2026 Volume 20 Issue 1 Article ID: ra.2025-0140
Middle meningeal artery embolization (MMAE) for chronic subdural hematoma (CSDH) is increasingly performed to reduce recurrence. Several recent randomized controlled trials, such as EMBOLISE, MAGIC-MT, and STEM, have shown significant interest in endovascular treatment of CSDH. Many studies have demonstrated clinical and radiographic outcomes following MMAE in different patient populations and treatment strategies, including stand-alone or adjunctive MMAE combined with surgical evacuation. Most studies have shown the efficacy of MMAE in reducing hematoma recurrence, reoperation rates, and even several adverse events compared with surgery alone. Moreover, stand-alone MMAE can be an efficacious and minimally invasive treatment option for some patients. However, its benefits on functional or radiographic outcomes vary depending on patient selection and treatment strategy. Furthermore, knowledge of periprocedural radiographic findings associated with hematoma resolution or reaccumulation after MMAE may contribute to a sufficient follow-up period after treatment, as well as to medical economics, and thus has clinical implications. In this review, we focus on the clinical and radiographic outcomes of MMAE for CSDH.
Chronic subdural hematoma (CSDH) is a common medical condition caused by the accumulation of watery blood and its degradation between the arachnoid and dura mater layers due to minor or unnoticed head trauma. Although the natural course of CSDH remains to be elucidated, an increase or attenuation has been observed.1) If progressive CSDH is left untreated, hematoma expansion can result in severe neurological symptoms. Surgical evacuation of the hematoma is the standard treatment for symptomatic CSDH. However, this approach is associated with a relatively high rate of CSDH recurrence and other potential complications, particularly in elderly patients with comorbidities. Consequently, alternative and less invasive treatment options, including embolization of the middle meningeal artery (MMA), have been developed based on the hypothesis that reducing continuous leakage from the MMA to the neomembrane of the CSDH may allow for increased resorption.
This article reviews middle meningeal artery embolization (MMAE) for CSDH, with a particular focus on the clinical and radiographic courses.
Recently, there has been a surge in studies demonstrating the effectiveness of MMAE as a stand-alone or adjunctive therapy in conjunction with surgical procedures. Previous case reports have described MMAE as a salvage therapy for managing recurrent CSDH after multiple surgical evacuations with favorable outcomes.2,3) A series of 20 patients treated with MMAE compared to 23 surgical controls reported significantly lower incidences of recurrence and earlier resolution of hematoma, resulting in statistical significance.4) Link et al.5) reported the first case series of 5 patients with relatively mild symptoms treated with stand-alone MMAE for CSDH. In this case series, all 5 patients were successfully managed with stand-alone MMAE and showed a significant reduction in the size or resolution of CSDH at 3–14 weeks after MMAE.5)
Based on a recent single-center cohort study focusing on the clinical outcomes of MMAE combined with surgery for CSDH, reoperation was required in no more than 7% of patients because of reaccumulation of CSDH during the study period, and almost 80% of patients showed improvement in the modified Rankin Scale (mRS) score for the treatment of primary or recurrent CSDH, with a 6.0% embolization-related complication rate.6) A Japanese nationwide survey demonstrated that hematomas decreased in 77.8% of cases at 30 days after MMAE with surgical evacuation in the patients treated with MMAE plus burr-hole surgery for recurrence.7) A favorable outcome (mRS 0–2) was observed in 71.4% of patients, and symptomatic CSDH recurrence occurred in 8.9%.7) MMAE-related complications were observed in 5.2% of the patients, with a morbidity of 0.9%.7) A German nationwide cohort study also demonstrated that treatment failure occurred in 16.2% of patients during a mean follow-up of 6.5 months after MMAE, including surgical evacuation, and the incidence of symptomatic MMAE-related complications was 2.5%.8)
Several comparative cohort studies have investigated the efficacy of adjunctive MMAE. Most cohort studies, including small-scale randomized controlled trials (RCTs), have revealed that adjunctive MMAE with surgical evacuation is associated with a lower risk of CSDH recurrence or progression resulting in reoperation than surgical treatment alone.9–15) Martinez-Gutierrez et al. demonstrated that adjunctive MMAE, especially for separated CSDH, was highly effective in preventing recurrence and reoperation.16) In terms of functional outcomes, several studies revealed that MMAE with surgical treatment was associated with good functional outcomes or lower mortality compared with surgery alone9,10,14,17); however, other studies did not show the efficacy of adjunctive MMAE.11,13,15) Accordingly, adjunctive MMAE seems to reduce hematoma recurrence or reoperation rates compared to surgery alone, although its influence on functional outcomes remains controversial.
