Reproducibility of quantitative measurements from intracoronary ultrasound images: Beat-to-beat variability and influence of the cardiac cycle

OBLECTIVES: The purpose of this study was to determine the variability ofquantitative measurements from intracoronary ultrasound images,and the influence of the cardiac cycle on this variability,as a basis for in vivo applications. METHODS: Methods Two observers analysed 30 MHz cross-sectional imagesfrom 96 in-vivo coronary arterial sites. By computer-assistedcontour tracing we determined lumen area, vessel area, lesionarea (vessel area minus lumen area) and percent obstruction(l00% x lesion area/vessel area). Intra- and inter-observerand beat-to-beat variability, and systolic to diastolic differenceswere calculated by paired analysis. RESULTS: Consistent intra- and inter-observer differences (bias) weresmall (0·9%). Random variations in the two direct parameterswere 21·l%, but for the two derived parameters they wereup to 40%. For all four parameters, random inter-observer variabilitywas significantly greater (up to 119% for vessel area: 19·3vs 8·8%) than intraobserver variability, but consistentvariability was similar. Consistent beat-to-beat differenceswere small (1·4%), random variations were 8·9%to l7·5%. Random beat-to-beat variability for all fourparameters was greater in diastolic than in systolic frames(up to 47·0% difference (11·9 vs 17·5%)for lesion area). Vascular dimensions were significantly greaterin systole (2%). We found an error of 0·24 mm (2SD) forintra-observer variability of calculated mean arterial diameters,which is similar to the error described in angiographic studies(0·22 mm). CONCLUSIONS: Quantitative measurements from intracoronary ultrasound imagesgenerally reproduce well. It is preferable to use directly measuredparameters as opposed to derived parameters, as they are lesssubject to variability. Variability can be reduced by selectingsystolic images. (Eur Heart J 1996; 17: 1593–1599)