Ultrasound guided microsurgical resection of cerebral arteriovenous malformations

Article information

Korean J Cerebrovasc Surg. 2025;.jcen.2025.E2025.08.005

Publication date (electronic) : 2025 December 31

doi : https://doi.org/10.7461/jcen.2025.E2025.08.005

Mohamed M Elsherbini, Mohamed A Kassem, Ashraf Ezzeldein, Ibrahim Eltantawy, Abdul Aziz Ismail, Mohamed Farouk

Department of Neurosurgery, Mansoura University, Mansoura City, Egypt

Correspondence to Mohamed M Elsherbini Department of Neurosurgery, Mansoura University, Elgomhouria St., Mansoura City, Egypt 35516 Tel +20 10 05731574 E-mail m_elsherbiny@mans.edu.eg

Received 2025 July 17; Revised 2025 November 19; Accepted 2025 November 21.

Abstract

INTRODUCTION

Intracranial arteriovenous malformation (AVM) is considered one of the most challenging lesions to manage, due to its complex pathophysiology and anatomy. Several management techniques have been adopted for AVM management, which depend on the specific characteristics of individual lesions [4]. Intracranial AVM can be either managed or observed; the choice can be controversial if they have not ruptured. According to the most widely used classification of intracranial AVMs, the Spetzler Martin grading system [16], two grades are mainly treated via surgical excision, and one grade is subject to multimodal treatment, including surgical excision. Therefore, despite numerous treatment methods being potentially useful for AVM, such as endovascular embolization and radiosurgery, surgical excision of intracranial AVM remains a preferred option in many cases, with the highest occlusion rate and most efficient prevention of future intracranial hemorrhage compared to alternatives, yet it needs to be further developed and perfected [7,12].

Ultrasonography was introduced to diagnose brain tumors in the 1950s [8], and by the 1980’s intraoperative ultrasound (IOUS) was being explored as a potential tool for guiding the resection of brain tumors [13]. IOUS is an ideal imaging tool, offering numerous compelling advantages. First of all, the device can be directly placed on the lesion, without interposed bone or soft tissue, to subsequently provide real-time imaging, regardless of brain tissue shifting during surgery (due to manipulation and cerebrospinal fluid (CSF) drainage). IOUS also reduces the surgical time and costs incurred with other imaging modalities, and sonography is free of harmful radiation, as well as being portable. It can be used to localize lesions such as a hematoma in AVMs [15]. Also, as an advantage of AVM surgery, Color Doppler in the B-mode image, identifying the presence and the direction of the flow, can be applied successfully for the surgical disconnection of an AVM [3].

In this research, the value and technical feasibility of IOUS are discussed and evaluated through retrospective study aiming at evaluating this tool in light of recent neurosurgical practice.

MATERIALS AND METHODS

This paper offers a retrospective observational descriptive case series study, based on analyzing databases and patients’ archives at Mansoura University Hospitals’ Department of Neurosurgery from 2021 to 2024. Institutional Review Board (IRB) approval was obtained for this study, and consent was waived due to the retrospective nature of the study. The medical records were reviewed, and all patients who underwent surgical excision of intracranial AVMs were studied. The study inclusion criteria were applied as follows:

- All ages.

- Both sexes.

- Complete medical records.

- Follow-up period of 3 months or more.

- Underwent postoperative digital subtraction angiography.

On the other hand, patients with incomplete medical records and those who dropped follow-up were excluded. Also, patients who underwent management plans other than microsurgical resection e.g., radiation and therapeutic endovascular embolization, were excluded, except patients who underwent mere preoperative embolization, who were included.

Preoperative imaging included initial computed tomography (CT) brain, CT angiography (CTA), magnetic resonance imaging (MRI), and digital subtraction angiography (DSA). Postoperative radiological follow-up included CT brain, MRI, CTA, and/or DSA.

The decision of the proper line of management in the studied context was based on multi-disciplinary team decision-making, including neurosurgery, neurology, and neuro-radiology, and thorough discussion with patients and their family members. The delivered management solutions encompassed four options: expectant therapy, endovascular embolization alone, direct surgical excision, and preoperative embolization followed by surgical excision. Patients who underwent surgical excision with or without preoperative embolization were included.

The ultrasound (US) device used in this study was the bk5000TM, BK Medical, General Electric (GE) Healthcare, USA (Fig. 1). The ultrasound probe is introduced to surgery in three basic steps: (1) after elevation of bone flap before durotomy, to detect the site and extent of the lesion (Fig. 2); (2) during resection, to evaluate extent of resection and nature of examined blood vessels and direction of blood flow; and (3) before closure, to assure total excision of the lesion. Doppler US was used to define feeding arteries and draining veins throughout the whole procedure.

