Highlights

  • LCBI value within the culprit plaque in the FNR group was significantly higher than in the no-FNR group

  • Multiple logistic regression revealed that T-LCBI/max-LCBI4mm ratio within the culprit plaque and attenuation grade were independent predictors for FNR

  • In future, the proper length of max–LCBI to evaluate vulnerable plaque should evaluated using next-generation NIRS-IVUS

Introduction

Percutaneous coronary intervention (PCI) may be complicated by peri-procedural myocardial infarction and reduced antegrade flow (“no-reflow”), which has been shown to be associated with worse short- and long-term clinical outcomes [1]. No-reflow is attributable to the distal embolization of atheromatous or thrombotic materials that derives from mechanical fragmentation of the culprit plaque by PCI. Several studies have demonstrated that thrombus formation, positive remodeling, and greater lipid plaque burden were independent predictors of the no-reflow phenomenon [2].

Furthermore, filter-no reflow (FNR) is a phenomenon wherein flow improves after the retrieve of distal protection. However, the usefulness of distal protection remains controversial [3, 4]. Conversely, some studies showed that distal protection devices may improve myocardial reperfusion among selected high-risk patient groups [5, 6].

While, combined imaging with near-infrared spectroscopy (NIRS) and intravascular ultrasound (IVUS) enables the identification and morphometric assessment of lipid-rich plaques in vivo [7]. This dual modality catheter was validated by prior studies in the setting of acute coronary syndrome (ACS) [8, 9]. A previous autopsy study demonstrated that NIRS was able to detect features associated with plaque vulnerability in humans [10]. In fact, a maximal lipid core burden index4mm (max-LCBI4mm) > 400 in NIRS accurately distinguished culprit from non-culprit segments within the artery [9].

Thus, the present study aimed to evaluate the prediction of the FNR phenomenon during PCI by using near-infrared spectroscopy with intravascular ultrasound (NIRS–IVUS).

Methods

Study population

Thirty-two patients who underwent PCI under distal protection device for ACS were prospectively enrolled. Patients had ischemic chest discomfort with ST-segment elevation or depression of > 0.5 mm or T-wave inversion in 2 or more leads. Acute myocardial infarction was diagnosed by increased troponin T, serum levels of creatine phosphokinase (CPK) (more than twice the upper limit of normal), and creatine phosphokinase-MB fraction (> 10% of total creatine kinase). Patients without elevation of the creatine kinase-MB fraction were classified as having unstable angina.

Patients with the following criteria were excluded: cardiogenic shock, left main coronary artery disease, extremely tortuous or heavily calcified vessels, intolerance to antiplatelet drugs, in-stent restenosis after DES implantation, and severe chronic kidney disease (an estimated glomerular filtration rate < 30 ml/min/1.73 m2).

Our study was approved by our institutional ethics review board and all participants provided written informed consent for the use of their data in our retrospective analysis.

PCI procedure

Patients who had undergone emergent PCI in this study presented to the emergency room with suspected ACS. Aspirin (200 mg) was administered orally in the emergency room, followed by a daily maintenance per oral dose of 100 mg thereafter. In addition, patients were also administered clopidogrel or prasugrel, at a loading dose of 300 or 20 mg, respectively, with a daily maintenance per oral dose of 75 mg for clopidogrel and 3.75 mg for prasugrel thereafter. Thereafter, dual anti-platelet therapy (DAPT) was continued for at least 1 year after PCI. Furthermore, unfractionated heparin was administered intravenously to achieve an activated clotting time of 200–250 s during PCI. A 6-French arterial sheath and a transradial or transfemoral artery approach were utilized. Angiography was performed using a 6-F guiding catheter to identify the site of the culprit lesion. A 0.014-inch coronary guidewire was passed through the lesion. After thrombus aspiration, a filter-based distal protection device (Filtrap®; NIPRO, Osaka, Japan), was placed distal to the culprit lesion. Based on operator’s discretion, a pre-dilatation procedure was also performed. Thereafter, after intravascular evaluation by NIRS–IVUS, a stent was implanted at the culprit plaque. Post-dilatation procedures were also performed based on NIRS–IVUS assessment with the objective of achieving optimal results. After completion of the PCI procedure, the Filtrap® was removed from the patient. Thereafter, the weight of the debris retrieved by the distal protection devices was evaluated. After stent implantation, thrombolysis myocardial infarction trial (TIMI) grade was evaluated and peak CPK was also followed.

