Optical Coherence Tomography Evaluation of In-Stent Restenotic Lesions with Visible Microvessels


Byeong-Keuk Kim, MD1, Jung-Sun Kim, MD1, Dong-Ho Shin, MD1, Young-Guk Ko, MD1, Donghoon Choi, MD1, Yangsoo Jang, MD1,2, Myeong-Ki Hong, MD1,2

Abstract: Objective. We sought to evaluate the characteristics of in-stent restenosis (ISR) lesions with microvessels, detected by an optical coherence tomography (OCT). Background. No sufficient in vivo data exist regarding microvessel characteristics in ISR lesions. Methods. Among 78 ISR lesions (drug-eluting stent, n = 72; bare-metal stent, n = 6) in our OCT registry database, visible microvessels were detected in 21 (27%). Microvessels were defined as low backscattering structures <200 μm in diameter on OCT. Clinical, angiographic, and OCT findings were compared between lesions with and without microvessels. Results. Lesions with microvessels had a larger reference vessel diameter (2.90 ± 0.47 mm vs 2.58 ± 0.42 mm; P=.009) and post-stent minimum lumen diameter (2.76 ± 0.29 mm vs 2.54 ± 0.39 mm; P=.033) than those without microvessels. From OCT findings at the segment with minimal lumen cross-sectional area (CSA), neointimal hyperplasia (NIH) CSA (5.4 ± 1.7 mm2  vs 4.2 ± 2.1 mm2; P=.024) and percent NIH CSA (NIH CSA × 100/stent CSA) were significantly greater in lesions with microvessels (79 ± 12% vs 67 ± 16%; P=.001). On multivariate analysis, reference vessel diameter (odds ratio [OR], 4.64; 95% confidence interval [CI], 1.05-20.4; P=.043) and percent NIH CSA at the segment with minimal lumen CSA (OR, 1.06; 95% CI, 1.01-1.12; P=.021) were independent predictors of microvessels. From receiver operating characteristic analysis, the cut-off values of reference vessel diameter and percent NIH CSA predicting the presence of microvessels were 3.1 mm and 74%, respectively. Conclusions. Visible microvessels in ISR lesions might be associated with increased vessel size and extent of NIH.

J INVASIVE CARDIOL 2012;24:116–120

Key words: optical coherence tomography, restenosis, stent


Many attempts have been made to find the mechanisms or factors related to excessive neointimal hyperplasia (NIH), which causes in-stent restenosis (ISR) after stent implantation.1-3 From an autopsy study, it was suggested that extensive neovascularization at sites of stent restenosis may play a key role in neointimal proliferation, progression, and further aggravation of neoatherosclerosis.4 However, minimal in vivo data exist regarding in-stent neovascularization or characteristics of the neointima. Optical coherence tomography (OCT), a newly developed intracoronary imaging modality, has provided high-resolution cross-sectional images of tissues and enabled identification of microstructures that could not be visualized by previous standard imaging tools such as intravascular ultrasound.5-7 A previous OCT study reported in vivo identification of visible microvessels that were suggestive of neovascularization within NIH.7 Therefore, we hypothesized that ISR lesions with microvessels may have different neointimal characteristics than those without microvessels. Clinical, angiographic, and OCT findings of lesions with microvessels were compared to those of lesions without microvessels.


This study included 78 lesions in 70 patients that were identified as ISR from the OCT follow-up registry database at our institute.8,9 Inclusion and exclusion criteria for follow-up OCT procedures have been previously reported.8,9 All available stents including drug-eluting stents (DES) and bare-metal stents (BMS) were evaluated in this OCT registry. Specifically, DES in this OCT follow-up registry were coated with sirolimus (Cypher; Cordis Corporation), paclitaxel (Taxus; Boston Scientific), everolimus (Xience, Abbott Vascular), or zotarolimus (Endeavor Sprint or Resolute; Medtronic). Stent selection at the time of coronary intervention was at the physician’s discretion. Stent implantation was performed using conventional techniques. In DES cases, dual antiplatelet therapy (aspirin and clopidogrel) was given to all patients until follow-up angiogram and OCT were performed. The study protocol was approved by the institutional review board of our institutions, and written informed consent was obtained from all patients before the procedure.

Quantitative coronary angiography analysis was performed using an offline quantitative coronary angiography system (CASS system II; Pie Medical Imaging) before and after stent implantation and at follow-up angiogram. The minimal lumen diameter (MLD) of the treated coronary segments and reference segment diameter were measured in the view that was the most severe and not foreshortened at each time. Angiographic late loss was defined as the difference between follow-up and postprocedure MLD. Angiographic restenosis was defined as diameter stenosis of more than 50% inside the stent or 5 mm segment proximal or distal to the stent.2 The patterns of angiographic ISR were classified as suggested by Mehran et al.10

