The reproducibility of volume measurements in intravascular ultrasound (IVUS) images derived from separate pull-back manoeuvres remains to be elucidated. Patients (n = 23) were imaged with IVUS prior to (first series) and following percutaneous transluminal angioplasty (PTA) (second series). In 15 patients, one matched vascular segment (3-4 cm in length), not subjected to PTA, was used for analysis of lumen, vessel and plaque volume using an automated contour analysis system. Volume measurements assessed by two independent observers and in the two separate series were compared. Interobserver differences in volume measurements were small (< or =0.4%), with low coefficients of variation (< or =1.7%) and high correlation coefficients (r = 1.00). Differences in volume measurements obtained in the two separate series were small (< or =2.6%), with low coefficients of variation (< or = 8.6%) and high correlation coefficients (r = 0.97-0.99). In conclusion, volume measurements derived from IVUS images are highly reproducible. Therefore, IVUS may be used to monitor the progression/regression of atherosclerotic plaque volume in a longitudinal study.