Preventing Foot Ulceration and Amputation by Decompressing Peripheral Nerves in Patients with Diabetic Neuropathy

Susceptibility to Compression

   Two metabolic changes in the peripheral nerves of the patient with diabetes render the nerve susceptible to chronic compression. The most critical metabolic change is the increased water content and subsequent increased volume within the nerve as a result of glucose being metabolized into sorbitol.¹ The second metabolic change is a decrease in the slow anterograde component of axoplasmic flow.²

This component of axoplasmic flow transports the lipoproteins necessary to maintain and rebuild the nerve. The peripheral nerve, as it crosses known areas of anatomic narrowing (eg, carpal tunnel at the wrist, cubital tunnel at the elbow, fibular tunnel at the outside of the knee, and the tarsal tunnel at the ankle), passes through a region of increased external pressure. Because the peripheral nerve in the patient with diabetes has an increased volume, pressure on the nerve in each of these anatomic regions is increased.

   This increased external pressure creates an increased intraneural pressure that decreases blood flow,³ resulting in a relative ischemic condition for the peripheral nerve. The neurophysiologic consequence of decreased blood flow in a peripheral nerve is the perception of paresthesias, interpreted centrally as numbness and tingling. The chronic pathophysiologic result of this area of increased pressure along a peripheral nerve is demyelination. The peripheral nerve with decreased axoplasmic flow, as in the person with diabetes, cannot transport sufficient protein structures to rebuild itself. Additionally, advanced glycosylation end (AGE) products reduce the normal gliding ability of the peripheral nerve. The nonenzymatic binding of glucose to the collagen within the nerve and in the epineurium are the basis for decreased elasticity. The combination of decreased peripheral nerve elasticity and its normal physiologic requirement to stretch as it glides across joints increases the stress and strain of the peripheral nerve within the areas of known anatomic narrowing. This increasing tension along the nerve further decreases blood flow within the nerve.⁴

   The hypothesis that the peripheral nerve in the person with diabetes has an increased susceptibility to compression has been tested in the rat model.⁵ A group of rats was rendered diabetic by administering streptozotocin. Silicone bands were placed around the sciatic nerves of the diabetic rats and a control group of nondiabetic, age-matched rats. After 6 months of banding (a time interval demonstrated to be sufficient to develop electrophysiologic and histologic changes consistent with chronic nerve compression in this model⁶), electrophysiology was tested in both groups. The diabetic rats had a statistically significant lower conduction velocity and a statistically significant lower amplitude for the sciatic nerve measured across the region of compression than did the nondiabetic banded rats, confirming that the peripheral nerve has an increased susceptibility to chronic nerve compression in the diabetic rat.

Chronic Nerve Compression and the Symptoms of Diabetic Neuropathy

   The patient with diabetic neuropathy has symptoms that include sensory complaints (ie, numbness and tingling, pain, and loss of sensation) and motor complaints (ie, weakness). The motor complaints extend to the autonomic system as well, and in the extremities include loss of pseudomotor function, resulting in dry, thicker skin. The sensory symptoms occur in a distribution that has been called "stocking and glove," with more severe symptoms in the lower extremity than the upper extremity. In contrast, the patient with a single chronic nerve compression will exhibit the same symptoms but only in the distribution of that particular nerve. For example, the patient with carpal tunnel syndrome with median nerve compression will only have sensory complaints in the palmar aspect of the thumb, index, and middle finger, and weakness in the muscles that control part of the thumb's function.⁷ In the upper extremity, chronic compression of the ulnar nerve at the elbow will result in paresthesias in the palmar and dorsal surfaces of the ring and little finger and weakness of pinch and grip strength. In advanced cases of ulnar nerve compression, intrinsic muscle weakness creates a "claw" deformity of the hand. Ulnar nerve compression at the elbow can be decompressed surgically.^8,9 If the radial sensory nerve were compressed in the forearm, numbness would be evident over the dorsoradial skin surface of the hand. The radial sensory nerve can be decompressed surgically as well.¹⁰ If a person were to have chronic compression of the median, ulnar, and radial nerves, a glove distribution of numbness would occur, with symptoms indistinguishable from the patient with symptomatic diabetic neuropathy of the upper extremity.