Recently, RCTs have compared the clinical outcomes of adjunctive MMAE with those of surgical evacuation. The EMBOLISE trial demonstrated that MMAE with surgical treatment was associated with a lower incidence of hematoma recurrence or progression resulting in reoperation than surgical treatment alone; however, there was no significant difference in functional deterioration and mortality at 90 days between the 2 groups.18) Another RCT, STEM, in which adjunctive MMAE with the liquid embolic agent Squid was randomly performed for either surgical or nonsurgical treatment, demonstrated that adjunctive MMAE resulted in a significantly lower risk of primary efficacy outcomes (recurrent or persistent CSDH, retreatment or rescue surgery, and major disabling vascular events within 180 days) than standard treatment alone, with a comparable incidence of disabling stroke or death within 30 days.19) As the follow-up period in this study (180 days as the trial endpoint) was 2 times longer than that of the EMBOLISE18) and MAGIC-MT20) (90 days), the higher progression or recurrence rates and adverse events might be explained.19) There was also a Chinese RCT, MAGIC-MT, in which MMAE with liquid embolic material was randomly performed for either surgical or nonsurgical treatment. While this trial demonstrated that symptomatic recurrence and functional outcomes were not significantly different between patients who underwent MMAE and controls at 90 days, a lower incidence of serious adverse events was observed in patients treated with MMAE.20) In this study, the subgroup analysis of the primary outcome revealed the potential effect of adjunctive MMAE in patients without burr-hole drainage and in those with a midline shift (MLS) of less than 10 mm.20) Among these large-scale RCTs, MMAE-related complications occurred in approximately 2% or less.18–20) An overview of the RCTs is presented in Table 1. The EMBOLISE trial was the only trial to examine the role of MMAE as an adjunct to surgery and had a highly significant result in reducing progression or recurrence. Based on these results, adjunctive MMAE could decrease the risk of recurrence, reoperation, and several adverse events, although it does not have an obvious benefit on functional outcomes.
| Trials | EMBOLISE18) | STEM19) | MAGIC-MT20) | |||
|---|---|---|---|---|---|---|
| Intervention | Control | Intervention | Control | Intervention | Control | |
| ST + MMAE (n = 197) |
ST alone (n = 203) |
MMAE with or without ST (n = 149) |
With or without ST (n = 161) |
MMAE with or without ST (n = 360) |
With or without ST (n = 362) |
|
| Inclusion criteria | Symptomatic CSDH, >15 mm thickness or >5 mm midline shift, focal motor deficit | Symptomatic CSDH, >10 mm in thickness | Symptomatic non-acute SDH, pre-mRS <3 | |||
| Primary outcome | Recurrence or progression at 90-day SAEs at 30 days |
Recurrent or residual, reoperation or surgical rescue, major disabling stroke, MI, or death from neurologic causes within 180-day SAEs at 180 days |
Recurrence or progression at 90-day SAEs at 90 days |
|||
| Primary outcome event | 8 (4.1%)† | 23 (11.3%)† | 19 (16%)‡ | 47 (36%)‡ | 24 (6.7%) | 36 (9.9%) |
| Serious adverse events | 32 (16.2%) | 34 (21.7%) | 60 (41.7%) | 71 (42.8%) | 24 (6.7%) | 42 (11.6%) |
| Embolization-related complication | 4 (2.0%) | NA | 3 (0.8%) | |||
| Functional outcome at 90 days (mRS 0–2) | 144 (81.4%) | 154 (83.7%) | NA* | NA* | 335 (93.1%) | 333 (92.0%) |
| Changes in hematoma thickness in 90 days (mm) | −7.15 | −7.24 | NA | NA | −17.7 ± 7.4 | −17.4 ± 6.9 |
| Change in MLS at 90 days (mm) | −3.19 | −3.29 | NA | NA | −9.8 ± 4.6 | −9.7 ± 4.6 |
| Change in hematoma volume at 90 days (mL) | −55.11 | −56.15 | NA | NA | −116.9 ± 39.6 | −115.8 ± 38.8 |
*mRS at 180 days did not significantly differ between the intervention group and the control group, though numeric values are not available.
† p = 0.008. ‡ p = 0.001.
