Introduction

The first ever report of successfully performed middle meningeal artery embolization (MMAE) for the management of chronic subdural hematoma (cSDH) was published by Mandai et al,1 and it described a case of a 59-year- old man with a hemorrhagic tendency due to liver cirrhosis, refractory to aggressive and repeatedly undertaken surgical interventions. Since then, MMAE has been gaining momentum as a promising treatment modality for cSDH, both alternative and adjunct to conservative management: surgical interventions (burr-hole craniostomy [BHC], twist-drill craniostomy [TDC], and craniotomy) or medications (eg, steroids2 or statins).3 Due to underlying pathomechanisms of cSDH formation and evidence of MMA contribution to its vascular supply,4,5 MMAE appears to be a promising technique, especially for reducing cSDH recurrence rates, which are relatively high in all surgical techniques.6 This comprehensive review aimed to provide insights into the current state of knowledge on MMAE and identify potential areas in need of further investigation at the time of rapidly increasing prevalence of cSDH.7-9

Epidemiology and pathogenesis of chronic subdural hematoma

With the exception of the COVID-19 pandemic period, which forced restrictions on social activities and thus, lowered the chances for head trauma and post-traumatic cSDH formation,10 the incidence of cSDH has been growing over the past decades, specifically among individuals aged 60 years and older.7 With reported cSDH incidence rates ranging from 1.72 to 20.6 per 100 000 persons per year in the general population,11 they nearly tripled among the population aged 80 years and older during the years 1990–2015, reaching the substantial 129.5 per 100 000 persons per year.8 This trend is likely to continue growing, given the aging world population and increasing life expectancy. By 2030, cSDH is expected to become the most common cranial neurosurgical disease in adults, surpassing brain tumors.9

The formation of cSDH is primarily thought to occur via 2 distinct pathways following a minor head trauma: a transformation from an initial acute SDH (aSDH) and, more commonly, de novo cSDH outset, often with baseline subdural hygroma.12 cSDHs of a nontraumatic origin have also been observed secondary to intracranial hypotension13,14 and rupture of an arachnoid cyst.15,16

The trigger for cSDH formation is splitting of the dural border cell layer, which initiates reparative and inflammatory responses engaging a variety of locally released cytokines, recruited inflammatory cells, and pro-angiogenic factors.17,18 Within the first 24 hours, the resultant proliferative tissue begins to form 2 neomembranes: the thinner internal and the thicker external one.19 The outer membrane houses abundant, excessively permeable, and prone-to-rupture capillaries, which permit continuous exudation of vascular contents into the subdural space.20 As the hematoma progresses, the leakage begins to outpace resorption and coagulation, eventually leading to a symptomatic mass effect. The expansion of cSDH is driven by both hemodynamics and biochemical contribution of nutrients delivered to the formed mass.21

Traditional chronic subdural hematoma management

Surgical evacuation and conservative treatment are the conventional methods of cSDH management. The main goal of surgery is a rapid alleviation of the mass effect and neurological symptoms. The principal surgical approaches are: BHC with drainage, TDC, and craniotomy. BHC, considered the first-line treatment with the best cure-to-complication ratio,22 typically involves drilling 2 burr holes 5–8 cm apart, with subsequent durotomy and irrigation of the cSDH with saline. This is followed by placement of a silicone drain, connected to a closed drainage system. TDC, which generally resembles BHC, involves performing small craniostomy using a handheld drill and implementing a closed-system drain. Its primary advantage is that it can be carried out under local anesthesia, even at the bedside, which may be beneficial for high-risk surgical patients.17 Craniotomy, the most invasive out of the surgical options, is usually reserved for cSDH cases with extensive membrane formation and multiple recurrences.22 Although it may reportedly reduce the risk of cSDH recurrence, concomitant membranectomy is a matter of controversy.17,23 In current practice, large craniotomies are being replaced by the so-called mini-craniotomies, with a level of invasiveness corresponding to that of BHC, but improved visualization of the subdural space.24 In addition, endoscope-assisted cSDH evacuation has emerged as an alternative technique for complex cases with thick vascular membranes, septations, and solid clots.25,26

