Differences and Advantages of Particles versus Liquid Material for Preoperative Intracranial Tumor Embolization: A Retrospective Multicenter Study

Abstract

Objectives

The superiority and usefulness of liquid material over particles for embolization have been a topic of debate due to differences in materials and techniques. This study aimed to identify the complications and outcomes associated with both embolization materials.

Methods

This retrospective multicenter cohort study included 93 patients from an endovascular treatment registry, treated from January 1, 2018 to May 31, 2022. It included patients who underwent preoperative embolization for meningioma, solitary fibrous tumor/hemangiopericytoma, and hemangioblastoma. Data for patient characteristics, procedural factors, complications, and outcomes were collected from medical records.

Results

A tortuous access route was the only factor independently associated with complications (p = 0.020). Although liquid material was more frequently used for embolization in relatively high-risk conditions, complication rates did not differ significantly between the groups (p = 0.999). In the liquid material group, the tip of the microcatheter could be guided closer to the tumor (p <0.001) using a distal access catheter and flow-guide microcatheters. The subgroup middle meningeal artery embolization had less operative bleeding in the liquid material group (p <0.001), whereas the particles group exhibited less intraoperative blood loss than the liquid material group (p = 0.006).

Conclusion

The vascular tortuosity of the access route was only associated with complications in preoperative tumor embolization. Liquid material and particles showed no difference in complication rates. The use of particles in embolization may reduce intraoperative bleeding, but not in all cases can it be used safely. Therefore, a thorough understanding of the characteristics of both approaches and their relative advantages in clinical practice is essential to opt for the appropriate material according to the case.

Introduction

Preoperative intracranial tumor embolization is performed for hypervascular tumors prior to surgical resection to reduce intraoperative blood loss and shorten the surgical time.¹⁾ The selection of embolization materials, such as liquid material, particles, or coils, depends on various factors, including the tumor type and location, the branch to be embolized, and the devices used. Complications, such as intratumoral hemorrhage, ischemia, cranial nerve palsy, as well as expected effects, vary depending on embolization materials. Several studies have demonstrated and compared the effectiveness of particles and glue in preoperative intracranial tumor embolization^2–4); however, no consensus on the choice of material has been reached. Borg et al. reported both embolic materials to be effective,⁵⁾ while Fang et al. reported that intraoperative blood loss and operative time were significantly lower using Glubran, which is composed of liquid material and methacryloxy-sulfolane, compared to using polyvinyl-alcohol.³⁾ Moreover, a previous study reported that liquid material use was a risk factor for preoperative embolization complications.⁶⁾

Preoperative tumor embolization using liquid material exhibits excellent vascular selectivity as the microcatheter can be freely inserted. In contrast, while a microcatheter with a larger inner diameter is required for preoperative tumor embolization using particles, the particles can penetrate deeper into the tumor vascular bed, reducing intraoperative blood loss. Nevertheless, no study to date has directly compared particles and liquid material under the conditions described above. Therefore, this study aimed to identify the advantages of both particles and liquid material to appropriately select an embolization material.

Materials and Methods

Study design and participants

This retrospective multicenter cohort study was conducted using data obtained from an endovascular registry of patients who underwent preoperative tumor embolization at three centers between January 2018 and May 2022. Owing to the retrospective study design, the Ethics Committee of our institution waived the requirement of obtaining written informed consent, offering participants an opt-out option in accordance with the Personal Information Protection Law and National Research Ethics Guidelines. The study protocol was reviewed and approved by the ethics committees of the three centers. The study was performed in accordance with the ethical standards of the 1964 Declaration of Helsinki and its later amendments.

The liquid material and particles used in this study were obtained from n-butyl cyanoacrylate: Histoacryl (B. Braun, Melsungen AG, Melsungen, Germany) and Embosphere (Nippon Kayaku Co. Ltd., Tokyo, Japan), respectively. The following data were collected from patients’ medical records: sex, age, medical history, blood test result, preoperative modified Rankin Scale (mRS) score, and symptomatology. Details of the embolic materials used, such as liquid material and particles of various sizes (100–300 μm, 300–500 μm), were recorded. Types of tumors evaluated included meningioma, solitary fibrous tumor/hemangiopericytoma (SFT/HPC), and hemangioblastoma. The volume of the tumor was calculated using the formula: 4/3 × π × (length/2) × (width/2) × (height/2), where length, width, and height represent the tumor dimensions. Target vessels involved (external carotid artery [ECA], internal carotid artery [ICA], vertebral artery [VA]), and the type of anesthesia were also documented. In addition, the procedural time for tumor embolization, time from embolization to surgery, intraoperative hemorrhage volume, and intraoperative transfusion requirements were recorded. Furthermore, cases requiring two or more embolization sessions, the employment of a provocation test, the use of a distal access catheter (DAC), and the use of a flow guide microcatheter were recorded. We also measured the distance between the tumor and the catheter tip, noting when this distance was within 10 mm.

