Orbital atherectomy (OA) is an alternative atherectomy for calcified coronary lesions, with acceptable outcomes1. OA is characterised by a drive shaft with an eccentrically mounted diamond-coated crown, which allows the combination of orbital and rotational motions and removes a layer of calcified plaque with antegrade and retrograde passes of the crown2. Recently, imaging-based algorithms have been proposed for atherectomy and lithotripsy345; however, the optimal device selection remains unclear. The burr of rotational atherectomy moves directly along the wire and shaves off calcified plaque by utilising wire bias; however, the movement of the OA crown can be unpredictable, particularly in cases of ending lesions. The objective of this study was to elucidate the anatomical setting for efficient OA on nodular calcium in bending lesions using in vitro experimental models in which the debulked volume can be measured quantitatively.
Methods
Model construction
In vitro vessel models were developed to evaluate atherectomy efficiency. The models consisted of two components: a bent coronary artery model − with a reference diameter of 3.5 mm and a 10 mm long calcified lesion causing 80% stenosis, located on the inner or the outer curve − and an aorta model (Figure 1A). In total, 40 in vitro stenotic coronary artery models with nodular calcium were developed, using two curvature radii (CR: 10 mm or 20 mm), two bending angles (BA: 100° or 60°), and five models with a combination of two different calcium positions (inner or outer curve) (Figure 1B). The calcium model was fabricated using a mixture of calcium sulphate hemihydrate powder, cement powder, and polyurethane resin. The concentrations of the mixture were adjusted to replicate the mechanical properties of in vivo calcification fractures. Each coronary artery model was connected to a whole aorta model. 
Figure 1. In vitro vessel models. The in vitro vessel model consists of a bent coronary artery model with calcium and an aorta model (A). Coronary percent stenosis: 80%; calcium length: 10 mm; reference diameter: 3.5 mm; and aortic diameter: 30 mm. B) Procedure setup. C) Model design for bending angle, calcium position, and curvature radius.
Simulation protocol
The procedures were performed using the Diamondback 360 Coronary Orbital Atherectomy System (Cardiovascular Systems, Inc. [now Abbott]) through a 6 Fr short-tip Amplatz 1.0 guiding catheter. A ViperWire Advance with Flex Tip Coronary Guide Wire (Cardiovascular Systems, Inc. [now Abbott]) was deployed within the vessel. The protocol encompassed two 30-second low-speed (80,000 rpm) ablations, followed by one 30-second high-speed (120,000 rpm) ablation. The OA technique principally involved gradual retraction of the crown and subsequent forward movement of the crown for models with calcium on the inner curve or direct forward movement of the crown for models with calcium on the outer curve. The crown was then manipulated back and forth in the lesion until frictional resistance was no longer perceived. During the experiments, the crown movement was recorded using a high-speed camera (FASTCAM-1024PCI [Photron]) (Moving image 1, Moving image 2).
Microcomputed tomography evaluation
Microcomputed tomography (micro-CT) analyses (TDM1300-IS [Yamato Scientific]) of the vessel models were performed before and after ablation to evaluate the treatment efficacy. The ablated calcium volume and maximal depth were analysed to compare the inner and outer curves in each model.
Statistical methods
Continuous variables are presented as mean±standard deviation (SD). The differences between the ablated calcium volumes in the models with calcium on the inner and outer curves were evaluated using a two-tailed t-test. Furthermore, the differences in ablated volume between the highly ablated and other models were evaluated across the inner and outer curves using a two-tailed t-test. Statistical analysis was performed using JMP 14.3.0 software (SAS Institute), and p<0.05 was considered statistically significant.
