Real-time In Vivo Monitoring And Dosimetry Optimizing Of Photodynamic Therapy For Port Wine Stain

Objective: Photodynamic therapy (PDT) has been applied in the treatment of Port wine stain (PWS) for nearly twenty years, but individualized PDT dosimetry still needs to be made according to the features of each PWS patient. This study aimed at improving its treatment efficiency by monitoring the photosensitizer concentration, light distribution and oxygen content in treated PWS skin during PDT, and analyzing their influence on the treatment efficiency.Methods: (1) Fluorescence spectra of PWS skin (thirty-one cases including forty-four areas) treated with PDT were monitored. An algorithm was developed to extract the fluorescence spectra of photosensitizer by correcting the distortions imposed by the absorption of melanin and hemoglobin, and the autofluorescence spectra of skin. Then an index representing the photosensitizer content throughout PDT was calculated according to the area under the curve of photosensitizer content-treatment time. The correlation between therapeutic efficiency and the values of this index was analyzed using Spearman's rank correlation analysis. (2) Twenty-one of the above thirty-one cases were monitored in the second treatment, and divided into three groups according to the way of photosensitizer administration. In the first group (eight areas), photosensitizer was injected intravenously quickly before irradiation as in the previous treatment. In the second group (ten areas), photosensitizer was injected intravenously slowly throughout the first 15min of irradiation. In the third group (seven areas), hot fomentation was made on the treated area for 30min, then photosensitizer was injected quickly as in the first group. The calculated photosensitizer content indexes were compared with those in the previous treatment. (3) An optical model of PWS tissue was established according to the pathological features of PWS, light distribution of 532nm laser in PWS was simulated using Monte Carlo simulation method. Light absorption in each skin layer and light fluence rate distribution in targeted vessels were analyzed. The effect of PWS structure and oxygen content variation on the light distribution was also analyzed. (4) A scheme of light dosimetry optimization by adjusting laser wavelength was proposed and analyzed by simulation. The molar extinction coefficient of PSD-007 at 488nm, 510nm, 532nm and 578nm was calculated by measuring the absorption spectra in PSD-007 solution (albumin buffer) at different concentrations. Light distribution and singlet oxygen yield at these four wavelengths in PWS were simulated using the method established in our lab. (5) Tissue oxygen content in PWS skin was analyzed preliminary by monitoring the reflection spectra of PWS during PDT, based on the different spectral features of HbO₂ and Hb.Results: (1) An apparent individual variation in the measured photosensitizer content throughout PDT was observed. The measured photosensitizer content index correlated well with the therapeutic efficiency of PDT (P=0.0005). (2) In the first group (photosensitizer was administrated in the same way in two treatments), the measured photosensitizer content index had no evident difference. In the second group (photosensitizer was administrated slowly in second treatment), the photosensitizer content index raised in seven of the ten treated areas, while in the third group (hot fomentation), the photosensitizer content index elevated only in two of the seven treated areas. (3) 532nm laser was absorbed mainly in stratum corneum, epidermis, and vessels. When the melanin content in epidermis was high or the target vessels located deeply, the fluence rate in vessels was low. Variation of oxygen content during PDT had no evident effect on fluence rate distribution in PWS. (4) With the same power density and irradiation time, singlet oxygen yield of 510nm PDT was greater than that of 532nm PDT, singlet oxygen yield of 488nm PDT was fair with 532nm, and the singlet oxygen yield of 578nm PDT was lower than that of 532nm PDT. For PWS with high melanin content or deeply located vessels, 510nm PDT generated more singlet oxygen in target vessels than 532nm PDT. (5) Before PDT, the tissue oxygen content was different in different types of PWS. During PDT, the tissue oxygen content was not evidently declined in majority of the P4 and P5 PWS, but declined apparently in all of the P6 PWS.Conclusions:(1) The algorithm of fluorescence spectra analysis established in this study can correct the distortions on photosensitizer fluorescence spectra imposed by the absorption of melanin and hemoglobin, and the autofluorescence spectra of skin. There was an apparent individual variation in the measured photosensitizer content throughout PWS treatment, and the measured photosensitizer content index correlated well with the therapeutic efficiency, suggesting that it can bring some help to predict the treatment outcome.(2) The measured photosensitizer content index can be raised when photosensitizer is injected intravenously slowly during irradiation. Hot fomentation before irradiation can only raise photosensitizer content temporally, but can't increase the photosensitizer content throughout the treatment.(3) Melanin content in epidermis and depth of vessels in dermis are the major factors that influence 532nm laser distribution in PWS. Oxygen content variation during PDT has no evident effect on the fluence rate distribution in PWS.(4) 510nm PDT has the potential to increase singlet oxygen yield in target vessels of PWS with a high melanin content or deeply located vessels.(5) Skin oxygen content is different in various types of PWS before PDT. During PDT, skin oxygen content does not evidently decline in majority of the P4 and P5 PWS, but declines apparently in P6 PWS.