Deep-seated tumors were defined in the following locations: anterior cranial fossa, orbital, sphenoid ridge, cavernous, petroclival, tentorial, and posterior fossa. A tortuous access route was defined by the presence of a type 3 aorta,⁷⁾ acute bifurcation angles of the brachiocephalic trunk and the left common carotid artery (CCA) from the aorta,⁸⁾ bovine origin to the left CCA, a coiled and kinked extracranial ICA,⁹⁾ and a tortuous iliofemoral artery with acute angulation.¹⁰⁾

Consecutive treatment cases during the observation period were included. The exclusion criteria were age <18 years and embolization with coils alone or using both particles and liquid material.

Preoperative embolization procedure

The protocol used was similar to that of a recent study on embolization with patients treated using liquid material.¹¹⁾ All patients requiring vessel embolization were selected based on diagnostic angiography. The timing of the embolization procedure was standardized to the day before surgery. All surgical interventions were conducted within approximately one week following the embolization. The indications for preoperative embolization included large tumors (at least 40 mm in the greatest diameter), hypervascular tumors, and difficulty in processing or inability to process the tumor-feeding artery early in the surgery. Embolization was performed when the operator determined that it was safe, i.e., when there was no dangerous anastomosis, the provocation test exhibited negative results, and the tip of the catheter reached the tumor. The devices used for embolization were selected based on the discretion of the operator at each institution. Typically, embolization of the ECA was performed under local anesthesia. Conversely, when embolizing feeders originating from the ICA or VA, the procedures were carried out under general anesthesia using a DAC. The microcatheter was inserted into the feeding artery as close as possible to the tumor, and embolization was performed using Embosphere or n-butyl cyanoacrylate. The embolic material was selected at the operator’s discretion depending on several factors, such as the arterial anatomy, the position of the microcatheter, and the selected feeding vessels. We used low-concentration n-butyl cyanoacrylate, or Embosphere sized 100–300 µm if the microcatheter was guided far enough distally and there was no dangerous anastomosis or vasa nervorum. On the contrary, high-concentration n-butyl cyanoacrylate or Embosphere larger than 300 µm in size were used if the microcatheter was located in a proximal vessel or if a high-flow or large feeder was to be embolized. In cases where there were concerns regarding embolic material migration into dangerous anastomosis or vasa nervorum, provocation tests were performed before embolization. The provocation test was performed using 1 mL (20 mg) of lidocaine diluted twofold, slowly infused over 1–2 minutes. Subsequent to the infusion, patients were monitored for the emergence of neurological symptoms. If any neurological symptoms appeared, the procedure was limited to proximal occlusion using coils.

Outcome measures

Complications were defined as any adverse event associated with the embolization procedure occurring in the period between post-embolization and pre-surgery. This includes both symptomatic and asymptomatic complications. The types of complications included symptomatic and asymptomatic bleeding, ischemia, cranial nerve paralysis, peritumoral edema, and others. Complications that caused and did not cause a deterioration in the mRS score were classified as major and minor, respectively, according to the classification of neurosurgical complications by Landriel et al.¹²⁾ The embolization outcomes evaluated included the degree of the disappearance of tumor staining, operative time, amount of blood loss during resection, presence or absence of blood transfusion, and postoperative tumor recurrence. Furthermore, we considered the impact of intraoperative bleeding during craniotomy and determined whether a sinus-crossing craniotomy was performed or not by examining the presence of liquid materials and particles. Since staining disappearance is subjective, two measures were compared: disappearance of 60% or more staining and complete disappearance of staining. In meningiomas, the surgical outcome was evaluated by Simpson grade classification based on surgical records and follow-up magnetic resonance imaging scans.¹³⁾ For other tumors, the surgical outcome was evaluated based on whether the tumors were totally resected or not. In addition, patient outcomes were assessed based on the time to recurrence at the last follow-up.

