Abstract
Cardiac toxicity is a well-recognized potential complication of thoracic radiation (RT), with recent studies demonstrating risk of cardiac injury that occurs with lower doses and at earlier timepoints than previously believed. Thus, an important goal of treatment planning is to minimize dose delivered to the heart. However, treatment planning assumes a static position of the heart on simulation CT. We hypothesized that variation in heart position during RT delivery due to normal respiratory and cardiac function impacts the dose delivered to the heart and that the delivered dose is significantly different from the planned dose. We conducted a study utilizing cine MRI to quantify cardiac motion and to determine its dosimetric impact. Cine MR images (1.5T) were acquired in both free-breathing (FB) and deep inspiration breath hold (DIBH) for 9 healthy volunteers and 8 breast cancer patients treated with whole breast RT. Balanced gradient echo MR images were acquired in four evenly spaced positions through the heart volume, in both sagittal and coronal planes, to quantify cardiac motion in the superior-inferior (SI), anterior-posterior (AP), and right-left (RL) directions. Frame rates were approximately 6.5 and 3.5 frames per second for sagittal and coronal acquisitions respectively. Images were acquired continuously for 15-17 seconds for DIBH and for 61 seconds for FB. The heart was contoured and a SIFT-based tracking algorithm was used to map pixels from a reference image to each frame. The displacement of each pixel relative to its reference position was recorded for every frame. The mean displacements in the SI, AP, and RL directions were determined for each patient and for the cohort. In the treatment planning system, each patient’s observed mean displacement in the SI, AP, and RL directions was applied as an asymmetric expansion on the heart. The mean and max dose to this structure, representing the volume of the heart when accounting for intrafractional motion, were recorded for each patient. Mean heart displacements in the R-L, S-I, and A-P directions are shown in Table 1, which summarizes the observed cardiac motion over 53,040 total frames.Tabled 1Abstract 2117; Table 1DIBH, mm (±SD)FB, mm (±SD)RL5.5 ± 1.17.9 ± 1.2AP5.0 ± 1.37.1 ± 1.4SI7.6 ± 2.514.0 ± 2.5 Open table in a new tab The mean delivered dose to the heart was 161% of the planned dose. When accounting for motion, delivered mean dose to the heart on average in DIBH and FB increased by 82 and 176 cGy, respectively. Delivered Dmax to the heart increased by a mean of 716 cGy. The position of the heart is affected by respiratory and cardiac function, with greater motion observed in free-breathing than in DIBH. The greatest displacements were in the S-I direction, but displacements of dosimetric significance were noted in all three planes for both FB and DIBH. Therefore, treatment planning for thoracic RT which involves steep dosimetric gradients such as tangential beams or IMRT should take the expected cardiac motion into account during treatment planning.
Published Version (Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have