CSDH, chronic subdural hematoma; MI, myocardial infarction; MLS, midline shift; MMAE, middle meningeal artery embolization; mRS, modified Rankin Scale; NA, not available; SAE, serious adverse event; SDH, subdural hematoma; ST, surgical treatment
Cohort studies have assessed the efficacy of stand-alone MMAE in patients with mild symptoms. Josko et al. reported the feasibility of stand-alone MMAE for both primary and recurrent CSDH in mildly symptomatic or asymptomatic patients, with 14.4% treatment failure.21) Lakhani et al.22) demonstrated that stand-alone MMAE had a lower risk of surgical rescue and lower mortality rates than medical treatments, although the difference was not statistically significant. Chen et al.23) also demonstrated that stand-alone MMAE was associated with a lower risk of surgery or death at 180 days than conservative management. These investigators also compared MMAE combined with surgical treatment with stand-alone MMAE and demonstrated that stand-alone MMAE resulted in durable clinical outcomes similar to those of MMAE combined with surgery in patients with moderate-sized CSDH and mild neurological symptoms.24) Moreover, Gajjar et al. and Mesina-Estarrón et al. demonstrated that stand-alone MMAE had a significantly lower risk of complications and adverse medical events without an increased risk of recurrence than surgical treatment.25,26) In the aforementioned STEM trial, while 61% of patients underwent standard surgical treatment, the benefit of MMAE with respect to the primary efficacy outcome seemed to be greater in patients receiving standard nonsurgical treatment than in those receiving standard surgical treatment (37% vs. 9%), suggesting that MMAE would also be efficacious as a primary or stand-alone therapy to reduce the risk of requiring evacuation, although the trial was not designed to compare those strata.19) These results suggest that stand-alone MMAE may be beneficial for CSDH with mild symptoms or in asymptomatic patients, with similar clinical outcomes to those combined with surgical treatment and a lower risk of complications and medical adverse events than surgical treatment alone.
For recurrent CSDH, a German nationwide cohort study demonstrated that treatment failure after MMAE, including a combination with surgical evacuation, was more frequent in recurrent CSDH than in primary CSDH,8) while Josko et al. demonstrated no significant difference in treatment failure between primary and recurrent CSDH after stand-alone MMAE for asymptomatic or mildly symptomatic patients.21) A multicenter RCT conducted in France assessed the efficacy of MMAE by injecting microparticles for CSDH recurrence or primary CSDH at a high risk of recurrence after surgical evacuation. However, in this trial, MMAE did not lead to a significantly lower rate of recurrence or better functional outcomes at 6 months than standard care alone.27) We also failed to demonstrate an add-on effect of MMAE in preventing CSDH recurrence in patients treated with Japanese Kampo Goreisan for recurrent CSDH.28) These reports suggest that MMAE does not reduce the rate of recurrence in patients at high risk of recurrence, whereas stand-alone MMAE for recurrent CSDH with mild symptoms may be feasible. However, further studies are necessary to determine the efficacy of MMAE for recurrent CSDH because the number of patients with recurrent CSDH was much lower than that of patients with primary CSDH.
Cohort studies have assessed the optimal timing for MMAE. Salim et al. demonstrated that preoperative MMAE was associated with reduced odds of repeat surgery compared to postoperative MMAE at 6 months, whereas no significant difference was observed in 6-month all-cause mortality.29) In contrast, Chen et al. demonstrated no significant differences in discharge disposition, in-hospital complications, death, or 180-day clinical outcomes between groups based on the timing of MMAE (before, on the same day, or after surgery).30) However, in this study, MMAE concurrent with surgical treatment was associated with a shorter length of hospital stay and lower costs than in other groups.30)
In summary, adjunctive MMAE for primary symptomatic CSDH could decrease the risk of recurrence, reoperation, and several adverse events, although it does not have an obvious benefit on functional outcomes compared with surgical treatment alone. Furthermore, stand-alone MMAE seems to be efficacious for clinical outcomes similar to those of adjunctive MMAE, whereas evidence for CSDH with clinical deterioration is not sufficient. The effect of MMAE on the clinical outcomes of recurrent CSDH or CSDH with a high risk of recurrence remains elusive.