Statistical analyses

Quantitative data are presented as mean ± standard deviation or median, and interquartile ranges and categorical data as frequencies (percentages). Pearson’s χ² or Fisher’s exact and Mann–Whitney U tests were used for comparisons between the groups. In addition, multivariate analysis was performed using a logistic regression model for significant variables in the univariate analysis. Statistical significance was set at p <0.05. All statistical analyses were performed using SPSS version 25 software (SPSS, Armonk, NY, USA).

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Results

Patient characteristics

The registry contained data for 93 (39 [41.9%] men and 54 [58.1%] women) consecutive embolizations performed between January 1, 2018 and May 31, 2022. The median patient age was 65 years (interquartile range [IQR], 50–73 years). The preoperative mRS score was 0–2 in 85 patients (91.4%). Embolization with liquid material and particles was performed in 47 (50.5%) and 46 (49.5%) patients, respectively. Particles sized 100–300 μm were used in 19 patients (41.3%), while those sized 300–500 μm were used in 27 patients (58.7%). The types of tumors embolized were meningiomas (77 patients, 82.8%), SFT/HPC (seven patients, 7.5%), and hemangioblastomas (nine patients, 9.7%). Deep-seated tumors were identified in 43 (46.2%) patients. Embolization of the ECA and branches of the ICA, or VA, were performed in 81 (87.1%) and 12 (12.9%) patients, respectively. Only 10 (10.8%) patients were embolized under general anesthesia, while the remaining underwent local anesthesia. The patients’ background data are summarized in Table 1.

Table 1. Patient characteristics.

Factors	Total (n = 93) n (%)
Age (Med [IQR]) (years)	65 [50, 73]
Male (%)	39 (41.9)
Preoperative mRS 3–5	8 (8.6)
Symptomatic	65 (69.9)
Embolic materials
Liquid material	47 (50.5)
Particles	46 (49.5)
Particles 100–300 μm	19 (17.9)
Particles 300–500 μm	27 (25.5)
Type of tumors
Meningioma	77 (82.8)
SFT/HPC	7 (7.5)
Hemangioblastoma	9 (9.7)
Tumor volume (Med [IQR]) (mm3)	22608.0 [10823.1, 34365.0]
Deep-seated location	43 (46.2)
Target vessels
ECA	81 (87.1)
ICA	2 (2.2)
VA	10 (10.8)
General anesthesia	10 (10.8)
Result of embolization*
Complete disappearance	8 (8.7)
More than 60% disappearance	58 (63.0)
Procedure time in tumor embolization (Med [IQR]) (min)	102.50 [78.75, 130.00]
Median time from embolization to surgery (Med [IQR]) (day)*	1 [1, 2]
Intraoperative hemorrhage (Med [IQR]) (mL)*	150 [100, 300]
Intraoperative transfusion*	10 (10.9)
Complication
Any complication	9 (9.7)
Major complication (change mRS)	2 (2.2)
Simpson grade 1–3 or gross total removal*	70 (76.1)
Postoperative recurrence**	14 (15.1)

*One missing value.

**Two missing values.

ECA: external carotid artery; ICA: internal carotid artery; Med [IQR]: median [interquartile range]; mRS: modified Rankin Scale; SFT/HPC: solitary fibrous tumor/ hemangiopericytoma; VA: vertebral artery

Univariate and multivariate analysis of complications in preoperative tumor embolization

The results of the univariate analysis of preoperative embolization-related complications are presented in Table 2. Complication rates did not differ significantly between the liquid material and particle groups. A tortuous access route was the only factor independently associated with complications (p = 0.020) (Table 2). Complications were observed in nine patients: seven involved minor complications, while two had major complications. The details of the complications are presented in Table 3. The complications included cerebral infarction (n = 3), cranial nerve palsy (n = 3), dural arteriovenous fistula due to vessel perforation (n = 1), peritumoral edema (n = 1), and transient facial dysesthesia (Table 3). Postoperative complications were categorized as grade Ia in seven patients and grade Ib in two patients in the classification of neurosurgical complications (Table 3).

Table 2. Univariate and multivariate analysis of complications in preoperative tumor embolization.

