Review Thecontribution of small and largesensory afferentsto postural control in patientswith peripheral neuropathy

Li Li*,Shuqi Zhang ,John Doson

a Collegeof Physical Education,Hunan Normal University,Changsha 410012,China

b Department of Health Sciencesand Kinesiology,Georgia Southern University,Statesboro,GA 30460,USA c Department of Kinesiology and Physical Education,Northern Illinois University,DeKalb,IL 60115,USA

Abstract Peripheral neuropathy(PN)isamultifariousdisorder that iscaused by damageto theperipheral nerves.Although thesymptomsof PN vary with the etiology,most cases are characterized by impaired tactile and proprioceptive sensation that progresses in a distal to proximal manner.Balance also tends to deteriorate as the disorder becomes more severe,and those aff licted are substantially more likely to fall while walking compared with thosewho arehealthy.Most patientswith PNwalk morecautiously and with greater stride variability than age-matched controls,but the majority of their fallsoccur when they must react to aperturbation such asa slippery or uneven surface.The purpose of thisstudy wasto f irst describe the role of somatosensory feedback in the control of posture and then discuss how that relationship is typically affected by the most common typesof PN.A comprehensivereview of thescientif ic literaturewasconducted using MEDLINE,and therelevant information wassynthesized.The evidence indicates that the proprioceptive feedback that is conveyed primarily through larger type Iafferents is important for postural control.However,the evidence indicatesthat the tactile feedback communicated through smaller type IIafferents isparticularly critical to the maintenance of balance.Many formsof PN often lead to chronic tactile desensitization in the soles of the feet and,although the central nervous system seems to adapt to this smaller type II afferent dysfunction by relying on more larger type I afferent ref lex loops,the result is still decreased stability.We propose a model that is intended both to help explain the relationship between stability and the smaller type IIafferent and thelarger type Iafferent feedback that may beimpaired by PN and to assist in thedevelopment of pertinent rehabilitativeinterventions.2095-2546/©2019 Published by Elsevier B.V.on behalf of Shanghai University of Sport.This is an open access article under the CCBY-NC-ND license.(http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords:Balance;Biomechanics;Neuroplasticity;Peripheral neuropathy;Postural control;Rehabilitation

1.Introduction

Peripheral neuropathy(PN)is a complex disorder that arisesfrom damageto≥1 peripheral nerves,and it isestimated to affect as much as 2.4%of the adult population and 8%-10%of those over the age of 55.1The majority of cases of PN aresecondary to a preexisting illness,themost common of which is diabetes mellitus,2but as many as 30%of cases are idiopathic.3In short,researchers have identif ied＞100 typesof PN,and the term,therefore,describesa highly diverse set of diseases that are characterized by awide variety of etiologies and pathologies.4However,many of the most common typesof PN,including diabetic PN(DPN),frequently result in specif ic functional impairment,which is a loss of balance that greatly enhances the risk of falling.5-13That is,although PN is a heterogeneous set of diseases that lead to many different forms of clinical presentation,the scope of this review is limited to the majority of types of the disorder that often result in functional impairmentsto balance.Indeed,the purposesof the review are to describe how and why balance is typically impaired with PN and to propose aconceptual model that may assist in the development of rehabilitative interventions for those with decreased postural stability that is caused by PN.To help unveil the nature of these impairments in individuals with PN,it is necessary to f irst elucidate the effect of PN on peripheral nerve function and the risk of falling and to then summarize postural control and how it is typically impacted by thedisorder.

2.PN and nerveconduction velocity

The def ining characteristic of PN is damage to the axons and/or myelin of peripheral nerves in a manner that typically results in abnormal conduction velocities and amplitudes.14Althoughα-motoneuron dysfunction frequently occurs,as is indicated by symptoms like muscle atrophy and strength loss,4,14chronic damage to the sensory nervous system occurs in＞85%of documented casesof PN.15It istherefore common for those with PN to experience both positive and negative sensory symptoms with the disease.Positive symptoms include the presence of sensations such as burning,tingling,and exaggerated pain responses(e.g.,allodyniaand hyperalgesia),whereas negative symptoms include the loss of tactile sensation,proprioception,and temperature sensitivity.14,16The nerve damage associated with most types of PN typically progresses in a distal to proximal manner,4such as from the foot sole to the ankle to the leg,which helps to explain why positive symptoms are regularly worse after long periods of weightbearing activity and negative symptoms are often described as numbness or“feet feel dead”.17Still,the clinical presentation of PN is often highly inconsistent,which is why nerve conduction velocity(NCV)is the leading assessment of sensory nerve impairment used in clinics4,18and epidemiologic studies.19-24

The standard sensory NCV test assesses the velocity and amplitude of action potentials in the sural,or short saphenous,nerve,which innervates the skin along the posterior aspect of the lower legs,ankles,and feet.25Given that nearly 30%of people with diabetes over the age of 40 have impaired sensation in their feet and hands,26it isnot surprising that amajority of thestudiesthat haveused sural NCV to monitor the progression of PN have done so in those with DPN.For example,Clauset al.20demonstrated that sural NCV diminished approximately 0.5 m/s each year in those with DPN,and Jarmuzewska and Ghidoni21reported that sural NCV decreased an average of 3.9 m/s every 10 years in patients diagnosed with type IIdiabetes.21Decreased sural NCV has also been linked with impaired glycemic control,24abnormal sensations,22and decreased quality of life23in this population.In light of these and other important pertinent studies that used sensory NCV,it isimportant to remember that PN is a disease unto itself that has many different causes and is associated with pathologic processes in various combinations of sensory nerve f ibers.However,most cases of PN do involve pathology in the smaller sensory f iberslike the types II,III,and unmyelinated 4 that transmit cutaneous sensations like touch,sharp pain,and temperature.27,28By contrast,sensory NCV is a measure that is limited to large diameter nerves,and at least 1 study29hasdemonstrated that PN can result in signif icant degeneration in sural nerve f iber density without a decrease in sural NCV.Therefore,small f iber involvement is at least to some degree independent of large f iber involvement with most cases of PN.Sensory NCV,which isconsidered the gold standard diagnostic technique,may not adequately assess the degeneration of the smaller diameter nerves that often occurs at the earliest stages of the disorder.30

3.PN and therisk of falling

Balance may be described as the dynamics of body posture to prevent falling,31and the risk of falling,in turn,can be predicted by one's ability to control postural sway and center of pressure(COP)while standing.32A particular concern with many of the most common types of PN is that balance tends to deteriorate as the disease becomes more severe.8,9,11,12The resulting def icits in sensory feedback lead to well-documented increases in postural sway while standing,5,10,33including exaggerated COP outcomes such as 95%area of COPand velocity of COPmovement.34Furthermore,many individuals with PN tend to perform more poorly ontestslikethe6-Minute Walk and Timed Up-and-Gotests,which are tests of functional mobility that highly correlate with standing balanceand areused clinically to predict therisk of falling.10,35

