Characteristics and Pattern of Calcified Nodule and/or Nodular Calcification Detected by Intravascular Ultrasound on the Device-Oriented Composite Endpoint (DoCE) in Patients with Heavily Calcified Lesions Who Underwent Rotational Atherectomy-Assisted Percutaneous Coronary Intervention

Objectives This study aimed to determine characteristics and pattern of a calcified nodule (CN) and/or nodular calcification (NC) detected by intravascular ultrasound (IVUS) on the device-oriented composite endpoint (DoCE) in patients with calcified lesions who underwent rotational atherectomy (RA)-assisted percutaneous coronary intervention (PCI). Background The characteristics and pattern of a CN and/or NC on clinical outcome remain unknown. Methods We retrospectively enrolled patients who underwent RA-assisted PCI at Siriraj Hospital during August 2016 to April 2020. Preprocedural IVUS imaging was mandatory. CN/NC was defined as convex shape of luminal surface and luminal side of calcium with protrusion into the coronary artery lumen as assessed by IVUS. The primary outcome was cumulative of DoCE, defined as the composite of cardiovascular death, myocardial infarction, and clinically-driven target lesion revascularization. Results Two hundred patients were included. Primary outcome occurred in 14%. The cumulative DoCE was significantly higher in the CN/NC group than that in the non-CN/NC group (20.7% vs. 8.8%, p = 0.022). CN/NC (p = 0.023) and MSA ≤ 5.5 mm2 (p = 0.047) were correlated with a significantly higher cumulative DoCE. CN/NC was the independent predictor for the cumulative DoCE (HR = 2.96, 95% CI 1.08–8.11, p = 0.035). Pattern and characteristic of CN/NC have a prognostic value. Patients with an eccentric CN/NC had a significantly higher cumulative DoCE compared to those CN/NC with concentric calcification (p = 0.014). Conclusion The presence of a CN/NC in patients with heavily calcified lesions who underwent RA-assisted PCI was found to be associated with increased cumulative 5 year DoCE, especially in patients with an eccentric CN/NC. The clinical trial is registered with TCTR20210616001.


Introduction
A calcifed nodule (CN) is defned as an eruptive nodular calcifcation that protrudes into the coronary artery lumen. Eruptive CN is the cause of acute coronary syndrome (ACS) in about 2-8% of ACS patients [1]. Te outcomes of patients with a CN in native coronary artery disease (CAD) were frst reported by Xu et al. [2]. Nodular calcifcation (NC) is benign in nature that does not lead to coronary thrombosis, but it could limit the device cross ability, stent strut damage, stent malapposition, and stent under expansion. To diferentiate between these 2 pathological entities is very difcult. In the setting of acute coronary syndrome (ACS), the appearance of CN at the culprit lesion is suggestive of eruptive CN. Nonculprit CN was observed in 30% of ACS. Morofuji et al. found a 41% prevalence of CN in patients with heavily calcifed lesion that required rotational atherectomy (RA)assisted percutaneous coronary intervention (PCI) [3]. Tey also reported CN to be associated with an unfavorable post-PCI outcome and signifcantly increased incidence of 5 year major adverse cardiovascular event (MACE). However, they did not identify the characteristics of CN/NC that associate with adverse clinical outcome. Plaque modifcation is an important step when managing a heavily calcifed lesion, and plaque modifcation in a CN/NC can be a challenge. Intravascular ultrasound (IVUS) has a signifcant role in assisted coronary and noncoronary intervention [4][5][6][7][8][9][10][11][12]. Intravascular imaging guidance reduce 1 year MACE in the patients with heavy calcifed lesion undergoing RA-assisted PCI [13]. Information obtained from intravascular ultrasound (IVUS) specifc to the presence and characteristics of a CN/NC within a heavily calcifed lesion that impact on device-oriented composite endpoint (DoCE) will improve patient management and clinical outcomes.
Te aim of this study was to determine the impact of a CN/NC detected by IVUS on the DoCE in patients with heavily calcifed lesions who underwent RA-assisted PCI. Our secondary objective was to determine the characteristics of a CN/NC identifed by IVUS that associate with unfavorable clinical outcomes.