Fig. 1.

Bk5000 Device, BK medical, GE healthcare, USA

Fig. 2.

Use of an ultrasound device intraoperative.

RESULTS

The study included 20 patients who matched the inclusion criteria, 11 (55%) of whom were males. Patients’ ages ranged from 5 to 55 years. The majority (65%) had intracranial hemorrhage (ICH) on presentation, and presented with headache (n=16, 80%) and seizures (n=13, 65%). Smaller proportions presented with syncope (n=8, 40%), vomiting (n=2, 10%), and motor deficit (n=6, 30%). Thirteen patients (65%) had ICH on presentation.

Six patients (30%) had an AVM with deep drainage, while 14 (70%) had an AVM with superficial drainage. As shown in (Table 1), the SM grading was as follows: grade 1 in one case (5%), grade 2 in five cases (25%), grade 3 in ten cases (50%), and grade 4 in four cases (20%). Three-quarters (n=15, 75%) of patients did not need preoperative embolization, while five (25%) underwent preoperative embolization prior to surgery. The embolic material used for preoperative embolization for all cases was ethylene-vinyl alcohol (EVOH) copolymer. The follow-up period ranged from four to 43 months, with a mean follow-up period of 16.55 months.

Table 1.

Characteristics of AVM

Intraoperative: For all cases, IOUS was successful in confirming the resection of the lesion. In 13 cases in which an intracerebral hematoma took place, the hematoma was visualized easily with grey-scale B-mode ultrasound, and its relation to the nidus could be appreciated. In all cases, complete resection was confirmed by colored Doppler ultrasound, with no flow signal in the lesion’s bed by the end of the surgery, as shown in Table 2.

Table 2.

IOUS findings in AVM cases

Color Doppler US could detect the vessels related to the nidus in 17 (85%) patients (i.e., feeding arteries, draining veins, and vessels en passage), but it failed for a single patient (i.e., 5%), and this step was not attempted for another two patients (10%).

Postoperative functional outcome was evaluated as per conscious level and neurological status in comparison to the preoperative functional status; the results are shown in Table 3. It can be seen that over half of patients (n=11, 55%) were fully conscious, with no deficit preoperatively and postoperatively. Four patients (20%) remained with the same neurological deficit presented preoperatively, while three (15%) developed postoperative deficit due to motor cortex traction, which subsequently resolved within three months of follow-up. One patient (5%) developed a motor deficit without improvement after three months and was still in the physiotherapy program at the time of data collection.

Table 3.

Functional status preoperative and postoperative of AVM cases

Two cases (10%) developed postoperative complications, which required secondary surgery: case no. 4 had CSF rhinorrhea, which required dural repair; and case no. 6 developed postoperative epidural hematoma, which was evacuated surgically (as shown in Table 3). Both complications were related to inadequate closure and were emphatically not attributed to IOUS or AVM excision. One (5%) patient (case no. 10) had a thin rim of acute subdural hematoma, which was managed conservatively without intervention, and another patient (no. 15) developed hydrocephalus upon presentation, which indicated temporary CSF diversion.

Illustrative case

A 30-year-old female patient presented with a history of headache (sudden severe onset), neck rigidity, and blurred vision; she was intact neurologically. The MRI of the brain revealed a Left Frontal AVM with Spetzler-Martin AVM grade 4 (Fig. 3). Digital subtraction angiography showed an AVM nidus with feeding arteries from the anterior cerebral artery, while draining veins into the superior sagittal sinus (SSS) and the straight sinus. The patient underwent preoperative endovascular embolization (Fig. 4), followed by microsurgical excision.

Fig. 3.

Non-contrast MRI: Left frontal lobe AVM. Spetzler-Martin AVM grade 4. MRI, magnetic resonance imaging; AVM, arteriovenous malformation

Fig. 4.

(A) DSA showing AVM during preoperative embolization, (B) CT after embolization, before surgical resection. DSA, digital subtraction angiogram; AVM, arteriovenous malformation; CT, computed tomography

IOU findings: IOU was valuable for the precise localization of the lesion, as well as for the appropriate approach and as an aiding tool to define the lesion by the multiple serpiginous color flow signals of high intensity, indicating a high-flow AVM. Also, IOU defined the residual hemosiderin from the previous hematoma as hyperechogenicity surrounding the AVM (Figs. 5, 6).

Fig. 5.

