Massimiliano Del Bene*§, Francesco Prada*°#, and Francesco DiMeco*^s
* Department of Neurosurgery, Fondazione IRCCS Istituto Neurologico Carlo Besta, Milan, Italy
§ Department of Experimental Oncology, IEO, European Institute of Oncology IRCCS, Milan, Italy
° Department of Neurological Surgery, University of Virginia, Charlottesville, VA, United States
# Focused Ultrasound Foundation, Charlottesville, VA, United States
^ Department of Pathophysiology and Transplantation, University of Milan, Milan, Italy
s Department of Neurological Surgery, Johns Hopkins Medical School, Baltimore, MD, United States
Ultrasound (US) is an imaging modality that is diffusely used in medicine, encompassing almost every clinical specialty. The reasons underlying such a large adoption rely on the ease of use, availability, relative low-cost and the ability to generate real-time, high-resolution images without ionizing radiation or cumbersome apparatuses.
US’s wide diffusion outside radiology started in medical specialties such as cardiology, pediatrics, gynecology and neurology, when years ago physicians realized the usefulness of operating a real-time, effective imaging tool whenever necessary to facilitate daily routine; these medical areas, building experience with day by day practice, soon became independent subdisciplines in clinical US.
The use of intraoperative ultrasound (ioUS) during neurosurgical procedures was first described in 1978 by Reid (27). The brain and medulla, because of their specific viscoelastic properties, represent an optimal target for ultrasound imaging, once exposed from bony structures (11). In fact, several applications have been described in spine and brain surgery, such as lesions recognition, guidance in surgical resection, catheter placement, vessels study, aspiration of abscess, posterior fossa decompression (2, 10, 11, 14, 27, 28, 35), and so forth. However, the initial interest towards US as an imaging tool in neurosurgery progressively weaned from the 80’s until the early 2000’s in favor of other imaging techniques such as CT and MRI which, at the time, were providing better image resolution, a comprehensive view of the whole organ, and multimodal imaging including contrast options. In addition, they were the mainstay for diagnostic imaging, thus becoming very well familiar to all neurosurgeons. As a consequence, and also because of poor image quality, difficulty in image interpretation and orientation, lack of specific knowledge and training, and often use of low-end US apparatus, US progressively lost interest from the majority of neurosurgeons (11, 35).
In recent years, thanks to substantial technical improvements, we have been witnessing a growing clinical and scientific attention towards the use of US as an intraoperative neurosurgical guidance tool. Nowadays, US B-mode offers excellent spatial and temporal resolution, providing superb imaging quality - in selected cases superior to MRI (25). Furthermore, many US modalities other than B-mode and Doppler-modes have been developed, making US a multimodality imaging tool. Nevertheless, still most neurosurgical studies conducted in the field of ioUS refer to B-mode as the only imaging modality, therefore limiting US’s full potential.
The most relevant and appealing features of present US scanners include their multimodality options, represented by technical advancements such as image fusion between real-time US and pre-operative imaging for surgical guidance, advanced Doppler imaging, contrast-enhanced ultrasound (CEUS) and elasto-sonography (25). All these features, together with training requirements, have already been deeply exploited in other disciplines and their use is regulated by different user groups such as the European Federation for Ultrasound in Medicine and Biology (EFSUMB) (8).
Indeed, in order to take advantage of the “adequate for the case” US imaging setup it would be advisable to use a high-end ioUS scanner, which in any case remains by far more affordable than other intra-operative imaging instruments. Multiple probes should also be available in order to cover different applications. High-frequency linear probes are needed to study small superficial lesions in great detail, while low-frequency convex probes are more suitable for bigger and deeper lesions. Currently, dedicated probes are under development for specific applications such as pituitary probes to perform trans-nasal surgery under ioUS guidance (13).
Semiotics and navigated ultrasound
In strict terms of image acquisition, a complete ioUS exam usually starts with a B-mode scan to orient in the surgical field and to identify major landmarks (5). B-mode provides mainly anatomical information such as lesion location, relationships with neighbor structures and residual tumor site and volume (4, 20). At the same time, B-mode can describe the nature of the tissue by virtue of its echogenicity (13).
One main limitation of B-mode is its specific semiotics, which require experience to be fully understood (4, 20). This is even more challenging in consideration of the US planes that, not constrained to standard axial, coronal and sagittal views, depend on probe orientation (20). Navigation systems can be coupled to US in order to overcome this limitation, facilitating the understanding of B-mode anatomy. Tracking systems, either optical or magnetic, allow US imaging navigation in the 3D space, and are able to fuse ioUS images to the co-planar pre-operative magnetic resonance imaging (MRI). The direct, real-time comparison between ioUS and pre-operative MRI allows better understanding of anatomical features and facilitates orientation (15, 18). Furthermore, the tracking system makes possible the acquisition of an updated 3D US volume to be used as a new reference volume for intra-operative navigation - therefore adjusting for brain shift phenomena (15, 33, 34).
B-mode semiotics is also dynamic and subjected to continuous modifications, showing surgical progression as demonstrated by artifacts occurring at the end of surgical removal. At that time, discriminating between surgical induced alterations, artifacts, or residual tumor can be particularly challenging. Several mechanisms have been proposed to explain this phenomenon, and further solutions have been consequently proposed: specific fluid to fill the cavity (29), mini-probes to scan from inside the cavity (31, 32), tangential trans-cortical scan (4), and also contrast enhanced ultrasound (17). This continuous detection of changes occurring during surgery is made possible by the readiness and ease of use of US as an intraoperative imaging modality: it is in fact possible to scan the surgical field multiple times during surgery, each scan lasting minutes or less (5).
