The primary aim of the present study was to assess the gray scale intravascular ultrasound (IVUS) findings that might be associated with late drug-eluting stent restenosis. The study included 47 patients (54 lesions) who had undergone either baseline IVUS-guided stent implantation or IVUS-guided repeat stenting to treat in-stent restenosis and then had IVUS follow-up data for ≥1.5 years afterward without any intervening procedures. The left anterior descending artery was the culprit in 59% of cases, and 50% of the lesions were at bifurcation sites. Quantitative and qualitative IVUS analyses showed a decreased minimum lumen area at follow-up from 6.0 ± 1.8 to 3.8 ± 1.4 mm 2 (p <0.0001) that was mainly due to neointimal hyperplasia with chronic stent recoil (defined as a >15% decrease in minimum stent area) in only 2 lesions and stent fracture in only 5 lesions. Calcified neointima appeared in 12 lesions, mostly in the form of macrocalcification, and was associated with increased calcium both behind the stent and in the reference segment. In conclusion, late drug-eluting stent restenosis showed neointimal calcification in 20% of cases, and chronic stent recoil was rare.
In-stent neointimal hyperplasia (NIH) most commonly peaks 6 months after stent implantation and on intravascular ultrasound (IVUS) studies typically appears hypoechoic and concentric. However, recent pathologic, optical coherence tomographic, and angioscopic studies have suggested that atherosclerosis can also develop within stents, albeit >1 year after the appearance of typical NIH-related in-stent restenosis, and includes a spectrum of findings from lipidic to calcified neointima. Serial IVUS changes with late in-stent restenosis remain an interesting issue for evaluation. The primary objective of the present study was to assess the various IVUS findings that might be associated with late drug-eluting stent restenosis, including NIH or neointimal calcification, underexpansion or chronic stent recoil, stent fracture, late stent malapposition, muscle bridges, and persistent calcium progression.
Methods
We searched the IVUS database at Columbia University Medical Center from January 2005 to December 2010 to identify 47 patients (54 lesions) who either had IVUS-guided stent implantation with IVUS follow-up data available ≥1.5 years after implantation without an intervening intervention (42 lesions in 35 patients) or who had IVUS-guided repeat stenting to treat in-stent restenosis, also with IVUS follow-up data available ≥1.5 years later without an intervening intervention (12 lesions in 12 patients). All patients presented for assessment of late in-stent restenosis, defined as the presence of recurrent angina or objective evidence of myocardial ischemia with ≥50% luminal narrowing on angiographic follow-up. Clinical and procedural data were collected from the hospital records. Hypertension was defined as systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg, and/or treatment with medication. Although the definition of hyperlipidemia included previously documented hypercholesterolemia (serum total cholesterol ≥230 mg/dl or serum triglycerides ≥200 mg/dl) or treatment with lipid-lowering drugs.
Coronary angiograms before intervention (either before de novo stent implantation or before treatment of in-stent restenosis), immediately after intervention, and at >1.5 years of follow-up were reviewed to evaluate the lesion characteristics at baseline and the patterns of stent failure at the follow-up examination, including the Mehran classification and the presence of stent fracture.
All baseline and follow-up (>1.5-year) IVUS studies were performed after administration of 200 μg of intracoronary nitroglycerine using motorized pullback at 0.5 mm/s to include the stent and >5-mm-long segments proximal and distal to the stent. The IVUS catheters and consoles used during the study period included the 45-MHz Revolution imaging catheter interfaced with the In-Vision Gold or s5 Imaging Systems (Volcano, San Diego, California) or 40-MHz Atlantis or i-Cross IVUS imaging catheters interfaced with ClearView, Galaxy, or iLab imaging systems (Boston Scientific, Natick, Massachusetts). The same IVUS catheter and console were used at both baseline and follow-up, in all but 1 patient. Baseline IVUS imaging was performed as the last step after stent deployment; follow-up IVUS imaging was performed before any intervention, with the exception of 2 lesions in which the restenosis pattern was total occlusion, and a small (1.5-mm) balloon was inflated to achieve Thrombolysis In Myocardial Infarction flow 2 or 3 before IVUS-guided catheter insertion. The images were archived onto s-VHS tape, CD-ROM, or DVD for off-line analysis.
