Diagnosis and Prognostic Value of the Underlying Cause of Acute Coronary Syndrome in Optical Coherence Tomography–Guided Emergency Percutaneous Coronary Intervention

Abstract

Background

The prognostic impact of optical coherence tomography–diagnosed culprit lesion morphology in acute coronary syndrome (ACS) has not been systematically examined in real‐world settings.

Methods and Results

This investigator‐initiated, prospective, multicenter, observational study was conducted at 22 Japanese hospitals to identify the prevalence of underlying ACS causes (plaque rupture [PR], plaque erosion [PE], and calcified nodules [CN]) and their impact on clinical outcomes. Patients with ACS diagnosed within 24 hours of symptom onset undergoing emergency percutaneous coronary intervention were enrolled. Optical coherence tomography–guided percutaneous coronary intervention recipients were assessed for underlying ACS causes and followed up for major adverse cardiac events (cardiovascular death, myocardial infarction, heart failure, or ischemia‐driven revascularization) at 1 year. Of 1702 patients with ACS, 702 (40.7%) underwent optical coherence tomography–guided percutaneous coronary intervention for analysis. PR, PE, and CN prevalence was 59.1%, 25.6%, and 4.0%, respectively. One‐year major adverse cardiac events occurred most frequently in patients with CN (32.1%), followed by PR (12.4%) and PE (6.2%) (log‐rank P<0.0001), primarily driven by increased cardiovascular death (CN, 25.0%; PR, 0.7%; PE, 1.1%; log‐rank P<0.0001) and heart failure trend (CN, 7.1%; PR, 6.8%; PE, 2.2%; log‐rank P<0.075). On multivariate Cox regression analysis, the underlying ACS cause was associated with 1‐year major adverse cardiac events (CN [hazard ratio (HR), 4.49 [95% CI, 1.35–14.89], P=0.014]; PR (HR, 2.18 [95% CI, 1.05–4.53], P=0.036]; PE as reference).

Conclusions

Despite being the least common, CN was a clinically significant underlying ACS cause, associated with the highest future major adverse cardiac events risk, followed by PR and PE. Future studies should evaluate the possibility of ACS underlying cause‐based optical coherence tomography–guided optimization.

Nonstandard Abbreviations and Acronyms

BARC

Bleeding Academic Research Consortium

CN

calcified nodule

MACE

major adverse cardiac events

PE

plaque erosion

PR

plaque rupture

TIMI

Thrombolysis in Myocardial Infarction

TLR

target lesion revascularization

Clinical Perspective.

What Is New?

• Optical coherence tomography–based frequencies of underlying acute coronary syndrome (ACS) causes were plaque rupture 59%, plaque erosion 26%, and calcified nodules 4%, including only patients with acute onset (symptoms within 24 hours).

• The underlying ACS causes defined by optical coherence tomography were associated with 1‐year major adverse cardiac events, which were highest among patients with calcified nodules, followed by those with plaque rupture and plaque erosion.

What Are the Clinical Implications?

• Optical coherence tomography–guided emergency percutaneous coronary intervention for patients with ACS was useful to differentiate the underlying causes of ACS and enable future major adverse cardiac events risk stratification.

• Further studies are warranted to determine whether optical coherence tomography can optimize emergency percutaneous coronary intervention according to each underlying cause of ACS.

Acute coronary syndrome (ACS) can cause sudden cardiac death, which mainly occurs due to acute coronary thrombosis as a consequence of atherosclerotic plaque accumulation. Cadaver studies of sudden coronary death have identified the major underlying plaque morphologies producing thrombosis as plaque rupture (PR), plaque erosion (PE), and calcified nodules (CN). ¹ Optical coherence tomography (OCT) is a light‐based imaging modality allowing high‐resolution imaging of intraluminal coronary artery structures. It has been used to diagnose the 3 most common in vivo causes of ACS, and the prevalence of each has been reported. ² , ³ , ⁴ Several OCT studies have reported that plaque morphology affected long‐term outcomes of patients with ACS following index percutaneous coronary intervention (PCI); patients with PE had better prognoses than those with PR. ⁵ , ⁶ However, the prevalence of OCT‐identified causes has been reported in retrospective studies with limited patient numbers and images recorded at varying time points. ⁴ In addition, the reported prognostic impacts of the underlying causes did not include those for patients with CN alone ⁵ , ⁶ ; therefore, they may not be generalizable to patients with ACS in real‐world settings. Furthermore, the proportion and characteristics of patients who did not undergo OCT assessment were not investigated; hence, the overall population of ACS was unknown.

Intravascular imaging is also used to optimize PCI. ⁷ The noninferiority of OCT‐guided versus intravascular ultrasound (IVUS)‐guided PCI has been reported according to target vessel failure at 1 year. ⁸ The clinical use of OCT for emergency PCI in ACS is gradually increasing ⁹ ; however, data that reflect daily clinical practice regarding the feasibility of OCT‐guided PCI, especially for ACS, are lacking.

This study aimed to investigate the prevalence of underlying ACS causes using OCT‐defined assessment of the culprit lesion morphology in patients diagnosed within 24 hours of symptom onset and their impact on clinical outcomes, from a prospective registry in real‐world ACS settings. We additionally aimed to assess the penetration and feasibility of OCT‐guided emergency PCI.

