Shortened refractory periods in human diabetic neuropathy.

Abstract

In routine nerve conduction studies, either a compound muscle or nerve action potential is recorded to measure the speed of the fastest fibers and the number of conducting axons. Activity-dependent excitability change is another aspect of neural conduction that may provide additional information on the ionic mechanisms underlying the pathophysiology of human neuropathies (Bostock et al., 1998; Burke et al., 2001). The refractory period has been shown to be a useful measure of subtle nerve dysfunction (Shefner and Dawson, 1990). Following an action potential, large myelinated axons are absolutely refractory for 0.5–1.0 ms, during which time they cannot generate another action potential no matter how strong the stimulus. The axons are then relatively refractory for another 3–4 ms, during which time a stronger than normal stimulus is required to generate an action potential (Burke et al., 2001). The absolute refractory period is very short, and has not proved to be a sensitive index when compound potentials are tested. Moreover, it is difficult in human subjects to measure the extent or duration of absolute refractoriness because there is a limitation on the stimulus that can be administered to a patient. Therefore, most of the previous studies have analyzed relative refractory periods in patients with peripheral neuropathy. In the current issue of the journal, a paper by Mackel and Brink (2003) investigated the absolute refractory periods of single sensory axons by using microneurography, and demonstrated that the refractory periods were significantly shorter, rather than longer as expected, for diabetic patients than for normal subjects. Responses of single sensory units in the median nerves were analyzed in 9 patients with Type I or Type II diabetes (range of disease duration, 3–18 years), most of whom had mildly symptomatic polyneuropathy. Conduction velocities did not differ between diabetic and normal subjects, but absolute refractory periods of diabetic axons ranged from 0.9 to 2.2 ms (mean, 1.3 ms), and were significantly shorter than the normal axons’ range of 0.9–3.5 ms (mean, 1.8 ms). The results were somewhat surprising, given that previous studies measuring the relatively refractory periods of compound sensory nerve potentials showed ‘prolonged’ refractory periods in diabetic patients (Schutt et al., 1983; Braume, 1999). Prolonged refractory periods have also been shown in patients with uremic, alcoholic, or toxic neuropathy (Schefner and Dawson, 1990). In patients with carpal tunnel syndrome, the refractory period of transmission across the lesion was shown to be prolonged (Gilliatt and Meer, 1990). Accordingly, almost of all the previous results showed ‘longer’ refractory periods in diseased nerves. However, the ‘shorter’ refractory periods in diabetic nerves may be consistent with reduced nodal Na currents, which have been shown in many experimental studies (Quasthoff, 1998). Diabetic neuropathy (distal symmetric polyneuropathy) results from a complex interplay between metabolic factors directly related to hyperglycemia and structural changes such as axonal degeneration and demyelination that result from microangiopathy (Sima, 1996). The major metabolic hypotheses proposed to explain diabetic neuropathy include activation of the polyol pathway and decreased Na-K ATPase activity. The conversion of excess glucose to sorbitol by the enzyme aldose reductase, and the resulting depletion of myo-inositol that leads to inactivation of Na-K ATPase, are postulated to play an important role in the pathophysiology of diabetic neuropathy (Sima, 1996; Quasthoff, 1998). Several studies using aldose reductase inhibitors in diabetic rodents have demonstrated beneficial preventive effects on nerve function, blood flow, and subsequent structural abnormalities (Sima et al., 1990), attesting to the importance of the polyol pathway in acute diabetic neuropathy. Intra-axonal sodium accumulation due to increased intra-axonal sorbitol concentration and decreased Na-K pump function would decrease the Na gradient across the axolemma, resulting in reduced Na currents when generating an action potential. The refractory period primarily depends on inactivation of Clinical Neurophysiology 114 (2003) 169–170