Objective To describe our early experience in treating cerebral arteriovenous malformations (AVMs) with the new Precipitating Hydrophobic Injectable Liquid (PHIL) embolic material.
Materials and methods Between June and August 2015 five patients with cerebral AVMs were treated at two tertiary university hospitals. PHIL was used as complementary treatment to Onyx liquid embolic material or as the sole endovascular treatment.
Results Five patients (average age 39 years (range 19–73)) with ruptured plexiform AVMs were treated. The group included one patient with Spetzler–Martin grade II AVMs, three grade III, and one grade IV. One grade II and two grade III AVMs were cured. A total of nine pedicles were embolized with an average of two pedicles per session. There were no procedural complications. One patient had a seizure after embolization but a brain CT scan showed no hemorrhage.
Conclusions PHIL is a new embolic agent that can be used for the treatment of cerebral AVMs.
- Arteriovenous Malformation
- Liquid Embolic Material
Statistics from Altmetric.com
Liquid embolic materials are the main neuroendovascular tool for treating arteriovenous malformations (AVMs), dural arteriovenous fistulas (AVFs), and some parenchymal vascularized tumors. Until the recent introduction of the Precipitating Hydrophobic Injectable Liquid (PHIL) (Microvention, USA) only n-butyl cyanoacrylate (n-BCA) and ethylene vinyl alcohol copolymer (EVOH) were used commercially under the trade names Onyx liquid embolic (Medtronic, USA) or Squid (Emboflu, Switzerland). The agent chosen is based on the operator’s preference, availability, angioarchitecture of the lesion, and level of expertise with each embolic material.
Herein, we aim to describe our initial experience in treating cerebral AVMs using the new liquid embolic material, PHIL, and discuss its similarity and advantages in comparison with the present liquid embolic agents.
Materials and methods
Five cerebral ruptured AVMs were treated in two university hospitals between June and August 2015. An institutional review board at each institution approved retrospective analysis of prospectively acquired data. Operators with expertise in treating cerebral AVMs were trained in an animal laboratory and with in vitro flow models before using the new liquid embolic material PHIL in patients. Written consent was obtained from the patient and their families before each procedure. A multidisciplinary team comprising neurological surgeons, neurologists, and radiologists decided the best treatment approach.
Pre-embolization was aimed as a complement to surgical resection and/or radiosurgery in AVMs of Spetzler–Martin grade (SM) III or more.1 Endovascular embolization of AVMs SM I or II aimed for radiological cure.
We recorded the following baseline data: age, sex, neurologic examination before and after the procedure, angioarchitecture of the AVM, and clinical presentation. Five patients with ruptured brain AVMs were identified and treated with PHIL 25%. All procedures were performed under general anesthesia. In general, a 6F access catheter was positioned in the pre-petrous portion of the internal carotid artery or V3 vertebral artery segment. A detachable 1.5 cm tip microcatheter—Apollo (Medtronic, USA)—was then navigated coaxially and used to selectively catheterize one or more nidal afferents. After super selective contrast injections demonstrated catheterization of the nidus, a ‘plug’ was created with PHIL in the detachable portion of the microcatheter. The PHIL plug was created in similar fashion to an EVOH plug, allowing some reflux into the detachable portion of the microcatheter and waiting in between injections of liquid embolic to allow solidification. PHIL was then slowly injected to improve nidal penetration. A Headway duo microcatheter (Microvention, USA) was also used in the embolization of some pedicles using the same technique but allowing less reflux. In some instances, a double lumen balloon, such as Scepter C (Microvention, USA) or Eclipse 2L (Balt, France), was used to improve nidus penetration using the ‘pressure cooker’ technique with arterial feeder flow arrest.2
Intravenous heparin was not used during the procedure and a calcium channel blocker was administered intra-arterially before pulling the microcatheter in order to prevent vasospasm.
In general, we waited at least 30 days from rupture to proceed with embolization, allowing clot reabsorption and normalization of the angioarchitecture. However, in cases of impending increased intracranial pressure due to mass effect from the clot and the need for urgent neurosurgical resection, we acutely embolized the nidus to facilitate open surgery.
Five ruptured symptomatic plexiform AVMs were treated (table 1). All the AVMs were treated in one session using PHIL 25% concentration, except for one patient who had had a previous embolization of one pedicle with Onyx. A total of nine pedicles were embolized with an average of two pedicles per session. The median time from rupture to endovascular treatment was 30 days, with a range of 1–90 days.
Three AVMs were SM grade III, one grade II, and one grade IV. Three AVMs were completely closed and at the end of the embolization procedure there was no visualization of arterial afferents or draining veins. One AVM SM III was referred for radiosurgery after embolization of approximately 50% of the nidus and an AVM SM IV was referred for surgical resection after embolization of 60% of the nidus, but this patient died due to a pulmonary embolism 3 months after treatment before surgical resection could be completed. One patient with an AVM SM grade III died during hospitalization from complications of the initial hemorrhage at presentation. In this patient the AVM was embolized 100% after one session, achieving radiological cure. However, the patient's neurological status did not improve over the course of 7 days and the family decided to withdraw care.
