Thermally determined excitation power limits for fiber-based fluorescence diagnosis based on ICG and PpIX

Objective: For fluorescence-guided stereotactic brain tumor biopsy via fiber-based fluorescence detection of protoporphyrin IX (PpIX) (typ. excitation at 405nm, emission at 633nm) which selectively accumulates in tumor tissue, an additional method for simultaneous blood vessel detection ahead of the distal fiber end shall be developed in order to minimize the risk of blood vessel perforation. For this purpose, the infrared fluorescence of the intraveneously administered intraluminal fluorophore indocyanin-green (ICG) (typ. excitation at 785nm, emission at 835nm) shall be detected through the same fiber-optic system. To determine which fluorescence signal levels can realistically be expected, the maximum tolerable excitation light power was derived from simulations of the thermal heat load on the tissue. The results were compared to the respective threshold values resulting from legal regulations.

Material and methods: Using the LITCIT simulation software (Roggan & Müller 1995; Roggan 1997, 2001; LMTB – Laser- und Medizin-Technologie GmbH, Berlin, 2001) the temperature distribution in human brain tissue was calculated as a function of time for a fixed focus diameter (here 0.29mm), and the asymptotic maximum temperature in the entire illuminated tissue region was investigated as a function of the illumination power. Worst case values were assumed for all involved parameters, such as optical parameters of blood and brain tissue, the blood perfusion of brain tissue etc. In the case of ICG, typical initial ICG concentrations after a bolus were asssumed. In addition to homogeneous brain tissue, we also investigated models with incorporated blood vessels, considering different diameters (100µm and 10µm) and distances from the distal fiber end (0.25mm and 1.25mm).

Results: For ICG/785nm (PpIX/405nm) we obtain a maximum tolerable laser power of 37 mW (5.7 mW) at the distal fiber end if thermal damage of unstructured normal brain tissue should be excluded without catheter cooling. In the case of ICG, the incorporation of blood vessels reduced the tolerable laser power to 25 mW. In the case of PpIX the tolerable laser power remained unchanged, since here the penetration depth of the excitation light is very small. From legal regulations for human skin tissue one would obtain threshold values of 28.5 mW for 785nm and 19.2 mW for 405nm (Lasernorm 60825–1, BGV B2, BGI 832), considering the prescribed power averaging protocols.

Conclusion: For the envisaged applications of fluorescence-guided stereotactic brain tumor biopsy via fiber-based fluorescence detection of PpIX and detection of larger blood vessels via fluorescence detection of ICG, we have obtained light power limits for an application-relevant fiber configuration. In the case of PpIX, we obtain a power limit significantly below the one derived from legal regulations.

Acknowledgement: This work was carried out with the support of Bundesministerium für Forschung und Technologie (FKZ 13N10172– NEUROTAX).