Confocal Neurolasermicroscopy in Human Brain – Perspectives for Neurosurgery on a Cellular Level (including additional Comments to this article)

Abstract

Background: During neurosurgery intraoperative imaging of vital neural structures on a cellular level would facilitate the development of new strategies for diagnosis and treatment. In vivo imaging would permit the detection of the tumour centre and infiltration zone. With targeted biopsies the lesion of interest could be determined before performing the biopsy, facilitating the final pathological diagnosis. In this study we present confocal neurolasermicroscopy as a new method in neurosurgery.

Methods: A miniaturised confocal neurolasermicroscope (NLM) was used ex vivo immediately after tumour resection of glioblastoma multiforme (GBM). NLM was performed with subcellular magnification up to a tissue depth of 100 μm. NLM images were compared to conventional histological images of the same tumour.

Results: The application of the method in nine patients allowed adequate diagnosis of a malignant glioma fulfilling the WHO criteria when compared to conventional histology. In one patient with glioblastoma multiforme NLM allowed the correct diagnosis of GBM to be made, demonstrating the high mitotic rate and cell pleomorphy of the tumour cells. Additional characteristics such as pleomorphic cells, mitotic figures, fibrillary matrix and the distinction between tumour centre and infiltration zone could be shown.

Conclusions: NLM is a tool which could be adapted for neurosurgical intraoperative applications with the potential to diagnose tumours and recognise the tumour centre and infiltration zone in vivo. Further applications of NLM to characterise subcellular structures and vascular architecture are possible.

Key words

endomicroscopy - brain tumour - histology - intraoperative diagnosis - neurolasermicroscopy