   This same relationship applies to the lower extremity. Compression of the sciatic nerve's common peroneal nerve at the lateral aspect of the knee occurs in the fibular tunnel. The symptoms include paresthesias or pain from the knee to the top of the foot. The motor component, when complete, results in a "drop foot," just as complete compression of the motor branch of the radial nerve, at the elbow, results in a "drop wrist." More commonly, in the leg, weakness of the long toe extensor occurs, so this toe is positioned lower than the other toes and is weak on manual muscle testing.¹¹ Compression of the common peroneal nerve in this location requires a neurolysis by division of the fascial coverings above and below the peroneus longus muscle. Over the dorsum of the foot, the deep peroneal nerve can be entrapped between the extensor digitorum brevis and the junction of the first metatarsal and the cunieform bone.¹² This entrapment is corrected by excision of the tendon of this small muscle, which has no functional significance in the foot. Entrapment of the deep peroneal nerve is similar to the radial sensory nerve entrapment in the forearm.

   In the foot, the analogy to carpal tunnel syndrome is tarsal tunnel syndrome.^13,14 However, the tarsal tunnel is really the analogy of the human forearm; therefore, to restore sensation to all toes and to the plantar aspect of the foot, the medial and lateral plantar nerves and the calcaneal nerve must each be released in their own separate tunnels, just distal to the tarsal tunnel. Severe compression of the lateral plantar nerve creates hyperextended toes at the metatarsal phalangeal joints. These appear as "hammer toes" in the person with diabetes, but they are actually "clawed toes" due to intrinsic muscle paralysis in the foot, just as the "clawed hand" results from intrinsic muscle paralysis in the hand.¹⁵ Relief of paresthesias and pain in the foot and, often, correction of the intrinsic muscle wasting, can be accomplished by decompression of the four medial ankle tunnels.¹⁶ If a person were to have chronic compression of the peroneal and tibial nerves, stocking distribution of numbness would occur, with symptoms indistinguishable from those of the patient with symptomatic diabetic neuropathy of the lower extremity.

   How can the clinician identify compression of the peripheral nerve? The most reliable clinical finding of nerve compression is tenderness of the nerve at the site of anatomic narrowing. This sensitivity of the nerve at the site of chronic compression may be manifested simply by the nerve being tender at that site, but most often is manifested by a distally radiating paresthesia in the distribution of the nerve when the nerve is gently percussed. This is referred to as a positive Tinel sign.¹⁷ In a patient with a neuropathy, where a systemic cause exists for the nerve dysfunction, no localizing sign along the course of the peripheral nerve should be present. However, if the neuropathy causes the nerve to be susceptible to nerve compression, superimposed compression of the peripheral nerve may have occurred in addition to the underlying neuropathy. Traditionally, electrodiagnostic testing is used to diagnose peripheral nerve compression, neuropathy, or nerve root compression. Often, however, when the peripheral neuropathy is so advanced that no conduction can be measured in the peripheral nerve or the conduction velocity and amplitude are already so reduced, identifying superimposed nerve compression in the patient with neuropathy is just not possible technically. In these situations, physical examination is critical in making this distinction.

Treatment of Neuropathy

   Treating the symptoms of neuropathy presupposes that the clinician knows the cause of the neuropathy. If the neuropathy is due to lead poisoning, chelation of the lead would be the starting point. If the neuropathy is due to a vitamin deficiency, dietary supplementation is the treatment. If the neuropathy is secondary to hypothyroidism, thyroid replacement is indicated. Clinicians can identify the etiology of the neuropathy in each of these cases using appropriate laboratory tests. Treatment of the metabolic problem corrects the cause of the neuropathy and is usually sufficient to relieve patient symptoms.

   Laboratory tests can identify the patient with diabetes and the patient who is not maintaining glycemic control. Unfortunately, even when maintaining strict glycemic control, patients may develop symptoms of neuropathy. In addition to maintaining glycemic control, management is usually limited to a combination of non-narcotic neuropathic pain medications and, when these are insufficient, narcotics. The classic triad of neuropathic medications includes carbamazepine, phenytoin, and amitriptyline. Many patients are unable to tolerate the side effects of carbamazepine, and phenytoin is often not effective. Because many patients with neuropathy have trouble sleeping, amitriptyline was often the most effective choice because the drug's most common side effect is drowsiness. Currently, gabapentin is the drug of choice for treating the symptoms of neuropathy. Unfortunately, many patients do not tolerate the effective doses of non-narcotic neuropathic pain medications or simply cannot accept the decrease in cognitive function these drugs induce.