Variate	Occurrence of complication		Univariate	Multivariate
	Yes = 9 (9.7) n (%)	No = 84 (90.3) n (%)	p-Value	OR (95% CI)	p-Value
Age (Med [IQR]) (years)	57 [48, 66]	65 [50, 73]	0.377
Male (%)	6 (66.7)	33 (38.3)	0.158
Preoperative mRS 3–5	1 (11.1)	7 (8.3)	0.572
Hypertension	6 (66.7)	30 (35.7)	0.084
Diabetes mellitus	3 (33.3)	8 (9.5)	0.070	4.972 (0.920–26.88)	0.063
Dyslipidemia	3 (33.3)	18 (21.4)	0.417
Tortuous access route	4 (44.4)	10 (11.9)	0.027	6.277 (1.338–29.457)	0.020
Type 3 aortic arch	0 (0.0)	5 (6.0)	0.999
Blood test results (mean ± SD)
Hb	14.16 ± 1.18	13.76 ± 1.53	0.454
Plt	23.30 ± 4.79	24.02 ± 6.02	0.731
Cre (Med [IQR])	0.83 [0.68, 1.04]	0.685 [0.60, 0.78]	0.050	1.144 (0.612–2.140)	0.674
eGFR (Med [IQR])	70.1 [54.0, 80.3]	74.5 [66.3, 85.9]	0.264
Two or more embolization sessions	3 (33.3)	32 (38.1)	0.999
Provocation test	0 (0.0)	18 (21.4)	0.198
Using distal access catheter	5 (55.6)	28 (33.3)	0.272
Using flow guide microcatheter	5 (55.6)	43 (51.2)	0.999
Distance between tumor and catheter tip within 10 mm	4 (44.4)	30 (35.7)	0.720
Symptomatic	6 (66.7)	59 (70.2)	0.999
Embolic materials			0.999
Liquid material	5 (55.6)	42 (50.0)
Particles	4 (44.4)	42 (50.0)
Particles in size (μm)			0.234
100–300	0 (0.0)	19 (22.6)
300–500	4 (44.4)	23 (27.4)
Type of tumors
Meningioma	7 (77.8)	70 (83.3)	0.650
SFT/HPC	0 (0.0)	7 (8.3)	0.999
Hemangioblastoma	2 (22.2)	7 (8.3)	0.209
Deep-seated location	2 (22.2)	41 (48.8)	0.170
Tumor volume (Med [IQR]) (mm3)	11487.8 [8921.0, 23854.1]	23095.5 [12902.35, 38196.2000]	0.157
Target vessels
ECA	7 (77.8)	74 (88.1)	0.327
ICA	0 (0.0)	2 (2.4)	0.999
VA	2 (22.2)	8 (9.5)	0.248
General anesthesia	2 (22.2)	8 (9.5)	0.248
Procedure time in tumor embolization (Med [IQR])	90 [67, 93]	105 [85, 130]	0.148
Result of embolization*
Complete disappearance	1 (11.1)	7 (8.4)	0.576
More than 60% disappearance	7 (77.8)	51 (61.4)	0.477

*One missing value.

ECA: external carotid artery; ICA: internal carotid artery; Med [IQR]: median [interquartile range]; mRS: modified Rankin Scale; SFT/HPC: solitary fibrous tumor/ hemangiopericytoma; VA: vertebral artery

Table 3. Embolization-related complications.

Types of complication	Total (n = 9)	Liquid material (n = 5)	Particles (n = 4)	Classification of neurosurgical complications
				Glade Ia (n = 7)	Glade Ib (n = 2)	Glade IIa–IV (n = 0)
Hemorrhage	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
Ischemia	3 (33.3)	3 (60)	0 (0)	2 (28.6)	1 (50)	0 (0)
Cranial nerve palsy	3 (33.3)	0 (0)	3 (75)	2 (28.6)	1 (50)	0 (0)
Dural AVF due to vessel perforation	1 (11.1)	1 (20)	0 (0)	1 (14.3)	0 (0)	0 (0)
Peritumoral edema	1 (11.1)	1	0 (0)	1 (14.3)	0 (0)	0 (0)
Transient facial dysesthesia	1 (11.1)	1 (20)	1 (25)	1 (14.3)	0 (0)	0 (0)
Outcomes of complications
Minor	7 (77.8)	4 (80)	3 (75)	7 (100)	0 (0)	0 (0)
Major	2 (22.2)	1 (20)	1 (25)	0 (0)	2 (100)	0 (0)