IOUS finding of AVM case. (A) Color Doppler image showing bidirectional, high blood flow within the tubular structures, suggestive of blood flow within an AVM (red arrow). Note: different grades of red and blue color signals which indicates different flow velocities and directions in the nidus. It also gives an idea about the depth of resection needed (here up to 3.8 cm depth) (white arrow). (B) Major draining vein (white arrow) indicated by color-coded signals, which indicates flow away from the probe and high flow signal intensity in close relation to the falx cerebri, which is hyperechoic (red arrow), indicating drainage in SSS. (C) A large feeding artery is shown (red arrow) indicated by high signal intensity of red color, which indicates flow towards the probe. IOUS, intraoperative ultrasound; AVM, arteriovenous malformation; SSS, superior sagittal sinus

Fig. 6.

IOUS finding of AVM case post-excision. (A) An image post excision showing the absence of the nidus. No evidence of residual AVM. The hypoechoic cavity (red arrow) represents the corridor for resection filled with saline, and the hyperechoic walls of the cavity (white arrow) represent the result of cauterization. (B) Another image post excision with artifact (abnormal color flow signal) produced by the metallic retractor (green arrow). The hypoechoic cavity (red arrow) represents the corridor for resection, filled with saline, and the hyperechoic walls of the cavity white arrow) represents the result of cauterization. IOUS, intraoperative ultrasound; AVM, arteriovenous malformation

Postoperative imaging: No residual nidus was present (Figs. 7, 8).

Fig. 7.

MRI 4 months postoperative of AVM case. (A) sagittal view, (B) coronal view, (C) axial view. A large well-defined area of CSF SI is seen in the left frontal lobe. There are multiple foci of high SI on T1W1, likely hemorrhagic (subacute). The lesion contains multiple signal void structures related to its medial wall with a few small enhancing vessels. AVM, arteriovenous malformation; CSF, cerebrospinal fluid, SI, signal intensity

Fig. 8.

DSA postoperative of AVM case: (A) lateral view (B) AP view. Showing total excision of the AVM with no residual nidus. DSA, digital subtraction angiogram; AVM, arteriovenous malformation; AP, anteroposterior

DISCUSSION

AVM represents one of the most complex and potentially life-threatening neurovascular lesions. Management strategies vary significantly and must be tailored according to the lesion’s angioarchitecture, location, and clinical presentation. While endovascular embolization and radiosurgery are increasingly used as adjuncts or alternatives, microsurgical resection remains the most definitive treatment option in selected patients, especially those with Spetzler-Martin grades I to III [1,4,7].

In this retrospective series analysis of 20 patients, IOUS proved to be a pivotal imaging modality, offering real-time, dynamic guidance during AVM resection. The 100% success rate in nidus localization affirms the high sensitivity of IOUS, especially for lesions situated near eloquent or deep brain regions. This finding aligns with previous literature, which emphasized the ability of IOUS to compensate for intraoperative brain shift, which is a critical limitation in static neuronavigational systems [13,15].

The use of color Doppler ultrasonography was particularly valuable in differentiating feeding arteries and draining veins in 85% of cases. This vascular recognition was vital in tailoring the surgical strategy, which allowed safer and more controlled resection especially, in eloquent areas. The ability to alter probe angulation intraoperatively also enhanced the surgeon’s capacity to assess AVM boundaries and adjacent vascular structures without the need for additional radiation or imaging breaks [9,17,19].

Interestingly, 13 patients (65%) presented with intracerebral hemorrhage (ICH), and for all these cases, IOUS successfully identified the hematoma and its relationship to the nidus. This aspect is of paramount importance in AVM surgery, as hematoma evacuation is often performed concurrently. The simultaneous visualization of both components supports more efficient and targeted surgical maneuvers, decreasing operative time and potential complications.

Furthermore, IOUS confirmed resection in all 20 patients, which is consistent with its well-documented role in intraoperative decision-making and resection assessment. Unlike intraoperative angiography or postoperative DSA, which are resource-intensive and sometimes delayed, IOUS provides instant confirmation of resection status, which is crucial in minimizing the risk of residual nidus and subsequent hemorrhage.

In terms of clinical outcomes, 55% of patients were fully intact postoperatively, and another 20% showed stable neurological status without deterioration. Importantly, three cases experienced transient neurological deficits that resolved within three months, suggesting that these may have been due to temporary intraoperative traction rather than permanent damage. Only one case developed a persistent deficit, highlighting the relative safety of IOUS-guided resections. Two cases required reoperation for complications (CSF rhinorrhea and epidural hematoma), but these were unrelated to IOUS, affirming its procedural safety.

