Perfusion changes associated with intratumoral embolization and feeder occlusion for meningiomas: an arterial spin labeling study

Abstract

Background

Preoperative embolization for meningiomas has been used as an adjuvant therapy for surgery, although its clinical benefits remain controversial. Among embolization strategies, intratumoral embolization has been suggested to reduce tumor blood flow (TBF) more effectively than feeder occlusion; however, objective perfusion-based evaluations remain limited. The purpose of this study was to explore perfusion changes associated with feeder occlusion and intratumoral embolization using arterial spin labeling (ASL).

Methods

Forty-three consecutive patients who underwent preoperative embolization for meningiomas and pre- and post-embolization ASL were classified into intratumoral embolization and feeder occlusion groups. The TBF and cerebral blood flow (CBF) of each patient were calculated from the ASL. The TBF/CBF (T/N) ratios and the flow reduction rate were evaluated for both intratumoral embolization and feeder occlusion groups.

Results

The postoperative T/N ratio significantly decreased in the intratumoral embolization group compared with the preoperative T/N ratio (P = 0.003). Meanwhile, no significant change was observed in the feeder occlusion group (P = 0.36). The flow reduction rate was significantly higher in the intratumoral embolization group than in the feeder occlusion group (50.6% vs. 24.1%; P = 0.001).

Conclusions

In this retrospective cohort, intratumoral embolization was associated with a greater reduction in tumor blood flow than feeder occlusion as assessed by ASL.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00701-026-06790-x.

Introduction

Meningiomas are the most common extraparenchymal tumors, accounting for 13–26% of all intracranial tumors [15]. Most meningiomas are benign and often asymptomatic; however, depending on their location, size, and growth rate, they can become symptomatic and occasionally life-threatening [22, 27]. The standard treatment for meningiomas is surgical resection. As meningiomas are hypervascular tumors, intraoperative blood loss can be a significant concern [25].

Since it was first reported in 1973, preoperative embolization of meningiomas, wherein the blood flow from tumor-feeding vessels is embolized, has been recognized as a useful adjuvant therapy in terms of intraoperative blood loss reduction, tumor softening, and shortened operative time [14, 30]. Recent advances in endovascular therapy and improvements in embolization materials have been reported to enhance the safety and efficacy of endovascular therapy; however, concurrent problems such as complications from the procedure and high risk of embolization of feeding arteries from the ICA system have been noted [4, 24].

There are two types of embolization techniques: feeder occlusion and intratumoral embolization. In the feeder occlusion technique, the feeding vessels proximal to the tumor are embolized, and in intratumoral embolization, the embolizing materials are delivered to the intratumoral vessels. Reportedly, the intratumoral embolization technique reduces tumor blood flow and softens the tumor more than feeder occlusion does; therefore, we aim to perform intratumoral embolization whenever possible [1, 20].

Few studies have objectively quantified the reduction in tumor blood flow after embolization [10, 32]. Furthermore, to our knowledge, only one prior study directly compared feeder occlusion and intratumoral embolization, using dynamic susceptibility contrast perfusion-weighted imaging (DSC-PWI) [2]. Although DSC-PWI allows relative perfusion quantification using contrast bolus tracking, several error sources—including arterial input function variability, bolus delay and dispersion, and blood–brain barrier leakage—limit its reproducibility and absolute quantification accuracy [6, 7]. In addition, susceptibility artifacts and partial volume effects can reduce consistency, and DSC-PWI requires gadolinium contrast administration, which restricts repeatability within a short interval. Conversely, arterial spin labeling (ASL) provides noninvasive, contrast-free, and repeatable perfusion measurement by using magnetically labeled arterial water as an endogenous tracer [7]. ASL is unaffected by contrast leakage or injection variability and can be safely repeated in the peri-procedural period. While ASL is influenced by arterial transit time and generally exhibits a lower signal-to-noise ratio than DSC-PWI, it allows reproducible evaluation of perfusion changes under consistent acquisition conditions [7]. Therefore, ASL serves as a practical and complementary imaging modality to DSC-PWI for assessing hemodynamic changes associated with embolization.

Accordingly, this study aimed to investigate perfusion changes associated with feeder occlusion and intratumoral embolization for meningioma using ASL before and after preoperative embolization.

