Optimization of antihypertensive drug therapy continues to be a clinical challenge in patients with diabetes mellitus and its complications. We assessed the interference of autonomic neuropathy with drug effects on heart rate variability in 13 hypertensive diabetic subjects (mean age 48.4 years) during treatment with two blood pressure lowering drugs, metoprolol and enalapril. The baseline findings were compared with those obtained in 24 diabetic subjects without hypertension (mean age 32.5 years) and in 24 non-diabetic hypertensive patients (mean age 47.6 years). Cardiovagal autonomic neuropathy was present in 10/13 (77%) of the hypertensive diabetic group, 14/24 (58%) the non-hypertensive diabetic group and 17/24 (71%) the non-diabetic hypertensive group. Heart rate variation was studied by power spectral analysis using total variability and three different frequency bands (low-frequency 0.025-0.075 Hz, mid-frequency 0.075-0.15 Hz and high-frequency 0.15-0.40 Hz). At baseline, the two hypertensive groups showed significantly smaller mid- and high-frequency heart rate variability compared with the diabetes only group. Age and the presence of cardiovagal autonomic neuropathy were important determinants of variability. Both metoprolol and enalapril reduced blood pressure comparably in hypertensive diabetics while metoprolol also reduced heart rate. Metoprolol decreased heart rate variability at the low-frequency and mid-frequency bands even after correction for the change in heart rate. Heart rate variability was not significantly altered by enalapril. In subjects with hypertension, diabetes and autonomic neuropathy, metoprolol almost abolished all heart rate variability. Therefore, an ACE inhibitor is a more neutral treatment alternative in such patients from the point of view of autonomic cardiac control.