Intraoperative Ultrasound For Identification Of Vascular Structures During Intracranial Tumour Resection And Minimisation Of Intraoperative Haemorrhagic Complications: A Narrative Review

Abstract

Background. Injury to arteries, perforators and draining veins is among the most consequential events in intracranial tumour surgery, and the resulting haemorrhage or infarction accounts for a substantial share of new postoperative deficits. Intraoperative ultrasound (ioUS) is the only widely available imaging modality that depicts both tumour and flowing blood in real time, at the operative site, without ionising radiation and without a major interruption of workflow.

Objective. To review how the individual ioUS modalities — B-mode, Doppler and power Doppler, contrast-enhanced ultrasound (CEUS) and navigated three-dimensional acquisition — contribute to the identification of vascular structures during tumour resection, and to examine critically what the published evidence does and does not establish about the effect of ioUS on intraoperative haemorrhagic complications.

Findings. ioUS reliably demonstrates flowing vessels down to the calibre of the lenticulostriate perforators, and navigated three-dimensional power Doppler allows their displacement to be re-assessed as the brain shifts during resection. CEUS adds a temporally resolved arterial and venous phase and improves delineation of perfused tumour. Pooled diagnostic accuracy for residual tumour is moderate: sensitivity 72.2% and specificity 93.5% in one meta-analysis of 409 diffuse gliomas, and sensitivity 75% (95% CI 62–84) with specificity 88% (95% CI 79–94) in another of 542 patients, with consistently better performance in low-grade than in high-grade lesions. For extent of resection the evidence is stronger: a randomised trial reported complete resection in 8 of 23 (35%) ultrasound-guided glioblastoma operations versus 2 of 24 (8%) with standard neuronavigation (p = 0.036), and a meta-analysis of 732 patients found a risk ratio for gross total resection of 2.02 (95% CI 1.31–3.10) in favour of ioUS. By contrast, no controlled study has been designed to test whether ioUS reduces intraoperative haemorrhage. The available complication data are uncontrolled: a single-arm meta-analysis reports a pooled complication rate of 15% (95% CI 7–23) without a comparator, and the randomised trial found no difference in complication rates between arms.

Conclusions. The rationale for using ioUS to identify and preserve vessels is anatomically sound and supported by technical series, and the benefit for extent of resection is now supported by randomised and pooled data. The specific claim that ioUS lowers the rate of intraoperative haemorrhage remains, however, unproven; it is a plausible mechanism that has not been the primary endpoint of any adequately powered study. Trials reporting vascular injury and intraoperative bleeding as prespecified outcomes are needed.