EDITORIAL

Hyperesthesia: An Early Manifestation of Diabetic Polyneuropathy

Key words: diabetic neuropathy, , hyperpathia, necessary. Since development of diabetic neuropathy is an current threshold extremely slow process, longitudinal follow-up study is nec- essary for manyyears and years. Furthermore, other causes of neurological symptoms and signs should be properly diag- nosed and strictly ruled out during the study. It maynot be Diabetic polyneuropathy is a specific form of peripheral an easy task. In this issue, Takemuraet al (2) report an inter- nerve disorder, in which distal nerve axons degenerate insidi- esting cross-sectional community-based study on sensory ously under hyperglycemia of mellitus. Although all changes in a glucose-intolerant cohort, which was carried out types of peripheral nerve fibers are involved, it is usually as a part of large longitudinal study on aging. sensory dominant with eventual involvement of autonomic and motor nerve fibers. Its manifestation ranges from See also p 1124. subclinical changes in nerve conduction to painful and se- vere motor and autonomic disabilities. Once established, it is A tool the authors employed for the detection of sensory irreversible. Early identification is, therefore, of clinical im- change was a current perception threshold (CPT) measure- portance. ment. During the test, subjects report threshold perception to Amongvariety of sensory symptoms, numbnessand dimi- sinusoidal electric test current delivered to the skin with 5 nution of sensation in the distal limbs appears first. Decrease Hz, 200 Hz, and 2,000 Hz, which are said to specifically in density of sensory nerve fibers due to destruction of the stimulate C-, A8-, and Ap-fibers, respectively. Although peripheral nerve inevitably leads to loss of sensation of the CPTmeasurement itself still has some physiological uncer- distal limbs. Hypoesthesia and anesthesia of the skin are tainty in receptor selectivity, fiber selectivity, and pathologi- therefore understood as 'the negative sensory phenomenon' cal correlation, their results are quite informative. They in a Jacksonian , since it is a defect of normal function. confirmed the presence of early to electric cur- On the other hand, tingling sensation, "pins and needle sen- rent, and have found a prevalence of hyperesthesia of about sation", and painful are something different from 20%in the diabetic population. It seems a bit lower than a loss of function, since they seemto be due to over-activity of previous hospital-based study using a similar technique (3), the surviving structure. Such 'positive sensory phenomenon' however, this community-based study clearly demonstrated or manifestation of an excess of sensory activity is physio- the fact that hyperesthesia is commonnot only in the diabetic logically due to generation of ectopic impulses at various population but also in the non-diabetic insulin-resistant levels in sensory nerve fibers (1). Although the precise group. Takemura et al (2) actually found that hyperglycemic mechanism is still obscure, both chemical and morphological hyperesthesia is more frequent in 2,000 Hz and 250 Hz abnormalities may be involved. Hypersensitivity and cross- rather than in 5 Hz. Aquestion is that then raised: if the CPT talk of the surviving and regenerating axonal membrane,sen- frequency theory is right, clinical hyperesthesia should be sitization of receptors, and central hyperexcitability are detected as large-fiber hyperesthesia of vibratory or pressure important candidates for the positive phenomenon. touch sense as well as smaller fiber hyperesthesia of - In patients with established diabetic polyneuropathy, posi- temperature sensation. tive and negative symptoms often are intermingled. In fact, Hyperesthesia and associated with diabetes manyphysicians have been confused with such complex have long attracted the attention of manycareful investiga- combination of paresthesia, dysesthesia, hypoesthesia, tors. It is well known that patients with newly diagnosed dia- hyperpathia, allodynia, and even pain, to various sensory betes frequently show numbness and reduced nerve testing stimuli. Howshould we understand such mixed clini- conduction velocity that improves rapidly with the establish- cal picture? It is still obscure how the positive and negative ment of euglycemia (4). Alteration in fast potassium current symptoms develop and regress, and influence each other dur- has been shown in experimental hyperglycemic (5), ing the natural course of diabetes mellitus, since there are which maycause instability