- Email: [email protected]
Grip force control and hand dexterity are impaired in individuals with diabetic peripheral neuropathy
Accepted Manuscript Title: Grip force control and hand dexterity are impaired in individuals with diabetic peripheral neuropathy Authors: Kauˆe Carvalho de Almeida Lima, Leandro Borges da Silva, Elaine Hatanaka, Luiz Clemente Rolim, Paulo Barbosa de Freitas PII: DOI: Reference:
S0304-3940(17)30716-4 http://dx.doi.org/10.1016/j.neulet.2017.08.071 NSL 33072
To appear in:
Neuroscience Letters
Received date: Revised date: Accepted date:
15-4-2017 24-8-2017 24-8-2017
Please cite this article as: Kauˆe Carvalho de Almeida Lima, Leandro Borges da Silva, Elaine Hatanaka, Luiz Clemente Rolim, Paulo Barbosa de Freitas, Grip force control and hand dexterity are impaired in individuals with diabetic peripheral neuropathy, Neuroscience Lettershttp://dx.doi.org/10.1016/j.neulet.2017.08.071 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Grip force control and hand dexterity are impaired in individuals with diabetic peripheral neuropathy
Kauê Carvalho de Almeida Limaa; Leandro Borges da Silvaa; Elaine Hatanakaa; Luiz Clemente Rolimb; Paulo Barbosa de Freitasa
a
Institute of Physical Activity and Sports Sciences and Interdisciplinary Graduate
Program in Health Sciences, Cruzeiro do Sul University, Sao Paulo, Brazil b
Diabetic Neuropathy and Foot Section of the Diabetes Center at Federal University
of Sao Paulo (UNIFESP), Sao Paulo, Brazil Corresponding Author: Paulo Barbosa de Freitas Universidade Cruzeiro do Sul Rua Galvão Bueno, 868, Liberdade, Sao Paulo, SP, Brazil, 01506-000 Phone: +55 (11) 3385-3103, Fax: +55 (11) 3385-3009 E-mail: [email protected] Highlights:
The effect of diabetic neuropathy on hand function and grip force control was assessed. Individuals with neuropathy have no apparent deficits in maximum grip strength. Hand and digits dexterity are impaired in individuals with neuropathy. Individuals with neuropathy set lower safety margin than healthy ones when holding. A low safety margin for individuals with neuropathy means a poor grip force control.
Abstract Diabetic peripheral neuropathy (DPN) affects the sensory function of the hands and, consequently, may negatively impact hand dexterity, maximum grip strength (GS Max), and hand grip force (GF) control during object manipulation. The aims of this study were to examine and compare the GF control during a simple holding task as well as GSMax and hand dexterity of individuals with DPN and healthy controls. Ten type 2 diabetic individuals diagnosed with DPN and ten age- and gender-matched healthy controls performed two traditional timed hand dexterity tests (i.e., nine-hole peg test and Jebsen-Taylor hand function test), a GSMax test, and a GF control test (i.e., hold a
2
instrumented handle). The results indicated that individuals with DPN and controls produce similar GSMax. However, individuals with DPN took longer to perform the hand dexterity tests and set lower safety margin (exerted lower GF) than controls when holding the handle. The findings showed that mild to moderate DPN did not significantly affect maximum hand force generation, but does impair hand dexterity and hand GF control, which could impair the performance of daily living manipulation tasks and put them in risk of easily dropping handheld objects. Keywords: diabetes mellitus; tactile sensitivity; sensory; feedback; motor control; manipulation
Introduction Diabetes mellitus (DM) is an endocrine and metabolic disorder often associated with comorbidities such as retinopathy, nephropathy, and neuropathy [1]. Signs of neuropathy are presented in approximately fifty percent of individuals with DM and the most common is the diabetic peripheral neuropathy (DPN) [2]. The DPN can affect both sensory and motor neurons and is characterized by diminished tactile sensitivity, nerve conduction velocity, and motor function in more advanced stages of the disease [3-4]. Although the worst DPN outcomes arise in the lower extremities (e.g., amputation), studies have showed that the hands of individuals with DM with and without diagnosed DPN are affected by deficits in the sensory and motor functions [58]. During manipulation tasks, individuals need to exert an adequate magnitude of grip force (GF) perpendicularly against the object contact surface to prevent slippage caused by external and self-generating forces acting tangentially (load force – LF) between the digits/hand skin and the object surface [9-10]. Slippage is prevented when the exerted GF is equal or higher than the ratio between LF and the static coefficient of friction (COF) of the skin-object interaction. Nevertheless, during manipulation individuals tend to be more precautious adopting a safety margin (SM), applying higher GF than the minimum needed to prevent slippage (GF Min). However, the amount of GF should not be much higher than the GF Min because it could cause fatigue, damage fragile objects, and affect the fine object control [9-10].
3
Previous studies have showed that individuals with central and peripheral neurological conditions present deficits in GF control. For example, individuals with Parkinson’s disease [11-12], stroke [13], cerebellar dysfunction [14], and mild multiple sclerosis [15,16] exert greater magnitude of GF when performing manipulation tasks than healthy individuals. Furthermore, individuals with carpal tunnel syndrome (CTS), a peripheral nervous system disorder, also exert greater amount of GF while manipulating objects than control [17-19]. Therefore, someone would expect that individuals with DPN would set higher SM and, consequently, exert higher magnitudes of GF when holding and transporting an object. Before investigating GF control of individuals with DPN, we studied individuals with DM but with no medical diagnosis and clinical signs of DPN [6]. Surprisingly, although individuals with DM did not show reduction in maximum power grip strength (GSMax) and impaired hand dexterity, they set much lower SM, using around half of GF exerted by the controls when simply holding an object. We argued that this probably occurred because the deficits in cutaneous sensitivity would be very mild in these diabetic individuals and that such deficits would not be sufficient to trigger the use of a compensatory GF control strategy by the GF controller, but would be enough to disrupt the GF control and cause error in the estimation of the GF needed during a object holding task [6]. Considering that DPN individuals present greater sensory and motor deficits than individuals with DM with no signs of DPN, we would expect they would produce lower GSMax, present impaired hand and digits dexterity (hypothesis 1), and set greater SM (i.e., exert greater magnitude of GF) (hypothesis 2) when compared to age- and sex-matched health controls when holding an object. This study is relevant in terms of extending the knowledge about the effects of the DPN on motor control in general and about the use of the somatosensory information to control the forces exerted against the objects’ surface during its manipulation in particular. Method
Participants The sample size of this study was determined after a pilot study performed with five individuals of each group (DPN and controls). The alpha level and power were set at 0.05 and 0.8, respectively, and the mean and standard deviation of the variable
4
relative safety margin (SMRel) were used to this calculation. Therefore, 10 individuals with type 2 DM (T2DM) diagnosed and with clinical signs of DPN (mean±SE, 58.3±1.3 years old) and 10 age- and sex-matched controls (58.2±1.2 years old) participated in the study. All participants were right-handed as indicated by their answers to the Edinburgh Handedness Inventory [20]. The participants signed an informed consent form before starting the participation in the study. The experimental procedures were conduct in accordance with the Declaration of Helsinki and the Universidade Cruzeiro do Sul Research Ethics Committee approved both the study protocol and the consent form. The T2DM individuals diagnosed with DPN were recruited from centers specialized in the treatment of individuals with DPN. They should be between 50–65 years old and have at least 10 years of diagnosis of the T2DM. They should also walk independently and have no foot amputation, no diagnostic and symptoms of other neuropathic syndromes or neurological diseases, and should present a score equal or higher than six in the Michigan Neuropathy Screening Instrument (MNSI) [21,22]. Finally, the DPN and control individuals should be able to understand and follow simple instructions and have no history of musculoskeletal injury or disease affecting their hands (e.g., CTS). Table 1 depicts the participants’ demographic, physical, metabolic, and biochemical profiles, used medication, time of diagnosis of T2DM and neuropathy, smoking behavior, blood pressure, BMI, body fat, lean mass, basal metabolic rate, and glucose and fructosamine levels. The fructosamine level, which indicates 2-3 weeks of glycemic control, showed that seven out of ten individuals with DPN presented fructosamine above 280 mmol/L and two presented fructosamine concentration below 200 mmol/L, indicating well-controlled T2DM.
