An open-source FACS automation system for high-throughput cell biology

Recent advances in gene editing are enabling the engineering of cells with an unprecedented level of scale. To capitalize on this opportunity, new methods are needed to accelerate the different steps required to manufacture and handle engineered cells. Here, we describe the development of an integrated software and hardware platform to automate Fluorescence-Activated Cell Sorting (FACS), a central step for the selection of cells displaying desired molecular attributes. Sorting large numbers of samples is laborious, and, to date, no automated system exists to sequentially manage FACS samples, likely owing to the need to tailor sorting conditions ("gating") to each individual sample. Our platform is built around a commercial instrument and integrates the handling and transfer of samples to and from the instrument, autonomous control of the instrument's software, and the algorithmic generation of sorting gates, resulting in walkaway functionality. Automation eliminates operator errors, standardizes gating conditions by eliminating operator-to-operator variations, and reduces hands-on labor by 93%. Moreover, our strategy for automating the operation of a commercial instrument control software in the absence of an Application Program Interface (API) exemplifies a universal solution for other instruments that lack an API. Our software and hardware designs are fully open-source and include step-by-step build documentation to contribute to a growing open ecosystem of tools for high-throughput cell biology.

Characterization of lipidic plaque materials at calcified atheroma: its association with calcification thickness evaluated by optical coherence tomography and near-infrared spectroscopy imaging

Introduction

The degree of calcification and its thickness have been considered to affect stent expansion, leading to an increases risk of repeat revascularization in patients receiving PCI. Pathophysiologically, accumulation of lipidic materials within vessel wall could trigger the formation of plaque calcification. Elucidating characteristics of lipidic plaque components at calcified atheroma may enable to identify phenotypes with thick calcification which less likely responds to PCI.

Purpose

This study investigated the relationship of calcification thickness with lipidic plaque materials at calcified atheroma by using OCT and near-infrared spectroscopy (NIRS) imaging.

Methods

We analyzed 52 calcified lesions (culprit/non culprit lesions=44/8) in 47 CAD patients (stable CAD/ACS=36/11) from the REASSURE-NIRS registry (NCT04864171). OCT and NIRS imaging evaluated 4-mm segment exhibiting maximum superficial calcification arc. Calcification thickness on OCT imaging, its arc on IVUS imaging, and NIRS-derived lipid arc were analyzed at every 1-mm interval cross-sectional images. In addition, yellow-calcification ratio (YCR = lipid arc/calcification arc) was calculated (Figure 1).

Results

53% of study subjects exhibited chronic kidney disease and 70% of them received a statin (averaged on-treatment LDL-C =89mg/dL). Throughout OCT and NIRS/IVUS imaging analysis of 260 cross-sectional images, the averaged calcification arc, its maximum thickness, lipid arc and YCR were 210° (167–285°), 0.78mm (0.62–0.95mm), 95° (31–169°) and 0.33 (0.09–0.59), respectively. As expected, thicker calcification more likely exhibited a greater calcification arc (r=0.30, p<0.001). Furthermore, a greater thickness of calcification was associated with smaller lipidic plaque burden, reflected by yellow arc (r=−0.36, p<0.001) and YCR (r=−0.36, p<0.001) (Figure 2). After adjusting age, gender and ACS, calcification arc (p<0.001) and YCR (p<0.001) continued to predict thicker calcification.

Conclusion

Thickening of calcification was associated with severer calcification arc, which was accompanied by the shrinkage of lipidic plaques. Our findings suggest the evaluation of lipidic plaque component as a potential tool to identify calcified atheroma harbouring thick calcification, which may cause a greater risk of stent underexpansion.

Funding Acknowledgement

Type of funding sources: None.

Deterioration of cardiogenic shock after acute myocardial infarction defined by the society for cardiovascular angiography and intervention cardiogenic shock classification scheme

Background

Cardiogenic shock (CS) in patients with AMI presents worse cardiovascular outcomes, which suggests the need for better risk stratification and management. The Society for Cardiovascular Angiography and Intervention (SCAI) has recently proposed CS classification scheme, which stratifies CS into 5 groups according to hypotension and hypoperfusion. While stage A and B exhibits CS without hypotension and/or hypoperfusion, their clinical condition could rapidly deteriorate into stage C-E. However, clinical characteristics and in-hospital outcomes of CS exhibiting its deterioration remains uncertain.