Hematoma thickness and volume are frequently used in the radiographic assessment of CSDH.31) Multiple techniques for measuring the volume and thickness have been proposed.32,33) Initial evaluation of CSDH volume and thickness was effective in predicting CSDH prognosis after surgical treatment.34) Based on a recent single-center cohort, 93.3% of cases achieved at least a 50% reduction in the maximum width of the hematoma on follow-up CT imaging (mean 113.2 days) after MMAE combined with surgical evacuation.6) Catapano et al. analyzed the radiographic hematoma clearance in patients with CSDH treated with MMAE and surgical evacuation. In this study, the mean hematoma thickness significantly reduced to 8.8 mm at 30 days, 3.4 mm at 90 days, and 1.0 mm at 180 days, whereas the thickness before treatment was 16.9 mm.35) They also showed that 63% of CSDHs following MMAE had complete or near-complete resolution by 90 days and 92% achieved this stage by 180 days, suggesting that a 90-day follow-up may be insufficient to observe complete or near-complete resolution of CSDH following MMAE, particularly compared with surgical evacuation alone.35) Gomes-Paz et al. demonstrated that the time course of both hematoma and MLS showed gradual normalization between 2 and 4 weeks after stand-alone MMAE.36) The average time-to-resolution of MLS was 46 days in patients with less than 5 mm MLS and 51 days in those with 5 mm or more, suggesting that stand-alone MMAE would be beneficial in select patients who are asymptomatic or mildly symptomatic even in the presence of an MLS greater than 5 mm.36)
Several cohort studies have included small-scale RCTs on long-term radiographic outcomes after MMAE. While some of these studies demonstrated that adjunctive MMAE was associated with good radiographic outcomes at the final imaging compared with surgical evacuation alone,9,12,37) others reported that adjunctive MMAE contributed to radiographic outcomes only in the early period after treatment.15,17) Two recent RCTs demonstrated radiographic outcomes after adjunctive MMAE compared with surgical evacuation alone. The EMBOLISE trial demonstrated no significant differences in the changes in hematoma volume, hematoma thickness, or MLS at 90 days between MMAE plus surgery and surgery alone.18) The Chinese RCT MAGIC-MT demonstrated no significant differences in changes in hematoma volume, hematoma thickness, or MLS at 90 days between MMAE plus surgery and surgery alone.20) An overview of the RCTs is presented in Table 1. On the basis of these results, adjunctive MMAE combined with surgical evacuation may not provide evidence of long-term radiographic efficacy.
Several studies have been conducted on stand-alone MMAE. MAGIC-MT demonstrated that MMAE combined with surgical evacuation resulted in faster hematoma resolution than MMAE alone during the early follow-up period, although no significant difference was observed at 6 months.20) Furthermore, Chen et al.23) demonstrated that stand-alone MMAE led to similar radiographic outcomes at follow-up periods >30 days after stand-alone MMAE compared with MMAE with surgical evacuation. Schmolling et al.38) also showed that hematoma thickness on follow-up imaging after MMAE with an injected Squid agent showed a significant decrease in both groups as a stand-alone procedure and as an adjunct to surgery, although there was no significant difference between the groups, even in the early or late periods. Housley et al. compared radiographic outcomes between surgical evacuation-only and stand-alone MMAE.39) In this study, surgical evacuation resulted in better radiographic outcomes in the immediate postoperative and early follow-up periods, although the difference was not significant at more distant follow-up periods with a higher recurrence rate in the surgical evacuation group.39) Based on these results, surgical evacuation would be beneficial only in the early period after treatment, suggesting that stand-alone MMAE might be an efficacious and minimally invasive treatment, except for aggressive CSDH, which requires urgent surgical evacuation.
In summary, adjunctive MMAE may not provide sufficient evidence for long-term radiographic efficacy, whereas stand-alone MMAE may be similarly efficacious for long-term radiographic outcomes as MMAE combined with surgical evacuation.
Several studies have demonstrated the radiographic predictors of radiographic outcomes after MMAE (Table 2). Focusing on the distribution of liquid embolisates, Ma et al.40) demonstrated that angiographic non-opacification of the frontal and parietal branches after stand-alone MMAE was associated with a decreased rate of surgical rescue and a higher rate of hematoma resolution. Ganga et al.41) demonstrated that deeper compartment penetration of the embolisate visualized on follow-up CT was associated with faster resorption of the hematoma, and liquid embolic membrane penetration during angiography was associated with nearly 4 times the incidence of complete resolution of the hematoma. Others have also reported that penetration of a liquid embolic agent into the skull, distal MMA embolization, or contralateral penetration was associated with higher rates of hematoma resolution, although the benefits of the final radiographic results remain to be elucidated.35,42,43)
| Hematoma resolution | Hematoma reaccumulation |
|---|---|
| • Angiographic non-opacification of the frontal and parietal branches after MMAE41) • Deeper compartment penetration of the embolisate on CT after MMAE42) • Liquid embolic penetration to the membrane42) and the skull43) during MMAE • Distal35) or contralateral44) MMA liquid embolic penetration during MMAE • Low hematoma density on CT at 1 week after MMAE49) |
