Striners RLM, Bertelsman FW: Clinical examination versus neurophysiological examination in the diagnosis of diabetic polyneuropathy. Diabet Med 9:716–721, 1992[Medline]
- Valk GD, Sonnaville JJ, Houtum WHV, Heine RJ, vanEijk JTM, Bouter LM, Bertelsman FW: The assessment of diabetic polyneuropathy in daily clinical practice: reproducibility and validity of Semmes Weinstein monofilaments examination and clinical neurological examination. Muscle Nerve 20:116–118, 1997[Medline]
- McGraw P, Winn B, Whitaker D: Reliability of the Snellen card. BMJ 10:1481–1482, 1995
- Uccioli L, Giacomini P, Monticone G, Magrini A, Durols A, Bruno E, Parisi L, Di Girolamo S, Menzinger G: Body sway in diabetic neuropathy. Diabetes Care 18:339–344, 1995[Abstract]
- Uccioli L, Giacomini P, Pasqualetti P, Di Girolamo S, Ferrigno P, Monticone G, Bruno E, Boccasena P, Magrini A, Parisi L, Menzinger G, Rossini PM: Contribution of central neuropathy to postural instability in IDDM patients with peripheral neuropathy. Diabetes Care 20:929–934, 1997[Abstract]
 CiteULike  Del.icio.us  Digg  Reddit  Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
J. J. Lee, J. A. Low, E. Croarkin, R. Parks, A. W. Berman, N. Mannan, S. M. Steinberg, and S. M. Swain
Changes in Neurologic Function Tests May Predict Neurotoxicity Caused by Ixabepilone
J. Clin. Oncol.,
May 1, 2006;
24(13):
2084 - 2091.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
TOP
ABSTRACT
INTRODUCTION