Results
Figure 2 shows the volume of ablated calcium in the inner and outer models. The ablated calcium volume differed depending on the calcium position: the mean ablated volume of nodular calcium on the inner curve of the bend was significantly higher than that on the outer curve of the bend (6.6±0.4 mm3 vs 2.7±0.4 mm3; p<0.0001). On the inner curve, the ablated calcium volume was highest in the CR20/BA60 model (8.3±1.4 mm3) compared with the CR10/BA100 (6.3±1.9 mm3; p=0.14), CR10/BA60 (5.6±1.4 mm3; p=0.04), and CR20/BA100 (6.1±1.3 mm3; p=0.09) models (Figure 3). On the outer curve, the ablated calcium volume was highest in the CR20/BA100 model (5.2±0.9 mm3) compared with the CR10/BA100 (2.0±0.6 mm3; p<0.001), CR10/BA60 (2.1±0.8 mm3; p=0.002), and CR20/BA60 (1.7±1.2 mm3; p<0.001) models (Figure 3). The maximum depth after ablation showed a trend similar to that of the ablated volume (Figure 4). The mean maximum depths were highest for the inner curve in the CR20/BA60 model (1.0±0.1 mm) and for the outer curve in the CR20/BA100 model (0.5±0.1 mm). A comparison of the maximum widths between the models revealed no significant disparities, both in the inner and outer models (Figure 5). Figure 6 shows representative photographs during ablation. In the CR20/BA60 model for the inner curve calcium, the proximal and the distal portions of the wire were firmly anchored to the vessel wall, and the crown exhibited a greater degree of linearity at the lesion site than in the other models. Conversely, in the CR20/BA100 model designed for the outer curve calcium, the wire exhibited a substantial arc, and the crown was correctly aligned with the lesion. 
Figure 2. Comparison of ablated calcium volume between the inner and outer curves of the lesion. The ablated calcium volume on the inner curve of the lesion was significantly higher compared with that on the outer curve (6.6±0.4 mm3 vs 2.7±0.4 mm3; p<0.0001). The black horizontal line indicates the mean value across the two groups. 
Figure 3. Comparison of ablated calcium volume among four combinations of CR and BA. For inner curve calcium, the ablated calcium volume was highest in the CR20/BA60 model (8.3±1.4 mm3). For outer curve calcium, the ablated calcium volume was significantly higher in the CR20/BA100 model (5.2±0.9 mm3), compared to the other models. CR: curvature radius; BA: bending angle 
Figure 4. Comparison of maximal depth after ablation among four combinations of CR and BA. For inner curve calcium, the maximum depth of ablated calcium was highest in the CR20/BA60 model (1.0±0.1 mm). For outer curve calcium, the maximum depth of ablated calcium was highest in the CR20/BA100 model (0.5±0.1 mm). CR: curvature radius; BA: bending angle 
Figure 5. Comparison of maximal width after ablation among four combinations of CR and BA. For inner curve calcium, the maximum width of ablated calcium in the CR20/BA60 model (1.8±0.2 mm) was not the highest, unlike the ablated calcium volume. For outer curve calcium, the maximum width of ablated calcium in the CR20/BA100 model (1.3±0.1 mm) was not the highest, unlike the ablated calcium volume. CR: curvature radius; BA: bending angle 
Figure 6. Representative images during ablation. Representative images illustrating how the crown on the guidewire touched the lesion in each of the inner curve calcium models (A), and representative images illustrating how the crown on the guidewire touched the lesion in each of the outer curve calcium models (B). CR: curvature radius; BA: bending angle
Discussion
In this study, we developed in vitro vessel models with eight different anatomies and quantitatively evaluated the ablated volume following OA. The results of this study provided quantitative evidence that (1) nodular calcium on the inner curve was ablated more effectively than that on the outer curve; (2) the volume of calcium ablation differed depending on the anatomical combination of CR and BA; (3) on the inner curve, the CR20/BA60 group exhibited the greatest ablated calcium volume and depth compared with the other groups; and (4) on the outer curve, the CR20/BA100 group showed the highest calcium ablation volume and depth among all groups. The unique nature of OA is characterised by the combination of orbital and rotational movements of an eccentrically mounted diamond-coated crown6. As the crown rotates, its orbital diameter expands radially via a centrifugal force to modify the calcified plaque and increase the luminal size7. However, this is not the case with the treatment of a bending vessel. Notably, a ViperWire Advance