Clinically shaped fields: relative outputs for Varian multileaf collimators and tray-mounted cerrobend blocks

The purpose of this study was to determine the variation in relative output factors for fields shaped using a multileaf collimator (MLC) versus tray-mounted cerrobend blocks. A total of 21 different clinical fields from 16 patients were compared by casting cerrobend blocks from the same films that had generated the MLC shapes. Output measurements were made on a Varian 2300CD (26 1 cm leaves per side) in solid water for 6 and 20 MV at source-to-axis-distance (SAD = 100 cm) and at two depths, 5 cm (d5) and the clinically prescribed depth (d(clinical)). All measurements were taken under the central axis of the beam unless the shaped field was obviously skewed off axis or the beam was split. Several cases were repeated on a Varian 2100CD (40 1 cm leaves per side) MLC. For the 21 cases studied, the average overall MLC factor (MLC reading divided by the block reading) was found to be 0.9952 with a standard deviation of 0.0024. Repeat measurements performed on the 2100CD were within 0.2% of the MLC factors measured (for the same fields) on the 2300CD. When the data are broken down in terms of clinical applications (type of case at a clinically prescribed depth and energy), the resulting total clinical average is 0.9957 with a standard deviation of 0.0022. Both the overall average and the total clinical average MLC factors (0.995-0.996) agree with previous literature, which used abstract or generic field shapes. Further investigation of the data from this study may find a correlation between the MLC factor and the percent of open field.