Ites, Katherine I. PT, DPT; Anderson, Elizabeth J. PT, DPT; Cahill, Megan L. PT, DPT; Kearney, Jenny A. PT, DPT; Post, Emily C. PT, DPT; Gilchrist, Laura S. PT, PhD
Diabetic peripheral neuropathy (DPN) is a serious complication of both type 1 and type 2 diabetes.1 In 2008, it was estimated that 18.82 million Americans, or 6.29% of the population, had diagnosed diabetes.2 A total of 1.6 million new cases of diabetes are diagnosed each year in the United States, a number that is on the rise as the incidence of obesity continues to increase. Although diabetes can affect people at any stage of development, its prevalence increases with age and it is estimated that 23.8% of individuals 60 years of age and older have diabetes.3
One of the many side effects of diabetes mellitus is diabetic peripheral neuropathy. It is estimated that 60% to 70% of individuals with diabetes have mild to severe forms of nervous system damage.2 A simple definition of DPN is “the presence of symptoms and/or signs of peripheral nerve dysfunction in people with diabetes after exclusion of other causes.”4 Although there are several types of DPN associated with diabetes, this article will focus on chronic sensorimotor DPN, the most common type of neuropathy in diabetes mellitus.4 Symptoms of peripheral neuropathy include numbness or insensitivity to pain or temperature, paresthesias, sharp pains or cramps, and extreme sensitivity to touch. In addition, as many as 30% of people with DPN experience muscle weakness, loss of ankle reflexes, and decreased balance and coordination.2 This can significantly impair physical function by limiting walking and standing activities and may also increase the risk for falls in people with DPN.
Peripheral neuropathy is caused by microvascular abnormalities resulting in nerve damage.5 Chronic hyperglycemia impairs microvascular circulation by disrupting normal cellular communication and initiating signaling cascades. Through the production of advanced glycation end products and protein kinase C signaling cascade, chronic hyperglycemia causes damage to nerves.5 Specifically, it results in axonal thickening and decreased capillary blood flow resulting in poor nerve perfusion and endonural hypoxia.5 These cellular-level impairments are manifested as loss of ankle reflexes, decreased position and vibratory sense, and sensory ataxia.6 In addition, patients with DPN often demonstrate delayed reflex responses to postural perturbations as a result of decreased nerve conduction velocity and are subsequently more likely to demonstrate balance impairments and an increased risk for falls.6
Management of DPN is generally multifaceted. Avoidance of complications through excellent blood glucose control appears to be the most effective strategy for prevention.4 Pharmacological interventions for symptom management are available. Some medications have been found effective but side effects often limit their use.4 Physical therapy interventions utilized to reduce the balance dysfunction can range from restoration of the health of the neurons to sensory integration to compensatory strategies.7 Examples of such interventions include improving circulation, guided practice of integrating internal and external sensory input, education on sensation loss and fall risk, instruction on home modifications, and introduction of assistive devices to minimize balance dysfunction.7,8
Although physical therapists can play an important role in the management of balance dysfunction as a result of DPN, there has been no comprehensive review published on this topic. Therefore, we performed a systematic review of the literature to determine the effectiveness of physical therapy interventions on balance dysfunction in adults with DPN. This study was conducted in line with the PRISMA Statement for systematic reviews.9
A comprehensive computerized database search of 4 electronic databases was conducted from inception to June 2009: CINAHL, starting at 1981, EMBASE starting at 1972, Medline starting at 1966, and Cochrane Review starting at 1988. Search terms included a combination of the terms “peripheral neuropathy,” “balance,” and “diabetes mellitus.” Bibliographies of retrieved articles were also searched for additional studies.
The article titles from the search were each reviewed independently by 2 authors. Titles were accepted if they contained the topics of diabetes and neuromuscular balance, and/or falls risk. In this review, balance was defined as the ability to maintain a steady position in a weight-bearing standing posture.7 When the 2 individual reviewers did not agree, a decision for inclusion or exclusion was made by a third independent reviewer. Abstracts of the included titles were obtained and analyzed by using the same method. Abstracts often contained words such as “falls risk” and “fear of falling” in addition to balance.
