Optical Characterization of Neurosurgical Operating Microscopes: Quantitative Fluorescence and Assessment of PpIX Photobleaching

Abstract

Protoporphyrin IX (PpIX) induced by 5-aminolevulinic acid (5-ALA) is increasingly used as a fluorescent marker for fluorescence-guided resection of malignant gliomas. Understanding how the properties of the excitation light source and PpIX fluorescence interact with the surgical microscope is critical for effective use of the fluorescence-guided tumor resection technique. In this study, we performed a detailed assessment of the intensity of the emitted blue light and white light and the light beam profile of clinical grade operating microscopes used for PpIX visualization. These measurements revealed both recognized fluorescence photobleaching limitations and unrecognized limitations that may alter quantitative observations of PpIX fluorescence obtained with the operating microscope with potential impact on research and clinical uses. We also evaluated the optical properties of a photostable fluorescent standard with an excitation-emission profile similar to PpIX. In addition, we measured the time-dependent dynamics of 5-ALA-induced PpIX fluorescence in an animal glioma model. Finally, we developed a ratiometric method for quantification of the PpIX fluorescence that uses the photostable fluorescent standard to normalize PpIX fluorescence intensity. This method increases accuracy and allows reproducible and direct comparability of the measurements from multiple samples.

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1

Optical power measurement experiments. Incident BLUE 400 optical power profile across the field of view of the operating microscope. Unless otherwise specified, measurements were performed at 20-mm focus distance and 100% microscope light power settings. (a) Diagram showing where the measurements were taken. (b) Light intensity profile across the horizontal axis of the field of view. (c) Light intensity profile across the vertical axis of the field of view. (d–f) Results of optical power assessment with different operating microscopes. (d) Graph showing absence of correlation between the remaining lamp hours and optical power measured at the center of the field of view. (e) Graph showing results of repetitive optical power measurements in operating microscopes spanning 83 days. There was no clear association with use time. Pentero microscopes 2 and 9 were assessed only once because of their limited availability. (f) Graph showing association between the microscope optical power setting and the measured incident optical power at a 20-cm focus distance in 8 operating microscopes. (g) Graph showing association between the focus distance and the measured incident optical power at 5 different microscope light power settings. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

Figure 2

Fluorescence in dye-in-polymer standards. (a) Fluorescent intensity in each of the 9 standards at time zero, labeled by dye concentration and thickness of the sheet. The intensity value increased in a linear fashion as thickness was increased for each specific concentration. This rate of linear increase varied depending on the concentration of the fluorescent dye in a standard. (b) Three-dimensional diagram of the fluorescence intensity of standard 7 under the beam of BLUE 400 light. The intensity is plotted in Z axis and color coded, the X and Y axes are on the surface. (c) Rate of photobleaching of 10 regions over the course of 30 min for standard 7. (d) Image of the dye-in polymer standard sheet under the operating microscope BLUE 400 light showing 10 uniform size regions (1–10) that were selected for intensity analysis (shown in c). Ppt, parts-per-thousand by weight; A.U., arbitrary units. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

Figure 3

Fluorescence decay in dye-in-polymer standards. Logarithmic fluorescence decay at location 1 on standard 7 over 30 min. Logarithmic regression equation with R2 value is included. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

Figure 4

Tumor-to-background ratios calculated from averages of all tumor and background fluorescence values captured using the external CMOS camera or the internal CCD camera of the operating microscope. The internal camera was not sensitive enough to record meaningful fluorescence data. The averages were calculated from 6 different samples for each time point. PpIX, protoporphyrin IX; CMOS, complementary metal–oxide–semiconductor; CCD, charge-coupled device. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

Figure 5

Dynamics of PpIX photobleaching in the tumor and dye-in-polymer standards. (a) Raw data. (b) Data were normalized, using standard 1 as a 10% reference point. Presented as mean ± SD. A.U., arbitrary units. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

Figure 6

Quantitative assessment of PpIX fluorescence in six brain tumors under the operating microscope. Observation within 30 min of continuous BLUE 400 light. (a) Raw tumor fluorescence values. (b) Tumor-to-background ratios calculated from the raw data. (c) Calculated tumor fluorescence values after normalization to the standard 6. Normalization, which allowed for direct comparison of the PpIX intensities among specimens, revealed that tumor 4 had significantly lower absolute tumor fluorescence. A.U., arbitrary units. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

Figure 7

Experimental setup. (a) Fluorescence spectra of the dye-in-polymer standard and 5-ALA–induced PpIX in mouse glioma. All standards showed perfect alignment of the emission spectra. Results from standard 1 are presented on the graph. Measurements performed on a Zeiss LSM710 laser scanning spectral confocal microscope, EC Plan-Neofluar 10×/0.3 objective, and 405 laser using the lambda mode with 32-channel detector covering range 419–721 nm. Mouse glioma imaging was performed ex vivo immediately after brain harvesting, 2 hours after 5-ALA administration. Data are presented as normalized % values to the maximum peak value. (b) Fluorescence spectra of protoporphyrin IX (PpIX) dimethyl ester dissolved in chloroform and the organic europium pigment dispersed in polymer. PpIX data were downloaded from http://omlc.org/spectra/PhotochemCAD/html/149.html which used data from Photochem CAD package, version 2.1a (Du 1998, Dixon 2005). (c) Setup for quantitative measurement of the PpIX signal and photobleaching in glioma tissue. Operating microscope is shown at left, external camera at right. (d) Operating room setup for quantitative measurement of PpIX fluorescence and normalization to the standard. Yellow rectangles show regions of interest selected for analysis of standards. (e) Two regions of interest were selected on 6 mouse brains: 1 over the tumor and 1 over the normal brain. Left image shows raw red channel, right image shows corresponding channel in a heatmap lookup view. A.U., arbitrary units; 5-ALA, 5-aminolevulinic acid; PpIX, protoporphyrin IX. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.