Article Text

Download PDFPDF

Original research
Middle meningeal artery embolization as an adjuvant treatment to surgery for symptomatic chronic subdural hematoma: a pilot study assessing hematoma volume resorption
  1. Sam Ng1,
  2. Imad Derraz2,
  3. Julien Boetto1,
  4. Cyril Dargazanli2,
  5. Gaëtan Poulen1,
  6. Gregory Gascou2,
  7. Pierre-Henri Lefevre2,
  8. Nicolas Molinari3,
  9. Nicolas Lonjon1,
  10. Vincent Costalat2
  1. 1 Department of Neurosurgery, Montpellier University Medical Center, Montpellier, Occitanie, France
  2. 2 Department of Neuroradiology, Montpellier University Medical Center, Montpellier, Occitanie, France
  3. 3 Department of Statistics, Montpellier University Medical Center, Montpellier, Occitanie, France
  1. Correspondence to Dr Sam Ng, Department of Neurosurgery, Hôpital Gui de Chauliac, Montpellier, France80, France; s-ng{at}


Background Chronic subdural hematoma (CSDH) is a common condition requiring surgical treatment; however, recurrence occurs in 15% of cases at 1 year. Middle meningeal artery (MMA) embolization has recently emerged as a promising treatment to prevent CSDH recurrence.

Objective To investigate the effect of MMA embolization on hematoma volume resorption (HVR) after surgery in symptomatic patients.

Methods From April 2018 to October 2018, participants with CSDH requiring surgery were prospectively randomized in a pilot study, and received either surgical treatment alone (ST group) or surgery and adjuvant MMA embolization (ST+MMAE group). The primary outcome was HVR measured on the 3 month CT scan compared with the immediate pre-embolization CT scan. Secondary outcomes were clinical recurrence of CSDH and safety measures.

Results 46 patients were randomized and 41 of these achieved a 3 month follow-up . Twenty-one patients received MMA embolization. At 3 months, the HVR from postsurgical level was higher in the ST+MMAE group (mean difference 17.5 mL, 95% CI 3.87 to 31.16 mL; p=0.015). Two participants presented a CSDH recurrence (one in each group). One patient died (ST group). No MMA embolization-related adverse events were reported.

Conclusion The addition of MMA embolization to surgery led to an increase in CSDH resorption at 3 months. One recurrence of CSDH was reported in each group, and there were no treatment-related complications.

  • chronic subdural hematoma
  • embolization
  • middle meningeal artery
  • surgery

Statistics from

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.


Chronic subdural hematoma (CSDH) is a one of the most frequent entities requiring neurosurgical treatment. As the aging population continues to grow, CSDH prevalence is increasing, becoming one of the most common cranial neurosurgical conditions and a significant public health problem.1 Furthermore, traditional treatments, including surgical evacuation, have proven disappointing with a high incidence of recurrence at 1 year, averaging approximately 15%.2

Surgical removal is effective in reducing the mass-effect induced by the hematoma,3 4 but it does not treat the underlying pathophysiologic mechanisms. The pathophysiology of CSDH development remains a matter for debate and is not fully understood. Beyond its classic traumatic etiology, the pathophysiology of CSDH involves a self-sustaining process of neoangiogenesis and fibrinolysis, leading to a progressive increase in hematoma volume.5 The resulting neovascularization leads to the formation of fragile capillaries, which may explain the high prevalence of rebleeding phenomena.6–8 CSDH also has an outer membrane, derived from the dura mater, that is currently considered to be a cause of repeated bleeding.9–11 Histological findings have shown that small vessels penetrate from the middle meningeal artery (MMA) through the dura mater and connect to sinusoidal neovessels in an outer membrane of the CSDH. These neovessels are suspected to cause intermittent bleeding and enlargement of the hematoma cavity.9–11

MMA embolization has been proposed as an endovascular approach to decrease the arterial supply to the vascularized membrane.12 Several recent publications describe the technical aspects and the feasibility of this procedure.13–21 Three populations could potentially benefit from MMA embolization: patients with de novo CSDH requiring initial surgical treatment where MMA embolization could be an adjunctive prophylactic treatment; patients who have failed conservative treatment where MMA embolization could be an alternative to surgery; and patients with recurrent CSDH that has already been surgically treated.22 Since post-therapeutic volumes were reported to be predictors of CSDH recurrence,23 hematoma volume resorption (HVR) based on repeated volume measurements may be a pertinent and reproducible method that has not been previously used to investigate a potential effect of MMA embolization. Furthermore, no previous studies have focused exclusively on a cohort of symptomatic de novo CSDH patients requiring surgery.

