The goal of this study was to examine the hypothesis that increases in pulse pressure produce hypertrophy of cerebral arterioles, even in the absence of increases in mean pressure. Sprague-Dawley rats underwent creation of an arteriovenous fistula and clipping of one carotid artery at 1 month of age. Rats that underwent exposure of the abdominal aorta without fistula production and unilateral carotid clipping served as controls. At about 6 months of age, the mechanics of sham and clipped pial arterioles were examined in vivo in anesthetized rats. Stress-strain relations were calculated from measurements of pial arteriolar pressure (servo null) and diameter and cross-sectional area of the arteriolar wall. Point counting stereology was used to quantify individual components in the arteriolar wall. Before deactivation of smooth muscle with EDTA, cross-sectional areas of the vessel wall and pulse pressures in sham pial arterioles were significantly greater (P < .05) in arteriovenous fistula rats than in control rats (cross-sectional area, 1468 +/- 100 versus 1129 +/- 104 microns 2; pulse pressure, 26 +/- 1 versus 14 +/- 1 mm Hg). In contrast, systolic and mean pressures in sham arterioles were not significantly different and diastolic pressure was significantly less in arteriovenous fistula rats (systolic pressure, 69 +/- 1 versus 67 +/- 4 mm Hg; mean pressure, 52 +/- 2 versus 57 +/- 3 mm Hg; diastolic pressure, 43 +/- 2 versus 53 +/- 3 mm Hg). Carotid clipping normalized cross-sectional area of the vessel wall (1083 +/- 86 microns 2) and pulse pressure (12 +/- 1 mm Hg) in pial arterioles of arteriovenous fistula rats. During maximal dilatation, the stress-strain curve in sham arterioles of arteriovenous fistula rats was shifted to the right of the curve in control rats, which indicates that arteriovenous fistulae increase passive distensibility of cerebral arterioles. The proportion of distensible components in the vessel wall (smooth muscle, elastin, and endothelium) was increased in sham arterioles of arteriovenous fistula rats. These findings (1) suggest that increases in pulse pressure, even in the absence of increases in mean pressure, are sufficient to produce hypertrophy of cerebral arterioles and (2) provide support for the concept that increases in distensibility of cerebral arterioles in association with hypertrophy of the vessel wall may be related to alterations in wall composition.