Autologous Fibrin Glue in Microvascular Decompression for Trigeminal Neuralgia and Hemifacial Spasm

Authors: Ken MATSUSHIMA, Kyosuke MATSUNAGA, Nobuyuki NAKAJIMA, Michihiro KOHNO

Abstract

Adhesive materials are widely used in microvascular decompression for treating neurovascular compression syndromes. They play an important role in the critical step of vessel fixation. Recently, completely autologous fibrin glue produced solely from a patient's own plasma was developed. It can theoretically reduce the risks of viral or prion transmission and allergic reactions, while potentially providing more stable long-term adhesion. This study reports our initial experience with completely autologous fibrin glue in microvascular decompression and evaluates its efficacy and safety. Twenty-seven consecutive patients (12 with trigeminal neuralgia and 15 with hemifacial spasm) underwent microvascular decompression using autologous fibrin glue, including 8 trigeminal neuralgia treated using a fully endoscopic supracerebellar infratentorial approach. Autologous cryoprecipitate and thrombin were prepared (average: 4.8 mL each) from preoperative blood collection (average: 388.9 mL) without adverse events. All offending arteries, including 5 vertebral arteries, were securely transposed using polytetrafluoroethylene and autologous fibrin glue. The fibrin glue was also used to support dural closure with a collagen matrix. All patients experienced complete symptom relief, with no need for postoperative medication, except for 1 patient with mild recurrence despite maintained vascular fixation confirmed on magnetic resonance imaging. No new permanent neurological deficits, cerebrospinal fluid leakage, infections, allergic reactions, or systemic complications were observed. These findings suggest that completely autologous fibrin glue is safe and effective for both vessel fixation and dural sealing in microvascular decompression, without complications related to its use.

Introduction

Microvascular decompression (MVD) is a well-established treatment for neurovascular compression syndromes, such as trigeminal neuralgia, hemifacial spasm, and glossopharyngeal neuralgia.^1-10^^)^ To achieve permanent separation of the offending vessel, several adhesives, such as cyanoacrylate adhesives and commercially available (allogeneic) fibrin glue, have been used to date.^11^^,^^12^^)^ However, cyanoacrylate adhesives are no longer used in MVD because several experimental studies have reported their vascular toxicity, including the formation of dissecting aneurysms.^13-17^^)^ Commercially available fibrin glue, which is produced from human plasma pooled from multiple donors and contains bovine aprotinin, is widely used and considered relatively safe, but it involves risks of viral and prion transmission and allergic reactions.^18-21^^)^ Recent technological advancements have enabled the production of completely autologous fibrin glue derived solely from the patient's own plasma. It can reduce the risks of infection and allergic reactions, and is expected to provide more effective long-term adhesion.^22-24^^)^ However, there are few studies on the use of autologous fibrin glue in MVD. In this study, we present our initial experience with completely autologous fibrin glue in MVD, and evaluate its efficacy and safety.

Materials and Methods

The data of 27 consecutive patients with neurovascular compression syndromes (12 with trigeminal neuralgia, and 15 with hemifacial spasm) who underwent surgical treatment using completely autologous fibrin glue under either a microscope or endoscope between March 2024 and April 2025 were retrospectively reviewed. The patients comprised 9 men and 18 women, with a mean age of 56.6 years (range: 32-76 years). Among the 12 patients with trigeminal neuralgia, 7 with superior cerebellar artery (SCA) compression and 1 with trigeminocerebellar artery compression were treated with a fully endoscopic procedure through the supracerebellar infratentorial approach, without exploring the cerebellopontine angle below the superior petrosal vein. For the remaining 4 trigeminal neuralgia cases, we selected the microscopic retrosigmoid suprafloccular approach, which provided adequate exposure of the entire cisternal course of cranial nerve (CN) V in the cerebellopontine angle through the horizontal fissure opening.^25^^,^^26^^)^ All surgeries for hemifacial spasm were performed using the microscopic retrosigmoid infrafloccular approach, which allowed for elevation of the flocculus and wide exposure around the root exit zone of CN VII by opening the lateral cerebellomedullary fissure.^26^^,^^27^^)^ Auditory brainstem responses (ABRs) were carefully monitored during all intradural procedures, and abnormal muscle responses (AMRs) were additionally monitored in all surgeries for hemifacial spasm. In cases requiring mobilization of the vertebral artery (VA), bilateral extraocular electromyography was also employed to avoid injury to the ipsilateral or contralateral CN VI by the mobilized vertebrobasilar artery system beyond the visible surgical field. Intraoperative indocyanine green angiography was utilized to confirm the patency of the perforators after vessel mobilization. A total of 31 surgeries for neurovascular compression syndrome were performed during the study period; however, the following 4 patients were excluded from this 1 patient from whom autologous fibrin glue could not be produced due to blood coagulation during collection, 2 patients who hesitated to undergo blood collection owing to a positive result in the Treponema pallidum latex agglutination test or Brugada-type electrocardiographic pattern, and 1 patient with trigeminal neuralgia but no offending vessel, who was cured by arachnoid dissection and internal neurolysis (nerve combing).