Methods

Patients

This retrospective study included 43 consecutive patients who underwent preoperative embolization for meningiomas at our institution between 2020 and 2025. The study protocol was approved by the Ethics Committee of Osaka Medical and Pharmaceutical University (Osaka, Japan; Approval No. 2928–3). All procedures adhered to the ethical standards of the institutional and national research committees and were conducted in accordance with the Declaration of Helsinki and its later amendments or comparable ethical standards. Written informed consent was obtained prior to this study in consideration of the use of anonymized clinical data.

Tumor blood flow was evaluated using ASL imaging both before and after embolization. Patients were classified into two groups based on postoperative CT findings: the intratumoral embolization group, in which embolic material was identified within the tumor, and the feeder occlusion group, in which no intratumoral embolic material was detected and embolization was effectively limited to proximal feeder occlusion. Group assignment was thus determined retrospectively and reflected anatomical and technical feasibility rather than a predefined comparative design. During the study period, a total of 160 patients underwent surgical resection for meningiomas at our institution. Among these, preoperative embolization was performed in 67 patients. Of these, 43 patients were included in the present study because both pre- and post-embolization ASL data were available for analysis (Table 1). Patients lacking complete ASL datasets were excluded.

Table 1.

Demographic data of patients

		Intratumoral	Feeder	P-value
	All patients	Embolization	Occlusion
	(n = 43)	(n = 30)	(n = 13)
Age (Avg. ± SD)	61.0 ± 15.3	58.1 ± 15.6	67.9 ± 12.5	0.05
Sex
Male	17	10	7	0.21
Female	26	20	6
Max diameter (Avg. ± SD, cm)	45.4 ± 14.1	45.3 ± 14.8	45.5 ± 12.9	0.98
WHO grading
Grade Ⅰ	34	22	12	0.16
Grade Ⅱ	9	8	1
Grade Ⅲ	0	0	0
Materials
Coil	2	0	2	0.03
NBCA	22	14	8
Coil + NBCA	19	16	3
Intraoperative blood loss (Avg. ± SD, ml)	230.7 ± 184.4	214.5 ± 155.4	268.1 ± 241.8	0.39
Operative time (Avg. ± SD, min)	547.9 ± 185.0	554.6 ± 187.7	532.3 ± 185.0	0.72

Avg. average, NBCA N-butyl-2-cyanoacrylate, SD standard deviation, WHO World Health Organization

Image acquisition

All MRI examinations were performed using a 3.0-T scanner (Ingenia Elition; Philips Healthcare, Best, Netherlands). ASL perfusion imaging was conducted using a three-dimensional spiral fast spin-echo sequence with background suppression covering the whole brain. The imaging parameters were as follows: repetition time/echo time/number of excitations, 4105 ms/16.5 ms/3; post-label delay, 1500 ms; slice thickness, 6 mm; number of slices, 22; field of view, 24 cm; and the labeling plane, at the level of the petrous portion of the internal carotid artery (ICA).

Pre-embolization ASL was performed 0–99 days before embolization (average 14.7 days), and post-embolization ASL was performed 1–5 days after embolization (average 1.7 days).

All digital subtraction angiography procedures were performed using a biplane angiographic system (Artis Zee BA Twin; Siemens, Berlin, Germany).

Imaging analysis

Tumor blood flow (TBF) (mL/min/100 g) was calculated from a region of interest (ROI) manually drawn on the ASL map at the maximal, axial, and cross-sectional areas of each tumor using 3D imaging software (SYNAPSE VINCENT, FUJIFILM Medical Co., Tokyo, Japan), with reference to the findings on contrast-enhanced T1-weighted imaging (Fig. 1A and B) [12]. Additionally, cerebral blood flow (CBF) was calculated by placing another circular ROI of 100 mm² within the contralateral cerebellar hemisphere (in the central cerebellum at the level of the fourth ventricle, 1.5–2 cm away from the ventricle) (Fig. 1C). ROIs were manually delineated by two independent neurosurgeons experienced in ASL interpretation, who were blinded to the embolization technique. Discrepancies in ROI placement were resolved by consensus. The normalized value of TBF to CBF was calculated (T/N ratio = TBF/CBF) for each patient [16]. The flow reduction rate was calculated using the following formula using pre- and postoperative T/N ratios:

$$\text{Flow reduction rate}(\text{\%})=(1-\text{postoperative T}/\text{N ratio}/\text{preoperative T}/\text{N ratio})\times 100$$

Fig. 1.

