Effect of geometrical optimization on the treatment volumes and the dose homogeneity of biplane interstitial brachytherapy implants

Background and purpose: The isodose distributions of HDR stepping source brachytherapy implants can be modified by changing dwell times and this procedure is called optimization. The purpose of this study is to evaluate the effect of geometrical optimization on the brachytherapy volumes and the dose homogeneity inside the implant and to compare them with non-optimized counterparts. Material and methods: A set of biplane breast implants consisting of 84 different configurations have been digitized by the planning computer and volumetric analysis was performed for both non-optimized and geometrically optimized implants. Treated length (T-L), treated volume (V-100). irradiated volume (V-50), overdose volume (V-200) and quality index (QI) have been calculated for every non-optimized implant and compared to its corresponding geometrically optimized implant having a similar configuration and covering the same target length. Results: The mean T-L was 74.48% of the active length (A(L)) for non-optimized implants and was 91.87% for optimized implants (P < 0.001). The mean QI was 1.83 for non-optimized implants and 2.17 for optimized implants (P < 0.001). The mean V-50/V-100 value was 2.71 for non-optimized implants and 2.65 for optimized implants (P < 0.001) and the mean V-200/V-100 value was 0.09 for non-optimized implants and 0.10 for optimized implants (P < 0.001). Conclusions: By performing geometrical optimization it is possible to implant shorter needles for a given tumour to adequately cover the target volume with the reference isodose and thus surgical damage is reduced. The amount of healthy tissues outside the target receiving considerable radiation is significantly reduced due to the decrease in irradiated volume. Dose homogeneity inside the implant is significantly improved. Although there is a slight increase of overdose volume inside the implant, this increase is considered to be negligible in clinical applications. (C) 1997 Elsevier Science Ireland Ltd.