Comparison of coronary arterial lumen dimensions on angiography and plaque characteristics on optical coherence tomography images and their changes induced by statin

This was a retrospective study of a single blind and randomized clinical trial registry (ClinicalTrial.gov registered number: NCT01023607), designed to assess the effects of statins on lipid-rich plaques. The inclusion criteria were as follows: 1) were statin-naïve (i.e., those who were not on statins) patients with one or more mild-to-moderate lesion(s); 2) had non-culprit lesions identified as lipid-rich plaques by OCT; 3) were to receive statin treatment during the following 12–month follow-up period; and 4) were to undergo CAG and three-vessel OCT imaging procedures during the 12–month follow-up period. The exclusion criteria were as follows: 1) a life expectancy ≤12 months; 2) contraindications to oral statins; 3) severe hepatic or renal dysfunction; or 4) congestive heart failure. Fifty-one patients including those who didn’t have the follow-up information (n = 31), had poor-quality images from OCTs (n = 13), and mismatched imaging (n = 7) were excluded. After screening, 69 patients (62.3% male; age: 55.7 ± 9.4 years) with 97 coronary non-culprit, lipid-rich plaques (30.1% located in the left anterior descending artery, 46.6% in right coronary artery, 17.5% in left circumflex artery) were analyzed in the current study. This study was approved by the institutional review board of the Second Hospital of Harbin Medical University (Harbin, China), and all patients provided written informed consent before any procedures were initiated.

According to the study protocol, follow-up angiography and three-vessel OCT images were performed at baseline and 12 months. Quantitative angiographic of the coronary arteries was performed to determine the minimal lumen diameter (MLD), diameter stenosis (DS) and lesion length as previously described in detail [17]. The interpreter was blinded to OCT findings. Analyses of the baseline and follow-up coronary angiograms were conducted using the Cardiovascular Angiography Analysis System (CAAS 5.10, Pie Medical Imaging B.V., Maastricht, the Netherlands).

After detecting by CAG and dealing with the culprit/target lesions, intracoronary OCT images were performed as previously described [4, 5]. In short, with the TD OCT system, a 3-F occlusion balloon catheter was advanced proximal to the lesion and inflated at 0.5 atm. to 0.7 atm. During image acquisition, lactate Ringer’s solution was infused intra-coronary from the distal tip of the occlusion balloon catheter at 0.5 ml/s to 2.0 ml/s by a high-pressure injector. The OCT wire was pulled back from a distal to a proximal position automatically at a rate of 1.0 mm/s. The fibrous-cap thickness (FCT) was measured 3 times at the thinnest place, and the values were averaged. A lipid-rich plaque was defined as previously that plaque with fibrous-cap thickness < 120 μm and lipid arc > 100° [5]. In lipid-rich plaque, the lipid arc was measured at 1 mm interval, the maximum value was recorded as maximum lipid arc, and the mean value was calculated as mean lipid arc. Lipid length was measured on a longitudinal view. Lipid index, was calculated as the mean lipid arc multiplied by lipid length [18]. A thin-cap fibroatheroma (TCFA) has rich lipid core and FCT < 65 μm, while a thick-cap fibroatheroma (ThCFA) was a LRP with a thick fibrous cap (≥65 μm) [3, 19]. Macrophage accumulation was characterized as increased signal intensity, combined with heterogeneous backward shadows within the lesion. Microvessel, also named neovascularization, appears as signal-free tubuloluminal structure with a diameter of 50–300 μm that was observed as ≥ 3 consecutive cross-sectional OCT images [20]. Beside lipid pool, the thin and linear region of high intensity was cholesterol crystal [21]. Intraluminal thrombus was identified as protruding masses attached to the arterial wall [22]. Calcification was recorded when an area showed low backscatter signal and a sharply delineated border. Plaque disruption was a discontinued fibrous membrane with underlying communication between the lumen and the cavity.

All OCT images were analyzed using proprietary offline software (LightLab Imaging, Inc., Westford, MA, USA) by two independent, experienced observers. In cases in which was disagreement between them, a third independent investigator helped reach a consensus on image matching and quality. All investigators of CAG and OCT images were blinded to the patient’s conditions and the study protocol, and all images were digitally stored for future analyses. In the follow up studies, repeated measurements from the same reference points were performed, and the measurements of CAG and OCT between the two points were determined using computer software.

All statistical analysis was performed by an independent statistician using IBM SPSS version 17.0 (SPSS Inc., Chicago, Illinois, USA). Data are expressed as mean ± standard deviation (SD) or counts and proportions. Quantitative data at baseline and at follow-up are compared using paired t test and qualitative data using McNemar test. Continuous variables between TCFAs and ThCFAs were compared by using unpaired t test. The categorical variables were analyzed using the Chi-square or Fisher exact test. Pearson or Spearman ranks test was used to investigate the relationship between CAG findings and OCT measurements, and also the correlations between changes in CAG and changes in OCT measurements. A p-value < 0.05 was considered to be statistically significant.

First, it was inevitable that the retrospective study from a clinical registry introduced a potential selection bias. Second, because of the short duration and small size of population of our study, the changes in CAG data did not correlated with the changes in OCT measurements. Additional studies with larger populations and longer term follow-up might be needed. Third, the definition of OCT-derived TCFAs does not correspond exactly with the histological definition of TCFAs because OCT cannot measure the necrotic core size; the macrophage infiltration, micro-vessels, cholesterol crystals, thrombus, calcification, and disruption were not quantified in the present study. Finally, the general results of non-culprit, lipid-rich plaques in this study cannot be applied to other types of plaques.