AVF: arteriovenous fistula

Comparison between liquid material and particle embolization

The findings of the univariate analysis of the factors related to liquid material and particle embolization are presented in Table 4. Liquid material was used in 47 patients (50.5%), and particles in 46 patients (49.5%). The two groups demonstrated no differences in patient age (p = 0.051), sex (p = 0.472), tumor volume (p = 0.735), tumor stain disappearance rate (complete disappearance: p = 0.268; more than 60%: p = 0.785), complications (p = 0.751), and postoperative recurrence rate (p = 0.227). Particles were used frequently in meningiomas (particles vs. liquid material: 91.3% vs. 74.5%; p = 0.031), while NBCAs were used frequently in hemangioblastomas (particles vs. liquid material: 2.2% vs. 17.0%; p = 0.030). Liquid material use involved multiple sessions (two or more embolization sessions: particles vs. liquid material: 26.1% vs. 48.9%; p = 0.023), a higher percentage of provocation tests (particles vs. liquid material: 6.5% vs. 31.9%; p = 0.002), and more frequent use of the DAC (particles vs. liquid material: 4.3% vs. 66.0%; p <0.001). Moreover, almost all liquid material cases (95.7%) involved the use of flow-guided microcatheters and embolization closer to the tumor than the cases that used particles (distance between tumor and catheter tip within 10 mm: particles vs. liquid material: 17.4% vs. 55.3%; p <0.001). In all cases involving the use of particles, the embolization was performed under local anesthesia. Liquid material was used under general anesthesia in 21.3% of cases. The particle group involved less intraoperative blood loss than the liquid material group (particles vs. liquid material: median, 100 mL vs. 245 mL; p = 0.009). There is a possibility that craniotomies crossing a sinus may lead to increased intraoperative blood loss. Therefore, to evaluate the effect of embolic materials on blood loss, we compared the incidence of craniotomies crossing a sinus between groups using liquid material and particles. There was no significant difference in the rates of crossing a sinus between the two groups (sinus-crossing craniotomy: particles vs. liquid material: 12.8% vs. 8.7%; p = 0.740). Additionally, the percentage of Simpson grades Ⅰ–Ⅲ (including total resection) was higher in the liquid material group than in the particle group (particles vs. liquid material: 62.2% vs. 89.4%; p = 0.002).

Table 4. Comparison between liquid materials and particles.

	Liquid material n = 47 (%)	Particles n = 46 (%)	p-Value
Age (mean ± SD) (years)	57.5 [47, 70.5]	68 [55, 73]	0.051
Male (%)	18 (38.3)	21 (45.7)	0.472
Type of tumors
Meningioma	35 (74.5)	42 (91.3)	0.031
SFT/HPC	4 (8.5)	3 (6.5)	0.999
Hemangioblastoma	8 (17.0)	1 (2.2)	0.030
Two or more embolization sessions	23 (48.9)	12 (26.1)	0.023
Provocation test	15 (31.9)	3 (6.5)	0.002
Using distal access catheter	31 (66.0)	2 (4.3)	<0.001
Using flow guide microcatheter	45 (95.7)	3 (6.5)	<0.001
Distance between tumor and catheter tip within 10 mm	26 (55.3)	8 (17.4)	<0.001
Tumor volume (mm3)	23112.60 [8180.85, 32659.20]	22447 [15860, 34604]	0.735
Target vessels
ECA	36 (76.6)	45 (97.8)	0.002
ICA	2 (4.2)	0 (0)	0.495
VA	9 (19.1)	1 (2.2)	0.015
Deep-seated location	25 (55.6)	18 (39.1)	0.174
General anesthesia	10 (21.3)	0 (0)	0.001
Result of embolization (tumor stain disappearance rate)*
Sinus-crossing craniotomy*	4 (8.7)	6 (12.8)	0.740
Complete disappearance	6 (12.8)	2 (4.4)	0.268
More than 60% disappearance	29 (61.7)	29 (64.4)	0.785
Procedure time in tumor embolization (Med [IQR]) (min)	97 [73.5, 124.5]	105 [85, 130]	0.450
Intraoperative blood loss (Med [IQR]) (mL)*	245 [100, 400]	100 [100, 200]	0.009
Intraoperative transfusion*	8 (17.0)	2 (4.4)	0.091
Complication
Any complication	5 (10.6)	4 (8.7)	0.751
Major complication (change mRS)	1 (2.1)	1 (2.2)	0.999
Simpson grade I–III (including total resection)*	42 (89.4)	28 (62.2)	0.002
Postoperative recurrence**	5 (10.6)	9 (19.6)	0.227

*One missing value.

