Methods

The Massachusetts General Hospital OCT registry is a multicenter registry of patients undergoing OCT of the coronary arteries and includes 20 sites across 6 countries. Any patient who underwent OCT was eligible for the registry, regardless of symptoms. For the present study, we screened 989 patients in the Massachusetts General Hospital OCT registry. From those patients, we selected 138 stents from 124 patients that had ≥3 consecutive cross-sections at 1-mm intervals with a mean neointimal thickness >100 μm, as identified by OCT.

The stents were classified into 3 groups according to the duration since stent implantation: an early phase (<9 months, n = 52), an intermediate phase (≥9 to <48 months, n = 61), and a delayed phase (≥48 months, n = 25). The optical coherence tomographic findings were compared between the BMSs and DESs in each phase. Symptomatic restenosis was defined as a restenotic stent that was the culprit lesion of an acute coronary syndrome, including ST-segment elevation myocardial infarction, non–ST-segment elevation myocardial infarction, or unstable angina pectoris, or a target lesion of percutaneous coronary intervention for stable angina pectoris with Canadian Cardiovascular Society class II-IV symptoms. Hypertension was entered on the electronic case report form when the following definitions were met: a history of hypertension, the use of an antihypertensive drug, systolic blood pressure of ≥140 mm Hg, or diastolic pressure of ≥90 mm Hg. Hyperlipidemia was entered on the form if there was a history of hyperlipidemia, the presence of lipid-lowering therapy, or newly diagnosed hyperlipidemia according to institutional guidelines.

Coronary angiograms during follow-up were analyzed using off-line quantitative coronary angiography (Quantcor QCA, version 5.0, Pie Medical Imaging BV, Maastricht, The Netherlands). The reference diameter, minimum lumen diameter, diameter stenosis, and lesion length were measured.

Both the time-domain OCT system (M2/M3 Cardiology Imaging System, LightLab Imaging, Westford, Massachusetts) and the frequency-domain OCT system (C7-XR OCT Intravascular Imaging System, St. Jude Medical, St. Paul, Minnesota) were used in the present study. The technique of intracoronary OCT has been previously described. In brief, with the M2/M3 system, an occlusion balloon (Helios, LightLab Imaging) was advanced proximal to the lesion and inflated ≤0.4 to 0.6 atm during image acquisition. The imaging wire was automatically pulled back from distally to proximally at 1.0 to 3.0 mm/s, and saline was continuously infused from the tip of the occlusion balloon. With the C7 system, a 2.7F optical coherence tomographic imaging catheter (Dragonfly, LightLab Imaging) was advanced distal to the lesion, and automatic pullback was started as soon as the blood was cleared. All images were digitally stored, de-identified, and submitted to Massachusetts General Hospital (Boston, Massachusetts) for analysis.

Cross-sectional optical coherence tomographic images of good quality were analyzed with a 1.0-mm interval for qualitative and quantitative evaluation. Qualitative analysis included the presence of lipid, macrophage, and neovascularization. Lipid was defined as a diffusely bordered signal-poor region with rapid signal attenuation. Lipid-laden intima was defined as a neointima with lipid ( Figure 1 ). For lipid-laden intima, the lipid arc and fibrous cap thickness were measured. Lipid-rich plaque (LRP) was defined as a plaque with a lipid arc >90°, and the %LRP was calculated as follows: %LRP = number of cross-sections with LRP/total number of analyzed cross-sections of plaque × 100. When the thinnest fibrous cap thickness was ≤65 μm in LRP, the plaque was defined as thin-cap fibroatheroma. Macrophage infiltration was defined as increased signal intensity accompanied by heterogeneous back shadows within the fibrous cap, as previously reported ( Figure 1 ). Neovascularization was defined as small vesicular or tubular structures with a diameter of 50 to 300 μm. The location of neovascularization was divided into intraintima or peristrut, as previously described ( Figure 1 ). The presence of disrupted neointima and thrombus was also recorded.

Tissue characteristics of neointima. (A) Lipid-laden neointima showing diffusely bordered signal-poor region with overlying signal-rich homogeneous band (white arrows). (B) Macrophage infiltration with high signal accumulation accompanied by heterogeneous back shadows (arrowheads). (C) Neovascularization showing small vesicular or tubular structures with 50- to 300-μm diameter (white arrow) differentiated from branch vessels.

A quantitative assessment was performed using the OCT off-line analysis software (LightLab Imaging). Tissue property analysis software was used for quantitative assessment ( Figure 2 ). The analysis was performed only for cross-sections with a mean neointima thickness >100 μm. Images with insufficient blood clearance were excluded. The stent and lumen areas were traced, and the thickest neointima was automatically indicated. The region of interest was placed at a quadrant section of neointima that was centered at the portion with the maximum neointimal thickness. Previously reported tissue property parameters, including signal attenuation, backscatter, and normalized standard deviation (NSD), were automatically calculated by the software. In brief, signal attenuation represents the degree of attenuation of the signal according to the depth of the tissue, backscatter represents the signal strength of the backscattering, and NSD represents the signal variability of the backscattering within the sample. All optical coherence tomographic images were analyzed by 2 independent investigators (T.Y. and K.K.) who were unaware of the patients’ information. When discordance was present between the readers, a consensus reading was obtained from a third independent investigator (S.K.). The inter- and intraobserver reliability for tissue property analyses were assessed using a random subset of 50 patients from all 3 groups.