   For many patients with neuropathy, the pain and paresthesias are not as troublesome as the loss of sensation in their feet. Reduced sensation causes loss of balance, inability to perceive hot water in the bathtub or to feel the pedals required for driving a car, and unsteadiness when walking down a flight of stairs. These symptoms of neuropathy also increase the risk for ulcer formation, infection, and amputation. No medication is currently available to treat the symptoms of sensory loss and this may create a sense of hopelessness, leading to depression and a sense of futility. Against this background, the concept that relieving symptoms of diabetic neuropathy by decompressing superimposed peripheral nerve compressions is possible is a source of optimism for this difficult clinical problem.¹⁸

   If the presence of a tight anatomic structure causes the symptoms of neuropathy in the person with diabetes, the absence of such a structure should result in the absence of symptoms of neuropathy. This hypothesis has been tested, and was confirmed, in the streptozotocin-induced diabetic rat model.¹⁹ That model was also the first to demonstrate that a neuropathic walking tract pattern occurs in the rat with diabetes. Specifically, a group of rats with serum glucose levels of 400 was followed for 1 year (about half the rat's lifetime). Half of the rats with streptozotocin-induced diabetes had the tarsal tunnel decompressed before the onset of diabetes; the other half were not treated by decompression of the tarsal tunnel and their anatomic site of compression was left intact. Both groups of rats were followed for 1 year with walking track analysis. The group without a tarsal tunnel had walking track patterns that were the same as weight-matched nondiabetic rats, while the diabetic rats with intact tarsal tunnels developed a progressive neuropathic walking track pattern (see Figures 1A and 1B). The results of this study suggested that in the rat model, in the absence of an anatomic site of narrowing, even the poorly controlled diabetic did not develop evidence of diabetic neuropathy in the feet. This study also suggested that even if the underlying metabolic neuropathy could not be corrected medically, treating symptoms related to superimposed chronic nerve compressions by decompressing the peripheral nerve might be possible. This research model has been applied recently to investigate cisplatin-related neuropathy.²⁰ The later study confirmed the observation that decompressing the tarsal tunnel in the rat with neuropathy, including cisplatin neuropathy, restores a normal walking track pattern.

Selecting Patients for Decompression

   Over the past 20 years, an approach to the selection of patients for surgical decompression of peripheral nerves has been developed. This approach begins with measurement of peripheral nerve function in order to stage the degree of nerve impairment.²¹ The model developed for staging the degree of nerve impairment in patients with chronic nerve compression but without a neuropathy also has been found to be valid in patients with neuropathy. Since 1989, measurement has included computer-assisted sensorimotor testing. Although vibrometry is useful for evaluating a single patient and comparing that patient to a group of patients, it does not help the surgeon interested in decompressing a particular nerve. This is because the vibration travels as a wave. The physician still needs to ascertain why the vibration is not perceived well. For instance, in the index finger, vibration may not be perceived because of a lesion of the median or the radial nerve. Similarly, if the vibration is not perceived well in the big toe, is it because of a lesion of the tibial or of the peroneal nerve? While the Semmes-Weinstein monofilament number 5.07 (10 g of force) may be useful in identifying an individual with diabetes who has lost protective sensation and is therefore at risk of ulceration in the foot,^22-29 this filament represents a cutaneous pressure threshold greater than 90 g/mm². At this advanced stage of chronic nerve compression, the patient has lost two-point discrimination, has severe axonal loss, and is most often past the point at which surgical intervention to restore sensation and relieve pain is still possible.³⁰

   In contrast, the Pressure-Specified Sensory Device™ (PSSD) (Sensory Management Services, LLC, Baltimore, Md.) can identify the earliest degree of chronic nerve compression by measuring the pressure required to distinguish one from two points touching the skin. Normative values for the Pressure-Specified Sensory Device™ have been reported for the upper extremity and for patients with carpal and cubital tunnel syndrome,³¹ for the lower extremity, and for patients with tarsal tunnel syndrome.³² The PSSD is at least as sensitive as traditional electrodiagnostic studies³³ and is not invasive and, therefore, not painful. No electric shocks are used.

   For the past 7 years, the American Diabetes Association has indicated that even people at low risk for ulceration should have a yearly somatosensory (quantitative sensory testing) measurement of the foot.^34-36 A guideline, based upon a cross-sectional study of people with diabetes with and without foot ulceration,³⁷ is available for application of measurement with neurosensory testing for the diabetic foot (see Figure 2). As the cutaneous pressure threshold for the big toe increases above the 99% confidence limit for normal (but axonal degeneration has not yet occurred), the patient with diabetes is referred to a diabetes educator for dietary advice, as well as to a podiatrist for evaluation of orthotic use and fitting/fabrication of special shoes. Once the 99% confidence limit is exceeded for the distance at which one from two points can be distinguished, indicating that axonal degeneration has occurred, a referral to a surgeon knowledgeable in peripheral nerve decompression is appropriate to determine whether the patient is a candidate for restoration of sensation and relief of pain.

