Article Text
PDF
Abstract
Background Zotarolimus-eluting stents (ZES) have a higher rate of neointimal coverage than the first-generation drug-eluting stents on optical coherence tomography (OCT).
Objective To determine whether neointimal coverage of stent struts detected by OCT can be used as a surrogate for endothelial function after ZES implantation.
Design Cross-sectional observational study.
Setting Three months after ZES implantation.
Patients and methods OCT was performed in 20 patients with a ZES at 3 months after stent implantation to evaluate strut coverage. Endothelium-dependent coronary vasomotion was estimated by infusing incremental doses of acetylcholine into the coronary ostium. The vascular response was measured in the 10 mm segments proximal and distal to the stent.
Results Of 20 ZES, 15 (75%) were covered completely with neointima, but the remaining 5 ZES had exposed struts. The high-dose acetylcholine infusion produced significant vasoconstriction in the proximal (−9.8±10.1%) and the distal stent segment (−29.7±22.7%). However, the degree of vasoconstriction to acetylcholine varied between individuals (from −0.6% to −77%). Although no relationship was observed between coronary vasomotor response (percentage change in diameter after acetylcholine administration) and average neointimal thickness, the number of cross-sections with uncovered struts showed an inverse correlation with coronary vasomotor response in proximal and distal stent segments (r=−0.57, p=0.007 and r=−0.83, p<0.001, respectively).
Conclusions The existence of exposed struts was associated with abnormal vasoconstriction to acetylcholine at 3 months after ZES implantation. The findings suggest that complete neointimal coverage of stent struts assessed by OCT could be used as a surrogate for vasomotion impairment at 3 months after ZES implantation.
- Drug-eluting stents
- acetylcholine
- optical coherence tomography
- intravascular ultrasound
- coronary stenting
- coronary vasomotion
Statistics from Altmetric.com
- Drug-eluting stents
- acetylcholine
- optical coherence tomography
- intravascular ultrasound
- coronary stenting
- coronary vasomotion
Introduction
First-generation drug-eluting stents remarkably reduced the rate of in-stent restenosis and subsequent target lesion revascularisation as compared with bare-metal stents in randomised clinical trials.1 2 However, excessive inhibition of neointimal formation may cause delayed vascular healing with incomplete endothelialisation, which has been associated with an increase risk of late and very late stent thrombosis.3 4 Observational data using optical coherence tomography (OCT), which has a higher resolution than any other available imaging modality and can assess tissue coverage of drug-eluting stents, demonstrated that stent struts of the first-generation drug-eluting stents still remain uncovered 2 years after deployment.5 Finn et al showed that the first-generation drug-eluting stents impaired the normal healing processes of the injured arterial wall, even over a period of 40 months after implantation, and the heterogeneity of healing in the stents was associated with late stent thrombosis.6
Zotarolimus-eluting stents (ZES) (Endeavor, Medtronic CardioVascular, Santa Rosa, California, USA), a cobalt chromium-based thin-strut stent with a phosphorylcholine biocompatible polymer and shorter drug-elution time within 2 weeks, have been developed as second-generation drug-eluting stents and reported to promote rapid and uniform healing of the endothelium.7 Recent intravascular ultrasound (IVUS) and OCT studies reported that ZES were associated with a higher rate of neointimal coverage at 8–9 months than the first-generation drug-eluting stents.8 9 Furthermore, a recent OCT study indicated that most of the stent struts were covered with neointima even 3 months after ZES implantation.10 Similarly, previous studies have demonstrated that coronary vasomotor response to acetylcholine at 6 months is preserved in ZES compared with the first-generation drug-eluting stents.11 However, it is not clear whether neointimal strut coverage detected by OCT can be used as surrogates for vessel healing after drug-eluting stent implantation. Therefore, this study was undertaken to evaluate the relationship between neointimal strut coverage detected by OCT and coronary vasomotion in response to acetylcholine at 3 months after ZES implantation.