In this prospective pilot study, we aimed to evaluate whether MMA embolization has an effect on HVR after surgery in symptomatic patients with de novo CSDH.


Study design and participants

This was a single-center, prospective, randomized, pilot study. It was designed as a prerequisite study before conducting a full-scale randomized trial to assess the efficacy of MMA embolization. The major null hypothesis was that no difference would be observed between the HVR at 3 months in the surgery plus embolization group versus the surgery control group. HVR was defined as the difference between the baseline CSDH volume (pre-embolization) and the 3 month CSDH volume. An arbitrary 6 month period of inclusion was decided upon. Institutional review board approval was obtained, and all patients provided written informed consent.

CSDH diagnosis was based on CT scans at admission. Participants were eligible if they were aged 18 years or older and required surgery because of one or more symptoms attributable to CSDH, including headache, cognitive impairment, ataxia, seizure, focal neurologic deficit, or decreased consciousness. Patients were excluded if they fulfilled one of the following criteria: a past history of CSDH surgery; CSDH of 10 mm or less in thickness; mixed densities with an acute component representing more than 50% of the collection; CSDH developed with underlying conditions (including vascular lesions and arachnoid cysts) and poor medical condition with less than 6 months life expectancy. Bilateral CSDHs were included because they were assumed to respond independently to embolization. After completion of the initial surgery, participants were randomly assigned by the on-call neuroradiologist and neurosurgeon at a ratio of 1:1 using a blocked randomization list. All randomized participants underwent a CT scan and received either surgical treatment alone (ST group) or surgery and adjuvant MMA embolization (ST+MMAE group).


Patients receiving antiplatelet or anticoagulation therapy were prepared with proper reversal methods using platelet or antagonist administration. The surgical technique employed was either a twist drill craniostomy or a wide craniotomy depending on the surgeon’s discretion. It has been commonly accepted that wide craniotomies need to be performed if the CSDH has multiple layers and/or mixed densities. In all cases, subdural drains were applied and connected to a continuous closed-system. The removal of the drains was performed in the first 48 hours, before completing the baseline CT scan.

If applicable, embolization was performed after the surgery during the same hospital admission. Biplane fluoroscopy was used, and the procedure was performed under conscious sedation. Systemic heparinization was not used. A 5 French guiding catheter was placed in the proximal external carotid artery and a microcatheter (Headway 21; MicroVention, Inc, Tustin, CA) was navigated by using a guidewire (Synchro-14; Stryker Neurovascular, Fremont, CA). A selective angiography through the microcatheter was performed to select target MMA branches and to detect potentially dangerous collateral vessels. The embolization was performed with polyvinyl alcohol particles (150–250 μm diameter) under a blank fluoroscopic roadmap. When flow stasis of the MMA was confirmed, the procedure was concluded. In cases with dangerous collaterals, the embolization was performed after the microcatheter was advanced more distally or collaterals were occluded with coils. The guide catheter and femoral sheath were then removed and femoral hemostasis obtained with a closure device and manual compression.


The primary outcome was HVR measured on the CT scan performed at 3 months after the pre-embolization CT scan (which was done immediately after surgery). Other outcomes were the percentage of resorption of the hematoma, the requirement for surgery due to CSDH recurrence, and treatment-related complications including all surgical and endovascular complications. Recurrence of CSDH was defined as any radiologically persistent subdural collection with persistent or new symptoms attributable to CSDH. A complication was defined as any adverse event related to the procedure. Patients subsequently underwent a clinical follow-up at 6 weeks and 3 months after surgery.