   The most valid prognostic indication for a good result from decompression of a nerve in the person with diabetes with symptoms of neuropathy is the presence of a positive Tinel sign. This test is performed by tapping the region of known anatomic tightness (eg, the tarsal tunnel) with the examiner's finger (not with a percussion hammer). A "positive" test occurs when the patient can feel a radiating sensation, painful or not, into the territory supplied by that nerve (eg, the arch of the foot, the heel, or the big toe) when the percussion is done over the tarsal tunnel. The simple perception by the patient that thumping occurred is not a positive sign. Tapping over several "control" sites (ie, areas of skin without a known anatomic region of compression beneath them) should be performed. The common peroneal nerve at the fibular head may often just be tender and distally radiating perception does not occur. Tenderness of this nerve is sufficient to suggest entrapment at this location. Experience with patients with diabetic neuropathy has demonstrated that when a superimposed nerve compression is identified by a positive Tinel sign, the chance of a good-to-excellent result (relief of pain and restoration of sensation to the feet) is 80%.

Results of Decompressing Peripheral Nerves

   Since 1992, several studies have been conducted to evaluate the effect of decompressing peripheral nerves in people with diabetes. For the purpose of this publication, the studies have been reviewed and their patient populations regrouped to permit comparison of nerve-specific results - specifically, carpal tunnel decompression (Table 1), cubital tunnel decompression (Table 2), and tarsal tunnel decompression (Table 3).

   Decompressing the median nerve in the carpal tunnel in the person with diabetes gives excellent relief of sensory symptoms in about 95% of patients and good results in the remaining 5%, with 95% of the patients recovering useful two-point discrimination.^38,39 These results are similar to those observed in people without diabetes who have carpal tunnel decompression.⁷

   The results of anterior submuscular transposition of the ulnar nerve at the elbow in the person with diabetes, using the musculofascial lengthening technique, gives excellent relief of sensory symptoms in about 77% of patients and good results in the another 22%, with about 95% of the patients recovering useful two-point discrimination.^38,39 These results are what one would expect in the person without diabetes who also has this type of ulnar nerve surgery for moderate-to- severe degree of ulnar nerve compression.⁷ Recovery of motor function is not as good - 55% of the patients recover normal grip strength and 40% recover normal pinch strength.

   The results of decompression of the four medial ankle tunnels, related to the tibial nerve and its medial and lateral plantar and calcaneal branches, are determined by restoration of sensation to the sole of the foot and relief of pain in the foot. For all four reported groups of patients, each of whom was decompressed using the same surgical technique, pain was relieved in 86% of patients and 72% recovered useful two-point discrimination.^38-41 Two studies included patients who had a history of ulceration, and the percentage of patients having relief of pain was the same in these patients; however, many of these patients recovered just protective sensation (no two-point discrimination).^40,41 Among the 62 patients in this combined series who had never had an ulcer or amputation, none reported an ulceration or an amputation during the follow-up period of observation. Among the 24 patients in this combined series who had a previous ulcer or amputation, one (4%) reported a recurrent ulceration during the follow-up period of observation.

   The ability to restore sensation to the feet of the person with diabetes holds the promise of prevention of ulceration and amputation. During the period of time the author has been doing this type of nerve decompression, a group of patients has been followed who had surgery performed on one leg and foot. These patients have been followed for a mean of 4.5 years. Figure 3 shows a patient who had the right leg decompressed 7 years before the photograph was taken. Sensation had been recovered in this foot but, because of the distance between his home and the physician's office, he never came back for nerve decompression surgery on the opposite foot. He developed an ulcer in the contralateral foot and eventually required amputation of two toes on that foot. The present series includes 43 patients. None of these patients has had an ulcer or an amputation on the side that was decompressed. In contrast, seven ulcerations and two amputations occurred in contralateral limbs. This difference was statistically significant (P = 0.002).⁴²

