Effect of Aldose Reductase Inhibition on Nerve Conduction Velocity and Resistance to Ischemic Conduction Block in Experimental Diabetes

Abstract

The relationships between nerve polyol levels and both nerve conduction velocity (NCV) and resistance to ischemie conduction block (RICB) in streptozocin-induced diabetic rats were examined in two studies. In the first study, sciatic NCV and RICB of the tail nerve, assessed by measuring the time to disappearance of the nerve action potential after the tail was rendered ischemie, were measured in nondiabetic rats, untreated diabetic rats, and diabetic rats given Statil, an aldose reductase inhibitor (ARI). Sciatic NCV was lower in the untreated diabetic animals than in control animals (P < .05), and RICB of the tail nerve was greater (P < .001). Treatment with the ARI completely prevented the slowing of NCV but had no significant effect on the increase in RICB. In the second study, similar groups of rats were treated with either ARI, insulin, or myo-inositol. Sciatic NCV was lower in the untreated diabetic rats than in the nondiabetic rats (P < .001). In diabetic rats treated with the ARI and in those treated with insulin, NCV was greater than in the untreated diabetic rats (P < .05 and P < .001, respectively) and was not significantly different from the nondiabetic rats. NCV in the myo-inositol-treated rats was not significantly different from that in the untreated diabetic rats. RICB was assessed by measuring the decline in sciatic nerve action potential amplitude at minute intervals after death. The time to a 50% fall in amplitude (t50) in the untreated diabetic rats was increased by 93%, and the time to complete disappearance (t0) was increased by 100% compared with the nondiabetic control rats. The increase in t50and t0 was prevented by insulin but was unaffected by myo-inositol. After the onset of ischemia, sciatic nerve function in the diabetic rats given the ARI initially fell rapidly: t50 was 29% lower than that found in the untreated diabetic rats (P < .02). The rate of decline then slowed, and t0 was not significantly different from that found in the untreated diabetic rats. t50 strongly correlated with nerve sorbitol levels among the insulin-treated diabetic rats (r = .94) and the untreated diabetic rats (r = .68) but not with nerve or plasma glucose levels. Thus, in experimental diabetes, RICB may be related to polyol-pathway overactivity. To explain ARI's reduction of t50 but not t0 in diabetic rats, we hypothesize that ARI may have had an effect on the RICB of ischemia-sensitive large-amplitude myelinated nerve fibers but little effect on ischemia-resistant low-amplitude unmyelinated fibers.