Methods
Study population and study protocol
Between May and October 2009, a prospective but non-consecutive series of 20 patients who were scheduled for coronary intervention for stable angina pectoris were enrolled in this study. Inclusion criteria were presence of typical stable effort angina or positive stress test, and indication for percutaneous coronary intervention of a single de novo lesion >50% diameter stenosis in a native coronary artery. The exclusion criteria were acute coronary syndrome, ≥50% coronary vasoconstriction in response to acetylcholine infusion during the pre-intervention test, a history of vasospasm, renal insufficiency with baseline creatinine ≥2.0 mg/dl, lesions located at the ostium of each artery (<10 mm from the ostium), or diffuse lesion. The ethics committee at Hyogo College of Medicine approved the protocols, and written informed consent was obtained before all procedures from all patients.
Study procedure
After enrolment, all patients were treated with a single ZES (Endeavor, Medtronic Vascular Inc, Santa Rosa, California) at the baseline procedure. After intervention, all patients received oral aspirin (100 mg/day) and clopidogrel (75 mg/day) for at least 3 months. Three months after ZES implantation, neointimal strut coverage was assessed with both IVUS and OCT, and vessel healing was estimated by vasomotor responses to an intracoronary infusion of acetylcholine and isosorbide dinitrate.
Evaluation of vasomotor response
Drugs with potential effects on vasomotor response, including nitrates, calcium channel blockers and β blockers were discontinued at least 72 h before the follow-up angiography. Endothelium-dependent vasomotion was estimated by measuring the coronary vasoreactivity in response to acetylcholine infusion into the coronary ostium of each coronary artery at the pre-intervention and 3-month follow-up. Before wiring the target lesion, vasomotor reactivity was estimated by infusing incremental doses of acetylcholine, 10 μg, 20 μg, 50 μg and 100 μg over 1 min into the coronary ostium through a guiding catheter.11 A 3 min interval was allowed between each concentration. Subsequently, endothelium-independent vasomotion was assessed after an intracoronary bolus of isosorbide dinitrate (3–5 mg).
Angiographic analysis
Quantitative coronary angiography was performed from two orthogonal projections that well delineated the coronary artery of interest, without overlapping of side branches and with less foreshortening, using a computer-assisted, automated edge-detection algorithm (CMS, MEDIS). Two segments in study vessel were chosen for analysis, 5 mm proximal and distal to the site of stenting. Changes in coronary diameter in response to acetylcholine and isosorbide dinitrate coronary infusion were expressed as percentage changes versus baseline angiograms.
IVUS imaging and analysis
IVUS examinations were attempted for all patients after intracoronary administration of 5 mg isosorbide dinitrate using a commercially available system (Atlantis Pro, Boston Scientific, Natick, Massachusetts, USA). The IVUS catheter was advanced >10 mm beyond the stent and an imaging run (using automated transducer pullback at 0.5 mm/s) was performed to a point >10 mm proximal to the stent. The IVUS images were analysed with planimetry software (Echoplaque 3, INDEC Systems, Mount View, California, USA) by an independent single observer who was blinded to the clinical presentations. IVUS examinations were carried out both after percutaneous coronary intervention and 3 months later, and qualitative and quantitative IVUS measurements were made according to the criteria of the American College of Cardiology Clinical Expert Consensus document on IVUS.12 Neointimal volume was calculated as stent volume minus lumen volume, and neointimal volume index was the value of neointimal volume divided by stent length. The percentage neointimal coverage was defined as the measurement of circumferential stent length covered with neointima divided by stent perimeter at every 1 mm cross-sectional image through the stented segment.8
OCT imaging and analysis
All OCT examinations were performed after intracoronary administration of 5 mg isosorbide dinitrate. A time domain M2 OCT system (M2 Cardiology Imaging System, LightLab Imaging, Inc, Westford, Massachusetts, USA) combined with a 0.016 inch wire-type imaging wire (ImagingWire, LightLab Imaging Inc) was used in this study. During image acquisition, an over-the-wire occlusion balloon catheter (Helios, Avantec Vascular Corp, Sunnyvale, California, USA) was inflated to within 0.4–0.6 atm while lactated Ringer's solution was infused from the distal tip of the occlusion balloon catheter at 0.5 ml/s to flush blood from the imaging field. An imaging run was performed from the distal to proximal area of the stented segment using automated transducer pullback at 1.0 mm/s. OCT images were analysed by two independent observers, who were blinded to clinical presentations and lesion characteristics, using proprietary offline software. The ZES strut condition was analysed every 1 mm along the stented segment. For neointimal hyperplasia (NIH) thickness, the distance between the centre reflection of the strut and the vessel wall was measured. The stent and lumen areas were measured by a manual trace and the percentage of NIH was calculated as (stent − lumen area) divided by stent area multiplied by 100.9 13 A cross-section with uncovered struts was defined if one or more stent strut was uncovered on cross-section.6 9 The number of cross-sections with uncovered struts was counted for each ZES. The contiguous strut uncovered index was calculated as contiguous frames with uncovered struts divided by the actual length of the ZES. Stent incomplete apposition was defined as a distance between the centre reflection of the strut and the vessel wall >110 μm.14
Statistics
Continuous variables were reported as mean±SD and categorical variables were reported as frequency percentages. Comparisons were carried out by analysis of variance for repeated measurements. When the test was significant, Bonferroni tests for paired comparisons were used. Analyses for the associations were performed using the multivariate linear regression technique. Multiple linear regression analysis was used to determine independent predictors of vasomotion impairment. Candidate variables included age, gender, diabetes, hypertension, current smoker, statin used, ACE inhibitor/angiotensin-receptor blocker used, β blocker used, and the number of cross-sections with uncovered struts. A p value of <0.05 was considered statistically significant.
