This systematic review follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (8).

An electronic literature search of the PubMed database was performed on 28 May 2023 using the following key terms: chronic subdural hematoma, meningeal artery, and embolization. All articles published from January 1976 to May 2023 were identified. Two reviewers (Y.O. and T.I.) independently screened the articles based on their title and abstract. One reviewer (Y.O.) then reviewed the full text of all relevant articles in detail to further assess the eligibility of the studies.

The following inclusion and exclusion criteria were used for the systematic review.

Data extraction was performed by one reviewer (Y.O.) using a predefined data extraction form. The following data were collected and analyzed: study design, sample population and size, patients’ baseline characteristics, antithrombotic therapy use, management strategy (including surgical and endovascular treatment), endovascular treatment success, outcome, follow-up duration, complications, cSDH recurrence rate, and the need for subsequent surgery.

Several definitions of resolution after MMAE were used among the studies, including complete resolution, near complete resolution (reduction of ≥90%), suboptimal resolution (reduction of ≥50%), and partial reduction, the latter two of which were ambiguous. Among these definitions, there were further differences in measurement methods, with some researchers using the hematoma volume for measurement and others using the hematoma thickness. In this study, we focused on the necessity of rescue surgery and defined resolution as radiological improvement.

Some reports included in this review defined recurrent cSDH as only symptomatic re-accumulation after surgical intervention, while others defined recurrent cSDH as also asymptomatic re-accumulation after surgical intervention. Both were included in recurrent cSDH in this review.

In MMAE for cSDH treatment, there are differences in the purpose of treatment and the characteristics of patients between MMAE as a sole treatment and MMAE as prophylaxis, in which surgery is performed before and after MMAE. In this study, we compared the data between reports in which all patients were treated with MMAE alone and reports in which all patients were treated with MMAE combined with surgery.

MMAE treatment success was defined as the successful embolization of the target vessel. All patients who failed to achieve embolization, including MMAE abort, were considered treatment failures. Complications of treatment were counted after excluding those considered to be complications related to surgery (evacuation), those with unknown details, and those with an unknown relationship to MMAE.

Our search strategy yielded 245 articles. Of these, 202 articles were excluded based on the exclusion criteria of the present review. A flow diagram of this study shown in Figure 1.

Forty-three articles were included in the final analysis, including 2 prospective uncontrolled studies, 1 prospective randomized study, and 40 retrospective studies involving 7 to 530 patients per series.

A total of 2,783 patients underwent MMAE in the selected studies (Table 1). Their mean age was 71.2 years, and 71.1% were male. The available data showed that 90.6% of patients had symptomatic cSDH and that 19.9% had bilateral cSDH. Concurrent antiplatelet or anticoagulant use at the time of cSDH treatment was reported in 50.3% of patients.

In total, 3,027 MMAE procedures were performed on 2,783 patients. The MMAE treatment success rate was 98.8%. In principle, treatment for bilateral cSDH was considered two embolization procedures, but some studies counted it as one treatment. This is clearly indicated in Table 2.

MMAE was performed as the sole treatment in 69.1% of patients and prophylactically after or before surgical evacuation in 30.7%. The details in six (0.2%) patients were not reported.

Among all cases, 23.7% were recurrent cSDH after previous surgical evacuation, and 52.7% were treated as upfront MMAE.

The embolic agents included particles, liquid agents, and coils and were used alone or in combination. Details are shown in Table 1. The mean follow-up duration ranged from 20 days to 5 years.

Following MMAE, the rates of cSDH resolution, recurrence, and surgical rescue at the last follow-up were 83.8%, 6.2%, and 6.4%, respectively.

In most studies, no serious complications occurred after MMAE. The rate of severe complications associated with MMAE was 1.0% (28 of 2,783 patients), including 10 cases of cerebral infarction, 5 cases of visual loss, 4 cases of facial palsy, 2 cases of cerebral hemorrhage, 3 cases of MMA arteriovenous fistula, 1 case of MMA rupture, 1 case of aortic dissection, 1 case of femoral artery occlusion, and 1 case of catheter entrapment. The rate of overall complications associated with MMAE was 2.3% (58 of 2,783 patients), including seizures, headache, renal dysfunction, transient neurological symptoms (diplopia and aphasia), and puncture site hematoma. Two overall treatment-related mortalities due to cerebral hemorrhage and femoral artery occlusion were observed.

The hematoma resolution, recurrence, and retreatment rates in the MMAE-alone treatment group (n = 815) were 86.7%, 6.3%, and 9.6%, respectively, whereas those in the prophylactic MMAE with combined surgery group (n = 370) were 95.6%, 4.4%, and 3.4%, respectively. The percentage of symptomatic patients in the MMAE-alone group and prophylactic MMAE group was 94.0% and 97.7%, respectively. These results are shown in Table 3.

The three initially reported indications for MMAE alone were failure of conservative treatment or asymptomatic or mildly symptomatic patients (to avoid surgery), advanced age and use of antiplatelet or anticoagulant drugs (as an alternative treatment considering the invasiveness of surgery), and prevention of recurrence in patients with recurrent disease after surgical treatment (53).