Hand function assessment Two tests aiming to assess tactile sensitivity and mechanoreceptors spatial resolution were used in DPN and controls. We performed the Semmes-Weinstein Monofilaments Examination (SWME) and the static two-point discriminator test (2-PD)
5
in the hands (tips of thumb, index, and little fingers). In the SWME, we adopted a score based on the range of monofilaments from the thinnest (i.e., 0.05 gf) to the thickest (i.e., 300 gf) in a total of six monofilaments. Thus, when the participant was able to feel the thinnest monofilament they received a score of one, when they were able to feel 0.2 gf received a score of two, and so on. When the participant was not able to feel any monofilaments, they received a score of seven. The sum of the scores obtained from each of the three tested fingers (thumb, index and little) was used as the dependent variable to describe the tactile sensitivity for each group. Therefore, the score of tactile sensitivity ranged from 3 to 21, with 3 meaning that the participant has no sensory loss and 21 meaning that they have a sensory loss and no tactile information can be detected at the tip of their tested digits. In addition, the static 2-PD score was defined as the value in millimeters (1-20 mm) in which the participant was able to feel the presence of two points and the sum of those values obtained for each finger was used as the dependent variable to describe mechanoreceptor spatial resolution. Next, the participants were seated on a chair and instructed to perform three tests commonly used to evaluate hand function: nine-hole peg test (9HPT, RolyanTM 9-Hole Peg Test, Model A8515); Jebsen-Taylor hand function test (JTHFT, Sammons Preston Jebsen-Taylor Hand Function Test); and GSMax test using a hydraulic Jamar hand dynamometer (Sammons Preston). The tests were performed with the dominant and non-dominant hands. Half of the DPN individuals and their respective controls started the tests with their dominant while the other half started with their non-dominant hand.
Grip force control assessment The instrumented handle used in this study is shown in Figure 1A. This handle is formed by two aluminum bars linked with a compression-tension load cell (LPM530, Cooper Instruments and Systems, USA) and two aluminum pieces and a multiaxis force and torque (F/T) transducer (Mini40, ATI, USA) in between forming the basis of the handle. A cylindrical mass was attached underneath the handle to increase the total mass of the handle, which was in total 647 g (W=6.35 N). The handle aperture was of 5 cm and the external surface of it was covered with extra fine sandpaper (320
6
grit), which provided a moderate to high COF between the digits skin and the object surface. _____________________ Insert Figure 1 around here _____________________ To perform the holding task the seated participants were firstly asked to grasp and hold the free and vertically oriented handle using the tip of all digits (i.e., the thumb and the opposing four fingers). Participants were asked to remain with the arm positioned beside the trunk and shoulder with an internal rotation of 60 °, elbow flexed at 90°, forearm pronated to 90°, wrist in a neutral position, metacarpophalangeal joint bent at 90°, and interphalangeal joints in the neutral position. This body configuration enabled the participant to maintain the center of the handle aligned with their body midline and positioned just above the umbilicus. Posteriorly, the experimenter instructed the participants to hold the handle as stable as possible for 10 seconds as they would hold a glass of water. The experimenter started the data collection immediately after the participant was able to keep a stable handle position. Also, the participants were instructed to slowly reduce the magnitude of the GF until the handle drop off of the hand (slip test) after hearing a beep sound, which was provided at the 10th second of the trial recording. The slip test was performed in order to identify the minimum amount of GF required to hold the object (GF Min) and calculate the slip ratio (SR) [6,10,23]. Participants performed five trials. The first two was used for familiarization, principally due to the slip test, and the last three were used for posterior analyses. They performed this task only with their dominant hand (right) since we have evidence that there is no effect of handedness in this kind of task [6].
Data processing and analyses Two customized LabVIEW codes (National Instruments, USA) were used for data acquisition and processing. The force signals were recorded at 200 Hz and stored for later analyses. The raw force signals were low-pass filtered with a 4th-order, zero lag, Butterworth filter with a cut-off frequency of 20 Hz. LF was simply the weight of
7
the handle (i.e., 6.35 N) because the handle was kept stationary and the horizontal component acting tangentially was negligible. Moreover, GF exerted against the object surface was estimated by compression force recorded from the load cell (F C) fixed between the two flat aluminum plates. The holding phase lasted 10 seconds, but we analyzed just the central six, skipping the first two and last two seconds [6]. For holding phase, we computed the within trial coefficient of variation of the GF (CV-GF) shown in percentage of the averaged GF during the selected time interval (GF Mean). Next, the GF control was assessed by GFMean and the relative safety margin (SMRel) using the following equation: SMRel=100×((GFMean – GFMin)/GFMin), in which GFMin was obtained in the slippage phase [6,24]. In the slippage phase, we estimated GF Min and calculated the SR (SR=GFMin/LF). The point of slippage was determined as the point in time in which a sudden reduction of the vertical force component (F Z) recorded from the F/T transducer was detected due to the handle’s acceleration. Specifically, the point of slippage was automatically detected as the point in time in which F Z was lower than the FZ mean minus two FZ standard deviations, both calculated during holding phase. The GFMin was estimated from the point immediately before the moment of slippage.
Statistical Analyses After assuring normal distribution of all variables (Shapiro-Wilk test), we completed analyses of variance (ANOVA). For hand function assessment, we performed three two-way ANOVA, with the last factor (hand) treated as repeated measure, to test for differences between groups (DPN and controls) and hand (dominant and non-dominant) in the performance of 9HPT and JTHFT and in the GSMax. For the holding task, four one-way ANOVA were performed to test for differences between groups in SR, CV-GF, GFMean, and SMRel. In addition, two MannWhitney U tests were performed to test for group differences in the SWME and the static 2PD test scores. The level of significance for all analyses was set at 0.05.
8
Results
Tactile sensitivity and spatial resolution The results of the SWME revealed that individuals with DPN presented higher scores (i.e., lower tactile sensitivity) in the digits from the dominant hand when compared to controls (median±interquartile range, DPN: 9±1 – Controls: 5±3.5 | U=3.5, p<.001). Also, results of the 2-PD test revealed that individuals with DPN presented lower tactile spatial resolution than controls in the digits of the dominant hand (DPN: 16.5±7.75 – Controls: 10±4.5 | U=15, p<.01).
Strength and dexterity tests Table 2 depicts mean (SE) of the performance in 9HPT and JTHFT, and GS Max. The results revealed that individuals with DPN had worst performance in both dexterity tests
than
controls,
9HPT
[F(1,18)=16.3,p<.005,η2=.48]
and
JTHFT
[F(1,18)=14.4,p<.005,η2p=.45], but revealed no differences for GSMax (p>.05). Furthermore, no group by hand interaction was revealed (p>.05). The results also revealed an effect of hand on all three tests. Specifically, participants performed better with their dominant when compared to their non-dominant hands [9HPT: F(1,18)=29.1,p<.001,η2p=.62;
JTHFT:
F(1,18)=20.8,p<.001,η2p=.54;
and
GSMax:
F(1,18)=12.3,p<.005,η2 p=.41].