Image acquisition by NIRS (Fig. 1)

Fig. 1
figure 1

A representative data. A 98-year-old female patient with ST elevated myocardial infarction who had 100% occlusion in the left anterior descending artery is shown (a). NIRS–IVUS (b, c) shows the attenuation of grade 4, and T-LCBI, max-LCBI4mm and T-LCBI/max–LCBI4mm ratio in culprit plaque (T-LCBI, max-LCBI4mm and T-LCBI/max–LCBI4mm ratio were 532, 897 and 0.59, respectively). Stent implantation (d) was performed. Coronary angiogram after stent implantation (e) shows filter-no flow. Coronary angiogram after the retrieve of distal protection (f) shows improvement in antegrade flow

Full size image

In the present study, the TVC catheter (InfraReDx, Inc., Burlington, Massachusetts), a dual-modality intravascular imaging device that coregisters a NIRS chemogram to a gray-scale IVUS image, was used to analyze culprit lesion. NIRS–IVUS was performed with a 3.2-F TVC catheter from the distal target vessel to the guide catheter using a motorized pullback system at 0.5 mm/s. In the present study, the culprit plaque was defined as the site of stent implantation; the site was identified using any markers such as side branches and calcification and observed before and after stent implantation. The total lipid core burden index (T-LCBI) and max-LCBI4mm within the culprit plaque were then automatically calculated as previously described [11]. In addition, to express the longitudinal expansion of the lipid within the culprit plaque, the T-LCBI/max–LCBI4mm ratio within the culprit plaque was also calculated. If the lipid arc in the entire culprit plaque is the same as that in the max-LCBI4mm, the T-LCBI/max–LCBI4mm ratio results in a value of 1. NIRS signal was measured before balloon dilatation of any type. However, based on the discretion of the clinical operator, the NIRS measurement was performed after pre-dilatation. NIRS data were stored digitally for subsequent analysis.

Gray-scale intravascular ultrasound image analysis (Fig. 1)

Quantitative analysis was performed according to the American College of Cardiology clinical expert consensus document on IVUS [12] cross-sectional area (CSA) of the lumen and the vessel was determined at the reference as well as the culprit lesion site manually. Vessel CSA was determined with the external elastic membrane (EEM) as the outer margin of the vessel, and lumen CSA, with the lumen border as the outer margin of the lumen. The difference between vessel CSA and lumen CSA thus provided the CSA of the plaque + media. The percentage of CSA of plaque + media was calculated as an index of plaque burden by dividing the CSA of the plaque + media by the vessel CSA. The proximal and distal reference segments were defined as the segments with the largest lumen with the least plaque and located within 5 mm proximal and distal to the lesion, but proximal to a major side branch [13]. The remodeling index (RI) was calculated by dividing the EEM CSA at the culprit lesion by the mean of the proximal and distal reference CSA. Remodeling with an RI > 1.05 was defined as positive remodeling, whereas remodeling with an RI < 0.95 was defined as negative remodeling [14].

Attenuation was defined as the backward signal attenuation without dense calcium [15] and the grade was scored based on the extent of attenuation occupying the number of quadrants as follows: grade 0, no attenuation; grade 1, 1-quadrant (< 90°); grade 2, 2-quadrant (90°–179°), grade 3, 3-quadrant (180°–269°); grade 4, 4-quadrant (270°–360°) attenuation.

Statistical analyses

All statistical analyses were performed using SPSS version 22 (IBM Japan, Tokyo, Japan). Continuous data with non- normal distribution are presented as means ± standard deviation, and categorical data as count percentages. The patients were divided into the FNR (N = 8) and no-FNR groups (N = 24). Comparisons between the FNR group and the no-FNR group were performed by Mann–Whitney’s U test for continuous data and Fisher’s exact test for categorical data. In addition, the area under curve (AUC) by receiver-operating characteristic (ROC) curve analysis using T-LCBI, max-LCB4mm, and the T-LCBI/max–LCBI4mm ratio was calculated. This was done to clarify a marker with high accuracy for predicting the occurrence of FNR. As a result, the AUC of T-LCBI/max–LCBI4mm ratio was greatest among T-LCBI, max-LCB4mm, and the T-LCBI/max–LCBI4mm ratio (AUC = 0.868, AUC = 0.864, and AUC = 0.885, respectively). Therefore, using the T-LCBI/max–LCBI4mm ratio, multiple logistic regression analysis was performed to determine the independent predictor for the occurrence of FNR. In addition, ROC curve analysis was conducted to identify the best cut-off value of T-LCBI/max–LCBI4mm ratio for predicting the occurrence of FNR. The best cut-off value was determined using the maximum sum of sensitivity and specificity. A two-sided p-value of < 0.05 was considered statistically significant for all analyses.