Purpose

To characterize AMI patients who deteriorated their CS status from stage A and B into stage C-E.

Methods

This single-center observational study included 326 consecutive AMI patients receiving primary PCI who presented CS stage A and B on arrival (2019.09.01–2021.09.30). Deterioration of CS (D-CS) was defined as the progression from stage A and B on arrival to stage C-E after primary PCI. Clinical characteristics and outcomes were compared in those with and without D-CS.

Results

D-CS was identified in 16.0% of entire subjects (=52/326). Of these, 94.2 and 5.8% of them exhibited stage C and E, respectively (Figure). Patients with D-CS more likely presented STEMI (84.6 vs. 67.9%, p=0.01) with a lower systolic BP (sBP) level (130±31 vs. 148±26mmHg, p<0.001) and a reduced LVEF (43±13 vs. 51±9%, p<0.001), whereas there was no significant difference in lactate level (1.5±0.4 vs. 1.2±0.3 mmol/L, p=0.22). Pre-TIMI flow grade 0–1 (69.2 vs. 47.8%, p=0.006), left main trunk stenosis (9.6 vs. 1.5%, p=0.007) and chronic total occlusion (21.2 vs. 8.4%, p=0.01) were more frequently observed in those with D-CS. Despite achieving a shorter onset-to-reperfusion time (199 vs. 276 minutes, p=0.002), D-CS was associated with in-hospital all-cause mortality after adjusting clinical characteristics (HR=33.6, 95% CI: 2.2–502.0, p=0.01). Furthermore, mechanical circulatory support (MCS) (30.8 vs. 0%, p<0.001) was more frequently required in patients with D-CS (IABP: 28.8 vs. 0%, p<0.001, ECMO: 11.5 vs. 0%, p<0.001, Impella: 3.8 vs. 0%, p=0.02). Further analysis identified sBP (HR=0.98, 95% CI: 0.97–1.00, p=0.008), LVEF (HR=0.94, 95% CI: 0.90–0.97, p<0.001) and pre-TIMI flow grade 0–1 (HR=0.41, 95% CI: 0.19–0.86, p=0.01) as independent contributors to D-CS. ROC analysis demonstrated sBP <135 mmHg (AUC=0.65) and LVEF <50% (AUC=0.69) as best cut-off values to predict D-CS. Of note, a risk of D-CS increased in association with the number of these three factors (p<0.001), and 44.0% of those with all of these factors presented D-CS (Figure).

Conclusion

16.0% of AMI without any hypotension/hypoperfusion on arrival exhibited deterioration of CS status on SCAI classification. The combination of sBP, LVEF and pre-TIMI flow grade could help to identify AMI subjects with a risk of D-CS, who may benefit from early adoption of intensified management including MCS prior to PCI.

Funding Acknowledgement

Type of funding sources: None.

Substantially elevated thromboembolic and bleeding risks in patients with AMI following acute/subacute stroke events

Introduction

AMI infrequently but concomitantly occurs after stroke events. Current guideline recommends primary PCI with DAPT in the setting of AMI. However, this approach is not necessarily applicable in AMI subjects following acute/subacute stroke events due to its bleeding risk. Clinical management and outcomes of these AMI subjects following remains uncertain.

Purpose

To characterize management and clinical outcomes in patients with AMI following acute/subacute stroke events (=post-stroke AMI).

Methods

The current study retrospectively analyzed 2041 AMI patients hospitalized at our institute from 2007 to 2018. Post-stroke AMI was defined as its occurrence within 14 days after ischemic/hemorrhagic stroke. The use of reperfusion and anti-thrombotic therapies, and the occurrence of major adverse cardiovascular events (=CV death, non-fatal MI and non-fatal stroke) and major bleeding events (BARC type 3 or 5) were compared in post-stroke and non-post-stroke AMI patients.

