The influence of tissue and applicator heterogeneities on brachytherapy dose distributions is not well understood, despite widespread use of shielded applicators in intracavitary therapy. Heterogeneity correction factors (HCF) have been measured using a silicon diode detector arising from bounded heterogeneities consisting of lead, steel, titanium, silver, aluminum, and air cylinders near brachytherapy sources of I-125, Cs-137, and Ir-192. In addition, transverse-axis dose distributions for the three sources in homogeneous water were measured for distances of 0.2 to 16.0 cm. For each point of measurement, relative diode readings were simulated by a Monte Carlo photon transport code utilizing accurate models of the source internal structure, the experimental measure geometry and the source-strength calibration geometry. Comparison of measured and calculated HCF's reveals excellent agreement (1%-3% average) over a wide range of materials, diameters, and thicknesses. In addition, Monte Carlo simulation not only accurately reproduced the relative transverse-axis dose distributions in homogeneous medium, but was able to predict the variation of diode response with photon energy with an accuracy of 3% over the range of 30-662 keV. Our measurements demonstrate that HCF's vary by as much as 60%-100% with distance and heterogeneity diameter for a fixed thickness. Finally, silicon diode measurements of HCF (defined as reading with heterogeneity/reading in homogeneous medium) is shown to lead to errors of 5%-30% for Cs-137 and Ir-192 sources in the presence of high-atomic number shielding materials. This paper concludes, that Monte Carlo simulation is a powerful, convenient and accurate tool for investigating the long-neglected area of brachytherapy heterogeneity corrections.