with Flex Tip Coronary Guide Wire and its driveshaft with crown caused vessel straightening and “wire bias” at the bend, facilitating enhanced contact with the wall. The crown can ablate the focal plaque bidirectionally as it advances back and forth through a target lesion, depending on the degree of wire bias. The experimental data demonstrated that significantly more calcium could be ablated from the inner curve of the vessel than from the outer curve. One potential explanation for this result is the enhanced contact of the crown with the inner curve when the crown is pulled back. This is a feature of OA that is absent in rotational atherectomy. A previous case report has discussed the successful treatment of tortuous lesions with pullback OA, rather than rotational atherectomy, although the mechanism has not been clarified8. The most noteworthy outcome derived from the data was that the combination of curve and angle was associated with the efficiency of OA for calcium on both the inner and outer curves of the bend. The maximum depth after ablation, rather than the maximum width, was comparable to the ablated volume. In the inner curve calcium models, the volume of calcium ablated in the CR20/BA60 model was the highest among the models. Conversely, in the outer curve calcium models, the volume of calcium ablated in the CR20/BA100 model was the highest among the models. These differences may be attributed to the fixed wire bias, which is affected by the CR and BA of the vessel. For lesions located on the outer curve with a large CR and BA, the outward force remains stable. This is because the guidewire’s low bending moment reduces the restoring force that would otherwise straighten the wire. The in vitro transparent bending models and micro-CT accurately demonstrated the effectiveness of OA. To evaluate the debulking effect of OA for calcified nodules (CNs), Shin et al quantified ablation volume using optical coherence tomography imaging9. The authors demonstrated that OA may be preferable for CNs with smaller baseline lumen areas. However, the analysis was based exclusively on the lumen area, and the effect of anatomy-related wire bias was not considered. The present study revealed that the efficacy of OA was enhanced when employed on a lesion located on the inner curve, characterised by a large radius of curvature and a small BA. The most significant indication for OA may be the presence of CNs on the inner curve. The utilisation of bidirectional ablation, involving a gradual retraction followed by a subsequent forward movement of the crown, resulted in significant removal of thick calcium.
Limitations
This study highlights the role of anatomical features in determining the efficacy of OA. However, the evaluation was conducted using simplified two-dimensional coronary artery models to analyse the effects of curvature radius, bending angle, and calcium position. Consequently, the findings may not fully reflect the complexity of real patient anatomies. Further studies using three-dimensional vessel models that more accurately represent the anatomical diversity of coronary curves would be more clinically relevant. To better assess the risk of injury to healthy vessel walls – especially in patients with more intricate vascular structures – intracoronary imaging should be employed in clinical practice.
Conclusions
The anatomical setting for efficient OA is the nodular calcium on the inner curve of the vessel. The volume of calcium ablation, predominantly determined by depth, varies depending on the anatomical combination of the CR and BA.
Impact on daily practice
Orbital atherectomy (OA) is a unique device characterised by the combination of orbital and rotational movements of an eccentrically mounted diamond-coated crown. The process of bidirectional ablation enhances the contact between the crown and the calcium on the inner curve, thereby leading to substantial calcium removal. This study quantitatively clarifies that OA more effectively ablates nodular calcium on an inner curve with a large radius and a small bending angle.
Acknowledgements
The orbital atherectomy devices, including the Diamondback 360 coronary orbital atherectomy classic crown and the ViperWire Advance with Flex Tip Coronary Guide Wire, were provided by Cardiovascular Systems, Inc (now Abbott).
Funding
This research was supported in part by a grant from the Japan Agency for Medical Research and Development (AMED, Grant Number 24mk0121298h0001).
Conflict of interest statement
The authors have no conflicts of interest to declare.