Once an article was included on the basis of abstract content, it was considered appropriate for this systematic review if the following criteria were met.
All study designs were eligible for review, with the exception of case reports. Lack of available research on this subject prevented us from limiting articles to randomized controlled trials alone. Only full reports were included; abstracts, unpublished articles, and studies in a language other than English were excluded.
A study was included if the study population consisted predominantly of adults with DPN. Studies that included subgroups of patients with other types of neuropathy were excluded if the data for patients with DPN could not be analyzed independently of the other groups.
All studies were required to include interventions that are within the physical therapy scope of practice. Excluded interventions included pharmacological treatment, surgical interventions, and acupuncture.
The studies were required to report a primary outcome measure of balance.
Each of the characteristics mentioned earlier was extracted from the articles for analysis.
A statistical analysis was conducted to determine the magnitude of the treatment effect. Effect size (ES) and number needed to treat (NNT) statistics were used to determine the magnitude of the treatment effect for each intervention. Effect size is reported with 95% confidence intervals. Insufficient evidence for each intervention prevented pooling of data for meta-analysis.
Assessment of Individual Study Quality
Once the articles were selected for complete review, 5 authors independently read each article and analyzed the study methods using the Delphi criteria (Table 1). The Delphi criteria has been shown to be a valid and reliable tool for the assessment of clinical trials and has also been used in systematic reviews.10 A recent systematic review by Olivo et al,11 which assessed the scales used to evaluate the methodological quality of randomized controlled trials in health care research, concluded that the Delphi List had good validity and was relevant for use in the field of physical therapy.
The Delphi criteria is composed of 9 questions to assess treatment allocation, patient population, eligibility criteria, subject and assessor blinding, outcome measures, and patient dropouts. For an answer of “yes,” the question was given a score of 1. For an answer of “no” or “don't know,” the question received a score of zero. Using this scale, an overall score out of 9 was given to each article, with a score of 6 of 9 considered to be high quality evidence.10
Group consensus was reached between all 5 reviewers to determine the Delphi score for each article. The independent reviewers were not blinded in regard to article title, author, journal, or funding source. None of the authors have a conflict of interest or previous publications in this subject manner that would bias the results of this study.
Assessment of Intervention Categories
First a numeric grade was given to each intervention type on the basis of the quality of available research, according the guidelines seen in Table 2.12 The numeric level, using a 5-point scale, was based on the design of the studies for the intervention category only.
Second, a letter grade (Table 3) was assigned to the intervention category that evaluated each intervention on the basis of magnitude of treatment effect on the intended outcome of functional balance. An intervention received a higher grade if the outcome measure was based on balance or function/quality of life. In addition, outcome measures were required to have demonstrated validity and reliability to obtain a higher grade.
Each intervention was given both a numeric level of evidence, based on the design of the available studies, and a letter grade reflective of the magnitude of intervention effect on the intended outcome of functional balance. Therefore, it is possible to have a well-designed trial that indicated no effect of interventions.
Individual reviewers determined both the numeric level of evidence and the letter grade appropriate for each intervention. Both were discussed, and group consensus was used to finalize the level of evidence and the clinical recommendations grade for each intervention.
Selection of Studies
Figure 1 provides an overview of the literature search and study selection. From the total of 2213 article titles considered for review, 2131 were excluded on the basis of the title alone because they did not include neuromuscular balance. Of the 82 abstracts, 69 articles were excluded because they focused solely on improvements in sensation without direct measurement of balance, or they did not discuss interventions that are within the physical therapy scope of practice, such as medication management. Articles that focused on ionic balance were also excluded. Thirteen potential articles8,13–24 were assessed for inclusion in our review, with 7 articles being excluded on the basis of our study's inclusion criteria. None of these articles met our search criteria; all lacked specific interventions and results for subjects with diabetic peripheral neuropathy.
