Failure of improved glycaemic control to reverse diabetic autonomic neuropathy. The St Thomas's Diabetic Study Group.

A prospective trial was conducted in 20 insulin-treated diabetic patients with established autonomic dysfunction to test the effect of a 2-year improvement in glycaemic control. On entry, the patients aged 34.6 +/- 10.8 years (mean +/- SD), duration 20.2 +/- 10.4 years, had evidence of poor control (HbA1 greater than 9.5% on at least two consecutive tests, and at least one abnormality in the 11 different neuropathic function tests used. These included four of pupillary function, six of cardiovascular function and one of vibration sensitivity. The method used to improve control was conventional insulin treatment combined with home monitoring of blood glucose and regular adjustment of insulin dosage under the supervision of a diabetic specialist nursing sister. The HbA1 level on entry was 13.0 +/- 0.7% (mean +/- SEM) which fell to 10.4 +/- 0.4% at 3 months and to 9.3 +/- 0.4% at 24 months. The neuropathy tests were performed at 0, 3, 6, 12, 18, and 24 months and significant changes were recorded in five. These changes (in two pupillary and three cardiovascular tests) all indicated worsening function which exceeded that explicable by aging. One cardiovascular test showed slower rates of deterioration in patients achieving the greatest improvements in control. It was concluded that a prolonged period of improved control failed to reverse established autonomic dysfunction of diabetes.