This paper, which is divided into parts I and II, describes the physical aspects of work on total-body irradiation (TBI) at the Middlesex Hospital, London, from 1988 to 1993. Irradiation is fractionated and bi-lateral with horizontal accelerator photon beams of 8 MV (1988 - 1992) at a source - surface distance (SSD) of 3.36 m and 10 MV (1992 - 1993) at an SSD of 4.62 m. The main aims were maximum patient comfort, a simple, accurate set-up with overall times per fraction of 30 min or less, dose homogeneity throughout the body within to , pre-irradiation treatment planning on nine CT slices using our commercial IGE RTplan (1988 - 1992) and Target 2 (1992 - 1993) treatment planning systems and, most important, verification of the plans by in vivo dosimetry to within . Verification of the planned lung doses, which are distributed over five CT slices, was given special attention.
In part I of this paper we describe the preliminary work, most of which was done prior to patient treatment. This consisted of standard dosimetric measurements (central axis depth doses, beam profiles at several depths, build-up and build-down curves, beam output calibrations, effect of body compensators, etc), in evaluating silicon diode dosimeters for in vivo dosimetry and of adapting and verifying the methods of treatment planning for TBI conditions. The results obtained with phantoms, including a Rando body phantom, showed that, in principle, our aims could be achieved. The final proof depended, however, on an analysis of the results of the in vivo work and this forms the subject of part II of this paper.