Dosimetric Influence of Oscillatory Spinal Cord Motion on the Cord Doses for Cervico-Thoracic Spine SBRT

Abstract

Oscillatory spinal cord motion is due to physiologic phenomena, such as cardiac and respiratory cycles which gets accentuated when the patient swallows during treatment. Random Bulk shifts which happen due to the patient movement can be mitigated with rigid immobilisation. The foci of the present study was to assess physiological spinal cord oscillatory motion specifically the Anterior/posterior (A/P) motion, consequent dosimetric influence on the cord doses & PRV implications when single fraction SBRT is used for cervico-thoracic spinal segments. Twelve patients who underwent single fraction spine SBRT for metastases at vertebral segments from C2-D4 were analysed for study purpose. All patients underwent rigid immobilization, 4D-MRI and 4D-CT simulation. MR Image acquisition focussing on the involved vertebra was performed during 30sec breath-holding (To assess cardiac influence), normal regular breathing and forced breathing. Spinal cord motion within the spinal canal was assessed by cardiac-triggered and ECG-gated gradient echo pulse sequence MRI. Following the fusion of 4-D CT (index bin) with the Axial-T2 MR sequence, target volume and the normal structures (including thecal sac) were delineated and intensity based deformable image registration algorithm was used to sequentially deform the structures to nearest-neighbour phases among the 10 phases. A moving voxel dose from unflat beam VMAT plan generated on the original bin is propagated through the 4D dose space using DIR and the fraction dose to that voxel on thecal sac is accumulated. Parameters analysed for each voxel of 4D Imaging included (1) the Motion patterns of SC - magnitude of the A/P displacements (mean) in different breathing patterns (2) Median percentage increase in reconstructed motion- perturbed thecal sac Max dose, threshold doses(0.03cc), 1 cc dose and 2 cc dose were analysed in 2 scenarios – Static and total accumulated. Computation was done using descriptive statistics & Wilcoxon-log-rank test. Mean, SD Antero-posterior shifts of the cord (mm) in 4D- MRI acquired during breath hold, normal breathing and moderate forced breathing were 0.2±0.9mm; 0.3±0.3mm, 1.2±1mm ( p<0.01 each). In free breathing mode, the median percentage increase in accumulated doses compared to static doses for cord Maximum dose, threshold dose, 1 cc and 2cc were 2.4%, 2.6%, 4.3% & 3.2% respectively. Cord motion studies reveal respiration to be a considerable cause of motion compared to heart-related motion. Spinal cord motion during breath-holding was highest at the levels of the first and second thoracic vertebral segments. It would be a safe practice to increase the Cord PRV by 0.5mm while targeting segments from C2 through D4 where the oscillatory motion amplitude is very high due to regional CSF flow differences.To the best of our knowledge this is the first reported study in literature which looked into the dosimetric influence of cord motion at specific spinal segments.