A total of 6 articles were reviewed for their methodological quality using the Delphi criteria (Table 4). Of the 6 articles, only 1 article received a Delphi score of 7/9 (Leonard et al13) indicating high methodological quality; this was the only randomized controlled trial found. The remaining 5 articles scored between a 1/9 to 5/9 on the Delphi criteria, indicating poor methodological quality.
This systematic review includes studies that assess 4 different physical therapy interventions with potential to improve balance in patients with DPN. These interventions include monochromatic infrared energy (MIRE) therapy, vibrating insoles, lower extremity strengthening exercises, and the use of a cane as a compensatory strategy. There was a wide variation in the type and quality of the outcome measures used to measure balance dysfunction across studies (Table 4).
Description of the Individual Trials Categorized by Intervention
Monochromatic Infrared Energy Therapy
The search of the literature resulted in 3 articles that investigated the effects of MIRE on balance and falls risk in patients with DPN. The methodological quality of the studies varied greatly, with scores on the Delphi criteria ranging from 1/9 to 7/9. Although all 3 studies utilized an Anodyne Therapy System, we were unable to compare dosages and parameters across studies, as this information was not consistently provided by the authors. Each study found similar results regarding the use of MIRE to improve balance deficits in patients with DPN, yet all had major methodological limitations, limiting generalization of the results.
The study by Leonard et al13 scored a 7/9 on the Delphi criteria. In this double-blinded, randomized, placebo-controlled study, the intervention group received 4 weeks of MIRE treatment, while the control group received 2 weeks of sham treatment followed by 2 weeks of active treatment. Strengths of this study included high methodological quality, clearly defined inclusion and exclusion criteria, and clearly explained treatment parameters. The primary balance measure was the question “Do you ever feel off balance or like you are going to fall?” Limitations of this study included a small sample size, the use of subjects as their own control, the control intervention being administered to only one leg with active treatment to the other leg at the same time, and the lack of a reliable and valid balance outcome measure. Although there were positive results with this study, the absence of a true control for balance dysfunction prevented computation of ES or NNT. It must be noted that the primary aim of this article was to measure improvements in sensation, and that balance dysfunction was a secondary outcome.
Kochman8 examined the effects of combining MIRE therapy with more traditional physical therapy interventions with proven effectiveness, such as neuromuscular reeducation and therapeutic exercise. This study was a nonrandomized, noncontrolled, and retrospective chart review of 38 patients and received a score of 1/9 on the Delphi criteria. Patients participated in an average of 12 physical therapy sessions consisting of MIRE treatment along with balance, strengthening, and stretching exercises. At the conclusion of the study, subjects demonstrated a significant decrease in risk for falls, as demonstrated by improvements on the Tinetti balance assessment scores (ES = 2.3 ± 0.459). A 93% decrease in the number of falls was reported at a 3-month follow-up interview (ES = 1.7 ± 0.459). Strengths of this study included the use of a valid and clinically relevant outcome measure and a 3-month follow-up period for fall occurrence. Limitations of this study included a small sample size, lack of randomization and control groups, absence of inclusion and exclusion criteria, and lack of specified treatment parameters. Because of the mixed intervention and lack of a control group, improvements seen in this study cannot be exclusively attributed to MIRE.
Powell et al14 reported a retrospective cohort study, which received a score of 1/9 on the Delphi criteria. Subjects consisted of 252 community-dwelling individuals with documented DPN, loss of protective sensation, and a history of improved sensation following MIRE therapy. Using telephone interviews, subjects were asked 5 questions regarding number of falls, balance, and fear of falling recalling back to the period prior to treatment and comparing it to the period following treatment for DPN. In the period following MIRE therapy, subjects reported a 55% reduction in number of falls and 79% improvement in fear of falling, both of which were statistically significant differences (NNT = 6.29). Strengths of this study include a large sample size, clear methodological outline, and clinically relevant outcomes. This study is limited by outcomes based solely on subject recollection to a time period prior to and after treatment for DPN, using perceived improvement as a measure, and poor study design. Also, this study lacked a control group to minimize the effect of confounding variables, such as exercise and diet recommendations.