Discussion

   The realization that the peripheral nerve in the patient with diabetes is susceptible to compression can give patients who suffer with unrelieved symptoms of neuropathy a source of optimism.¹⁸ Experience gained during the past 20 years, progressing from clinical observations³⁸ to basic science research¹⁹ and back again to clinical treatment of patients with diabetes and symptomatic lower extremity neuropathy⁴² has resulted in practical patterns of care for patient with diabetes. Independent surgical centers report that decompression of the tibial nerve and its branches at the ankle and foot level can relieve pain and restore sensibility in about 80% of the patients.^40,41

   As with the treatment of most diseases, the earlier a patient can be referred for treatment, the better the chance that symptoms can be reduced. With regard to diabetic neuropathy, once the patient has developed an ulcer, sufficient sensory axons have usually degenerated to the point that restoring protective sensation is only possible by decompressing the peripheral nerve. By contrast, if abnormal sensibility can be determined earlier, the ability to restore sensation can be offered at an earlier stage in the pathophysiology. The reason the results of decompression of the median nerve in the carpal tunnel have a higher success rate than decompression of the tibial nerve in the tarsal tunnel is that patients usually present earlier with hand than with foot problems.^38,39 This phenomenon may be related to the general pessimism that accompanies teaching the patient with diabetes that neuropathy is "progressive and irreversible."

Conclusion

   The observation that patients who have had restoration of sensation to their feet through decompression of peripheral nerves have not developed ulcers or had an amputation suggests that the natural sequelae of diabetic neuropathy may be amenable to change. To effect this change, clinicians responsible for the care of the patient with diabetes will need to measure sensibility in the foot, evaluate the foot for the presence of a Tinel sign over known sites of peripheral nerve compression, and refer the patient to a surgeon trained in lower extremity peripheral nerve decompression techniques. If this concept can be introduced into clinical practice, a significant decrease in foot ulcerations and amputations should be observed.

1. Jakobsen J. Peripheral nerves in early experimental diabetes. Expansion of the endoneurial space as a cause of increased water content. Diabetologia. 1978;14:113.

2. Jakobsen J, Sidenius P. Decreased axonal transport of structural proteins in streptozotocin diabetic rats. J Clin Invest. 1980;66:292.

3. Lundborg G, Rydevik B. Effects of stretching the tibial nerve of the rabbit: a preliminary study of the intraneural circulation and the barrier function of the perineurium. J Bone Joint Surg. 1973;55B:390-401.

4. Rydevik B, Lundborg G. Effects of graded compression on intraneural blood flow. J Hand Surg. 1981;6A:3-12.

5. Dellon AL, Mackinnon, SE, Seiler WA. Susceptibility of the diabetic nerve to chronic compression. Ann Plast Surg. 1988;20:117-119.

6. Mackinnon SE, Dellon AL, Hudson AR, Hunter D. Chronic nerve compression - an experimental model in the rat. Ann Plast Surg. 1984;13:112-120.

7. Mackinnon SE, Dellon AL. Surgery of the Peripheral Nerve. New York, NY: Thieme;1988.

8. Dellon AL. Operative technique for submuscular transposition of the ulnar nerve. Contemporary Orthopedics. 1988;16:17-24.

9. Dellon AL. Techniques for successful management of ulnar nerve entrapment at the elbow. Neurosurg Clin N Am. 1991;2:57-73.

10. Dellon AL, Mackinnon SE. Radial sensory nerve entrapment in the forearm. J Hand Surg. 1986;11A:199-205.

11. Mont MA, Dellon AL, Chen F, Hungerford MW, Krackow KA, Hungerford DH. Operative treatment of peroneal nerve palsy. J Bone Joint Surg. 1996;78A:863-869.

12. Dellon AL. Entrapment of the deep peroneal nerve on the dorsum of the foot. Foot and Ankle. 1990;11:73-80.

13. Keck C. The tarsal tunnel syndrome. J Bone Joint Surg. 1962;44A:180.

14. Lam SJS. A tarsal tunnel syndrome. Lancet. 1962;2:1354.

15. Mackinnon SE, Dellon AL. Homologies between the tarsal and carpal tunnels: implications for treatment of the tarsal tunnel syndrome. Contemporary Orthopedics. 1987;14:75-79.

16. Dellon AL. Computer-assisted sensibility evaluation and surgical treatment of tarsal tunnel syndrome. Advances in Podiatry.1996;2:17-40.

17. Dellon AL. Tinel or not Tinel. J Hand Surg. 1984;9B:216.

18. Dellon AL. Optimism in diabetic neuropathy. Ann Plast Surg. 1988;20:103-105.

19. Dellon AL, Dellon ES, Seiler WA. Effect of tarsal tunnel decompression in the streptozotocin-induced diabetic rat. Microsurgery. 1994;15:265-268.