Results
Baseline patient and lesion characteristics
Baseline patient and lesion characteristics are shown in table 1. Locations of target lesions were left anterior descending (n=11), left circumflex (n=3) and right coronary arteries (n=6). The average interval between the baseline and follow-up studies was 97±12 days.
Baseline patient and lesion characteristics
Angiographic and IVUS analysis
Angiographic and IVUS data are shown in table 2. Late loss at the 3-month follow-up was 0.3±0.6 mm. At the 3-month follow-up, there were no lesions showing angiographic in-stent restenosis and no patients required target lesion revascularisation. Although an incomplete stent apposition was detected by IVUS in two (10%) patients at post percutaneous coronary intervention, these two incomplete appositions were completely resolved at 3 months. Late acquired incomplete apposition was not detected at 3 months after ZES implantation by IVUS. Approximately 24% of all stent surfaces were covered with neointima in IVUS analysis. No thrombus was detected during serial IVUS examinations.
Angiographic, IVUS, and OCT analysis
OCT analysis
Clear OCT images of all 20 ZES were obtained without any serious complication at 3 months. In total, 5170 of 5205 struts (99.3%) were completely covered with neointima. The average neointimal thickness was 217 μm, which results in approximately 17% of neointima over stent struts <100 μm. OCT data are shown in table 2. Stent edge dissection, intracoronary thrombus within stents, or incomplete stent apposition was not detected in any patients at the 3-month follow-up. Of the 20 stents evaluated, 15 (75%) were completely covered with neointima. The remaining five partially covered stents had only 3 (0.9%), 5 (2.5%), 6 (2.1%), 8 (3.6%) and 13 (4.3%) uncovered struts.
Coronary vasomotor response 3 months after ZES implantation
The coronary vasomotor response to incremental doses of acetylcholine and isosorbide dinitrate is shown in figure 1. Although intracoronary acetylcholine infusion at 10 μg and 20 μg produced no coronary vasoconstriction (−1.4±2.3%, p=0.4 and −2.7±3.0%, p=0.1; vs baseline, respectively), 50 μg and 100 μg acetylcholine infusion produced significant vasoconstriction (−6.5±6.7%, p<0.01 and −9.8±10.1%, p<0.01; vs baseline, respectively) in the proximal stent segments. Similarly, in the distal stent segment, acetylcholine infusion at 50 μg and 100 μg produced significant vasoconstriction (−21.3±18.0%, p<0.01 and −29.7±22.7%, p<0.01; vs baseline, respectively). As shown in figure 1, the coronary vasomotor response to acetylcholine varied between individuals. However, intracoronary isosorbide dinitrate induced a similar grade of vasodilatation in all patients at both the proximal and distal stent segments.
Changes in vasomotor reaction 3 months after zotarolimus-eluting stent (ZES) implantation. Mean coronary artery segment luminal diameter change at 3 months' follow-up after intracoronary infusion of acetylcholine with incremental doses of 10, 20, 50 and 100 μg and isosorbide dinitrate (ISDN) for individual patients. Thick line is mean diameter.
