IMPORTANCE OF IN-VIVO DOSIMETRY AS PART OF A QUALITY ASSURANCE PROGRAM IN TANGENTIAL BREAST TREATMENTS

Purpose: The investigation of the accuracy and reproducibility in the daily dose delivered in tangential breast treatments with in vivo dose measurements. Methods and Materials: In vivo dose measurements performed on the tangential treatment fields of 35 breast cancer patients are analysed for three units: a 6 MV linear accelerator, an old Cobalt unit and a new Cobalt unit. The results are plotted in frequency distributions. Deviations on the mean are often the expression of a systematic error in one of the core procedures of a department. A large spread of the results around the mean indicates a high burden of random set-up errors and/or systematic errors in individual patients. The reproducibility in dose delivery is studied by comparing repetitive checks to their respective mean for investigation of random day-to-day variation. Results: A small systematic error on the entrance dose (+ 1.4%) is detected on the old Cobalt unit due to a discrepancy between measured and published percentage depth dose values. An unexpected systematic overdosage (+ 6%) is detected after implementation of a new software for dose calculation, proving that treatment quality is a process needing continuous monitoring. The transmission measurements demonstrate a systematic error in dose delivery of 1.5 to 3% due to the assumption that the breast is water equivalent when calculating the dose. The large spread of the transmission measurements (s(a) = 7.7%) shows that the weakest point in the treatment preparation chain is inaccurate acquisition of external body contours, leading to systematic errors in dose delivery for specific patients. The standard deviation for the reproducibility is 3.1% for the old Cobalt unit, vs. 1.6% on the other units, demonstrating the influence of staffing and mechanical characteristics of the units on daily precision in dose delivery. Conclusion: In vivo dosimetry is an important tool in a departmental quality assurance program to detect systematic errors in dose delivery, to identify inadequate treatment situations, to investigate weak points in the chain of treatment preparation and to ensure accurate dose delivery for individual patients. The predictive value of a single check for the accuracy in dose delivery during the whole treatment series is high for reproducible treatment methodologies.