Region of interest (ROI) placement for arterial spin labeling (ASL) analysis. A A ROI (green outline) was manually delineated on a contrast-enhanced axial T1-weighted image at the maximal cross-sectional area of the tumor. B The same ROI (green outline) was transferred to the corresponding location on the coregistered, postprocessed cerebral blood flow map derived from ASL MRI. C For normalization, a separate circular ROI (green outline) was placed in the contralateral cerebellar hemisphere on the ASL map, avoiding major vessels and imaging artifacts. Min and Max indicate the minimum and maximum pixel values of the ASL perfusion map

The location, embolized vessel, material used for embolization, and flow reduction rate for all patients are summarized in Table 2. The pre- and postoperative T/N ratios were compared between the intratumoral embolization and feeder occlusion groups to determine whether there was a difference in the flow reduction rate (Fig. 2A, B, and C).

Table 2.

Summary of patients who underwent preoperative embolization

Group	Case Number	Location	Feeder		Embolization		Material		NBCA	Flow reduction rate (%)
			ICA	ECA	ICA	ECA	NBCA	Coil	Volume (ml)
Intratumoral embolization	1	Convexity		MMA		MMA	○	○	0.48	34.8
	2	Tentorial		MMA, OA		MMA, OA	○	○	not available	66.9
	3	Cavernous sinus	ILT	AFR, MMA	ILT	AFR, MMA	○		0.39	15.0
	4	Sphenoid ridge	MHT	AMA, MMA		AMA, MMA	○	○	not available	13.9
	5	Clinoidal	ILT	MMA		MMA	○	○	0.23	61.7
	6	Sphenoid ridge		MMA		MMA	○	○	0.26	73.3
	7	Sphenoid ridge	MCA	AMA, MMA, ADTA		AMA, MMA, ADTA	○	○	0.61	26.9
	8	Sphenoid ridge		MMA		MMA	○	○	0.28	26.7
	9	Falx		MMA		MMA	○		0.97	58.5
	10	Convexity		MMA		MMA	○		0.33	62.9
	11	Convexity	MCA	MMA		MMA	○	○	0.52	51.6
	12	Parasagittal	MCA	MMA		MMA	○		0.49	53.7
	13	Sphenoid ridge		AFR, IMA, MMA		IMA, MMA	○	○	0.13	11.2
	14	Parasagittal		MMA		MMA	○		0.40	47.2
	15	Parasagittal	ACA	MMA		MMA	○	○	0.85	16.8
	16	Petrous		APA		APA	○		0.12	71.6
	17	Parasagittal		MMA		MMA	○	○	0.23	87.5
	18	Clinoidal	ILT	AFR	ILT	AFR	○	○	0.10	54.2
	19	Clinoidal	MHT	AFR		AFR	○		0.13	23.3
	20	Convexity		MMA		MMA	○		0.17	70.4
	21	Sphenoid ridge	ILT	AMA, MMA		AMA, MMA	○	○	0.28	48.0
	22	Clinoidal	ACA, MHT	AFR, AMA		AFR, MMA	○	○	0.54	35.5
	23	Convexity		MMA		MMA	○		0.21	57.0
	24	Sphenoid ridge	OpA	AFR, AMA, MMA, STA		AFR, AMA, MMA, STA	○		0.69	54.9
	25	Parasagittal		MMA		MMA	○		0.12	76.0
	26	Clinoidal		AMA		AMA	○		0.17	63.5
	27	Convexity		MMA		MMA	○		0.51	75.9
	28	Petrous	MHT	APA		APA	○		0.18	32.2
	29	Convexity		MMA		MMA	○	○	0.08	61.3
	30	Convexity		MMA		MMA	○	○	0.64	87.2
	Avg									50.6
Feeder occlusion	1	Tentorial	MHT		MHT			○	0	55.7
	2	Petrous		APA		APA		○	0	0.5
	3	Falx		MMA, PMA		MMA, PMA	○		0.26	25.0
	4	Sphenoid ridge		AEA, AMA		AEA, AMA	○		0.29	12.4
	5	Parasagittal		MMA		MMA	○	○	0.20	41.6
	6	Sphenoid ridge	MCA	AMA, MMA		AMA, MMA	○		0.45	0.4
	7	Parasagittal		MMA, OA		MMA	○		0.12	1.8
	8	Planum sphenoidal		AEA, MMA		AEA, MMA	○		0.10	29.8
	9	Petrous	MHT	MMA		MMA	○	○	not available	2.4
	10	Convexity		MMA		MMA	○		0.13	41.1
	11	Petrous		MMA		MMA	○		0.03	11.5
	12	Petrous		APA, AFR, MMA		APA, AFR, MMA	○	○	0.34	38.4
	13	Tentorial		OA		OA	○		0.13	53.0
	Avg									24.1

