Article Text

Download PDFPDF

Case series
3D-guided direct puncture therapeutic embolization of intracranial tumors
  1. Jildaz Caroff1,
  2. Nidhal Benachour1,
  3. Léon Ikka1,
  4. Jérôme Nevoux2,
  5. Fabrice Parker3,
  6. Valerio Da Ros1,4,
  7. Cristian Mihalea1,5,
  8. Marta Iacobucci1,6,
  9. Jacques Moret1,7,
  10. Laurent Spelle1,7
  1. 1Department of Interventional Neuroradiology, NEURI Brain Vascular Center, Bicêtre Hospital, Le Kremlin-Bicêtre, France
  2. 2Department of Otorhinolaryngology, Le Kremlin-Bicêtre, France
  3. 3Department of Neurosurgery, Bicêtre Hospital, Le Kremlin-Bicêtre, France
  4. 4Department of Diagnostic Imaging and Interventional Radiology, Molecular Imaging and Radiotherapy, Policlinico Tor Vergata, Rome, Italy
  5. 5Department of Neurosurgery, University of Medicine and Pharmacy “Victor Babes”, Timisoara, Romania
  6. 6Department of Bioimaging and Radiological Sciences, Policlinico “A. Gemelli”. Rome, Italy
  7. 7Faculté de Médecine, Université Paris Sud, Le Kremlin-Bicêtre, France
  1. Correspondence to Dr Jildaz Caroff, Department of Interventional Neuroradiology, NEURI Brain Vascular Center, Bicêtre Hospital, 78 Rue du General Leclerc, Le Kremlin-Bicêtre 94270, France; Jildaz.caroff{at}aphp.fr

Abstract

Background Direct punctures of intracranial tumors have rarely been described in the literature.

Objective To assess the feasibility, safety, efficacy, and advantages of using 3D DSA-guided direct puncture rather than the traditional transarterial route to preoperatively devascularize intracranial lesions in particular clinical situations, paying special attention to any correlation with surgical observations; we present the largest series to date.

Methods Between July 2015 and July 2016, data from all presurgical embolizations performed in our institution were prospectively collected. Information on tumor type, location, size, eventual bone erosion, complications, devascularization percentage, and estimated blood loss was analyzed.

Results Tumors of four patients (two meningioma, two endolymphatic sac tumor) were embolized using direct puncture. 3D XperGuide planning software was used in all procedures. Embolization was feasible in all cases. In one case, a small craniotomy was specifically performed to allow needle positioning. In all cases n-butyl cyanoacrylate was used. No ischemic or hemorrhagic complications related to embolization occurred. Complete or near complete devascularization was obtained in all cases. In one case, surgery was not performed and the patient was monitored. Resection was complete without significant blood loss in two cases, and resection was incomplete but satisfactory in one case.

Conclusions In selected cases, 3D-guided direct puncture of intracranial tumors appears safe, feasible, and efficient for preoperative embolization.

  • Embolic
  • Intervention
  • Liquid Embolic Material
  • Malignant
  • Tumor
View Full Text

Statistics from Altmetric.com

Introduction

Preoperative transarterial embolization of central nervous system tumors is known to be a safe and efficient technique that reduces blood loss, operative time, and surgical morbidity.1–5 Indeed, Kim et al6 stated that “it is hard to justify nonembolization of a hypervascular skull base tumor”.

However, the transarterial approach can sometimes be impossible, owing to tortuous and/or distal access through tiny arteries. It can also sometimes be a high-risk strategy, either because of the presence of dangerous anastomosis or because the arterial tumor feeder also vascularizes normal parenchyma (which occurs mostly in cases of internal carotid artery-originated feeders).

Indeed, the risks associated with preoperative embolization are not to be overlooked. In a recent meta-analysis, Shah et al7 found a 4.6% rate of immediate complications.

Direct puncture embolization has achieved good clinical results in the treatment of head and neck tumors.8–11 According to some, “this method may play an increasing role in the preoperative treatment of highly vascularized tumors in various locations”.12 However, there are few reports of intracranial lesions treated using this technique.12–16

We sought here to assess the feasibility, safety, efficacy, and advantages of using 3D DSA-guided direct puncture rather than the transarterial route to preoperatively devascularize intracranial lesions in particular clinical situations, with special attention paid to correlations with surgical observations.