Grip force control Figure 2 depicts GF and LF time-series for a representative individual with DPN and his respective control. As can be observed, GF was much lower and closer to the GFMin for the individual with DPN when compared to his matched control. The holding
9
task was separated in two phases, holding and slippage. During the slippage phase, the only variable analyzed was the SR that did not differ between DPN and controls (p>.05). Regarding GF steadiness, the results revealed no difference between groups (p>.05) for CV-GF. However, in terms of GF control, the results revealed that both, the GFMean and the SMRel was smaller for individuals with DPN when compared to controls [GFMean: F(1,18)=7.72,p<.05,η2p=.3 | SMRel: F(1,18)=13.1,p<.005,η2p=.42]. Table 3 depicts means (SE) of SR, GFMean, SMRel, and CV-GF. _____________________ Figure 2 ______________________
Discussion The aims of this study were to evaluate and compare hand function (strength and dexterity) and GF control of DPN and healthy individuals. Regarding strength and dexterity, we hypothesized that GSMax and hand dexterity would be sensible to detect differences between DPN and controls. This hypothesis was partially confirmed since the results showed that individuals with DPN presented worst performance in JTHFT and 9HPT, but showed that individuals with DPN and controls produced similar GS Max. Recently, we found no difference between diabetic individuals with no signs of neuropathy and controls for hand and digits dexterity (JTHFT and 9HPT, respectively) and strength tests (i.e., GSMax) [6]. In the present study, we found that hand and digits dexterity is negatively affected by DPN but not strength. Unsurprisingly, deficits in hand and digits dexterity could be associated with the reduction in tactile sensitivity shown by individuals with DPN assessed in this study. The lack of significant difference between individuals with DPN and controls for the GS Max could be a sign that individuals with DPN assessed in this study still have a preserved and functional motor system. Also, the number of individuals with DPN tested in each group (n=10) could be responsible for the lack of difference between groups. A sample size calculation using DPN and control group GSMax mean and standard deviation, alpha
10
level of 0.05 and power of 0.8 indicated that 44 individuals in each group would be necessary to detect a significant lower GSMax in the DPN than in the control group. However, it is important to notice that the sample size calculation for this study was based on SMRel not in the GSMax. Regarding GF control, it was hypothesized that, due to their great sensory loss and the possible use of a compensatory GF control strategy, individuals with DPN would increase their SM by increasing GF while holding an object. This hypothesis was rejected as the results showed that individuals with DPN set much lower SM (≈44%) than controls during the free holding task. Based on the results of this study and on the previous one [6], it could be suggested that the individuals with no signs of neuropathy tested in our prior study would present mild neuropathy, but it would be concealed by the lack of signs and symptoms and by the preservation of hand function. Corroborating this assumption, we found no difference between the relative difference between individuals with DPN and their controls and diabetic individuals with no DPN and their controls (i.e., 43,34% and 48.64%, respectively) [6]. Both groups exerted a little less than half of GF exerted by the age- and sex-matched controls when steadily holding an object. Additionally, we could also claim that the DPN has a different effect on GF control than would have digital anesthesia [9,25,26], sensory deafferentation [27,28], and central neurological deficits such as multiple sclerosis, Parkinson’s disease, stroke, and cerebellar dysfunction [11-16], which are known to lead to an increase in the exerted GF. It would not be realistic to expect that individuals with DPN would have similar GF control than anesthetized controls. While DPN causes a chronic and mild to moderate sensory impairment, anesthesia causes an acute and severe loss on the ability to temporarily detect any tactile stimulus. Due to the chronicity of the DPN, it would be expected that individuals with DPN would compensate this sensory impairment and reduce its effects. Therefore, one would expect a stronger effect of anesthesia than DPN in GF control, but still would expect a greater SM in individuals with DPN than in controls. Similarly, it would be unfair to compare the GF exerted by a deafferented patient with no sense of touch, vibration and pressure and individuals with mild to moderate neuropathy. All individuals with DPN tested in this study were able to feel the 10 gf (approximately 1 N) monofilaments when it was applied to the digits. GF applied by individuals with DPN against the object was around 5 N. Thus,
11
they would be able to detect any significant changes in GF exerted against the surface of a hand-held object. A condition comparable to DPN would be carpal tunnel syndrome (CTS), which is considered an entrapment neuropathy. As individuals with DPN, individuals with CTS also present reduced tactile sensitivity and in more severe cases, they present muscle weakness. Some studies have shown that individuals with CTS apply higher GF than controls [17-19], but others have not observed this increase in GF in this population [29,30]. Therefore, the reduction in SM observed in individuals with DPN in this study would be unexpected. Hence, what would be the underlying cause of a lower GF exertion and consequent lower SM set by individuals with DPN when holding an object? In our previous study [6], we suggested that the mild sensory impairment of diabetic individuals with no DPN would not be sufficient to elicit the use of a compensatory GF control strategy, which would lead to a increased GF exertion, but would be enough to disrupt the proper use of sensory information in this diabetic individuals causing an error of the estimation of GF needed while steadily holding an object. It would be convenient to extrapolate this suggestion to diabetic individuals with DNP, with moderate and severe sensation loss in the tip of the digits. As mentioned earlier, despite having a reduced tactile acuity, the magnitude of this reduction in individuals with DPN would be lower than the reduction seen in healthy anesthetized and deafferented individuals. There would be still tactile stimulus providing residual information that could be used by the GF controller to detect changes in the digitsobject interaction, but yet not enough to assure an adequate estimation of the forces between the digits skin and object surface and, consequently, a proper GF control. Corroborating, a person from the DPN group who was the most affected in terms of tactile sensitivity, who presented the higher score in the 2-PD test, the worst performance in 9HPT and JTHFT, and the lower level of GS Max showed the GF-LF ratio and SMRel a little higher than the group averages, but her GF-LF ratio and SM was much lower than the averaged ratio and SM presented by the control group. Thus, we speculate that even if we had tested only diabetic individuals with severe neuropathy we would have found that individuals with severe neuropathy would present a lower ratio and SM than controls.
12
Despite being considered a peripheral neurological disease, evidences show that DPN also affects parts of the central nervous system. Specifically, studies found that individuals with DPN also have reduction of the spinal cord area in places such C2/3 [31] and C4/5 and T3/4 [32] when compared to individuals with no DPN and controls. Moreover, a recent study showed a reduction in gray matter volume in the primary somatosensory (S1), parts of the cingulate cortex and in the supramarginal gyrus in individuals with DPN when compared to controls [33]. Those cortical areas are closely related to the control and coordination of GF and LF [34]. Therefore, impaired tactile stimulus detection and neural conduction to the CNS, and altered tactile information processing in S1 could be detrimental to the formation of a new sensorimotor representation of a new object in individuals with DPN [35]. It could take longer to the GF controller of diabetic individuals with and without DPN to form a proper sensorimotor representation when dealing with novel objects and to implement a conservative GF control strategy. In order to test this hypothesis, a new study should be done in which participants with DPN and controls would perform a large number of trials using the same object. There are still some aspects of GF control in individuals with DM with and with no DPN that should be explored. For instance, the effect of performing dynamic tasks, such as lifting and shaking a handheld object, on GF control should be investigated in order to examine if the reduction in SM observed in diabetic individuals during the holding task remains similar or is task-dependent. It is important to mention that most of the studies investigating the effect of neurological diseases on GF control used dynamic tasks. Likewise, trial-to-trial GF adaptation as well as the GF adaptation to changes in object’s friction and weight should also be tested to assess the effect of the sensory loss on GF adaptation in diabetic individuals without and with diagnosis of DPN. Finally, the relationship between GF control variables and electrophysiological variables, such as sensory and motor neural conduction velocity, should be explored to a better understanding of the GF control and to determine the feasibility of using GF control variables in an early detection of DPN.
Conclusion
13
Based on the results of the present study we conclude that individuals with DPN present poorer hand and digits dexterity tests in spite of having their GS Max preserved. In addition, we conclude that individuals with DPN can still maintain a steady GF while holding an object, although, they present impairment in GF scaling. Specifically, they set a relatively low SM, which put them in risk of easily dropping handheld objects. This impairment of GF scaling may be predominantly caused by the reduction in tactile sensitivity, but the deficits in the sensorimotor transformation should also be considered. Taken together, these findings indicate that the study of GF control during object manipulation can be a promising venue for further understanding of the effect of DPN on movement control. More important, SM obtained from GF control studies and, specifically, during a simple holding task, could be seen as an easily obtained behavioral biomarker for early detection of neuropathy in individuals with DM even in situation in which no clinical signs of neuropathy is present [6]. In order to support this statement a larger and prospective study should be conducted.
Acknowledgments: The authors acknowledge the Sao Paulo State Research Foundation (FAPESP #2010/02939-4 and #2014/26397-7) and National Counsel of Technological and Scientific Development (CNPq, #479903/2013-1) for their financial support. Lima and Borges are thankful for their scholarships provided by Coordination for the Improvement of Higher Education Personnel (CAPES – Brazil).
References [1] ADA. (2013) Diagnosis and Classification of Diabetes Mellitus, American Diabetes Association. Diabetes Care;36.
14
[2] Van Acker K, Bouhassira D, De Bacquer D, Weiss S, Matthys K, Raemen H, Mathieu C, Colin IM (2009) Prevalence and impact on quality of life of peripheral neuropathy with or without neuropathic pain in type 1 and type 2 diabetic patients attending hospital outpatients clinics. Diabetes Metab. 35(3):206-13. [3] Watkins PJ, Thomas PK (1998) Diabetes mellitus and the nervous system. J Neurol Neurosurg Psychiatry 65 620-32. [4] Ramji N, Toth C, Kennedy J, Zochodne DW (2007) Does diabetes mellitus target motor neurons? Neurobiol Dis. 26:301-11. [5] Dahlin LB, Thrainsdottir S, Cederlund R, Thomsen NOB, Eriksson KF, Rosen I, et al (2008) Vibrotactile sense in median and ulnar nerve innervated fingers of men with Type 2 diabetes, normal or impaired glucose tolerance. Diabetic Med 25:543-9. [6] de Freitas PB, Lima KCA (2013) Grip force control during simple manipulation tasks in non-neuropathic diabetic individuals. Clin Neurophysiol 124:1904–10. [7] Gorniak SL, Khan A, Ochoa N, Sharma MD, Phan CL (2014) Detecting subtle fingertip sensory and motor dysfunction in adults with type II diabetes. Exp Brain Res. 232(4):1283-91 [8] Ochoa N, Gorniak SL (2014) Changes in sensory function and force production in adults with type II diabetes. Muscle Nerve 50(6):984-90. [9] Johansson RS, Westling G (1984) Roles of glabrous skin receptors and sensorimotor memory in automatic control of precision grip when lifting rougher or more slippery objects. Exp Brain Res 56:550-64. [10] Westling G, Johansson RS. (1984) Factors influencing the force control during precision grip. Exp Brain Res 53:277-84. [11] Fellows SJ, Noth J, Schwarz M (1998) Precision grip and Parkinson's disease. Brain. 1998;121:1771-84 [12] Nowak DA, Hermsdorfer J (2002) Coordination of grip and load forces during vertical point-to-point movements with a grasped object in Parkinson's disease. Behav Neurosci 116:837-50. [13] Hermsdorfer J, Hagl E, Nowak DA, Marquardt C (2003) Grip force control during object manipulation in cerebral stroke. Clin Neurophysiol 114:915-29. [14] Brandauer B, Timmann D, Hausler A, Hermsdorfer J (2010) Influences of load characteristics on impaired control of grip forces in patients with cerebellar damage. J Neurophysiol 103:698-708.