Association between stent strut coverage and coronary vasomotor response
There was no significant relationship between coronary vasomotor response to incremental doses of acetylcholine and percentage neointimal covered area over stent surfaces measured by IVUS. Also, there were no relationships between coronary vasomotor response in proximal stent segments at any dose of acetylcholine and average NIH area, percentage NIH area, average NIH thickness and heterogeneity score measured by OCT. Similarly, no relationship was seen between coronary vasomotor response in distal stent segment at any dose of acetylcholine and average NIH area, percentage NIH area, average NIH thickness and heterogeneity score measured by OCT. However, the number of cross-sections with uncovered struts showed a significant inverse correlation with coronary vasomotor response in proximal stent segments at both high- (100 μg) and low-dose (10 μg) acetylcholine infusion (r=−0.51, p=0.02 and r=−0.57, p=0.007, respectively) (figure 2A). Also, there was a linear correlation between the number of cross-sections with uncovered struts and coronary vasomotor response in distal stent segments at both high- (100 μg) and low-dose (10 μg) of acetylcholine infusion (r=−0.49, p=0.03 and r=−0.83, p<0.001, respectively) (figure 2B). Similarly, an inverse linear correlation was observed between the contiguous strut uncovered index and coronary vasomotor response both in proximal and distal stent segments at low-dose acetylcholine infusion (r=−0.61, p<0.001 and r=−0.41, p=0.03, respectively).
Association between the number of cross-sections with uncovered strut and vasomotor response to acetylcholine infusion. (A) The number of cross-sections with uncovered strut inversely correlated with coronary vasomotor response to incremental doses of acetylcholine in the proximal stent segment (r=−0.51, p=0.02 for high dose and r=−0.57, p=0.007 for low dose acetylcholine, respectively). (B) In the distal stent segment, the number of cross-sections with uncovered strut was associated with coronary vasomotor response to incremental doses of acetylcholine (r=−0.49, p=0.03 for high dose and r=−0.83, p<0.001 for low dose, respectively).
Although coronary vasoconstriction to high-dose acetylcholine was more prominent in patients with diabetes mellitus than in patients without diabetes mellitus at both the proximal and distal stent segments (−13.8±10.9% vs −3.9±4.9%, p=0.03 at the proximal segments, and −39.9±21.8% vs −14.5±14.3%, p<0.01 at the distal segments), no correlations were identified between vasomotion impairment and other coronary risk factors and/or any drug treatments. By multiple linear regression analysis, the number of cross-sections with uncovered struts was the only independent predictor of vasomotion impairment at both the proximal and distal stent segments (p=0.01 and p<0.01, respectively).
Discussion
This is the first study to evaluate the association between stent strut coverage and coronary vasomotor response to intracoronary acetylcholine infusion at 3 months after ZES implantation. This study demonstrated that the degree of vasoconstriction in response to acetylcholine was associated with the number of uncovered struts detected by OCT at 3 months after ZES implantation.
The mechanisms of stent thrombosis after drug-eluting stent implantation are multifactorial and include patient, lesion and procedural factors, as well as compliance with, and response to, antiplatelet therapy.6 15 16 However, all currently available drug-eluting stents may be susceptible to late stent thrombosis because of the delayed arterial healing and impaired endothelialisation that accompany the drugs used to inhibit NIH. Accordingly, a previous pathological report demonstrated an association between the lack of neointimal coverage in stent struts and thrombus formation.17 This study showed a discrepancy in the rate of neointimal coverage 3 months after ZES implantation between IVUS and OCT (24.5% in IVUS versus 99.3% in OCT), although the methods of measurement was somewhat different. Kim et al demonstrated that neointimal coverage of stent struts was almost complete (99.7%) in ZES at 9 months.9 In accordance with our study, a previous IVUS and OCT study reported that the neointimal coverage at 3 months after ZES deployment was 25.8% in IVUS and 99.9% in OCT.10 Our study confirmed the finding that a 3-month period would be sufficient for complete neointimal coverage of stent struts for ZES. However, the question remains as to whether complete neointimal coverage of stent struts can be used as a surrogate for vessel healing after drug-eluting stent implantation.
