An effect on microvascular function has been proposed as a possible mechanism explaining the association of acute smoking with increased blood pressure and decreased insulin sensitivity. However, the effects of smoking on microvascular function have not been studied. We have investigated the acute effects of smoking on microvascular function in 12 healthy smokers. Before and after smoking a cigarette, we measured heart rate, blood pressure and capillary recruitment during peak reactive hyperaemia. We also measured endothelium-dependent and endothelium-independent vasodilatation of the skin microcirculation with iontophoresis of acetylcholine and sodium nitroprusside respectively combined with laser Doppler fluxmetry. To exclude non-specific changes, a control study with sham smoking was performed. The smoking and sham smoking studies were conducted in a randomized order. Compared with sham smoking, acute smoking caused increases in heart rate (smoking, 9.3+/-4.1 beats/min; sham, -1.3+/-3.0 beats/min; P < 0.001) and systolic blood pressure (smoking, 6.3+/-8.8 mmHg; sham, 0.8+/-4.4 mmHg; P < 0.05); decreases in absolute (smoking, -4.9+/-6.9 per mm(2); sham, 0.8+/-2.1 per mm(2); P = 0.01) and relative (smoking, -13.8+/-21.4%; sham, 1.9+/-6.9%; P = 0.02) capillary recruitment during peak reactive hyperaemia; and decreases in absolute [smoking, -62.4+/-47.7 perfusion units (PU); sham, -30.8+/-32.6 PU; P = 0.04] and relative (smoking, -147+/-163%; sham, 32+/-225%; P = 0.07) vasodilatation caused by acetylcholine. Absolute (smoking, -31.6+/-58.5 PU; sham, -8.4+/-44.0 PU; P = 0.3) and relative (smoking, -50.2+/-219.0%; sham, -17.1+/-139%; P = 0.7) vasodilatation caused by sodium nitroprusside were not affected. Thus acute smoking is associated with impaired capillary recruitment during peak reactive hyperaemia and impaired microvascular endothelium-dependent vasodilatation. These findings may explain the increased blood pressure and decreased insulin sensitivity that have been observed after acute smoking.