Background: Caffeine intake increases walking endurance in patients with peripheral arterial disease (PAD). This study evaluated whether the caffeine effect was correlated with the degree of PAD and whether revascularization affected the walking endurance enhancing effect following caffeine. Patients and methods: We used data from a previously reported randomized crossover study of 83 outpatients with intermittent claudication. Patients received placebo and caffeine at an interval of 1 week and undertook a treadmill test to measure pain-free walking distance (PWD) and maximum walking distance (MWD). This was repeated 3 months later in 47 nonoperatively treated patients and 36 patients who underwent revascularization. The degree of PAD was estimated by ankle-brachial index (ABI). Data were analyzed using a mixed model with subject-specific random intercepts. Results: For MWD, the correlation between caffeine effect and ABI was a mean [95% confidence interval; CI] of -0.14 [CI -0.71-0.44], and for PWD it was -0.02 [CI -0.75-0.70]. Caffeine increased MWD by 20% [CI 7-32] and 15% [CI -0.04-29] in nonoperative and operative patients, respectively. This difference in caffeine effect between nonoperative and operative patients was not significant ( p = 0.6). Similarly, the difference in caffeine effect on PWD between the nonoperative and operative group was not statistically significant ( p = 0.12). Conclusions: The walking-enhancing effect of caffeine was not significantly associated with the severity of PAD.