Drs. Fennis and Vredeveld suspect that our ultrasound method was incapable of detecting a 0.08-mm difference between the exercise and reference groups. They rely on basic ultrasound physics to justify their concern. They would be correct, if we had dealt only with repeated measurements in a single patient. However, our study was based on several hundred repeated measurements over 6 years in the 2 groups. Statistical analyses were based on the intention-to-treat principle. We know of no evidence for Fennis and Vredeveld's strong statement that "it is commonly accepted that the standard measurement error is 2 to 3 times [the] wavelength." During the 6-year period in which we also performed quality control scans, the mean difference between repeated scans was 0.0037 mm. We hope that this explanation will allow correct interpretation of our data. Finally, any inaccuracy in the method used to measure an outcome variable tends to underestimate the true effect of an intervention. Therefore, the antiatherosclerotic effect of regular aerobic physical exercise is likely to be even stronger than we observed in the DNASCO (DNA Polymorphism and Carotid Atherosclerosis) Study.