Results

Patient characteristics (Table 1)

Table 1 Baseline characteristics
Full size table

The patients’ mean age was 67 ± 10 years. Men comprised 67% of patients. Overall, the frequency of FNR was 25%. There were no significant differences in age, body mass index, and medical history, including hypertension, dyslipidemia, diabetes mellitus, and smoking between 2 groups.

Difference between both groups (Table 2)

Table 2 The data during PCI
Full size table

Regarding PCI procedure, the use of thrombus aspiration, and frequency of pre-dilatation and post-dilatation were comparable between groups. Regarding IVUS data, the attenuation grade was significantly higher in the FNR group than in the no-FNR group (1.6 ± 0.6 vs. 0.9 ± 0.42, p < 0.01). Plaque burden was comparable between groups (66.3 ± 7.4 vs. 66.4 ± 3.8%, p = 0.98). However, RI in the FNR group tended to be higher than in the no-FNR group (1.18 ± 0.22 vs. 1.06 ± 0.04, p = 0.08). Lumen area and vessel area were also similar between both groups.

As for, NIRS, max-LCBI4mm, T-LCBI, and the T-LCBI/max–LCBI4mm ratio were significantly higher in the FNR group than in the no-FNR group (559 ± 160 vs. 351 ± 140, p = 0.01; 285 ± 125 vs. 116 ± 105, p = 0.01; 0.50 ± 0.10 vs. 0.33 ± 0.13, p < 0.01). In addition, the weight of the retrieved debris was significantly heavier in the FNR group than in the no-FNR group (11.1 ± 3.2 vs. 7.0 ± 2.4 mg, p < 0.01).

TIMI grade at final angiography was also similar between groups. In addition, there was no significant difference in peak CPK value between groups (2825 ± 2016 vs. 2329 ± 1109 IU/L, p = 0.73).

The predictor for the occurrence of FNR (Table 3)

Table 3 The data after PCI
Full size table

To identify predictors for the occurrence of FNR and parameters with a p-value for univariate analysis < 0.10, i.e., attenuation grade, RI and T-LCBI/max–LCBI4mm ratio were selected for the multivariate logistic regression analyses. The variables were stratified according to their average value (attenuation grade > 1.04, RI > 1.07, T-LCBI/max–LCBI4mm ratio > 0.42). Multivariate logistic regression analysis revealed that attenuation grade > 1.04 (odds ratio [OR] 18.4, 95% confidence interval [CI] 1.5–215.7, p = 0.02) and T-LCBI/max–LCBI4mm ratio > 0.42 (OR 14.4, 95% CI 1.2–176.8, p = 0.03) were significant positive predictors for the occurrence of FNR (Table 4).

Table 4 Multiple logistic regression analysis for the independent predictor of the occurrence of FNR
Full size table

The cut-off value of T-LCBI/max–LCBI4mm ratio for predicting the occurrence of FNR (Fig. 2)

Fig. 2
figure 2

Receiver operating characteristic curve analysis for predicting FNR. The cut-off value of T-LCBI/max–LCBI4mm ratio for predicting the occurrence of FNR is shown. See text for details

Full size image

Receiver operating characteristic curve analysis revealed a cut-off value of 0.47 for T-LCBI/max–LCBI4mm ratio predicting the occurrence of FNR (AUC = 0.88; sensitivity, 87.5%; specificity, 74.0%).

Discussion

The main findings of the present study are as follows: (1) LCBI value within the culprit plaque in the FNR group was significantly higher than in the no-FNR group, and (2) multiple logistic regression revealed that T-LCBI/max–LCBI4mm ratio within the culprit plaque and attenuation grade were independent predictors for FNR.

Validation between NIRS and lipid plaque

Vulnerable plaques are typically characterized by large necrotic cores with either a nonexistent or a thin cap fibrous (< 65 µm) and active macrophages near or within the fibrous cap [16, 17]. The necrotic core of the atherosclerotic plaque is highly thrombogenic and contains fragile tissues such as lipid depositions within foam cells as well as intramural bleeding and/or cholesterol crystals [18].

To validate the accuracy of the NIRS catheter for detection of lipid core plaque, two pivotal studies including the SPECTACL clinical study were carried out. These studies validated the NIRS catheter for accuracy in detection of lipid core plaque [19, 20]. In fact, a study using coronary angiography and NIRS showed that the culprit lesions of ACS were lipid core plaques (84.4%) in most cases [21]. Furthermore, in a study of 20 patients with ST-segment elevation myocardial infarction, max-LCBI4mm > 400 in NIRS accurately distinguished culprit segment from non-culprit segments within the artery [19]. A larger patient group during a multicenter trial reported similar results [22]. Moreover, our study also showed that mean max-LCBI4mm of culprit plaque in the patients with ACS was nearly 400, which is compatible with above-mentioned data. Thus, it is expected that this novel capability of NIRS will be of assistance in the management of patients with coronary artery disease.

