Background and purpose. We assessed whether intentional undersized dilatation of targeted lesions during carotid artery stenting (CAS) carried a higher risk of in-stent restenosis (ISR) and correlation to subsequent ischemic stroke in qualifying arteries in the follow-up period. Methods. Consecutive patients undergoing CAS between April 2003 and May 2010 were retrospectively reviewed. The use of a filter device as a distal embolic protection device (EPD) was first approved by Japanese governmental health insurance in April 2008; previously, transient balloon occlusion was used off-label. Until March 2008 (Group A), the target diameter of balloon dilatation was 80–100% of the normal vessel diameter just distal to the stenotic lesion. Moderately undersized dilatation (70–80% of the normal vessel diameter) using the distal EPD was adopted in April 2008 (Group B) in an attempt to reduce the amount of released plaque debris. Results. We analyzed 132 CAS procedures (125 patients) in Group A and 53 CAS procedures (52 patients) in Group B. The mean follow-up period was 35.4 months (35.3 months in Group A and 36.0 months in Group B). Eight lesions (4.3%; 7 in Group A and 1 in Group B) developed ISR. None of the patients had symptomatic ISR, and ISR did not increase in Group B (odds ratio, 0.34; 95% confidence interval, 0.04–2.86; p = 0.32). Conclusions. Undersized dilatation of targeted lesions did not increase the risk of developing ISR, and we suggest it as a viable treatment option to prevent ischemic events during CAS.