METHODS

Study Design and Population

The authors declare that all supporting data are available within the article and its online supplementary files. The TACTICS (Tokyo, Kanagawa, Chiba, Shizuoka, and Ibaraki active OCT applications for ACS) registry is an investigator‐initiated, prospective, multicenter, observational study conducted at 22 Japanese hospitals between November 2019 and April 2021. Patients with ACS diagnosed within 24 hours of symptom onset who underwent OCT‐guided emergency PCI were enrolled. ACS diagnoses included ST‐segment–elevation myocardial infarction (STEMI), non‐ST‐segment–elevation myocardial infarction, and unstable angina. Myocardial infarction (MI) was defined according to the Fourth Universal Definition of MI. ¹⁰ Unstable angina was defined as angina with (1) onset within 24 hours of the latest symptoms, (2) chest pain at rest or with minimal exertion, and (3) ST‐segment deviation of >1 mm in ≥2 contiguous electrocardiographic leads. ¹¹ The inclusion and exclusion criteria are listed in Table S1. The main exclusion criteria were non‐de novo lesions, contraindications to dual antiplatelet therapy, and life expectancy of <2 years. The study rationale and design, including prespecified analysis planning, were published previously. ¹² The protocol was approved by each institution's ethics committee and registered in the University Hospital Medical Information Network Clinical Trials Registry of Japan (UMIN‐CTR, ID 000039050). This study was performed in accordance with the principles of the Declaration of Helsinki, and all patients provided written informed consent. A simultaneous, parallel observational cohort called the TACTICS Background registry comprised patients with ACS diagnosed within 24 hours of symptom onset who underwent emergency PCI but were not enrolled in TACTICS. This was created to further explore the penetration and feasibility of OCT‐guided emergency PCI and clarify selection bias of the TACTICS registry. Data management and analysis were performed by the Division of Clinical Pharmacology, Department of Pharmacology, Showa University School of Medicine (Tokyo, Japan).

End Points

The primary end point of the study was to identify the prevalence of underlying causes of ACS using OCT‐defined morphological assessment of the culprit lesion. The key secondary clinical end points were hazard ratios of composited major adverse cardiac events (MACE) in patients with underlying ACS causes at 12‐month follow‐up. MACE included cardiovascular death, nonfatal MI, heart failure (HF), or ischemia‐driven revascularization. Cardiovascular death was defined as death from cardiovascular causes ¹³ ; MI according to the Fourth Universal Definition of MI ¹⁰ ; and HF as prolonged hospitalization (inpatient) or new hospitalization (postdischarge) with new or worsening symptoms with objective signs of HF and treatment initiation or escalation specifically for HF. ¹⁴ Ischemia‐driven revascularization was defined as any revascularization with at least 1 of the following: (1) angina symptoms despite optimal medical therapy, (2) new ischemic electrocardiogram changes, (3) positive noninvasive test, or (4) positive invasive physiological test. Target lesion revascularization was defined as repeat PCI of the target lesion (the treated segment including the 5‐mm margins proximal and distal to the stent) or bypass surgery of the target vessel. Other clinical events included all‐cause death, stroke, stent thrombosis, ¹³ and bleeding, defined as Bleeding Academic Research Consortium types 2, 3, or 5. ¹⁵ All clinical event data were collected by scheduled hospital visit or telephone contact. An independent committee that consisted of clinicians who were unaware of OCT findings adjudicated all clinical events.

PCI Procedures

The PCI operator identified the culprit lesion based on combined findings of the coronary angiogram, electrocardiographic assessment, regional wall motion abnormalities on echocardiogram, and intraluminal thrombus formation identified using OCT images. Standard emergency PCI was performed aiming to achieve prompt Thrombolysis in MI (TIMI) III grade recanalization, mainly followed by second‐ or newer‐generation drug‐eluting stent implantation. OCT images were acquired using the ILUMIEN OPTIS system and Dragonfly OPTIS or Dragonfly OpStar imaging catheter (Abbott Vascular, Santa Clara, CA) after recanalization. The catheter was advanced beyond the target lesion and withdrawn to the guide catheter at a motorized pullback speed of 36 mm/s. During image acquisition, blood was displaced by the continuous injection of contrast medium or low‐molecular‐weight dextran. Initial preprocedural OCT recording was recommended; in cases of a low TIMI grade, thrombus aspiration or gentle predilatation using a balloon ≤2.0 mm were permitted to obtain prompt recanalization if necessary. When recording OCT, if available, the OPTIS Mobile system (Abbott Vascular) was used to co‐register the coronary angiogram and OCT. Targets for PCI optimization using OCT were decided according to the details in Table S2, ¹⁶ and left to the operator's discretion. Briefly, the reference site was identified adjacent to the culprit lesion, which was minimally diseased with the least lipid plaque. When the stent was applied, stent diameter was determined using either the mean lumen diameter at the distal reference site rounded up by 0.25 to 0.5 mm or the smaller mean external elastic lamina diameter of the proximal or distal reference rounded down to the nearest 0.25 mm. ⁷ To avoid no‐reflow phenomena from excessive poststent balloon dilatation of ACS culprit lesions, which were commonly rich in lipid plaques and thrombi, procedure termination was at the operator's discretion. The operator was required to document the timing of OCT image acquisition and how the OCT information changed the PCI strategy at each timepoint. After PCI, all patients received standard dual antiplatelet therapy (aspirin and any P2Y12 receptor inhibitor) for at least 1 month at the physician's discretion, considering both ischemic and bleeding risks, followed by single antiplatelet therapy. Guideline‐directed medical treatment and lifestyle management of coronary risk factors for secondary prevention, including cardiac rehabilitation, were recommended for all patients. ¹⁷

Angiographic Imaging and Analysis

Cine angiograms were analyzed offline at an angiographic core laboratory (Cardiocore Japan, Tokyo, Japan), blinded to clinical and OCT findings. Lesion complexity was assessed using the American College of Cardiology/American Heart Association classification. ¹⁸ The SYNTAX score was calculated before and after emergency PCI. ¹⁹ Postprocedural myocardial perfusion was graded using the TIMI criteria.