The majority of falls in those with PN occur while they are walking,7and individuals with PN are 15 times more likely to experience an injury while walking than age-matched participants with intact sensation.5The predictive factors that are associated with an increased risk of falling in the elderly are the relative measures of dynamic stability in walking,including variability of stride-to-stride,step lengths,and step widths.36Although these increased measures of variability associated with PN are dueto slow walking speedsand arenot directly related to sensory loss,it is possible that years of loss of peripheral sensation and fear of falling cause thoseindividuals to self-select slower walking speeds.7Those aff licted with the most common types of PN often do walk cautiously,as is indicated by their signif icantly decreased speed,7,37-39step lengths,39ankle moments,ankle powers,and ground reaction forces.40Similar walking alterations have also been observed in healthy individuals with experimentally decreased plantar cutaneous sensation(e.g.,using ice immersion),but individuals with PN exhibit persistent variability on those measures.41However,it is important to emphasize that most patients with PN can generate relatively normal and stable locomotory behavior,and the majority of the falls they experience occur when they need to quickly react to perturbationssuch asirregular surfaces or unexpected objects.6,7,13That is,the ability to detect postural changes and make corrections to COP after a perturbation is diminished because it depends on a complex response involving cutaneous and proprioceptive sensory receptors,as well as both small and large sensory f ibers,which may be impaired in those with PN.Those diff iculties in responding to perturbations,along with the observed differences between individuals with pathologically versus experimentally reduced sensation,suggest that the most common causes of PN impair not just specif ic cutaneous receptors or sensory f ibers,but all peripheral sensory systems.42In light of that information,we f irst provide an overview of the relationship between postural control and somatosensation and then describe how that relationship is typically impaired in those with PN.

4.Overview of postural control

Postural control may be def ined as the act of achieving,maintaining,or restoring a state of balance during any posture or activity,43and it depends on a combination of both passive and active mechanical controls.31,44Passive control refers to the stiffness and kinematic proprieties associated with the pertinent anatomical structures(e.g.,bones and other components of the joints),as well as the effect that gravity exerts on them,whereas active control describes the nervous regulation of skeletal muscle in a manner that requires energy expenditure.45Passive control helps to explain phenomena like the consistency of postural control observed across many different types of tasks and the decline in postural stability that may occur with muscle fatigue in the lower extremities.However,active control is responsible for sway detection and postural correction,45,46and it is critical to our ability to stabilize and maintain balance while standing and walking.47,48

Active postural control depends on a complex interaction between the joints,skeletal muscles,and both the peripheral nervous system and the central nervous system(CNS).The functional role of the nervous system in active control may be subdivided into 4 components:stimulation collection via sensory receptors,afferent signaling via sensory neurons,CNS control of information processing and decision making in the CNS,and efferent signaling to skeletal muscles viaα-motoneurons.The latter 2 are the sole components used in feedforward control, which is accomplished using internal preprogrammed models that are based on anticipation.49By contrast,feedback control involves modif ication of ongoing movement using the information that is gathered by sensory receptorsand transmitted to the CNSby sensory neurons.Consequently,feedback control allowsfor ahigher degreeof accuracy becauseit isbased on error detection and correction,but it is also necessarily slower than feedforward control.Optimal postural control dependson acombination of both feedforward and feedback processes.50

Themechanicsof postural control during standing areoften described using an inverted pendulum model,and the goal of control is to maintain the COP,the weighted average of all pressures over the area that is contacting the ground,about the base of support.31Simply put,we naturally sway as we stand,and our stability depends on our ability to sense,control,and correct those movements.Postural control during walking is,of course,quite different because the goal is to actually move outside the base of support and yet maintain stability from one stride to the next.Two popular theories of postural control of gait are passive dynamic walking(PDW)and a central pattern generator.PDW develops from a simple mechanistic model in which gait is a natural repetitive motion that is generated by gravity and inertia.51Under the frame of PDW theory,segmental inertia and joint stiffness account for most of the control for walking,and the role of the nervous system is to provide more guidance than overt control.47By contrast,the central pattern generator theory depends heavily on the nervous system and feedforward control,because walking is considered a rhythmic movement that is preprogrammed at the upper level of the spinal cord.52According to thistheory,sensory feedback isimportant to the control of posture during the stance phase of walking while just 1 foot is on the ground,but it is less important during the swing phase.42In fact,studies in quadrupeds suggest that locomotion can occur in the absence of afferent inputs53,54or even a cerebrum.Nevertheless,bipedal gait is consistently less stable than that in quadrupeds,and it is generally presumed to require some level of feedback control,55particularly during perturbations.Similarly,active nervous control and sensory feedback are also required within the PDWmodel to optimizelateral balancein thegait45and to correct errors.56To summarize,the evidence indicates that sensory feedback is required to respond to perturbations and maintain posture while standing and walking.

5.Postural control and somatosensation

The sensory receptors and afferent neurons that are most critical to providing information about the difference between current posture and upright position are those associated with the visual,vestibular,and somatosensory systems.57All 3 of those sensory systems contribute to the maintenance of balance at all times,57but they are weighted differently according to the specif ic task.58-61For example,the somatosensory and visual systems provide suff icient sensory information to maintain balance during quiet standing with eyes open and feet shoulder width apart,62whereas the vestibular system is more signif icantly involved while balancing on an unstable platform.63That said,the somatosensory system is of particular interest to this review because it provides the most accurate information to assist postural control,57and it is the sensory system that ismost often impaired by PN.14,27,28

Somatosensation refersto feedback from the body surface and its interaction with the external environment,and it includes the proprioceptive and tactile subdivisions.The tactile subdivision pertains mostly to cutaneous sensations such as touch,pressure,and vibration,while the proprioceptive includes muscle spindles and Golgi tendon organs that contribute to the detection of joint position and joint motion.64,65More specif ically,Golgi tendon organs monitor muscle loading and their information isconveyed through type Ib sensory neurons,whereas muscle spindles provide feedback about both dynamic and static muscle length through large type Ia and IIsensory neurons,respectively.By comparison,smaller diameter sensory neurons are responsible for all tactile sensations,including some information about touch that uses the type III neurons that are particularly susceptible to PN.27,28The 4 main tactile receptors in the skin include Merkel's cells,Pacinian corpuscles,Meissner'scorpuscles,and Ruff ini endings.66,67Meissner's and Pacinian corpuscles are rapidly adapting receptors that are responsible for vibrotactile sensation,whereas Merkel'scellsand Ruff iniendingsadapt slowly and are responsible for touch and pressure sensitivity.Becauseslowly adapting receptorsbetter retain their sensitivity throughout continuous stimulation,Merkel's cells and Ruff ini endings likely provide more important tactile feedback for postural control during slow movements and quiet standing.68,69