Results

Post-stroke AMI was identified in 1.1% of entire subjects (=23/2041). Of these, 65% of them (=15/23) had AMI within 3 days from the onset of stoke event. Over 60% of them was due to cardioembolic stroke, followed by hemorrhagic (9%), atherothrombotic ones (8%) and other causes (22%). Post-stroke AMI patients were more likely to exhibit Af (p=0.02) and a history of hemodialysis (p=0.009), and have a lower BMI (p=0.04) and hemoglobin level (p=0.02). They were less likely to receive emergent coronary angiography, and primary PCI was conducted in only 65% of post-stroke AMI patients (Table). Furthermore, they more frequently received thrombectomy (p=0.04) alone rather than stent implantation (p=0.002) (Table). With regard to anti-thrombotic therapy, the proportion of DAPT use was significantly lower in post-stroke AMI subjects (52 vs. 89%, p=0.0001), and 17% of them did not receive any anti-thrombotic agents. Of note, only 48% (p=0.04) and 43% (p=0.0001) of post-stroke AMI patients were treated with other established medical therapies including β-blocker and statin, respectively. During the observational period (median = 2.9 years), post-stroke AMI was associated with a greater likelihood experiencing major adverse cardiovascular events (log-rank p<0.001, Figure), CV death (log-rank p<0.0001) and stroke events (log-rank p<0.0001). Furthermore, the frequency of their major bleeding events was substantially elevated (log-rank p<0.001, Figure).

Conclusions

In our real-world data, the adoption of guideline-recommended reperfusion and anti-thrombotic therapies were considerably low in AMI subjects following acute/subacute stroke events. Given their elevated risk of cardiovascular and bleeding events, it is required to establish better therapeutic management for mitigating their thrombotic/bleeding risks.

Funding Acknowledgement

Type of funding sources: None. Table 1Figure 1

The relationship of the underlying lipidic plaque at the implanted newer-generation drug-eluting stents with future stent-related events: insights from the REASSURE-NIRS registry

Background

Lipid-rich plaque is an important substrate causing acute coronary events. Near-infrared spectroscopy (NIRS) imaging has been shown to visualize lipidic coronary plaque at non-culprit site associated with future coronary events. Given that histopathological studies reported that the unstable plaque underlying the implanted drug-eluting stent (DES) could cause neoatherosclerosis formation, we hypothesized that NIRS-based evaluation of lipidic plaque burden behind the implanted DES may clinically predict the occurrence of stent failure in patients with CAD receiving PCI.

Purpose

We aimed to investigate the relationship of stent-related events' risk with lipidic plaque materials behind the implanted DES imaged by NIRS/intravascular ultrasound (NIRS/IVUS) imaging.

Methods

The REASSURE-NIRS registry is an on-going multi-center registry to enroll CAD subjects receiving NIRS/IVUS-guided PCI. In this registry data, 406 lesions in 379 CAD subjects (ACS/non-ACS=150/229) receiving new-generation DES were analyzed. Minimum stent area (MSA) after PCI and maximum lipid-core-burden index in any 4mm-segment within the implanted stents (in-stent maxLCBI4mm) were measured. A 3-year lesion-oriented composite outcome [LOCO: culprit lesion-related MI + ischemia-driven target lesion revascularization (ID-TLR)] was compared in subjects stratified according to the tertile of in-stent maxLCBI4mm.

Results

The mean value of in-stent maxLCBI4mm was 221, and 17% of lesions exhibited in-stent maxLCBI4mm >400. Patients with a greater in-stent maxLCBI4mm were more likely to exhibit a higher LDL-C level (p=0.026) with a longer stent length (p<0.001) and a smaller MSA (p=0.033) (Picture 1). Over 95% of entire study subjects received a statin. During the observational period (median=726 days), the frequency of LOCO up to 3 years was 3.4% in entire study subjects (culprit lesion-related MI=1.0%, ID-TLR=2.8%). Kaplan-Meier curve analysis demonstrated that the occurrence of LOCO did not increase in association with in-stent maxLCBI4mm (log-rank p-value=0.25, Picture 2). In addition, in-stent maxLCBI4mm did not associate with each component of LOCO (culprit lesion-related MI: p=0.502, ID-TLR: p=0.872). Receiver Operating Characteristic analysis revealed that the predictive ability of in-stent maxLCBI4mm for the occurrence of LOCO was unsatisfactorily (c-statistics=0.486).

