Establishing radiation therapy treatment planning effects involving implantable pacemakers and implantable cardioverter-defibrillators

Synthetic Megavoltage Cone Beam Computed Tomography Image Generation for Improved Contouring Accuracy of Cardiac Pacemakers

In this study, we aimed to enhance the contouring accuracy of cardiac pacemakers by improving their visualization using deep learning models to predict MV CBCT images based on kV CT or CBCT images. Ten pacemakers and four thorax phantoms were included, creating a total of 35 combinations. Each combination was imaged on a Varian Halcyon (kV/MV CBCT images) and Siemens SOMATOM CT scanner (kV CT images). Two generative adversarial network (GAN)-based models, cycleGAN and conditional GAN (cGAN), were trained to generate synthetic MV (sMV) CBCT images from kV CT/CBCT images using twenty-eight datasets (80%). The pacemakers in the sMV CBCT images and original MV CBCT images were manually delineated and reviewed by three users. The Dice similarity coefficient (DSC), 95% Hausdorff distance (HD95), and mean surface distance (MSD) were used to compare contour accuracy. Visual inspection showed the improved visualization of pacemakers on sMV CBCT images compared to original kV CT/CBCT images. Moreover, cGAN demonstrated superior performance in enhancing pacemaker visualization compared to cycleGAN. The mean DSC, HD95, and MSD for contours on sMV CBCT images generated from kV CT/CBCT images were 0.91 ± 0.02/0.92 ± 0.01, 1.38 ± 0.31 mm/1.18 ± 0.20 mm, and 0.42 ± 0.07 mm/0.36 ± 0.06 mm using the cGAN model. Deep learning-based methods, specifically cycleGAN and cGAN, can effectively enhance the visualization of pacemakers in thorax kV CT/CBCT images, therefore improving the contouring precision of these devices.

Current Clinical Practice in Patients With Cardiac Implantable Electronic Devices (CIED) Undergoing Radiotherapy (RT)

Patients with cardiac implantable electronic devices (CIED) undergoing radiotherapy (RT) are more common due to ageing of the population. With newer CIEDs implementing the complementary metal-oxide semiconductor (CMOS) technology which allows the miniaturisation of CIED, it is also more susceptible to RT. Effects of RT on CIED ranges from device interference, device operational/memory errors of permanent damage. These malfunctions can cause life threatening clinical effects. Cumulative dose is not the only component of RT that causes CIED malfunction, as neutron use and dose rate effect also affects CIEDs. The management of this patient cohort in clinical practice is inconsistent due to lack of a consistent guideline from manufacturers and physician specialty societies. Our review will focus on the current clinical practice and the recent updated guidelines of managing patients with CIED undergoing RT. We aim to simplify the evidence and provide a simple and easy to use guide based on the recent guidelines.

Management of radiotherapy patients with implanted cardiac pacemakers and defibrillators: A Report of the AAPM TG-203†

Managing radiotherapy patients with implanted cardiac devices (implantable cardiac pacemakers and implantable cardioverter-defibrillators) has been a great practical and procedural challenge in radiation oncology practice. Since the publication of the AAPM TG-34 in 1994, large bodies of literature and case reports have been published about different kinds of radiation effects on modern technology implantable cardiac devices and patient management before, during, and after radiotherapy. This task group report provides the framework that analyzes the potential failure modes of these devices and lays out the methodology for patient management in a comprehensive and concise way, in every step of the entire radiotherapy process.

Radiolucent Skull Clamps for Intra-Operative Imaging: A Technical Note

Conventionally, surgery and procedural-based radiology are performed on different premises. With advances in imaging technology, the operating room is rapidly being transformed into an intraoperative imaging suite. Diagnostic imaging in conjunction with surgery has great utility and by all accounts has great future potential. During the last decade, cone beam computed tomography (CT) scanners have been introduced and have made intraoperative imaging more feasible because these scanners can be made less bulky. The current usefulness of intraoperative magnetic resonance imaging (MRI) or CT for neurosurgery, however, is impaired by the lack of completely radiolucent skull clamps, causing image artifacts. Metal artifacts are particularly problematic, given that they lead to a considerably higher image quality degradation factor for cone beam CT scanners than for conventional CT scanners. Here, we describe our experience with near-radiolucent skull clamps and their associated problems and discuss future improvements to facilitate high-quality image guidance in the field of neurosurgery.

Measuring dose from radiotherapy treatments in the vicinity of a cardiac pacemaker

This study investigated the dose absorbed by tissues surrounding artificial cardiac pacemakers during external beam radiotherapy procedures. The usefulness of out-of-field reference data, treatment planning systems, and skin dose measurements to estimate the dose in the vicinity of a pacemaker was also examined. Measurements were performed by installing a pacemaker onto an anthropomorphic phantom, and using radiochromic film and optically stimulated luminescence dosimeters to measure the dose in the vicinity of the device during the delivery of square fields and clinical treatment plans. It was found that the dose delivered in the vicinity of the cardiac device was unevenly distributed both laterally and anteroposteriorly. As the device was moved distally from the square field, the dose dropped exponentially, in line with out-of-field reference data in the literature. Treatment planning systems were found to substantially underestimate the dose for volumetric modulated arc therapy, helical tomotherapy, and 3D conformal treatments. The skin dose was observed to be either greater or lesser than the dose received at the depth of the device, depending on the treatment site, and so care should be if skin dose measurements are to be used to estimate the dose to a pacemaker. Square field reference data may be used as an upper estimate of absorbed dose per monitor unit in the vicinity of a cardiac device for complex treatments involving multiple gantry angles.