Our search of the literature produced an article by Priplata et al15 that examined the effects of subsensory mechanical noise on balance in individuals with DPN (Table 4). This single-blinded prospective study with 15 subjects scored 5/9 on the Delphi criteria. Individuals with DPN participated in trials of quiet, undisturbed standing with eyes closed, both with and without vibrating insoles. Eight sway parameters were measured, all of which decreased in the vibrating insoles trials (ES = 0.513 ± 0.385 for AP sway and 0.256 ± 0.321 for ML sway). Strengths of this study included clearly defined inclusion and exclusion criteria, randomization of intervention presentation, and a low dropout rate. Limitations of this study included a small sample size and limited applicability of the outcome measure to functional balance. In addition, raw data was not available to calculate ES or NNT; calculations were estimated from graphs. Although this article provided evidence to support the use of mechanical noise to reduce sway in static standing in patients with DPN, the transferability of these results to balance activities such as walking and climbing stairs is limited because the study was conducted in a laboratory setting with patients in quiet-standing with eyes closed.
Lower Extremity Strengthening Exercises
The search of the literature produced 1 article by Richardson et al16 that examined the effects of a focused exercise regimen on balance in 20 subjects with peripheral neuropathy. This prospective, single blind cohort study received a score of 4/9 on the Delphi criteria. Participants in the intervention group participated in lower extremity exercises consisting of open and closed chain ankle exercises, wall slides, and single-leg stance for 3 weeks. Participants in the control group performed neck flexion and scapular stabilization exercises. Subjects in both groups performed 3 trials of each outcome measure (tandem stance, single-leg stance, and functional reach) and completed the Activities-specific Balance Confidence (ABC) scale before and after the intervention. At the end of the study, the intervention group showed significant improvements in all 3 functional outcome measures when compared with the control group, which showed no significant improvements (ES = 1.32 ± 0.63 for single-leg stance, 0.476 ± 0.553 for functional reach, and 0.448 ± 0.539 for tandem stance). In contrast, there were no significant differences noted in ABC scores for either group. These results showed that a brief, intense lower extremity exercise regimen could lead to improvements in 3 clinical parameters of balance in a group of older persons with peripheral neuropathy. Strengths of this study included the use of 3 clinical measures of balance and clearly defined exercise regimens for both groups. Weaknesses of this study included a small sample size and lack of randomization and matched-control subjects. Although the article scored poorly on the Delphi criteria indicating low methodological quality, the outcome measures were of clinical relevance.
A search of the literature resulted in 1 article that examined the effects of a cane on balance in individuals with DPN.17 This nonrandomized controlled trial by Ashton-Miller et al17 scored 2/9 on the Delphi Criteria. Subjects consisted of 8 patients with DPN and 8 age- and gender-matched controls. All subjects stood in a frame with a computerized base and hand railings. Subjects were cued to shift weight to 1 leg and balance for a minimum of 3 seconds, while experiencing a perturbation. Participants completed 28 trials in each of 4 conditions: eyes open, eyes closed, eyes open with cane, and eyes closed with cane. Failure rate, defined as the inability to maintain single-leg stance for 3 seconds without touching the handrails or placing the other foot on the ground, was measured during each of the conditions. Results of the study showed that patients with DPN had a significantly higher failure rate in all 4 conditions. In the DPN population, the use of a cane significantly reduced the failure rate in both the eyes open and eyes closed conditions (ES = 3.56 ± 0.707; NNT = 1.78). Strengths of this study included use of an appropriate control group and selection of a challenging balance situation. Limitations were a small sample size and lack of randomization or blinding. Although the article revealed good statistical evidence supporting the use of a single-end cane (SEC) for patients with balance dysfunction secondary to DPN, it lacked a clinically applicable outcome measure and high methodological quality.
Considering articles that met the criteria for this systematic review, the intervention of lower extremity strengthening exercise presents the best clinical evidence for treating balance dysfunction in patients with DPN (Table 5). Monochromatic infrared energy, vibrating insoles, and use of a cane do not have research-based outcomes to support their use at this time because they lack quality studies with strong methods and clinically important outcomes. Although the lower extremity strengthening exercise demonstrated the best clinical evidence when addressing balance impairments for patients with DPN, the study had methodological flaws and small sample sizes resulting in low scores on the Delphi criteria.