OCT Image Acquisition and Analysis

OCT images were analyzed using offline software at an independent OCT core laboratory (Kobe Cardiovascular Core Analysis Laboratory, Kobe, Japan). Two independent experienced interventional cardiologists performed qualitative analysis of the baseline OCT and assessed culprit‐plaque morphology; they were blinded to all clinical data except those of the PCI procedure and angiography. In case of discrepancy, a consensus reading was obtained via discussion at a conference attended by at least 6 experienced analysts. Quantitative and qualitative evaluations were performed frame by frame as previously described. ²⁰ , ²¹ , ²²

PR was identified by a disrupted fibrous cap overlying lipid plaque, with or without intraplaque cavity formation. PE was identified by attached thrombus overlying a visualized plaque without evidence of fibrous cap disruption, evaluated on multiple adjacent frames. CN was defined as calcific nodules erupting in the lumen with disruption of the fibrous cap and the presence of substantive calcium proximal or distal to the lesion. The culprit lesions not meeting the preceding definitions were classified as “others,” which included the following (Data S1): (1) spontaneous coronary artery dissection; (2) ectasia/aneurysm; (3) significant stenosis; (4) vasospasm; or (5) coronary embolism. The underlying tissue characteristics of the culprit lesion, including lipid plaque, thin‐cap fibroatheroma, calcified plaque, thrombus, macrophage accumulation, cholesterol crystal microchannel, and layered plaque, were assessed (Data S1).

On quantitative OCT analysis, the culprit lesion length was defined from the distal to proximal reference site, generally the stented segment. Areas of the minimum lumen, minimum stent, and proximal and distal reference lumens (the largest area within 5 mm of the stent edge before a major side branch) were recorded, and the mean reference lumen area (the averaged proximal and distal reference lumen areas) and stent expansion (minimum stent area divided by the mean reference lumen area) were calculated. Stent eccentricity index was defined as the minimum stent diameter divided by the maximum stent diameter at a single location. The maximum in‐stent tissue protrusion and stent‐malapposed areas were measured. Stent‐edge dissection was defined as the disruption of the vessel luminal surface with a visible flap at the stent edge or the 5‐mm proximal and distal reference segments. ²³

Statistical Analysis

Statistical analysis was performed using R statistical software (version 4.1.2, R Foundation for Statistical Computing, Vienna, Austria). Continuous variables are presented as mean±SD and were compared using an unpaired t test when normally distributed or are presented as median (interquartile range) when non‐normally distributed. Categorical variables are expressed as frequency (%) and were compared using χ ² statistics or Fisher exact test. Interobserver agreement among the main 3 underlying causes of ACS (PR, PE, and CN) was assessed using Fleiss Kappa (κ) statistics in 100 randomly selected plaques. Bonferroni adjustment was used to control for multiple comparisons between the 3 groups (PR, PE, and CN) and performed only when the P value for the overall 3‐group test was <0.05. Time‐to‐event data are presented as Kaplan–Meier estimates and were compared using the log‐rank test. Hazard ratio (HR) and 95% CI were estimated using a Cox proportional hazards regression model. Multivariate Cox models were used to adjust for baseline variables considered clinically relevant based on proven associations with adverse cardiovascular outcomes. The covariates entered into these models included age, sex, hemodialysis, ACS presentation (STEMI, non‐ST‐segment–elevation myocardial infarction, or unstable angina), diabetes, left ventricular ejection fraction, mean reference lumen area by final OCT, stent expansion, and post‐SYNTAX score. The Proportionality of the Cox regression model was assessed by Schoenfeld individual test. All tests were 2‐sided, and statistical significance was set at P<0.05. Missing values were evaluated as null values.

RESULTS

Patient Enrollment and Characteristics

Figure 1A summarizes the patient enrollment. Of the 1724 patients with ACS undergoing emergency PCI, 702 (40.7%) underwent OCT‐guided PCI and were eligible in the TACTICS registry. Of the TACTICS Background registry patients, 72% underwent IVUS‐guided PCI. The main clinical reasons for avoiding OCT were cardiogenic shock (14.5%), lesion anatomy (9.6%), and renal failure (6.9%) (Table S3). Compared with TACTICS registry patients, Background registry patients were older, had more frequent left main disease, and had higher in‐hospital mortality rates (1.0% versus 8.3%, P<0.001) (Table S4).

Figure 1. Flowchart of patient enrollment and prevalence of underlying causes of ACS.

A, Flowchart of patient enrollment; B, Representative OCT images of PR, PE, and CN; C, Prevalence of underlying causes of ACS. ACS indicates acute coronary syndrome; CN, calcified nodule; OCT, optical coherence tomography; PCI, percutaneous coronary intervention; PE, plaque erosion; and PR, plaque rupture.