OCT examination using a conventional OCT system (Model M2 Cardiology Imaging System; LightLab Imaging) with a motorized pull-back system at 1.0 mm/s was previously described.8,9 OCT analysis was performed by an independent investigator blinded to patient and procedural information. The region of interest included the stented segments and adjacent segments (5 mm proximal and distal to the stent margins). Cross-sectional OCT images were analyzed at 1 mm intervals (every 15 frames). Inadequate images, including noncircumferential cross-sections, poor image quality, or cross-sections with major side branches (diameter ≥2.0 mm), were excluded from this analysis. Stent and luminal cross-sectional areas (CSAs) were measured at 1 mm intervals, and NIH CSA was calculated as the stent CSA minus the luminal CSA. Percent NIH CSA was calculated as NIH CSA × 100/stent CSA. Mean values are reported in this study. The thickness of NIH was measured as the distance between the endoluminal surface of the neointima and the stent strut.11 An uncovered strut was defined as having an NIH thickness of 0 µm.11 A malapposed strut was defined as a strut that had detached from the vessel wall (Cypher ≥160 μm; Taxus ≥130 μm; Endeavor Sprint or Resolute ≥110 μm; Xience ≥100 μm).11-13 The percentage of malapposed or uncovered struts in each stented lesion was calculated as (number of malapposed or uncovered struts/total number of struts in all cross-sections of the lesion) × 100.

In qualitative OCT assessment, visible microvessels were defined as the well-delineated low backscattering structures <200 μm in diameter and showing a trajectory within the vessel.7 Representative cases of microvessels are shown in Figure 1. Neointimal tissue morphology was classified into homogenous, heterogeneous, or layered appearance based on optical properties and backscattering pattern in the segment with maximal luminal narrowing, as defined in a previous OCT study.7 Thrombus was defined as a signal rich, low or high backscattering, irregular mass protruding into the lumen; or an intraluminal mass unconnected from the surface of vessel wall that had signal-free shadowing on the OCT image (more than 250 μm at the thickest point).8 All lesions were evaluated by OCT before any interventional procedure to avoid injury to restenotic tissues.

Statistical analyses were performed using Statistical Analysis System software Version 9.1.3 (SAS Institute). Categorical data are presented as numbers and percentages, and were compared using Chi-square statistics or Fisher’s exact test. Continuous data are presented as mean ± standard deviation and compared using the Student’s t-test. If the distributions were skewed, a nonparametric test was used. To determine the independent predictors of microvessel presence in neointimal tissue, multivariate logistic regression analysis was conducted. The variables with P<.1 on univariate analysis were included. For the cut-off values of the independent predictors of microvessel presence on multivariate logistic regression analysis, receiver operating characteristic (ROC) curve analysis was performed.


Microvessels were detected in 21 of 78 ISR lesions (27%). Baseline clinical characteristics of ISR lesions with and without microvessels are summarized in Table 1. There were no significant differences in baseline clinical characteristics between the two groups. Baseline angiographic and procedural characteristics are shown in Table 2. Reference vessel diameter, postprocedural MLD, and implanted stent diameter were significantly larger in lesions with microvessels. The incidences of ISR lesions with microvessels were not statistically different between BMS (3 of 6 lesions, 50%) vs DES (18 of 72 lesions, 25%; P=.335) or between paclitaxel-eluting stents (7 of 29 lesions, 24%) vs other DES (11 of 43 lesions, 26%; P=1.000). Results of OCT analysis are shown in Table 3. There was a tendency toward a longer time interval from stent implantation to follow-up OCT in the lesions with microvessels. From OCT findings at the segment with minimal lumen CSA, NIH CSA (5.4 ± 1.7 mm2 vs 4.2 ± 2.1 mm2; P=.024) and percent NIH CSA (79 ± 12% vs 67 ± 16%; P=.001) were significantly greater in lesions with microvessels. On multivariate logistic regression analysis, reference vessel diameter (odds ratio [OR], 4.64; 95% confidence interval [CI], 1.05-20.4; P=.043) and percent NIH CSA at the segment with minimal lumen CSA (OR, 1.06; 95% CI, 1.01-1.12; P=.021) were independent predictors of microvessel presence. From the receiver operating characteristic curve analysis, the cut-off values of reference vessel diameter and percent NIH CSA to predict the presence of microvessels were 3.1 mm (area under the curve [AUC], 0.698; 95% CI, 0.540-0.857; P=.013; sensitivity, 53%; specificity, 89%) and 74% (AUC, 0.731; 95% CI, 0.612-0.850; P=.002; sensitivity, 76%; specificity, 68%), respectively (Figures 2A and 2B). Based on the cut-off values, the incidences of microvessels on OCT were compared. Lesions with reference vessel diameter ≥3.1 mm or percent NIH CSA ≥74% showed significantly higher incidence of microvessels than those with reference vessel diameter <3.1 mm (65% vs 16%; P<.001) or percent NIH CSA <74% (46% vs 12%; P<.001), respectively. When the incidences of microvessels were compared among the four groups based on the cut-off values of reference vessel diameter and percent NIH CSA, the lesions with reference vessel diameter ≥3.1 mm and percent NIH CSA ≥74% (80%) showed the highest incidence of microvessels compared to the other lesions (P<.001; Figure 2C).

Post new comment

  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.

More information about formatting options

Enter the characters shown in the image.