Study Design.
Tis single-center retrospective cohort study was conducted at the Interventional Cardiology Unit of the Division of Cardiology, Department of Medicine, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Tailand. Two hundred and ninety-three consecutive patients who underwent RA-assisted IVUS-guided PCI to treat a moderate to severe calcifed lesion during August 2016 to April 2020 were evaluated. Preprocedural IVUS was mandatory, and preprocedural IVUS imaging could have been performed either before or after RA. Patients with signifcant valvular heart disease or severe comorbidities, such as malignant neoplasm requiring chemotherapy, surgery, or radiation, were excluded. We also excluded patients with cardiogenic shock before procedure, and patients who were lost to follow-up. Two hundred patients who met inclusion and exclusion criteria were enrolled. Te Siriraj Institutional Review Board approved this study (COA no. 220/2021).
Histological section classifcation of calcifed nodule was described by Virmani et al. [14,15]. Calcifed nodule (CN) is defned as a lesion with fbrous cap disruption from eruptive calcifc nodules associated with an occlusive or nonocclusive platelet/fbrin thrombus. Nodular calcifcation (NC) is the least common form of calcifcation in the coronary vasculature. NC is accompanied by fbrin with a thick, intact fbrous cap. In patients with heavily calcifed lesions undergoing RA, the advancement of the IVUS catheter before intervention is impossible. Te resolution of IVUS in a detected fbrous cap disruption is limited, especially in IVUS with lower MHz. Te study objective is focused on impact of these CN/NC on DoCE in patients with heavily calcifed lesions who underwent RA-assisted PCI. We opt to combine these two entities for analysis. CN/NC was defned as convex shape of luminal surface and luminal side of calcium with protrusion into the coronary artery lumen [1,16]  An experienced intravascular imaging technician and experienced interventionists who were blinded to the clinical outcome reviewed the angiographic and IVUS imaging data. Intraobserver and interobserver variability yielded good concordance for the diagnosis of CN (k � 0.95 and k � 0.90, respectively).

Angiographic Analysis.
Qualitative and quantitative analysis was performed by an experienced intravascular imaging technician and 2 experienced interventional cardiologists who were blinded to the clinical information. Quantitative coronary angiography (QCA) analysis was performed using a cardiovascular measurement system (QAngio XA 7.2, MEDIS, Leiden, Te Netherlands). Coronary angiograms were performed in at least 2 orthogonal views. Angiographic calcifcation grading was as follows: (1) none: no radiopacity; (2) mild calcifcation: faint radiopacities noted during the cardiac cycles; (3) moderate calcifcation: dense radiopacities noted only during the cardiac cycle; and (4) severe calcifcation: dense radiopacities noted without cardiac motion before contrast injection generally compromising both sides of the arterial lumen.

