In kilovoltage X-ray treatment, beam shaping and shielding normal tissue are accomplished by thin sheets of lead cutout, the thickness of which is selected based upon either published data or measurements. Available broad beam attenuation (BBA) data are found to be unsatisfactory and are the subject of this investigation. BBA is defined as the ratio of intensity with (I) and without (I0) attenuating medium for a large field in a phantom. BBA = I(x,t,E)/I0(x,0,E), where x is the depth of measurement, t is the thickness of attenuator, and E is the beam energy. The depth x should be zero for kilovoltage beams and dmax for megavoltage beams. Unfortunately, x is limited by the window thickness which is the core of this study. A Farmer-type cylindrical ion chamber and three parallel plate ion chambers (Capintec, PS-033; Markus; and Holt) were used to measure BBA for kilovoltage beams from a Siemens Stabilipan unit. Results indicate that attenuation is strongly dependent on the window thickness. For the 240 kVp beam, the thickness of lead for 5% and 1% transmissions are 3.1 mm, and 5.2 mm, respectively, with the Capintec chamber. The corresponding values of lead thickness for the Markus chamber are 2.3 mm and 4.0 mm; for the Holt chamber the values are 1.1 mm and 2.2 mm; and for the cylindrical chambers the values are 1.1 mm and 2.3 mm, respectively. Similar variabilities in lead thickness with ion chambers were also noted for the other kilovoltage beams. The large differences in lead thicknesses produce enormous clinical errors, especially for shielding eye and other critical structures. For small thickness of lead (< 0.1 mm), a 20-fold increase in surface dose could be observed instead of usual beam attenuation. This is due to intense low energy photoelectrons liberated from lead sheets in the contact with tissue. It is concluded that the lead thickness required to shield normal tissue varies with ion chamber. Until national or international guidelines for broad beam transmission measurements are established, the shielding materials in contact with skin should be coated with a thin (> or = 0.3 mm) low atomic number medium. In such a situation, transmission measurements will be independent of the choice of an ion chamber.
Read full abstract