Conclusion

The amount of underlying lipidic materials at culprit lesions receiving new-generation DES implantation did not necessarily predict future stent-related events. Clinical significance of maxLCBI4mm behind the implanted DES may be different from that at naïve non-culprit plaques.

Funding Acknowledgement

Type of funding sources: None. Background and lesion characteristicsKaplan-Meier analysis for LOCO

Characterization of cholesterol efflux capacity in diabetic and non-diabetic patients with coronary artery disease: comparison between acute coronary syndrome and stable coronary artery disease

Introduction

Type 2 diabetic patients more likely exhibit a lower high-density lipoprotein (HDL) level. Given a greater glycation and oxidative stress in diabetic subjects, these atherogenic characteristics could cause dysfunctional HDL including a reduced cholesterol efflux capacity (CEC), which may account for an increased risk of diabetic macrovascular disease including acute coronary syndrome (ACS). However, it remains to be fully elucidated characteristics of HDL-mediated CEC in type 2 diabetic patients, in association with clinical presentation of coronary artery disease (CAD).

Purpose

To characterize CEC in CAD subjects with type 2 diabetes mellitus.

Methods

The current study prospectively analyzed 87 statin-naive patients with CAD. CEC was measured by using the collected apolipoprotein B-depleted serum. Liquid scintillation counting (Perkin-Elmer Analytical Sciences, MA, US) was used to quantify the efflux of radioactive cholesterol from J774 cells. Clinical characteristics and CEC were compared in diabetic and non-diabetic subjects.

Results

The averaged HbA1c in diabetic patients was 6.7±1.2, and 66.7% of them achieved HbA1c <7.0%. Diabetic subjects more likely exhibited a history of hypertension and dyslipidemia, and multi-vessel disease (Table). Moreover, a lower CEC level was observed in diabetic patients, accompanied by a lower HDL-C and apolipoprotein A-I levels with a higher level of triglyceride (Table). HDL-C (r=0.62, p-value<0.01) and Apolipoprotein A-I (r=0.70, p-value <0.01) were associated with CEC, whereas there was no significant difference in CEC between subjects with HbA1c <7.0% vs. ≥7.0% (0.74±0.07 vs. 0.78±0.08, p=0.22). On multivariate analysis, type 2 diabetes mellitus was an independent contributor to CEC <0.79 (median) (HR=2.75, 95% CI: 1.11–6.82, p=0.03). Interestingly in particular, CEC was substantially lower in diabetic patients with ACS compared to those with stable CAD (Figure). By contrast, clinical presentation of CAD did not affect CEC in non-diabetic subjects (Figure).

Conclusions

A lower CEC level was observed in subjects with type 2 diabetes mellitus. In particular, this HDL functionality was profoundly diminished in those presenting ACS. Our findings suggest functionality of HDL as a potential therapeutic target in diabetic patients experiencing ACS.

Funding Acknowledgement

Type of funding sources: None. Table 1Figure 1

Comparison of coronary atherosclerotic features in response to achieving LDL-C <55 mg/dl between non-diabetic and diabetic patients: insights from the REASSURE-NIRS registry

Introduction

Current ESC guideline recommends achieving LDL-C <1.4 mmol/l in very high-risk subjects. Despite fabvourable anti-atherosclerotic effects of lowering LDL-C, its efficacy is diminished in type 2 diabetic patients. Whether response of coronary atheroma to on-treatment LDL-C <1.4 mmol/l differs in diabetic and non-diabetic subjects has not been elucidated yet.

Methods

The REASSURE-NIRS registry is an on-going multi-center registry to enroll CAD subjects receiving PCI under the guidance of near-infrared spectroscopy/intravascular ultrasound (NIRS/IVUS: DualProTM, Nipro, Tokyo, Japan) imaging. Culprit lesions in 557 CAD patients who already received a statin were evaluated by NIRS/IVUS. Maximum 4-mm-lipid-core burden-index (maxLCBI4mm) and plaque calcification grade at culprit sites were measured. Calcification grade at each 1-mm cross-sectional image was defined as follows: calcium arc 0° = 0, 0–90° = 1, 90–180° = 2, 180–270° = 3, 270–360° = 4. MaxLCBI4mm and the averaged calcification grade were compared in diabetic and non-diabetic subjects stratified according to on-treatment LDL-C level, respectively.