Dosimetric review of cardiac implantable electronic device patients receiving radiotherapy

A formal communication process was established and evaluated for the management of patients with cardiac implantable electronic devices (CIEDs) receiving radiation therapy (RT). Methods to estimate dose to the CIED were evaluated for their appropriateness in the management of these patients. A retrospective, institutional review board (IRB) approved study of 69 patients with CIEDs treated with RT between 2005 and 2011 was performed. The treatment sites, techniques, and the estimated doses to the CIEDs were analyzed and compared to estimates from published peripheral dose (PD) data and three treatment planning systems (TPSs) — UMPlan, Eclipse's AAA and Acuros algorithms. When measurements were indicated, radiation doses to the CIEDs ranged from 0.01–5.06 Gy. Total peripheral dose estimates based on publications differed from TLD measurements by an average of 0.94 Gy (0.05–4.49 Gy) and 0.51 Gy (0–2.74 Gy) for CIEDs within 2.5 cm and between 2.5 and 10 cm of the treatment field edge, respectively. Total peripheral dose estimates based on three TPSs differed from measurements by an average of 0.69 Gy (0.02–3.72 Gy) for CIEDs within 2.5 cm of the field edge. Of the 69 patients evaluated in this study, only two with defibrillators experienced a partial reset of their device during treatment. Based on this study, few CIED‐related events were observed during RT. The only noted correlation with treatment parameters for these two events was beam energy, as both patients were treated with high‐energy photon beams (16 MV). Differences in estimated and measured CIED doses were observed when using published PD data and TPS calculations. As such, we continue to follow conservative guidelines and measure CIED doses when the device is within 10 cm of the field or the estimated dose is greater than 2 Gy for pacemakers or 1 Gy for defibrillators.

PACS number: 87.55.N‐

Evaluation of a ventricular assist device: stability under x-rays and therapeutic beam attenuation

Improved outcomes and quality of life of heart failure patients have been reported with the use of left ventricular assist devices (LVADs). However, little information exists regarding devices in patients undergoing radiation cancer treatment. Two HeartWare Ventricular Assist Device (HVAD) pumps were repeatedly irradiated with high intensity 18 MV x-rays to a dosage range of 64-75 Gy at a rate of 6 Gy/min from a radiation oncology particle accelerator to determine operational stability. Pump parameter data was collected through a data acquisition system. Second, a computerized tomography (CT) scan was taken of the device, and a treatment planning computer estimated characteristics of dose scattering and attenuation. Results were then compared with actual radiation measurements. The devices exhibited no changes in pump operation during the procedure, though the titanium components of the HVAD markedly attenuate the therapy beam. Computer modeling indicated an 11.8% dose change in the absorbed dosage that was distinctly less than the 84% dose change measured with detectors. Simulated and measured scattering processes were negligible. Computer modeling underestimates pretreatment dose to patients when the device is in the field of radiation. Future x-ray radiation dosimetry and treatment planning in HVAD patients should be carefully managed by radiation oncology specialists.

An opposed matched field IMRT technique for prostate cancer patients with bilateral prosthetic hips

Intensity‐modulated radiation therapy (IMRT) has gained wide‐spread use for treating patients with prostate cancer, yet developing a plan for patients with bi‐lateral metal hip prostheses implants may be challenging. The high atomic number of the metallic hips not only gives rise to streak artifacts that obscure anatomy but also attenuates laterally directed fields by a significant amount that cannot be reliably ascertained from the CT dataset. A common approach to planning directs five IMRT fields such that incidence through the metal hips is avoided. While this technique generally gives adequate PTV coverage, it may escalate the rectal dose if beams, which would otherwise be incident from a lateral direction, are angled toward a posterior direction in order to avoid the prosthesis. In this work, we propose and investigate a new technique which alleviates this problem by introducing asymmetric opposed fields that are edge‐matched along a plane that is tangent to the metal prostheses. With this approach, a posterior oblique field is oriented closer to the lateral direction but does not irradiate the ipsilateral prosthesis. The portion of the target eclipsed by the prosthesis is irradiated by the opposed matched anterior oblique field which, again, avoids the corresponding ipsilateral prosthesis. While the proposed technique may improve rectal sparing and PTV coverage, the dose along the match plane is sensitive to intrafraction motion. In the worse case of intrafraction motion perpendicular to the plane occurring in the time interval between the deliveries of successive fields of the opposed matched pair, the induced error is typically about 5 cGy per mm of target motion for a 200 cGy fraction. To reduce the induced error, several approaches to broadening the penumbra at the match plane were investigated and compared to conventional IMRT plans for three patients. Phantom measurements were performed to evaluate the effectiveness of these approaches. Match‐plane shifts of 4 mm in a single step, in two 2 mm steps, and in four 1 mm steps, were effective in reducing the worse case induced error to 2.8 cGy per mm. Imposing match‐plane shifts precludes the use of intensity modulation for the opposed matched field pairs. Therefore, we favor an approach whereby the opposed matched fields overlap by 4 mm. Since both fields contribute fluence to the overlap region, the worse case induced error was observed to be typically within 2.9 cGy per mm. In conclusion, the use of this technique should be considered for patients with bilateral metal hip implants who do not meet dose‐volume criteria by conventional IMRT techniques.

PACS number: 87.55.‐x