In general,somatosensory information is thought to inf luence static stability in standing and dynamic stability primarily by affecting the activities of lower leg muscleslike the tibialis anterior and soleus,aswell asby mediating gait patternsat the ankle,knee,and hip joints.70Wediscusstheseitemsin thefollowing text:the specif ic roles that proprioceptive and tactile sensations play in feedback control of posture and how those relationships may be affected by PN.However,it is worth f irst noting that many of the studies that have investigated the effect of somatosensory dysfunction on postural control during standing and walking have not done so with the elderly population,who are more likely to have decreased postural stability with degenerative neurologic disorders.6,71Instead,many of those studies have used healthy adultswhose somatosensation wasdecreased using soft or moving supporting surfaces,ischemic injections,mechanical vibratory stimuli,or inf lated blood pressure cuffsat the ankle or thigh.72-76

6.Roleof ankleproprioception and stretch ref lex postural control

Much of the research on the proprioceptive inf luence of postural control has focused on the ankle proprioceptors and pertinent stretch ref lexes because the ankle-foot complex is the only part of the body that contacts the ground and it is the site in which most postural sway occurs.77In addition to the inf luence of proprioception,ankle joint ligaments and the surrounding muscles could also contribute to ankle joint stability.44,78,79The 3 most likely contributors to stability reduction and enhanced sway at theanklejointsareadecreasein muscular strength of the ankle evertors,an increase in ligamentous laxity,and proprioceptive def icits resulting from a disruption in the integrity of the receptors.80-83Because the stiffness of muscles and ligaments around the ankle joint alone cannot achieve joint stabilization,78ankle proprioception is believed to be a critical determinant of functional joint stability,78,84which,in turn,may inf luence postural stability in standing and walking.Indeed,studies by Fu and Hui-Chan85and Jerosch and Prymka,86who conducted joint reposition tests after ankle injury,demonstrated a high correlation between joint stability and ankle proprioception.In addition,Lee and Lin87reported that 12 weeksof biomechanical ankleplatform system training improved joint and postural stability in conjunction with enhancements in ankle proprioception.

In light of the correlations discussed,very few studies have directly investigated theimportanceof ankleproprioception in postural control owing,in part,to the diff iculty of experimentally inducing temporary dysfunction in thepertinent receptors without affecting other sensory receptors.In 1 such study by Hertel et al.,88the investigators anesthetized portions of the ankle and then analyzed postural sway under both static and dynamic conditions.The results indicated that the anesthesia treatment did adversely affect joint proprioception,but the reduction of joint sensory input did not affect postural sway.In a similar study,De Carlo and Talbot89examined dynamic stability using a multiaxial platform,and they also reported no difference between anesthetized and unanesthetized ankles.One could argue that the observations of both of those studies were limited by the fact that the injections they used produced uneven or incomplete decreased in ankle proprioception.Most types of PN are not like acute desensitization because those aff licted with progressive forms of the disease for a longer period adapt to ankle joint proprioception through neuroplasticity.A more recent study did study patients with DPN with conf irmed lower leg proprioceptive dysfunction,and those investigators reported no differences in balance-correcting responses between patients and healthy controls.90However,that study did not provide any information regarding participants'foot sole cutaneous sensation,which is known to be an important component of postural control.91It is possible that some compensation from other sensory divisions(e.g.,foot sole cutaneous sensation)masked the importance of proprioception in the studies described,and proprioceptive feedback continuesto be considered an important component of postural control.9,77,92Clearly,the precise role that ankle and lower leg proprioceptors play in the control of balance during standing and walking hasyet to befully elucidated.35

Another important consideration with the relationship between ankle proprioceptors and postural control is the stretch ref lex and the information it can provide about the connection between large afferent f ibers(LAF),the CNS,and α-motoneuron stimulation of skeletal muscle.Interneurons within the spinal cord elicited this ref lexive stimulation of muscle contraction in response to feedback from the muscle spindles.93Proprioceptive feedback also travelsup to the cerebellum that,in turn,can modify the sensitivity and excitation of the spinal interneurons in a manner that helps to control muscle tension to maintain posture and locomotion.94,95The sensory feedback provided by spindles and their contribution to the stretch ref lex arc is divided into primary and secondary components.Primary spindle f ibers convey feedback about the velocity of muscle length changes using large-diameter type Iαsensory neurons,whereassecondary f ibersprovideinformation about static musclelength using smaller type IIneurons.

Among the more important musclesfor postural control are those that dorsif lex(tibialis anterior)and plantar f lex(gastrocnemius)the ankle,including the soleus muscles that are critical agonists during both standing and the push-off phase of gait.96The soleus stretch ref lex is necessary to both inhibit plantar f lexion during the swing phase of locomotion and provide excitation during the stance phase,and it is thought to help correct balance when responding to perturbations and unexpected stretching of the plantar f lexors.93,95,97When stretching the soleus in a seated position(i.e.,with unloaded soleus),the resulting stretch ref lex produces 2 bursts of afferent activity with different latencies.The burst with the shorter latency has an onset of approximately 40 ms and is attributed to the excitation of the primary spindle f ibers and type Iαsensory afferents,97,98which is why it has been described as the LAFref lex loop.The other burst has a latency of about 70 ms and is associated with the type IIafferents that originate from secondary spindle endings;99therefore,it is often called the small afferent f iber(SAF)ref lex loop.Both stretch ref lex loops and typesof sensory afferentsarethought to contributeto postural control during standing and walking,93,100,101but the SAF and ref lex loop are thought to be more important.66,99,100,102

The Hoffman ref lex(H-ref lex)is a ref lective skeletal musclecontraction that occursin response to an electrical stimulation of the sensory afferents that are associated with the spindles.Although H-ref lex and stretch ref lex arenot identical,the H-ref lex is a common tool used to estimate the function of the stretch ref lex because they are both dependent on the same afferent neurons andα-motoneurons,as well as the interneurons that connect them.103,104As compared with the stretch ref lex,the latency of the H-ref lex indicates the eff iciency of the synaptic transmission between the afferents andα-motoneurons,and the amplitude of the H-ref lex ref lects the excitation level of theα-motoneurons.Also like the stretch ref lex,the CNSaltersthe latency and amplitudeof the H-ref lex when the brain modif ies the sensitivity and threshold of excitability of the spinal interneurons.105,106One of the advantages of the H-ref lex is that it is less inf luenced by joint motions and the activities of other peripheral sensory receptors;consequently,it is often used to investigate central adaptive neuroplasticity during interventional studies.107The ratio between the amplitude of the H-ref lex(H-wave)and the amplitude of the depolarization in theα-motoneuron that is distal to the electrical stimulation(M-wave)is also commonly used as an index for estimating the level of ref lex excitability of the motor pool.106,108

Capaday and Stein105investigated the inf luence of posture on the H-ref lex in the soleus,and they reported variances between standing and walking that indicated differences in CNScontrol.While standing with relatively small leg muscle activity,body sway results in relatively larger H-wave amplitudes and intense stretch ref lexes to counteract the sway and maintain stability.By contrast,the amplitudes of the H-ref lex are generally smaller during walking,but they do vary between the swing and stance phases.Walking requires more complianceand lessrigid control of theanklethan standing,109and the smaller amplitudes of the H-ref lex during walking are partly due to the relaxation of the soleus throughout the swing phase.Thestronger modulation of the H-ref lex during walking is not simply a passive effect of theα-motoneuron excitation level,it indicatesthat sensory feedback modif iesthe CNScontrol at the mean time.