All of the surgical modalities for cSDH management are associated with a substantial risk of periprocedural complications, mortality, and cSDH recurrence. Studies report outcomes as follows: complication rates, 0%–25%, 0%–18%, and 15%–25%; mortality rates, 0%–13%, 5%–8% and 0%–17%; recurrence rates, 2%–31%, 3%–33%, and 0%–28% for BHC, TDC, and craniotomy, respectively.6

As for the conservative medical management alternative to surgery, the role of steroids is the most extensively studied so far, but the evidence of their impact is not sufficient to draw firm conclusions. A study from 2009 showed favorable outcomes in a group of 101 mildly symptomatic patients (Markwalder Grading Score 1–2) treated exclusively with dexamethasone—over two-thirds of them were managed successfully, without further interventions.2 A Chinese randomized clinical trial (RCT) comprising 200 nonsurgically treated cSDH patients diagnosed between February 2014 and November 2015 showed significantly higher resolution of hematoma volume and clinical improvement in the individuals receiving atorvastatin in comparison with the placebo group.3 Two cases of bevacizumab use have been described, both of which resulted in total27 or near total resolution of cSDH.28 There are also indications of positive response to tranexamic acid (TXA)29,30 and etizolam.31 Adding corticosteroids or TXA as postoperative adjuncts might be associated with lower cSDH recurrence rates.32,33

Contribution of the middle meningeal artery embolization to chronic subdural hematoma formation and relevant aspects of its anatomy

Branches of the MMA are known to provide major neovascular supply to cSDHs. Comparison of MMAs between hemispheres with and without cSDH led to a conclusion that diameters of MMAs on the cSDH side are considerably larger.4 The MMA diameter of less than 1.5 mm has also been identified as an independent predictor of failure of MMAE as a treatment modality for cSDH.5 Given that, the underlying mechanism of MMAE offers a potential to strike the right balance between surgical and conservative management. Namely, MMAE addresses both the hemodynamic and biochemical pathways of cSDH formation. Unlike BHC, it does not decrease the extrinsic intracranial tamponade and simultaneously prevents shifting of the cSDH growth phenotype into a primarily biochemically-driven one, thus allowing for progressive involution.21

Understanding the anatomical variations of MMA branches is crucial for avoiding serious complications of MMAE. Unless necessary, it is advised not to perform proximal embolizations near the foramen spinosum, where the MMA enters the skull base, for the sake of petrosal and cavernous branches—usually well-visualized on frontal digital subtraction angiography views.34 The petrosal branch of the MMA, along with the stylomastoid branch of the posterior auricular artery, provides major vascular supply to the geniculate ganglion of the facial nerve. The cavernous branch, anastomosing with the posterior branch of the inferolateral trunk, makes a possible source of supply to the Meckel cave and the trigeminal nerve.35 Also of critical importance, is the identification of MMA-orbital anastomoses, including: the meningo-lacrimal branch (the more common variant [approx. 5%], supplying the lateral orbit), emerging through the connection of the sphenoid branch of the MMA with the lacrimal artery, and the meningo-ophthalmic branch (the rarer variant [<⁠1%], supplying the entire orbit), a collateral between the sphenoid branch and the ophthalmic artery.34 The MMA itself can also originate from the ophthalmic artery. This variation, often a determinant of disqualification from MMAE, can be observed up to 3.42% of cSDH patients.36 Embolic agent in MMA after embolization is visualized in Figure 1.

Figure 1. Branches of the right middle meningeal artery during embolization

Current use of middle meningeal artery embolization in clinical practice

MMAE can be used both as a pre- or postoperative adjunct to surgical drainage, as well as an alternative treatment for certain groups of cSDH patients. A multidisciplinary consensus–based statement published in November 2024 recommends MMAE as an adjunct to surgery in all recurrent cSDHs and as a standalone treatment in both recurrent and de novo cSDH in patients requiring intervention but deemed unsuitable for surgery due to coagulopathies or continuous use of antithrombotics which cannot be suspended.37 In a systematic review of literature, Di Cristofori et al38 identified 2 more indications for MMAE: adjunctive (as prophylaxis of cSDH recurrence after initial surgery) and standalone (attempted at preventing surgery in patients with paucisymptomatic cSDHs).