The relationship among NIRS, lipid and FNR

As mentioned, the necrotic core plaque is highly thrombogenic, containing fragile tissues such as lipid depositions within foam cells [18]. These elements may be related to in-stent thrombus formation [23] or can be easily released into the blood stream during coronary interventions [24,25,26]. One cause of peri-procedural myocardial infarction is attributable to distal embolization of these lipid core plaque contents and/or intracoronary thrombus [27].

In addition, the no-reflow phenomenon is originally described as an angiographic phenomenon that occurs when inadequate myocardial perfusion is observed after opening an epicardial coronary artery with stenosis; regardless of the underlying causes, no-reflow phenomenon is related to long-term mortality in patients with STEMI [28, 29]. In general, this phenomenon is now understood to be caused by multiple factors including endothelial dysfunction, myocyte edema, neutrophil infiltration, microvascular spasm, oxygen-free radicals, and distal embolization of plaque and/or thrombus at culprit coronary lesion [30]. The distal atherothrombotic embolization has been reported to contribute to microvascular injury to some extent, particularly during PCI [15]. So far, there are various reports on the predictor of FNR, slow flow, and no-reflow using various imaging device such as IVUS, magnetic resonance imaging, and OCT [2, 15, 31,32,33,34,35]. Several reports have shown that atherosclerotic plaque with ultrasound attenuation might be related to the deterioration of coronary flow in patients with ACS. In fact, in several autopsy studies, ultrasound attenuation was associated with fibro-fatty tissue containing scattered cholesterol clefts and Von Kossa-positive granules corresponding to microcalcification [36, 37]. Therefore, in the present study, the high attenuation grade and high LCBI value relationship to FNR agrees with these reports. Among these risk factors for the deterioration of coronary flow, distal embolization of the contents of the plaque and/or thrombus might be prevented. Therefore, the effect of distal protection devices has been studied. However, the effect of distal protection devices was controversial [3, 4, 38, 39]. Therefore, it is important to predict the necessity of distal protection device during PCI because various imaging modalities have demonstrated an association between PCI of lipid core plaque and periprocedural MI [18, 40,41,42,43,44]. As with these imaging devices, several NIRS–IVUS studies has also described that high LCBI value, including T-LCBI and max-LCBI4mm, was associated with a high risk of periprocedural MI, presumably due to distal embolization [11, 24, 45]. In addition, the dilation of a large, longitudinal and circumferential lipid core lesion was associated with a high risk of periprocedural MI [24]. Because the T-LCBI/max–LCBI4mm ratio within the culprit plaque adopted in our study indicates the longitudinal and circumferential lipid expansion, the high T-LCBI/max–LCBI4mm ratio causing the occurrence of the FNR seems to agree with these reports. Additionally, the change in LCBI value before and after stent implantation was significantly greater in the FNR group than in no-FNR group. In addition, the present study has shown that T-LCBI/max–LCBI4mm ratio within the culprit plaque was also associated with heavy weight of debris retrieved from distal protection. Therefore, a randomized study would be required to determine whether NIRS can prospectively identify lipid core plaque prone to a high rate of distal embolization and periprocedural MI, and whether the use of a distal protection device is effective in preventing this complication.

Limitations

Limitations of this study are as follows. First, the number of study patients was small, although this is a prospective study. Second, as it had been already reported that the attenuation plaque on IVUS, T-LCBI, and max-LCBI4mm values were associated with a high incidence of transient no-reflow; we conducted a study using distal protection device to investigate the predictors such as T-LCBI/max–LCBI4mm ratio associated with FNR phenomenon, regardless of the lack of periprocedural myocardial enzyme examinations. In addition, NIRS cannot evaluate the depth of the lipid pool [46]. Therefore, another modality like optical coherence tomography should be utilized in future study. Third, distal protection device and NIRS–IVUS catheter were not used in all lesions for the following reasons: severe tortuousness, small vessel diameter, severity of stenosis, and the presence of severe calcification. Consequently, our results were limited by selection bias and may not apply to such lesions. Therefore, no definite conclusion can be drawn from the results of the present study. These limitations warrant future studies involving larger populations.

Conclusion

The use of T-LCBI/max–LCBI4mm ratio within the culprit plaque by NIRS–IVUS might be an effective marker to predict FNR during PCI.