Data collection

Radiological and clinical data were collected prospectively. CSDHs were classified into four types (homogenous, laminar, separated, trabecular) and three stages (stage 1, homogeneous or laminar type; stage 2, separated type; stage 3, trabecular type) according to the Nakaguchi classification.24

Volume measurements were made on the baseline CT scan and on the 3 month CT scan by two independent blinded observers using Horos software (Nimble Co LLC d/b/a Purview in Annapolis, MD). The region-of-interest based volume was obtained from each CT slice on the workstation. All CT scans were non-contrasted and had a contiguous slice thickness of 1.25 mm. The region-of-interest was delineated using the freehand selection tool on each axial slice containing CSDH. The contoured volume included all of the subdural hematoma space (from inner membrane to the skull), including potential pneumocephalus. Then, the region-of-interest volume was computed in multiple planes by summing the volumes of all regions-of-interest from all selected slices. The thickness measurement was produced on a single axial slice corresponding to the maximum size of the subdural collection. The HVR was then obtained in each group by calculating the difference between the baseline CSDH volume (pre-embolization) and the 3 month CSDH volume.

Statistical analyses

Continuous variables were expressed as mean±SD and categorical variables were expressed as numbers and percentage. Categorical variables were compared using χ2 test, and continuous variables were compared with Student’s t-test or the Mann-Whitney test, according to the normality of the distribution, assessed with the Shapiro-Wilk test. For all statistical analyses, p<0.05 was considered statistically significant. Statistical tests were performed using R 3.5.0 software (R Foundation for Statistical Computing, Vienna, Austria).


Patient inclusion

Fifty-seven participants with radiological CSDH were screened in our institution between April 2018 and October 2018. The flowchart is presented in figure 1. Nine participants were assessed as asymptomatic by the on-call neurosurgeon. Two participants were excluded because of a CSDH that had been already surgically treated or because they had an estimated life expectancy of under 6 months. Forty-eight patients were eligible and gave consent to participate in the study. Twenty-five participants (29 CSDH including bilateral presentations) underwent surgical treatment alone and 21 participants (25 CSDH) underwent surgical treatment and the MMA embolization procedure as an adjuvant treatment. Among the ST+MMAE group, one participant was excluded because MMA embolization was not technically feasible due to the tortuosity of the vessels, and one participant required a surgical rescue procedure 6 weeks after admission. Forty-one participants were present at 3 month follow-up: 22 participants (25 CSDH) in the ST group and 19 participants (22 CSDH) in the ST+MMAE group.

Figure 1

Flowchart of study. CSDH, chronic subdural hematoma; ST, surgical treatment alone; ST+MMAE, surgical treatment and middle meningeal artery embolization.

The patient demographic characteristics and radiological data are summarized in tables 1 and 2.

Table 1

Baseline characteristics of ST+MMAE group and ST group

Table 2

Radiological data comparison between ST+MMAE group and ST group

The primary and secondary outcome results are presented in table 3. At 3 months, the HVR was higher in the ST+MMAE group compared with the ST group (52.6±24.9 mL vs 35.1±21.0 mL, respectively; p=0.015). The difference in mean HVR observed between the two groups was 17.5 mL (95% CI 3.87 to 31.16 mL) in favor of the ST+MMAE group (figure 2).

Figure 2

Resorbed hematoma volume (mL) from postsurgical to 3 months. Data are presented as median, quartiles and extreme values. P<0.05 was considered statistically significant. ST, surgical treatment alone; ST+MMAE, surgical treatment and middle meningeal artery embolization.

Table 3

Outcomes results

The difference in HVR observed between groups can be possibly explained by a greater CSDH volume at baseline in the ST+MMAE group compared with the ST group (respectively, 65.2±27.1 mL vs 51.2±27.4 mL; p=0.14) and a greater CSDH volume in the ST group than in the ST+MMAE group (respectively, 16.2±18.4 mL vs 12.6±18.5 mL; p=0.24).

A surgical rescue procedure was required for one patient out of 21 in the ST+MMAE group due to CSDH recurrence, and similarly for one patient out of 25 in the ST group (in this case further complicated by pneumonia and death). In the ST+MMAE group, seizures occurred in one patient 3 days after the intervention (the patient then improved after antiepileptic drug administration) and one groin hematoma occurred, requiring simple manual compression without any transfusion. Among the ST group, one patient underwent an early acute subdural re-bleed on removal of the subdural drain, without clinical consequences.