Quantitative tissue property analysis. (A) Thickest portion of neointima that is semiautomatically calculated and indicated by software (yellow arrow). (B) Region of interest positioned toward thickest portion of neointima and traced with the arc of 90° and the depth from the lumen to stent struts (green area).

Categorical data are expressed as counts and percentages and were compared with the chi-square test or Fisher’s exact test, depending on the data type. Continuous measurements are expressed as the mean ± SD and were analyzed with the Student t test and analysis of variance. The %LRP comparisons between the 2 stent types at each phase and among the 3 phases within the same stent type were performed using analysis of variance and Duncan’s multiple range post hoc test using the natural log-transformed values, because the raw data were not normally distributed. We also performed the nonparametric tests (i.e., Mann-Whitney U test and Kruskal-Wallis test), and the resulting statistically significant findings were consistent with those from the log-transformed analysis of variance. We chose to report the log-transformed analysis of variance results, because they allowed us to then proceed to the post hoc test. The inter- and intraobserver reliabilities were estimated using the κ coefficient for binary outcomes and the intraclass correlation coefficient for continuous measurements. All statistical analysis was performed with SAS, version 9.1.3 (SAS Institute, Cary, NC). A p value <0.05 was considered statistically significant.

Results

Of all 124 patients, 1 had both BMSs and DESs (1 BMS and 2 DESs). The patient characteristics of the remaining 123 patients are listed in Table 1 . The 2 groups had a similar prevalence of risk factors. No significant difference was seen in the mean follow-up time from stent implantation. Most of the patients presented with stable angina, and no significant difference was present in the prevalence of acute coronary syndrome between the 2 groups during the follow-up period.

The angiographic findings during follow-up are summarized in Table 2 . No significant difference was seen in the location of the stents. In the qualitative comparative analysis, the reference diameter and minimum lumen diameter tended to be smaller in the BMSs than in the DESs. The restenotic lesion length was significantly longer in the BMSs. No significant difference was seen in diameter stenosis between the 2 groups.

Of all 2,965 cross-sections of optical coherence tomographic images in 138 stents, 124 cross-sections (4.2%) were excluded from the analysis because of suboptimal image quality, and 2,841 cross-sections were analyzed. Quantitative tissue property analysis was performed in 1,464 cross-sections with a mean neointimal thickness >100 μm. The estimated inter- and intraobserver κ coefficient for the presence of lipid-laden plaque was 0.90 and 0.93, respectively, and the intraclass correlation coefficient for attenuation was 0.94 and 0.95, respectively.

The optical coherence tomographic findings are summarized in Table 3 . Of all 138 stents (56 BMSs and 82 DESs), 52 stents were in the early phase, 61 in the intermediate phase, and 25 in the delayed phase. In the early phase, a greater frequency of lipid-laden neointima and significantly greater %LRP were found in the DESs than in the BMSs. Neovascularization, especially in the peristrut area, was significantly more frequent in the BMSs than in DESs. No significant difference was seen in the other categorical values. The quantitative tissue property analysis did not show any significant differences between the BMSs and DESs. In the intermediate phase, the prevalence of lipid-laden neointima and the %LRP remained significantly greater in the DESs compared to the BMSs, but no significant difference was observed in macrophage infiltration, thin-cap fibroatheroma, disruption, or thrombus formation between the 2 groups. The tissue property analysis showed greater attenuation and NSD in the DESs, a finding consistent with the changes in neointima tissue characteristics. In the delayed phase, most stents showed lipid-laden neointima in both BMSs and DESs, and the %LRP tended to be greater in the BMSs. In the tissue property analysis, the signal attenuation was significantly greater in the BMSs than in the DESs.

The incidence of lipid-laden neointima and %LRP in the 3 phases are depicted in Figure 3 , and the interperiod comparison is summarized in Table 4 . In the BMS ( Table 4 ), the incidence of lipid-laden intima, %LRP, and tissue property values did not show a significant difference between the early and intermediate phases. However, the incidence of lipid-laden neointima, %LRP, attenuation, and NSD sharply increased between the intermediate and delayed phases and also became significantly greater in the delayed phase compared to the early and intermediate phases. In DESs ( Table 4 ), the incidence of lipid-laden intima and %LRP showed constant trends toward an increase over time. Neovascularization was significantly more frequent in the delayed phase compared to the early and intermediate phases. However, no significant difference was observed among the 3 phases in the BMSs.

Comparison of different parameters of neoatherosclerosis between the 2 types of stents over time. (A) Incidence of lipid-laden plaque and (B) %LRP. *Significant difference using Duncan’s multiple range post hoc test at a 5% significance level. E = early phase (<9 months); I = intermediate phase (≥9 to <48 months); D = delayed phase (>48 months).

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. Outcomes of Preoperative Angiotensin-Converting Enzyme Inhibitor Therapy in Patients Undergoing Isolated Coronary Artery Bypass Grafting
2. Value of Three-Dimensional Speckle-Tracking Longitudinal Strain for Predicting Improvement of Left Ventricular Function After Acute Myocardial Infarction
3. Usefulness of Bidimensional Strain Imaging for Predicting Outcome in Asymptomatic Patients Aged ≤ 16 Years With Isolated Moderate to Severe Aortic Regurgitation
4. Insulin Sensitivity, but Not Hyperglycemia, May Be Superior as a Prognostic Marker and Therapeutic Target for Diabetes in Heart Failure

Stay updated, free articles. Join our Telegram channel

Join