Interventional procedure overview of middle meningeal artery embolisation for chronic subdural haematomas
Closed for comments This consultation ended on at Request commenting lead permission
Evidence summary
Population and studies description
This interventional procedure overview is based on approximately 66,088 patients with CSDHs from 2 systematic reviews and meta-analyses (Haldrup 2020; Ironside 2021), 3 cohort studies (Dicpinigaitis 2021; Nia 2022; Scoville 2022), 2 case series (Catapano 2021; Khorasanizadeh 2022) and 2 case reports (Piergallini 2019; Wilseck 2022). Of these 66,088 patients, only 1,703 patients had MMAEs, with most patients having primary MMAEs (about 60%), followed by rescue MMAEs (26%) and then prophylactic MMAEs (14%).
This is a rapid review of the literature, and a flow chart of the complete selection process is shown in figure 1. This overview presents 9 studies as the main evidence in table 2 and table 3, and lists other 78 relevant studies in table 5.
Of the 9 studies included in the main evidence, 6 primary studies were done in the US, 1 systematic review and meta-analysis included studies which were done in Japan (n=8), USA (n=7), Korea (n=2), France (n=2), and China (n=1), 2 articles did not report the countries (Haldrup 2020; Piergallini 2019). Four articles reported the outcomes based on number of patients (Haldrup 2020; Dicpinigaitis 2021; Nia 2022; Scoville 2022), 4 articles presented the outcomes using number of treated CSDHs (Catapano 2021; Khorasanizadeh 2022; Piergallini 2019; Wilseck 2022), and 1 systematic review by Ironside (2021) mainly considered number of patients (only 1 included study reported the outcome data using number of cases of CSDHs). Patients were recruited from 2012 to 2021. Five primary studies had follow-up durations of 3 to 6 months, 2 systematic reviews included studies with follow-up periods ranging from 6 weeks to 5 years, and 1 study (Dicpinigaitis 2021) did not report the follow-up duration.
The systematic review and meta-analysis by Ironside (2021) compared MMAE with conventional management (surgical drainage or observation). Another systematic review and meta-analysis by Haldrup (2020) investigated the effect of MMAE for primary and recurrent CSDH. Two cohort studies compared MMAE with surgery (Nia 2022; Dicpingigaitis 2021) and 1 cohort study (Scoville 2022) compared outcomes after particle or liquid MMAE. Two case reports were included because of unique safety outcomes identified (Piergallini 2019; Wilseck 2022). Table 2 presents study details.
Study no. | First author, date country | Patients (male: female) | Age (mean, years) | Study design | Inclusion criteria | Intervention | Follow up |
|---|---|---|---|---|---|---|---|
1 | Ironside 2021 Japan (n=8), USA (n=7), Korea (n=2), France (n=2), and China (n=1). | 1,386 (20 studies: 5 double-arm studies and 15 single-arm studies) (MMAE: 74% male) | MMAE:72.7 | Systematic review and meta-analysis | Studies with 3 or more patients having MMAE for CSDH; postembolisation outcomes data describing CSDH recurrence being reported; and English language. | MMAE using particles, liquid, coils, microspheres, or Onyx (n=688 patients with 714 MMAEs)
Conventional management (n=698); observation with serial neuroimaging or surgical drainage | Mean 1.5 to 26.3 months |
2 | Haldrup 2020 Not reported for individual studies | 191 (18 single-arm studies) | 45 to 91 years (apart from 1 case report of patient being 13 years old) | Systematic review and meta-analysis | Not reported | MMAE using PVA, NBCA or coils for:
| 6 weeks to 5 years |
3 | Nia (2022) US (31 HCOs) | Primary MMAE: 72.0; Adjunct MMAE: 67.9;Rescue MMAE: 69.8; Primary surgery; 62.4. | Cohort study (TriNetX Analytics Network - national registry) | A confirmed diagnosis of nontraumatic CSDH and nontraumatic SDH in patients who had embolisation or surgical procedures for evacuation of haematoma within 1 month of the CSDH diagnosis. | MMAE:
Primary surgery: n=4,050 (94.8%) | 6 months | |
4 | Scoville 2022 US (15 academic centres) | 208 (155:53) | Mean 61.8 to 73.5 years | Cohort study | Presence of NASH and treatment with MMAE. | MMAE using particles or liquid embolisates:
| 90 days |
5 | Dicpinigaitis (2021) US (NIS) | 60,045 (42,115:17,930) | 62.1 | Cohort study | Adult patients with primary admission diagnoses of non-traumatic SDH; treatment modalities including craniotomy, burr hole craniostomy, and MMAE. | MMAE for CSDH: n=390 (0.6%)
Craniotomy or burr hole craniostomy: n=59,655 (99.4%) | Not reported |
6 | Catapano (2021) US | 66 (84 treated CSDHs) (51:15) | 70 | Case series | Not reported | MMAE using liquid (Onyx), particles, coils or NBCA:
| 180 days |
