The integrity of endothelium-dependent vasodilation in the skin of patients with insulin-dependent diabetes mellitus (IDDM) is unclear, especially with respect to the role of nitric oxide. To examine this, forearm skin blood flow by laser Doppler flowmetry and total blood flow by venous occlusion plethysmography was measured in response to brachial artery infusions of an endothelium-dependent (methacholine) and -independent (sodium nitroprusside) vasodilator. Peak hyperemic forearm blood flow, following 5 min of arterial occlusion, was also determined. Responses were compared in 11 control subjects and 16 patients with insulin-dependent diabetes mellitus. In ten normal subjects, co-infusion of NG-monomethyl-L-arginine with methacholine produced a significant reduction in total forearm blood flow response to methacholine (p < 0.002), measured by venous occlusion plethysmography, as well as vascular conductance (p < 0.001), confirming that nitric oxide contributes to this response. In contrast, NG-monomethyl-L-arginine had no significant effect on the methacholine-induced increase in forearm skin blood flow measured by laser Doppler flowmetry indicating that factors other than nitric oxide may be involved. Increases in forearm skin blood flow in response to methacholine, sodium nitroprusside and to an ischemic stimulus were not significantly different between the normal subjects and patients with IDDM. Dose-related increases in total forearm blood flow and vascular conductance were not significantly different between control subjects and diabetic patients during infusions of methacholine. The increases in these parameters during infusions of sodium nitroprusside, however, were significantly less in the diabetic group than in the control group (p < 0.05) as was the peak reactive hyperemic blood flow (p < 0.05). Since skin blood flow was not affected, the reduced vasodilator responses to sodium nitroprusside and an ischemic stimulus in the diabetic group are in forearm skeletal muscle. The reduced muscle blood flow does not reflect a decreased vasodilatory capacity, but rather a functional impairment in response to nitric oxide and ischemia since the methacholine dilation was normal. The normal vasodilator responses in the forearm skin, which is predominantly capillary as opposed to arteriovenous anastomatic blood flow, indicate that the response to nitric oxide and an ischemic stimulus in this vascular bed is intact in patients with IDDM. This is, therefore, an unlikely cause of diabetic skin, complications in these areas.