Prone belly board device training improves geometric setup accuracy in lower GI radiotherapy

Abstract

AbstractBackgroundPatients having a course of radiotherapy (RT) must be appropriately immobilised for stability and accuracy. Having opened a new cancer service in June 2009 and commenced treating lower gastrointestinal cancers in 2010, a prone belly board device (BBD) was introduced as the standard radiotherapy immobilisation. A training package was created to aid clinical skills retention of therapeutic radiographers and manage setup quality. Setup reproducibility using the BBD was retrospectively assessed with electronic portal image (EPI) verified geometric displacements as the main outcome measure both before and after the introduction of training.MethodTwenty retrospective Pinnacle computed tomography-planned patients and their geometric displacements on treatment were evaluated between 2010 and 2011—ten prior to (Patient Group A) and ten following training (Patient Group B). The only inclusion criterion was that patients were immobilised for RT on the Medtec ContouraTM carbon fibre BBD. Patients were prone and were treated to 45–50·4 Gy in 25–28 fractions on a 6–10 MV LinAc equipped with EPI. Reproducibility was assessed by comparing geometric measurement of the bony pelvis on the Pinnacle digitally reconstructed radiograph (DRR) with an EPI captured at day 0, 1, 2 and weekly during treatment for each patient. Systematic and random errors were analysed with respect to the average geometric displacement with standard deviation per patient between the Pinnacle DRR and the EPI.ResultsThe age range was 41–77 years and there were 15 male and five female patients with diagnosed rectal cancers (T3–T4, N0–N2, M0). Three hundred and seventy one images were analysed. An improvement in population systematic and random error was most notable in the superior–inferior direction (Patient Group A Σpop = 3·1 mm, σpop = 3·6 mm to Patient Group B Σpop = 2·0 mm, σpop = 2·3 mm, respectively).Discussion/ConclusionThere is evidence that the use of the BBD is more reproducible when accompanied by a task-specific training package. Based on the results of this study, further work will be carried out on training standardisation for patient positioning with a BBD for reducing systematic and random geometric displacements.