OCT‐Guided Emergency PCI and PCI Strategy Changes Based on OCT

The procedural results of OCT‐guided emergency PCI are shown in Table S5. For the 702 patients undergoing OCT‐guided emergency PCI, the median onset to door (admission) time was 3.2 (1.3–10.2) hours. The median door to balloon time was 1.4 (1.0–3.2) hours and 1.1 (0.8–1.7) hours for STEMI cases only. TIMI 3 flow grade was achieved in 96.9% of patients. In post‐PCI OCT analysis, minimum lumen area ≥4.5 mm² or ≥70% of the mean reference lumen area was achieved in 81.9% of patients.

Changes in OCT‐based PCI strategy are depicted in Figure 2. Based on OCT, the operator changed the PCI strategy in 408/702 (58.1%) patients. The changes were (1) stentless PCI in 34/702 (4.8%), (2) additional lesion preparation (thrombus aspiration, atherectomy, or excimer laser angioplasty) in 39/702 (5.6%), (3) stent size change in 105/668 (15.7%), (4) stent length change in 107/668 (16.0%), (5) additional stent in 17/570 (3.0%), and (6) additional postdilatation in 311/570 (54.6%) patients. Of those who underwent additional postdilatation, 45% did so due to suboptimal stent expansion and 55% for incomplete stent apposition.

Figure 2. Frequency of OCT‐based PCI strategy changes.

More than 1 strategy change could coexist in each individual patient. OCT indicates optical coherence tomography; and PCI, percutaneous coronary intervention.

Prevalence of Underlying Causes

Of the 702 patients, 7 could not be evaluated because of inadequate imaging (n=5) or lack of culprit lesion on images (n=2). Eventually, 695 patients underwent OCT‐defined assessment of culprit lesion morphology. The prevalence of PR, PE, and CN was 411 (59.1%), 178 (25.6%), and 28 (4.0%), respectively (Figure 1B and 1C). Seventy‐eight patients (11.2%) were classified as “others,” comprising 64 (9.1%) lesions with significant stenosis, 7 (1.0%) coronary spasm, 3 (0.4%) ectasia/aneurysm, 3 (0.4%) embolism, and 1 (0.1%) spontaneous coronary dissection. The interobserver kappa coefficient for PR, PE, and CN was 0.890.

Patient characteristics between PR, PE, and CN are compared in Table 1. Patients with CN were the oldest, followed by those with PR and PE. Male to female ratio differed in CN versus PR and PE. The prevalence of STEMI and post‐PCI peak creatine kinase values were higher in patients with PR versus those with PE and CN. More patients with PE were current smokers than those with PR or CN. Compared with patients with PR and PE, those with CN had higher incidence of comorbidities, including Killip III/IV HF, diabetes, chronic kidney disease, hemodialysis, previous coronary bypass surgery, stroke, and lower creatinine clearance. The 3 groups had similar left ventricular ejection fraction.

Table 1.

Patient Characteristics

	PR	PE	CN	P value	P value
	(n=411)	(n=178)	(n=28)		PR vs PE	PE vs CN	PR vs CN
Age, y	66.5±12.3	63.7±13.4	75.0±11.3	<0.001	<0.001	0.002	0.049
Male	332 (80.8)	150 (84.3)	18 (64.3)	0.042	1.000	0.069	0.190
BMI	24.6±4.0	25.0±4.4	23.0±4.0	0.056
Clinical presentation				0.014	0.015	1.000	0.396
STEMI	299 (72.7)	106 (59.6)	16 (57.1)
NSTEMI	94 (22.9)	58 (32.6)	9 (32.1)
Unstable angina	18 (4.4)	14 (7.9)	3 (10.7)
Killip III/IV	30 (7.3)	5 (2.8)	6 (21.4)	0.001	0.162	0.001	0.068
Hypertension	274 (66.7)	112 (62.9)	20 (71.4)	0.552
Hyperlipidemia	235 (57.2)	104 (58.4)	16 (57.1)	0.960
Use of statin at admission	97 (41.3)	35 (33.7)	11 (68.8)	0.025	0.683	0.048	0.178
Diabetes	135 (32.8)	41 (23.0)	14 (50.0)	0.005	0.066	0.017	0.298
Use of insulin	14 (3.4)	3 (1.7)	2 (7.1)	0.240
Current smoker	143 (34.8)	77 (43.3)	4 (14.3)	0.007	0.190	0.020	0.131
Peripheral artery disease	7 (1.7)	2 (1.1)	1 (3.6)	0.618
Chronic kidney disease*	41 (10.0)	19 (10.7)	11 (39.3)	<0.001	1.000	<0.001	<0.001
Hemodialysis	6 (1.5)	4 (2.2)	6 (21.4)	<0.001	1.000	<0.001	<0.001
Family history of premature coronary artery disease	72 (17.5)	22 (12.4)	2 (7.1)	0.129
Previous myocardial infarction	17 (4.1)	5 (2.8)	2 (7.1)	0.493
Previous PCI	32 (7.8)	9 (5.1)	4 (14.3)	0.175
Previous coronary bypass surgery	2 (0.5)	0 (0.0)	1 (3.6)	0.041	1.000	0.861	1.000
Previous stroke	15 (3.6)	4 (2.2)	7 (25.0)	<0.001	1.000	<0.001	<0.001
Previous cancer	27 (6.6)	4 (2.2)	1 (3.6)	0.087
LVEF, %	55.2±10.3	55.7±10.5	56.8±8.8	0.706
Laboratory data at admission
Creatinine clearance, mL/min	81.1±34.3	89.6±42.2	45.7±29.1	<0.001	<0.001	<0.001	0.055
LDL cholesterol, mg/dL	130±40	130±37	103±36	0.002	0.002	0.002	1.000
HDL cholesterol, mg/dL	50±15	49±16	52±15	0.638
Triglyceride, mg/dL	137±96	148±119	103±52	0.082
HbA1c, %	6.4±1.2	6.3±1.1	6.6±1.3	0.568
BNP, pg/mL	107±180	116±185	348±436	<0.001	1.000	0.060	0.076
Peak CK, U/L	1518±2137	1092±1784	648±1081	0.010	0.223	0.001	0.038
Medication at discharge				0.706
Dual antiplatelet therapy	386 (93.9)	168 (94.4)	24 (88.9)	0.538
Single antiplatelet therapy†	24 (5.8)	8 (4.5)	2 (7.4)	0.732
Anticoagulation therapy	33 (8.0)	12 (6.7)	2 (7.4)	0.863
Statin	405 (98.5)	174 (97.8)	26 (96.3)	0.597
ACE‐I/ARB/ARNI	313 (76.2)	131 (73.6)	15 (55.6)	0.056
β‐blocker	291 (70.8)	129 (72.5)	18 (66.7)	0.803