ACA Anterior cerebral artery, ADTA Anterior deep temporal artery, AEA Anterior ethmoidal artery, AFR Artery of foramen rotundum, AMA Accessory meningeal artery, APA Ascending pharyngeal artery, Avg. Average, ECA External carotid artery, ICA Internal carotid artery, ILT Inferolateral trunk, IMA Internal maxillary artery, MCA Middle cerebral artery, MHT Meningohypophyseal trunk, MMA Middle meningeal artery, NBCA N-butyl-2-cyanoacrylate, OA Occipital artery, OpA Ophthalmic artery, PMA Posterior meningeal artery, STA Superficial temporal artery

Fig. 2.

Changes in tumor blood flow before and after embolization assessed by ASL. A In the intratumoral embolization (IE) group, the tumor-to-normal (T/N) ratio significantly decreased after embolization compared with before embolization (pre, 2.64 ± 1.87; post, 1.38 ± 1.19; *P = 0.003). B In the feeder occlusion (FO) group, no significant difference was observed between pre- and post-embolization T/N ratio (pre, 2.88 ± 2.01; post, 2.21 ± 1.67; P = 0.36). C The flow reduction rate was significantly higher in the IE group than in the FO group (IE, 50.6 ± 22.2; FO, 24.1 ± 20.5; *P = 0.001).Data are presented as mean ± standard deviation. Pre, pre-embolization; post, post-embolization

Endovascular therapy

Preoperative embolization was considered for meningiomas that were relatively large (generally > 30 mm in maximum diameter) and demonstrated marked hypervascularity on preoperative angiography. Tumors with a dominant blood supply from the external carotid artery (ECA) system, such as the middle meningeal artery, were preferentially selected. Tumor location and clinical presentation were also considered, particularly in cases where substantial intraoperative blood loss was anticipated. The indication was discussed and determined by multiple neurosurgeons at our institution.

All procedures were performed by experienced neurointerventionalists under general anesthesia via a transfemoral approach. N-butyl-2-cyanoacrylate (NBCA; Histoacryl, B. Braun Melsungen AG, Melsungen, Germany) and bare platinum coils were used as embolic materials. In all cases, NBCA was used at a fixed concentration of 13%, which was standardized across patients and not adjusted according to feeder size or flow dynamics. Systemic heparinization (60–80 units/kg) was administered intra-arterially, maintaining the activated clotting time at approximately twice the baseline value. A 6- or 7-Fr guiding catheter was positioned in the external carotid artery (ECA) or ICA, and a microcatheter (Marathon®, Medtronic, Tokyo, Japan) was navigated as close as possible to the tumor using a distal access catheter (Cerulean® 4-Fr, Medikit, Tokyo, Japan).

Target vessels for embolization were determined based on angiographic findings. Priority was given to branches of the ECA, such as the middle meningeal artery (MMA) and occipital artery. In selected cases, feeders from the inferolateral trunk (ILT) or meningohypophyseal trunk (MHT) of the ICA were also targeted when they were judged to contribute significantly to tumor perfusion.

The selection of embolic material and technique depended on the vascular anatomy and proximity to cranial nerve–supplying arteries. Neurovascular branches (e.g., the petrosal branch of the MMA and the neuromeningeal branch of the ascending pharyngeal artery) were treated with coils to avoid cranial nerve palsy caused by NBCA entering these branches. Non-neurovascular branches of the ECA were embolized with NBCA whenever feasible to achieve intratumoral penetration. ICA feeders were generally avoided because they often supply normal brain parenchyma, posing a high risk of cerebral infarction. However, when embolization of the ILT or MHT was considered essential, proximal feeder occlusion using coils was performed if catheterization could not advance beyond the origin of these trunks to prevent NBCA reflux into the ICA. When the microcatheter could be advanced closer to the tumor, NBCA was carefully injected to achieve intratumoral embolization. Therefore, the embolization strategy was individualized for each case and not selected for the purpose of comparative evaluation. Specifically, feeder occlusion was chosen when distal microcatheter navigation close to the tumor was technically difficult, when feeders originated from the ICA with a risk of reflux, or when the feeding vessels supplied cranial nerves. In some cases, intratumoral embolization was attempted with NBCA; however, feeder occlusion was ultimately achieved because the embolic agent failed to reach the tumor due to unfavorable flow dynamics. For the purpose of analysis, embolization patterns were retrospectively classified based on postoperative computed tomography findings: intratumoral embolization was defined as the presence of embolic material within the tumor parenchyma, indicating intratumoral penetration, whereas feeder occlusion was defined as embolization limited to proximal feeding vessels without evidence of intratumoral embolic material. All identified tumor feeders and those actually embolized are detailed in Table 2.