IVUS Quantitative Analysis.
Qualitative and quantitative analysis was performed at the target lesion according to and following an expert consensus document on the standards for acquisition, measurement, and reporting of intravascular ultrasound studies [17].
(1) Maximum calcium arch grade at the lesion is as follows: (i) Eccentric calcifcation was defned as a calcium arch <180 degrees (ii) Concentric calcifcation was defned as a calcium arch ≥180 degrees (2) Calcium location is as follows: (i) Superfcial calcifcation was defned as the leading edge of the acoustic shadowing appears within the shallowest 50% of the plaque plus media thickness (ii) Deep calcifcation was defned as the leading edge of the acoustic shadowing appears within the deepest 50% of the plaque plus media thickness (3) Calcium characteristics are as follows: (i) Napkin ring calcifcation was defned as severe circumferential calcifcation. (ii) Eccentric calcifed nodule/nodular calcifcation was defned as calcifed nodule/nodular calcifcation without calcifcation at the opposite site of calcifed nodule/nodular calcifcation. (iii) Calcifed nodule/nodular calcifcation with concentric calcifcation was defned as calcifed nodule/nodular calcifcation with calcifcation at the opposite site of calcifed nodule/nodular calcifcation. Mix lesion was identifed as concentric calcifcation. (iv) Reverberation was defned as an artifact represented by secondary false echoes of the same structure and reverberation from the leading edge of calcium [18]. (v) Calcium fracture was defned as a gap of calcium and direct exposure of calcium to the lumen at the gap [19]. Lumen measurements were performed using the interface between the lumen and the leading edge of the intima. (vi) Lumen cross-sectional area (CSA) was defned as the area bounded by the luminal border.
(4) Reference vessel is as follows: (i) Te proximal and distal reference segments with the maximum lumen and least amount of plaque within 5 mm proximal or distal to the lesion (4) Stent measurement is as follows: (i) Minimal stent area (MSA) was defned as the minimal area bounded by the stent border (ii) Minimum stent diameter was defned as the shortest diameter through the center of mass of the stent (iii) Maximum stent diameter was defned as the longest diameter through the center of mass of the stent (iv) Asymmetry index [20] was defned as follows: ((maximum stent diameter minus minimum stent diameter) divided by maximum stent diameter) measured at the MSA (v) Asymmetrical stent expansion was defned as a stent symmetry index >0.3 (vi) Symmetrical stent expansion was defned as a stent symmetry index ≤0.3

PCI Procedure and Clinical Follow-Up.
All patients received a bolus injection of heparin 100 unit/kilogram to maintain an activated clotting time of >250 seconds. Dual antiplatelet therapy with 81 mg/day aspirin with 75 mg/day clopidogrel or 10 mg/day prasugrel or 90 mg twice a day ticagrelor was continued for at least one year after the procedure. Te conventional 0.014 inch guidewire was replaced with a 0.009 inch ROTAWire ™ foppy guidewire (Boston Scientifc) or a 0.009 inch ROTAWire ™ extra support guidewire (Boston Scientifc). Te initial burr size selection was based on preprocedural IVUS imaging indicating whether or not the lesion can be passed before RA. A burr speed of 140,000-200,000 revolutions per minute (RPM) with a run duration of 10-15 seconds was used. Te target fnal burr-to-artery ratio was within 0.4 to 0.6. Noncompliance balloon or cutting balloon was routinely used for predilatation before stent implantation. Te stent diameter was determined by measuring the external elastic lamina diameter at the proximal and distal reference sites. Postprocedure imaging was performed after stent implantation to evaluate stent apposition, stent optimization, and procedural complication, such as dissection or tissue protrusion. Duration and type of antiplatelet and antithrombotic were prescribed according to operator discretion.

Study Outcome.
Te primary outcome was the deviceoriented composite endpoint (DoCE), defned as the composite of cardiovascular death, myocardial infarction, and clinically-driven target lesion revascularization.

Statistical Analysis.
Te sample size was calculated using data from a study by Morofuji, et al. [3]. Tey reported an event rate of 24% in the CN group, and an 8.3% event rate in the non-CN group. We estimated that a sample size of 170 patients would give our study 80% power to detect a 15% diference in DoCE between the CN/NC group and non-CN/ NC group. Continuous variables were reported as median (interquartile range 25-75) or mean ± standard deviation depending on the distribution of data. Categorical data were reported as number and percentage. Chi-square test or Fisher's exact test was used to compare categorical data, and Student's t-test (normally distributed data) or Mann-Whitney U test (non-normally distributed data) was used to compare continuous data. Kaplan-Meier survival analysis was used to estimate the cumulative DoCE and Cox proportional hazard ratio was used to identify independent predictors of the cumulative DoCE. Factors with p value <0.2 after using a univariate model were included in multivariable Cox proportional hazard analysis. A p value less than 0.05 was considered to be statistically signifcant. All statistical analyses were performed using SPSS Statistics version 18 (SPSS, Inc., Chicago, IL, USA).

. Results
Te study reviewed 293 consecutive patients who underwent RA-assisted IVUS-guided PCI at the Faculty of Medicine, Siriraj Hospital, during August 2016 to April 2020. Two hundred patients who met inclusion and exclusion criteria were enrolled. A fow diagram showing the patient enrollment protocol is shown in Figure 1.

