To study the determinants of smooth muscle injury during balloon angioplasty, we conducted a series of experiments to examine the effects of degree of arterial stretching, duration of balloon inflation, and arterial precontraction on smooth muscle injury after balloon angioplasty in isolated, perfused whole-vessel segments of rabbit aortas and dog carotid arteries. Freshly dissected rabbit thoracic aortas and dog carotid arteries were mounted in a muscle bathperfusion chamber and perfused at 80 mm Hg. The proximal half of each aorta was dilated with a 5-mm (41+6%o stretch), 6-mm (64±6% stretch), or 8-mm (97±9% stretch) balloon angioplasty catheter, and the uninjured half of each vessel served as the control. The vasoconstrictor behavior of the dilated segment was then assessed by dose-response testing; long-axis, ultrasonic imaging combined with computerized edge-detection image processing was used to measure changes in segmental internal vessel diameters that were induced by phenylephrine. A similar series of experiments was performed in dog carotid arteries with 5-mm balloon catheters (42 ±2% stretch) to compare the susceptibility to smooth muscle injury between elastic (aortic) and muscular (carotid) arteries. Additional experiments were performed to determine the roles of prolonged (30 minutes) balloon inflation and arterial precontraction on smooth muscle injury after balloon angioplasty. In rabbit aortas, the dilated arterial segments demonstrated normal reactivity to phenylephrine after dilatation with 5-and 6-mm balloons (p=NS versus control). Severe smooth muscle injury (histopathologically) with "arterial paralysis" was observed after severe stretch (8-mm balloon) and after 5-mm balloon dilatation (46±5% stretch) in precontracted vessels. Prolonged balloon inflations did not impair aortic vasoconstrictor behavior. Dog carotid (muscular) arteries demonstrated angioplasty-induced smooth muscle injury with less severe degrees of stretch (47-52% stretch). Geometric modeling suggests that medial stretching during balloon angioplasty of diseased vessels in vivo is in the range of 15-41%. We conclude that 1) relatively severe arterial stretching is required to injure smooth muscle when balloon angioplasty is performed in relaxed arteries, 2) prolonged balloon inflation does not alter the severity of smooth muscle injury after balloon dilatation in isolated arteries, 3) muscular arteries appear more susceptible to stretch-induced smooth muscle injury than do elastic arteries, 4) precontraction serves as a catalyst in promoting stretch-induced smooth muscle cell lysis and arterial paralysis, and 5) these findings are consistent with the clinical observations that coronary angioplasty rarely results in arterial paralysis. (Circulation 1990;82:2170-2184 B alloon angioplasty is a widely accepted modality for the treatment of obstructive peripheral and coronary artery disease. Despite the widespread application of this therapy, there is still uncertainty regarding the mechanism(s) of successful angio-