Reproducibility of neointima quantification with motorized intravascular ultrasound pullback in stented coronary arteries

Abstract

Intravascular ultrasound (IVUS) imaging has shown excellent reproducibility immediately after coronary stent implantation. However, the variability of measurements in lesions late after stent implantation, when neointima formation is present, has not been studied. Neointimal tissue is generally low echogenic and thus difficult to quantify. We therefore sought to analyze the reproducibility of morphometric measurements late after stent implantation. Fifty consecutive patients were investigated 6 months after Palmaz-Schatz stent implantation (motorized catheter pullback 0.5 mm/s). Two experienced investigators independently identified the stent area, lumen area, and neointimal area at different sites within the stent. Planimetric measurements were performed with commercially available software. Correlation coefficient and mean difference for corresponding measurements were calculated for the intraobserver and interobserver comparisons. Variability for the intraobserver and interobserver comparisons was similar. Observer agreement regarding the presence of neointimal hyperplasia was as high as 71% (interobserver comparison 62%). The mean difference for neointima area was 0.06 +/- 1.5 mm(2) (-0.6 +/- 1.5 mm(2)); mean differences for lumen area were 0.02 +/- 0.19 mm(2) (0.03 +/- 0.17 mm(2)) and for stent area 0.01 +/- 0.09 mm(2) (-0.02 +/- 0.12 mm(2)) (values for interobserver comparison are given in parentheses). Correlation between measurements was high for all structures: correlation coefficients were 0.66 (0.69) for neointima, 0.94 (0.95) for lumen, and 0.95 (0. 91) for stent area. Morphometric measurements of IVUS investigations with motorized IVUS pullback late after stent placement show good reproducibility. Intraobserver variability and interobserver variability are low. Differences for corresponding measurements were more pronounced for neointima area. Motorized catheter pullback guarantees high reliability of IVUS measurements and should be used routinely for clinical IVUS studies.