White Thrombosis on Optical Coherence Tomography after Rotational Atherectomy of Severely Coronary Calcied Lesions

Purpose Rotational atherectomy (RA) has improved percutaneous treatment of severe coronary calcied lesions, but the "no-reow" phenomenon remains a serious complication. Platelet activation by RA may contribute to the no-reow, we use optical coherence tomography (OCT) to test the effect of RA on white thrombosis that could conrm platelet activation indirectly. We analyzed 53 consecutive patients with coronary angiographic severe calcied lesions. All the patients were examined by OCT. Twenty patients received RA and OCT imaging performed before and after RA and stent implantation. The remaining patients were as a control group and OCT imaging performed before balloon dilatation and after stent implantation.


Introduction
Coronary calci cation plaque is common in patients undergoing percutaneous coronary interventions (PCI). The incidence ranges between 17% and 35% [1]. Coronary calci cation affects the surgery success rate and long-term clinical outcome [2,3]. Calci ed lesions can lead to stent implantation failure or incomplete stent expansion, thus affecting the long-term e cacy of stents. It also increases the risk of perforation and coronary dissection during operation [4,5]. Rotational atherectomy (RA) plays a key role in the therapy of severely calci ed lesions [6,7]. It can reduce the incidence of complications and improve the operation success rate. European expert consensus on rotational atherectomy recommend that when heavily calci ed lesions that cannot be adequately dilated by a balloon before stenting, RA is a reasonable choice to su cient pretreatment of calci ed lesions [7].
Platelet activation is an important determinant of acute outcomes of PCI. Some studies had con rmed that RA could cause platelet aggregation in vitro models [8,9], which lead to slow ow or no re ow during RA [10].
Optical coherence tomography (OCT) is a high resolution (10 μm axial), light-based imaging modality, it can accurately display the microstructure characteristics of atherosclerotic plaque, identify vulnerable plaque, thrombus, calci cation and dissection, which has been demonstrated to show excellent correlation with histology [11]. To our knowledge, there has been no imaging study in vivo to assess the effect of RA on thrombogenesis by OCT. The purpose of our research was to use OCT to examine the effect of RA on thrombogenesis which could con rm platelet activation indirectly in vivo.

Study population
This is a single-center observational study, including 53 consecutive patients who underwent OCT imaging due to severe coronary calci cation on angiography from July 2019 to December 2019. The de nition of severe calci cation on angiography was de ned as clear calci ed shadows that can be seen in both beating and resting hearts before contrast injection. If the OCT image shows a maximum calcium angle>180°, maximum calcium thickness>0.5mm and maximum calcium length>5mm, the patients will receive RA therapy before stent placement [12]. If the OCT catheter did not pass the lesion, OCT was performed immediately after dilation with a small balloon. Finally, 20 patients met the above criterion and were included in the RA group, and they received RA therapy before stent placement. The other 33 patients were included in the control group, and they received routine interventional therapy. The demographic and clinical characteristics of all patients were derived from the electronic record system. This study was approved by the ethics at Peking University People's Hospital. All patients provided written informed consent before participation.

RA details
Before procedural, all patients received an oral loading dose of 300 mg aspirin and 600mg clopidogrel.
During procedural, all patients received unfractionated heparin at a dose of 70-100 U/kg or bivalirudin to maintain an activated clotting time (ACT) >300 seconds. The choice of vascular access, burr size was left at the operators' discretion. RA was performed by using the Rotablator (Boston Scienti c Scimed, Inc., Maple Grove, MN, USA). The burr size was selected to reach a burr/vessel ratio of 0.5-0.6. RA speed ranged between 140,000 and 160,000 rotation per minute. Each ablation time was 10-15 seconds. During RA, a continuous intracoronary infusion of a cocktail with unfractionated heparin and nitroglycerin was employed. The success of RA was de ned as the complete expansion of the balloon of the target lesion after RA.
OCT Image Acquisition and Analysis OCT image acquisition was performed before RA, after RA and after stent placement in the RA group, and it was performed before balloon dilatation and after stent implantation in the control group with C7-XR or ILUIMEN OPTIS OCT Intravascular Imagining System (Abbott, USA). All OCT data were analyzed using Abbott Vascular O ine Review Workstation by two independent interventional cardiologists who were blinded to the clinical and angiographic information.
The minimum lesion lumen area (MSA) was de ned as the smallest cross-sectional area (CSA) in the lesion segment. Calcium lesion was evaluated using three parameters: maximum angle, maximum thickness, and length. When calcium was extremely thick, and its border was not clear due to attenuation, the maximum visible thickness was measured [12].
Thrombus was one consequence of platelet aggregation, so we con rmed platelet aggregation by thrombogenesis indirectly. White thrombus was de ned as a mass attached to the luminal surface or oating within the lumen, which is less backscattering, homogeneous, and has low attenuation [13]. The thrombotic load was evaluated by the number of white thrombus on OCT.

Statistical Analysis
Descriptive statistics are presented as mean±SD and compared with the t-test. Categorical data are expressed as frequency and percentage and was compared with chi-square or Fisher exact test where appropriate. Pearson Correlation Analysis was used to analyze the correlation between continuous variables.

Patient Characteristics
We nely included 53 consecutive cases. baseline clinical characteristics were summarized in Table 1.
The patients were older and had a higher incidence of Diabetes mellitus in the RA group. In the RA group, the average age of the patients was 70.3±9.5 years. Most patients were male (75.0%), and clinical manifestations were acute coronary syndrome (ACS).