Values are mean±SD or n (%). ACE‐I indicates angiotensin‐converting enzyme inhibitor; ARB, angiotensin II receptor blocker; ARNI, angiotensin receptor neprilysin inhibitor; BMI, body mass index; BNP, brain natriuretic peptide; CN, calcified nodule; CK, creatine kinase; HbA1c, glycosylated hemoglobin; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein; LVEF, left ventricular ejection fraction; NSTEMI, non‐ST‐segment–elevation myocardial infarction; PCI, percutaneous coronary intervention; PE, plaque erosion; PR, plaque rupture; and STEMI, ST‐elevation myocardial infarction.

^*Defined as estimated glomerular filtration rate <60 mL/min per 1.73 m² for ≥3 mo.

^† All patients on single antiplatelet therapy had concomitant anticoagulation.

Angiographic and Procedural Characteristics

Angiographic and procedural characteristics are summarized in Table 2. PR was observed almost equally in the left anterior descending and right coronary arteries, PE was evident predominantly in the left anterior descending artery, and more than half of the observed CNs were seen in the right coronary artery. Patients with CN had higher proportions of type B2/C culprit lesions and pre‐PCI SYNTAX scores versus those with PR and PE. Compared with patients with PR and PE, those with CN required more pre‐OCT small balloon dilatation, underwent more frequent stentless procedures, and experienced longer procedure times. While the final TIMI 3 flow achievement rate was similar in the 3 groups, the post‐PCI SYNTAX score was highest in patients with CN, followed by those with PR and PE.

Table 2.

Angiographic and Procedural Characteristics

	PR	PE	CN	P value	P value
	(n=411)	(n=178)	(n=28)		PR vs PE	PE vs CN	PR vs CN
Culprit vessel				0.001	0.006	0.009	0.198
Left main	2 (0.5)	1 (0.6)	1 (3.6)
Left anterior descending artery	198 (48.2)	112 (62.9)	12 (42.9)
Left circumflex	38 (9.2)	20 (11.2)	0 (0.0)
Right	173 (42.1)	45 (25.3)	15 (53.6)
Type B2/C	270 (65.7)	104 (58.5)	23 (82.2)	0.032	0.336	0.086	0.342
TIMI flow				0.101
0	187 (45.5)	70 (39.3)	9 (32.1)
1	64 (15.6)	21 (11.8)	3 (10.7)
2 or 3	160 (38.9)	87 (48.9)	16 (57.1)
Pre‐PCI SYNTAX score	14.0±8.2	13.0±7.5	20.4±13.6	<0.001	0.387	0.030	0.071
Any intervention before initial OCT
None	118 (28.7)	79 (44.4)	9 (32.1)	0.001	0.001	0.935	1.000
Thrombus aspiration	206 (50.1)	65 (36.5)	6 (21.4)	<0.001	0.009	0.533	0.018
Small balloon (≤2 mm) dilatation	128 (31.1)	45 (25.3)	15 (53.6)	0.009	0.545	0.014	0.075
Atherectomy	0 (0.0)	0 (0.0)	4 (18.2)	<0.001	<0.001	0.004	NA
Number of stent(s)	1.1±0.4	1.1±0.4	1.0±0.7	0.245
Stentless	3 (0.7)	4 (2.2)	5 (17.9)	<0.001	0.755	0.003	<0.001
Postdilatation	262 (96.7)	109 (96.5)	18 (94.7)	0.904
Maximum balloon size, mm	3.54±0.60	3.47±0.66	3.36±0.67	0.360
Duration of PCI procedure, min	73.5±40.5	69.3±39.1	105.3±45.7	<0.001	0.001	0.004	0.714
TIMI 3 flow grade after PCI	394 (95.9)	173 (98.3)	28 (100.0)	0.192
Post‐PCI SYNTAX score	4.5±6.5	3.0±4.8	10.9±12.2	<0.001	0.006	0.008	0.036

Values are the mean±SD or n (%). CN indicates calcified nodule; OCT, optical coherence tomography; PCI, percutaneous coronary intervention; PE, plaque erosion; PR, plaque rupture; and TIMI, Thrombolysis in Myocardial Infarction.

