Preliminary clinical experience with laser angioplasty has suggested that arterial spasm may complicate attempts to employ laser light to accomplish vascular recanalization. The present study was designed to investigate the role of energy profile on the development of arterial spasm during laser angioplasty. Laser irradiation was delivered percutaneously in vivo to New Zealand white rabbits and to Yucatan microswine with or without atherosclerotic lesions induced by a combination of balloon endothelial denudation and atherogenic diet. Continuous wave (CW) laser irradiation from an argon ion gas laser (wavelength 488 to 514 nm) was applied to 23 arteries, while 16 arteries were irradiated using a pulsed xenon chloride (308 nm) or xenon fluoride (351 nm) excimer laser. Arterial spasm, defined as greater than 50% reduction in luminal diameter narrowing, complicated delivery of laser light to 17 (74%) of the 23 arteries irradiated with the CW argon laser. Spasm was consistently observed at powers greater than 2 W, at cumulative exposures greater than 200 seconds, and at total energy greater than 200 joules. Spasm was typically diffuse (including the length of the vessel) and protracted (lasting up to 120 minutes). Intra-arterial nitroglycerin (up to 300 micrograms) produced only temporary and incomplete resolution of laser-induced spasm. In contrast, spasm was never observed in any of the 16 arteries in which laser angioplasty was performed using a pulsed laser (0.95 to 6.37 joules/cm2, 10 to 50 Hz, 48 to 370 seconds). Thus CW but not pulsed laser angioplasty may be complicated by arterial spasm.(ABSTRACT TRUNCATED AT 250 WORDS)