Materials and methods

Data collection and analysis

Between July 2015 and July 2016, data from all presurgical embolizations performed in our institution were prospectively collected in a database. Information on tumor type, location, size, eventual bone erosion, complications, devascularization percentage, and estimated blood loss was analyzed.

Endovascular technique

Informed consent was obtained in each case. All procedures were performed by two experienced interventional neuroradiologists. All patients were treated under general anesthesia and systemic heparinization. Endovascular treatment was performed using a biplane flat panel angiographic system (Allura Xper 20/15, Philips Healthcare, Best, The Netherlands). Percutaneus arterial transfemoral access was always used to allow angiographic pre-embolization anatomical analysis and permanent control of both tumoral blush and normal brain vascularization through a 5F catheter.

XperGuide software

Lesions were punctured with a 22-gauge needle under fluoroscopic guidance using the XperGuide software (Philips Healthcare). This software allows the interventionalist to define, using an intraprocedural cone-beam CT scan (XperCT, Philips Healthcare), a precise needle entry point on the skin's surface and a needle trajectory to access the tumor while avoiding the normal brain parenchyma (figures 1 and 3). The C-arm is then automatically positioned in one of the two orthogonal positions (entry point view and progression view), with a superimposition of the cone-beam CT images and the virtual trajectory onto the real-time fluoroscopic image, allowing guidance up to the defined depth.10

Figure 1

A middle-aged patient was referred to our institution with an anterior skull base meningioma with a frontal sinus invasion, revealed by a subcutaneous tumefaction. (A) Preoperative MRI showing a traditional transarterial embolization performed in the external territory with particulate embolics, with unsatisfactory results. In a second session, a direct puncture embolization was performed through the bone defect. (B) Follow-up MRI demonstrating almost complete resection. (C) XperGuide Application progression view, corresponding to the superimposition of the DSA on the CT image and the predefined trajectory to the target point. (D) Fluoroscopic view of the positioning of the first needle. Eleven puncture sites allowed a good devascularization (>80%) but a thin peripheral blush remained. (E and F) Fluoroscopic view of the final glue cast. (G) Pre-embolization left internal carotid angiogram. (H) Final common left carotid angiogram.

Figure 3

A middle-aged patient was monitored in our institution for von Hippel-Lindau disease with cerebellar and bulbar hemangioblastoma. (A) Post-gadolinium MR demonstrating a hypervascular endolymphatic sac tumoral remnant after a partial resection due to massive operative bleeding. (B) CT showing the craniotomy bone defect (arrow). During the following year, the patient presented with worsening clinical symptoms, including a loss of right-sided hearing, sixth and seventh nerve palsies, and intense increasing headaches. (C and D) Subsequent vertebral and external carotid lateral angiograms revealing an intense tumoral blush. The patient was embolized with direct puncture through the previous craniectomy. (E and F) Both entry point and target were defined using the XperGuide application (arrows). The needle was then positioned using 3D fluoroscopic guidance with the superimposed trajectory from the entry point view (G) and progression view (H). (I and J) Lateral and anteroposterior views of the final cast of glue. (K and L) Final lateral DSA of the vertebral and external carotid arteries demonstrating a very good angiographic result (devascularization >90%). Postoperatively, headaches improved and the neurological status stabilized; it was therefore decided to monitor the patient without surgical removal of the tumor.

Reflux of blood was obtained before direct injection of the contrast agent into the tumor, to confirm good positioning and to analyze the tumoral parenchymography and local venous drainage.

Embolization

Embolization was performed with a mixture of 15% n-butyl cyanoacrylate formulated as Glubran (GEM, Viareggio, Italy) and 85% of lipiodol under permanent fluoroscopic roadmapping. The liquid embolic material was injected with a 3 mL luer lock syringe. The duration of the injections varied from a few seconds to several minutes. If required, multiple puncture sites were used, in cases of intratumoral interstitium extravasation, or when the embolic material was no longer progressing sufficiently to allow complete filling of all different compartments of the lesion. All needles were removed only at the end of the procedure, to ensure complete polymerization and hemostasis. Following the procedure, a brain angiogram was obtained to evaluate the extent of the devascularization and to rule out any normal artery occlusion. A C-arm Xpert CT scan was also repeated to analyze the penetration of the embolic material into the tumor and to detect potential bleeding.