A histopathological study indicated that re-endothelialisation was complete 3 months after bare-metal stent implantation.18 Shin et al showed that endothelium-dependent vasomotion after intracoronary infusion of acetylcholine was preserved 6 months after bare-metal stent implantation, although drug-eluting stent implantation was associated with endothelial dysfunction at 6 months, predominantly in the distal segment of the treated vessel.19 Acetylcholine evokes a nitric oxide (NO)-mediated vasodilatory response in healthy arteries via muscarinic endothelial membrane receptors, but this effect is blunted, and paradoxical vasoconstriction is observed with endothelial dysfunction.20 Therefore, quantitative coronary angiography after intracoronary acetylcholine infusion is one of the most used invasive methods for assessing coronary endothelial function after drug-eluting stent implantation.8 11 Previous studies showed abnormal coronary vasoconstrictive responses to acetylcholine after the first-generation drug-eluting stent implanration.21 22 It has been reported that both sirolimus and paclitaxel inhibit smooth muscle cell proliferation and endothelial regeneration in vitro.23 24 Also, the synthetic polymer may be an important trigger of local coronary inflammation. Coronary inflammation is responsible for delayed re-endothelialisation of stent and vessel walls, which eventually results in a delayed vascular healing response.
Another potential explanation for delayed vessel healing may include acute or late hypersensitivity reaction to the polymer and/or drug. In contrast, it has been reported that ZES, second-generation drug-eluting stents, are associated with fewer coronary vasoconstrictive responses to acetylcholine than first-generation drug-eluting stents.11 21 Although the mechanism underlying the early restoration of endothelial function after ZES implantation has not been elucidated, rapid elution of the drug may be one of the potential mechanisms. The majority (approximately 95%) of zotarolimus is released from the stent by 14 days.25 This rapid elution kinetics of ZES may reduce the local toxicity to the endothelium.26
Another possible explanation may include the unique characteristics of the phosphorylcholine polymer. An animal study reported that a phosphorylcholine polymer-coated stent had excellent biocompatibility as shown by mild inflammatory reactions, near-complete endothelialisation and near absence of thrombus.27
These findings suggest that although a larger amount of NIH leads to a need for target lesion revascularisation, the coverage of ZES struts with NIH may offer a protective advantage for early vessel healing. However, in this study, there was an individual variation of vasomotor responses to an intracoronary infusion of acetylcholine and the degree of neointimal stent strut coverage at 3 months after ZES implantation. This finding suggests that the rate of vessel healing varies from patient to patient at 3 months after ZES implantation. Our data demonstrated that the number of cross-sections with uncovered struts had an inverse correlation with coronary vasomotor response to acetylcholine infusion. Although most patients had complete strut neointimal coverage and minor vasoconstriction in response to acetylcholine, some patients had uncovered stent struts with significant vasoconstriction at 3 months after ZES implantation. In accordance with our data, a histopathological study demonstrated that stent strut coverage correlates highly with endothelialisation.6 This finding suggests that neointimal coverage of stent struts might be used as a surrogate for vascular healing at 3 months after ZES implantation.
Limitations
This study had several limitations. First, the number of enrolled patients was relatively small. However, over 5000 stent struts were evaluated for neointimal coverage, and vasomotor response was estimated by infusing four different doses of acetylcholine. Second, OCT examinations were not performed immediately after stent implantation. However, no incomplete apposition was identified at follow-up, suggesting that no late acquired incomplete apposition was observed in this study. Third, because we assessed vessel healing 3 months after ZES implantation, our results refer to this specific time point. Finally, the clinical relevance, such as preferable duration of dual antiplatelet treatment, remains unknown. Local coronary vasoconstriction is only an indirect sign of endothelial dysfunction, probably caused by delayed re-endothelialisation in the segments proximal and distal to the stent.
Conclusions
The existence of exposed struts was associated with abnormal vasoconstriction in response to acetylcholine infusion. Our findings suggest that complete neointimal coverage of stent struts assessed by OCT could be used as a surrogate for vasomotion impairment at 3 months after ZES implantation.
Acknowledgments
The authors thank the staff in the cardiac catheterisation laboratory in Hyogo College of Medicine for their excellent assistance during the study.
References
Footnotes
See Editorial, p 953
Patient consent Obtained.
Ethics approval This study was conducted with the approval of the Hyogo College of Medicine.
Provenance and peer review Not commissioned; externally peer reviewed.