According to several studies,25–30 decreased lower extremity strength is a significant risk factor for falls in the geriatric population. Diminished ankle strength and rate of force production may lead to balance impairments, as normal recovery from perturbation involves rapid production of adequate muscle force to maintain the body's center of mass over its base of support. The article supporting lower extremity strengthening exercise scored 4/9 on the Delphi criteria suggesting low to moderate methodological quality.16 However, it used a clinically important outcome measure and showed statistically and clinically significant results and was thus given a grade of B for clinical recommendation. Despite the low to moderate level of evidence, the article demonstrated good clinical relevance and utilized clinically valid outcome measures. Lower extremity strengthening exercise was the only intervention found that can be supported with research evidence to treat balance dysfunction in patients with DPN.
The use of a single-end cane in patients with DPN and balance dysfunction does not address the sensation loss itself, but teaches compensation in hopes of minimizing fall risk. Previous research has shown that biomechanically, a cane functions to increase a person's base of support, allowing a greater range of center of mass motion without compromising stability.31 In addition, a cane allows the hand to be an additional point of somatosensory feedback.31
Utilizing 2 separate scales for grading methodology and clinical importance, the use of a single-ended cane received a grade of level II-2 evidence for methodological quality and a grade C for clinically relevant findings as seen in Table 5. Although the study on cane use reported both statistically significant and clinically relevant results, suggesting a beneficial effect, it scored a 2/9 on the Delphi criteria and does not have research evidence to support its clinical use in the treatment of balance dysfunction in patients with diabetic peripheral neuropathy. One considerable limitation is that this study was conducted in an artificial laboratory setting, leading us to question the generalizability of the findings to a patient's daily environment. The intervention of cane use received a grade C, a poor recommendation for clinical use, based on the quality of evidence, methodology, and strong effect size and NNT.
Vibrating insoles demonstrate the potential for improving balance dysfunction in patients with DPN. Previous research has shown that the presence of subsensory noise can enhance sensory and motor function and improve balance in patients with somatosensory deficits.32 Patients with DPN display significantly elevated thresholds for detecting sensory information such as tactile, vibratory, and joint angle and muscle force (proprioceptive) input, leading to an increased risk for falls.33 Thus, it is thought that low-level noise applied directly to sensory neurons enhances their ability to detect weak stimuli.
Although the Priplata et al15 study demonstrated an improvement in sway parameters with the use of subsensory vibrating insoles, this intervention received a grade of level II-2 evidence and grade C. Despite receiving a 5/9 on the Delphi criteria, the study lacked a functional outcome measure applicable in clinical situations. In addition, the effect size calculations for this study were small to moderate. The use of vibrating insoles is not likely a feasible intervention for physical therapists at this time as such insoles are not commercially available. Thus, vibrating insoles to treat balance dysfunction in patients with DPN cannot be recommended for clinical use at this time.