Our findings are consistent with those reported by Della Pepa et al. [6], who demonstrated that IOUS, particularly when combined with color Doppler and contrast-enhanced ultrasound (CEUS), significantly enhanced intraoperative identification of AVM architecture. Their study showed successful identification of feeding arteries and draining veins in over 90% of cases, which contributed to improved resection outcomes. Similarly, in this study, IOUS enabled vascular identification in 85% of patients, and confirmed the successful evaluation of the nidus bed after resection in 100% of cases, indicating that IOUS is a reliable and effective tool during AVM surgery.

These results are in alignment with the outcomes of a major meta-analysis by Gulino et al. [10], which analyzed the use of intraoperative microvascular Doppler in vascular neurosurgery. Their analysis highlighted that intraoperative microvascular Doppler is a reliable and straightforward tool for assessing blood flow velocity during surgery, aiding in the identification of vessel compromise and ensuring complete lesion obliteration. In the current study, the application of IOUS facilitated the identification of feeding arteries and draining veins in 85% of cases, contributing to complete resection in all patients. These parallels underscore the efficacy of IOUS techniques in enhancing surgical precision and patient outcomes in AVM.

The selective use of preoperative embolization in 25% of patients also deserves comment. Although embolization can facilitate resection by reducing intraoperative bleeding and nidus size, it also carries potent risks and may not always be necessary. Our findings suggest that with the real-time vascular assessment provided by IOUS, complete resection is achievable in many cases without embolization. This is in line with studies advocating for judicious, case-by-case application of embolization, rather than routine or default use. The embolic material used for this study was concurrent with previous results, showing the feasibility and efficiency of Onyx for such steps [2,5,11].

One of the major advantages of IOUS is its adaptability and portability, which is particularly valuable in resource-limited settings. It requires minimal setup, is repeatable during the same procedure, and avoids dangerous radiation exposure. However, the technique is operator-dependent and demands a steep learning curve. The effectiveness of IOUS is enhanced by familiarity with ultrasound physics, probe manipulation, and the ability to correlate grayscale and Doppler findings with surgical anatomy.

It is also important to emphasize that successful total resection of all lesions in this study is above the expected outcome in comparison to those reported in literature, but also to the nature of lesions, as all lesions which are amenable to surgical resection in the study are of SM grade 4 or lower, which matches the guidelines in light of the SM grading system. This result is attributed to the location of the lesions, as all lesions involved in this study had cortical presentation; lesions with grade 4 were classified as such due to deep drainage rather than location. Also, exclusion of cases which underwent multi-modal management and the small number of cases included in this study explain these optimistic results.

Additionally, IOUS may have limitations for larger AVMs with irregular architecture, or in cases with deep nidus location obscured by overlying brain edema or hematoma. In three cases (no. 5, 6, and 10) in this study, the visualization of related vessels was not attempted or failed, underlining the importance of experience and standardized protocols in IOUS use.

Compared with intraoperative MRI or DSA, IOUS systems are substantially less expensive, widely available, and do not require a dedicated hybrid operative room (OR), making them a cost-effective navigation option, especially in resource-limited settings [14]. Although IOUS is inherently operator-dependent and has a steeper learning curve than static preoperative imaging, structured training in ultrasound principles and standardized acquisition protocols can flatten this curve. Overall, IOUS offers a low-cost, effective, and progressively more user-friendly modality that can enhance the safety and completeness of cerebral AVM resection when incorporated into routine microsurgical workflow [18].

The limitations of the study are the relatively small number of cases (i.e., 20) and the retrospective nature of the study. Prospective studies with larger cohorts and direct comparison with intraoperative angiography or navigation systems would provide deeper insights into IOUS’s true diagnostic value and cost-effectiveness with specific analysis of duration of surgery, blood loss, and complications. Multicenter trials and inclusion of longterm outcomes could help refine its role in AVM treatment algorithms globally.

CONCLUSIONS

Given its safety, efficiency, and diagnostic precision, IOUS should be considered a fundamental tool in AVM microsurgery. It may be especially valuable in pediatric and young adult populations, where long-term radiation avoidance is crucial. The integration of IOUS with advanced techniques such as CEUS, 3D reconstruction, and elastosonography could further improve lesion characterization and resection control in future practice [3,15].

The findings of this analysis support the routine incorporation of IOUS in AVM microsurgical resection. Its ability to dynamically guide the surgeon throughout all stages of the operation—from initial localization to confirmation of resection—makes it a valuable adjunct in improving surgical precision, reducing morbidity, and ensuring complete nidus obliteration.

Article information Continued