Optical technologies for intraoperative neurosurgical guidance

Abstract

Biomedical optics is a broadly interdisciplinary field at the interface of optical engineering, biophysics, computer science, medicine, biology, and chemistry, helping us understand light-tissue interactions to create applications with diagnostic and therapeutic value in medicine. Implementation of biomedical optics tools and principles has had a notable scientific and clinical resurgence in recent years in the neurosurgical community. This is in great part due to work in fluorescence-guided surgery of brain tumors leading to reports of significant improvement in maximizing the rates of gross-total resection. Multiple additional optical technologies have been implemented clinically, including diffuse reflectance spectroscopy and imaging, optical coherence tomography, Raman spectroscopy and imaging, and advanced quantitative methods, including quantitative fluorescence and lifetime imaging. Here we present a clinically relevant and technologically informed overview and discussion of some of the major clinical implementations of optical technologies as intraoperative guidance tools in neurosurgery.

Keywords: ALA-PpIX = 5-aminolevulinic acid-induced protoporphyrin IX; BBB = blood-brain barrier; CARS = coherent anti-Stokes Raman scattering; CBF = cerebral blood flow; CCD = charge-coupled device; GBM = glioblastoma multiforme; GTR = gross-total resection; ICG = indocyanine green; LSCI = laser speckle contrast imaging; NIR = near-infrared; OCT = optical coherence tomography; Raman spectroscopy and imaging; SRS = stimulated Raman scattering; biomedical optics; diffuse reflectance spectroscopy and imaging; fluorescence lifetime; fluorescence-guided surgery; image-guided surgery; laser speckle contrast imaging; optical coherence tomography; λem = emission wavelength; λex = excitation wavelength.

FIG. 1

Intraoperative paired white light and fluorescence mode images during multiple stages of PpIX fluorescence-guided resection of a GBM, showing strong visible PpIX fluorescence at the beginning (A and B), middle (C and D), and near end (E and F) of surgery, and no visible fluorescence at the end (G and H) (heads up display contour lines in green). Heads up display corresponding to segmented contrast-enhancing tumor demonstrates strong PpIX fluorescent tissue outside (F) the segmented region near the end of the resection and no visible PpIX fluorescent tissue at the end (H). The patient underwent fluorescence-guided resection under a protocol approved by the institutional review board at Dartmouth-Hitchcock Medical Center.

FIG. 2

A and B: Intraoperative images of fluorescein sodium in human gliomas showing intraoperative green fluorescence in the tumor bed with their corresponding T1-weighted contrast-enhanced MR images (C and D). Green fluorescence in the tumor bed (arrows) corresponds to tumor tissue but nonspecific diffuse fluorescence is also visible at the dural flap. Reproduced with permission from Acerbi et al: Neurosurg Focus 36(2):E5, 2014.

FIG. 3

In vivo quantitative fluorescence imaging in gliomas showing ALA-PpIX fluorescence-guided surgery at the beginning (A–C), middle (E–G), and end of surgery (I–K), with visible fluorescence images (B, F, and J) and corresponding quantitative fluorescence images following correction for tissue optical properties noting improved tumor tissue and fluorophore detection (C, G, and K). D and H: Spectra showing presence (D) and absence (H) of PpIX signal during and after surgery. L: Histologically confirmed tumor tissue corresponding to location in panel G with white arrow. Reproduced with permission from Valdés et al: Sci Rep 2:798, 2012.

FIG. 4

Diagram of the intraoperative spectroscopy dual fluorescence system and reflectance probe showing the handheld contact probe, collection of dual fluorescence and reflectance signals, advanced light transport processing algorithms, and machine learning technique for tissue classification. Reproduced with permission from Valdés et al: J Biomed Opt 16:116007, 2011.

FIG. 5

Images showing label-free brain tumor imaging with stimulated Raman scattering (SRS) microscopy. A: SRS imaging of lipid (at 2854 cm⁻¹), protein (at 2930 cm⁻¹), and red blood cells (at 2800 cm⁻¹) of a fresh brain tumor tissue from an oligodendroglioma surgical case (WHO Grade II). Myelin fibers were visualized based on the strong lipid signals (green), cell nuclei and blood vessels were visualized mainly due to their strong protein signals (blue), and red blood cells were imaged based on the non-Raman 2-color 2-photon absorption signals (magenta). B: SRS and H & E imaging of the same frozen section of an entire tissue specimen from an oligoastrocytoma surgical case. Lipid/protein mapping of the brain tumor tissue with SRS provided very similar information to H & E staining with cell-to-cell correlation (B lower zoom images), demonstrating that SRS imaging could be used for label-free neuropathology. Scale bars = 100 μm (A and B upper), 1 mm (B lower). Brain tissue samples were obtained under an institutional review board–approved protocol from the Brigham and Women’s Hospital and Dana-Farber Cancer Institute Neuro-oncology Program Biorepository collection.