Result

The proportion of diabetic (n=293, HbA1c; 6.9±0.9%) and non-diabetic patients (n=264) with on-treatment LDL-C <1.4 mmol/l was 8.54 and 16.67%, respectivey (p=0.01). In non-diabetic patients, achieving LDL-C <1.4mmol/L was associated with a lower maxLCBI4mm, whereas, in diabetic patients, maxLCBI4mm was numerically smaller under achieving LDL-C <1.4 mmol/l, but this comparison did not meet statistical significance (Figure 1). Furthermore, a greater degree of calcification grade in non-diabetic patients was observed in association with on-treatment LDL-C level (Figure 2). However, plaque calcification at diabetic coronary atheroma was not necessarily induced under achieving stricter LDL-C goal. Subgroup analysis demonstrated that diabetic patients with body mass index ≥25 (odds ratio = 0.15; 95% CI: 0.18–1.19, p=0.04), estimated glomerular filtration rate <60 (mL/min/1.73m2) (odds ratio = 0.31; 95% CI: 0.10–0.90, p=0.03) and non-insulin use (odds ratio = 0.36; 95% CI: 0.14–0.87, p=0.02) benefit from achieving LDL-C <1.4 mmol/l.

Conclusion

Achieving LDL-C <1.4 mmol/l was associated with more stabilized atheroma in non-diabetic patients with CAD, whereas these favourable effects were not observed in diabetic subjects. Our findings suggest the potential need to modify additional atherogenic risks for stabilizing diabetic coronary atheroma under achieving LDL-C <1.4 mmol/l.

Funding Acknowledgement

Type of funding sources: None. Figure 1Figure 2

Characterization of residual lipid-rich plaques despite achieving LDL-C <1.8mmol/l with a statin in patients with coronary artery disease: insights from the REASSURE-NIRS registry

Introduction

Recent studies have demonstrated favourable modification of lipidic plaque materials under achieving LDL-C <1.8mmol/l with a statin, which potentially accounts for its clinical benefit. However, coronary events still occur even under optimal LDL-C management. This may suggest the presence of residual lipid-rich coronary plaque despite on-treatment LDL-C <1.8mmol/l. Given that near-infrared spectroscopy (NIRS) enables quantitative evaluation of lipidic plaque in vivo, we employed this imaging modality to investigate characteristics and drivers of residual lipid-rich plaques in statin-treated patients with coronary artery disease (CAD) who achieved LDL-C <1.8mmol/l.

Purpose

To clarify the frequency, clinical demographics and factors associated with residual lipid-rich plaques under LDL-C <1.8mmol/l.

Methods

The REASSURE-NIRS registry is an on-going multi-center registry to enroll CAD subjects receiving NIRS/intravascular ultrasound-guided PCI. The current analysis included 133 statin-treated stable CAD patients with on-treatment LDL-C <1.8mmol/l from August 2015 to December 2020. The maximum 4-mm lipid core burden index (maxLCBI4mm) at culprit lesions was measured by NIRS imaging prior to PCI. Clinical characteristics were compared in patients with and without maxLCBI4mm ≥400 at culprit lesions.

Results

In the current study, 45% (=58/128) of study subjects exhibited maxLCBI4mm ≥400 at culprit lesions under on-treatment LDL-C <1.8 mmol/l. They were more likely to be female, whereas there were no differences in age and the frequency of risk factors. Most of study subjects received moderate to high-intensity statin (p=0.79), and over one-fourth of them were treated with ezetimibe (p=0.56). Under these lipid-lowering therapies, LDL-C level was significantly higher in patients with maxLCBI4mm ≥400 (Table). Additionally, a lower frequency of LDL-C <1.4mmol/l was observed in those exhibiting maxLCBI4mm ≥400 (31.0 vs. 45.7%), but this comparison failed to meet statistical significance (p=0.09). Despite LDL-C control with a statin, deterioration of coronary flow after PCI with stent implantation more frequently occurred in patients with maxLCBI4mm ≥400 (Table). Multivariate analysis demonstrated that an independent factor associated with maxLCBI4mm ≥400 was LDL-C level (OR=1.05; 95% CI=1.00–1.10, p=0.03), but not other lipid and clinical parameters.

