Ex vivo light dosimetry and Monte Carlo simulations for endobronchial photodynamic therapy

The light distribution during photodynamic therapy of the bronchial tree has been estimated by measuring the fluence rate in ex vivo experiments on dissected pig bronchi. The trachea was illuminated (630 nm) with a cylindrical diffuser and the fluence rate was measured with a fibre optic isotropic probe. The experiment with the diffuser on the central axis was also simulated with Monte Carlo techniques using the optical properties that were determined with a double-integrating-sphere set-up. The results from ex vivo experiments and the Monte Carlo simulations were found to agree within the error of measurement (15%), indicating that the Monte Carlo technique can be used to estimate the light distribution for varying geometries and optical properties. The results showed that the light fluence rate in the mucosa of the tracheal tract may increase by a factor of six compared to the fluence rate in air (in the absence of tissue). This is due to the scattering properties of the tissue and the multiple reflections within the cavity. Further ex vivo experiments showed that the positioning of the diffuser is critical for the fluence rate in the lesion to be treated. When the position of the diffuser was changed from the central axis to near the lesion, the fluence rate in the mucosa increased significantly by several orders of magnitude as compared to the initial (central) illumination. The inter- and intraspecimen variations in this increase were large (+/- 35%) because of variations in optical and geometrical properties and light source positioning, respectively. These variations might cause under- or overdosage resulting in either insufficient tumour necrosis or excessive normal tissue damage.