Clinical experience with routine diode dosimetry for electron beam radiotherapy

Electron beam radiotherapy is frequently administered based on clinical setups without formal treatment planning. We felt, therefore, that it was important to monitor electron beam treatments by in vivo dosimetry to prevent errors in treatment delivery. In this study, we present our clinical experience with patient dose verification using electron diodes and quantitatively assess the dose perturbations caused by the diodes during electron beam radiotherapy.

A commercial diode dosimeter was used for the in vivo dose measurements. During patient dosimetry, the patients were set up as usual by the therapists. Before treatment, a diode was placed on the patient's skin surface and secured with hypoallergenic tape. The patient was then treated and the diode response registered and stored in the patient radiotherapy system database via our in-house software. A customized patient in vivo dosimetry report showing patient details, expected and measured dose, and percent difference was then generated and printed for analysis and record keeping. We studied the perturbation of electron beams by diodes using film dosimetry. Beam profiles at the 90% prescription isodose depths were obtained with and without the diode on the beam central axis, for 6-20 MeV electron beams and applicator/insert sizes ranging from a 3-cm diameter circular field to a 25 x 25 cm open field.

In vivo dose measurements on 360 patients resulted in the following ranges of deviations from the expected dose at the various anatomic sites: Breast (222 patients) -20.3 to +23.5% (median deviation 0%); Head and Neck (63 patients) -21.5 to +14.8% (median -0.7%); Other sites (75 patients) -17.6 to +18.8% (median +0.5%). Routine diode dosimetry during the first treatment on 360 patients (460 treatment sites) resulted in 11.5% of the measurements outside our acceptable +/-6% dose deviation window. Only 3.7% of the total measurements were outside +/-10% dose deviation. Detailed investigations revealed that the dose discrepancies, overwhelmingly, were due to inaccurate diode orientation and positioning, especially in the regions with rapidly changing contours and/or sloping surfaces. The presence of a diode in the treatment field was found, in some cases, to cause significant dose reduction, most noticeable with smaller fields and lower energy beams. The reduction in dose ranged from 16% (for a 6 MeV beam and a 3 cm diameter circular field) to 4% (for a 12 MeV beam and a 10 x 10 cm field).

Diode dosimetry was found to be convenient and valuable for verifying in real time the dose delivery accuracy of electron beam treatments, but with some caveats. When treating a small field by low energy electrons, frequent use of diodes is undesirable, because it might result in appreciable reduction of dose to the target volume.