15
[15] Krishnan V, De Freitas PB, Jaric S (2008) Impaired Object Manipulation in Mildly Involved Individuals with Multiple Sclerosis. Motor Control12:3-20. [16] Iyengar V, Santos MJ, Ko M, Aruin AS (2009) Grip force control in individuals with multiple sclerosis. Neurorehabil Neural Repair 23:855-61. [17] Lowe BD, Freivalds A (1999) Effect of carpal tunnel syndrome on grip force coordination on hand tools. Ergonomics 42:550-64. [18] Zhang W, Johnston JA, Ross MA, Smith AA, Coakley BJ, Gleason EA, et al. (2011) Effects of carpal tunnel syndrome on adaptation of multi-digit forces to object weight for whole-hand manipulation. PLoS One. 6:e27715. [19] Chen P-T, Jou I-M, Lin C-J, Chieh H-F, Kuo L-C, Su F-C (2015) Is the Control of Applied Digital Forces During Natural Five-digit Grasping Affected by Carpal Tunnel Syndrome? Clin Orthop Relat Res 473:2371–82. [20] Oldfield RC (1971) The assessment and analysis of handedness: The Edinburgh Inventory. Neuropsychologia. 9:97-113. [21] Feldman EL, Stevens MJ, Thomas PK, Brown MB, Canal N, Greene DA (1994) A practical two-step quantitative clinical and electrophysiological assessment for the diagnosis and staging of diabetic neuropathy. Diabetes Care 17:1281-9. [22] Valk GD, Grootenhuis PA, Bouter LM, Bertelsmann FW (1994) Complaints of neuropathy related to the clinical and neurophysiological assessment of nerve function in patients with diabetes mellitus. Diabetes Res Clin Pract 26:29-34. [23] de Freitas PB, Uygur M, Jaric S. (2009) Grip force adaptation in manipulation activities performed under different coating and grasping conditions. Neurosci Lett 457:16-20. [24] Danion F. (2008) Grip force safety margin in rock climbers. Int J Sport Med 29: 168-72. [25] Monzee J, Lamarre Y, Smith AM (2003) The effects of digital anesthesia on force control using a precision grip. J Neurophysiol 89:672-83. [26] Carteron A, McPartlan K, Gioeli C, Reid E, Turturro M, Hahn B, Benson C, Zhang W (2016). Temporary Nerve Block at Selected Digits Revealed Hand Motor Deficits in Grasping Tasks. Frontiers in Human Neuroscience, 10, 596. [27] Hermsdorfer J, Elias Z, Cole JD, Quaney BM, Nowak DA. (2008) Preserved and impaired aspects of feed-forward grip force control after chronic somatosensory deafferentation. Neurorehabil Neural Repair 22:374-84.
16
[28] Nowak DA, Glasauer S, Hermsdorfer J (2004) How predictive is grip force control in the complete absence of somatosensory feedback? Brain127:182-92. [29] Thonnard J, Saels P, Van den Bergh P, Lejeune T (1999) Effects of chronic median nerve compression at the wrist on sensation and manual skills. Exp Brain Res 128:61-4. [30] Nowak DA, Hermsdorfer J, Marquardt C, Topka H (2003) Moving objects with clumsy fingers: how predictive is grip force control in patients with impaired manual sensibility? Clin Neurophysiol. 2003;114:472-87. [31] Selvarajah D, Wilkinson ID, Emery CJ, Harris ND, Shaw PJ, Witte DR, et al (2006) Early involvement of the spinal cord in diabetic peripheral neuropathy. Diabetes Care 29:2664-9. [32] Eaton SE, Harris ND, Rajbhandari SM, Greenwood P, Wilkinson ID, Ward JD, et al (2001) Spinal-cord involvement in diabetic peripheral neuropathy. Lancet. 2001;358:35-6. [33] Selvarajah D, Wilkinson ID, Maxwell M, Davies J, Sankar A, Boland E, et al (2014) Magnetic resonance neuroimaging study of brain structural differences in diabetic peripheral neuropathy. Diabetes Care 37:1681-8. [34] Ehrsson HH, Fagergren A, Johansson RS, Forssberg H (2003) Evidence for the involvement of the posterior parietal cortex in coordination of fingertip forces for grasp stability in manipulation. J Neurophysiol 90:2978-86. [35] Gordon AM, Westling G, Cole KJ, Johansson RS. (1998) Memory representations underlying motor commands used during manipulation of common and novel objects. J Neurophysiol 69:1789-96.
17
Figure Captions Figure 1: Free holding task illustration. (A) The participants held the handle as stable as possible for 10 seconds; (B) after hearing a beep, they slowly reduced the magnitude of GF until the object slippage (the slip test); (C) compression force (FC) recorded by the single-axis force transducer, and the vertical (F Z) and horizontal (FY and FX) force components recorded by the multi-axis F/T transducer.
Figure 2: Load (LF) and grip force (GF) time-series obtained during a free holding task from representative DPN and control participants. Dashed vertical line indicates the moment in which the beep sound was heard and GF started being reduced till slippage.
18
Figr-1
19 Figr-2
20
Table 1. Medical, physical, and anthropometric characteristics of the participants with diabetic peripheral neuropathy (DPN) and healthy controls. DPN (means ± SE) 10
Controls (means ± SE) 10
0
6
5.7 ± 0.9 7 7 3 3
0.5 ± 0.2 0 2 1 0
Physical and Clinical Aspects Age (years) Time since diagnosis of T2DM (years) Time since diagnosis of DPN (years) Smokers (n) Former smokers (n) Alcoholism (n) Carpal tunnel syndrome (n)
58.3 ± 1.3 16 ± 1.7 4.1 ± 0.5 1 4 2 0
58.2 ± 1.2 ----1 0 0 0
Blood Pressure Systolic (mmHg) Diastolic (mmHg)
129 ± 8 75 ± 3
115.1 ± 4 79.4 ± 2.3
Anthropometric Measurements Body mass (kg) Stature (m) BMI (kg/m2) Body fat (%) Lean mass (kg) Basal metabolic rate
81.5 ± 3.9 1.61 ± 0.03 30.8 ± 1 31.3 ± 2.5 56.6 ± 3.1 1721 ± 94
80.5 ± 4.2 1.67 ± 0.04 29.1 ± 1.4 35.8 ± 2.2 49.5 ± 2.8 1505 ± 85
Glycemic Control (n) Fasting capillary glycemia (mg/dL) Fructosamine (mmol/L)
9-10 141.6 ± 11.6 274.2 ± 34.8
4-10 113.8 ± 9.4 244.17 ± 17
N Medication Participants without medication Number of medications prescribed per person Participants on medication for diabetes Participants on medication for hypertension Participants on medication for dyslipidemia Participants on medication for neuropathy
21
Table 2. Mean (SE) values representing the performance of DPN and healthy controls in the nine-hole peg test (9HPT) and in the Jebsen-Taylor hand function test (JTHFT) and the maximum power grip strength (GSMax) of the dominant (DH) and non-dominant (NDH) hand.
Hand
Group
9HPT(s)
JTHFT(s)
GSMax(kgf)
DPN
19.39** (0.61)
33.63** (1.50)
32.90 (2.90)
Controls
15.73 (0.41)
26.56 (0.54)
37.90 (3.02)
DPN
20.61* (0.75)
35.70** (1.79)
31.40 (2.64)
Controls
17.72 (0.62)
29.41 (1.02)
34.90 (2.20)
DH
NDH
Note: * p<0.01 and ** p<0.001.
22
Table 3. Mean (SE) of slip ratio (SR), averaged GF exerted (GF Mean), relative safety margin (SMRel, in % of the GFMin), and coefficient of variation of the exerted GF (CVGF) when the DPN and control participants held the free moving handle with their dominant hand.