**Two missing values.

ECA: external carotid artery; ICA: internal carotid artery; mRS: modified Rankin Scale; SFT/HPC: solitary fibrous tumor/hemangiopericytoma; VA: vertebral artery

Discussion

This was a retrospective, multicenter, observational study of consecutive patients who underwent preoperative tumor embolization for meningiomas, SFT/HPC, and hemangioblastomas. We compared patients embolized with liquid material or particles and identified the characteristics of the different embolic materials. We hypothesized that understanding the characteristics of each material would allow for optimal embolization material selection. To the best of our knowledge, this is the first study to compare liquid material and particles in such detail in relation to the type of tumor, endovascular devices, distance between the tumor and catheter tip, and target vessels.

Liquid material embolization

Hishikawa et al. reported that the embolization of feeding vessels other than the ECA and the use of liquid materials could increase the complication rate in intracranial tumor embolization.⁶⁾ In this study, as shown in Tables 3 and 4, infarction occurred more frequently in the liquid material group because it was used more frequently for embolization in relatively high-risk patients, such as those involving branches other than the ECA, deep-seated locations, and hemangioblastoma. However, the liquid material group had the same complication rate as the particle group. This may have been due to the use of more provocation tests, DAC, and flow-guide microcatheters to guide the microcatheter tip as close to the tumor as possible in embolization using liquid material rather than particles. Accordingly, embolization using liquid material may be safe and effective in cases involving branches other than the ECA or hemangioblastomas.

Particles embolization

The particle group exhibited significantly less intraoperative blood loss and required fewer sessions than the liquid material group. These characteristics of particle embolization remained consistent regardless of the type of branch or tumor. As for intraoperative blood loss, particles may be transported farther and penetrate deeper into the tumor vascular bed with liquid material.⁴⁾ Therefore, particles may contribute to a greater reduction in intraoperative blood loss compared to liquid material. Accordingly, in cases where minimal intraoperative blood loss is required, particles may be a better choice. In addition, particle embolization has the advantage of reducing treatment costs because effective embolization can be achieved in a single procedure, and the same catheter can be used several times.

Nonetheless, the main complication associated with particle usage is cranial nerve deficit. To prevent cranial nerve deficits, anatomical knowledge of the vasa nervorum and dangerous anastomosis, as well as analysis of preoperative angiography data, are crucial. If the microcatheter cannot be guided distal to the site of dangerous anastomosis or vasa nervorum, larger-sized particles should be used, which are less likely to cause neurological symptoms if they migrate to potential anastomotic connections or feeding vessels supplying cranial nerves. Smaller particles can lead to more effective preoperative embolization and intratumoral necrosis of meningiomas than larger particles because vascular beds within meningiomas have been reported to range from 10 to 50 μm.¹⁴⁾ However, small particles can cause complications such as intratumoral hemorrhage and neurological deficits. Sluzewski et al. reported that using large particles, such as 400-μm calibrated microspheres, is safer than using small polyvinyl-alcohol particles, as no hemorrhagic complications occurred with the large, calibrated microspheres.¹⁵⁾ In this study, particles sized 300–500 μm were used in 58.7% of cases, and intraoperative blood loss (p = 0.009) was significantly less than that observed with liquid material. Therefore, particles sized 300–500 μm may be sufficient to reduce intraoperative blood loss. In addition, Okada et al. reported safe and effective embolization using particles sized 500–700 μm, which were diluted sufficiently to allow the relatively large particles to reach the tumor without clogging the microcatheter or causing feeder occlusion.⁴⁾ They were able to prevent complications of preoperative embolization for skull-base meningiomas.

Based on these findings, 300–500 μm-sized particles should be used to prevent complications and facilitate effective preoperative embolization when using particles if the microcatheter can be guided close enough to the tumor and there is no risk of migration to the vasa nervorum or dangerous anastomosis. In addition to the present study, the results of embolization with 100–300 μm-sized particles and 300–500 μm-sized particles were also compared (Supplementary Table 1). There was no difference in technique or complications. It might be possible to safely embolize with 100–300 μm-sized particles if they are sufficiently close to the tumor. Conversely, particles having 500–700 μm in size should be used if the distance between the microcatheter and the tumor is long and if there are branches between the catheter and the tumor that may lead to dangerous anastomosis or vasa nervorum. The particle group demonstrated a significantly lower tumor resection rate because this group may have had more cases wherein the tumors were in the convexity and involved venous sinuses, precluding tumor resection to preserve the veins.