7.Role of foot sole sensation in postural control

Because at least 1 foot isalways in contact with the ground during standing and walking,thecutaneoustactilereceptorsin the soles of the feet provide constant feedback about the surface characteristics of the terrain and whether it becomes slippery,unstable,irregular,and so on.Additionally,foot sole sensation(FSS)is important to postural control because it helpsto inform the CNSasto how thebody massand the COP aremoving relativeto thebase(s)of support.Plantar cutaneous feedback isalso alogical placeto investigate theenhanced risk of falls that occur with PN because the lossof FSSis often one of the earliest and most obvious clinical signs of the disease.91,106,110

Numerous investigators have reported that the feedback from the cutaneous receptors in the soles helps to regulate postural sway,41,65,91,106,111-114but Nardone et al.66-68have conducted some of the key studies.These investigators examined body sway area during quiet stance in patientswith either Charcot-Marie-Tooth(CMT)type 1A,CMT type 2,or DPN.CMT type 1A is a neurologic disease that impairs the function of type Iaand larger diameter type IIsensory neurons,whereas CMT type 2 and DPN both cause additional impairment to the smaller type IIβneurons.The investigators reported that the patients with CMT type 1 were able to stand upright normally,but those with DPN or CMT type 2 had decreased postural stability.These observations indicate that tactile sensory feedback is critical to postural control during standing,66,68especially feedback about touch and pressure that is detected by Merkel's cells and Ruff iniendings and then conveyed through smaller diameter type IIneurons.67

Another important measure that is used to help understand theroleof FSSin thecontrol of posture,and how that relationship may be affected by PN,is the distribution of force over the foot sole,or plantar pressure distribution.Numerous studies have demonstrated that plantar pressure distribution is altered in healthy individuals with experimentally reduced FSS,11,41,110-112,115,116as well as in patients with PN.91,106,115,117Those alterations have typically consisted of shifts in COP away from the toes and toward the midfoot,41,110,111,115but shiftsaway from specif ic regionsof insensitivity have also been described.91,112Still,it is important to note that not all pertinent studies have produced similar results.For example,somemorerecent investigationsreported that targeted decreases in FSS using anesthetic injections failed to affect plantar pressure distribution,118and,perhaps more important,did not impair dynamic stability.119,120Although the inconsistent observations across these studies may be explained by differences in experimental methods and theextent to which sensation wasdecreased,34,116,118it isclear that more investigation is required to fully understand how changes in FSSaffect the plantar pressure distribution and the basic characteristics of gait.Nevertheless,studies involving both patientswith PN10,121and healthy individualswith experimentally decreased sensation41havedemonstrated that reductions in FSS do lead to slower and more cautious patterns of walking.It is also relevant that Perry et al.69have shown that FSSis important to the maintenance of posture when perturbation evokescompensatory stepping.

8.Sensory reweighting and PN

An intact somatosensory system is thought to provide the most accurate information to assist postural control,57but it has been established that alternative sources of sensory information can be used to compensate for those who have been impaired by disease or destabilizing environments.122-124Regarding postural control,sensory reweighting occurs when the CNS uses one type of sensory stimulus that is coupled to the control of balance(upweighted)to compensate for another weakened stimulus.125,126Somatosensory reweighting can occur acutely,such as while walking blindfolded or with experimentally decreased somatosensation,or it may be prolonged by neuroplastic changes to the CNSin response to chronic impairments occurring with diseases such as PN.Although the exact nature of the neuroplastic adaptations is not yet clear,studies have demonstrated that they occur in the spine,127supraspinal areas,128,129and cerebellum.37What is clear is that there are differences between acute and chronic sensory reweighting;consequently,we should be careful when comparing postural responses to acutely versus chronically decreased somatosensation because they may involve different compensatory strategies.For example,the distinctions between thetactile and proprioceptivesystemsthat areevident in healthy individuals with experimentally decreased somatosensation are not present in those with the chronic sensory adaptations that are caused by the most common types of PN.34

One measure that can help to elucidate the impact of somatosensory reweighting on postural control in many of those with PN versus healthy individuals with experimentally reduced sensation isthe H-index.The H-index isavariation of the H-ref lex,and it providesa normalized time course between the onset of the M-wave and the onset of the H-wave relative to an individual'sheight:130

The H-index represents the entire arc of the type I LAF ref lex loop,including the synapsesof thespinal cord that integrate peripheral sensory information and are affected by chronic reweighting.130,131The H-index has been shown to correlate with other measures of balance,and it is considered to be both a helpful tool for diagnosing neurologic impairments132and areliablemeasurefor individualswith PN.117

Although both the LAF ref lex loop and the smaller type II afferent(SAF)ref lex loop are thought to be important to the control of posture,93,100,101the latter is generally considered to play a more signif icant role.66,99,102Furthermore,the decrease in FSSthat often occurswith themore common typesof PN is associated with impairment to the SAF ref lex loop and is thought to diminish postural control.11,66,67What is less well-known is how the decrease in FSS and chronic sensory adaptations that may occur with chronic forms of PN affect the relationship between SAFand LAFref lex loopsin the control of posture.We have recently investigated this relationship106by comparing postural control and the H-index in patientswith theplantar cutaneoussensation that wasimpaired by chronic PN versus age-matched controls.The results indicated that the individuals with PN had a decreased H-index,greater postural sway,and impaired functional mobility.There was also a signif icant correlation between the H-index and postural sway in those with PN,but not in the controls that exhibited normal FSS.These observations indicate that the LAF ref lex loop moderates postural control for those with impaired plantar cutaneous sensation.That is,balance control may depend more on LAF ref lex loops in those with PN,and sensory reweighting may allow their LAFloop and proprioception to compensate for their smaller f iber degeneration and impaired cutaneous sensation.For example,Dixit et al.133recently provided indirect evidence of sensory reweighting in individuals with DPN.After 8 weeks of aerobic exercise training,participants had improved control of the COP movement while quietly standing on a foam surface with their eyes closed,which indicates the proprioceptive adaptation occurred without visual feedback.

9.Conceptual model based on thisliterature

To encapsulate the common effect of chronic PN on the postural control that is discussed above,we propose the following conceptual model to describe the relationship between stability and the SAFand LAFfeedback that isoften impaired by thedisease(Fig.1).Imagine that an individual is standing on a platform that is supported by both LAFs near the center and SAFs around theperimeter.Thisconceptual model showsthe relationship between the functions of LAFs and SAFs in the development of PN.Impaired LAF function will not threaten the balance of the system,as long as the SAFs remain healthy and function normally,because the balance of the system is mainly supported by the columns in the perimeter(the SAFs).In contrast,the system will be less stable if the SAFs(pillars at the perimeter)are impaired,but the stability decrease may not be clinically evident if the LAFs(support columns in the center of the platform)are healthy.However,the decrease in stability would become apparent if the system with SAF impairment is challenged by an external or internal perturbation,because the LAFsprovideamuch smaller baseof support.Such aconceptual model can help us to explain why individuals with PN-induced impairments in SAF function and tactile sensation,compared with people with only LAF impairments,are far more likely to becomeunstableand fall when they encounter perturbations.