So far, the most optimal timing for MMAE after surgical intervention has remained unclear. A small study of 17 recurrent cSDH cases treated with a combination of surgery and subsequent MMAE showed a correlation of a short interval between initial surgical treatment and MMAE with a higher risk of another post-MMAE recurrence.39 A multicenter analysis of 237 patients did not report any significant differences in terms of reoperation and procedural success rates between the groups undergoing MMAE either within 2 days or 3–7 days after surgery.40

Procedural efficacy of middle meningeal artery embolization in comparison with other treatment modalities

When comparing MMAE with solely medical management of cSDH, reports of propensity-matched analyses regarding surgical rescue rates vary. Chen et al41 found no differences between the 2 groups derived from nontraumatic cSDH patients. Lakhani et al42 reported lower need for surgical rescue (odds ratio [OR], 0.472; 95% CI, 0.235–0.946; P = 0.03) in the MMAE cohort. Catapano et al43 concluded that conservative management alone was predictive of treatment failure defined as surgical rescue, residual cSDH, or reaccumulation above 10 mm (OR, 13; 95% CI, 1.7–99; P = 0.01), as well as incomplete cSDH resolution. MMAE, on the other hand, was correlated with greater extent of cSDH volume reduction and lower all-cause mortality at 300-day follow-up (hazard ratio, 0.55; 95% CI, 0.35–0.87; P = 0.01).41 Full resolution of chronic subdural hematoma after MMA embolization is visualized in Figure 2.

Figure 2. A – computed tomography of the brain of a 73-year-old asymptomatic man with a history of head trauma, showing a chronic subdural hematoma (cSDH) 3–4 mm in diameter over the right hemisphere; B – partial hemolyzation and expansion of the cSDH up to 12 mm in diameter after 3 months of conventional management; C – complete cSDH resolution after 4 months of treatment exclusively with right-sided middle meningeal artery embolization

Patients treated exclusively with MMAE had lower cSDH recurrence rates in comparison with the ones undergoing surgery alone (4.2% vs 22.9%; P = 0.01).44,45 As may have been expected, Housley et al44 observed significantly greater mean cSDH volume reduction in the surgically managed group in the direct postprocedural period; however, this difference was not found at distant follow-up. Catapano et al43 found that, similar to conservative management, surgery alone was predictive of treatment failure and incomplete cSDH resolution.

In the comparison of groups of patients treated with surgery with adjunctive MMAE vs surgery alone, the conclusions are as follows: adjunctive MMAE reduced cSDH recurrence (7.7% vs 30.8%; P = 0.04) and reintervention (3.8% vs 23.1%; P = 0.049)46 or 30-day readmission rates (4.2% vs 8%; P <⁠0.01),47 and enhanced hematoma resolution at follow-up.48 Mortality rates of both approaches were either similar47,49 or lower in individuals receiving combined management.50 Results on procedural complication rates were unclear.47,49,51

A few studies also compared surgery with adjunctive MMAE and standalone MMAE.52,53 There were no differences between surgical rescue and radiographic failure rates between the 2 groups in a study conducted by Chen et al.52 Gajjar et al53 reported markedly higher functional improvement in the modified Rankin Scale in patients undergoing combined management (surgery with complementary MMAE), along with comparable technical success and procedural complication rates.