Ultrasound additional features
Once a first morphological appraisal of the surgical field has been performed, it is possible to evaluate tissue perfusion with Doppler imaging, which relies on the Doppler effect to describe blood flow to indirectly study the vascular tree. Combining B-mode and various Doppler modes (color- spectral-Doppler), it is possible to study in great detail the blood flow in a vessel and at same time the anatomy (1, 5, 19). In vascular surgery, this kind of examination provides information on target location and a real-time feedback on the vascular effects of a surgical manoeuvre (9). In oncological surgery, the quantitative-anatomical doppler analysis can be beneficial to localize neighbor vessels and to assess flow preservation during removal progression (5). Power doppler represents a different approach to doppler imaging relying on the amplitude of doppler signal to detect blood flow. It does not provide information regarding flow speed and direction, whereas it quantifies the entity of blood flow with higher sensitivity and less angle dependency than other modalities. It is useful to localize vessels, to identify small and/or low-flow, and to study tissue perfusion - achieving an overview of blood flow in the region of interest (19, 33). A further evolution of power doppler is the high sensitivity doppler, which has considerably improved the degree of flow sensitivity even in very small vessels and slow flow detection (12).
Ultrasound contrast agents (UCA), in combination with contrast-specific imaging techniques, are increasingly accepted for diagnostic imaging and post-operative work-up in several organs. The use of contrast agents, as for CT and MRI where contrast is almost compulsory, is becoming well documented also for US (CEUS). UCAs consist of micron-sized bubbles stabilized by a shell. They are purely intravascular and are not able to extravasate – unlike, for instance, MRI contrast agents (6, 30). CEUS allows the dynamic and continuous representation of microbubbles distribution, relying on their harmonic oscillation without disruption (6, 24, 30). Therefore, CEUS provides a dynamic and continuous representation of the tissue perfusion and of the whole vascular tree. Its use has become the mainstay for diagnostic purposes in many organs such as the liver where, among other indications, it allows characterization of focal lesions, thrombosis or even monitoring the outcome of ablative procedures for malignant tumors (7).
Recently, EFSUMB guidelines on the use of CEUS in non-hepatic applications have introduced the use of CEUS in two areas of neurosurgery: tissue (organ/tumor) perfusion and vessel visualization (angiosonography) (30). As performed in other organs, CEUS allows for tumor visualization, characterization, and evaluation of residual tumor (5, 17, 23-26). CEUS highlighted the same volume target in glioblastoma as shown by gadolinium-enhanced MRI, thus playing a pivotal role in guiding surgical resection (26). In the case of highly heterogeneous tumors, CEUS is able to discriminate areas of necrosis and areas of viable tissue by virtue of differences in vascularization - thus influencing surgical decision making (5, 23). Furthermore, CEUS can provide a continuous and dynamic feedback on the vascular tree and its modifications that can take place during surgical resection - thus aiding the assessment of feeders, venous drainage, and confirmation of tumor de-vascularization (5, 17, 19, 24-26). CEUS clearly highlights high- and low-flow as well as large and small vessels at the same time, describing all vessels in the surgical field from an anatomical and functional standpoint - thus allowing dynamic real time angiosonography (19, 22, 25). CEUS can demonstrate blood flow presence, modification, direction, velocity and pattern - allowing inference about surgically-induced changes, e.g. in arteriovenous malformation or aneurysm surgery. This can be useful to approach complex vascular lesions, in order to understand their architecture and to protect vital structures (22, 25).
Elasto-sonography represents another US modality to study the mechanical properties of a tissue in a qualitative or quantitative way. It is based on observing how a tissue distorts after a mechanical stimulus. Currently, two techniques are available to study the elastic properties of tissue in brain: shear wave and strain elastography. The basic principle is the same: with shear wave elastography the stimulus is provided by the probe as an US impulse that provides quantitative information on tissue stiffness (3, 21), whereas with strain elastography the mechanical stimulus derives from brain pulsation and probe motion thus leading to qualitative evaluation (3, 21). Similar to CEUS, the use of elasto-sonography has already been well established and codified in numerous organs, while its adoption in neurosurgery is still limited. In our experience, elasto-sonography highlights most lesions with more contrast than B-mode, and discriminates between high-grade and low-grade gliomas with high specificity and sensitivity (21). It holds potential for residual tissue evaluation but this still remains under scrutiny.
In conclusion, ioUS is different from all other intra-operative imaging techniques. ioUS is a true real-time imaging tool: it is repeatable and allows in certain conditions to operate under direct guidance. ioUS has superb spatial resolution and it is a tomography, thus allowing the study of tissue depth. It is complementary to other imaging modalities such as fluorescence which requires tissue exposure to visualize the fluorophore. Thanks to multimodal examination, it is possible to obtain a comprehensive characterization of the surgical field and to obtain morphological and functional information. Another considerable point of value is the cost-effectiveness; if compared to other techniques such as intra-operative MRI which, in addition, are not real-time. ioUS is by far more affordable both in terms of time (exam duration, patient preparation, interruption of surgery) and economy (costs of scanner, consumables, total surgery time).
Thanks to its versatility and relative low cost, ioUS appears to be the optimal tool to be adopted worldwide, including in low- and middle-income countries where financial resources are not readily available. On the other hand, ioUS is a demanding technique that requires study and experience: training and practice are mandatory. To this end, we want to stress the need for attending dedicated courses, to practice every time possible, and also to invest in next generation technologies such as simulation. In our experience, simulation has proved to significantly help neurosurgeons became familiar with ioUS (16).
We firmly believe that ioUS represents a significant adjunct that should be part of a synergistic approach, including neurophysiological monitoring and other imaging techniques, in an effort to improve the effectiveness and safety of neurosurgery.