Statistical analysis

Given the small sample size, all statistical analyses were performed in an exploratory manner.

Continuous variables were compared using Student’s t-test or the Mann–Whitney U test, as appropriate. Within-group pre/post comparisons of the T/N ratio were performed using paired t-tests. Categorical variables were compared using Fisher’s exact test. Statistical significance was set at P < 0.05.

As an additional analysis, a subgroup analysis restricted to patients who underwent post-embolization ASL on the day following embolization was performed to assess the potential impact of ASL timing on perfusion measurements (Supplementary Table 1 and Supplementary Fig. 1).

Results

There were no significant differences in age, sex, maximum tumor diameter, or WHO grading between the two groups. No WHO grade 3 cases were observed in either group. There were no significant differences in intraoperative blood loss and operative time between the two groups (Table 1).

The tumors were located in the sphenoid ridge (n = 9, 20.9%), convexity (n = 9, 20.9%), parasagittal (n = 7, 16.3%), petrous (n = 6, 14.0%), clinoidal (n = 5, 11.6%), tentorial (n = 3, 7.0%), falx (n = 2, 4.7%), planum sphenoidal (n = 1, 2.3%), and cavernous sinus (n = 1, 2.3%). Most of the embolized vessels were branches of the ECA, with only three cases of embolization of the ILT or MHT branches of the ICA. The flow reduction rates ranged from 11.2% to 87.5% in the intratumoral embolization group and from 0.4% to 55.7% in the feeder occlusion group, reflecting variable degrees of reduction in intratumoral blood flow across patients in both groups (Table 2).

In the intratumoral embolization group, embolization significantly decreased the T/N ratio (P = 0.003), whereas no significant differences were observed in the feeder occlusion group (P = 0.36) (Fig. 2A and B). The flow reduction rate was significantly higher in the intratumoral embolization group than in the feeder occlusion group (50.6% vs. 24.1%, respectively; P = 0.001) (Fig. 2C).

In a subgroup analysis restricted to patients who underwent post-embolization ASL on the day following embolization, a similar trend in flow reduction was observed (Supplementary Table 1 and Supplementary Fig. 1).

Neurological deficit

Among all cases, postoperative neurological deficits were observed in five patients, all of which were transient. Postoperative MRI showed no ischemia or hemorrhage in any case. Mild postoperative cerebral edema was observed in one patient with a falx meningioma, who experienced transient mild weakness of the right lower extremity that resolved within 4 days. Cranial nerve-related symptoms occurred in three patients with clinoidal or sphenoid ridge meningiomas. Facial dysesthesia was observed in two patients (one sphenoid ridge and one clinoidal meningioma) and was resolved at 1 month and 2 years, respectively. In addition, one patient with a clinoidal meningioma developed transient oculomotor nerve palsy accompanied by facial sensory disturbance, both of which resolved within 4 months. Another patient with a clinoidal meningioma experienced transient motor aphagia, which resolved within 4 days.

All cases of transient neurological deficits were in the intratumoral embolization group, in which embolization was performed using NBCA. In contrast, no postoperative complications were observed in any patient in the feeder occlusion group.

Representative cases

Representative cases of intratumoral embolization and feeder occlusion are shown in Figs. 3 and 4, respectively.

Fig. 3.

Representative case of intratumoral embolization for a convexity meningioma. A Pre-embolization brain computed tomography (CT) showing a left temporal convexity lesion with surrounding edema. B Pre-embolization contrast-enhanced T1-weighted imaging (CE-T1WI) demonstrating left convexity meningioma. C Pre-embolization arterial spin labeling (ASL) showing a markedly increased perfusion signal within the tumor. D Pre-embolization digital subtraction angiography (DSA) revealing the left middle meningeal artery as the primary feeder. E Post-embolization DSA demonstrating disappearance of the tumor blush following intratumoral embolization. F Post-embolization CT confirming deposition of embolic material within the tumor. G Post-embolization CE-T1WI showing partial loss of contrast enhancement within the tumor. H Post-embolization ASL demonstrating a marked reduction in the intratumoral hyperperfusion signal compared with pre-embolization

Fig. 4.

