Clinical application of intraoperative ultrasound superb microvascular imaging in brain tumors resections: contributing to the achievement of total tumoral resection

Siman Cai, Hao Xing, Yuekun Wang, Wenbin Ma, Yuxin Jiang, Jianchu Li, Hongyan WangBMC Medical Imaging 2024 · DOI

TL;DR

This study investigates the effectiveness of Superb Microvascular Imaging (SMI), an advanced ultrasound technique, in improving the accuracy of brain tumor boundary identification during surgery. The findings suggest that SMI can significantly enhance visualization of tumor margins, particularly in high-grade gliomas, thereby contributing to more complete tumor resections and potentially improving patient outcomes.

What Problem Does This Paper Address?

Achieving a "gross-total resection" (GTR) of brain tumors, meaning complete removal of the tumor, is a critical goal in neurosurgery. GTR is strongly linked to improved patient survival and reduced risk of malignant transformation. However, the infiltrative nature of many brain tumors, where cancer cells spread diffusely into surrounding healthy tissue, makes precise identification of tumor boundaries a significant challenge. Traditional methods like neuronavigation systems can be affected by brain shift during surgery. While fluorescence imaging and contrast-enhanced ultrasound (CEUS) offer improvements, they have limitations such as adverse reactions or the need for contrast agents. This paper addresses the need for a real-time, non-invasive, and effective imaging technique to accurately delineate tumor margins during surgery to facilitate GTR.

Methods

The study enrolled 57 consecutive patients undergoing brain tumor surgery. Before and within 48 hours after surgery, MRIs were performed to assess the extent of resection. During the operation, both conventional B-mode ultrasound and the Superb Microvascular Imaging (SMI) technique were used to evaluate tumor boundaries. SMI, a Doppler technique that visualizes microvascular blood flow without contrast agents, was employed in both monochrome (mSMI) and color (cSMI) modes. An experienced ultrasound doctor and a neurosurgeon independently assessed the clarity of tumor boundaries in both B-mode and SMI. Statistical analysis, including binary logistic regression, was used to identify factors associated with GTR.

Datasets

The study utilized data from 57 patients undergoing brain tumor surgery at Peking Union Medical College Hospital. The primary "dataset" consists of the clinical and imaging data collected from these patients, including pre- and post-operative MRI scans, and intraoperative ultrasound (B-mode and SMI) findings. No new public benchmark datasets were introduced.

Key Insights and Ideas

  • SMI's Superiority in High-Grade Gliomas: The most striking finding is SMI's significant advantage in defining the boundaries of high-grade gliomas (HGGs), which are highly vascularized and notoriously difficult to delineate.
  • Non-Invasive Visualization: SMI offers a compelling alternative to contrast-enhanced methods, providing detailed microvascular information without the need for contrast agents, thus avoiding potential adverse reactions and associated costs.
  • Enhanced Consistency: The study suggests that SMI leads to greater consistency in the assessment of tumor boundaries between sonographers and neurosurgeons, potentially reducing inter-observer variability.
  • Predictors of GTR: Tumor size (smaller tumors being more amenable to GTR) and the clarity of the tumor boundary (clearer boundaries facilitating GTR) were identified as significant predictors of achieving gross-total resection.

Key Results

  • Out of 57 patients, 37 (63.9%) achieved gross-total resection (GTR), while 20 (35.1%) had subtotal resection (nGTR).
  • SMI improved the definition of tumor margins in 8 out of 23 HGG cases, a statistically significant improvement (P=0.033).
  • Tumor size ≥ 5 cm and unclear ultrasonic boundaries were identified as independent risk factors for nGTR (OR > 1, P < 0.05).
  • The likelihood of GTR increased significantly for tumors smaller than 5 cm (OR=12.317) and for those with clear boundaries (OR=4.597).
  • Consistency in lesion boundary assessment was high for both B-mode (Cohen's kappa=0.860) and SMI (Cohen's kappa=0.684), with SMI showing slightly lower but still substantial agreement.

Limitations and Future Work

The authors acknowledge several limitations:

  • Sample Size: The study included 57 patients, which they suggest is a relatively small sample size. They recommend expanding the number of cases for future research.
  • Follow-up Time: The study did not extensively detail long-term follow-up, and they suggest extending follow-up periods to compare the effects of different intraoperative methods on patient quality of life and clinical outcomes.
  • Identification of Residual Tumors: While SMI aids in boundary demarcation, the paper notes that intraoperative detection of residual tumors and pathological biopsy remain areas for further research.
  • Comparison with Other Methods: The study primarily compares SMI to B-mode ultrasound. Further comparative studies with other advanced techniques like CEUS or intraoperative MRI could be beneficial.

Future research directions suggested include:

  • Investigating the role of SMI in identifying residual tumors post-resection.
  • Exploring new technologies, such as novel liquid acoustic couplers, to further enhance ultrasound's identification capabilities.
  • Expanding the application of SMI to a wider range of brain tumor types and surgical scenarios.

Summary updated

BibTeX
@article{cai2024clinical,
  author = {Siman Cai and Hao Xing and Yuekun Wang and Yu Wang and Wenbin Ma and Yuxin Jiang and Jianchu Li and Hongyan Wang},
  title = {{Clinical application of intraoperative ultrasound superb microvascular imaging in brain tumors resections: contributing to the achievement of total tumoral resection}},
  journal = {BMC Medical Imaging},
  year = {2024},
  doi = {10.1186/s12880-024-01321-5},
}
Figure