Contrast-Enhanced Ultrasound Assisted Surgery of Intramedullary Spinal Cord Tumors: Analysis of Technical Benefits and Intra-operative Microbubble Distribution Characteristics

Abstract

Intra-operative contrast-enhanced ultrasound (CEUS) is a relatively standardized procedure in brain neurosurgery, but it is still underused in spinal cord and intramedullary tumor evaluation. We reviewed and analyzed the intra-operative data from a surgical series of patients harboring intramedullary spinal cord tumors who underwent surgery under CEUS guidance. CEUS was performed in 12 patients (age range: 13–55 y); all lesions had ill-defined boundaries or peritumoral cysts at preliminary intra-operative B-mode ultrasound. CEUS highlighted the tumors in all cases. The contrast agent's spinal distribution revealed different phases (arterial, peak, washout), as observed in the brain, but these appeared to be slower and less intense. In our experience, intra-operative CEUS allows surgeons to assess spinal cord perfusion and highlight intramedullary tumors in real time. As for other imaging modalities, ultrasound contrast agents add valuable information over baseline imaging, and their use should be fostered to better understand microbubble distribution dynamics.

Introduction

Intramedullary spinal cord tumors remain a major surgical challenge despite the continuous technological improvement of surgical tools. Among these tumors, ependymomas are the most common overall, particularly in the adult population, whereas astrocytomas have higher occurrence rates in children (Jallo et al. 2003; Schellinger et al. 2008; Hsu et al. 2011; Klekamp 2015); high-grade gliomas (HGGs) are, instead, quite rare (Behmanesh et al. 2017). Even though surgical resection is the treatment of choice, it is not always safely achievable as it could yield neurologic sequelae (Jallo et al. 2001; Sandalcioglu et al. 2005; Shrivastava et al. 2005; Costa et al. 2013; Verla et al. 2016) because of the intrinsic infiltrative nature of the lesions, especially when spinal cord manipulation is necessary to reach the tumor (Behmanesh et al. 2017). In such cases, the combination of direct parenchymal and vascular damage can have a relevant impact on the outcome. On the other hand, tumor histotype and the extent of resection are the two most important prognostic factors for tumor progression and overall survival, respectively (Garcés-Ambrossi et al. 2009).

Incomplete visualization or underestimation of the dimensions of intramedullary tumors could, however, induce neurosurgeons to prematurely stop the resection to avoid possible spinal cord injuries and related neurologic deficits. Magnetic resonance imaging (MRI) represents the gold standard for diagnosis and assessment of intramedullary tumors; however, its intra-operative use is not yet widely diffused because it is an expensive and time-consuming technique (Duprez et al. 2008; Netuka et al. 2011). On the contrary, intra-operative ultrasound (IOUS) has proven to be a valuable tool in visualizing both the spinal cord and intramedullary lesions before durotomy (Bozinov et al. 2012; Prada et al. 2014c; Vetrano and Prada 2016; Ivanov et al. 2017). Nevertheless, the distinction between the lesion and infiltrated spinal cord, or other structures such as peritumoral cysts and ependymal canal dilation, on the basis of B-mode only, remains challenging. Even pilocytic astrocytoma boundaries, in which the solid and the cystic portions of the tumor seem to displace rather than infiltrate the spinal cord, are ill defined at B-mode.

Furthermore, HGGs tend to have indistinct, blurred margins at IOUS, as illustrated in brain malignant gliomas (Prada et al. 2016). Also, B-mode and color Doppler cannot adequately estimate the local blood flow impairment caused by the tumor or secondary to the surgical maneuvers during resection. However, this information is paramount for reducing the risks in spinal surgery, as illustrated in traumatic spinal cord injuries, where a significant blood flow reduction in the peri-traumatized tissue determines secondary injury, including inflammation, swelling and cell death (Khaing et al. 2020). Most of the previous studies on contrast-enhanced ultrasound (CEUS) evaluation of spinal cord damage are based on experimental and animal models (Huang et al. 2013, 2017; Khaing et al. 2018, 2020). The use of CEUS for spinal tumors, both intra- and extramedullary, is still minimal, with only a few anecdotal cases so far presented (Vetrano et al. 2015a, 2015b; Della Pepa et al. 2018a; Han et al. 2020). This aspect is owing mainly to the steep learning curve of intra-operative CEUS, which requires a specific knowledge of US physics and US cerebral semiotics and specific training in ultrasound contrast agent (UCA) analysis. Moreover, differently from the brain's patterns of distribution, the dynamics of microbubbles (MBs) within the spinal cord is still to be explored. Based on our previous experience with the use of CEUS in brain tumors (Prada et al. 2014a, 2015, 2017b), we performed CEUS imaging during intramedullary tumor surgery.

We describe a series of patients in whom we used CEUS to highlight intramedullary tumors to enhance the surgical guidance and thoroughly describe the contrast-enhancing characteristics of different lesions to better understand the pathophysiology of microbubble dynamics during surgery.