Group
SR
GFMean
SMRel (%of GFMin) CV-GF (%)
DPN
0.45 (0.01)
5.08* (0.31)
77.15** (8.43)
5.11 (0.63)
Controls
0.43 (0.02)
7.63 (0.86)
178.03 (21.93)
3.66 (0.5)
Note: *p<0.01 and **p<0.005.
S0304-3940(17)30716-4 http://dx.doi.org/10.1016/j.neulet.2017.08.071 NSL 33072
To appear in:
Neuroscience Letters
Received date: Revised date: Accepted date:
15-4-2017 24-8-2017 24-8-2017
Please cite this article as: Kauˆe Carvalho de Almeida Lima, Leandro Borges da Silva, Elaine Hatanaka, Luiz Clemente Rolim, Paulo Barbosa de Freitas, Grip force control and hand dexterity are impaired in individuals with diabetic peripheral neuropathy, Neuroscience Lettershttp://dx.doi.org/10.1016/j.neulet.2017.08.071 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Grip force control and hand dexterity are impaired in individuals with diabetic peripheral neuropathy
Kauê Carvalho de Almeida Limaa; Leandro Borges da Silvaa; Elaine Hatanakaa; Luiz Clemente Rolimb; Paulo Barbosa de Freitasa
a
Institute of Physical Activity and Sports Sciences and Interdisciplinary Graduate
Program in Health Sciences, Cruzeiro do Sul University, Sao Paulo, Brazil b
Diabetic Neuropathy and Foot Section of the Diabetes Center at Federal University
of Sao Paulo (UNIFESP), Sao Paulo, Brazil Corresponding Author: Paulo Barbosa de Freitas Universidade Cruzeiro do Sul Rua Galvão Bueno, 868, Liberdade, Sao Paulo, SP, Brazil, 01506-000 Phone: +55 (11) 3385-3103, Fax: +55 (11) 3385-3009 E-mail: [email protected] Highlights:
The effect of diabetic neuropathy on hand function and grip force control was assessed. Individuals with neuropathy have no apparent deficits in maximum grip strength. Hand and digits dexterity are impaired in individuals with neuropathy. Individuals with neuropathy set lower safety margin than healthy ones when holding. A low safety margin for individuals with neuropathy means a poor grip force control.
Abstract Diabetic peripheral neuropathy (DPN) affects the sensory function of the hands and, consequently, may negatively impact hand dexterity, maximum grip strength (GS Max), and hand grip force (GF) control during object manipulation. The aims of this study were to examine and compare the GF control during a simple holding task as well as GSMax and hand dexterity of individuals with DPN and healthy controls. Ten type 2 diabetic individuals diagnosed with DPN and ten age- and gender-matched healthy controls performed two traditional timed hand dexterity tests (i.e., nine-hole peg test and Jebsen-Taylor hand function test), a GSMax test, and a GF control test (i.e., hold a
2
instrumented handle). The results indicated that individuals with DPN and controls produce similar GSMax. However, individuals with DPN took longer to perform the hand dexterity tests and set lower safety margin (exerted lower GF) than controls when holding the handle. The findings showed that mild to moderate DPN did not significantly affect maximum hand force generation, but does impair hand dexterity and hand GF control, which could impair the performance of daily living manipulation tasks and put them in risk of easily dropping handheld objects. Keywords: diabetes mellitus; tactile sensitivity; sensory; feedback; motor control; manipulation
Introduction Diabetes mellitus (DM) is an endocrine and metabolic disorder often associated with comorbidities such as retinopathy, nephropathy, and neuropathy [1]. Signs of neuropathy are presented in approximately fifty percent of individuals with DM and the most common is the diabetic peripheral neuropathy (DPN) [2]. The DPN can affect both sensory and motor neurons and is characterized by diminished tactile sensitivity, nerve conduction velocity, and motor function in more advanced stages of the disease [3-4]. Although the worst DPN outcomes arise in the lower extremities (e.g., amputation), studies have showed that the hands of individuals with DM with and without diagnosed DPN are affected by deficits in the sensory and motor functions [58]. During manipulation tasks, individuals need to exert an adequate magnitude of grip force (GF) perpendicularly against the object contact surface to prevent slippage caused by external and self-generating forces acting tangentially (load force – LF) between the digits/hand skin and the object surface [9-10]. Slippage is prevented when the exerted GF is equal or higher than the ratio between LF and the static coefficient of friction (COF) of the skin-object interaction. Nevertheless, during manipulation individuals tend to be more precautious adopting a safety margin (SM), applying higher GF than the minimum needed to prevent slippage (GF Min). However, the amount of GF should not be much higher than the GF Min because it could cause fatigue, damage fragile objects, and affect the fine object control [9-10].
3
Previous studies have showed that individuals with central and peripheral neurological conditions present deficits in GF control. For example, individuals with Parkinson’s disease [11-12], stroke [13], cerebellar dysfunction [14], and mild multiple sclerosis [15,16] exert greater magnitude of GF when performing manipulation tasks than healthy individuals. Furthermore, individuals with carpal tunnel syndrome (CTS), a peripheral nervous system disorder, also exert greater amount of GF while manipulating objects than control [17-19]. Therefore, someone would expect that individuals with DPN would set higher SM and, consequently, exert higher magnitudes of GF when holding and transporting an object. Before investigating GF control of individuals with DPN, we studied individuals with DM but with no medical diagnosis and clinical signs of DPN [6]. Surprisingly, although individuals with DM did not show reduction in maximum power grip strength (GSMax) and impaired hand dexterity, they set much lower SM, using around half of GF exerted by the controls when simply holding an object. We argued that this probably occurred because the deficits in cutaneous sensitivity would be very mild in these diabetic individuals and that such deficits would not be sufficient to trigger the use of a compensatory GF control strategy by the GF controller, but would be enough to disrupt the GF control and cause error in the estimation of the GF needed during a object holding task [6]. Considering that DPN individuals present greater sensory and motor deficits than individuals with DM with no signs of DPN, we would expect they would produce lower GSMax, present impaired hand and digits dexterity (hypothesis 1), and set greater SM (i.e., exert greater magnitude of GF) (hypothesis 2) when compared to age- and sex-matched health controls when holding an object. This study is relevant in terms of extending the knowledge about the effects of the DPN on motor control in general and about the use of the somatosensory information to control the forces exerted against the objects’ surface during its manipulation in particular. Method
Participants The sample size of this study was determined after a pilot study performed with five individuals of each group (DPN and controls). The alpha level and power were set at 0.05 and 0.8, respectively, and the mean and standard deviation of the variable
4
relative safety margin (SMRel) were used to this calculation. Therefore, 10 individuals with type 2 DM (T2DM) diagnosed and with clinical signs of DPN (mean±SE, 58.3±1.3 years old) and 10 age- and sex-matched controls (58.2±1.2 years old) participated in the study. All participants were right-handed as indicated by their answers to the Edinburgh Handedness Inventory [20]. The participants signed an informed consent form before starting the participation in the study. The experimental procedures were conduct in accordance with the Declaration of Helsinki and the Universidade Cruzeiro do Sul Research Ethics Committee approved both the study protocol and the consent form. The T2DM individuals diagnosed with DPN were recruited from centers specialized in the treatment of individuals with DPN. They should be between 50–65 years old and have at least 10 years of diagnosis of the T2DM. They should also walk independently and have no foot amputation, no diagnostic and symptoms of other neuropathic syndromes or neurological diseases, and should present a score equal or higher than six in the Michigan Neuropathy Screening Instrument (MNSI) [21,22]. Finally, the DPN and control individuals should be able to understand and follow simple instructions and have no history of musculoskeletal injury or disease affecting their hands (e.g., CTS). Table 1 depicts the participants’ demographic, physical, metabolic, and biochemical profiles, used medication, time of diagnosis of T2DM and neuropathy, smoking behavior, blood pressure, BMI, body fat, lean mass, basal metabolic rate, and glucose and fructosamine levels. The fructosamine level, which indicates 2-3 weeks of glycemic control, showed that seven out of ten individuals with DPN presented fructosamine above 280 mmol/L and two presented fructosamine concentration below 200 mmol/L, indicating well-controlled T2DM.
