Methods

From the Massachusetts General Hospital OCT Registry, 689 patients with previously implanted stents were identified from August 2010 to September 2012. Patients with incomplete demographic data (n = 18), unknown type of stent (n = 173), endothelial progenitor cell–capturing stents (n = 87), and polytetrafluoroethylene-covered stents (n = 1) were excluded from the study. Cases with any interventional procedure before imaging (n = 61), overlapping stents (n = 38), and stents with poor imaging quality (n = 40) were also excluded. Of the remaining 437 stents in 271 patients, we selected stents with a mean neointimal thickness (NIT) >100 μm for ≥3 consecutive 1-mm cross-sections to obtain a sufficient amount of neointima to detect tissue characteristics, as previously reported. Therefore, 252 stents in 212 patients were included in the final analysis: 100 bare metal stents (BMSs) and 152 drug-eluting stents (DESs). Hypertension was diagnosed when ≥1 of the following criteria was met: history of hypertension, use of antihypertensive drugs, systolic blood pressure ≥140 mm Hg, or diastolic blood pressure ≥90 mm Hg. Hyperlipidemia was diagnosed in case of history of hyperlipidemia, use of lipid-lowering drugs, or newly diagnosed hyperlipidemia according to institutional guidelines. Clinical and angiographic characteristics were compared between patients with and without NA. Based on the values of mean NIT, stents were divided into tertiles, and incidence and characteristics of NA were compared among the 3 groups. According to the follow-up duration, stents were assigned to an early (<9 months), intermediate (9 to 48 months), or late (>48 months) phase. The registry was approved by each institutional review board, and every patient provided informed consent.

OCT image acquisition was performed using commercially available frequency domain (C7-XR OCT Intravascular Imaging System; St. Jude Medical Inc., St. Paul, Minnesota) or time domain (M2/M3 Cardiology Imaging Systems; LightLab Imaging Inc., Westford, Massachusetts) OCT systems. The intracoronary OCT imaging technique has been previously described. All the images were de-identified, digitally stored, and submitted to the Massachusetts General Hospital OCT Registry Laboratory for analysis.

Cross-sectional OCT images were analyzed every 1 mm by 2 independent investigators (RV and TY) who were blinded to clinical and laboratory data. When there was discordance between the 2 observers, a consensus reading was obtained from a third investigator (KK). Quantitative and qualitative analyses were performed using OCT proprietary software for off-line analysis (LightLab Imaging Inc., Westford, Massachusetts). For each frame, lumen and stent areas were traced and minimum, maximum, and mean NIT were automatically calculated. NA was defined as the presence of lipid-laden intima and/or calcification inside the stent ( Figure 1 ). We defined lipid-laden intima as a diffusely bordered signal-poor region inside the stent ( Figure 1 ) and categorized it as lipid-rich intima (LRI) when lipid arc was wider than 90°. The percentage of LRI was calculated as (number of frames with LRI/total number of analyzed frames) × 100. When an LRI was covered by a fibrous cap thinner than 65 μm, the lesion was labeled as thin-cap fibroatheroma (TCFA)-like intima. In-stent calcifications were recorded as signal-poor or heterogenous regions with a sharply delineated border ( Figure 1 ). Neovascularization (NV) was defined as the presence of signal-poor holes or tubular structures with a diameter of 50 to 300 μm and categorized as peristrut and intra-intima, as previously reported. The presence of disrupted intima, defined as discontinuity of the fibrous cap connecting the lumen with the underlying lipid pool ( Figure 1 ), and thrombus, defined as a mass attached to the luminal surface or floating within the lumen, was also recorded.

Representative OCT images of NA. (A) Lipid-laden intima, appearing as a signal-poor region with diffuse borders (asterisks) , with disruption of the fibrous cap (arrow) . (B) In-stent calcification: signal-poor area with sharp borders (arrows) . Scale bar = 1 mm.

Off-line quantitative analysis of coronary angiograms recorded at the time of OCT imaging was performed using a validated edge-detection system (CAAS Version 5.1; Pie Medical Imaging, Maastricht, The Netherlands). End-diastolic frames were selected for the analysis, and the tip of the catheter was used for calibration. Reference lumen diameter, minimum lumen diameter, diameter stenosis, and lesion length were measured.

Categorical variables were expressed as counts and percentage and compared using the chi-square or Fisher’s exact test, as appropriate. Continuous variables were expressed as mean ± SD and compared using the independent samples Student t test for 2-group comparisons and 1-way analysis of variance followed by Bonferroni post hoc test for 3-group comparisons. For the comparisons of OCT findings expressed as categorical values, the observed significance was further adjusted for the multiplicity of simultaneously conducted tests by Bonferroni correction. Mean NIT was evaluated as a continuous variable, in tertiles and quartiles. All the approaches showed similar results, and we chose for simplicity to present the results by tertiles. Intra- and interobserver variability was estimated by the k measure of agreement. Multivariate logistic regression analysis was performed to determine the impact of mean NIT on the presence of NA (dependent variable), after adjusting for stent type, follow-up duration, conventional coronary risk factors, statin, and angiotensin-converting enzyme inhibitor or angiotensin II receptor blockade use. Univariate analysis was first performed, and all the variables exhibiting a p value <0.05 were entered en bloc in the multivariate model. A generalized estimating equations approach with exchangeable correlation structure was used to take into account the within-subject correlation because of the analysis of multiple stents in a single patient. A p value <0.05 was considered statistically significant. Statistical analysis was performed using SPSS 17.0 (SPSS Inc., Chicago, Illinois).

