Analysis of cardiac motion without respiratory motion for cardiac stereotactic body radiation therapy

Abstract Purpose/objective(s) To study the heart motion using cardiac gated computed tomographies (CGCT) to provide guidance on treatment planning margins during cardiac stereotactic body radiation therapy (SBRT). Materials/methods Ten patients were selected for this study, who received CGCT scans that were acquired with intravenous contrast under a voluntary breath‐hold using a dual source CT scanner. For each patient, CGCT images were reconstructed in multiple phases (10%–90%) of the cardiac cycle and the left ventricle (LV), right ventricle (RV), ascending aorta (AAo), ostia of the right coronary artery (O‐RCA), left coronary artery (O‐LCA), and left anterior descending artery (LAD) were contoured at each phase. For these contours, the centroid displacements from their corresponding average positions were measured at each phase in the superior–inferior (SI), medial–lateral (ML), and anterior–posterior (AP). The average volumes as well as the maximum to minimum ratios were analyzed for the LV and RV. Results For the six contoured substructures, more than 90% of the measured displacements were <5 mm. For these patients, the average volumes ranged from 191.25 to 429.51 cc for LV and from 91.76 to 286.88 cc for RV. For each patient, the ratios of maximum to minimum volumes within a cardiac cycle ranged from 1.15 to 1.54 for LV and from 1.34 to 1.84 for RV. Conclusion Based on this study, cardiac motion is variable depending on the specific substructure of the heart but is mostly within 5 mm. Depending on the location (central or peripheral) of the treatment target and treatment purposes, the treatment planning margins for targets and risk volumes should be adjusted accordingly. In the future, we will further assess heart motion and its dosimetric impact.

VT may also be found idiopathic in some patients. 2 A dangerous condition related to VT is ventricular fibrillation. With ventricular fibrillation, the diseased ventricle contracts in a very rapid and uncoordinated manner, resulting in heart failure, frequent fainting episodes, or sudden death by a cardiac arrest. Sustained ventricular arrhythmia is the most common cause of sudden cardiac death, accounting for 75-80% of cases. 3 Implantable cardioverter-defibrillators (ICDs) have been used, along with conventional and antiarrhythmic drug therapy, to prevent sudden cardiac arrest in patients with depressed left ventricular function. Ventricular tachycardia has a high recurrence rate, and catheter ablation can be used to prevent or reduce recurrent episodes of VT. 4,5 However, its invasive nature increases the risk of procedural complications. 6 A recent publication from Washington University in St. Louis reported a noninvasive cardiac radiation for ablation of VT with promising initial resultsin five patient with refractory VT, they reported a markedly reduction of VT. 7 Although this innovative treatment is not a current standard of care, it can be offered to patients who have treatment-refractory VT and have limited other treatment options. The radioablation procedure, referred to as stereotactic radiosurgery, has been a mainstay of treatment for non-malignant conditions such as trigeminal neuralgia and arteriovenous malformations. In the thorax, the radioablation procedure, referred as stereotactic body radiation therapy (SBRT), has been applied for patients with earlystage lung cancers, who are medically inoperable. 8 As reported by Cuculich et al., 7 SBRT was used as the noninvasive treatment method instead of catheter radiofrequency ablation to treat recurrent VT.
In SBRT, organ motion management is important. In the thorax and abdominal regions, breathing motion managements are well studied. 9,10 Breath-hold methods have been applied to patients with lung cancer, liver tumors, or breast cancer during SBRT and conventional radiotherapy. Comparing to the breathing motion, cardiac motion is secondary. If breathing motion can be minimized using either breath-hold method or breathing gated treatment method, the planning margin for cardiac radiation ablation can be drastically reduced. Studies of cardiac motion in radiation therapy are scarce.
Different from the breathing motion, cardiac motion is fast and asymmetrical. [11][12][13] The motion magnitude of each substructure of the heart varies in the rapid cycle of heart beat. Typical CT scanners used in radiation oncology departments are not suitable for study of the cardiac motion. In this work, separated from the impact of the breathing motion, we studied the heart substructure motion using gated cardiac computed tomography (CT) images acquired in the cardiovascular department of our institution to seek for planning margin guidance for cardiac SBRT treatment.
Recent studies showed that the doses to the substructures of heart, especially the left ventricle and the left anterior descending artery, were predictive to radiotherapy related cardiac toxicity. [14][15][16][17] The traditional treatment planning constraint, mean dose to whole heart, is not as good a predictor and does not correlate with the doses to the substructures. Therefore, studying the motion of the heart sub-  18 Figure 1 shows the example contours of these structures. To study the motion of the left anterior descending coronary artery (LAD), instead of contouring the entire LAD, we used an alternative method based on a published study. 19 The coronary arteries can be described by landmarks such as openings and bifurcation points. As shown in Figs. 2(a) 20  For example, SI n (n = 1, 2, . . ., 9) was the SI coordinate at the nth phase, and SI 0 was the average of SI n . The absolute displacement was therefore |SI n − SI 0 | for the nth phase.
The distance to average position for each phase is defined as shown in Eq. (1), where d is the distance; SI, ML, and AP are the position values; n is the phase number.
For data analysis, this work reports the mean value AE standard deviation (SD), minimum, maximum, as well as the distributions of the absolute displacement and distance with 0.5 or 1 mm intervals.
OUYANG ET AL.

| 49
To study the volume changes of the LV and RV, the contour volumes were recorded for each patient in each phase.
The average volumes were calculated for the LV and RV for each patient by averaging over the nine phases. The ratios of maximum to minimum volumes were also calculated for each patient.

| RESULTS
For the 10 patients, as shown in   Table 2 shows the average distance for the centroid of each structure as calculated using Eq. (1), and Fig. 4 shows the histograms of the distances within the defined motion ranges. While the majority of the distances were <5 mm, the RV, O-RCA, and LAD had slightly larger motions.
As shown in Table 3, the ratio of the maximum phase volume to the minimum phase volume ranged from 1.15 to 1.54 for the LV,   The composite distances ranged from 0.20 to 7.85 mm, with 91% of the distances <4 mm and 97% of the distances <5 mm. With the assumption that the breath-hold was perfect during the imaging process, the studied motion was the cardiac motion decoupled from the respiratory motion.
As previous studies found that cardiac motion was asymmetrical, [11][12][13] it is, therefore, important to study the motion of substruc-    heart motion with 20 breast cancer patients using respiratory phasecorrelated 4D-CT and concluded that the LAD motion was up to 9 mm during free breathing. 25 There is limited knowledge on how to incorporate true 4D heart motion into radiotherapy treatment planning. This is becoming increasingly important as radiotherapy, especially SBRT, is now being used to treat substructures of the heart. The high prescription doses and location precision demand better understanding of cardiac motionwith or without respiratory motion management. As a preliminary study, our results warrant further investigation in full motion and dosimetric modeling.

| CONCLUSION
In conclusion, cardiac motion is variable but mostly within 5 mm. When a cardiac substructure is the radiation treatment target, an ITV should be drawn using the best available 4D imaging modality. If advanced imaging equipment is not available or if cardiac substructures are considered as organs at risk during radiotherapy, it is recommended to have a 5 mm expansion to account for cardiac motion. Future study includes full assessment on heart motion and its dosimetric impact.

CONFLI CT OF INTEREST
There is no conflict interest.