Representative case of feeder occlusion for a petrous meningioma. A Pre-embolization brain contrast-enhanced T1-weighted imaging (CE-T1WI) showing a meningioma extending from the right cerebellopontine angle to Meckel’s cave. B Pre-embolization arterial spin labeling (ASL) demonstrating increased perfusion within the tumor. C Pre-embolization digital subtraction angiography (DSA) identifying the meningohypophyseal trunk as the main feeding artery. D Post-embolization DSA showing marked reduction of the tumor blush after proximal feeder occlusion. E Post-embolization CE-T1WI demonstrating slight reduction of contrast enhancement. F Post-embolization ASL showing a reduction in the hyperperfusion signal compared with pre-embolization

Case 1 (intratumoral embolization group)

A patient in their 60 s was referred to our department after an incidental meningioma was detected on head imaging performed in another department. CT and MRI showed a neoplastic lesion with a maximum diameter of 42 mm in the left temporal convexity (Fig. 3A and B). Tumor removal was scheduled after tumor embolization. Preoperative ASL showed a TBF of 1157 (mL/100 g/min), a CBF of 242.2 (mL/100 g/min), and a T/N ratio of 4.78 (Fig. 3C). External carotid angiography revealed a significant tumor blush (Fig. 3D). The MMA was the primary feeder for the tumor. The MMA parietal branch and posterior convexity branch were embolized with NBCA, and the petrosal branch was embolized with coils for feeder occlusion. External carotid angiography showed complete disappearance of tumor staining (Fig. 3E). Internal carotid angiography revealed that the middle cerebral artery was the main feeder, and no tumor embolization was performed in the branches of the ICA. Post-embolization CT showed embolic material within the tumor, and MRI showed a partial contrast defect in the tumor (Fig. 3F and G). Post-embolization ASL showed a TBF of 1086 (mL/100 g/min), a CBF of 469.9 (mL/100 g/min), and a T/N ratio of 2.31 (Fig. 3H). Based on the pre- and postoperative T/N ratios, the rate of blood flow reduction was 51.6%. A total tumor resection was performed 5 days after embolization. Intraoperative blood loss was 180 ml. Pathological examination revealed a clear-cell meningioma. The patient was discharged without postoperative complications and has had no recurrence since discharge.

Case 2 (feeder occlusion group)

A patient in their 70 s was followed for a suspected meningioma. As the tumor gradually enlarged and trigeminal neuralgia appeared, surgery was planned. MRI showed a neoplastic lesion with a maximum diameter of 30 mm in the right cerebellopontine angle extending to Meckel’s cave (Fig. 4A). Preoperative ASL showed a TBF of 2009.7 (mL/100 g/min) and a CBF of 346.6 (mL/100 g/min), with a T/N ratio of 5.80 (Fig. 4B). The main feeder was the tentorial artery of the MHT (Fig. 4C). Feeder occlusion was performed using coils, and tumor staining disappeared on internal carotid angiography (Fig. 4D). Post-embolization MRI showed no obvious contrast deficit within the tumor, although there was a slight decrease in tumor volume (Fig. 4E). Post-embolization ASL showed a TBF of 1122.2 (mL/100 g/min) and CBF of 437.2 (mL/100 g/min), with a T/N ratio of 2.57 (Fig. 4F). Based on the pre- and postoperative T/N ratios, the flow reduction rate was calculated to be 55.7%. Tumor resection was performed on the sixth day after embolization, and the tumor in Meckel’s cave remained; however, the remainder was removed. Intraoperative blood loss was 30 ml. Pathological examination revealed meningothelial meningioma. The patient was discharged without postoperative complications and has had no recurrence since discharge.

Discussion

This study is, to our knowledge, the first to examine tumor blood flow changes after intratumoral embolization and feeder occlusion for meningiomas using ASL. Traditionally, the efficacy of preoperative embolization has been evaluated based on intraoperative blood loss, operative time, and hemoglobin changes. However, these parameters are influenced by various confounding factors, such as tumor location, vascularity, and surgeon experience, making objective evaluation difficult. Additionally, intraoperative impressions such as tumor hardness and ease of removal are inherently subjective [23, 31].

To overcome these limitations, several studies have utilized ASL to objectively measure TBF before and after embolization [10, 32]. DSC-PWI has also been utilized for similar purposes [9, 11]. Although DSC-PWI can provide perfusion maps with a high signal-to-noise ratio, its quantitative accuracy may be affected by arterial input variability, susceptibility artifacts, and contrast leakage [6, 7]. ASL, on the other hand, offers a noninvasive, contrast-free, and repeatable method that can be safely applied in the periprocedural period. Despite the known limitations of ASL, such as potential underestimation due to arterial transit time effects, its reproducibility and low inter-session variability make it a practical tool for longitudinal evaluation of embolization efficacy [7, 10, 32].