Conclusion

Almost half of CAD subjects who achieved LDL-C level <1.8mmol/l still exhibited the accumulation of lipidic plaque materials within vessel wall. Given that LDL-C level was associated with this residual lipid-rich plaque features, our findings support current ESC-guideline recommended LDL-C goal (<1.4mmol/l) to optimize the secondary prevention in stable CAD patients.

Funding Acknowledgement

Type of funding sources: None.

Characterization of thromboembolic and bleeding risks in cancer patients with acute myocardial infarction under the use of guideline-recommended dual-antiplatelet therapy

Background

Atherosclerotic cardiovascular disease including acute myocardial infarction (AMI) has become one of major co-existing diseases in cancer patients due to their improved survival rate. Current guideline recommends dual-antiplatelet therapy (DAPT) in patients with AMI. Given that the presence of cancer elevates not only coagulability but bleeding risks, these substrate may further worsen cardiovascular outcomes and bleeding risks in cancer subjects with AMI receiving DAPT.

Methods

We retrospectively analyzed 712 AMI patients treated by primary PCI with drug-eluting stent and DAPT between 2007 and 2017. The diagnosis of cancer was determined through medical record review. Clinical characteristics, thromboembolic (=all-cause death+non-fatal MI+stroke) and bleeding events were compared in AMI subjects with vs. without cancer.

Results

Cancer was identified in 11.1% (=79/712) of study subjects. Of these, around 40% of them had gastrointestinal cancer (=35/79), followed by lung cancer (=5/79) and breast cancer (=8/79). Cancer patients were more likely to be older (77±7 v. 69±13 years, p<0.001) with a history of Af (25 v. 10%, p<0.001), CKD (eGFR<60: 60 v. 42%, p=0.002), anemia (hemoglobin: 12.8±1.8 v. 13.9±1.8 g/dl, p<0.001). Under anti-thrombotic (DAPT=86%, triple-antiplatelet therapy=14%) and optimal medical therapies (ACE-I=90%, beta-blocker=76%, statin=96%), more frequent occurrence of thromboembolic events was observed in patients with cancer (34.2 v. 12.6%, p=0.004, Picture). Furthermore, the presence of cancer was associated with more than four times greater risk of bleeding events compared to non-cancer subjects (18.9 v. 4.3%, p<0.001, Picture). In particular, the frequency of both major (10.1 vs. 3.3%, p=0.003) and minor (8.9 vs. 0.9%, p<0.001) bleeding events was significantly higher in patients with cancer. In multivariate analysis, cancer independently predicted bleeding events (Table).

Conclusions

Under the use of guideline recommended DAPT, the concomitance of cancer in AMI subjects was a predictor for thromboembolic as well as bleeding events. In particular, the relationship between cancer and bleeding was significant. These observations underscore the appropriate selection and duration of anti-thrombotic agents in AMI subjects with cancer.

Cardiac/Bleeding Events in AMI Subjects

Funding Acknowledgement

Type of funding source: None

Characterization of plaque features exhibiting physiological mismatch between fractional flow reserve and resting index: near-infrared spectroscopy imaging analysis

Background

In addition to fractional flow reserve (FFR), resting indexes (RI) have been shown as another physiological measure to evaluate myocardial ischemia. Despite the clinical usefulness of RI without the use of intravenous vasodilatory agent, discrepancy between FFR and RI infrequently occurs. Whether this physiological mismatch is derived by specific plaque feature remains unknown.

Purpose

To characterize coronary plaques associated with coronary physiological mismatch.

Methods

We analyzed 59 coronary arteries (LAD/RCA/LCX=49/4/6) with FFR≤0.80 in 57 stable CAD subjects receiving PCI. Following measurement of FFR and RI, culprit lesion was evaluated by near-infrared spectroscopy and intravascular ultrasound (NIRS/IVUS). The analyzed vessels were stratified according to FFR and RI values: FFR≤0.75+RI>0.89 (n=6: physiological mismatch), FFR>0.75+RI>0.89 (n=6), FFR≤0.75+RI≤0.89 (n=33) and FFR>0.75+RI≤0.89 (n=14).

