Palliative radiation therapy (RT) can be a necessary element of treatment for hospitalized cancer patients who may be critically ill. We have implemented patient-free simulation machine starts using existing diagnostic imaging for treatment planning, eliminating the need for simulation. We hypothesize machine starts reduce time in the department without detriment to clinical outcomes. An IRB-approved database of patients receiving RT was queried for hospitalized patients treated with a machine start. Variables including time on machine and in department, and clinical outcomes were collected. A binary clinical endpoint variable indicated whether the goal of RT was achieved (extubation, hemostasis, etc.). A similar cohort of inpatients receiving CT simulation and mediastinal RT was compared with the machine start group. Mann-Whitney U test was used for comparison between groups. Hypothesis testing for proportions was used to compare patients meeting clinical endpoints. A total of 23 patients were treated with 25 machine starts. The median age was 65 (range 43-82) and median KPS was 50. A total of 15 patients were in the intensive care unit of which 12 were mechanically ventilated (MV). The most common histology was NSCLC (14). RT was most commonly 8-10 Gy in 1 fraction (88%, range 7-30 Gy in 1-10 fractions). Field arrangement was equally weighted AP/PA fields in 84%. Treated sites were mediastinum (17), bone (3), abdomen/pelvis (2), neck (2), and brain (1). Diagnostic CT of the treatment area was obtained a median of 6 days prior to RT (0-57 days). Machine starts within 4 days of admission were significantly more likely to meet the intended clinical endpoint (87% vs 47%, p = 0.02). On day of machine start, patients' median time in the department was 32 minutes (10-141 min). Of machine starts with treatment times available (22), median time in the LINAC vault was 15 minutes (5-31 min). The median shifts from the initial CBCT were 0.9 cm in the superior-inferior direction (0-2.7 cm), left-right 0.9 cm (0-5.9 cm) and anterior-posterior 0.5 cm (0-3 cm). One patient required repeat CBCT after initial shifts. In a similar cohort of patients who underwent CT simulation followed by RT to the mediastinum, the average time in the department was 122 minutes, compared with 45 minutes for machine start patients treated to the mediastinum (p = 0.02). Primary clinical endpoint was achieved in 65% of machine starts. 61% of patients clinically improved and were stable for discharge. Of the MV patients, 42% were successfully extubated and 58% were discharged. Of patients who died (74%), average time to death was 53 days after RT (4-126 days). Among the MV subgroup, 83% died an average of 49 days from RT (4-126 days). Machine starts are a timely and accurate alternative to traditional planning with CT simulation that results in approximately 60% less time in the department. This is an acceptable RT planning technique for a population that demands greater medical attention and resources.
Read full abstract