Hand function assessment Two tests aiming to assess tactile sensitivity and mechanoreceptors spatial resolution were used in DPN and controls. We performed the Semmes-Weinstein Monofilaments Examination (SWME) and the static two-point discriminator test (2-PD)
5
in the hands (tips of thumb, index, and little fingers). In the SWME, we adopted a score based on the range of monofilaments from the thinnest (i.e., 0.05 gf) to the thickest (i.e., 300 gf) in a total of six monofilaments. Thus, when the participant was able to feel the thinnest monofilament they received a score of one, when they were able to feel 0.2 gf received a score of two, and so on. When the participant was not able to feel any monofilaments, they received a score of seven. The sum of the scores obtained from each of the three tested fingers (thumb, index and little) was used as the dependent variable to describe the tactile sensitivity for each group. Therefore, the score of tactile sensitivity ranged from 3 to 21, with 3 meaning that the participant has no sensory loss and 21 meaning that they have a sensory loss and no tactile information can be detected at the tip of their tested digits. In addition, the static 2-PD score was defined as the value in millimeters (1-20 mm) in which the participant was able to feel the presence of two points and the sum of those values obtained for each finger was used as the dependent variable to describe mechanoreceptor spatial resolution. Next, the participants were seated on a chair and instructed to perform three tests commonly used to evaluate hand function: nine-hole peg test (9HPT, RolyanTM 9-Hole Peg Test, Model A8515); Jebsen-Taylor hand function test (JTHFT, Sammons Preston Jebsen-Taylor Hand Function Test); and GSMax test using a hydraulic Jamar hand dynamometer (Sammons Preston). The tests were performed with the dominant and non-dominant hands. Half of the DPN individuals and their respective controls started the tests with their dominant while the other half started with their non-dominant hand.
Grip force control assessment The instrumented handle used in this study is shown in Figure 1A. This handle is formed by two aluminum bars linked with a compression-tension load cell (LPM530, Cooper Instruments and Systems, USA) and two aluminum pieces and a multiaxis force and torque (F/T) transducer (Mini40, ATI, USA) in between forming the basis of the handle. A cylindrical mass was attached underneath the handle to increase the total mass of the handle, which was in total 647 g (W=6.35 N). The handle aperture was of 5 cm and the external surface of it was covered with extra fine sandpaper (320
6
grit), which provided a moderate to high COF between the digits skin and the object surface. _____________________ Insert Figure 1 around here _____________________ To perform the holding task the seated participants were firstly asked to grasp and hold the free and vertically oriented handle using the tip of all digits (i.e., the thumb and the opposing four fingers). Participants were asked to remain with the arm positioned beside the trunk and shoulder with an internal rotation of 60 °, elbow flexed at 90°, forearm pronated to 90°, wrist in a neutral position, metacarpophalangeal joint bent at 90°, and interphalangeal joints in the neutral position. This body configuration enabled the participant to maintain the center of the handle aligned with their body midline and positioned just above the umbilicus. Posteriorly, the experimenter instructed the participants to hold the handle as stable as possible for 10 seconds as they would hold a glass of water. The experimenter started the data collection immediately after the participant was able to keep a stable handle position. Also, the participants were instructed to slowly reduce the magnitude of the GF until the handle drop off of the hand (slip test) after hearing a beep sound, which was provided at the 10th second of the trial recording. The slip test was performed in order to identify the minimum amount of GF required to hold the object (GF Min) and calculate the slip ratio (SR) [6,10,23]. Participants performed five trials. The first two was used for familiarization, principally due to the slip test, and the last three were used for posterior analyses. They performed this task only with their dominant hand (right) since we have evidence that there is no effect of handedness in this kind of task [6].
Data processing and analyses Two customized LabVIEW codes (National Instruments, USA) were used for data acquisition and processing. The force signals were recorded at 200 Hz and stored for later analyses. The raw force signals were low-pass filtered with a 4th-order, zero lag, Butterworth filter with a cut-off frequency of 20 Hz. LF was simply the weight of
7
the handle (i.e., 6.35 N) because the handle was kept stationary and the horizontal component acting tangentially was negligible. Moreover, GF exerted against the object surface was estimated by compression force recorded from the load cell (F C) fixed between the two flat aluminum plates. The holding phase lasted 10 seconds, but we analyzed just the central six, skipping the first two and last two seconds [6]. For holding phase, we computed the within trial coefficient of variation of the GF (CV-GF) shown in percentage of the averaged GF during the selected time interval (GF Mean). Next, the GF control was assessed by GFMean and the relative safety margin (SMRel) using the following equation: SMRel=100×((GFMean – GFMin)/GFMin), in which GFMin was obtained in the slippage phase [6,24]. In the slippage phase, we estimated GF Min and calculated the SR (SR=GFMin/LF). The point of slippage was determined as the point in time in which a sudden reduction of the vertical force component (F Z) recorded from the F/T transducer was detected due to the handle’s acceleration. Specifically, the point of slippage was automatically detected as the point in time in which F Z was lower than the FZ mean minus two FZ standard deviations, both calculated during holding phase. The GFMin was estimated from the point immediately before the moment of slippage.
Statistical Analyses After assuring normal distribution of all variables (Shapiro-Wilk test), we completed analyses of variance (ANOVA). For hand function assessment, we performed three two-way ANOVA, with the last factor (hand) treated as repeated measure, to test for differences between groups (DPN and controls) and hand (dominant and non-dominant) in the performance of 9HPT and JTHFT and in the GSMax. For the holding task, four one-way ANOVA were performed to test for differences between groups in SR, CV-GF, GFMean, and SMRel. In addition, two MannWhitney U tests were performed to test for group differences in the SWME and the static 2PD test scores. The level of significance for all analyses was set at 0.05.
8
Results
Tactile sensitivity and spatial resolution The results of the SWME revealed that individuals with DPN presented higher scores (i.e., lower tactile sensitivity) in the digits from the dominant hand when compared to controls (median±interquartile range, DPN: 9±1 – Controls: 5±3.5 | U=3.5, p<.001). Also, results of the 2-PD test revealed that individuals with DPN presented lower tactile spatial resolution than controls in the digits of the dominant hand (DPN: 16.5±7.75 – Controls: 10±4.5 | U=15, p<.01).
Strength and dexterity tests Table 2 depicts mean (SE) of the performance in 9HPT and JTHFT, and GS Max. The results revealed that individuals with DPN had worst performance in both dexterity tests
than
controls,
9HPT
[F(1,18)=16.3,p<.005,η2=.48]
and
JTHFT
[F(1,18)=14.4,p<.005,η2p=.45], but revealed no differences for GSMax (p>.05). Furthermore, no group by hand interaction was revealed (p>.05). The results also revealed an effect of hand on all three tests. Specifically, participants performed better with their dominant when compared to their non-dominant hands [9HPT: F(1,18)=29.1,p<.001,η2p=.62;
JTHFT:
F(1,18)=20.8,p<.001,η2p=.54;
and
GSMax:
F(1,18)=12.3,p<.005,η2 p=.41].