OCT Findings

Table 3 summarizes detailed OCT findings from culprit lesions. Lipid‐rich plaques (maximum lipid arc ≥180°), thin‐cap fibroatheroma, macrophage accumulation, and cholesterol crystals were frequently observed in PR lesions. Extensively calcified plaques (maximum calcium arc ≥180°) were rarely seen with PR and PE but were observed with all CNs (100%). Layered plaques were most common in PE lesions, followed by PR and CN. The stent length, minimum lumen area after PCI, and stent expansion were similar in the 3 groups, while the stent eccentricity index was lower and the frequency of stent edge dissection was higher in CN lesions and the maximum in‐stent tissue protrusion area was higher in PR versus that in the other groups.

Table 3.

OCT Findings

	PR	PE	CN	P value	P value
	(n=411)	(n=178)	(n=28)		PR vs PE	PE vs CN	PR vs CN
Findings before stenting
Lipid plaque	411 (100.0)	142 (79.8)	10 (35.7)	<0.001	<0.001	<0.001	<0.001
Maximum lipid arc ≥180°	376 (91.5)	93 (52.2)	5 (17.9)	<0.001	<0.001	0.004	<0.001
TCFA	340 (82.7)	46 (25.8)	5 (17.9)	<0.001	<0.001	1.000	<0.001
Calcification	254 (61.8)	102 (57.3)	28 (100.0)	<0.001	1.000	<0.001	<0.001
Maximum calcium arc ≥180°	35 (8.5)	18 (10.1)	28 (100.0)	<0.001	1.000	<0.001	<0.001
Thrombus	397 (96.6)	178 (100.0)	27 (96.4)	0.044	0.084	0.861	1.000
Macrophage accumulation	400 (97.3)	157 (88.2)	19 (67.9)	<0.001	<0.001	0.032	<0.001
Cholesterol crystals	251 (61.1)	88 (49.4)	9 (32.1)	0.001	0.034	0.400	0.015
Microchannel	194 (47.2)	94 (52.8)	8 (28.6)	0.050	0.738	0.088	0.257
Layered plaque	284 (69.1)	144 (80.9)	10 (35.7)	<0.001	0.013	<0.001	0.002
Final findings
Stent length, mm	25.6±10.0	24.6±10.0	29.2±13.5	0.113
Mean reference lumen area, mm2	8.67±3.49	8.49±3.48	7.54±3.20	0.308
Minimum lumen area, mm2	5.88±2.10	5.83±2.24	5.52±2.59	0.746
Minimum lumen area divided by the mean reference lumen area, %	70.8±17.3	71.5±16.1	74.3±18.4	0.608
Stent expansion, %	75.3±18.5	75.2±16.6	74.0±17.8	0.944
Stent eccentricity index	0.82±0.06	0.82±0.07	0.71±0.08	<0.001	1.000	<0.001	<0.001
Reference lumen narrowing	191 (49.2)	76 (46.1)	14 (63.6)	0.292
Maximum in‐stent tissue protrusion area, mm2	1.25±0.74	0.98±0.62	0.80±0.33	<0.001	<0.001	0.104	<0.001
Maximum stent malapposed area, mm2	0.91±0.75	0.91±0.72	1.15±0.57	0.315
Stent edge dissection
Proximal*	49 (12.8)	19 (11.6)	7 (35.0)	0.013	1.000	0.002	0.008
Distal†	28 (7.3)	8 (5.0)	6 (28.6)	0.001	1.000	0.038	0.040

Values are mean±SD or n (%). CN indicates calcified nodule; OCT, optical coherence tomography; PE, plaque erosion; PR, plaque rupture; and TCFA, thin‐cap fibroatheroma.

^*OCT images were available in 568 lesions (PR, 384, PE, 165, CN, 20).

^† OCT images were available in 568 lesions (PR, 384, PE, 164, CN, 19).

Clinical Outcomes

Table 4 summarizes patient outcomes. Clinical follow‐up data were available for 609 of 617 patients (99.0%). The median follow‐up period was 12.7 months (interquartile range, 12.0–13.6). One‐year MACE occurred most frequently in patients with CN (32.1%), followed by those with PR (12.4%) and PE (6.2%) (log‐rank P<0.0001, Figure 3), primarily driven by increased cardiovascular death (CN, 25.0%; PR, 0.7%; PE, 1.1%; log‐rank P<0.0001) and a trend for HF (CN, 7.1%; PR, 6.8%; PE, 2.2%; log‐rank P<0.075) (Figure S1). Bleeding Academic Research Consortium 2, 3, or 5 bleeding was most common in patients with CN (log‐rank P=0.047). Figure 4 depicts the unadjusted and adjusted Cox proportional HR values for MACEs. The proportionality of the Cox regression model was observed in each outcome (Table S6). After multivariate Cox regression analysis (Table S7), the underlying ACS cause (CN [HR, 4.49 [95% CI, 1.35–14.89], P=0.014]; PR [HR, 2.18 [95% CI, 1.05–4.53], P=0.036]; PE as reference), hemodialysis (HR, 3.2 [95% CI, 1.07–9.61], P=0.038), left ventricular ejection fraction (HR, 0.96 [95% CI, 0.93–0.98], P<0.001), and post‐SYNTAX score (HR, 1.03 [95% CI, 1–1.07], P=0.046) were noted as independent predictors of MACE at 1 year.

Table 4.