References

• 1 Ammirati M, Vick N, Liao YL. et al . Effect of the extent of surgical resection on survival and quality of life in patients with supratentorial glioblastomas and anaplastic astrocytomas.  Neurosurgery. 1987;  21 201-206
• 2 Asthagiri AR, Pouratian N, Sherman J. et al . Advances in brain tumor surgery.  Neurol Clin. 2007;  25 975-1003 , , viii–ix
• 3 Bojarski C, Günther U, Rieger K. et al . In vivo diagnosis of acute intestinal graft-versus-host disease by confocal endomicroscopy.  Endoscopy. 2009;  41 433-438
• 4 Burger PC, Green SB. Patient age, histologic features, and length of survival in patients with glioblastoma multiforme.  Cancer. 1987;  59 1617-1625
• 5 Chou SN. Embolectomy of middle cerebral artery: Report of a case.  J Neurosurg. 1963;  20 161-163
• 6 Ferenc T, Janik-Spiechowicz E, Bratkowska W. et al . Mutagenic activity of 3,6-diamino-10-methyl-9,10-dihydroacridine in Salmonella typhimurium cells.  Int J Occup Med Environ Health. 1999;  12 67-72
• 7 Goetz M, Kiesslich R, Dienes HP. et al . In vivo confocal laser endomicroscopy of the human liver: a novel method for assessing liver microarchitecture in real time.  Endoscopy. 2008;  40 554-562
• 8 Goetz M, Thomas S, Heimann A. et al . Dynamic in vivo imaging of microvasculature and perfusion by miniaturized confocal laser microscopy.  Eur Surg Res. 2008;  41 290-297
• 9 Goetz M, Vieth M, Kanzler S. et al . In vivo confocal laser laparoscopy allows real time subsurface microscopy in animal models of liver disease.  J Hepatol. 2008;  48 91-97
• 10 Hoffman A, Goetz M, Vieth M. et al . Confocal laser endomicroscopy: technical status and current indications.  Endoscopy. 2006;  38 1275-1283
• 11 Jacobson 2nd JH, Wallman LJ, Schumacher GA. et al . Microsurgery as an aid to middle cerebral artery endarterectomy.  J Neurosurg. 1962;  19 108-115
• 12 Lacroix M, Abi-Said D, Fourney DR. et al . A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival.  J Neurosurg. 2001;  95 190-198
• 13 Lademann J, Otberg N, Richter H. et al . Application of optical non-invasive methods in skin physiology: a comparison of laser scanning microscopy and optical coherent tomography with histological analysis.  Skin Res Technol. 2007;  13 119-132
• 14 Leong R, Nguyen N, Meredith C. et al . In vivo confocal endomicroscopy in the diagnosis and evaluation of celiac disease.  Gastroenterology. 2008;  135 1870-1876
• 15 Lipson BK, Yannuzzi LA. Complications of intravenous fluorescein injections.  Int Ophthalmol Clin. 1989;  29 200-205
• 16 Lougheed WM, Marshall BM, Hunter M. et al . Common carotid to intracranial internal carotid bypass venous graft.  Technical note. J Neurosurg. 1971;  34 114-118
• 17 Lougheed WM, Tom M. A method of introducing blood into the subarachnoid space in the region of the circle of Willis in dogs.  Can J Surg. 1961;  4 329-337
• 18 Louis DN, Ohgaki H, Wiestler OD. et al . The 2007 WHO classification of tumours of the central nervous system.  Acta Neuropathol. 2007;  114 97-109
• 19 Makale M. Intravital imaging and cell invasion.  Methods Enzymol. 2007;  426 375-401
• 20 Pool JL, Colton RP. The dissecting microscope for intracranial vascular surgery.  J Neurosurg. 1966;  25 315-318
• 21 Rand RW, Jannetta PJ. Micro-neurosurgery for aneurysms of the vertebral-basilar artery system.  J Neurosurg. 1967;  27 330-335
• 22 Read TA, Farhadi M, Bjerkvig R. et al . Intravital microscopy reveals novel antivascular and antitumor effects of endostatin delivered locally by alginate-encapsulated cells.  Cancer Res. 2001;  61 6830-6837
• 23 Stummer W, Pichlmeier U, Meinel T. et al . Fluorescence-guided surgery with 5-aminolevulinic acid for resection of malignant glioma: a randomised controlled multicentre phase III trial.  Lancet Oncol. 2006;  7 392-401
• 24 Stummer W, Stocker S, Wagner S. et al . Intraoperative detection of malignant gliomas by 5-aminolevulinic acid-induced porphyrin fluorescence.  Neurosurgery. 1998;  42 518-525 , ; discussion 525–526
• 25 Suihko C, Swindle LD, Thomas SG. et al . Fluorescence fibre-optic confocal microscopy of skin in vivo: microscope and fluorophores.  Skin Res Technol. 2005;  11 254-267
• 26 Vajkoczy P, Ullrich A, Menger MD. Intravital fluorescence videomicroscopy to study tumor angiogenesis and microcirculation.  Neoplasia. 2000;  2 53-61
• 27 Wessels JT, Busse AC, Mahrt J. et al . In vivo imaging in experimental preclinical tumor research – a review.  Cytometry A. 2007;  71 542-549
• 28 Wood JR, Green SB, Shapiro WR. The prognostic importance of tumor size in malignant gliomas: a computed tomographic scan study by the Brain Tumor Cooperative Group.  J Clin Oncol. 1988;  6 338-343
• 29 Woringer E, Kunlin J. Anastomosis between the common carotid and the intracranial carotid or the sylvian artery by a graft, using the suspended suture technique.  Neurochirurgie. 1963;  200 181-188
• 30 Zheng W, Harris M, Kho KW. et al . Confocal endomicroscopic imaging of normal and neoplastic human tongue tissue using ALA-induced-PPIX fluorescence: a preliminary study.  Oncol Rep. 2004;  12 397-401

Comment to the article:

H.-G. Schlosser et al. Confocal Neurolasermicroscopy in Human Brain – Perspectives for Neurosurgery on a Cellular Level. Cent Eur Neurosurg 2010; 71: 13 – 19

Neurosurgery for gliomas of adults (WHO grades II – IV) requires both maximal tu mor mass reduction and sparing of local functional anatomy – in the face of a diffusely infiltrating neoplasm that, almost by definition, tends to lack clear-cut boundaries. Individual resection procedures are, therefore, influenced by and dependent on intraoperative decision making, especially as it comes to (i) identifying the tumor as glial in nature, (ii) grading (low- vs. high-grade), as well as (iii) pinpointing peripheral areas of significant drop in cellularity that will be elected as technical resection “margins”.