The role of MMAE in cSDH management has also been the area of interest in a number of RCTs. Table 1 summarizes the details and main findings of those that have already been completed. Related RCTs currently recruiting are: MEMBRANE (Middle Meningeal Artery Embolization After Burr Hole Drainage for Subdural Hematoma, NCT05327933), CHESS (Chronic Subdural Hematoma Embolization versus Surgery Study, NCT06347796), EMMA-Can (Embolization of the Middle Meningeal Artery in Canada, NCT04923984), LEADH (Less Invasive Embolization Approach for the Treatment of Chronic Subdural Hematoma, NCT05374681), SWEMMA (Swedish Middle Meningeal Artery Embolization Trial, NCT05267184), and STORMM (Subdural Treatment with or without Middle Meningeal Artery Embolization, NCT06163547). OTEMACS (Onyx Trial for the Embolization of the Middle Meningeal Artery for Chronic Subdural Hematoma, NCT04742920) has currently been terminated.

Table 1. Completed clinical trials for middle meningeal artery embolization in chronic subdural hematoma patients

Study (period)

Number of patients

Control arm

Treatment arm

Primary outcomes

Key findings (treatment vs control groups)

EMBOLISE (2020–2025)54

400

Surgical evacuation

Surgical evacuation, with adjunctive Onyx MMAE

  • Recurrence or progression requiring surgery;
  • Poor clinical outcome;
  • Clinical deterioration at 90 days

RR, 0.36 (95% CI,

0.11–0.8; P = 0.008)

MAGIC-MT (2021–2024)55

722

Burr-hole drainage or medical management

Burr-hole drainage or medical management, with adjunctive Onyx MMAE

  • Symptomatic recurrence or progression within 90 days

Between-group difference, −3.3% (95% CI, −7.4 to 0.8; P = 0.1)

STEM

(2020–2024)56

310

Surgical evacuation or medical management

Surgical evacuation or medical management, with adjunctive Squid MMAE

  • Recurrence at 180 days;
  • Reoperation or surgical rescue within 180 days;
  • New, major disabling stroke, myocardial infarction or death from any neurological cause within 30 days

OR, 0.36 (95% CI, 0.2−0.66; P = 0.001)

EMPROTECT (2020–2023)57

342

Medical management

Burr-hole drainage or medical management, with adjunctive Squid MMAE

  • Recurrence within 6 months

OR, 0.64 (95% CI, 0.36−1.14; P = 0.13)

Abbreviations: EMBOLISE, Embolization of the Middle Meningeal Artery with Onyx Liquid Embolic System in the Treatment of Subacute and Chronic Subdural Hematoma; EMPROTECT, Embolization of the Middle Meningeal Artery for the Treatment of Chronic Subdural Hematoma: A Prospective, Multicenter, Observational Registry; MAGIC-MT, Middle meningeal Artery embolization with Glubran 2 versus burr hole trephination In the treatment of Chronic subdural hematoma: a Multicenter randomized controlled Trial; MMAE, middle meningeal artery embolization; OD, odds ratio; RR, relative risk; STEM; Squid Trial for the Embolization of the Middle Meningeal Artery for the Treatment of Chronic Subdural Hematoma

Embolic agents for middle meningeal artery embolization

Considerable variability regarding the choice of embolic agents and endovascular techniques employed for MMAE is observed among institutions. The primarily used permanent embolic agents are: liquid embolic agents (ethylene vinyl alcohol copolymer Onyx (Medtronic, Minneapolis, Minnesota, United States), Squid (Balt USA LLC, Branford, Connecticut, United States) a low-viscosity equivalent of Onyx), n-butyl cyanoacrylate (nBCA), polyvinyl alcohol (PVA) particles, tris-acryl gelatin microspheres (TAGMs), and coils. Onyx is a polymer comprised of ethylene-vinyl alcohol copolymer with dimethyl sulfoxide (DMSO) as a solvent and tantalum powder added for opacity. While MMAE is generally feasible under local anesthesia or conscious sedation, the toxic DMSO content necessitates general anesthesia. From a technical perspective, the primary merit of liquid agents is enabling embolization of distal MMA branches, which reduces the probability of developing collateral flow supplying cSDH and potentially providing access to branches of the contralateral MMA. As they are nonadhesive, Onyx and Squid also allow for longer injection times and temporary pauses midprocedure, if needed. Unlike liquid embolic agents, PVA particles tend to aggregate within proximal vessels, limiting the degree of distal penetration. Additionally, PVA itself lacks opacity, which may impede intraprocedural assessment of its distribution and identification of potential reflux into eloquent branches. On the other hand, contrary to PVA particles of equal size, TAGMs may penetrate to smaller branches, as they are unlikely to aggregate. Their porcine gelatin content might have allergic potential. The coils, either steel or platinum, are also targeted for proximal embolization and cause vessel occlusion through thrombosis. Therefore, any coagulopathies may preclude complete embolization. Possible adverse effects of coil deployment include coil migration and vessel rupture. However, unlike liquid agents and particles, coils do not carry the risk of ischemic complications to the cranial nerves or retina. Numerous techniques of deployment and various coil types are used.58-60

