Common brain tumors include glioma, metastasis, meningioma, etc., and glioma is the most common intracranial malignant tumor. Cancer cells invade surrounding tissues by diverse modes of dissemination, such as expansive growth, collective invasion, mesenchymal migration, and amoeboid migration(29). The most important hallmark of glioma is its aggressive behavior, and its extensive invasion of healthy brain tissue is the main reason for the poor prognosis and the difficulty in finding curative therapies(30). In terms of the main invasion patterns of metastases, although the main pattern is well-demarcated growth(50%), vascular co-option(18%) and diffuse infiltration(32%) are also common modes of tumor invasion. At the same time, there was no particular correlation between the different modes of invasion of brain metastases into the brain parenchyma and the type of primary tumor(31). Tumors arising in the dura, arachnoid, and pia meninges are called meningiomas, 80% of which are benign. However, 20% of benign, atypical and anaplastic meningiomas have pathophysiological manifestations of invasion to peripheral healthy tissues, and the postoperative recurrence rate is high(32–34).
Surgical resection is the primary treatment strategy, and EOR has been identified as a major independent predictor of overall survival (OS) (35). Specifically, for glioblastoma, GTR is directly related to OS. Lacroix et al. (36)found that the median survival time after primary surgery was improved by 4.2 months compared with the resection of 98% or more of the tumor volume being removed with or without resection. GTR has been identified as a target of brain tumor therapy, and every possible effort must be made to achieve maximum resection safely. The biological characteristics of brain tumors make it challenging for imaging to identify the exact lesion margin.
To achieve GTR, evaluation of tumor boundaries is essential. IOUS has been used for intraoperative navigation since 1980(37). IOUS is real-time, inherently convenient and distinctly cheaper and has incomparable advantages over other imaging techniques(38). This imaging method can contribute to accurately locating and delineating the boundary of the tumor, evaluating the extent of resection, and guiding the surgical approach, which has great guiding significance for surgery(39). The imaging information provided by IOUS gives neurosurgeons a new perspective to observe and study brain tumors and will help to adjust the surgical strategy and accurately guide tumor resection, maximize the EOR and improve the safety of patients(40, 41). Conventional two-dimensional ultrasound is affected by relatively low contrast and peritumoral edema of brain tumors, so it is difficult to define tumor boundaries.
CEUS enhances the display of microvessels through injection of contrast agent, which can dynamically display the blood flow of arteries, veins and capillaries in real time and reduce the signals from adjacent brain parenchyma(42). CEUS can make up for the disadvantages of conventional B-mode ultrasound, highlight the lesions and their margins, and contribute to distinguishing tumors from peritumoural edematous brain tissue(43). SMI does not require the injection of contrast media but has shown comparable diagnostic efficacy in many other organ diseases. For brain tumors, our study shows that SMI can effectively compensate for B-mode ultrasound in identifying tumor margins, especially for HGG(Fig. 2). HGG is a highly vascularized tumor, and angiogenesis is one of the most obvious characteristics of HGG, which is in remarkable contrast to normal brain tissue(44). In LGG and metastatic tumors, there were also some tumor boundaries that were clearer under SMI, but there was no significant difference. In meningiomas, both B-mode ultrasound and SMI can clearly identify the lesion boundary(Fig. 3). The peritumoral edema of LGG is significantly less than that of HGG, and the margin between the tumor and the adjacent normal brain is not covered by edema(45). Metastases are highly heterogeneous, and the metastatic nature retains the inherent morphology of the primary tumor histologically. Usually, the lesion boundary is clearer than glioma in B-mode ultrasound morphology. Most of them are well demarcated from the surrounding gliotic and rarefied brain parenchyma(46). Meningiomas have been clearly delineated in B-mode mode, but the application of SMI enables visualization of vascular modifications (such as destruction and remodeling) at the meningioma-brain interface. In conclusion, it is a convenient, noninvasive, objective and novel ultrasound Doppler technology. SMI does not reduce the quality of boundary imaging, especially for HGG boundary identification of highly aggressive tumors, which has guiding significance for determining surgical plans and adjusting treatment strategies(47). At the same time, we also found greater consistency in assessing lesion boundaries under SMI conditions. Sonographers are better at recognizing and interpreting ultrasound images than neurosurgeons. When SMI strengthens the delineation of the lesion boundary, the lesion is more prominently displayed, and the consistency of interpretation by different doctors is improved to a certain extent.
In Table 4, we further analyzed the correlation between ultrasound features and GTR and found that the size and demarcation of the tumor were correlated with GTR. Munkvold et al. analyzed the correlation between intraoperative 3D B-mode ultrasound manifestations and GTR and found that in 144 gliomas, small tumors was the factor significantly associated with GTR(48). However, in our study, the demarcation between tumor and healthy brain tissue in SMI was significantly associated with GTR, mainly because SMI improves the delineation of brain tumors and provides histological information about vascularization, which can help neurosurgeons distinguish tumors from edema, further confirming the advantages of SMI over B-mode ultrasound in boundary delineation.
Table 4
Logistic regression analysis explored possible predictors of total resection.
| OR(95%CI) | P value |
Tumors size(cm) | | 0.003* |
≥5cm | 1.00 | |
<5cm | 12.317(2.392–63.430) | |
Deep(cm) | | 0.191 |
≥2cm | 1.00 | |
< 2cm | 3.275(0.553–19.389) | |
Peritumor edema(cm) | | 0.147 |
≥ 2cm | 1.00 | |
< 2cm | 3.373(0.653–17.428) | |
Tumor interface in SMI | | 0.045* |
Unclear | 1.00 | |
Clear | 4.597(1.034–20.449) | |
*P<0.05 |
As mentioned above, as an innovative Doppler technique in neurosurgery, SMI can make a good judgment on the boundary of the tumor and observe the tumor in brain tumor resection. The internal and surrounding blood flow can guide the resection to a certain extent and finally help the patient achieve GTR as far as possible. With the continuous development of research and the continuous progress of ultrasound technology, intraoperative ultrasound and new ultrasound technology will play an increasingly important role in neurosurgery.