Results

Baseline patient characteristics are listed in Table 1 . Each value was compared between patients with NA in ≥1 stent and patients without NA. Serum creatinine levels were significantly higher in patients with NA, as well as total cholesterol and low-density lipoprotein cholesterol levels. Compared with patients with NA, those without NA were more frequently on statin and angiotensin-converting enzyme inhibitor or angiotensin II receptor blockade therapy. There was no significant difference in clinical presentation between patients with and without NA. However, a diagnosis of acute coronary syndrome was more frequent in patients with DES than in those with BMS (23% vs 12%, p = 0.044). In patients presenting with acute coronary syndrome, the stented segment was the culprit lesion in 47% of cases.

Angiographic findings are listed in Table 2 . There was no difference in stent location between the NA and the non-NA groups. Compared with stents without NA, those with NA had significantly smaller minimum lumen diameter and greater diameter stenosis. Lesion length and reference lumen diameter were not different between the two groups.

A total of 252 stents (100 BMSs and 152 DESs) were analyzed. NA was detected in 121 stents (48.0%): 37 BMSs (37.0%) and 84 DESs (55.3%). Mean and maximum NITs were significantly greater in stents with NA than those without NA ( Table 2 ). OCT analysis of incidence and characteristics of NA according to tertiles of mean NIT is shown in Figure 2 and Table 3 . In the BMS group, the prevalence of lipid-laden intima was progressively greater across the tertiles, with a significant difference between the lowest and the highest tertiles ( Figure 2 ). The percentage of LRI and the prevalence of calcification were not different between the tertiles (p = 0.124 and p = 0.053, respectively). NV, in particular with intra-intima location, was more frequent in the mid and highest tertiles than that in the lowest tertile. The prevalence of TCFA-like intima, disrupted intima, and thrombus was not different among the tertiles ( Table 3 ). In the DES group, the prevalence of lipid-laden intima and the percentage of LRI were progressively greater across the tertiles ( Figure 2 ), whereas no significant difference was observed in the prevalence of calcification (p = 0.252). The prevalence of NV, particularly with intra-intima location, TCFA-like intima, disrupted intima, and thrombus, was different among the tertiles, with significantly higher values in the highest tertile ( Table 3 ). Inter- and intraobserver agreements were k = 0.90 and k = 0.94 for lipid-laden intima and k = 0.92 and k = 0.93 for in-stent calcification, respectively.

OCT analysis of incidence and characteristics of NA according to tertiles of mean NIT. (A) In the BMS group, stents in the highest tertile had greater prevalence of lipid-laden intima than stents in the lowest tertile. NV, in particular intra-intima neovascularization (II-NV), was more frequent inside the stents in the mid and highest tertiles compared with those in the lowest tertile. Percentage of LRI (%LRI), calcification, and peristrut neovascularization (PS-NV) were not different among the tertiles; (B) In the DES group, the prevalence of lipid-laden intima and %LRI were progressively greater across the tertiles. Stents in the highest tertile showed significantly greater prevalence of NV (particularly, II-NV) than those in the other tertiles. No difference in prevalence of calcification and PS-NV was detected. † Indicates significance after adjustment with Bonferroni correction (For interpretation of references to color in this figure legend, the reader is referred to the web version of this article.).

Table 3

Optical coherence tomography analysis of neointima characteristics according to tertiles of mean neointimal thickness (NIT)

Variable	Lowest Tertile	Mid Tertile	Highest Tertile	p
				Pearson Chi-Square	Lowest vs Mid Tertile	Mid vs Highest Tertile	Lowest vs Highest Tertile
BMSs	33	33	34
TCFA-like intima	3 (9)	6 (18)	3 (9)	0.410	0.473	0.444	0.697
Disrupted intima	1 (3)	3 (9)	4 (12)	0.403	0.606	0.967	0.371
Thrombus	0	2 (6)	5 (15)	0.060	0.473	0.449	0.068
DESs	51	51	50
TCFA-like intima	1 (2)	9 (18)	12 (24)	0.005	0.020	0.588	0.003 ∗
Disrupted intima	1 (2)	1 (2)	9 (18)	0.002	0.475	0.018	0.018
Thrombus	0	6 (12)	10 (20)	0.004	0.035	0.389	0.002 ∗

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Race/Ethnic Disparities in Risk Factor Control and Survival in the Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) Trial
2. One-Year Outcomes of Percutaneous Coronary Intervention With the 38-mm Resolute Zotarolimus-Eluting Stent
3. Effectiveness and Safety of Spironolactone for Systolic Heart Failure
4. Trauma in Patients With Continuous-Flow Left Ventricular Assist Devices

Stay updated, free articles. Join our Telegram channel

Join