Effect of Glycemic Control on Heart Rate Variability in Type I Diabetic Patients With Cardiac Autonomic Neuropathy Andrew J. Burger, MD, Larry A. Weinrauch, MD, John A. D’Elia, MD, and Doron Aronson, MD Diabetic cardiac autonomic neuropathy (CAN) is associ- ated with a high risk of cardiovascular events. Previous studies have shown that strict glycemic control slows the deterioration of CAN as assessed by standard auto- nomic function tests but fails to show reversibility. The aim of this study was to evaluate the effect of glycemic control on early and advanced CAN in type I diabetic patients using power spectral analysis of heart rate variability (HRV). Ten patients with early and 13 patients with advanced CAN were enrolled in a program of intensified insulin treatment. Standard autonomic func- tion tests and 24-hour time and frequency domain HRV parameters were obtained at baseline, 3, 6, and 12 months. Hemoglobin A 1C decreased from 9.5  0.4% to 8.4  0.5% (p  0.02) in the early CAN group, and from 9.3  0.4% to 8.2  0.5% (p  0.006) in the advanced CAN group. In general, both time and frequency domain HRV indexes tended to improve in patients with early CAN but continued to deteriorate in patients with ad- vanced CAN. The low- and high-frequency power in- creased in patients with early CAN (229  95 to 626  563 ms 2 and 62  30 to 183  168 ms 2 , respectively). The high-frequency power significantly improved at 12 months over baseline (p  0.04), indicating increased parasympathetic tone. By contrast, these parameters continued to deteriorate in patients with advanced CAN (65  32 to 46  8 ms 2 and 193  75 to 144  33 ms 2 , respectively). Autonomic function tests showed no sig- nificant change in both groups. These data show that a reversible metabolic component of CAN exists in pa- tients with early CAN. Power spectral analysis of HRV allows early identification of potential reversibility as early as 1 year after the institution of strict glycemic control. 1999 by Excerpta Medica, Inc. (Am J Cardiol 1999;84:687– 691) A utonomic neuropathy is extremely common in diabetes, 1,2 and carries a considerably high risk of cardiovascular mortality and morbidity. 3–6 Cardiac autonomic neuropathy (CAN) is thought to promote the genesis of ventricular arrhythmias, 5,6 silent isch- emia, 7 and myocardial infarction. 8 Although the salu- tary effects of intensive diabetes therapy on peripheral nerve function have been shown in the Diabetes Con- trol and Complication Trial, 9 attempts to halt or re- verse the course of CAN through improvement in diabetic control have generally met with little suc- cess. 9 –14 Previous studies on the effect of glycemic control on autonomic function have evaluated auto- nomic function using a battery of instantaneous car- diac autonomic function tests. 9 –13 However, standard autonomic tests may not be sensitive enough to detect subtle effects of interventions on autonomic nerve function. 15,16 Power spectral analysis of heart rate variability (HRV) has proven useful in evaluating cardiovascular autonomic activity in diabetic pa- tients 15,16 and may be a sensitive method for evaluat- ing the effect of improved glycemic control on dia- betic CAN. In the current study, we determined the effect of stringent glycemic control on HRV in type I patients with CAN. METHODS Patients: The study group consisted of 23 type I diabetic patients (age range 21 to 65 years), who were prospectively enrolled as part of a multicenter study. The study compared the effects of a program of strin- gent 3 to 4 insulin injections per day with the effects of the same insulin regimen plus the addition of weekly intravenous insulin infusion therapy. The in- stitutional review board approved the study for human research. Exclusion criteria were as follows: (1) causes of neuropathy other than diabetes (e.g., chronic alcohol abuse, vitamin B12 deficiency, hypothyroidism, drug- induced neuropathy) and significant neurologic dis- ease (e.g., Parkinson’s disease, epilepsy, recent stroke); (2) coronary artery disease or congestive heart failure; (3) a creatinine clearance of 30 ml/min; (4) pregnancy; and (5) use of medications with potential influence on autonomic nerve function (e.g., - and -adrenergic) except for angiotensin-converting en- zyme inhibitors. The latter were administered to pa- tients with concomitant diabetic nephropathy (6 with early and 10 with advanced CAN). At entrance, a full history was obtained and phys- ical examination was performed. Routine electrolytes, blood urea nitrogen, creatinine, hematocrit, hemoglo- bin A 1C , liver function tests, serum proteins, and lipid profile were obtained. Standardized instantaneous au- tonomic tests were obtained as recommended for clin- From the Division of Cardiology, Beth Israel Deaconess Medical Center; and the Joslin Diabetes Center, Boston, Massachusetts. Manu- script received March 1, 1999; revised manuscript received and accepted April 30, 1999. Address for reprints: Andrew J. Burger, MD, Beth Israel Deaconess Medical Center, West Campus, Noninvasive Cardiology Laboratory, Baker-3, 1 Deaconess Road, Boston, Massachusetts 02215. E-mail: aburger@caregroup.harvard.edu. 687 ©1999 by Excerpta Medica, Inc. All rights reserved. 0002-9149/99/$–see front matter The American Journal of Cardiology Vol. 84 September 15, 1999 PII S0002-9149(99)00417-8