purpose: The arterial pressure waveform is derived from the complex interaction of the left ventricular stroke volume and the physical properties of the arterial circulation. Widespread abnormalities in the physical characteristics of the arterial vessels associated with diabetes mellitus can produce consistent changes in the shape of the pressure pulse waveform, providing information about arterial structure and tone that can be quantitated by pulse contour analysis. patients and methods: We analyzed intra-arterial branchial artery waveforms in 28 patients with non-insulin-dependent diabetes mellitus and 22 control subjects matched for age and sex. A computer-based assessment of the diastolic pressure decay and a modified Windkessel model of the circulation were employed to quantify changes in arterial waveform morphology in terms of the large-artery compliance (C1), the oscillatory diastolic waveform (C2), inertance, and systemic resistance. results: No differences were found in heart rate, mean arterial pressure, cardiac output, or stroke volume between groups. The mean oscillary arterial compliance estimate was significantly reduced in diabetic subjects versus controls: 0.02 (95% confidence interval [C1], 0.01 to 0.03) mL/mm Hg versus 0.08 (95% CI, 0.04 to 0.12) mL/mm Hg (p <0.001). Oscillatory compliance values were uniformly reduced in the diabetic subjects regardless of the presence or absence of physical complications of the disease. No differences in large-artery compliance, inertance, or systemic resistance were found between groups. No positive correlations were found between indices of glycemic control, the known duration of diabetes, and any of the hemodynamic variables. conclusions: Quantitative changes in the arterial pressure pulse waveform, reflected by a reduced oscillatory compliance estimate, were found in patients with non-insulin-dependent diabetes mellitus. This estimate appears to act as an early marker for the vascular abnormalities associated with diabetes before complications of the disease become clinically apparent. By contrast, no changes in large-artery compliance were found in this patient population free from clinically obvious macrovascular disease.