In the timely context of current biomicroscopic endeavours, the authors of this article propose that “neurolasermicroscopic” study of glioblastoma tissue ex vivo in the operating theatre may attain a diagnostic accuracy comparable to the examination of conventional histology slides. Specifically, it is shown that the quartet of diagnostic hallmarks of glioblastoma (i.e., a diffusely infiltrating astrocytoma with cellular atypia, mitotic figures, microvascular proliferation and necrosis) can reliably be identified by “neurolasermicroscopy“.

This is a small case series on a single neoplastic entity, yet with reasonably consistent and promising results. The authors are also to be congratulated on pertinently addressing some limitations and potential pitfalls inherent to relying on individual diagnostic clues of classical glioblastoma (i.e., relative lack of pleomorphism in the small cell variant of glioblastoma and in glioblastoma with oligodendroglial component). Conversely, those conversant with the daily practice of surgical neuropathology will also remark that atypical astrocytes – at times frighteningly so – may occur as a fixture of some low-grade gliomas (e.g., pleomorphic xanthoastrocytoma), non-glial tumors (e.g., cerebral lymphomas), and even non-neoplastic conditions (e.g., demyelinating diseases). On the long term, “neurolasermicroscopy” might possibly contribute additional novel diagnostic criteria in the evalualtion of neoplasms and tumor-like lesions of the central nervous system. From an immediate practical point of view, one major appeal of “neurolasermicroscopy” resides in its generating microscopic images for telepathological systems, thus neutralizing infrastructural barriers between neurosurgery and neuropathology.

Andreas Raabe
Istvan Vajtai
Department of Neurosurgery
University Bern
Bern, Switzerland
DOI: 10.1055/s-0029-1243090

Comment to the article:

H.-G. Schlosser et al. Confocal Neurolasermicroscopy in Human Brain – Perspectives for Neurosurgery on a Cellular Level. Cent Eur Neurosurg 2010; 71: 13 – 19

The authors present a manuscript on neurolasermicroscopy (NML) as a promising new method for intraoperative tumor diagnosis. I share their view that NLM carries potential for rapid tissue diagnosis. If this method were to work well, it would surely expedite intraoperative tissue diagnosis during open surgery or for stereotactic biopsies. However, more data would surely add to the investigation. These should be collected in a larger collective of patients, assessing the different tissue regions of glioblastoma patients, i.e. infiltration zone, solid tumor, necrosis and adjacent edematous tissue. A necessary assessment would also involve blinded review of pathology derived from confocal microscopy and compared to the conventional histological investigation. Finally, other pathologies need to be tested, such as lymphomas, abscesses and metastasis, to determine whether the method also discriminates malignant gliomas from these other common afflictions in this complex field.

Walter Stummer
Department of Neurosurgery
University Münster
Münster, Germany
DOI: 10.1055/s-0029-1243101

Comment to the article:

H.-G. Schlosser et al. Confocal Neurolasermicroscopy in Human Brain – Perspectives for Neurosurgery on a Cellular Level. Cent Eur Neurosurg 2010; 71: 13 – 19

Optical Targeting in Neurosurgery Two dimensional video imaging of the cortical surface or a resection cavity is the most frequent optical imaging technique intraoperatively. A variety of technologies are being developed that use biophotonic methods that do not require labeling to visualize morphology, reactivity and functional responses and that can reconstruct volumetric data on a cellular level. The term “optical imaging of intrinsic signals” (OIS) is also known because intrinsic optical properties such as absorption and scattering are detected which can be measured directly at the brain surface.

The manuscript of Schlosser et al. in the current issue of Central European Neurosurgery describes an endoscopy based miniature confocal laser microscope (neurolasermicroscope ?=? NLM) that can be introduced to analyze any anatomical surface or cavity which has obvious benefits over any other microscopy technique that requires larger optics but close proximity to the objects. With the current technique, morphological mapping of e.?g. a resection cavity or any other surface for neurosurgical purposes might be possible, including analyses of deep seated lesions. In addition to purely morphological studies, the nuclear density, cellular pleomorphy, mitotic rate and microvascular proliferation was assessed without any additional labels, as required for e.?g. multiphoton microscopy (MDM). The penetration depth of the NLM is up to 100?μm, which should allow for sufficient analysis of the infiltration zone.

