The pathophysiological mechanisms linking cardiovascular dysautonomy to mortality are unclear. The aim of this study was to investigate the pattern of 24-h ambulatory blood pressure (BP) monitoring (ABPM) in diabetic patients with cardiovascular autonomic neuropathy (CAN). We evaluated 391 type 2 diabetic patients in a cross-sectional study. Five clinical tests of CAN were performed: heart-rate variation during deep breathing, the Valsalva maneuver, and standing, and BP variation during handgrip and standing. Patients were considered to have initial CAN if one heart-rate test was abnormal or two were borderline, and to have definite or severe CAN if at least two tests were abnormal. Differences between patients with and without CAN were assessed by bivariate tests and ANCOVA. Of the 391 patients, 230 (59%) presented clinical CAN, of whom 53 had definite or severe involvement. Patients with CAN were older, had diabetes of longer duration, and had an equal prevalence of hypertension but used more antihypertensive drugs than those without CAN. On ABPM, patients with definite or severe CAN had higher systolic BP (SBP) and pulse pressures (PP) than those without CAN, particularly in the nighttime (SBP: 128 +/- 18 vs. 117 +/- 16 mmHg, p = 0.007; PP: 58 +/- 13 vs. 50 +/- 11 mmHg, p = 0.003) and early morning (SBP: 140 +/- 18 vs. 131 +/- 17 mmHg, p = 0.05) after adjustment for potential confounders, as well as a higher prevalence of the systolic nondipping pattern (75.5% vs. 50.9%, p = 0.021). In conclusion, type 2 diabetic patients with more severe CAN have higher SBP and PP, especially during the nighttime and early morning, as well as a higher prevalence of nondipping status. This unfavorable 24-h ABPM pattern may contribute to the increased cardiovascular risk of diabetic patients with dysautonomy.