• Membrane presence on CT or MRI before MMAE45) • Homogeneous and separated architectures in hematoma on CT before MMAE46) • Inner membrane enhancement within the hematoma and the presence of a contrast medium–blood fluid–fluid level on CT after MMAE47) • Aggravation of the rainbow sign on arterial spin labeling after MMAE48) • High hematoma density on CT at 1 week after MMAE49) • Hematoma membrane enhancement on DynaCT during MMAE50) • MMA diameter less than 1.5 mm51) |
DynaCT, Siemens Healthineers, Erlangen, Germany; MMA, middle meningeal artery; MMAE, middle meningeal artery embolization
Regarding imaging before MMAE, Wang et al.44) demonstrated that membrane presence on preprocedural CT or MRI was associated with decreased CSDH resolution following MMAE. Golub et al.45) focused on the internal hematoma architecture on CT to assess the predictors of treatment failure after stand-alone MMAE and showed that homogeneous and separated architectures were predictors of MMAE failure and were associated with hematoma resorption. Some studies have focused on the association between hematoma resolution and post-MMAE CT findings. Maurer et al.46) showed that inner membrane enhancement within the hematoma and the presence of contrast medium (blood fluid levels) were associated with the risk of persistent or progressive CSDH at follow-up. Li et al.47) demonstrated that aggravation of the rainbow sign on arterial spin labeling after stand-alone MMAE correlated with recurrence and worse neurological outcomes. Akamatsu et al.48) focused on the time-course density of hematoma on CT and demonstrated that the rapid resolution of hematoma following MMAE was associated with low hematoma density at 1-week post-embolization, suggesting that a 1-month follow-up would be sufficient in cases of low density. However, they also concluded that high hematoma density at 1-week post-embolization was associated with hematoma resolution and risk of recurrence and thus required a 3-month follow-up.48) Nakagawa et al.49) investigated hematoma membrane enhancement using angiography-DynaCT (cone-beam computed tomography; Siemens Healthineers, Erlangen, Germany) in patients with repeatedly recurrent CSDH during MMAE and demonstrated that the interval between CSDH recurrences was correlated with the amount of hematoma membrane enhancement on DynaCT. Consequently, these radiographic predictors may have contributed to a sufficient follow-up period after MMAE. In contrast, Salem et al.50) investigated the predictors of MMAE treatment failure and showed that a small MMA diameter (<1.5 mm) was the only factor independently associated with both hematoma reaccumulation and hematoma resolution.50)
In summary, there have been several reports on the periprocedural radiographic predictors of hematoma resolution and reaccumulation in MMAE. The former is related to angiographic outcomes, including the degree of branch occlusion and penetration of embolisates. The latter includes topographic findings, such as pre-MMAE homogeneous and separated architectures of the CSDH on CT, and post-MMAE findings, such as texture or density of the CSDH on CT and rainbow sign on arterial spin labeling imaging. There were also pre-MMAE angiographic findings related to the reaccumulation of CSDH, such as hematoma membrane enhancement on DynaCT angiography and a small MMA diameter (<1.5 mm). These predictors may contribute to identifying the risk of CSDH recurrence and determining appropriate follow-up periods.
Most patients with CSDH are elderly; therefore, less-invasive treatments with lower recurrence should be considered. Therefore, stand-alone or adjunctive MMAE may be an alternative for treating patients with CSDH. However, since Japan is ahead of the rest of the world with the largest proportion of people aged over 60, currently above 30%, and is set to reach almost 45% by 2050,51) appropriate indications for MMAE should also be discussed from the perspective of health economics. From this perspective, the development of a predictive risk for recurrence and clinical outcomes using artificial intelligence or machine learning would contribute.
MMAE has emerged as a promising treatment option for CSDH. Only adjunctive MMAE combined with surgical evacuation for symptomatic CSDH has sufficient evidence to reduce the risk of hematoma recurrence, reoperation rates, and even several adverse events compared with surgery alone, with a lower incidence of procedural omplications, although its benefit for long-term functional or radiographic outcomes remains controversial. Moreover, stand-alone MMAE could be an efficacious and minimally invasive treatment for asymptomatic or mildly symptomatic patients, although the evidence is insufficient. However, because of the relatively high mortality rate unrelated to CSDH up to 6 months and 1 year after treatment, CSDHs occur as sentinel health events related to a reduction in lifespan.52–54) Therefore, patient selection for MMAE should be based on the patient’s background. While radiographic follow-up would currently be sufficient for 180 days to observe complete resolution of CSDH, the period could be shortened with MMAE by considering several radiographic predictors, including hematoma density and angiographic outcomes.
We thank Editage (https://www.editage.jp/) for assistance with English language editing.
The authors declare no conflicts of interest.