Clinical Outcomes at 1 Year

	PR	PE	CN
	(n=411)	(n=178)	(n=28)
Major adverse cardiac events	51 (12.4)	11 (6.2)	9 (32.1)
Cardiovascular death	3 (0.7)	2 (1.1)	7 (25.0)
Myocardial infarction	8 (1.9)	3 (1.7)	1 (3.6)
Heart failure	28 (6.8)	4 (2.2)	2 (7.1)
Inpatient	21 (5.1)	4 (2.2)	2 (7.1)
Postdischarge	7 (1.7)	0 (0.0)	0 (0.0)
Revascularization	20 (4.9)	7 (3.9)	2 (7.1)
Target lesion revascularization	10 (2.4)	2 (1.1)	1 (3.6)
Nontarget lesion revascularization	10 (2.4)	5 (2.8)	1 (3.6)
Other events
All‐cause death	12 (2.9)	3 (1.7)	8 (28.6)
Stroke	5 (1.2)	2 (1.1)	1 (3.6)
Stent thrombosis	3 (0.7)	2 (1.1)	0 (0.0)
Bleeding	20 (4.9)	4 (2.2)	3 (10.7)

Values are mean±SD or n (%). BARC indicates Bleeding Academic Research Consortium; CN, calcified nodule; PE, plaque erosion; and PR, plaque rupture.

Figure 3. Kaplan–Meier time‐to‐event curve for the cumulative MACE rate.

CN indicates calcified nodule; HR, hazard ratio; MACE, major adverse cardiac events; PE, plaque erosion; and PR, plaque rupture.

Figure 4. Unadjusted and adjusted hazard ratios for MACE.

A, Unadjusted hazard ratios for MACE; B, Adjusted hazard ratios for MACE. CN indicates calcified nodule; HR, hazard ratio; MACE, major adverse cardiac events; PE, plaque erosion; and PR, plaque rupture.

DISCUSSION

This was a prospective multicenter registry enrolling patients with ACS (within 24 hours of symptoms) who underwent OCT‐guided emergency PCI to evaluate the underlying ACS causes, their impact on clinical outcomes, and the penetration and feasibility of OCT‐guided PCI. Our main findings were (1) OCT‐guided emergency PCI was performed in 40.7% of cases; (2) OCT‐guided PCI led to PCI strategy change in 58.1% of cases; (3) the prevalence of underlying ACS causes in acute phase was recorded (PR, 59%, PE, 26%, CN, 4%); and (4) the underlying ACS causes defined by OCT were associated with 1‐year MACE, which was highest among patients with CN, followed by those with PR and PE.

Prevalence and Patient Characteristics According to OCT‐Guided PCI for ACS

According to a Korean registry, imaging‐guided PCI was performed for 28.7% of patients with acute MI in 2017 (OCT guidance, 4.7%). ⁹ Use of IVUS and OCT for acute MI in the United States (2004–2014) was reported as 2.6% and 0.1%, respectively. ²⁴ Although a clear disparity exists in the proportion of intravascular imaging used in different countries, this proportion has increased every year in each cohort. ⁹ , ²⁴ While a previous Japanese registry ²⁵ from 2008 to 2014 reported that imaging guidance (mainly with IVUS) was utilized in 40% of cases of emergency PCI for ACS, in the current Japanese ACS registry from 2019 to 2021, imaging guidance was used in 84% of cases, and IVUS and OCT were applied equally. Yet, critically ill patients, such as those with cardiogenic shock, left main lesions, or renal failure, were more likely to be treated with IVUS‐ or angio‐guided PCI than with OCT. OCT guidance appears to be growing more common in real‐world settings of emergency PCI for patients with ACS without serious comorbidities.

OCT‐Based PCI Strategy Changes and Acute Results

Intravascular imaging facilitates PCI optimization by evaluating lesion morphology, optimizing stent implantation, and assessing postprocedural results. ⁷ As OCT has become generalized for ACS, PCI optimization benefits have been reported regarding larger balloon/stent use leading to increased luminal gains ²⁶ and higher postprocedural fractional flow reserves ²⁷ versus those by angiography‐guided PCI alone. In this registry, OCT guidance helped PCI operators to change PCI strategy in 58.1% of cases, and prespecified imaging goals were achieved in 81.9% of cases without compromising TIMI III flow restoration. This might be attributable to additional postdilatation by observing OCT images after stent implantation. The contribution of OCT to PCI optimization in patients with ACS needs further analysis in future studies.

Underlying Causes of ACS: Characteristics

Previously reported OCT‐based average frequencies of ACS causes were 50% for PR, 37% for PE, and 3% for CN. ⁴ Our data confirmed these findings with frequencies of 59% for PR, 26% for PE, and 4% for CN, which included only patients with acute onset (symptoms within 24 hours). However, in our registry, we observed a relatively high PR frequency. Higher PR frequency has been reported in patients without preceding symptoms (pre‐infarction angina) ²⁸ because not all PRs provoke clinical ACS, and some develop healed plaques and significant area stenosis. This may worsen angina and lead to delayed subsequent ACS onset, ²⁹ which could explain our higher observed PR prevalence. From this registry, with enrollment limited to patients with acute‐phase disease, our data may reflect a more precise prevalence of ACS causes.

Compared with patients with PR, those with PE were younger, more likely current smokers, less likely to have diabetes, and preponderant in the left anterior descending artery, which corresponds with previously reported findings. ³⁰ Patients with CN were older and had more comorbidities, similar to the results of a previous IVUS‐based study. ³¹ Moreover, our data showed higher pre‐ and post‐PCI SYNTAX scores in patients with CN compared with those without, implying that patients with CN had a predisposition to severe atherosclerosis at both the patient and lesion levels.

