Evaluation of intra-operative ultrasound imaging in brain tumor resection: a prospective study

The purpose of our study was to evaluate intra-operative ultrasound (IOUS) as a tool of resection control after brain tumor surgery. In addition, we looked for tumor species suitable for ultrasound representation.

Using a Siemens Omnia Sonoline Ultrasound, 36 tumors were examined, high-grade gliomas (62%), metastases (22%) and others (16%). We focused on tumor imaging by ultrasound with regard to its reliability of tumor expansion and margins. Evaluation of the images was carried out by correlating the ultrasound-based intra-operative measured tumor volume before and after resection with a pre- and post-operative (within 48 hours) measured volume by MRI. The IOUS measurements were performed by the neurosurgeon and the MRI measurements by the neuroradiologist. Thus, the measurement procedures were blinded. Corresponding to a deviation of the ultrasound volume by 10, 20 and > 20% from the MRI volume, the correlation was ranked good, moderate and poor. For assessing the agreement between these two methods of imaging, the statistical analysis was conducted using a method described by Bland and Altman.

High-grade gliomas mostly showed a moderate or poor correlation in comparing IOUS- and MRI-tumor volumetry resulting in incomplete resection. Metastases resulted in a good to moderate correlation with a satisfactory extent of resection. The other tumors had poor images with larger tumor residues. The MRI measured volumes tended to be larger on average; the deviation grew with tumor size .

The reliability of IOUS depends on tumor type. It is beneficial to use IOUS for the resection of metastases and a few high-grade gliomas. Concerning the volumetric accuracy, the value of IOUS is worse than its value of navigation and resection control.