Details of RA and Stent Placement
Detailed information on RA techniques and stents placement data was shown in Table 2. The most common RA target vessel was the left anterior descending branch (LAD). There was no left circum ex branch RA in our data. The most commonly employed burr size was 1.5mm (70.0%), mean burr size was 1.48±0.14mm. The mean rotation speed was 152,300±4,200rpm. The mean number of rotations per patient was 5.3±2.1 times and the mean ablation time per RA was 11.3±3.2 seconds. The mean ACT time was 300.3±35.7 seconds. There was one patient with no re ow after RA.
OCT Based Lesion Modi cation before and after RA OCT showed that the maximum calcium angle, maximum calcium thickness and calcium length was 322.5±58.8°vs. 309.4±63.6°, 1.67±0.27 vs. 0.86±0.19 mm, and 41.7±13.5 vs. 40.7±13.3 mm before and after RA, respectively. Although the difference was not statistically signi cant, the maximum calcium thickness was lower than pre-RA. We found that in the control group, there was no thrombogenesis during the procedure, but in the RA group, all the target vessels had only white thrombus on OCT after RA, which could con rm platelet activation indirectly, and red thrombus was not found. Representative OCT images of RA-related white thrombus are shown in Figure 1. RA led to calcium fracture in 80% lesions and dissection in 15% lesions. Representative OCT images of calcium crack and dissection are shown in Figure 2.

Impact Factors of RA-induced White Thrombus
OCT showed the average number of white thrombus per lesion after RA was 7.23±4.4, and the average length of every white thrombus was 0.51±0.33mm. Pearson Correlation was calculated between thrombotic load (number of white thrombus) and other variables (burr size, rotation speed, rotation number, ablation time per RA, ACT, maximum calcium angle, maximum calcium thickness, calcium length, and minimal lumen area). The result showed that the thrombotic load was related to burr size (r=0.575, P=0.040) and rotation number (r=0.599, P=0.031). Bigger burr size and higher number of rotation would result in higher thrombotic load.

Discussion
We reported for the rst time OCT images of RA induced white thrombosis of severely calci ed coronary lesions.
RA played an important role in pre-treat of severely calci ed lesions before stenting. Complications of RA included coronary dissection (10.5%), severe coronary spasm (1.6-6.6%), acute vascular occlusion (3.1%), slow-ow/no-re ow (1.2-7.6%) and coronary artery perforation (0-2%) [14.15]. Among these complications, no-re ow phenomenon remains serious and common. Potential mechanisms of no-re ow during RA included atheromatous debris embolism, platelet activation, microcirculatory vasospasm and so on [16]. A previous study had proved the effect of RA on platelet activation in an in vitro model [8]. To our knowledge, this is the rst report about the effect of RA on platelet aggregation in vivo indirectly proved by OCT imaging of white thrombosis. In one article (Kini et al.) which assessed of the mechanistic effects of rotational and orbital atherectomy in severely calci ed coronary lesions by OCT (RA details: burr size 1.68±0.11mm, rotation speed 150,000rpm) [17], the authors found the similar phenomenon, also in all the lesions treated by RA. They de ned this phenomenon as small intimal nodules. But according to the OCT image characteristics (different transmissivity and the clear boundary from intima), combined with the results of the in vitro test, we think it is white thrombus. Future histological studies will be necessary to prove this hypothesis.
In our research, only white thrombus was found after RA, which was same to Kini's article [17]. Red thrombus was not found. The reasons may be that red thrombogenesis was usually be seen after the mixed thrombus completely block the lumen. it often composes the tail of the continuous thrombus.
During the RA procedure, there was no enough times to generate red thrombus.
The thrombotic load is in uenced by several factors. In our study, we found bigger burr size and higher number of rotation would result in higher thrombotic load. In invitro model study, Reisman et al. tested freshly collected heparinized blood, which was exposed to a 2.0 mm Rotablator burr rotating at one of three speeds: 180,000, 140,000, or 0 rpm. There were signi cantly more platelet aggregates of >20microm diameter at higher RA speeds, compared with high-speed RA (180,000rpm), low-speed RA (140,000rpm) can reduce the platelet aggregation [8]. In our study, we did not nd the in uence of rotation speed on platelet aggregation. The reason may be that in our center, we use low-medium rotation speed (140,000-160,000rpm) in all the patients in clinical practice, so the relation of rotation speed and thrombotic load can't be found in the Pearson Correlation Analysis.
The mechanism of RA induced platelet aggregation is not entirely clear. It may be caused by heat generated by RA and RA induced cell damage [18]. These data suggest a proper RA strategy (burr size, rotation speed and ablation numbers) should be used to avoid no-re ow during RA. Williams et al. showed that pretreating platelet-rich plasma with abciximab could decrease the RA induced platelet aggregation [9]. Future work is necessary to the potential of therapeutic options, including additional pharmacotherapy agents with RA to mitigate this effect.
This work was an observational study. We generated a hypothesis from the observations. There were several limitations of our research. First, the study sample number was small. Second, the presence of white thrombus does not directly prove platelet aggregation. Future histological studies and randomized controlled study using different RA strategies will be needed to prove this hypothesis.

Conclusions
White thrombosis during RA is proved by OCT in vivo. Treating calci ed lesions with RA may enhance thrombogenesis. The bigger burr size, a higher number of ablation times, maybe together with higher rotation speed, can in uence the thrombotic load. These data suggest a proper RA strategy should be used to avoid no-re ow during RA.