Comparative data on MMAE outcomes with the use of different embolic agents are limited. In a propensity score–matched analysis comprising 1070 MMAE cases, Salem et al61 found no significant differences in radiographic success or surgical rescue and complication rates between the groups of patients undergoing MMAE for cSDH with particles, Onyx, and nBCA. Shehabeldin et al62 reported greater hematoma resolution at late follow-up in the Onyx group, as compared with PVA. The patients undergoing embolization with TAGMs reportedly experienced postprocedural headaches more often than those in the PVA MMAE cohort.63 Results of recent meta-analyses seem to slightly favor liquid embolic agents over particles and coils, demonstrating reduced all-cause mortality for nBCA,64 lowest rates of cSDH recurrence and complications for Onyx, as well as a reduced risk of reoperation for both.65,66 Except for that, the reported results for MMAE safety and efficacy are comparable, regardless of the embolic agents used.61-68 There is a need for prospective studies on this topic, but considering the aforementioned findings, they are likely to produce similar results.

Vascular access for middle meningeal artery embolization

Even though the choice of vascular access for MMAE still remains at a physician’s discretion, and preferences vary between institutions, transradial access (TRA) appears to be a safe and efficient alternative to the well-established transfemoral access (TFA; EMBOLISE [Embolization of the Middle Meningeal Artery with Onyx Liquid Embolic System in the Treatment of Subacute and Chronic Subdural Hematoma]). Most studies show superiority or at least noninferiority of TRA over TFA in certain aspects. Hung et al69 showed that in unilateral MMAE, TRA had significantly shorter procedural duration than TFA (median, 116.6 vs 154.9 min) and fewer site complications—all of them occurred in the TFA subgroup patients. Similarly, Yamamoto et al70 reported shorter procedural duration for bilateral TRA MMAE in comparison with TFA (median, 151 vs 174 min), also noting lower incidence of postoperative delirium in the elderly patients assigned for TRA. However, on a larger sample size, Salem et al71 presented slightly longer procedural duration for TRA (median 68.5 vs 59 min), as well as lower radiographic success rates (77.4% vs 87.3%). Surgical rescue rates, access-related and overall complication rates, favorable functional outcomes at the last follow-up, length of stay, and technical success rates were comparable. Besides that, 2 cases of a novel superficial temporal artery access for MMAE in patients with arterial changes precluding both TRA and TFA,72,73 as well as 1 case of a salvage maneuver in the form of direct carotid bulb access,74 have been recently described, though they are yet to be compared with the other 2 approaches.