Patients were scheduled to collect autologous blood (300 or 400 mL) 4 to 24 days before surgery (average: 14.8 days), after obtaining informed consent. A preoperative hemoglobin threshold of ≥11.0 g/100 mL was set for autologous blood collection. The collection volume was 400 mL for patients weighing ≥50 kg and 300 mL for those under 50 kg. In elderly patients or those with mild anemia, oral iron supplementation for 1 to 2 weeks prior to collection was recommended. Completely autologous fibrin glue was produced using the CryoSeal^Ⓡ^ System (Asahi Kasei Medical, Tokyo, Japan), an automated device designed to prepare cryoprecipitate and thrombin from autologous plasma. This cryoprecipitate mainly consists of fibrinogen, factors VIII and XIII, von Willebrand factor, fibronectin, and plasminogen. Cryoprecipitate and thrombin solution were produced and then transferred within a closed system into 2 syringes in approximately 90 mins. Each pair of syringes was stored below −18℃ until the day of surgery. At the start of the surgery, the prepared frozen cryoprecipitate and thrombin solution were thawed in hot water in the operating room and were mainly used during the decompression procedure and to support dural closure. The thawed cryoprecipitate and thrombin solution should be used within 6 hrs to ensure enzymatic activity. For vessel fixation, a tape- or ball-shaped piece of polytetrafluoroethylene (PTFE, Cirrax, CROWNJUN, Tokyo, Japan) soaked with cryoprecipitate was applied to the offending vessel for mobilization and stabilized using dot tips that were separately preloaded with the cryoprecipitate and thrombin. The dura was closed by itself with inlay and onlay support of a collagen matrix (DuraGen^Ⓡ^, Integra LifeSciences, Princeton, NJ, USA), and the cryoprecipitate and thrombin were applied simultaneously using a spray tip to prevent cerebrospinal fluid (CSF) leakage. Concentrated autologous red blood cells and cryoprecipitate-poor plasma from the production of cryoprecipitate and thrombin were reinfused during the surgery in all cases, not due to surgical bleeding, but as part of our standard protocol to reduce the risk of perioperative anemia. Approval for this study was obtained from the institutional research ethics board (study approval T2024-0025).

Results

The mean volume of autologous cryoprecipitate and thrombin produced was 4.8 mL each (range: 2.3-7.5 mL), which were prepared from patients' blood collected preoperatively without any adverse events (Table 1).

All offending arteries were transposed using PTFE and autologous fibrin glue, without requiring an additional allogeneic blood transfusion. Transposition of the VA was required in 5 patients. In 1 patient with trigeminal neuralgia after vestibular schwannoma surgery, the residual tumor, superior petrosal vein, and CN VII, which were adhered to CN V, were interposed in addition to the transposition of the anterior inferior cerebellar artery (AICA), to resolve all possible etiologies. Although exact volumes were not recorded for each case, based on surgical experience, < 1.5 mL of autologous fibrin glue was generally sufficient for vessel fixation, and the remaining volume was used to support dural closure. All patients experienced complete resolution of their symptoms without the need for medication after surgery. Postoperative magnetic resonance imaging (MRI) within 3 days confirmed successful decompression and stable vessel positioning in all patients. None of the patients developed new permanent neurological deficits, CSF leakage requiring spinal drainage or additional surgery, wound infection, allergic reactions, or systemic complications associated with the use of autologous fibrin glue. One patient with trigeminal neuralgia experienced mild symptomatic recurrence 4 months after surgery, although the separation of the offending artery was confirmed on follow-up MRI. He achieved pain relief with a small amount of medication.

Illustrative case 1

A man in his 60s presented with disabling and intractable right trigeminal neuralgia in the V2 and V3 distribution, which began 18 years previously and progressively worsened (Video 1). His pain had been refractory to medication, peripheral trigeminal nerve blocks, and pulsed radiofrequency treatment. Preoperative MRI revealed compression of the right CN V by the SCA (Fig. 1A). From a 400-mL autologous blood sample collected 22 days before surgery, 4.7 mL of autologous cryoprecipitate and thrombin were prepared. The fully endoscopic supracerebellar infratentorial approach, which enables transposition of the SCA without exploring the cerebellopontine angle, was planned under ABR monitoring. Fully endoscopic MVD also enables wet-field confirmation of vessel fixation by recreating the physiological condition in an artificial CSF-filled surgical field.