Complication of preoperative intracranial tumor embolization

In this study, only vascular tortuosity was significantly associated with complications. This was assumed because the microcatheter could not be guided distally in cases where the access route was tortuous, and embolization was performed when embolic materials were injected from the proximal site of branches of dangerous anastomosis and vasa nervorum. Therefore, aggressive embolization should be avoided when a tortuous access route prevents guiding the microcatheter close to the tumor to prevent preoperative complications. The overall and major complication rates were 9.7% and 2.2% in this study, respectively. In a previous study, for cases of meningiomas, the pooled complication rate associated with preoperative embolization was 4.3% (0%–25%), and serious complications occurred in approximately 2.3% of the cases.¹⁶⁾ In a study by Ilyas et al., overall complication, major complication, and mortality rates associated with preoperative embolization of skull-base meningiomas were 12%, 6%, and 0.2%, respectively.¹⁷⁾ For hemangioblastoma embolization, Ampie et al. previously reported an overall complication rate of 11.7% and a mortality rate of 1.8%. They showed that preoperative embolization of hemangioblastomas did not reduce the risk of resection and that embolization itself is not standard care.¹⁸⁾ Although the complication rate in this study cannot be easily compared with those reported previously because this study included intracranial tumors of various types and locations, the findings can be considered to be approximately equal to those reported previously.

Limitations

The present study showed no difference between the liquid material and the particle groups in terms of the degree of embolization on cerebral angiography. Although a previous study reported that approximately 80% of embolizations prolonged the period to recurrence, most of the embolization data in that study were obtained with particles.¹⁹⁾ Accordingly, further research is required to determine whether liquid material has the same effect of prolonging recurrence.

This study had several limitations: First, it was a retrospective study with a small number of patients. Accordingly, the amount of data and statistical power were limited. Second, the analysis had various biases owing to differences in patient baseline characteristics, such as tumor type and location. These biases were also observed during the surgery, particularly with regard to intraoperative blood loss. The precise assessment of the contribution of tumor embolization to intraoperative blood loss presents challenges, particularly when tumors are localized to parasagittal or cavernous regions. Since intraoperative bleeding is reflected by the amount of blood lost during skin incision and craniotomy; thus, we differentiated cases based on whether the craniotomy crossed the sinus and compared the embolic substances used, finding no significant difference. However, as we did not separately assess the blood loss from skin incision to craniotomy and the blood loss during tumor resection, it was challenging to accurately evaluate to what extent preoperative tumor embolization contributed to the reduction of intraoperative bleeding. Furthermore, the study does not address the impact of pathological findings on intraoperative blood loss. This is particularly pertinent for meningiomas, where different subtypes, such as meningothelial and angiomatous meningiomas, may influence the volume of hemorrhage differently. Thus, the absence of a detailed investigation into these pathological distinctions poses a limitation. Therefore, future research with a larger prospective cohort is warranted to address these limitations.

Conclusion

Liquid material and particles showed no difference in complication rates. The vascular tortuosity of the access route is associated with complications in preoperative tumor embolization. Therefore, aggressive embolization should be avoided in cases involving vascular tortuosity that cannot be guided sufficiently distally with a microcatheter. Preoperative embolization using particles may be more effective in reducing intraoperative blood loss in a single procedure. However, since the distance between the tumor and the microcatheter tip in embolization using particles is longer than that with liquid material, appropriate anatomical knowledge (e.g., an understanding of potential anastomotic connections) is essential.

Acknowledgments

We wish to thank all the staff at the participating institutions for their support in preoperative embolization, as well as the study participants and their families.

Author Contributions

Yu Iida and Taisuke Akimoto contributed to the conception and design of the study, acquisition of data, and analysis and interpretation of data. Yu Iida, Taisuke Akimoto, Shigeta Miyake, Ryosuke Suzuki, Wataru Shimohigoshi, Jun Suenaga, Yasunobu Nakai, and Katsumi Sakata collected clinical data. Tetsuya Yamamoto supervised all aspects of this study. All the authors have read the final manuscript and approved its submission for publication. All authors attested to meeting the four ICMJE authorship criteria.

Disclosure Statement

The authors declare that they have no conflicts of interest.