Fig.1.Current understanding of the functions of large afferent f ibers(LAF)and small afferent f ibers(SAFs)in relation to postural control.SAFs plays an important role in the feedback process for postural control,whereas LAFs becomemoreimportant with impaired SAFs.

10.Conclusion

To quickly review some of the highlights discussed herein,many types of PN typically include degeneration and dysfunction in the distal sensory neurons,4,15especially those that transmit tactile sensations like touch.10,27,28,116As the disease progresses and becomes more severe,balance deteriorates8-12and the risk of falling and sustaining an injury while walking increases substantially.5Individuals with these forms of PN walk more cautiously and with greater stride variability than those with intact somatosensation,7,37-39and most of their falls occur as a result of their impaired ability to react to perturbations such as slippery surfaces and unexpected obstacles.6,7,13One of the most important determinants of postural control is the cutaneous tactile feedback that is transmitted by SAFs,66-68particularly that at the solesof the feet.The decreased FSSthat typically occurs with PN17,114,134leads to cautious walking10,41,121and has been shown to inhibit the recovery of balance after perturbations.69Finally,the evidence indicates that patients with PN may compensate for their impaired FSSthrough a greater coupling of postural control to proprioceptivefeedback and the LAFref lexiveloop.106

PN is a complex disorder that is often diff icult to control,and most medical treatments are focused on decreasing pain rather than decreasing the increased risk of falling that frequently accompanies the most common types of the condition.Our recent observations106suggest that improved LAF ref lex function might enhance postural control in those who have impaired SAF function.Some intervention studies have already shown that exercise can improve the function of the LAF ref lex loop in athletes and elderly adults.135,136Studies have also demonstrated that routine exercise can help individuals with PN to improve strength and balance,137-139reaction time and the risk of falling,140and FSSand functional gait.141Future pertinent studies should continue to investigate the CNS adaptations that affect postural control,and they should continue to explore how exercise affects those adaptations and how it improvesbalancein thisclinical population.

Authors'contributions

LL initiated the concept and designed the basic structure of the review paper;SZ contributed to the main structure of the manuscript and provided most of the references;JD drafted the manuscript.All 3 authors contributed to the editing and f inalization of the manuscript.All authors have read and approved the f inal version of the manuscript,and agree with theorder of presentation of theauthors.

Competinginterests

The authorsdeclare that they haveno competing interests.

References

1.Martyn CN,Hughes RA.Epidemiology of peripheral neuropathy.JNeurol Neurosurg Psychiatry 1997;62:310-8.

2.Mold JW,Vesely SK,Keyl BA,Schenk JB,Roberts M.The prevalence,predictors,and consequences of peripheral sensory neuropathy in older patients.JAm Board Fam Pract 2004;17:309-18.

3.Smith AG,Singleton JR.Idiopathic neuropathy,prediabetesand themetabolic syndrome.JNeurol Sci2006;242:9-14.

4.National Institute of Neurological Disorders and Stroke.Peripheral neuropathy fact sheet.Available at:https://www.ninds.nih.gov/Disorders/Patient-Caregiver-Education/Fact-Sheets/Peripheral-Neuropathy-Fact-Sheet ;2012[accessed 13.05.2018].

5.Cavanagh PR,Derr JA,Ulbrecht JS,Maser RE,Orchard TJ.Problems with gait and posture in neuropathic patients with insulin-dependent diabetesmellitus.Diabet Med 1992;9:469-74.

6.DeMott TK,Richardson JK,Thies SB,Ashton-Miller JA.Fallsand gait characteristics among older persons with peripheral neuropathy.Am J Phys Med 2007;86:125-32.

7.Dingwell JB,Cavanagh PR.Increased variability of continuous overground walking in neuropathic patients is only indirectly related to sensory loss.Gait Posture2001;14:1-10.

8.Geurts AC,Mulder TW,Nienhuis B,Mars P,Rijken RA.Postural organization in patients with hereditary motor and sensory neuropathy.Arch Phys Med Rehabil 1992;73:569-72.

9.Kavounoudias A,Roll R,Roll JP.The plantar sole is a“dynamometric map”for human balancecontrol.Neuroreport 1998;9:3247-52.

10.Manor B,Li L.Characteristicsof functional gait among people with and without peripheral neuropathy.Gait Posture2009;30:253-6.

11.Meyer PF,Oddsson LI,De Luca CJ.Theroleof plantar cutaneoussensation in unperturbed stance.Exp Brain Res2004;156:505-12.

12.Rogers MW,Wardman DL,Lord SR,Fitzpatrick RC.Passivetactilesensory input improves stability during standing.Exp Brain Res 2001;136:514-22.

13.Stolze H,Klebe S,Zechlin C,Baecker C,Friege L,Deuschl G.Falls in frequent neurological diseases.JNeurol 2004;251:79-84.

14.Azhary H,Farooq MU,Bhanushali M,Majid A,Kassab MY.Peripheral neuropathy:differential diagnosis and management.Am Fam Physician 2010;81:887-92.

15.Padua L,Schenone A,Aprile I,Benedetti L,Caliandro P,Tonali P,et al.Quality of life and disability assessment in neuropathy:a multicentre study.JPeripher Nerv Syst 2005;10:3-10.

16.Apfel SC,Asbury AK,Bril V,Bruns TM,Campbell JN,Chalk CH,et al.Positiveneuropathic sensory symptomsasendpointsin diabetic neuropathy trials.JNeurol Sci2001;189:3-5.

17.Boulton AJ,Malik RA,Arezzo JC,Sosenko JM.Diabetic somatic neuropathies.Diabetes Care2004;27:1458-86.

18.Richardson JK.Factorsassociated with fallsin older patientswith diffuse polyneuropathy.JAm Geriatr Soc 2002;50:1767-73.

19.Behse F,Buchthal F,Carlsen F.Nerve biopsy and conduction studies in diabetic neuropathy.JNeurol Neurosurg Psychiatry 1977;40:1072-82.

20.Claus D,Mustafa C,Vogel W,Herz M,Neund¨orfer B.Assessment of diabetic neuropathy:def inition of normal and discrimination of abnormal nervefunction.Muscle Nerve1993;16:757-68.

21.Jarmuzewska EA,Ghidoni A.Study of the onset and progression of peripheral neuropathy and hypertension in NIDDM.Minerva Med 2000;91:1-5.

22.Løseth S,Lindal S,Sta°lberg E,Mellgren SI.Intraepidermal nerve f ibre density,quantitative sensory testing and nerve conduction studies in a patient material with symptoms and signs of sensory polyneuropathy.Eur JNeurol 2006;13:105-11.