A variety of biophotonic, optical imaging technologies are already in clinical use or nearing clinical application:

Laser speckle flow imaging: Laser speckle flow imaging of the exposed cortex can image the spatial distribution of the flow throughout the vascular network during functional activation. A laser diode excites the tissue which is recorded rapidly at a high resolution of the cortex which appear to be low-contrast images of the speckle pattern from the laser illumination. The speckle pattern is caused by the coherent laser light scattering within the brain. If red blood cells are moving within the image, they cause the speckle pattern to vary over time. The rate at which the speckles change relates to how fast the red blood cells are moving (Hecht et al. 2009, Paul et al. 2006).

Optical coherence tomography (OCT): OCT is similar to ultrasound imaging although reflections of near infrared radiation, rather than sound, is detected. In ultrasonography the time delay of reflected sound waves is translated into images. Since a time delay of reflected light cannot be measured directly, light is split into probe and reference light. The reference light and the probe light are combined by a beam splitter and registered by a detector. Interference of low coherence light only occurs when the optical path length of reference and probe are matched within the coherence length of the light source. This implies that the difference between the optical path of the reference and the probe light determines the depth in the sample at which the magnitude of reflection is translated into tissue contrast images. OCT imaging has a relatively high frequency, therefore, functional measurements of the cortical surface might be possible.

Recently, Vakoc et al. 2009 has demonstrated that three-dimensional microscopy of the tumor vasculature of both distinct blood and lymphatic vessels in vivo using optical frequency domain imaging, a second generation development of OCT imaging. Compared to multiphoton microscopy (MDM) of the same tissues, OFDI was similar in resolution, imaged a wider field (30-40x the area of MDM), was 12x faster. OFDI was contrast-free to delineate the vascular and lymphatic networks. A dedicated algorithm was introduced to delineate the vascular structures for automated analyses of the vascular network.

Confocal imaging: Confocal microscopy relies on the rejection of scattered light by isolating signals originating from the focus of the scanning beam. Confocal excitation wavelengths are typically in the visible spectrum, where tissue scatter and absorption are high. Therefore, when using confocal microscopy to image depths beyond 200 to 300 μm, the laser beam can no longer focus, and images become blurry and lack sensitivity. Confocal microscopy can also be combined with fluorescence emission. While it is assumed that the detected fluorescence light originates from the focus of the scanning beam, some of the detected signals are generated by excitation and emission light that is scattered or absorbed within the tissue above and below the focus.

Multiphoton microscopy (MDM): Two-photon microscopy has also been applied to in vivo imaging of the brain. While fluorescent contrast is required, two-photon microscopy provides an unprecedented view of in vivo brain activity on a cellular and microvascular level. Similar to confocal microscopy, two-photon microscopy requires a focused beam of laser light to be steered within the tissue, sensing the properties of each location and using them to form a 2-D or 3-D image. This technique overcomes many of the disadvantages of confocal microscopy. Scattering and absorption of NIR radiation in tissue is much lower than for visible light, therefore, the focus of the laser beam can be maintained at depths of >600 μm. Although autofluorescence can be measured and used for intrinsic contrast, confocal imaging and MDM mostly rely on fluorescent labeling of target structures.

Raman spectroscopy: Raman spectroscopy is based on inelastic scattering of monochromatic radiation which is a laser diode emitting light source. This process can excite molecular vibrations. As numerous vibrations of biomolecules are probed simultaneously more bands are observed in vibrational spectra of cells and tissue than in other optical spectra giving a fingerprint-like signature. These bands provide information about the biochemistry, structure and composition of the underlying sample. Among the main advantages is that this information is obtained without labels. The potential advantages of Raman spectroscopy in neuro-oncology include that brain tissue can be characterized label-free at the microstructural and/or molecular level with spatial resolution in the single cell range. This permits accurate delineation of tumor margins, and sensitive and specific identification of tumor remnants upon preservation of normal tissue. However, any area of interest can be investigated at one time and quantitatively analyzed: if one would choose to resect 80?% tumor cell bearing tissue or would choose to respect 80?% normal tissue with 20?% tumor cell tissue remaining would remain in the physicians decision. A Raman system can be coupled to a fiber optic probe similar to the NLM described here although mapping rather than imaging would be currently possible. This approach has been applied to detect metastasis in mouse brains (Krafft, Kirsch et al. 2007) and quantitatively dissect glioma cell contaminants (Steiner et al. 2008) by either Raman or Infrared spectroscopy.