7 | Khorasanizadeh (2022) US | 78 (94 hemispheres treated CSDHs) | 72 | Case series | Diagnosis of CSDH and treatment by MMAE | MMAE:
| Mean 113.9 days |
Piergallini (2019) Not reported | 2 (3 treated CSDHs) (0:2) | 82 | Case report | Not reported | Prophylactic MMAE using PVA at 2 and 4 days after burr hole surgery | 3 months | |
9 | Wilseck (2022) US | 1 (male) (bilateral acute on CSDH) | 62 | Case report | Not reported | Upfront MMAE: bilateral embolisation. | 90 days |
Efficacy outcomes | Safety outcomes | |
|---|---|---|
Ironside (2021) | MMAE compared with conventional management (n=698):
5 double-arm studies (n=902): MMAE (n=204) compared with conventional management (n=698)
Subgroup analysis of upfront (n=43; 4 studies) compared with postoperative MMAE (n=256; 15 studies):
| MMAE compared with conventional management:
5 double-arm studies (n=902): MMAE (n=204) compared with conventional management (n=698)
Subgroup analysis of upfront (n=43) compared with postoperative MMAE (n=256):
|
Haldrup 2020 | Recurrence rate: OR for recurrence - primary compared with recurrent CSDH: 1.7 (95% CI 0.3 to 11.1), p>0.05 | No complications were described (not defined in 6 studies and 0 complication in 12 studies) |
Nia (2022) | Hospital stay (mean, 95% CI):
6-month survival ratio: the difference is 1.25% for the primary MMAE compared with primary surgery (p=0.35) Treatment failure or need for surgical rescue:
| Occurrence of headaches or facial weakness, or mortality between primary MMAE and primary surgery: all p>0.05 |
Scoville 2022 | 50% reduction in maximal haematoma thickness at 90 days - particles compared with liquid embolisates:
Time to reach 50% reduction in haematoma size - particles compared with liquid embolisates:
Retreatments within 90 days - particles compared with liquid embolisates:
| Complications - particles compared with liquid embolisates:
Detailed complications:
|
Dicpinigaitis (2021) | Clinical outcome comparison – MMAE compared with surgical evacuation by craniotomy or burr hole craniostomy:
Propensity score-adjusted comparison – MMAE compared with surgical evacuation by craniotomy or burr hole craniostomy:
| Intracranial complications: MMAE compared with surgical evacuation by craniotomy or burr hole craniostomy
In-hospital mortality: less than 3% in MMAE |
Catapano (2021) | CSDH size (mean [SD], mm):
Complete or near-complete resolution:
Surgical rescue within 30 days after MMAE: 4% (3/84) In the multivariable analysis, only distal embolysate penetration (into the vertex) was independently associated with rapid clearance (OR 3.9, 95% CI 1.4 to 11.1; p=0.01) and resolution of CSDHs at 90 days (OR 5.0, 95% CI 1.7 to 14.6; p=0.003). | Complication: cerebrovascular accident, 1% (1/84) likely caused by instability of the access catheter (via a transfemoral route). |
Khorasanizadeh (2022) | Haematoma axial thickness (mean, mm):
Haematoma volume (mean [SD], ml):
Haematoma axial thickness decreased by at least 50%: coils only, 83.3% (n=10) compared with 72.0% (n=59), p=0.4 Haematoma volume decreased by at least 50%:
Rescue surgery:
mRS stabilisation or improvement:
| Procedure-related complications: n=2
The patient had emergent thrombectomy of the left M2 clot with successful recanalisation and neurologic recovery. |
Piergallini (2019) | Discharge: 1 week after MMA CT and angio-CT scans at 3 months: no recurrence of the SDH or fistula and no signs of intracranial bleeding. | DAVF after MMAE: n=2 Both patients remained completely asymptomatic after the development of the fistula, which was occluded by coiling the MMA at the level of the point of shunt. |
Wilseck (2022) | Discharge: 1 day after MMAE Sequential imaging showed decreasing density and size of the treated subdural collections leading to less mass effect and midline shift. There was no evidence of subsequent acute bleed. The patient's headache improved, and the patient remains neurologically intact. | Iatrogenic MMA pseudoaneurysm related to use of a dual-lumen balloon catheter: n=1 This event was successfully treated with ethylene vinyl alcohol embolisation. |
Procedure technique
The embolisation procedure was not thoroughly described in all studies, but in general, the procedures were done under general or local anaesthesia with light or no sedation. Endovascular access to the MMA was through the radial or femoral artery, and a microcatheter was advanced under roadmap guidance. Target branches were embolised using liquid embolics, particles, coils, or a combination. Particles and liquid embolics were used more often than coils, while coils were likely to be used in combination with particles. If dangerous anastomoses including ophthalmic collaterals were seen, coils were used exclusively.