23.Padua L,Saponara C,Ghirlanda G,Padua R,Aprile I,Caliandro P,et al.Lower limb nerve impairment in diabetic patients:multiperspective assessment.Eur JNeurol 2002;9:69-73.

24.Tkac I,Bril V.Glycemic control is related to the electrophysiologic severity of diabetic peripheral sensorimotor polyneuropathy.Diabetes Care1998;21:1749-52.

25.Umphred DA,Lazaro RT,Roller M,Burton G,editors.Neurological rehabilitation-E-book.Edinburgh:Elsevier Health Sciences;2013.

26.Centers for Disease Control and Prevention.Nation diabetes fact sheet:national estimates and general information on diabetes and prediabetes in the United States;2011.Available at:http://www.cdc.gov/diabetes/pubs/pdf/ndfs_2011.pdf ;[accessed 13.05.2018].

27.Baron R.Mechanisms of disease:neuropathic pain—a clinical perspective.Nat Rev Neurol 2006;2:95-106.

28.Talbot S,Couture R.Emerging role of microglial kinin B1 receptor in diabetic pain neuropathy.Exp Neurol 2012;234:373-81.

29.Periquet MI,Novak V,Collins MP,Nagaraja HN,Erdem S,Nash SM,et al.Painful sensory neuropathy prospective evaluation using skin biopsy.Neurology 1999;53:1641-7.

30.Lacomis D,Giuliani MJ,Steen V,Powell HC.Small f iber neuropathy and vasculitis.Arthritis Rheumatol 1997;40:1173-7.

31.Winter DA.Biomechanics of normal and pathological gait:implications for understanding human locomotor control.J Mot Behav 1989;21:337-55.

32.Sheldon JH.The effect of age on the control of sway.Gerontol Clin 1963;5:129-38.

33.Simoneau GG,Ulbrecht JS,Derr JA,Becker MB,Cavanagh PR.Postural instability in patients with diabetic sensory neuropathy.Diabetes Care 1994;17:1411-21.

34.Kars HJ,Hijmans JM,Geertzen JH,Zijlstra W.The effect of reduced somatosensation on standing balance:a systematic review.J Diabetes Sci Technol 2009;3:931-43.

35.Manor B,Doherty A,Li L.Thereliability of physical performancemeasuresin peripheral neuropathy.Gait Posture2008;28:343-6.

36.Maki BE.Gait changesin older adults:predictorsof fallsor indicatorsof fear.JAm Geriatr Soc 1997;45:313-20.

37.Manor B,Newton E,Abduljalil A,Novak V.The relationship between brain volumeand walking outcomesin older adultswith and without diabetic peripheral neuropathy.Diabetes Care2012;35:1907-12.

38.Menz HB,Lord SR,Fitzpatrick RC.A tactile stimulus applied to the leg improves postural stability in young,old and neuropathic subjects.Neurosci Lett 2006;406:23-6.

39.Richardson JK,Thies SB,DeMott TK,Ashton-Miller JA.A comparison of gait characteristics between older women with and without peripheral neuropathy in standard and challenging environments.J Am Geriatr Soc 2004;52:1532-7.

40.Mueller MJ,Minor SD,Sahrmann SA,Schaaf JA,Strube MJ.Differences in the gait characteristics of patients with diabetes and peripheral neuropathy compared with age-matched controls. Phys Ther 1994;74:299-308.

41.Eils E,Behrens S,Mers O,Thorwesten L,V¨olker K,Rosenbaum D.Reduced plantar sensation causes a cautious walking pattern.Gait Posture2004;20:54-60.

42.Sinkjær T,Andersen JB,Ladouceur M,Christensen LO,Nielsen JB.Major role for sensory feedback in soleus EMG activity in the stance phaseof walking in man.JPhysiol 2000;523:817-27.

43.Pollock AS,Durward BR,Rowe PJ,Paul JP.What is balance?Clin Rehabil 2000;14:402-6.

44.Winter DA,Patla AE,Prince F,Ishac M,Gielo-Perczak K.Stiffnesscontrol of balancein quiet standing.JNeurophysiol 1998;80:1211-21.

45.Bauby CE,Kuo AD.Activecontrol of lateral balancein human walking.JBiomech 2000;33:1433-40.

46.Qu X,Nussbaum MA.Evaluation of the rolesof passiveand activecontrol of balanceusing a balancecontrol model.JBiomech 2009;42:1850-5.

47.Garcia M,Ruina A,Coleman M,Chatterjee A.Some results in passivedynamic walking.In:Pfeiffer F,editor.Proceedings of the Euromech 375:Biology and Technology of Walking.M¨unchen:Technical University of Munich;March 23-25,1998.p.268-75.

48.Morasso PG,Schieppati M.Can muscle stiffness alone stabilize upright standing.JNeurophysiol 1999;82:1622-6.

49.Kawato M.Internal models for motor control and trajectory planning.Curr Opin Neurobiol 1999;9:718-27.

50.Desmurget M,Grafton S.Forward modeling allows feedback control for fast reaching movements.Trends Cogn Sci2000;4:423-31.

51.McGeer T.Passivedynamic walking.Int JRobotics Res1990;9:62-82.

52.Duysens J,Van de Crommert HW.Neural control of locomotion;part 1:the central pattern generator from cats to humans.Gait Posture 1998;7:131-41.

53.Grillner S,Zangger P.On thecentral generation of locomotion in thelow spinal cat.Exp Brain Res1979;34:241-61.

54.Pearson KG,Rossignol S.Fictive motor patternsin chronic spinal cats.J Neurophysiol 1991;66:1874-87.

55.Reisman DS,Wityk R,Silver K,Bastian AJ.Locomotor adaptation on a split-belt treadmill can improve walking symmetry post-stroke.Brain 2007;130:1861-72.

56.Kuo AD.Therelativerolesof feedforward and feedback in thecontrol of rhythmic movements.Motor Control 2002;6:129-45.

57.Manchester D,Woollacott M,Zederbauer-Hylton N,Marin O.Visual,vestibular and somatosensory contributions to balance control in the older adult.JGerontol 1989;44:M118-27.

58.Diener HC,Dichgans J.On the role of vestibular,visual and somatosensory information for dynamic postural control in humans.Prog Brain Res1988;76:253-62.

59.Mergner T,Rosemeier T.Interaction of vestibular,somatosensory and visual signalsfor postural control and motion perception under terrestrial and microgravity conditions—a conceptual model.Brain Res Rev 1998;28:118-35.

60.Peterka RJ,Benolken MS.Role of somatosensory and vestibular cues in attenuating visually induced human postural sway.Exp Brain Res 1995;105:101-10.

61.Simoneau GG,Ulbrecht JS,Derr JA,Cavanagh PR.Role of somatosensory input in the control of human posture.Gait Posture 1995;3:115-22.

62.Timmann D,Belting C,Schwarz M,Diener HC.Inf luence of visual and somatosensory input on leg EMGresponsesin dynamic posturography in normals.Electroencephalogr Clin Neurophysiol 1994;93:7-14.