Progress in neurosurgery depends on innovative technological developments. We need additional information for clinical decision making, in particular, we are interested in :

• Morphology generated by varying contrast,
  e.?g. nuclei to determine tumor characteristics or lipids to reveal fiber orientation and disruption

• High lateral resolution,
  subcellular resolution to obtain characteristics of pathology

• Sufficient depth
  beyond the limitations of multiphoton imaging to allow interventions within the possibilities of surgical dexterity

• Frequent measurements
  over extended periods of time and

• Real time temporal resolution
  to monitor blood flow and to detect functional differences (similar to MRI-generated bold effect, see Optical Imaging)

• Biochemical composition of any tissue or cell to classify pathologies, grade tumors, differentiate substructures of the CNS.

Currently, only two technologies are available for clinical use : 1) direct fluorescence imaging using systemic administration of 5-Aminolaevulinic-acid-(ALA) to detect glioblastoma tissue. In principal, this method would work on the cellular level as well. Using the fluorescent dye as a photosensitizing agent, the method might facilitate diagnostic imaging and therapeutic cellular targeting via photodynamic therapy in the same setting. 2) Recently, a conventional stereoscopic operating microscope with an integrated OCT imaging system was clinically certified (Böhringer et al. 2006, 2009).

In general, current neurosurgical techniques are seemingly limited to the neurosurgeons ability to visualize and maneuver. The promise of cellular resolution offered by the described optical systems require additional technologies for selective targeting.

The current manuscript and the many other emerging examples of label-free optical imaging demonstrate that molecular and cellular neurosurgery is advancing rapidly. We urgently need to develop tools to translate the new types of information into therapeutic approaches. Radicality and selectivity may become synonyms rather than contradictions if we can target pathologies beyond neurosurgical dexterity.

Matthias Kirsch
Gabriele Schackert
Klinik und Poliklinik für Neurochirurgie Carl Gustav Carus Universitätsklinikum Dresden
Technische Universität Dresden
Dresden, Germany
DOI: 10.1055/s-0029-1243102

References
1 Böhringer HJ, Lankenau E, Stellmacher F, Reusche E, Hüttmann G, Giese A. Imaging of human brain tumor tissue by near-infrared laser coherence tomography. Acta Neurochir (Wien) 2009: May; 151 (5): 507–17

2 Böhringer HJ, Lankenau E, Rohde V, Hüttmann G, Giese A. Optical coherence tomography for experimental neuroendoscopy. Minim Invasive Neurosurg 2006 Oct; 49 (5): 269–75

3 Steiner G, Küchler S, Hermann A, Koch E, Salzer R, Schackert G, Kirsch M. Rapid and label-free classification of human glioma cells by infrared spectroscopic imaging. Cytometry A 2008 Dec; 73A (12): 1158–64

4 Krafft C, Kirsch M, Beleites C, Schackert G, Salzer R. Methodology for fiber-optic Raman mapping and FTIR imaging of metastases in mouse brains. Anal Bioanal Chem 2007 Oct; 389 (4): 1133–42

5 Vakoc BJ, Lanning RM, Tyrrell JA, Padera TP, Bartlett LA, Stylianopoulos T, Munn LL, Tearney GJ, Fukumura D, Jain RK, Bouma BE. Three-dimensional microscopy of the tumor microenvironment in vivo using optical frequency domain imaging. 2009 Sep; 13 [Epub ahead of print] http://www.nature.com/nm/journal/vaop/ncurrent/abs/nm. 1971.html)

6 Schlosser HG, Suess O, Vajkoczy P, van Landeghem FKH, Zeitz M, Bojarski C. Confocal Neurolasermicroscopy in Human Brain – Perspectives for Neurosurgery on a Cellular Level. Cent Eur Neurosurg 2010; 71: 13 – 19

7 Hecht N, Woitzik J, Dreier JP. Vajkoczy P. Intraoperative monitoring of cerebral blood flow by laser speckle contrast analysis. Neurosurg Focus 2009: Oct; 27 (4): E11

8 Paul JS, Al Nashash H, Luft AR, Le TM. Statistical mapping of speckle autocorrelation for visualization of hyperaemic responses to cortical stimulation. Ann Biomed Eng 2006: Jul; 34 (7): 1107–18

Correspondence