In the present study, the T/N ratio significantly decreased after embolization in the intratumoral embolization group (P = 0.003), whereas no significant change was observed in the feeder occlusion group (P = 0.36), suggesting that simple proximal feeder occlusion may not consistently result in a measurable reduction in tumor perfusion (Fig. 2A and B). This may be explained by the fact that proximal feeder occlusion does not directly disrupt the intratumoral microvascular network and may allow residual perfusion through distal branches or collateral supply. In contrast, intratumoral embolization enables deeper penetration of embolic material into the tumor vasculature, leading to a more effective reduction in tumor blood flow. Furthermore, the mean flow reduction rate was significantly higher in the intratumoral embolization group than in the feeder occlusion group (50.6% vs. 24.1%, P = 0.001), confirming that deeper intratumoral penetration of embolic material more effectively decreases TBF (Fig. 2C). These findings align with previous DSC-PWI–based reports [2] and extend them by demonstrating that ASL can reliably detect perfusion changes without the need for contrast administration, reinforcing its utility for objective hemodynamic assessment in neurointerventional settings. In a subgroup analysis limited to patients who underwent post-embolization ASL on the day following embolization, a similar trend was observed. The flow reduction rate remained higher in the intratumoral embolization group than in the feeder occlusion group (53.4 vs. 24.7, P = 0.005), although the number of cases was limited and the group sizes were unbalanced (Supplementary Fig. 1). This finding suggests that the observed perfusion differences were not solely attributable to variability in ASL timing; however, the results should be interpreted with caution given the small sample size.

Although intratumoral embolization achieved greater reduction in TBF, there were no significant differences in intraoperative blood loss or operative time between the two groups. This discrepancy may be explained by several confounding factors, including tumor size, site, vascularity, and surgical complexity, as well as individual variability in surgical technique. Moreover, embolization may contribute more to qualitative factors such as tumor softening and ease of resection rather than directly reducing measurable blood loss, particularly in a limited sample size. Larger multicenter studies using standardized surgical parameters are warranted to further clarify the clinical impact of preoperative perfusion reduction.

Intratumoral embolization is often performed using particulate embolic agents because of their ability to penetrate small-caliber intratumoral vessels. However, particulate embolization is associated with limited controllability, heterogeneous distribution, and a potential risk of non-target embolization through dangerous anastomoses. In contrast, NBCA allows precise delivery under continuous fluoroscopic monitoring, provides permanent occlusion, and enables controlled intratumoral penetration when the microcatheter can be advanced sufficiently close to the tumor. For these reasons, and because of institutional experience and familiarity, NBCA was selected as the primary embolic material in this study. Nevertheless, the choice of embolic material may influence both the degree of flow reduction and the risk profile, and this should be considered when interpreting the present results.

Preoperative embolization carries a risk of procedure-related complications, although improvements in endovascular techniques and embolic materials have enhanced both safety and efficacy [5, 8, 26, 28]. In our study, transient neurological deficits occurred only in patients treated with NBCA. Postoperative MRI revealed no ischemia or hemorrhage, and these symptoms were attributed to post-embolization neurological syndrome (PENS). According to Tanaka et al., PENS is likely caused by transient tumor swelling or peritumoral inflammation following embolization [29]. Although prophylactic corticosteroids were not used in this study, previous reports have demonstrated their effectiveness in preventing PENS, suggesting that pre-embolization steroid administration should be considered when appropriate [17]. In our series, transient neurological deficits occurred mainly in patients with clinoidal meningiomas. Because these tumors are often located adjacent to cranial nerves, local inflammatory changes or transient tumor swelling after NBCA embolization may more readily result in cranial nerve–related symptoms. In addition, tumors located near eloquent cortical areas may cause transient neurological deficits due to peritumoral edema or inflammation following embolization, even in the absence of radiological ischemia.