Results

The median values of percent diameter stenosis, FFR and RI were 51%, 0.75 and 0.87, respectively. Physiological mismatch was observed in 10.1% (=6/59) of analyzed vessels. On IVUS imaging, maximum percent plaque area was greater than 70% in all groups (p=0.29). Furthermore, there were no significant differences in angiographic and IVUS-derived minimum lumen area across 4 groups (Table). However, culprit lesions exhibiting physiological mismatch contained a substantially larger amount of lipid plaque, reflected by a higher maximum 4-mm lipid-core burned index (maxLCBI4mm: p=0.04) on NIRS imaging (Table). Multivariate analysis demonstrated maxLCBI4mm as the only plaque feature associated with physiological mismatch (odds ratio=1.010, 95% CI: 1.001–1.019, p=0.02).

Conclusion

Plaque feature associated with coronary physiological mismatch was the extent of lipidic materials but not the quantity of coronary atheroma. Since the accumulation of lipidic plaque component is caused by endothelial dysfunction, this vascular substrate could impair baseline vasomotion, thereby causing a lower FFR despite preserved RI value. Evaluation of lipidic burden may be a potential option to avoid unnecessary deferral of revascularization in subjects with normal RI value.

maxLCBI4mm in each group

Funding Acknowledgement

Type of funding source: None

Ex Vivo assessment of competent strut coverage after coronary stenting by optical coherence tomography

Background

In many studies, struts coverage is defined as >0 mm of tissue overlying the stent struts by optical coherence tomography (OCT). However, this definition has never been validated using histology as the “gold standard”. The present study sought to assess the appropriate cut-off value of neointimal thickness of stent strut coverage by OCT using histology.

Methods

OCT imaging was performed on 39 human coronary arteries with stents from 25 patients at autopsy. A total of 165 cross-sectional images from 46 stents were co-registered with histology. The optimal cut-off value of strut coverage by OCT was determined. Strut coverage by histology was defined as endothelial cells with at least underlying two layers of smooth muscle cells. Considering the resolution of OCT is 10–20 μm, 3 different cut-off values (i.e. at ≥20, ≥40, and ≥60 μm) were assessed.

Results

A total of 2235 struts were evaluated by histology. Eventually, 1216 struts which were well-matched struts were analyzed in this study. By histology, uncovered struts were observed in 160 struts and covered struts were observed in 1056 struts. The broadly used definition of OCT-coverage which does not consider neointimal thickness yielded a poor specificity of 37.5% and high sensitivity 100%. Of 3 cut-off values, the cut-off value of >40 μm was more accurate as compared to >20 and >60 mm [sensitivity (99.3%), specificity (91.0%), positive predictive value (98.6%), and negative predictive value (95.6%)]

Conclusion

The most accurate cut-off value was ≥40 μm neointimal thickness by OCT in order to identify stent strut coverage validated by histology.

Funding Acknowledgement

Type of funding source: None

P5635Predictive ability of lipdic burden for FFR-derived physiological measures: insights from near-infrared spectroscopy imaging analysis

Background

Fractional flow reserve (FFR) has enabled to physiologically assess the myocardial ischemia of coronary artery with intermediate stenosis. Mechanistically, not only the severity of coronary stenosis but also the extent of maximal vasodilatation within the entire coronary artery potentially affects this physiological measure. Since the accumulation of lipidic materials within vessel wall increases vascular stiffness via inducing endothelial dysfunction, the presence of lipidic atheroma burden may affect physiological measures.

Purpose

To investigate the association of FFR with lipidic coronary atheroma by near-infrared spectroscopy/intravascular ultrasound (NIRS/IVUS) imaging, which quantitatively visualize lipidic burden in vivo.

Methods

We analyzed 61 coronary arteries (LAD/RCA/LCX=52/5/4) with FFR≤0.80 in 59 stable coronary artery disease subjects receiving PCI. Following FFR measurement, NIRS/IVUS imaging was conducted to evaluate the extent of atheroma burden (maximum percent plaque area=max%PA) and lipidic materials (lipid core burden index within the entire vessel=LCBIvessel). The analyzed vessels were stratified according to FFR: definite FFR group (FFR≤0.74, n=34) and gray-zone FFR group (0.75≤FFR≤0.80, n=27).