We used validated planimetry software (echoPlaque, INDEC Medical Systems, Santa Clara, California) for quantitative and qualitative analyses. Qualitative and quantitative analyses were performed according to the criteria of the American College of Cardiology Clinical Expert Consensus Document on Standards for Acquisition, Measurements, and Reporting of IVUS Studies. Quantitative analysis included measurement of the external elastic membrane, stent, and lumen cross-sectional areas (CSAs) within the stent and adjacent reference segments. The plaque and media CSAs were calculated as the external elastic membrane minus the stent CSA within the stented segment and the external elastic membrane minus the lumen CSA within the reference segments. The in-stent NIH CSA was calculated as the stent minus lumen CSA, and the percentage of NIH was calculated as the NIH divided by the stent CSA. Stent expansion was calculated as the minimum stent area compared to the average of the proximal and distal reference lumen CSAs.
Calcium was brighter than the adventitia, with acoustic shadowing of underlying tissue and was quantified by measuring the maximum calcium arc and localized to nonstented plaque, the plaque behind the stent, or the neointima. Neointimal calcification was classified as macrocalcification (hyperechoic material: density greater than that of the adventitia, with acoustic shadow and a calcium arc ≥10°); and microcalcification (hyperechoic material with acoustic shadow and a calcium arc <10°).
The baseline and follow-up IVUS images were analyzed side-by-side to identify the slices with the greatest decrease in stent CSA. To assess the restenosis process, the image slice with the smallest lumen area at the follow-up examination was identified and compared to the same image slice on the postintervention and preintervention studies using a fiduciary point or vascular or perivascular markings as reference points.
To define the threshold for chronic stent recoil, we randomly chose 200 stented lesions with both baseline and follow-up IVUS imaging studies from patients who were not a part of the present study to serve as a reference cohort. In these patients, the change in the minimum stent CSA measured 2.0 ± 9.7%. In a study by Hong et al, the maximum change in minimum stent CSA was observed in Cross-flex stents and measured −4.0 ± 6.0%. Therefore, because the mean minus 2 SDs was −17.4% in our reference cohort and −16% in the Cross-flex stents, we defined chronic stent recoil as a ≥15% decrease in stent CSA.
Stent malapposition was defined as a separation of multiple stent struts from the luminal border with the presence of blood speckle behind the struts. Malapposition was categorized as persistent (visible at both baseline and follow-up), resolved (only visible at baseline), and acquired late (only visible at follow-up).
Stent fracture was defined as the absence of stent struts and was divided into partial stent fracture (absence of stent struts for 1/3 of the stent circumference) or complete fracture (absence of stent struts for the entire circumference of the stent).
A muscle bridge was defined as a segment of epicardial coronary artery having systolic compression and an echocardiographically lucent muscle band surrounding the artery.
Statistical analysis was performed using StatView, version 5.0 (SAS Institute, Cary, North Carolina). Categorical variables are expressed as counts and percentages. Continuous variables are presented as the mean ± SD. Continuous variables were compared using the Student t test if normally distributed. Nonparametric tests (Wilcoxon signed rank test) were used for statistical analysis of non-normally distributed variables. p Values <0.05 were considered significant.