Baseline Characteristics.
Te prevalence of history of hypertension was signifcantly higher in the CN/NC group than in the non-CN/NC group. No other baseline characteristics were signifcantly diferent between groups. Fortythree percent of overall patients had chronic kidney disease. Te baseline characteristics of all patients and comparison between CN/NC and non-CN/NC patients are shown in Table 1.

Lesion and Procedure
Characteristics. Seventy-seven percent of overall patients had angiographic severe calcifcation. Te majority of patients had multivessel disease, and 71% of procedures were performed in the left anterior descending (LAD) artery. Lesion and procedural characteristics are summarized in Table 2. Final burr size, burr-toartery ratio, stent diameter, number of stents, and stent length were not signifcantly diferent between groups. Periprocedural complications (Table 2) occurred in 3.0% of overall patients, and they occurred more frequently in the CN group (4.6% vs. 1.8%, p � 0.407). IVUS analysis showed superfcial calcifcation and napkin ring calcifcation in 98% and 50% of overall cases, respectively. CN/NC was found in 43.5% (87 patients) of overall patients. Regarding calcifcation characteristics, superfcial calcifcation, concentric calcifcation, napkin ring calcifcation, and calcium fracture were not signifcantly diferent between the CN/NC and non-CN/NC groups. For lesion subtypes, concentric calcifcation without CN/NC were found most frequently in 102 patients (51%) followed by CN/NC with concentric calcifcation (80 patients, 40%). Eccentric calcifcation was found in 18 patients (9%). Eccentric CN/NC were the least common subtypes of calcifed plaque in this study (7 patients, 3.5%). Poststent IVUS analysis showed no signifcant diference in mean MSA between groups. Asymmetrical stent expansion was more frequently observed (numerically) in the CN/NC group (9.3% vs. 1.4%, p � 0.084) ( Table 3).
Wire bias based on IVUS data and DoCE in the patients with eccentric CN/NC is shown in Table 4. Te patient who had wire bias toward eccentric CN/NC had a good prognosis after RA-assisted PCI. In contrast, the patient who had wire bias opposite to eccentric CN/NC had increased 5 year DoCE. Sixty percent of the patients with eccentric CN/NC and wire bias opposite to CN/NC had 5 year DoCE.