Due to its strong pathophysiological rationale, MMA embolization has recently been proposed as an endovascular approach to decrease the arterial supply to the vascularized membrane of CSDHs.12 Results from the present study suggest an increase in HVR at 3 months in a group of patients treated with embolization, emphasizing the possible adjunctive effect of MMA embolization.

These findings are consistent with previous reports that have assessed MMA embolization in CSDH. Ban et al 14 compared 72 patients, enrolled prospectively, who were treated by MMA embolization to a retrospective cohort of patients who underwent conventional treatment, and demonstrated that MMA embolization significantly prevented re-accumulation of CSDH (thickness over 10 mm) and the need for surgical rescue after a 6 month follow-up. Interestingly, only one recurrence in 72 embolized patients was reported in this series, while in this study we observed one recurrence in 19 participants. In contrast to our study, both the conventional treatment group and the embolization group were composed of symptomatic and asymptomatic patients (40.7% and 37.5 %, respectively). In another study, Link et al 15 presented a retrospective case series of 49 patients without a control group. In this study only 10 patients underwent MMA embolization as an adjuvant treatment to surgery. Four patients (8.9%) who required surgical rescue, after a 6 week follow-up, were reported.

Three potential indications for MMA embolization remain to be investigated, namely surgically naïve CSDH, adjunctive postsurgical CSDH, and recurrent postsurgical CSDH.14 15 22 In the present study, patients presenting with recurrence of CSDH after previous surgery were excluded. We also excluded asymptomatic patients since the current literature does not provide sufficient data to justify an additional intervention in this population, especially as some asymptomatic CSDHs spontaneously resorb or resorb with conservative treatments.25 Some trials of MMA embolization as an alternative to surgery for symptomatic patients have been reported,14 15 26 but a detailed description of their symptomatic status is necessary to carefully assess the efficacy of MMA embolization in this specific cohort.

In order to evaluate the potential effect of MMA embolization, the present pilot study was based on repeated volume measurements, which we consider to be a more accurate and reproducible method than the measurement of maximum thickness on a single axial slice. Published studies have evaluated the efficacy of MMA embolization over the avoidance of surgery, in cases of CSDH recurrence, as their primary outcome measure.14 15 However, the definition of significant CSDH recurrence is controversial. Patient age, cognitive status, overall medical condition, radiological evolution (including the occurrence of multiple membranous subdural structures) and consideration of adjuvant medical therapy such as corticosteroids27 or atorvastatin,28 are other parameters generally considered before opting for a second surgery. Consequently, the indications for surgical removal of a recurrent, previously surgically evacuated CSDH remain subjective and may depend on individual or institutional practices.29 30


This study has several limitations. This is a single-institution study with a limited sample of patients; a multicenter study on a larger cohort is anticipated. In addition, a power calculation was not performed to determine the required sample size. Indeed, the current literature does not provide data to predict the effects of MMA embolization on HVR. Furthermore, endovascular navigation in this aged population is not straightforward. The embolization procedure may not be possible in every case due to the tortuosity of the vessels (as we reported for one case in our series) or due to an aberrant origin of the MMA or due to the presence of dangerous collaterals, although we did not encounter such cases in this study.


The addition of MMA embolization to surgical treatment led to an increase in CSDH resorption at 3 months. One recurrence of CSDH was reported in each treatment group, and no treatment-related complications were recorded. These preliminary results suggest that MMA embolization should be considered as a therapeutic option to be further investigated in larger full-scale trials.


We thank members from the Departments of Neurosurgery and Neuroradiology (Montpellier, France) who helped to enroll and manage patients during the study: Kifah Khouri, Nizar Suleiman, Marie-Ange Rodriguez, Marine Le Corre, Emilie Chan-Seng, Jérôme Cochereau, Alice Rolland, Emilie Aloy, William Haynes, Carlos Riquelme.



  • SN and ID contributed equally.

  • Contributors SN and ID contributed equally to this paper. SN, ID and JB: data collection, analysis and interpretation of the data. NM: statistical analysis. SN: paper writing. NL and VC designed the study and supervised the project. SN, JB, GP, NL: surgical management. ID, CD, GC, PHL, VC: neurointerventional management.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

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

  • Data availability statement Data are available from the corresponding author upon reasonable request