Results
The mean follow-up period between the baseline and follow-up IVUS studies was 2.8 years (minimum 1.6 and maximum 4.5). At stent implantation, the mean patient age was 66.6 years, 74% were men, 43% had diabetes mellitus, 17% had impaired renal function (estimated creatinine clearance <60 ml/min), 76% presented with acute coronary syndrome, 21% had previous myocardial infarction, and 53% had undergone previous percutaneous coronary intervention ( Table 1 ).
| Characteristic | Value |
|---|---|
| Age (yrs) | 66.6 ± 10.9 |
| Men | 35 (74%) |
| Acute coronary syndrome presentation of in-stent restenosis | 36 (76) |
| Diabetes mellitus | 20 (43%) |
| Hypertension | 34 (72%) |
| Hyperlipidemia | 33 (70%) |
| Smoker | 5 (11%) |
| Family history of coronary artery disease | 8 (17%) |
| Previous myocardial infarction | 10 (21%) |
| Previous percutaneous coronary intervention | 25 (53%) |
| Previous coronary bypass | 7 (15%) |
| Estimated creatinine clearance (<60 ml/min) | 8 (17%) |
| Low-density lipoprotein cholesterol at baseline (mg/dl) | 76 ± 37 |
| High-density lipoprotein cholesterol at baseline (mg/dl) | 38 ± 10 |
Lesions involved the left anterior descending artery in 59% of cases, were bifurcations in approximately 50%, and were type B2/C (American Heart Association and American College of Cardiology classification) in 70%. A sirolimus-eluting stent (Cordis, Bridgewater, New Jersey) was deployed in 80% of the lesions, and a paclitaxel-eluting stent (Boston Scientific) was deployed in 15% of the lesions ( Table 2 ). At the follow-up examination, 63% had focal in-stent restenosis, 20% had diffuse in-stent restenosis, 13% had proliferative in-stent restenosis, and 4% had total occlusion. In addition, 7% had the angiographic appearance of stent fracture.
| Characteristic | Value |
|---|---|
| Coronary artery | |
| Left anterior descending | 32 (59%) |
| Left circumflex | 10 (18%) |
| Right | 11 (21%) |
| Left main | 1 (2%) |
| In-stent restenosis lesion | 11 (20%) |
| Bifurcation lesion | 27 (50%) |
| Stent type | |
| Sirolimus-eluting stent | 43 (80%) |
| Paclitaxel-eluting stent | 8 (15%) |
| Other | 3 (5%) |
| Stent number per lesion | 1.2 ± 0.5 |
| Total stent length (mm) | 18.6 ± 7.1 |
The minimum lumen CSA at follow-up was significantly smaller than at baseline (3.8 ± 1.4 vs 6.0 ± 1.8 mm 2 ; p <0.0001) and was mostly the result of significant NIH (maximum percentage of NIH at follow-up was 34.1 ± 16.4%; Table 3 ). The minimum lumen CSA site was located at the maximum NIH site in 41 lesions (76%) and at the minimum stent CSA site in 35 lesions (65%). The maximum NIH site and minimum stent CSA site were co-located in 58%.
| Variable | Baseline (n = 54) | Follow-up (n = 54) | p Value |
|---|---|---|---|
| Reference segment | |||
| Lumen cross-sectional area (mm 2 ) | 7.3 ± 2.1 | 7.4 ± 2.1 | 0.0003 |
| External elastic membrane cross-sectional area (mm 2 ) | 13.6 ± 4.2 | 14.8 ± 4.6 | 0.002 |
| Stented segment | |||
| Minimum lumen cross-sectional area site | |||
| External elastic membrane cross-sectional area (mm 2 ) | 13.9 ± 4.1 | 15.1 ± 4.9 | 0.003 |
| Lumen cross-sectional area (mm 2 ) | 6.0 ± 1.8 | 3.8 ± 1.4 | <0.0001 |
| Plaque burden (%) | 56.1 ± 7.5 | 59.5 ± 11.0 | 0.005 |
| Minimum stent cross-sectional area (mm 2 ) | 6.0 ± 1.8 | 5.9 ± 1.9 | 0.034 |
| Stent expansion (%) | 83.4 ± 11.4 | 79.1 ± 11.5 | 0.0004 |
| Maximum neointimal hyperplasia (%) | 34.1 ± 16.4 | ||
| Location of maximum neointimal hyperplasia site (n) | |||
| Proximal stent edge | 20 (37%) | ||
| Stent body | 16 (30%) | ||
| Distal stent edge | 18 (33%) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