In an evaluation of the efficacy of MMAE alone, Housley et al. (37) compared the outcomes of 48 propensity-matched patients with cSDH who underwent either surgery alone or MMAE alone as initial treatment. There was a significant hematoma reduction in the surgery group immediately after surgery; after 12 weeks of treatment, however, there was no significant difference in hematoma reduction between the two groups. Furthermore, the recurrence rate was significantly lower in the MMAE group (22.9% vs. 4.2%) (37). Kim (5) compared the outcomes of 23 patients who received conventional treatment and 20 patients who received MMAE among 43 patients who developed recurrent cSDH after surgical treatment. The MMAE group showed better prevention of recurrence and earlier brain re-expansion despite the fact that patients of advanced age were significantly more likely to use antithrombotic drugs in the MMAE group (5). These studies show that the effect of MMAE alone for reducing a hematoma is equivalent to evacuation surgery alone in long-term follow-up, and the effect of preventing recurrence may surpass the effect of evacuation surgery alone.

Some studies showed a good hematoma reduction effect of MMAE alone, even for massive cSDH. Gomez-Paz et al. (28) reported that patients with massive cSDH with a midline shift of ≥5 mm were treated with upfront MMAE and showed good improvement in symptoms and imaging findings under careful follow-up.

In the present review, surgical evacuation was needed in 9.6% of patients in the MMAE-alone group, suggesting that some patients developed hematoma enlargement or neurological deterioration even after MMAE. Therefore, careful follow-up is important, especially in the early postoperative period after patients undergo MMAE alone.

Some reports also suggested that MMAE alone is superior to surgery alone in terms of cost-effectiveness because of the reduced need for additional therapeutic intervention in the treatment of cSDH (54, 55).

In this study, prophylactic MMAE showed a favorable hematoma reduction rate, recurrence rate, and reoperation rate. The major advantage of MMAE is its effect in preventing recurrence. In this review, the reoperation rate was 1.4% to 4.9% in the MMAE with surgical treatment group and 11.6% to 18.8% in the conventional surgical treatment group (9, 15, 32, 33).

Okuma et al. (4) reported the following predictive factors for refractory cSDH after burr-hole surgery: use of antiplatelet drugs or anticoagulants, blood coagulation disorder, hepatic dysfunction, hemodialysis, terminal malignancy, advanced age (>80 years), cerebral atrophy, large preoperative hematoma volume (>150 mL), niveau formation, post cerebrospinal fluid shunt placement, no placement of a drain during surgery, postoperative residual air (>20%), and multiple recurrences. The authors also stated that such predictive factors have the potential to be good indications for prophylactic MMAE.

The selection of appropriate cases with respect to the indication for prophylactic MMAE is important from the standpoint of medical economics. Further investigation of the indications for prophylactic MMAE is warranted.

The rate of MMAE-related complications in this review was 2.3%. Although serious complications such as cerebral infarction, visual loss, and facial palsy rarely occurred, their development suggests the possibility of stray embolic material entering high-risk anastomosis sites. Because of the potential for anastomosis in MMAs that involve the retinal artery and vasa nervorum of the facial nerve, embolization can lead to serious and permanent complications such as blindness and facial nerve palsy (29). During the embolization procedure, the microcatheter should be positioned distal to the branch, and attention should be paid to the findings of reflux during embolic agent injection (44). In addition, in patients with a high risk of migration of embolic material to a compromised anastomosis, it is important to perform a provocation test using lidocaine and abort the embolization based on the result of this test, if necessary (16). Although MMAE is a safe and easy procedure, close attention is needed when working with high-risk anastomoses to prevent complications during MMAE.

With the increase in reports of MMAE for treatment of cSDH, there has also been an increase in reports regarding factors that affect the efficacy of MMAE. Salem et al. (46) evaluated various factors in 530 patients treated with MMAE; they found that an MMA main trunk diameter of <1.5 mm and anticoagulant medication use were factors associated with higher retreatment rates and that the use of liquid embolic material was a factor associated with lower retreatment rates. Some reports have suggested that the use of liquid embolic material is as effective and safe as the use of particles (29, 44, 49).

In addition, distal (midline) penetration of the embolizing material has been cited as a factor that shortens the time to hematoma clearance (31, 41). Among all anticoagulants, only factor Xa inhibitors were reported to be associated with retreatment (38).

This review had three main limitations. First, most of the reports were retrospective, and they contained various indications for MMAE and definitions of resolution. Thus, the statistical examination was limited, leaving potential for important selection bias. Second, the content of the articles was mixed, with some studies limited to MMAE (whether upfront MMAE, prophylactic MMAE, or MMAE for recurrence after surgery), others comparing MMAE with surgery, and still others focusing on the clinical course or radiographic findings. Third, some studies lacked information on the timing of MMAE and surgical treatment and did not clearly indicate whether complications in cases of combined surgery and MMAE were caused by surgery or MMAE. These factors made statistical evaluation difficult in this review. Further prospective randomized trials are required to establish the clear indications for MMAE as an initial treatment or combined treatment with evacuation surgeries.