Grip force control Figure 2 depicts GF and LF time-series for a representative individual with DPN and his respective control. As can be observed, GF was much lower and closer to the GFMin for the individual with DPN when compared to his matched control. The holding
9
task was separated in two phases, holding and slippage. During the slippage phase, the only variable analyzed was the SR that did not differ between DPN and controls (p>.05). Regarding GF steadiness, the results revealed no difference between groups (p>.05) for CV-GF. However, in terms of GF control, the results revealed that both, the GFMean and the SMRel was smaller for individuals with DPN when compared to controls [GFMean: F(1,18)=7.72,p<.05,η2p=.3 | SMRel: F(1,18)=13.1,p<.005,η2p=.42]. Table 3 depicts means (SE) of SR, GFMean, SMRel, and CV-GF. _____________________ Figure 2 ______________________
Discussion The aims of this study were to evaluate and compare hand function (strength and dexterity) and GF control of DPN and healthy individuals. Regarding strength and dexterity, we hypothesized that GSMax and hand dexterity would be sensible to detect differences between DPN and controls. This hypothesis was partially confirmed since the results showed that individuals with DPN presented worst performance in JTHFT and 9HPT, but showed that individuals with DPN and controls produced similar GS Max. Recently, we found no difference between diabetic individuals with no signs of neuropathy and controls for hand and digits dexterity (JTHFT and 9HPT, respectively) and strength tests (i.e., GSMax) [6]. In the present study, we found that hand and digits dexterity is negatively affected by DPN but not strength. Unsurprisingly, deficits in hand and digits dexterity could be associated with the reduction in tactile sensitivity shown by individuals with DPN assessed in this study. The lack of significant difference between individuals with DPN and controls for the GS Max could be a sign that individuals with DPN assessed in this study still have a preserved and functional motor system. Also, the number of individuals with DPN tested in each group (n=10) could be responsible for the lack of difference between groups. A sample size calculation using DPN and control group GSMax mean and standard deviation, alpha
10
level of 0.05 and power of 0.8 indicated that 44 individuals in each group would be necessary to detect a significant lower GSMax in the DPN than in the control group. However, it is important to notice that the sample size calculation for this study was based on SMRel not in the GSMax. Regarding GF control, it was hypothesized that, due to their great sensory loss and the possible use of a compensatory GF control strategy, individuals with DPN would increase their SM by increasing GF while holding an object. This hypothesis was rejected as the results showed that individuals with DPN set much lower SM (≈44%) than controls during the free holding task. Based on the results of this study and on the previous one [6], it could be suggested that the individuals with no signs of neuropathy tested in our prior study would present mild neuropathy, but it would be concealed by the lack of signs and symptoms and by the preservation of hand function. Corroborating this assumption, we found no difference between the relative difference between individuals with DPN and their controls and diabetic individuals with no DPN and their controls (i.e., 43,34% and 48.64%, respectively) [6]. Both groups exerted a little less than half of GF exerted by the age- and sex-matched controls when steadily holding an object. Additionally, we could also claim that the DPN has a different effect on GF control than would have digital anesthesia [9,25,26], sensory deafferentation [27,28], and central neurological deficits such as multiple sclerosis, Parkinson’s disease, stroke, and cerebellar dysfunction [11-16], which are known to lead to an increase in the exerted GF. It would not be realistic to expect that individuals with DPN would have similar GF control than anesthetized controls. While DPN causes a chronic and mild to moderate sensory impairment, anesthesia causes an acute and severe loss on the ability to temporarily detect any tactile stimulus. Due to the chronicity of the DPN, it would be expected that individuals with DPN would compensate this sensory impairment and reduce its effects. Therefore, one would expect a stronger effect of anesthesia than DPN in GF control, but still would expect a greater SM in individuals with DPN than in controls. Similarly, it would be unfair to compare the GF exerted by a deafferented patient with no sense of touch, vibration and pressure and individuals with mild to moderate neuropathy. All individuals with DPN tested in this study were able to feel the 10 gf (approximately 1 N) monofilaments when it was applied to the digits. GF applied by individuals with DPN against the object was around 5 N. Thus,
11
they would be able to detect any significant changes in GF exerted against the surface of a hand-held object. A condition comparable to DPN would be carpal tunnel syndrome (CTS), which is considered an entrapment neuropathy. As individuals with DPN, individuals with CTS also present reduced tactile sensitivity and in more severe cases, they present muscle weakness. Some studies have shown that individuals with CTS apply higher GF than controls [17-19], but others have not observed this increase in GF in this population [29,30]. Therefore, the reduction in SM observed in individuals with DPN in this study would be unexpected. Hence, what would be the underlying cause of a lower GF exertion and consequent lower SM set by individuals with DPN when holding an object? In our previous study [6], we suggested that the mild sensory impairment of diabetic individuals with no DPN would not be sufficient to elicit the use of a compensatory GF control strategy, which would lead to a increased GF exertion, but would be enough to disrupt the proper use of sensory information in this diabetic individuals causing an error of the estimation of GF needed while steadily holding an object. It would be convenient to extrapolate this suggestion to diabetic individuals with DNP, with moderate and severe sensation loss in the tip of the digits. As mentioned earlier, despite having a reduced tactile acuity, the magnitude of this reduction in individuals with DPN would be lower than the reduction seen in healthy anesthetized and deafferented individuals. There would be still tactile stimulus providing residual information that could be used by the GF controller to detect changes in the digitsobject interaction, but yet not enough to assure an adequate estimation of the forces between the digits skin and object surface and, consequently, a proper GF control. Corroborating, a person from the DPN group who was the most affected in terms of tactile sensitivity, who presented the higher score in the 2-PD test, the worst performance in 9HPT and JTHFT, and the lower level of GS Max showed the GF-LF ratio and SMRel a little higher than the group averages, but her GF-LF ratio and SM was much lower than the averaged ratio and SM presented by the control group. Thus, we speculate that even if we had tested only diabetic individuals with severe neuropathy we would have found that individuals with severe neuropathy would present a lower ratio and SM than controls.
12
Despite being considered a peripheral neurological disease, evidences show that DPN also affects parts of the central nervous system. Specifically, studies found that individuals with DPN also have reduction of the spinal cord area in places such C2/3 [31] and C4/5 and T3/4 [32] when compared to individuals with no DPN and controls. Moreover, a recent study showed a reduction in gray matter volume in the primary somatosensory (S1), parts of the cingulate cortex and in the supramarginal gyrus in individuals with DPN when compared to controls [33]. Those cortical areas are closely related to the control and coordination of GF and LF [34]. Therefore, impaired tactile stimulus detection and neural conduction to the CNS, and altered tactile information processing in S1 could be detrimental to the formation of a new sensorimotor representation of a new object in individuals with DPN [35]. It could take longer to the GF controller of diabetic individuals with and without DPN to form a proper sensorimotor representation when dealing with novel objects and to implement a conservative GF control strategy. In order to test this hypothesis, a new study should be done in which participants with DPN and controls would perform a large number of trials using the same object. There are still some aspects of GF control in individuals with DM with and with no DPN that should be explored. For instance, the effect of performing dynamic tasks, such as lifting and shaking a handheld object, on GF control should be investigated in order to examine if the reduction in SM observed in diabetic individuals during the holding task remains similar or is task-dependent. It is important to mention that most of the studies investigating the effect of neurological diseases on GF control used dynamic tasks. Likewise, trial-to-trial GF adaptation as well as the GF adaptation to changes in object’s friction and weight should also be tested to assess the effect of the sensory loss on GF adaptation in diabetic individuals without and with diagnosis of DPN. Finally, the relationship between GF control variables and electrophysiological variables, such as sensory and motor neural conduction velocity, should be explored to a better understanding of the GF control and to determine the feasibility of using GF control variables in an early detection of DPN.
Conclusion
13
Based on the results of the present study we conclude that individuals with DPN present poorer hand and digits dexterity tests in spite of having their GS Max preserved. In addition, we conclude that individuals with DPN can still maintain a steady GF while holding an object, although, they present impairment in GF scaling. Specifically, they set a relatively low SM, which put them in risk of easily dropping handheld objects. This impairment of GF scaling may be predominantly caused by the reduction in tactile sensitivity, but the deficits in the sensorimotor transformation should also be considered. Taken together, these findings indicate that the study of GF control during object manipulation can be a promising venue for further understanding of the effect of DPN on movement control. More important, SM obtained from GF control studies and, specifically, during a simple holding task, could be seen as an easily obtained behavioral biomarker for early detection of neuropathy in individuals with DM even in situation in which no clinical signs of neuropathy is present [6]. In order to support this statement a larger and prospective study should be conducted.
Acknowledgments: The authors acknowledge the Sao Paulo State Research Foundation (FAPESP #2010/02939-4 and #2014/26397-7) and National Counsel of Technological and Scientific Development (CNPq, #479903/2013-1) for their financial support. Lima and Borges are thankful for their scholarships provided by Coordination for the Improvement of Higher Education Personnel (CAPES – Brazil).
References [1] ADA. (2013) Diagnosis and Classification of Diabetes Mellitus, American Diabetes Association. Diabetes Care;36.
14
[2] Van Acker K, Bouhassira D, De Bacquer D, Weiss S, Matthys K, Raemen H, Mathieu C, Colin IM (2009) Prevalence and impact on quality of life of peripheral neuropathy with or without neuropathic pain in type 1 and type 2 diabetic patients attending hospital outpatients clinics. Diabetes Metab. 35(3):206-13. [3] Watkins PJ, Thomas PK (1998) Diabetes mellitus and the nervous system. J Neurol Neurosurg Psychiatry 65 620-32. [4] Ramji N, Toth C, Kennedy J, Zochodne DW (2007) Does diabetes mellitus target motor neurons? Neurobiol Dis. 26:301-11. [5] Dahlin LB, Thrainsdottir S, Cederlund R, Thomsen NOB, Eriksson KF, Rosen I, et al (2008) Vibrotactile sense in median and ulnar nerve innervated fingers of men with Type 2 diabetes, normal or impaired glucose tolerance. Diabetic Med 25:543-9. [6] de Freitas PB, Lima KCA (2013) Grip force control during simple manipulation tasks in non-neuropathic diabetic individuals. Clin Neurophysiol 124:1904–10. [7] Gorniak SL, Khan A, Ochoa N, Sharma MD, Phan CL (2014) Detecting subtle fingertip sensory and motor dysfunction in adults with type II diabetes. Exp Brain Res. 232(4):1283-91 [8] Ochoa N, Gorniak SL (2014) Changes in sensory function and force production in adults with type II diabetes. Muscle Nerve 50(6):984-90. [9] Johansson RS, Westling G (1984) Roles of glabrous skin receptors and sensorimotor memory in automatic control of precision grip when lifting rougher or more slippery objects. Exp Brain Res 56:550-64. [10] Westling G, Johansson RS. (1984) Factors influencing the force control during precision grip. Exp Brain Res 53:277-84. [11] Fellows SJ, Noth J, Schwarz M (1998) Precision grip and Parkinson's disease. Brain. 1998;121:1771-84 [12] Nowak DA, Hermsdorfer J (2002) Coordination of grip and load forces during vertical point-to-point movements with a grasped object in Parkinson's disease. Behav Neurosci 116:837-50. [13] Hermsdorfer J, Hagl E, Nowak DA, Marquardt C (2003) Grip force control during object manipulation in cerebral stroke. Clin Neurophysiol 114:915-29. [14] Brandauer B, Timmann D, Hausler A, Hermsdorfer J (2010) Influences of load characteristics on impaired control of grip forces in patients with cerebellar damage. J Neurophysiol 103:698-708.
