Inter-institutional variability of small-field-dosimetry beams among HD120™ multileaf collimators: a multi-institutional analysis

Detector selection and technical problems can result in large variations in small-field-dosimetry data among institutions. In this study, we evaluated inter-institutional variability in the small-field-beam data of the Novalis Tx linear accelerator (Varian Medical Systems and BrainLAB) with an HD120 (TM) multileaf collimator. Beam data for modeling an iPLAN treatment planning system (BrainLAB) were collected from 19 institutions and median values of percentage depth doses (PDD), diagonal profiles, transversal profiles, and ratios of detector readings (detector output factors; OFdet) were calculated. Inter-institutional variability was defined as the difference between the median value and the value for each machine. PDD measured with a 100 mm square field size and diagonal profiles showed only small variations; however, when measured with a 5 mm square field size, the PDD variation from the median exceeded +/- 2%, especially for ionization chambers. With a 10 mm square field, the variation was within approximately +/- 1%. The OFdet variation was within +/- 2% for >= 20 mm square field sizes. The maximum variation exceeded 20% for 5 mm square fields. The ionization chambers' OFdet values were smaller than the median, whereas those for the EDGE detector (Sun Nuclear Corp) were larger. When the OFdet values were corrected by output factor correction factors, the variation was greatly reduced, with only a few machines showing variations greater than +/- 5% from the median value. In conclusion, this multi-institutional investigation of small field dosimetry for HD120 multileaf collimators demonstrated some large variations in the dosimetric parameters, especially for a 5 mm square field size. Most differences were detector-dependent, and the variation was reduced when output correction factors were applied. However, variations probably due to measurement errors were also observed, indicating that careful management is needed for small-field dosimetry.