In the park-bench position, a V-shaped retroauricular incision was made, and a small retrosigmoid craniotomy was performed, while exposing the caudal one-third of the transverse sinus and the transverse-sigmoid sinus junction. The dura was opened to enable superior retraction of the transverse sinus, and the ambient cistern was drained through the fully endoscopic supracerebellar infratentorial approach (Fig. 1B). Arachnoid dissection around CN V revealed the compression of CN V by a loop of the SCA (Fig. 1C). Subsequently, the SCA was wrapped around a tape-shaped PTFE soaked with cryoprecipitate, transposed superomedially to the tentorium, and fixed using the dot tips loaded with the cryoprecipitate and thrombin (Fig. 1D). An endoscopic panoramic view helped to confirm the absence of other offending vessels around CN V, and strong fixation of the SCA was also verified in the wet field after filling with artificial CSF (Fig. 1E). The dura was closed by itself with inlay and onlay support of a collagen matrix, and cryoprecipitate and thrombin were sprayed simultaneously.

The postoperative course was uneventful, and the patient experienced complete relief from trigeminal neuralgia immediately after the surgery, without any medication. His postoperative MRI scan confirmed complete decompression of the nerve, without any complications (Fig. 1F).

Illustrative case 2

A man in his 50s presented with right trigeminal neuralgia in the V2 distribution, which was refractory to medication, peripheral trigeminal nerve blocks, and intranasal lidocaine spray (Video 2). Preoperative MRI displayed compression of CN V superiorly by the SCA and inferiorly by the AICA, along with a dolichoectatic VA (Fig. 2A and B). A total of 400 mL of autologous blood was collected from the patient 14 days preoperatively, yielding 7.5 mL of autologous fibrin glue. For the safe and effective transposition of this ectatic VA, we planned the retrosigmoid suprafloccular approach with the horizontal fissure opening under a microscope. In addition to ABR monitoring, bilateral extraocular electromyography was used to detect potential injury to either the ipsilateral or contralateral CN VI caused by mobilization of the vertebrobasilar artery system beyond the visible surgical field. When abnormal spontaneous electromyographic activity was detected during manipulation of the VA, surgical procedures were temporarily paused or adjusted to avoid nerve injury.

The patient was placed in the park-bench position, and a retrosigmoid craniotomy was performed. The posterior half of the sigmoid sinus was exposed to enable retraction of the sinus, and the opened mastoid air cells were packed with bone wax. After opening the horizontal fissure (petrosal fissure and superior limb of the cerebellopontine fissure), the entire cisternal course of CN V was exposed (Fig. 2C and D). The SCA was first transposed to the tentorium via the supracerebellar infratentorial approach. Then, the VA was transposed and fixed to the petrous dura through the adequately expanded surgical field between CN V and CNs VII/VIII. After confirming secure fixation of the ectatic VA, the AICA was fixed to this transposed VA to achieve complete isolation of CN V (Fig. 2E).

The patient experienced complete symptomatic relief immediately after surgery, with an uneventful postoperative course. His postoperative MRI scan confirmed successful decompression of CN V, with no observable complications (Fig. 2F).

Illustrative case 3

A man in his 50s presented with an 8-year history of left hemifacial spasm that had been refractory to medication and botulinum toxin injections. Preoperative MRI showed compression of the root exit zone of CN VII by the tortuous VA and AICA (Fig. 3A). Autologous blood (400 mL) was collected 24 days before the surgery, and 4.0 mL of autologous fibrin glue was prepared. We planned the retrosigmoid infrafloccular approach under the microscope with ABR and AMR monitoring.

A small retrosigmoid craniotomy was performed in a park-bench position. The sigmoid sinus was retracted by exposing its posterior half, and bone wax was applied to seal the opened mastoid air cells. After opening the lateral portion of the cerebellomedullary fissure while exposing the junction of the lower CNs with the medulla, the VA was transposed and fixed to the petrous dura ventrorostral to the jugular foramen using a tape-shaped PTFE and autologous fibrin glue (Fig. 3B). Mobilization of the VA revealed a small AICA compressing CN VII (Fig. 3C). The AICA was subsequently fixed to the transposed VA (Fig. 3D), and good patency of its perforators was confirmed using intraoperative indocyanine green angiography (Fig. 3E). The AMR disappeared after VA transposition.

The patient was completely relieved of his hemifacial spasm immediately after the surgery and had an uneventful postoperative course. His postoperative MRI scan displayed complete decompression of the nerve without any complications (Fig. 3F).

