Purpose
For the estimation of the late complications of radiotherapy treatment in cervix cancer patients, the volume of a dose delivered to the organs at risk is significant. With regards to the computation of dose to points or volumes of organs at risk, in vivo dose measurements were performed with semiconductor detectors or with thermoluminescence dosimetry [1-3]. The accuracy of dose measurements with semiconductor detectors is influenced by several factors, such as: angular change in sensitivity of the diode, the temperature dependence [4], the change of sensitivity as a function of depth, the change in calibration factor, a dose/dose rate dependence [5], the energy dependence and variation of sensibility with accumulated dose. Additionally, the high gradient in dose distribution near brachytherapy applicator resulted in a large difference between computed and measured dose values which was probably due to a movement of a probe during the treatment fraction [6]. The aim of our investigation was to check the consistency of the Fletcher-Suit (FS) applicator geometry and the stability of position of the rectal probe’s during the course of brachytherapy treatment fraction.
Material and methods
We considered 14 treatment fractions of 10 patients treated with the Fletcher-Suit applicator with 15 degrees tandem (Nucletron®, Netherlands). Oval ovoid’s spacers of the same form and dimensions (round or D-profile) were applied for each treatment fraction. The in vivo dose measurements were performed with five-channel rectal probe (PTW, Germany). We inserted radio-opaque markers spaced 1 cm into the applicator and obtained post-anterior (PA) and posterior oblique (PO) reconstruction images. The possible applicator displacement was avoided due to transportation of the patient and in situ images in brachytherapy treatment room were acquired with the use of C-arm (Mobilgil AR7®, Giraldoni, Italy). The video signal from the C-arms monitor was recorded [7] and treatment plans with dicomized images using PLATO Brachytherapy Treatment Planning System (BPS) v13.7 (Nucletron®) were prepared. The dose distribution was normalized to the ICRU Report 38 reference points (A) and 7 Gy dose per fraction was administered. The patient lied in a lithotomic position during the treatment planning and dose administration. Pairs of post-treatment reconstruction images were obtained and a new treatment plan was prepared. The post-treatment images were acquired with the same reconstruction angles as the pre-treatment images, but with a slight difference in magnification and in the position of the central beam axis of the applicator. For comparison of corresponding images, we registered them with a landmark that initiated a mutual information registration method. Due to lack of applicator shift of the bony structures, the registered images showed both the FS applicator and the pelvis bones overlap, whereas in case of applicator movement, the bony structures appeared shifted. We estimated the extent of the shift with the radio-opaque markers spaced 1 cm in a horizontal position. For checking the relative position of the tandem and the ovoid’s, we estimated the reconstructed three – dimensional pre- and post-treatment coordinates of the first, the 7th radio-opaque markers inserted into the tandem, the second markers from the ovoid’s tip and the coordinates of the rectal probe’s detectors.
Results
The time stamp of the pre- and post-treatment reconstruction images differed between 25 and 50 minutes, the average 34 minutes. In all treatment fractions a good match in reconstructed pre- and post-treatment coordinates of the first and the 7th tandem markers and in the ovoid markers were experienced. The average three-dimensional difference in marker coordinates was 1.6 ±1.2 mm, while for the rectal probe’s detectors was 8 ±6 mm. Based on the evaluation of the registered pre- and post-treatment reconstruction images, we separated the treatment fractions into two groups (Table 1). In Group A, on the pre- and post-treatment reconstruction images the bony structures showed good match, and the shift in rectal probe’s position was smaller than 5 mm. The difference in reconstructed three-dimensional coordinates of pre- and post- treatment position of the rectal probe’s detectors was in average 6.5 ±2.4 mm. The standard deviation of computed dose to points of the semiconductor detectors was 5% of the reference dose (7 Gy). These values for insertions with a shift larger than 5 mm (Group B) were 10 ±9 mm and 11%, respectively.
Discussion and Conclusion
The lateral and sagittal position of the ovoid’s with regard to the tandem showed good consistency during brachytherapy treatment fraction. These parameters contain the main influence on the FS applicators dose distribution [8]. The three-dimensional difference of pre- and post-treatment applicator coordinates were within the error of reconstruction accuracy, comparing to those of larger extent in rectal probe’s semiconductor detectors. Despite the same position of a patient during the whole treatment, some significant applicator or rectal probe displacement with respect to bony structures were experienced during 30% of treatment fractions. In case of a difference bigger than 5% between the dose values at points of the semiconductor detectors computed by the treatment planning system and the readout from in vivo detectors, the applicator and rectal probe position with post-treatment reconstruction images was verified.
Acknowledgements
The author would like to thank Dr Kofi Agyemang-Prempeh, MD, for performing the insertions for the study.
References
1. Eich H, Haverkamp U, Micke O et al. Zur problematik dosimetrischer verfahren bei der bestimmung der strahlendosis am rektum in der HDR-brachytherapie des zervixkarzinoms. Strahlenther Onkol 1998; 174: 375-380.
2. Stuecklschweiger GF, Arian-Schad KS, Poier E et al. Bladder and rectal dose of gynaecologic high-dose rate implants: comparison of orthogonal radiographic measurements with in vivo and CT-assisted measurements. Radiology 1991; 181: 889-894.
3. Huh JS, Lim DH, Ahn YC et al. Comparison between in vivo dosimetry and Bariumcontrast technique for prediction for rectal complication in high dose rate intracavitary radiotherapy in cervix cancer patients. Strahlenther Onkol 2003; 179: 191-196.
4. Van Dam J, Leunens G, Dutreix A. Correlation between temperature and dose rate dependence of semiconductor response; influence of accumulated dose. Radiother Oncol 1990; 19: 345-351.
5. Wilkins D, Li X-A, Cygler J et al. The effect of dose rate dependence of p-type silicon detectors on linac relative dosimetry. Med Phys 1997; 24: 879-881.
6. Waldhausl C, Wambersie A, Pötter R et al. In-vivo dosimetry for gynaecological brachytherapy: physical and clinical considerations. Radiother Oncol 2005; 77: 310-317.
7. Pálvölgyi J. To what extent can digital images obtained with a non-isocentric C-arm be used for brachytherapy treatment planning in gynaecology. Radiother Oncol 2006; 67: 107.
8. Pálvölgyi J. Influence of different Fletcher-Suit applicator geometries on sagittal dose distribution. Phys Med 2009 [accepted for publication].