Monochromatic infrared energy therapy is a type of phototherapy that transmits heat to exposed surfaces via a low-level laser through pads placed on the skin.34 The exact mechanism of action remains unclear, but it has been proposed that MIRE triggers a release of nitric oxide from hemoglobin in exposed tissues. Nitric oxide leads to vasodilation, increased circulation, and decreased swelling. It may also stimulate angiogenesis, leading to accelerated tissue healing and ultimately decreasing sensory impairments and improving balance.34 Monochromatic infrared energy therapy is thus enticing, as the proposed mechanism of action purports to improve the nerve dysfunction itself. Monochromatic infrared energy received the highest level of evidence grade, level I, due to the inclusion of 1 randomized controlled trial. However, the methodological flaws in the balance dysfunction portion of the Leonard et al13 trial decreases the weight of the clinical evidence provided by this trial specifically for balance dysfunction. Although the Kochman8 study does not yield evidence on MIRE alone, it appears to have a meaningful clinical effect, because the MIRE therapy was paired with unspecified stretching, strengthening, and balance training. In fact, the ES of MIRE plus the exercise program appears to be close to that of the ES of the exercise alone trial in the single-leg stance outcome. The combination of the 3 studies, therefore, does not provide sufficient evidence to recommend MIRE intervention for balance dysfunction at this time and was given a grade C clinical recommendation. Of note, the Anodyne Therapy System (Anodyne Therapy, LLC, Tampa, FL), used to deliver MIRE, has not been approved by the FDA to be marketed for the treatment of DPN.35
Limitations of this systematic review include a change in databases while this study was being conducted. Following the first literature search, the databases EMBASE and CINAHL were combined into 1 database at our facility. Although this makes it difficult to replicate this study, it is believed that all articles from the original EMBASE and CINAHL searches can be found in the merged database
Weaknesses of this systematic review include a lack of available published research on this subject, which prevented us from conducting a meta-analysis. Also, initial search terms may not have found all potentially relevant articles. We attempted to address this limitation by also searching the bibliographies of retrieved articles. In addition, numbers for some of the NNT and/or ES calculations were estimated from published graphs secondary to the lack of data reported in some of the articles. A further limitation in this study is publication bias; only previously published research was included.
Lack of high-quality studies in this area of research is evidenced by only 1 randomized controlled study. Without randomization, studies are biased toward positive results. This systematic review revealed that there is a significant need for further research to investigate the effectiveness of physical therapy interventions to improve balance in patients with DPN. Currently, there is very little evidence to support such interventions. Of the evidence that is available, many studies are poorly designed and do not utilize reliable and valid outcome measures. Future research should investigate the effects of these physical therapy interventions on balance impairments in patients with DPN but should be designed and completed in a fashion that provides higher-level evidence. Research including randomized controlled trials with a greater number of subjects and meaningful outcome measures will enhance the quality of evidence to support interventions for improvement in balance. On the basis of the available research and this review, the intervention of lower extremity strengthening exercise can be given a fair recommendation for clinical use in addressing balance dysfunction in patients with DPN.
We would like to acknowledge John Schmitt, PT, PhD, for his assistance with reviewing this manuscript and providing statistical suggestions.
1. Goodman CC, Fuller KS. Pathology: Implications for Physical Therapist. 3rd ed. Philadelphia, PA: Saunders; 2009.
4. Boulton AJM, Vilnik AI, Arezzo JC, et al. Diabetic neuropathies: a statement by the American Diabetes Association. Diab Care. 2005;28:956–962.
5. Cade WT. Diabetes-related microvascular and macrovascular diseases in the physical therapy setting. Phys Ther. 2008;88:1322–1335.
6. Gutierrez EM, Helber MD, Dealva D, Ashton-Miller JA, Richardson JK. Mild diabetic neuropathy affects ankle motor function. Clin Biomech. 2001;16:522–528.
7. O'Sullivan SB, Schmitz TJ. Physical Rehabilitation. 4th ed. Philadelphia, PA: F.A. Davis Company; 2007.
8. Kochman AB. Monochromatic infrared photo energy and physical therapy for peripheral neuropathy: influence on sensation, balance, and falls. J Geriatr Phys Ther. 2004;27(1):18–21.
9. Moher D, Liberati A, Tetzlaff J, Altman DG; The Prisma Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009;151:264–269.
10. Verhagen AP, de Vet HCW, de Bie RA, et al. The Delphi list: a criteria list for quality assessment of randomized clinical trials for conducting systematic reviews developed by Delphi consensus. J Clin Epidmiol. 1998;51:1235–1241.
11. Olivo SA, Macedo LG, Gadotti I, et al. Scales to assess the quality of randomized Controlled trials: a systematic review. Phys Ther. 2008;88:156–175.