63.Lacour M,Borel L.Vestibular control of posture and gait.Arch Ital Biol 1993;131:81-104.

64.Hijmans JM,Geertzen JH,Dijkstra PU,Postema K.A systematic review of the effects of shoes and other ankle or foot appliances on balance in older people and people with peripheral nervous system disorders.Gait Posture2007;25:316-23.

65.McKeon PO,Hertel J.Diminished plantar cutaneous sensation and postural control.Percept Mot Skills2007;104:56-66.

66.Nardone A,Galante M,Pareyson D,Schieppati M.Balance control in sensory neuron disease.Clin Neurophysiol 2007;118:538-50.

67.Nardone A,Grasso M,Schieppati M.Balance control in peripheral neuropathy:are patients equally unstable under static and dynamic conditions?Gait Posture2006;23:364-73.

68.Nardone A,Tarantola J,Miscio G,Pisano F,Schenone A,Schieppati M.Loss of large-diameter spindle afferent f ibres is not detrimental to the control of body sway during upright stance:evidence from neuropathy.Exp Brain Res2000;135:155-62.

69.Perry SD,McIlroy WE,Maki BE.The role of plantar cutaneousmechanoreceptorsin the control of compensatory stepping reactions evoked by unpredictable,multi-directional perturbation.Brain Res 2000;877:401-6.

70.Erickson MA,Oliver T,Baldini T,Bach J.Biomechanical assessment of conventional unit rod f ixation versusaunit rod pediclescrew construct:a human cadaver study.Spine2004;29:1314-9.

71.Hausdorff JM,Cudkowicz ME,Firtion R,Wei JY,Goldberger AL.Gait variability and basal ganglia disorders:stride-to-stride variations of gait cycle timing in Parkinson's disease and Huntington's disease.Mov Disord 1998;13:428-37.

72.Anacker SL,Di Fabio RP.Inf luence of sensory inputs on standing balance in community-dwelling elders with a recent history of falling.Phys Ther 1992;72:575-81.

73.Horak FB,Nashner LM,Diener HC.Postural strategies associated with somatosensory and vestibular loss.Exp Brain Res1990;82:167-77.

74.Lord SR,Clark RD,Webster IW.Postural stability and associated physiological factors in a population of aged persons.J Gerontol 1991;46:M69-76.

75.Mauritz KH,Dietz V.Characteristics of postural instability induced by ischemic blocking of leg afferents.Exp Brain Res1980;38:117-9.

76.Teasdale N,Stelmach GE,Breunig A.Postural sway characteristics of the elderly under normal and altered visual and support surface conditions.JGerontol 1991;46:B238-44.

77.Fitzpatrick R,McCloskey DI.Proprioceptive,visual and vestibular thresholdsfor theperception of sway during standing in humans.JPhysiol 1994;478:173-86.

78.Morasso PG,Sanguineti V.Anklemusclestiffnessalonecannot stabilize balanceduring quiet standing.JNeurophysiol 2002;88:2157-62.

79.Winter DA,Patla AE,Rietdyk S,Ishac MG.Ankle muscle stiffness in the control of balance during quiet standing.J Neurophysiol 2001;85:2630-3.

80.Freeman MA.Instability of the foot after injuries to the lateral ligament of theankle.JBone Joint Surg Br 1965;47:669-77.

81.Garn SN,Newton RA.Kinesthetic awareness in subjects with multiple anklesprains.Phys Ther 1988;68:1667-71.

82.Lentell GL,Katzman LL,Walters MR.Therelationship between muscle function and anklestability.JOrthop Sports Phys Ther 1990;11:605-11.

83.Tropp H.Pronator muscleweaknessin functional instability of the ankle joint.Int JSports Med 1986;7:291-4.

84.Irrgang JJ,Whitney SL,Cox ED.Balanceand proprioceptivetraining for rehabilitation of thelower extremity.JSport Rehabil 1994;3:68-83.

85.Fu AS,Hui-Chan CW.Anklejoint proprioception and postural control in basketball players with bilateral ankle sprains.Am J Sports Med 2005;33:1174-82.

86.Jerosch J,Prymka M.Proprioception and joint stability.Knee Surg Sports Traumatol Arthrosc 1996;4:171-9.

87.Lee AJ,Lin WH.Twelve-week biomechanical ankle platform system training on postural stability and ankle proprioception in subjects with unilateral functional ankleinstability.Clin Biomech 2008;23:1065-72.

88.Hertel JN,Guskiewicz KM,Kahler DM,Perrin DH.Effect of lateral anklejoint anesthesiaon center of balance,postural sway,and joint position sense.JSport Rehabil 1996;5:111-9.

89.De Carlo MS,Talbot RW.Evaluation of ankle joint proprioception following injection of the anterior talof ibular ligament.J Orthop Sports Phys Ther 1986;8:70-6.

90.Bloem B,Allum JH,Carpenter M,Verschuuren JJ,Honegger F.Triggering of balance corrections and compensatory strategies in a patient with total leg proprioceptiveloss.Exp Brain Res2002;142:91-107.

91.Zhang S,Li L.The differential effects of foot sole sensory on plantar pressure distribution between balance and gait.Gait Posture 2013;37:532-5.

92.Simmons RW,Richardson C,Pozos R.Postural stability of diabetic patients with and without cutaneous sensory def icit in the foot.Diabetes Res Clin Pract 1997;36:153-60.

93.Clarac F,Cattaert D,Le Ray D.Central control components of a“simple”stretch ref lex.Trends Neurosci2000;23:199-208.

94.Hoffer JA,Andreassen S.Regulation of soleus muscle stiffness in premammillary cats:intrinsic and ref lex components.J Neurophysiol 1981;45:267-85.

95.Sinkjær T,Andersen JB,Larsen BI.Soleusstretch ref lex modulation during gait in humans.JNeurophysiol 1996;76:1112-20.

96.Di Giulio I,Maganaris CN,Baltzopoulos V,Loram ID.The proprioceptive and agonist roles of gastrocnemius,soleus and tibialis anterior muscles in maintaining human upright posture. J Physiol 2009;587:2399-416.

97.Matthews PB.The human stretch ref lex and the motor cortex.Trends Neurosci1991;14:87-91.

98.Taylor J,Stein RB,Murphy PR.Impulse rates and sensitivity to stretch of soleus muscle spindle afferent f ibers during locomotion in premammillary cats.JNeurophysiol 1985;53:341-60.

99.Schieppati M,Nardone A.Medium-latency stretch ref lexes of foot and leg muscles analysed by cooling the lower limb in standing humans.J Physiol 1997;503:691-8.

100.Grey MJ,Ladouceur M,Andersen JB,Nielsen JB,Sinkjær T.Group II muscle afferents probably contribute to the medium latency soleus stretch ref lex during walking in humans.J Physiol 2001;534:925-33.

101.Schieppati M,Nardone A.Group II spindle afferent f ibers in humans:their possible role in the ref lex control of stance.Prog Brain Res 1999;123:461-72.