15
[15] Krishnan V, De Freitas PB, Jaric S (2008) Impaired Object Manipulation in Mildly Involved Individuals with Multiple Sclerosis. Motor Control12:3-20. [16] Iyengar V, Santos MJ, Ko M, Aruin AS (2009) Grip force control in individuals with multiple sclerosis. Neurorehabil Neural Repair 23:855-61. [17] Lowe BD, Freivalds A (1999) Effect of carpal tunnel syndrome on grip force coordination on hand tools. Ergonomics 42:550-64. [18] Zhang W, Johnston JA, Ross MA, Smith AA, Coakley BJ, Gleason EA, et al. (2011) Effects of carpal tunnel syndrome on adaptation of multi-digit forces to object weight for whole-hand manipulation. PLoS One. 6:e27715. [19] Chen P-T, Jou I-M, Lin C-J, Chieh H-F, Kuo L-C, Su F-C (2015) Is the Control of Applied Digital Forces During Natural Five-digit Grasping Affected by Carpal Tunnel Syndrome? Clin Orthop Relat Res 473:2371–82. [20] Oldfield RC (1971) The assessment and analysis of handedness: The Edinburgh Inventory. Neuropsychologia. 9:97-113. [21] Feldman EL, Stevens MJ, Thomas PK, Brown MB, Canal N, Greene DA (1994) A practical two-step quantitative clinical and electrophysiological assessment for the diagnosis and staging of diabetic neuropathy. Diabetes Care 17:1281-9. [22] Valk GD, Grootenhuis PA, Bouter LM, Bertelsmann FW (1994) Complaints of neuropathy related to the clinical and neurophysiological assessment of nerve function in patients with diabetes mellitus. Diabetes Res Clin Pract 26:29-34. [23] de Freitas PB, Uygur M, Jaric S. (2009) Grip force adaptation in manipulation activities performed under different coating and grasping conditions. Neurosci Lett 457:16-20. [24] Danion F. (2008) Grip force safety margin in rock climbers. Int J Sport Med 29: 168-72. [25] Monzee J, Lamarre Y, Smith AM (2003) The effects of digital anesthesia on force control using a precision grip. J Neurophysiol 89:672-83. [26] Carteron A, McPartlan K, Gioeli C, Reid E, Turturro M, Hahn B, Benson C, Zhang W (2016). Temporary Nerve Block at Selected Digits Revealed Hand Motor Deficits in Grasping Tasks. Frontiers in Human Neuroscience, 10, 596. [27] Hermsdorfer J, Elias Z, Cole JD, Quaney BM, Nowak DA. (2008) Preserved and impaired aspects of feed-forward grip force control after chronic somatosensory deafferentation. Neurorehabil Neural Repair 22:374-84.
16
[28] Nowak DA, Glasauer S, Hermsdorfer J (2004) How predictive is grip force control in the complete absence of somatosensory feedback? Brain127:182-92. [29] Thonnard J, Saels P, Van den Bergh P, Lejeune T (1999) Effects of chronic median nerve compression at the wrist on sensation and manual skills. Exp Brain Res 128:61-4. [30] Nowak DA, Hermsdorfer J, Marquardt C, Topka H (2003) Moving objects with clumsy fingers: how predictive is grip force control in patients with impaired manual sensibility? Clin Neurophysiol. 2003;114:472-87. [31] Selvarajah D, Wilkinson ID, Emery CJ, Harris ND, Shaw PJ, Witte DR, et al (2006) Early involvement of the spinal cord in diabetic peripheral neuropathy. Diabetes Care 29:2664-9. [32] Eaton SE, Harris ND, Rajbhandari SM, Greenwood P, Wilkinson ID, Ward JD, et al (2001) Spinal-cord involvement in diabetic peripheral neuropathy. Lancet. 2001;358:35-6. [33] Selvarajah D, Wilkinson ID, Maxwell M, Davies J, Sankar A, Boland E, et al (2014) Magnetic resonance neuroimaging study of brain structural differences in diabetic peripheral neuropathy. Diabetes Care 37:1681-8. [34] Ehrsson HH, Fagergren A, Johansson RS, Forssberg H (2003) Evidence for the involvement of the posterior parietal cortex in coordination of fingertip forces for grasp stability in manipulation. J Neurophysiol 90:2978-86. [35] Gordon AM, Westling G, Cole KJ, Johansson RS. (1998) Memory representations underlying motor commands used during manipulation of common and novel objects. J Neurophysiol 69:1789-96.
17
Figure Captions Figure 1: Free holding task illustration. (A) The participants held the handle as stable as possible for 10 seconds; (B) after hearing a beep, they slowly reduced the magnitude of GF until the object slippage (the slip test); (C) compression force (FC) recorded by the single-axis force transducer, and the vertical (F Z) and horizontal (FY and FX) force components recorded by the multi-axis F/T transducer.
Figure 2: Load (LF) and grip force (GF) time-series obtained during a free holding task from representative DPN and control participants. Dashed vertical line indicates the moment in which the beep sound was heard and GF started being reduced till slippage.
18
Figr-1
19 Figr-2
20
Table 1. Medical, physical, and anthropometric characteristics of the participants with diabetic peripheral neuropathy (DPN) and healthy controls. DPN (means ± SE) 10
Controls (means ± SE) 10
0
6
5.7 ± 0.9 7 7 3 3
0.5 ± 0.2 0 2 1 0
Physical and Clinical Aspects Age (years) Time since diagnosis of T2DM (years) Time since diagnosis of DPN (years) Smokers (n) Former smokers (n) Alcoholism (n) Carpal tunnel syndrome (n)
58.3 ± 1.3 16 ± 1.7 4.1 ± 0.5 1 4 2 0
58.2 ± 1.2 ----1 0 0 0
Blood Pressure Systolic (mmHg) Diastolic (mmHg)
129 ± 8 75 ± 3
115.1 ± 4 79.4 ± 2.3
Anthropometric Measurements Body mass (kg) Stature (m) BMI (kg/m2) Body fat (%) Lean mass (kg) Basal metabolic rate
81.5 ± 3.9 1.61 ± 0.03 30.8 ± 1 31.3 ± 2.5 56.6 ± 3.1 1721 ± 94
80.5 ± 4.2 1.67 ± 0.04 29.1 ± 1.4 35.8 ± 2.2 49.5 ± 2.8 1505 ± 85
Glycemic Control (n) Fasting capillary glycemia (mg/dL) Fructosamine (mmol/L)
9-10 141.6 ± 11.6 274.2 ± 34.8
4-10 113.8 ± 9.4 244.17 ± 17
N Medication Participants without medication Number of medications prescribed per person Participants on medication for diabetes Participants on medication for hypertension Participants on medication for dyslipidemia Participants on medication for neuropathy
21
Table 2. Mean (SE) values representing the performance of DPN and healthy controls in the nine-hole peg test (9HPT) and in the Jebsen-Taylor hand function test (JTHFT) and the maximum power grip strength (GSMax) of the dominant (DH) and non-dominant (NDH) hand.
Hand
Group
9HPT(s)
JTHFT(s)
GSMax(kgf)
DPN
19.39** (0.61)
33.63** (1.50)
32.90 (2.90)
Controls
15.73 (0.41)
26.56 (0.54)
37.90 (3.02)
DPN
20.61* (0.75)
35.70** (1.79)
31.40 (2.64)
Controls
17.72 (0.62)
29.41 (1.02)
34.90 (2.20)
DH
NDH
Note: * p<0.01 and ** p<0.001.
22
Table 3. Mean (SE) of slip ratio (SR), averaged GF exerted (GF Mean), relative safety margin (SMRel, in % of the GFMin), and coefficient of variation of the exerted GF (CVGF) when the DPN and control participants held the free moving handle with their dominant hand.
Group
SR
GFMean
SMRel (%of GFMin) CV-GF (%)
DPN
0.45 (0.01)
5.08* (0.31)
77.15** (8.43)
5.11 (0.63)
Controls
0.43 (0.02)
7.63 (0.86)
178.03 (21.93)
3.66 (0.5)
Note: *p<0.01 and **p<0.005.