In this study, no significant difference in WHO grading was observed between the intratumoral embolization and feeder occlusion groups. However, several previous reports have suggested that preoperative embolization can induce ischemic or necrotic changes in tumor tissue, occasionally increasing MIB-1–positive cells and potentially complicating histopathological grading [13, 18, 19, 21]. Therefore, even when no apparent difference in WHO grade is detected, the potential influence of embolization-related tissue changes should be considered when interpreting postoperative pathological findings. Importantly, post-embolization necrosis can usually be distinguished from the micronecrosis characteristic of biologically aggressive meningiomas, allowing pathologists to differentiate iatrogenic changes from true malignant features [3, 19]. Close collaboration between neurosurgeons and pathologists remains essential for accurate pathological assessment after embolization.

This study has several limitations. First, it was a retrospective, nonrandomized, single-center study with a small sample size, limiting statistical power. In addition, several cases in the feeder occlusion group showed minimal reduction in tumor blood flow, which may have disproportionately influenced the between-group comparison. Such low-response cases likely reflect the inherent limitation of proximal feeder occlusion rather than technical failure, but they increase variability within the feeder occlusion group and complicate interpretation of between-group differences in a non-randomized cohort. Second, tumor location and vascular supply were heterogeneous and not analyzed separately, which may have affected intraoperative parameters. Third, differences in the proportion of ICA and ECA feeders, together with the use of a single post-labeling delay and post-embolization ASL acquisition within 1–5 days, may have introduced bias related to arterial transit time and temporal variation in perfusion. Although a subgroup analysis restricted to cases with post-embolization ASL performed on the following day yielded similar results, the limited number of patients and imbalance between groups preclude definitive conclusions. In particular, delayed or collateral tumor perfusion may not be adequately captured by single-delay ASL, potentially leading to underestimation of residual tumor blood flow. This limitation may be especially relevant in cases of proximal feeder occlusion, where alternative feeding pathways or collateral circulation remain despite embolization. Future studies employing multi-delay or vessel-selective ASL may allow more accurate characterization of tumor perfusion dynamics and separation of ICA and ECA contributions. Finally, the use of NBCA differed significantly between groups, meaning that embolic material itself could have influenced flow reduction independent of technique, representing an inherent confounding factor. Importantly, embolization techniques were selected based on vascular anatomy and technical feasibility rather than for the purpose of direct comparison.

Conclusions

Using ASL, this study evaluated changes in tumor blood flow before and after preoperative embolization for meningiomas. Intratumoral embolization was associated with a greater reduction in tumor blood flow than feeder occlusion in this retrospective cohort. Although intratumoral embolization achieved a more pronounced hemodynamic effect, no significant differences were observed in intraoperative blood loss or operative time between the two techniques. These findings suggest that ASL may serve as a useful, noninvasive tool for objectively assessing the hemodynamic impact of embolization strategies, while highlighting the need for further studies with larger, prospective cohorts to clarify the clinical implications.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

ASL

Arterial spin labeling

CBF

Cerebral blood flow

CT

Computed tomography

DSC-PWI

Dynamic susceptibility contrast-enhanced perfusion-weighted imaging

ECA

External carotid artery

ICA

Internal carotid artery

ILT

Inferolateral trunk

MHT

Meningohypophyseal trunk

MMA

Middle meningeal artery

MRI

Magnetic resonance imaging

NBCA

N-butyl-2-cyanoacrylate

PENS

Post-embolization neurological syndrome

ROI

Region of interest

TBF

Tumor blood flow

T/N ratio

Tumor-to-normal ratio

WHO

World Health Organization

Author Contribution

M.Fu. (Masao Fukumura) conceived the study, designed the research, performed data analysis, and drafted the manuscript. H.K., K.Y., T.Ko., Y.F., S.T., Y.T., N.O., G.F., R.Ya., R.H. contributed to clinical data collection, angiographic evaluation, and assisted with endovascular procedures. M.Ka., N.N., M.Fu., S.K., T.T., M.W. performed the neurosurgical operations and contributed to clinical management. M.Ka., N.N., M.Fu., S.K., T.T., M.W. also supervised the study, verified analytical methods, and critically revised the manuscript for important intellectual content. All authors reviewed and approved the final manuscript.

Funding

No funding was received for this research.

Data availability

The datasets generated and analyzed during the current study are not publicly available due to patient confidentiality but are available from the corresponding author on reasonable request.

Declarations

Ethical approval

This study was approved by the Ethics Committee of Osaka Medical and Pharmaceutical University (Approval No. 2928–3). All procedures were conducted in accordance with the ethical standards of the institutional and national research committees and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Consent to participate

Written informed consent was obtained from all individual participants included in the study.

Consent for publication

Not applicable (no identifiable personal data are included).

Competing interests

The authors declare no competing interests.