of the resting membranepoten- many limitations in a hospital-based approach, even if it is of tial. Morley et al (6) revealed hyperalgesia in diabetic pa- large scale. Careful longitudinal community-based study tients by applying simple square wave electric stimuli with using a large cohort with mild glucose impairment will be duration of 0.6 ms. They even found that a 50 g glucose Internal Medicine Vol. 41, No. 12 (December 2002) 1079 infusion resulted in a significant decrease in the threshold level of pain even in normal subjects. Changes in pain References threshold maybe thus functionally altered by plasma glucose level (7). Hyperexcitability of C-fiber is also a convincing 1) Sivak M, Ochoa J, Fernandez JM. Positive manifestation of nerve fiber finding in streptozotocin-induced hyperglycemic rats (8). dysfunction: Clinical electromyographic and pathologic correlates. In: On the other hand, Dyck et al (9) reported that Clinical Electromyography 2nd ed, Brown WF, Bolton CF, Eds, Butterworth-Heinemann, Boston, 1993: 1 17-147. hyperesthesia in diabetic polyneuropathy was found only for 2) Takekuma K, Ando F, Niino N, Shimokata H. Prevalence of heat-pain threshold and not for vibratory sense. Since they hyperesthesia detected by current perception threshold test in subjects found more neuropathic impairment and symptoms in with glucose metabolic impairments in a community. Intern Med41: hyperesthetic patients by using a sophisticated quantitative 1 124-1129,2002. sensory testing system (QST), they concluded that 3) Umezawa S, Kanamori A, Yajima Y, Aoki C. Current perception threshold in evaluating diabetic neuropathy. Tounyoubyou (Journal of hyperesthesia is a manifestation of hyperalgesia in diabetic the Japan Diabetes Society) 40: 711-718, 1997 ( in Japanese, Abstract patients. According to them, diabetic hyperesthesia is not an in English). excess of touch sensory activity, but a result of changes in 4) Gregersen G. Variations in motor conduction velocity produced by pain sensation. It is of great interest that they found acute changes in the metabolic state in diabetic patients. Diabetologia 4: 273-277, 1968. hyperesthesia/hyperalgesia only in the mildest spectrum of 5) Schneider U, Quasthoff S, Mitrovic N, Grafe P. Hyperglycemic diabetic neuropathy. In addition, in the mildest group hypoxia alters after-potential and fast-K+ conductance of rat axons by hypoesthesia to vibratory sense was simultaneously demon- cytoplasmic acidification. J Physiol (Lond) 465: 679-697, 1993. strated. In more severe cases all modalities of sensation were 6) Morley GK, Mooradian AD, Levine AS, Morley JE. Mechanism of pain in diabetic . Effect of glucose on pain per- hypoesthetic. ception in humans. Am J Med 77: 79-82, 1984. In conclusion, early hyperesthesia in diabetes is still a 7) Dobretsov M, Hastings SL, Stimers JR, Zhang JM. Mechanical confusing issue. What kind of fiber is responsible for it? Is hyperalgesia in rats with chronic perfusion of lumber dorsal root gan- it a kind of hyperalgesia? Isn't it merely functional? glion with hyperglycemic solution. J Neurosci Methods 110: 9-15, Nevertheless, some investigators have began to quantita- 2001. 8) Chen X, Levine JD. Hyper-responsibility in a subset of C-fiber tively study subclinical skin hyperesthesia and hyperalgesia in a model of painful diabetic neuropathy in the rat. in peripheral neuropathy by using quantitative thermotest Neuroscience 102: 185-192, 2001. (10) and QST (9). CPTmeasurement (ll) appears to be an- 9) Dyck PJ, Larson TS, Dyck PJB, O'Brien PC, Velosa JA. Patterns of other unique tool to detect 'subclinical positive phenome- quantitative sensation testing of hypoesthesia and hyperalgesia are pre- non' , although more pathophysiological validation of the test dictive of diabetic polyneuropathy: A study of three cohorts. Diabetes will be needed. Most conventional methods of sensory test- Care 23: 510-517, 2000. 10) Verdugo R, Ochoa JL. Quantitative somatosensory thermotest. A key ing such as sensory nerve conduction studies merely reveal method for functional evaluation of small calibre afferent channels. normality or negative phenomenon. Brain 115: 893-913, 1992. ll) Katims JJ, Naviasky EH, Rendell MS, Ng LKY, Bleecker ML. Masayuki Baba, MD, PhD Constant current sine wave transcutaneous nerve stimulation for the evaluation of peripheral neuropathy. Arch Phys Med Rehabil 68: 210- Department of Neurological Science, 213, 1987. Hirosaki University School of Medicine, 5 Zaifu-cho, Hirosaki, Aomori 036-8562

1080 Internal Medicine Vol. 41, No. 12 (December 2002)