102.Mazzaro N,Grey MJ,do Nascimento OF,Sinkjær T.Afferent-mediated modulation of the soleus muscle activity during the stance phase of human walking.Exp Brain Res2006;173:713-23.

103.Akazawa K,Aldridge JW,Steeves JD,Stein RB.Modulation of stretch ref lexes during locomotion in the mesencephalic cat.J Physiol 1982;329:553-67.

104.Burke D.Critical examination of the case for or against fusimotor involvement in disordersof muscletone.Adv Neurol 1983;39:133-50.

105.Capaday C,Stein RB.Amplitudemodulation of thesoleus H-ref lex in the human during walking and standing.JNeurosci1986;6:1308-13.

106.Zhang S,Manor B,Li L.H-index is important for postural control for people with impaired foot sole sensation.PloSOne 2015;10:e0121847.doi:10.1371/journal.pone.0121847.

107.Zehr PE.Considerations for useof the Hoffmann ref lex in exercisestudies.Eur JAppl Physiol 2002;86:455-68.

108.Schieppati M.The Hoffmann ref lex:a means of assessing spinal ref lex excitability and its descending control in man.Prog Neurobiol 1987;28:345-76.

109.Houk JC.An assessment of stretch ref lex function.Prog Brain Res 1976;44:303-14.

110.Pham H,Armstrong DG,Harvey C,Harkless LB,Giurini JM,Veves A.Screening techniques to identify people at high risk for diabetic foot ulceration:a prospective multicenter trial.Diabetes Care 2000;23:606-11.

111.Chen H,Nigg BM,Hulliger M,De Koning J.Inf luence of sensory input on plantar pressuredistribution.Clin Biomech 1995;10:271-4.

112.Eils E,Nolte S,Tewes M,Thorwesten L,V¨olker K,Rosenbaum D.Modif ied pressure distribution patterns in walking following reduction of plantar sensation.JBiomech 2002;35:1307-13.

113.Nurse MA,Nigg BM.The effect of changes in foot sensation on plantar pressureand muscleactivity.Clin Biomech 2001;16:719-27.

114.Wu G,Chiang JH.The signif icanceof somatosensory stimulationsto the human foot in the control of postural ref lexes.Exp Brain Res 1997;114:163-9.

115.Caselli A,Pham H,Giurini JM,Armstrong DG,Veves A.The forefoot-to-rearfoot plantar pressure ratio is increased in severe diabetic neuropathy and can predict foot ulceration.Diabetes Care 2002;25:1066-71.

116.Hong SL,Manor B,Li L.Stanceand sensory feedback inf luence on postural dynamics.Neurosci Lett 2007;423:104-8.

117.Zhang S,Holmes M,Li L.Reliability of nerve function assessments for peoplewith peripheral neuropathy.Int JNeurosci2015;125:201-7.

118.H¨ohne A,Stark C,Br¨uggemann GP.Plantar pressure distribution in gait isnot affected by targeted reduced plantar cutaneoussensation.Clin Biomech 2009;24:308-3.

119.England SA,Granata KP.The inf luence of gait speed on local dynamic stability of walking.Gait Posture2007;25:172-8.

120.H¨ohne A,Stark C,Br¨uggemann GP,Arampatzis A.Effects of reduced plantar cutaneous afferent feedback on locomotor adjustmentsin dynamic stability during perturbed walking.JBiomech 2011;44:2194-200.

121.Tsai YJ,Lin SI.Older adults adopted more cautious gait patterns when walking in socksthan barefoot.Gait Posture2013;37:88-92.

122.Curthoys IS,Halmagyi GM.Vestibular compensation:a review of the oculomotor,neural,and clinical consequences of unilateral vestibular loss.JVestib Res1995;5:67-107.

123.Merabet LB,Pascual-Leone A.Neural reorganization following sensory loss:theopportunity of change.Nat Rev Neurosci2010;11:44-52.

124.Nashner LM.Adaptation of human movement to altered environments.Trends Neurosci1982;5:358-61.

125.Oie KS,Kiemel T,Jeka JJ.Multisensory fusion:simultaneousre-weighting of vision and touch for thecontrol of human posture.Brain Res Cogn Brain Res2002;14:164-76.

126.Peterka RJ.Sensorimotor integration in human postural control.JNeurophysiol 2002;88:1097-118.

127.Bove M,Trompetto C,Abbruzzese G,Schieppati M.Theposture-related interaction between Ia-afferent and descending input on the spinal ref lex excitability in humans.Neurosci Lett 2006;397:301-6.

128.Lalonde R,Strazielle C.Brain regions and genes affecting postural control.Prog Neurobiol 2007;81:45-60.

129.Visser JE,Bloem BR.Role of the basal ganglia in balance control.Neural Plast 2005;12:161-74.

130.Scaglioni G,Ferri A,Minetti AE,Martin A,Van Hoecke J,Capodaglio P,et al.Plantar f lexor activation capacity and H ref lex in older adults:adaptationsto strength training.JAppl Physiol 2002;92:2292-302.

131.Knikou M.The H-ref lex as a probe:pathways and pitfalls.J Neurosci Methods2008;171:1-2.

132.Aiello I,Rosati G,Serra G,Manca M.The diagnostic value of Hindex in S1 root compression.J Neurol Neurosurg Psychiatry 1981;44:171-2.

133.Dixit S,Maiya A,Shastry BA,Guddattu V.Analysis of postural control during quiet standing in a population with diabetic peripheral neuropathy undergoing moderate intensity aerobic exercise training:a single blind,randomized controlled trial.Am J Phys Med Rehabil 2016;95:516-24.

134.Boulton AJ,Gries FA,Jervell JA.Guidelinesfor thediagnosisand outpatient management of diabetic peripheral neuropathy.Diabet Med 1998;15:508-14.

135.Gruber M,Taube W,Gollhofer A,Beck S,Amtage F,Schubert M.Training-specif ic adaptations of H-and stretch ref lexes in human soleus muscle.JMot Behav 2007;39:68-78.

136.Guan H,Koceja DM.Effects of long-term Tai Chi practice on balance and H-ref lex characteristics.Am JChin Med 2011;39:251-60.

137.Dobson JL,McMillan J,Li L.Benef its of exerciseintervention in reducing neuropathic pain.Front Cell Neurosci 2014;8:102.doi:10.3389/fncel.2014.00102.

138.Li L,Hondzinski JM.Select exercise modalities may reverse movement dysfunction because of peripheral neuropathy.Exerc Sport Sci Rev 2012;40:133-7.

139.Richardson JK,Sandman D,Vela S.A focused exercise regimen improvesclinical measuresof balancein patientswith peripheral neuropathy.Arch Phys Med Rehabil 2001;82:205-9.

140.Morrison S,Colberg SR,Mariano M,Parson HK,Vinik AI.Balance training reduces falls risk in older individuals with type 2 diabetes.Diabetes Care2010;33:748-50.

141.Li L,Manor B.Long term Tai Chiexerciseimprovesphysical performance among peoplewith peripheral neuropathy.Am JChin Med 2010;38:449-59.