We evaluated the prevalence of peripheral neuropathy by clinical and electrophysiological criteria and the prevalence of autonomic parasympathetic nerve dysfunction by heart-rate variation during deep breathing (expiration-to-inspiration ratio [E:I]) in 132 newly diagnosed non-insulin-dependent diabetic (NIDDM) subjects aged 45–64 yr and 142 randomly selected nondiabetic control subjects. The relationship of nerve dysfunction to the degree of hyperglycemia and insulin-secretion capacity were also investigated. Single and scattered symptoms and signs of peripheral neuropathy were found in both diabetic and control subjects. Symptomatic polyneuropathy was found in 1.5% of diabetic subjects but none of the control subjects. Polyneuropathy defined by clinical signs was found in 2.3% of the diabetic subjects and 1.4% of the control subjects. No subjects with both symptoms and signs were seen. Nerve conduction velocities (NCVs) were significantly slower in diabetic than control subjects. Polyneuropathy according to electrophysiological criteria was found in 15.2% of diabetic subjects but was not found in any control subjects. Electromyographic abnormalities were more common in diabetic than control women, but no significant differences were found in men. The resting heart rate was higher in diabetic than control women, but no significant difference was found in men. The mean E:I was significantly lower in diabetic men and women than control men and women. An abnormally low E: I was found in 9.2% of the diabetic men, 3.3% of the control men, 3.3% of the diabetic women, and none of the control women. NCV parameters, but not E:I, were inversely correlated with fasting blood glucose and glycosylated hemoglobin levels. A positive correlation between NCV and fasting and postglucose serum insulin levels was found in both groups. These results show that clinical diabetic neuropathy is not common at diagnosis of NIDDM, but disturbances in peripheral and autonomic nerve function, detected by electrophysiological and cardiovascular reflex methods, are often present at that stage of the disease. The prevalence of neuropathy is also highly dependent on the criteria used. The degree of peripheral nerve dysfunction was associated with the degree of hyperglycemia, but an association between glycemic control and autonomic dysfunction could not be demonstrated.
- Received February 5, 1988.
- Revision received June 7, 1989.
- Accepted June 7, 1989.
- Copyright © 1989 by the American Diabetes Association