Results

Over the study period, four patients (two male, two female, age range 7–74 years) underwent direct percutaneous puncture embolization of intracranial tumors. Tumor pathology was meningioma in two cases and endolymphatic sac tumor in the other two.

3D XperGuide planning software was used in all procedures. Embolization was feasible in all cases. In only one case, needle positioning required a specifically performed small craniotomy. Bone erosion allowed access in the two cases of meningioma; a previous craniotomy defect allowed access in the final case. The needle trajectory never penetrated the normal brain parenchyma. No ischemic or hemorrhagic complications related to embolization occurred. In one case, direct puncture embolization followed a partial and unsatisfactory transarterial embolization. Complete devascularization was obtained in two cases, and near complete devascularization (>80%) in two.

In one case, the patient was observed. In the remaining three cases, the mean interval between embolization and surgery was 3.7 days (range 1–7). Resection was complete without significant blood loss in two cases and resection was incomplete but satisfactory in one case. The effectiveness of devascularization was judged satisfactory by the surgeons in all three cases.

Illustrative cases

Case 1

A middle-aged patient was referred to our institution with an anterior skull base meningioma with a frontal sinus invasion, revealed by a subcutaneous tumefaction (figure 1). A traditional transarterial embolization performed in the external territory with particulate embolics resulted in an unsatisfactory devascularization. In a second session, a direct puncture embolization was performed through the bone defect. Eleven puncture sites allowed a good devascularization (>80%) but a thin peripheral blush remained. This may have been the cause of significant bleeding during surgical resection of the skull base insertion of the meningioma 3 days later. However, follow-up imaging demonstrated almost complete resection.

Case 2

An elderly patient presented with a parasagittal meningioma (figure 2). DSA revealed a hypervascular lesion with partial occlusion of the superior sagittal sinus. The feeders from the external carotid artery appeared extremely tortuous, and so direct puncture was performed through a skull thinning using a 14-gauge sternal bone marrow aspiration needle. Complete devascularization was obtained and complete resection was performed 7 days later without any complications.

Figure 2

An elderly patient presented with a parasagittal meningioma. (A) Coronal view of enhanced CT. DSA reveals a hypervascular lesion with partial occlusion of the superior sagittal sinus. (B) Feeders from the external carotid artery appeared extremely tortuous, so direct puncture was performed through a skull thinning with a 14-gauge sternal bone marrow aspiration needle. (C) XperGuide progression view. (D) Final position of the needle. (E) XperCT showing the final cast of glue. (F) Complete devascularization was obtained and complete resection was performed 7 days later without any complications.

Case 3

A middle-aged patient was monitored in our institution for von Hippel-Lindau disease with cerebellar and bulbar hemangioblastoma (figure 3). A large endolympatic sac tumor was also diagnosed. Surgery occurred after the patient had undergone a transarterial particulate embolization. However, a massive hemorrhage during surgery meant that only minimal resection could be performed. During the following year, the patient presented with worsening clinical symptoms, including a loss of right-sided hearing, sixth and seventh nerve palsies, and intense increasing headaches. Direct puncture embolization through the previous craniectomy resulted in a very good angiographic result (devascularization >90%). Postoperatively, the headaches improved and the neurological status stabilized; it was therefore decided to observe the patient without surgical removal of the tumor.

Case 4

A pre-teen patient presenting with an endolymphatic sac tumor associated with von Hippel-Lindau disease was referred to our institution for surgery (figure 4). This required preoperative embolization; however, during the session, supraselective catheterization of the distal ascending pharyngeal artery was impossible. In a second session, a retro-auricular craniotomy was created before direct puncture embolization in the angiosuite, which led to an almost complete angiographic devascularization of the tumor. The following day a complete surgical resection was performed through the same craniotomy without any complications.