Discussion
Te present study investigated the impact of a CN/NC on the DoCE in patients with a heavily calcifed lesion who underwent RA-assisted IVUS-guided PCI.   calcium score using (1) superfcial calcium angle >270°l onger than 5 mm, (2) 360°of superfcial calcium, (3) calcifed nodule, and (4) vessel diameter <3.5 mm [21]. Tese parameters of IVUS derived calcium score associated with stent underexpansion and requiring plaque modifcation such as RA before stent implantation. Even though napkin ring calcifcation is a predictor of stent under expansion [22] and an indication for plaque modifcation, such as RA, napkin ring calcifcation or superfcial calcium >270°longer than 5 mm has no impact on the long-term DoCE in our study because we performed RA in all patients. In the setting of RA-assisted PCI, we found the calcifed nodule and MSA < 5.5 mm 2 correlated with a signifcantly higher incidence of the cumulative DoCE. MSA ≤ 5.5 mm 2 is also the cut point for predicting angiographic in-stent restenosis in the DES era [23]. However, after multivariate analysis, the CN/NC was an independent predictor for the cumulative DoCE.
In our study, the prevalence of MSA ≤ 5.5 mm 2 and stent expansion were not signifcantly diferent between the CN/NC and non-CN/NC groups, but TVR was signifcantly higher in the CN/NC group [24]. CN also has known to increase the incidence of TVR. Nakamura et al. proposed the potential mechanism of progression of instent calcifed nodule to be a protrusion of calcifed nodule followed by calcifying fbrin thrombus in the early phase, fragments of calcifcation and fbrin deposition caused by mechanical destruction of sheath of calcifcation in midphase, and neocalcifed nodule in the late phase [25].   Asymmetrical stent expansion was also numerically higher in the CN/NC group. Suwannasom et al. [20] found asymmetrical stent expansion to be the independent factor associated with device-oriented composite endpoint (DoCE).
In the present study, we used pre-and post-PCI IVUS analysis to investigate pattern and characteristic of a CN/NC in predicting the cumulative DoCE. Te pattern and distribution of a CN/NC in relation to surrounding calcifed lesion obtained by prestent IVUS analysis helps in risk prediction. Pre-stent IVUS analysis showed an eccentric CN/NC to be associated with a higher cumulative DoCE compared to a CN/NC with concentric calcifcation. Te presence of concentric calcifcation or napkin ring calcifcation with a CN/NC would allow the CN/NC to be adequately ablated during RA. In contrast to an eccentric CN/ NC, RA would ablate in the opposite side of CN/NC particularly in the presence of vessel tortuosity or wire bias. Tis mechanism could afect suboptimal contact of RA burr to the CN/NC leading to inadequately plaque modifcation/ lesion preparation specifcally in an eccentric CN/NC lesion. Based in IVUS data in this study, we found that the patient with eccentric CN/NC and wire bias toward CN/NC still had favorable prognosis after RA-assisted PCI. Te patient with eccentric CN/NC and wire bias opposite to CN/NC had increased 5 year DoCE. Te character of wire bias opposite to CN/NC in eccentric CN/NC was probable unsuitable anatomy for RA-assisted PCI. We suggested paying attention on wire bias characteristics based on IVUS data in the patients with eccentric CN/NC in order to select appropriate atherectomy device during the procedure. Stent implantation in an eccentric CN/NC will preferentially limit expansion on calcifed side and overexpansion on the opposite noncalcifed site causing stent malapposition/asymmetrical stent expansion. Using of the high-pressure noncompliance or semicompliance balloon postdilatation may increase risk   Our study provides novel insights on the prognostic value of diferent coronary plaque calcifcation phenotypes. Our fndings suggest that careful attention should be paid to both pre-PCI and poststent IVUS analysis. When an eccentric CN/ NC is encountered, we recommend careful attention on the efect of wire bias and/or selection of the proper rota-wire type to facilitate optimal burr contact with the CN/NC. Periprocedural IVUS assessment for adequate plaque modifcation, lesion preparation before stent implantation is crucial to reduce the possibility of stent under expansion, asymmetrical expansion, and incomplete stent apposition. Other debulking techniques, such as orbital atherectomy and intravascular lithotripsy, should also be investigated in this subtype of lesion (CN/NC with eccentric calcifcation).

Limitations.
Tis study has some mentionable limitations. First, the retrospective design of our study increased its vulnerability to missing or incomplete data. Second, our data was collected from only a single center.
Tird, our center is a major urban national tertiary referral center for the complex PCI. As such, it is possible that our fndings may not be generalizable to other care settings. Fourth, we used routine RA as debulking therapy in calcifed lesion. We have no outcome information of other debulking therapy such as orbital atherectomy or intravascular lithotripsy on CN. Fifth, patients' data were analyzed from RA database. We cannot exclude selection bias in imaging characteristics which is suitable for RA. Eccentric CN/NC was found in the minority group in our study while it was found in the majority group in Watanabe study [27]. Sixth, because of the heavy calcifed lesion with signifcant stenosis, 80% of patients IVUS failed to cross lesion prior to RA. However, IVUS was immediately performed after RA, prior to angioplasty in all patients. Even though, the characteristic and pattern of calcifed nodule may be able to interpret. But we could miss the important information such as disrupt or intact fbrous cap, eruptive calcifed nodule from morphological change post RA. Seventh, IVUS has limited resolution. IVUS could miss to detect fbrous cap disruption and the presence of luminal thrombus on the surface of CN; hence, it cannot clearly diferentiate CN from NC. Optical coherence tomography (OCT) would provide a better diferentiation between CN and NC. However, this study focused on impact of CN/NC on the DoCE rather than ACS etiology.