The purpose of this study was to examine the possible difference in the 24-hr BP profile--including short-term BP variability, assessed as the standard deviation--between diabetic and non-diabetic hypertensives. We measured 24-hr ambulatory BP in 11 diabetic hypertensives (diabetic HT) and 10 non-diabetic hypertensives (non-diabetic HT) who were hospitalized for the educational program in our hospital and were under stable salt intake. Renal function and sleep apnea were also estimated. There were no significant differences in 24-hr systolic BP (141 mmHg vs. 135 mmHg, ns), daytime systolic BP (143 mmHg vs. 138 mmHg, ns), and nighttime systolic BP (135 mmHg vs. 130 mmHg, ns) between diabetic HT and non-diabetic HT. The values of 24-hr HR (69.7 beats/min vs. 65.2 beats/min, ns) and 24-hr HR variability (9.9 beats/min vs. 10.1 beats/min, ns) were also similar between the groups. Interestingly, diabetic HT had a significantly greater 24-hr systolic and diastolic BP variability than non-diabetic HT (18.2 mmHg vs. 14.5 mmHg, p < 0.05; 11.5 mmHg vs. 9.6 mmHg, p < 0.05, respectively). The values for creatinine clearance, urinary protein excretion, and apnea-hypopnea index were similar between the groups. Bivariate linear regression analysis demonstrated that fasting blood glucose was the primary determinant of 24-hr diastolic BP variability (r = 0.661, p < 0.01). Multiple stepwise regression analysis revealed that fasting blood glucose was a significant and independent contributor to 24-hr systolic BP variability (r = 0.501, p < 0.05). Taken together, these results demonstrate that BP variability is increased in diabetic hypertensives. Furthermore, it is possible that an elevation of fasting blood glucose may contribute to the enhanced BP variability in hypertensives.