Results

NIRS/IVUS imaging analysis (analyzed longitudinal length=77±7mm) was more likely to exhibit a significantly higher LCBIvessel and a larger max%PA in the definite FFR group (Table). Of note, FFR was significantly correlated to LCBIvessel (ρ=-0.299, p=0.02), but not max%PA (ρ=-0.255, p=0.07). Multivariate analysis demonstrated that an independent determinant of FFR≤0.74 was LCBIvessel [odds ratio (OR)=1.016, 95% confidential interval (CI)=1.002–1.031, p=0.02], but not max%PA [OR=1.084, 95% CI=0.994–1.182, p=0.07]. Area under the receiver-operating characteristic curve analysis elucidated that the addition of LCBIvessel to angiography- and IVUS-derived measures resulted in a significant improvement for detecting FFR≤0.74 (picture).

Definite FFR Group (FFR≤0.74, n=34) Gray-zone FFR Group (0.75≤FFR≤0.80, n=27) p value Fractional flow reserve (FFR) 0.68±0.05 0.78±0.02 <0.01 Percent diameter stenosis (%) 56.2±13.1 51.9±7.8 0.16 Maximum percent plaque area (max%PA, %) 84.3±6.9 79.9±7.2 0.01 Lipid core burden index within the entire vessel (LCBIvessel) 102.0±60.2 65.6±51.6 0.01

ROC analysis for detecting FFR≦0.74

Conclusion

The propagation of lipidic burden associates with the physiological measures. The present findings indicate the possibility that vessel characteristics or instability may have influence for causing ischemia on the coronary artery.

P1561An elevated risk of heart failure and stroke events in octogenarian Japanese patients with acute myocardial infarction who received percutaneous coronary intervention

Background

The proportion of the octogenarian population is expanding especially in Eastern society. Due to the clustering of risk factors, acute myocardial infarction (AMI) represents a major cardiovascular complication in octogenarian subjects. This suggests the need to further optimize their therapeutic management to prevent future cardiac events after AMI. However, analysis of clinical characteristics and cardiovascular outcomes in octogenarian subjects with AMI who received the current established medical therapies is limited.

Purpose

To investigate clinical features and prognosis in octogenarian AMI subjects treated with percutaneous coronary intervention (PCI).

Methods

We analyzed 1547 AMI subjects underwent PCI between 2007 and 2017. Baseline characteristics and the occurrence of composite major adverse cardiovascular events (cardiac death, non-fatal MI, revascularization, heart failure and stroke) were compared in octogenarian and non-octogenarian subjects.

Results

22.0% (340/1547) of study subjects was octogenarian. They were more likely to have chronic kidney disease (CKD) and a lower level of LDL-C on admission (Table). Moreover, a higher prevalence of severer Killip class and LVEF <30% were observed in octogenarians (Table). However, they were not optimally treated with the established medical therapies at discharge (Table). During the observational period (median=3.1 years), the composite of cardiovascular events more frequently occurred in octogenarian subjects. Of note, they exhibited a 2.15-fold and 3.01-fold increased risk for heart failure and stroke events, respectively (Figure).

Table 1 Non-Octogenarian (n=1207) Octogenarian (n=340) P-value CKD* (%) 33.8 63.2 <0.0001 LVEF <30% (%) 5.7 10.3 0.02 Killip class 1.33±0.03 1.55±0.05 <0.0001 LDL-C (mmol/L) 3.20±0.03 2.80±0.05 <0.0001 Statin (%) 86.3 78.2 0.0006 Beta-blocker (%) 74.0 65.8 0.005 ACE-I/ARB (%) 87.3 76.6 <0.0001 DAPT (%) 86.0 88.6 0.42 *CKD is defined as estimated glomerular filtration rate (eGFR) <60 ml/min/1.73 m2.

Figure 1

Conclusions

Octogenarian subjects with AMI were high-risk group associated with heart failure and stroke events. Their distinct clinical backgrounds may affect the adoption of optimal medical therapies, potentially resulting in worse cardiovascular outcomes. Further intensified management should be applied to octogenarian subjects with AMI.