Johannes P. A. Marijnissen

Light dosimetry in optical phantoms and in tissues: I. Multiple flux and transport theory

This is the first of two papers on the quantitative measurement of light energy fluence rates in optical phantoms and in tissues, in vitro and in vivo. The theory discussed in the present paper will be used in a forthcoming experimental paper to quantitatively check measurements of light energy fluence rates. A simple multiple flux model, which is equivalent to the diffusion approximation, is derived from the equation of transfer in a plane as well as in a spherical geometry. The equations obtained are similar to those of the Kubelka-Munk and related heuristic models. This permits conclusions regarding the limitations of these models and the values of their constants. The heuristic models are equivalent to diffusion theory for diffuse incident light, but not for collimated incident light. We also present a simple calculation of the radiance as a function of direction in the diffusion domain. This, together with the effective attenuation coefficient, permits indirect experimental determination of both the albedo and the anisotropy factor (g) of the scattering function. Similarity relations are discussed, as they result from the so called delta-Eddington approximation, leading to the conclusion that far from boundaries and sources light propagation characteristics do not change very much when g and omega s are varied, provided omega s (1-g) is kept constant (omega s = scattering coefficient). Therefore, only two optical constants are required to approximately describe light propagation in homogeneous and isotropic media in the diffusion approximation.

Measurements and calculations of the energy fluence rate in a scattering and absorbing phantom at 633 nm

We have studied the influence of absorption, scattering, and refractive index of a phantom medium in conjunction with various beam diameters on the penetration depth of light at 633 nm. We used mixtures of Intralipid 10% (scattering medium) and Evans blue (absorbing medium). Measurements were performed in media with a scattering coefficient of 1 mm(-1), an anisotropy factor of 0.71, absorption coefficients of 1.3 x 10(-3), 0.01, and 0.05 mm(-1), and a refractive index of 1.33. The experimental results were compared with an analytical solution of the fluence rate based on diffusion theory. We found good agreement (deviations of <10%) between theory and experiment for incident beam diameters between 10 and 60 mm.

New trends in photobiology light dosimetry: Status and prospects

This paper is a report on the state of the art of light dosimetry in photomedicine and photobiology. The basic quantity of interest is the radiant energy fluence rate, which can either be measured using a suitable probe, or calculated theoretically from measured optical constants. First, theoretical models used to analyse experimental transmission and reflection data are briefly discussed. It is shown that a two-flux model derived from the transport equation in the diffusion approximation resembles the Kubelka-Munk and other heuristic models. This illustrates the limitations of these models and suggests their abandonment in favour of transport theory. For theoretical energy fluence rate calculations at least three optical constants are needed, namely the absorption coefficient, the scattering coefficient and the average cosine of the scattering angle. These three constants have been measured for very few tissues. In principle only two of the three constants can be measured directly on thin samples, independent of a theoretical model. The energy fluence rate can be measured quantitatively with a miniature fibre optic probe with isotropic response. Such measurements allow indirect determination of the three optical constants. It appears that we are just beginning to understand the distribution of light energy fluence rate in tissues. Tasks for the near future are comparison of methods to measure optical constants, quantitative checks of calculated and measured energy fluence rates in model tissues and optical phantoms and further development of theoretical models. Particular attention is required for boundary conditions, with and without refractive index matching.

Calculating the response of isotropic light dosimetry probes as a function of the tissue refractive index

The response of an isotropic light dosimetry probe as function of the tissue refractive index (n) has been calculated using diffusion theory. The response has also been measured in a collimated light beam with the probe in air, water (n = 1.33), ethylene glycol (n = 1.43), and glycerin (n =1.46). For a 3.2-mm diam probe with little light absorption, the theoretical result depends only on n and differs from the experimental data by not more than 6%. For a 0.8-mm diam probe with some light absorption, excellent agreement between theory and experiment is obtained by adjusting the absorption and the reduced scattering coefficients of the probe material.

PILOT STUDY ON LIGHT DOSIMETRY FOR ENDOBRONCHIAL PHOTODYNAMIC THERAPY

Endobronchial photodynamic therapy (EB‐PDT) using photofrin as the photosensitizer is currently being evaluated as a new treatment modality for inoperable endobronchial tumors. One of the current problems with EB‐PDT is the lack of adequate light dosimetry, which hampers proper interpretation of treatment results. In this study exploratory light dosimetry experiments were performed in plastic bronchus models using either a microlens‐tipped fiber (suitable for illumination of small superficial tumors) or a cylindrical diffuser fiber (suitable for intraluminal illumination or interstitial illumination of partially obstructing tumors). It is shown that the light fluence prescriptions of current clinical protocols yield a different fluence in tissue for each illumination modality. Depending on the actual placement of the cylindrical diffuser within the lumen, the light fluence at 5 mm depth in the homogeneous tissue model may vary by a factor of 3. The results were confirmed by in vivo experiments in the trachea of a pig. There is a possibility of enhanced tissue response by accidental hyperthermia induced during EB‐PDT. The temperature rise was therefore estimated in vivo using a rat tumor model to mimic clinical EB‐PDT. Temperature rises of at least 5°C and 10°C can be expected for intraluminal and intratumoral illumination, respectively, at 3.5 ± 1 mm depth in tissue and 400 mW/cm diffuser output. Light fluence and its distribution in the bronchus strongly depend on the geometry and the optical properties of the tissue as well as on the technique of illumination. As a result of inadequate dosimetry, significant variations in treatment response between patients may be expected.

Use of sheet color filters for video‐endoscopic observation during intraluminal photodynamic therapy

Photodynamic therapy (PDT) is currently evaluated in clinical studies for the treatment of bronchial and oesophageal tumors.

Whole bladder wall photodynamic therapy with in-situ light dosimetry for carcinoma in situ of the bladder

We report on 15 patients with multifocal carcinoma in situ of the bladder, treated with whole bladder wall photodynamic therapy (PDT). The total light dose, measured in situ (scattered plus nonscattered light) was 100 J/cm2 in the first six patients and 75 J/cm2 in the remaining nine patients. Follow-up ranges were from 6 to 27 months (average 15 months). Two cystectomies had to be performed in the first treatment group because of permanent shrunk bladders. Pathology of the resection specimens showed extensive granulation and fibrosis throughout the whole bladder wall. In the second treatment group, the maximal bladder capacity measured three months after PDT had increased on the average of 63% compared to the initial pretreatment values. No increased fibrosis could be detected on microscopical examination of random biopsies. Four recurrences necessitated cystectomy after 5 to 9 months, two in each treatment group. Three out of these originated in patients with a previous history of invasive bladder cancer. The preliminary data demonstrate the importance of in-situ light dosimetry for minimizing local side effects of PDT as well as the importance of strict inclusion criteria to optimize the therapeutic ratio.

Light dosage in whole bladder wall photodynamic therapy at 532 and 630 nm

The optical absorption, scattering and anisotropy coefficient of piglet and diseased human bladder tissue were determined in vitro in the wavelength range of 500 - 650 nm. Monte Carlo simulations for whole bladder wall (WBW) photodynamic therapy (PDT) have been performed using the optical parameters determined in vitro. The calculated light dose rate values are in agreement with those measured in clinical WBW-PDT, previously performed at 630 nm. The light dose rate at the bladder wall can differ by a factor of 4, due to variations in optical properties of the tissue. This demonstrates the necessity of in situ light dosimetry during clinical WBW-PDT. WBW-PDT with red light (630 nm) will be technically more advantageous than with green light (532 nm), because of a higher integrating sphere effect.