12. Philadelphia Panel Evidence-Based Clinical Practice Guidelines. Phys Ther. 2001;81:1629–1640.
13. Leonard DR, Farooqi MH, Myers S. Restoration of sensation, reduced pain, and improved balance in subjects with diabetic peripheral neuropathy: a double-blinded, randomized, placebo-controlled study with monochromatic near-infrared treatment. Diab Care. 2004;27(1):168–172.
14. Powell MW, Carnegie DH, Burke TJ. Reversal of diabetic peripheral neuropathy with phototherapy (MIRE) decreases falls and the fear of falling and improves activities of daily living in seniors. Age Ageing. 2006;35:11–16.
15. Priplata AA, Patritti BL, Niemi JB, et al. Noise-enhanced balance control in patients with diabetes and patients with stroke. Ann Neurol. 2005;59:4–12.
16. Richardson JK, Sandman D, Vela S. A focused exercise regimen improves clinical measures of balance in patients with peripheral neuropathy. Arch Phys Med Rehabil. 2001;82:205–209.
17. Ashton-Miller JA, Yeh MWL, Richardson JK, Galloway T. A cane reduces loss of balance in patients with peripheral neuropathy: results from a challenging unipedal balance test. Arch Phys Med Rehabil. 1996;77:446–452.
18. White CM, Pritchard J, Turner-Stokes L. Exercise for people with peripheral neuropathy. Cochrane Database Syst Rev. 2004;4:1–20. Updated May 2010.
19. Murphy DR. Diagnosis and manipulative treatment in diabetic polyneuropathy and its relation to intertarsal joint dysfunction. J Manipulative Physiol Ther. 1994;17(1):29–37.
20. Prasansuk S, Siriyananda C, Nakorn AN, et al. Balance disorders in the elderly and the benefit of balance exercise. J Med Assoc Thai. 2004;87:1225–1233.
21. Rao N, Aruin AS. Automatic postural responses in individuals with peripheral neuropathy and ankle-foot orthoses. Diab Res Clin Pract. 2006;74:48–56.
22. Graham RC, Hughes RAC, White CM. A prospective study of physiotherapist prescribed community based exercise in inflammatory peripheral neuropathy. J Neurol. 2007;254:228–235.
23. Balducci S, Iacobellis G, Parisi L, et al. Exercise training can modify the natural history of diabetic peripheral neuropathy. J Diab Compl. 2006;20:216–223.
24. Richerson S, Rosendale K. Does tai chi improve plantar sensory ability? A pilot study. Diab Technol Ther. 2007;9:276–286.
25. Rantanen T, Guralink JM, Sakari-Rantala R, et al. Disability, physical activity, and muscle strength in older women. The women's health and aging study. Arch Phys Med Rehabil. 1990;80:130–135.
26. Lord SR, Ward JA, Williams P, Anstey KJ. Physiological factors associated with falls in older community-dwelling women. J Am Geriatr Soc. 1994;42:1110–1117.
27. Lord SR, Clark RD, Webster IW. Physiological factors associated with falls in an elderly population. J Am Geriatr Soc. 1991;39:1194–1200.
28. Lord SR, McLean D, Stathers G. Physiological factors associated with injurious falls in older people living in the community. Arch Gerontol Geriatr. 1992;38:338–346.
29. Hess JA, Woollacott M, Shivitz N. Ankle force and rate of force production increase following high intensity strength training in frail older adults. Aging Clin Exp Res. 2006;18(2):107–115.
30. Thelen DG, Ashton-Miller JA, Schultz AB, Alexander NB. Do neural factors underlie age differences in rapid ankle torque development? J Am Geriatr Soc. 1996;44:804–808.
31. Bateni H, Maki B. Assistive devices for balance and mobility: benefits, demands, and adverse consequences. Arch Phys Med Rehabil. 2005;86:134–145.
32. Priplata AA, Niemi JB, Harry JD, et al. Vibrating insoles and balance control in elderly people. Lancet. 2003;362:1123–1124.
33. Moss F, Ward LM, Sannita WG. Stochastic resonance and sensory information processing: a tutorial and review of application. Clin Neurophysiol. 2004;115:267–281.
balance; diabetes; peripheral neuropathy; systematic review