Discussion

Adhesives for vessel mobilization in MVD

Adhesives are widely used for vessel fixation in MVD, although some authors have reported techniques that do not use adhesives, such as aneurysm clip application or stitched sling retractions.^28^^,^^29^^)^ Regarding adhesives, cyanoacrylate adhesives were widely used for MVD in the past, but their use is now mainly limited to intravascular embolization glue due to their vascular toxicity, which can cause dissecting aneurysms, as well as their potential neural toxicity.^13-17^^,^^30^^)^ Recently, commercially available allogeneic fibrin glue has become widely used and is regarded as relatively safe. However, it involves the risks of viral and prion transmission and allergic reactions. Several authors have documented such cases, with more than 10 cases resulting in anaphylactic shock, including one fatal case.^18-21^^,^^31-34^^)^ Wakamoto et al.^21^^)^ reported that 2 patients developed aseptic meningitis after MVD, and an allergic reaction to fibrinogen was confirmed by a lymphocyte stimulation test in 1 of them, similar to a patient reported by Iuchi et al.,^18^^)^ in which an allergic reaction to thrombin was confirmed by a skin test.

Autologous fibrin glue

Autologous fibrinogen was first described in 1983.^35^^)^ In the conventional system with autologous fibrinogen, cryoprecipitate was produced manually from a patient's own blood plasma, but thrombin was prepared from bovine or pooled human plasma, in which the associated risks could not be eliminated. However, recent technological advancements have enabled the production of completely autologous fibrin glue, containing not only the fibrinogen but also the thrombin. This autologous fibrin glue can theoretically reduce the risks of infection and allergic reactions, as it is produced solely from the patient's own plasma. Building on more than 2 years of our own experience using autologous fibrin glue in various neurosurgical procedures, such as aneurysm clipping, bypass, and tumor resection, we introduced it for MVD, in which adhesives play an essential role in the main vessel fixation. In our initial experience with 27 cases of MVD using autologous fibrin glue, we found that it was effective for vessel fixation and did not cause any recurrence due to prosthesis slippage, including for patients requiring transposition of the dolichoectatic VA. Furthermore, we found that it could also be used for dural closure without causing CSF leakage, requiring additional treatment. Although no quantitative assessment of viscosity was performed, the autologous cryoprecipitate appeared to have slightly lower viscosity than commercial fibrinogen solutions, facilitating better absorption into PTFE material. The safety of autologous fibrin glue was also confirmed, as it caused no adverse events, including infection and allergic reactions. Yoshida et al.^24^^)^ quantitatively evaluated the adhesive strength of autologous fibrin glue in an experimental model, showing that although its initial strength was lower than that of allogeneic glue, it became comparable or superior after 24 hrs. Additionally, Hitomi et al.^22^^)^ reported increased tensile strength at wound sites treated with autologous fibrin glue in an animal model, which may reflect enhanced tissue stabilization over time, although this is not a direct measure of adhesive strength. Lower thrombin activity and multiple active hemostatic ingredients included in autologous fibrinogen were suggested to contribute to its long-term stronger adhesion.^22^^,^^24^^)^ This long-term stable adhesion may contribute to the mechanical stability of vessel fixation, potentially reducing the risk of symptomatic recurrence due to vessel recompression. In addition, some in vivo and in vitro studies suggested the potential antibacterial properties and wound-healing effects of autologous fibrin glue, owing to its immunoglobulins, complement, transforming growth factor-beta, and vascular endothelial growth factor.^36-38^^)^

Kinoshita et al.^36^^)^ reported activity of the autologous fibrin glue against Escherichia coli, while Staphylococcus aureus is the most common pathogen in neurosurgical site infections.

Disadvantages of completely autologous fibrin glue and study limitations

Completely autologous fibrin glue has several disadvantages that should be noted. First, although autologous fibrin glue can be automatically prepared from the patient's own plasma using the CryoSeal^Ⓡ^ System in approximately 90 mins, it poses a risk of post-collection anemia. Blood collection should be scheduled with consideration of a recovery period, and its use should be considered carefully in patients with chronic anemia. Another disadvantage is the interindividual variability in the volume of autologous fibrin glue that can be produced. However, in all patients in this study, autologous up to 400 mL blood was sufficient to produce an amount of autologous fibrin glue equivalent to or greater than that of commercially available fibrin glue.

This preliminary study has several important limitations, including a small sample size, a short follow-up duration, and its retrospective design as a surgical case series conducted at a single institution. Further clinical studies are required to confirm the safety and efficacy of completely autologous fibrin glue in MVD, particularly to verify its potential for reducing recurrence rates through improved long-term adhesion.

Conclusions

Completely autologous fibrin glue is a safe and effective option in MVD, not only for vessel fixation but also for reinforcing dural closure, without any cases of adverse events during preoperative blood collection, symptomatic recurrence due to vessel recompression, CSF leakage, infection, or allergic reactions.

Conflicts of Interest Disclosure

All authors have no conflict of interest, and COI declarations for the past 3 years have been submitted to the Japan Neurosurgical Society office.

Supplementary Material