There were no procedural complications. The patient with an AVM SM II, which was completely closed with embolization of the main afferent, had a seizure in the recovery the room. Head CT after the procedure showed no evidence of hemorrhagic complications.
Cerebral AVMs are complex lesions that usually require a multimodality approach to achieve cure. Adding another liquid embolic agent with different properties from those of the endovascular armamentarium might improve outcomes and chances of cure. The preliminary experience in treating a dural AVF (dAVF) with PHIL has been recently described.3 However, to our knowledge, reports of experience with this new embolic agent in treating cerebral AVMs are limited and have not been published.
PHIL is a non-adhesive iodine copolymer which, contrary to Onyx and Squid, advances into the vasculature as a ‘block’ instead of forming layers. This property appears to provide a ‘true’ visualization of the treated vessel when PHIL occupies the lumen, instead of progressively occluding the inner lumen through the accumulation of layers. The layering effect or ‘lava-like’ flow pattern of other embolic agents can make it difficult to determine if the vessel is completely occupied by the embolic agent or if there is an inner area of the lumen that still needs to be filled.
Despite being less radiopaque than Onyx and Squid, PHIL was easily visualized on fluoroscopy or magnified substrated roadmap and we had no difficulty in determining the penetration of the agent into the nidus (figures 1⇓–3). Leyon et al,3 described similar findings in the treatment of dAVFs.
In our experience, PHIL behaves similarly to Onyx and Squid, in that it required a proximal plug to ensure forward flow and deeper penetration.4 ,5 Several minutes were needed before a plug was adequately formed with PHIL, similar to the time needed with Onyx or Squid. Once the microcatheter tip was distally positioned into the arterial feeders, optimal penetration into the nidus was invariably achieved. The pressure cooker technique was used in two cases to facilitate anterograde embolization and impede reflux. In one pedicle, suboptimal penetration was seen, mainly because the tip of the microcatheter was far from the nidus and we encountered reflux close to parenchymal branches. Similarly to other EVOH-based liquid embolic agents, PHIL appears to fill different nidus compartments on different injections. It also reaches a forward flow ‘momentum’ that should be maximized to occlude most of the nidus. Leyon et al,3 also described easily achieving forward flow in treating dAVFs.
n-BCA is a free-flowing monomer that rapidly polymerizes within vessels through an ionic reaction. We prefer to use n-BCA in high-flow fistulous aferents.6 Its polymerization rate and visualization depend on the final concentration after mixing it with radiopaque ethiodized oil, a polymerization retardant.7 In very high-flow fistulas, additional 1–3 μm tantalum powder could be added in order to improve radiopacity and lower the concentration of ethiodized oil to ranges of 10–30%. We found no fistulous afferents and it is yet to be determined how PHIL 25% would behave in high-flow shunts in comparison with n-BCA. The company provides two other concentrations of PHIL: 30% and 35%, which might be better suited for embolization of shunting systems. Leyon et al,3 did not report any problems in filling fistulous pouches with PHIL 25%, except when tortuous anatomy prevented distal navigation of the microcatheter to enable the embolic material to penetrate the fistula.
One major difference of PHIL, in comparison with other liquid embolic agents, is that we did not experience any microcatheter occlusion (nine pedicles). In this case series and other non-published embolizations of dAVFs we were able to continue injecting PHIL despite prolonged waiting times between injections. Moreover, it appears that PHIL offers less resistance to forward penetration than Onyx and Squid, despite prolonged injections. Therefore, it is very important to monitor propagation into the draining vein or physiological arterial branches that may inadvertently occlude with PHIL.
Another minor technical difference using PHIL is that it comes in glass syringes, which can make the initial injection more challenging owing to decreased control in comparison with plastic syringes. In contrast to other liquid embolic agents (Onyx and Squid), PHIL does not have to be shaken before injection since it does not have suspended tantalum but does require purging of the delivery microcatheter with dimethyl sulfoxide before injection.
This small case series describes our preliminary experience with a new embolic agent that proved to be useful in the treatment of AVMs. Experienced operators who routinely use other embolic agents should not have a steep learning curve with PHIL. Similarly to Onyx and Squid, with PHIL it appears that one of the main keys for success in penetrating the nidus is to create a robust plug followed by a controlled injection. Its progression as a block allows deeper penetration, but caution is needed when the material is reaching the venous drainage. The use of detachable tip microcatheters and flow arrest techniques are feasible and might enhance the efficacy of PHIL embolization in selected patients.
Contributors EAS and SO-G: study concept and design; all authors: acquisition, analysis, and interpretation of the data; administrative, technical, and material support; EAS and SO-G: drafting of the manuscript: EAS: study supervision and critical revision of the manuscript for important intellectual content.
Competing interests None declared.
Ethics approval Institutional Review Board.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Treatment protocols are available upon request.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.