Technical considerations for middle meningeal artery embolization

No consensus on the solely technical nuances of MMAE has been reached yet. Research indicates that: 1) embolization of both anterior and posterior MMA branches is associated with higher rates of complete cSDH resolution than single-branch embolization (76% vs 33%)75; 2) penetration of the embolic agent into distal MMA segments is independently associated with rapid clearance (OR, 3.9; 95% CI, 1.4–11.1; P = 0.01) and resolution of cSDHs at 90 days (OR, 5; 95% CI, 1.7–14.6; P = 0.003)76; and 3) penetration of nBCA into the falx (the “bright falx” sign) is associated with faster volumetric improvement of cSDH.77 A number of solutions have been proposed over the past few years, attempting to refine the procedure. The “sugar rush” technique uses concomitant injection of diluted nBCA and 5% dextrose in water to allow deeper penetration of the embolic agent. In a study group of 61 patients, Majidi et al78 reported an obliteration rate to the distal MMA branches of 100%, with no incidence of permanent cranial nerve palsy or ischemic complications. The Single Pedicle Embolization of the Distal Middle Meningeal Artery using nBCA technique involves distal penetration of the MMA using diluted nBCA for delayed polymerization and a single catheter, leveraging natural MMA anastomoses and ensuring minimal risk of nontarget embolization, as well as shorter procedural duration.79 Helical coils can be deployed into the MMA following initial PVA administration to further augment the blood flow and enable repeated injection of the embolizate, allowing more thorough embolization.80 Adjunctive proximal coil placement has been proposed as a method minimizing the risk of nontarget embolization of proximal MMA branches and preventing its recanalization.81 Different types of catheters and the feasibility of their use for MMAE are also being tested.82,83

Safety profile and procedural complications

Current literature suggests MMAE is a generally safe procedure. A recent meta-analysis comprising 921 patients undergoing MMAE reported 35 procedure-related adverse events.84 One of the rare but possible procedural complications is facial nerve palsy, most likely caused by inadvertent embolization of the vasa nervorum of the facial nerve. Among the 7 cases reported to date, the onset of symptoms was observed within a time frame ranging from direct postoprocedural period to up to 7 days after the procedure. In 4 cases, partial resolution of the symptoms was observed over time. One of the patients showed no improvement throughout follow-up. In the 3 cases identified in our institution, the exact mechanism of the complication remains unclear—no reflux to the petrosal branch of the MMA, which contributes to the vascular supply of the facial nerve, was observed on control angiographies.85

Another possible complication of MMAE is vision loss due to occlusion of the ophthalmic artery. Jauregui et al86 reported a case of a 60-year-old man who suffered such a complication, with suggestive findings on fundoscopy. A diffuse disc, along with retinal pallor and attenuation of retinal vasculature indicated ischemia, while the retinal pigmentary plaques were compatible with damage to the outer retina.

Unintended migration of the embolic agent to the frontal and parietal branches of the superficial temporal artery may lead to cutaneous necrosis of the scalp. In a patient with sudden onset of alopecia 5 days after MMAE, histopathologic examination showed presence of microspheres in deep dermal vessels.87

Considering neurological complications, the literature acknowledges 3 cases of cerebrovascular accidents, including small stroke, caused by embolization of a cortical vein via an initially unnoticed microscopic fistulous connection from the MMA, and thromboembolic stroke.83,88,89 In a case series by Khorasanizadeh et al,67 1 patient developed postprocedural global aphasia and hemiplegia attributed to the occlusion of the middle cerebral artery anterior division. Emergency thrombectomy allowed for a neurologic recovery of the patient.

Catheters used for MMAE may cause injuries to vessel walls and potentially lead to formation of iatrogenic arteriovenous fistulas77 or small hemorrhages. Another potential catheter-related complication is its inadvertent retention inside the vessel.90 In order to remove it, a stent retriever with continuous aspiration can be used.91

Similarly to other endovascular interventions, MMAE carries a risk of access-site complications. These may include a groin hematoma,78 common femoral artery pseudoaneurysm,77 or femoral artery occlusion requiring bypass surgery92 in the case of TFA, whereas for TRA—radial artery occlusion, avulsion, extravasation, wrist hematoma and compartment syndrome.93

Conclusions

MMAE appears to be a fairly safe and efficient, minimally-invasive treatment modality for cSDH. It may serve as a good alternative for patients unsuitable for surgery and those experiencing cSDH recurrences. There is a variety of embolic agents and feasible procedural techniques for MMAE, with limited comparative data. In the light of that, and due to the heterogeneity of patient inclusion criteria among the existing studies, there is a need for further evaluation and standardization of both technical aspects of MMAE and its outcomes.