Figure 4

A pre-teen patient was referred to our institution for surgery on an endolymphatic sac tumor associated with von Hippel-Lindau disease. (A and B) MRI depicting a tumoral lesion with both a cystic part and an enhanced solid component (arrow) requiring preoperative embolization. (C and D) Left common carotid and ascending pharyngeal arteries angiograms demonstrating an intense tumoral blush. (E) Distal supraselective catheterization of the ascending pharyngeal artery appeared impossible. In another session, a retro-auricular craniotomy was created in the operating room before direct puncture embolization in the angiosuite. (F) Trajectory was defined using the XperGuide application (arrow). (G) Test injection through the needle (arrow) confirming the correct positioning of the needle. (H) Preoperative oblique external carotid artery angiogram showing the vascular blush (circle). (I) Final cast of glue. (J) Final external carotid angiogram demonstrating an almost complete angiographic devascularization of the tumor. A complete surgical resection was performed the following day through the same craniotomy without any complications.

Discussion

Transarterial preoperative embolization is typically used for the preoperative devascularization of hypervascular intracranial lesions.6 ,7 In some cases, it can be impossible to perform this procedure safely, owing to excessive vessel tortuosity and small vessel caliber, or because of the potential for normal brain ischemic lesion.

In those cases, the use of direct puncture under 3D fluoroscopic guidance may well result in effective devascularization. In our experience of extra-axial and intracranial lesions, puncture was always possible and resulted in satisfactory outcomes.

For direct puncture, access through the skull can be achieved either via bone erosion, previous craniectomy, or a dedicated burr hole that will be used again during tumor removal.

To obtain effective devacularization, the goal is to obliterate all of the tumor capillary bed, and not just sacrifice the feeding vessel. The advantage of direct puncture over transarterial embolization is that different compartments of the tumor can be punctured to obtain complete filling, which can sometimes be impossible with a single arterial feeder.

Some studies have used Onyx injections for direct punctures, particularly in cases of head and neck tumors.14 We chose to use n-butyl cyanoacrylate, because it results in excellent penetration of the tumor bed without the need to create a large reflux plug, as with Onyx. Its almost immediate polymerization may also be useful in cases of extravasation. Furthermore, the low concentration of glue allows for prolonged injection of relatively large quantities of glue.

The widespread use of this technique has probably been limited by neuroradiologists' fear of causing bleeding complications and damage to normal brain parenchyma during puncture. However, with good collaboration with surgeons to define the access point, a 3D guidance software tool such as XperGuide now allows precise and safe access to any intracranial target point. It is also important to highlight that needles used for direct puncture embolizations are much smaller than stereotaxic biopsy needles.17 We did not encounter any postprocedural bleeding in this series.

As far back as 1994, Casasco et al11 stated that, in the case of craniofacial tumors, “the direct punctures approach [was] initially used to devascularize tumors with difficult or dangerous intravascular access, but in view of the hemodynamic and surgical results obtained, we believe that the indications for this technique can be extended to hypervascular tumors accessible to conventional embolization”. Single or small case series highlight the number of theoretical advantages of direct punctures,10 ,13 ,18 and in view of the excellent devascularization results we obtained, we agree that this technique may play an increasing role in the preoperative management of highly vascularized intracranial lesions.12 Even if this technique has already been described, we report the largest series to date, also highlighting the increased safety brought by the 3D XperGuide. This will also possibly allow access to small and deep lesions such as in case 4.

Although we did not encounter any complications in our series, Elhammady et al14 described a case of inadvertent transtumoral migration of Onyx into the superior sagittal sinus during the embolization of a parasagittal meningioma, which resulted in a venous infarction. In that case, embolization was performed under single-plane fluoroscopy and highlights the need to use biplane fluoroscopy with efficient guiding and blank roadmapping tools.

The major limitation of our study is clearly the small number of cases and we cannot, of course, reach conclusions about the safety of the procedure from a study of four cases. We hope to benefit from more cumulative reported experience to define the place of this exciting approach, which requires expertise from the operator in multiple fields, including vascular anatomy, conventional embolization techniques, and use of 3D image-guidance software.

Conclusion

In selected cases, 3D-guided direct puncture of intracranial tumors appears safe, feasible, and efficient for preoperative embolization.

References

View Abstract

Footnotes

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval Assistance Publique - Hôpitaux de Paris.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement The authors are willing to share spreadsheets from their data extraction on request.

Request Permissions

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.