Underlying Causes of ACS and Clinical Outcome

Although adequate PCI optimization was achieved equally in this study, the 1‐year clinical outcomes differed between patients with CN, PR, and PE. Since OCT introduction into clinical ACS settings, PR has been focused upon as a main cause of ACS, and its relationship to clinical outcomes has been addressed. ⁵ In this study, patients with PR presented with a large proportion of STEMI and the highest peak creatine kinase elevation, which might have led to the development of inpatient and postdischarge HF. ³² In addition, PR has been strongly correlated with pancoronary inflammation and high thin‐cap fibroatheroma prevalence, ³³ which may produce future events.

According to previous reports, CN is a rare but important cause of ACS with the worst clinical outcomes, ³¹ , ³⁴ which showed that the higher incidence of MACE in patients with CN was mainly driven by target lesion revascularization and ACS recurrence. Additionally, in the analysis of nonculprit left anterior descending artery (from both patients with ACS and stable angina), ³⁵ CN was associated with a high 1‐year incidence of cardiac death (13.3%) and target lesion MI (6.7%). In this study, patients with CN developed 1‐year MACE as high as 32.1%, including 25% of cardiovascular deaths; this was significantly worse than with PR and PE. Patients with CN had more advanced atherosclerosis, with the highest pre‐ and post‐PCI SYNTAX scores, and concomitant comorbidities, including advanced age, diabetes, chronic kidney disease, hemodialysis, and Killip III/IV HF, which was likely to relate to poor clinical outcomes. However, confounder‐adjusted multivariate analysis found that OCT‐defined CN in patients with ACS was a robust marker to stratify risks for a cardiac event postemergency PCI. Although we did not observe increased target lesion revascularization in CN versus in other lesions, some cardiovascular deaths might have occurred without a target lesion revascularization event or clear evidence of stent thrombosis or MI, even if caused by the originally treated lesion factor. Considering previous reports regarding high target lesion revascularization rates in CN lesions, we assume that the poor prognosis of this small cohort of patients was probably due to both lesion‐ and patient‐related factors. In this study, we used relatively conservative imaging goals (ie, >70% lumen area of mean reference lumen area) in order to secure TIMI III flow recanalization. The optimization strategy for each underlying cause of ACS remains uncertain, and therefore, further studies are warranted to evaluate the clinical impact of OCT‐guided PCI optimization and subsequent tailored medical management according to the underlying ACS cause.

Study Limitations

There are several limitations to the precise interpretation of this registry's results. First, this was an observational study, and the indication for OCT‐guided PCI was at the operator's discretion. Thus, selection bias was possible. OCT‐guided PCI was less likely to be performed in patients who were older, with cardiogenic shock or renal failure, and with left main coronary artery lesions. Additionally, long‐term clinical follow‐ups were not collected for patients without OCT‐guided PCI. Therefore, the study results may not be applicable to all patients with ACS. Second, for prompt recanalization, thrombus aspiration or gentle predilatation using a balloon ≤2.0 mm was allowed before initial OCT; this may have affected the precise assessment of lesion morphology. Third, OCT‐based assessment of underlying ACS causes was not supported by the histological definition of these mechanisms. Fourth, the direct impact of the diagnosis of the underlying cause of ACS during PCI by each operator on PCI strategy was not investigated. Fifth, while ischemia‐driven revascularization was prespecified, the possibility of occulostenotic reflex due to follow‐up angiography was not completely eliminated. Sixth, we carefully selected covariates for the multivariable Cox proportional hazards regression models; however, confounding bias cannot be entirely eliminated.

CONCLUSIONS

This multicenter ACS registry found that the penetration rate of OCT use in emergency PCI was 40% and showed the specific prevalence of underlying cause of ACS in acute‐phase patients. Using OCT‐based assessment of culprit lesions enables the stratification of future risks of MACE. The contribution of OCT guidance to PCI optimization in each ACS cause warrants further study.

Sources of Funding

This study was sponsored by Abbott Medical Japan LLC (Minato‐ku, Tokyo, Japan). Other than financial sponsorship, the company had no role in study protocol development or implementation, management, data collection, or analysis. The authors and colleagues were solely responsible for the design and execution of this study.

Disclosures

Dr Takuya Mizukami received consultancy fees from Zeon Medical Inc., research grants from Boston Scientific, and speaker fees from Abbott Vascular, Cathworks, and Boston Scientific. Dr Mamoru Nanasato received lecture fees from Boston Scientific. Dr Ken Kozuma received lecture fees and research funds from Abbott Medical Japan LLC and Boston Scientific. Dr Junichi Yamaguchi was endowed by Abbott Medical Japan LLC, Boston Scientific, Medtronic, and Terumo. Dr Hiroshi Suzuki received grant support from Abbott Medical Japan LLC and Daiichi‐Sankyo. Dr Hiromasa Otake received lecture fees from Abbott Medical Japan LLC and Terumo. Dr Masahiko Ochiai received lecture fees from Abbott Medical Japan LLC, Asahi Intecc, Boston Scientific, and Terumo. Toshiro Shinke received personal fees and research grants from Abbott Medical Japan LLC. The remaining authors have no disclosures to report.

Supporting information

Data S1

Table S1–S7

Figure S1