OBJECTIVE This study sought to assess the coronary flow reserve (CFR) in patients with pure vasospastic angina (VSA). RESULTS The phasic flow velocities of both spasm-positive and spasm-negative coronary arteries of the left anterior descending artery (LAD) were recorded at rest and during hyperaemia (50 microg of adenosine triphosphate infusion intracoronary) using a 0.014 inch, 15 MHz Doppler guide wire in 42 patients with pure VSA and acetylcholine (ACh)-induced coronary artery spasms (20-100 microg), and 23 controls with normal coronary arteries without ACh-induced vasospasm. These 42 patients had 16 vessels with focal spasms (>99%), 17 vessels with diffuse spasms (>90%) in the LAD, and nine vessels with ACh-induced spasms in the right coronary artery, but not the LAD. Coronary flow reserve was obtained from the ratio of the hyperaemic/baseline time-averaged peak velocity. Coronary flow reserve did not differ between patients with VSA and the controls (2.9+/-0.8 versus 3.2+/-0.7, NS). Moreover, CFR did not differ among the four cases (focal: 2.8+/-0.7; diffuse: 3.0+/-0.9; non spasm: 2.9+/-0.7 versus controls: 3.2+/-0.7, respectively, NS). Coronary flow reserve in vessels with proximal spasms was significantly higher than that in vessels with mid or distal spasms (3.4+/-0.8 versus 2.6+/-0.6, 2.6+/-0.9, p<0.05). The only significant correlation was between CFR and age (p=0.0275) or the duration of angina before admission (p=0.0405). CONCLUSIONS There was no difference in CFR in patients with ACh-induced spasms between the spasm-positive and spasm-negative vessels. Moreover, CFR was maintained normally in vessels with diffuse spasms, as in those with focal spasms. The most important determinant factors for CFR in patients with VSA were age and the duration of angina before admission.