20. Tassler PL, Dellon AL, Lesser G, Grossman S. Utility of decompressive surgery in the prophylaxis and treatment of cisplatin neuropathy in adult rats. J Reconstr Surg. 2000;16:457-463.

21. Dellon AL. Clinical grading of peripheral nerve problems. Neurosurg Clin N Am. 2001;12:229-240.

22. Holewski JJ, Stess RM, Graf PM, et al. Aesthiometry: quantification of cutaneous pressure sensation in diabetic peripheral neuropathy. J Rehabil Res Dev. 1988;25:1.

23. Sosenko JM, Kato M, Soto R, et al. Comparison of quantitative sensory threshold measures for their association with foot ulceration in diabetic patients. Diabetes Care. 1990;13:1057.

24. Kumar S, Fernando DJS, Veves A, et al. Semmes-Weinstein monofilaments: a simple, effective and inexpensive screening device for identifying diabetic patients at risk of foot ulceration. Diabetes Res Clin Pract. 1995;13:63.

25. Mueller M.J. Identifying patients with diabetes mellitus who are at risk for lower extremity complications: use of Semmes-Weinstein monofilaments. Phys Ther. 1996;76:68.

26. Klenerman L, McCabe C, Cogley D, et al. Screening for patients at risk of diabetic foot ulceration in a general diabetic outpatient clinic. Diabetes Medicine. 1996;13:561.

27. Ronnemaa T, Hamalainen H, Toikka T, et al. Evaluation of the impact of podiatrist care in the primary prevention of foot problems in diabetic subjects. Diabetes Care. 1997;20:1833.

28. deSonnaville JJ, Colly LP, Wijkel D, et al. The prevalence and determinants of foot ulceration in type II diabetic patients in a primary health care setting. Diabetes Res Clin Pract. 1997;35:149.

29. Armstrong DG, Lavery L, Vela SA, et al. Choosing a practical screening instrument to identify patients at risk for diabetic foot ulceration. Arch Intern Med. 1998;158:289.

30. Dellon AL. Somatosensory Testing and Rehabilitation. Baltimore, Md.: Institute for Peripheral Nerve Surgey; 2000.

31. Dellon AL, Keller KM. Computer-assisted quantitative sensorimotor testing patients with carpal and cubital tunnel syndrome. Ann Plast Surg. 1997;38:493-502.

32. Tassler PL, Dellon AL. Pressure perception in the normal lower extremity and in tarsal tunnel syndrome. Muscle Nerve. 1996;19:285-289.

33. Tassler PL, Dellon AL. Correlation of measurements of pressure perception using the Pressure-specified Sensory Device with electrodiagnostic testing. Journal of Occupational Medicine.1995;37:862-866.

34. Asbury AK, Porte D Jr. American Diabetes Association Clinical Practice Recommendations 1995; Consensus Statement on Diabetic Neuropathy. Diabetes Care. 1995; Suppl 1(18):53-58.

35. Asbury AK, Porte D Jr. American Diabetes Association Clinical Practice Recommendations 1995; Consensus Statement on Standardized Measures in Diabetic Neuropathy. Diabetes Care. 1995;Suppl 1(18):59-82.

36. Clinical Practice Recommendations 1997: S31; Foot Care in Patients with Diabetes Mellitus. Diabetes Care. 1997;Suppl 1(19):580-581.

37. Tassler PL, Dellon AL, Sheffler N. Cutaneous pressure thresholds in ulcerated and non-ulcerated diabetic feet, measured with the Pressure-specified Sensory Device. J Am Podiatr Med Assoc. 1995;85:679-684.

38. Dellon AL. Treatment of symptoms of diabetic neuropathy by peripheral nerve decompression. Plast Reconstr Surg. 1992;89:689-697.

39. Aszmann OA, Kress KM, Dellon AL. Results of decompression of peripheral nerves in people with diabetes: a prospective, blinded study. Plast Reconstr Surg. 2000;106:816-822.

40. Wieman TJ. Patel VG. Treatment of hyperesthetic neuropathic pain in people with diabetes; decompression of the tarsal tunnel. Ann Surg. 1995;221:660-665.

41. Chafee H. Decompression of peripehral nerves for diabetic neuropathy. Plast Reconstr Surg. 2000;106:813-815.

42. Aszmann OA, Tassler P, Dellon AL. Changing the natural history of diabetic neuropathy. Data on file, 2002.