MMAE was used to treat primary or recurrent CSDHs, and when combining with surgical evacuation, MMAE was also applied to prevent the recurrence of CSDHs. Therefore, MMAE has been done in 3 treatment stages:
upfront or primary embolisation in patients with a previously untreated CSDH so embolisation was used as the first treatment without concomitant surgical intervention.
prophylactic or adjunct embolisation after surgical evacuation without evidence of interval postoperative CSDH recurrence (surgery and embolisation usually within 7 days).
rescue embolisation for recurrent CSDH after previous surgical evacuation (more than 7 days after primary surgery).
Of the 9 studies, 3 studies presented the outcome data for different treatment purposes separately (Ironside 2021; Haldrup 2020; Nia 2022), 1 study focused on different embolic agents in each treatment stage (Scoville 2022), and 3 studies (Dicpinigaitis 2021; Catapano 2021; Khorasanizadeh 2022) reported the outcomes of MMAE used in different treatment stages as a whole. Where possible, the outcomes for MMAE used in specific treatment stage will be described separately.
Efficacy
Haematoma resolution
Complete or near-complete resolution of haematoma was described in the case series by Catapano (2021). This study reported that the rate of complete or near-complete resolution was in 18% (15/84) of treated CSDHs at 30 days, 63% (45/72) at 90 days and 92% (67/73) at 180 days after MMAE. This study also found that complete or near-complete resolution was statistically significantly more at 180 days than that at 90 and 30 days after MMAE.
Haematoma reduction
Data on haematoma size before and after MMAE was presented in 3 studies. Catapano (2021) reported that the mean CSDH size statistically significantly reduced from 16.9 mm (SD 4.8) at baseline to 8.8 mm (SD 4.3) at 30 days, 3.4 mm (SD 3.0) at 90 days, and 1.0 mm (SD 1.7) at 180 days after MMAE. Khorasanizadeh (2022) found that the mean haematoma axial thickness and volume reduced from 16.0 mm and 92.3 ml to 5.7 mm (mean reduction, 59.2%) and 29.6 ml (mean reduction 71.4%) at a mean of 114 days after MMAE. Authors did not find statistically significant differences between MMAE using coil only and MMAE using coils and particles in haematoma axial thickness reduction by at least 50% (83% compared with 72%) and haematoma volume reduction by at least 50% (82% compared with 76%). Similarly, Scoville (2022) reported that the proportions of patients who reached 50% hematoma thickness reduction were not statistically significantly different between particles and liquid embolisates at 90 days after upfront MMAE (62% compared with 61%), rescue MMAE (61% compared with 56%) or prophylactic MMAE (73% compared with 20%).
Haematoma recurrence and surgical rescue
Data for haematoma recurrence and surgical rescue was described in 6 studies. When comparing MMAE with standard management, Ironside (2021) reported that the pooled recurrence rate of CSDH recurrence from 20 studies was 5% (95% CI 3.2 to 6.5%) in the MMAE group and 22% (95% CI 0.6 to 42.4%) in the conventional management group. MMAE was associated with a lower rate of CSDH recurrence compared with conventional management (OR 0.15, 95% CI 0.03 to 0.75, p=0.02, I2=38%; n=639, 4 studies). In terms of surgical rescue, authors found that the pooled rate was 4% (95% CI 2.8 to 5.9%) in the MMAE group and 16% (95% CI 5.95 to 27.0%) in the conventional management group. MMAE was also associated with a lower rate of surgical rescue than conventional management (OR 0.21, 95% CI 0.07 to 0.58, p=0.03, I2=19%; n=902, 5 studies). Nia (2022) reported a lower rate of treatment failure or need for surgical rescue after primary MMAE than primary surgery (OR 0.47, 95% CI 0.25 to 0.90, p=0.02).
In 2 case series, the rate of surgical rescue was from 4% (3/84) within 30 days after MMAE (Catapano 2021) to 8.5% (8/94) after a mean follow up of 33.1 days (Khorasanizadeh 2022).
When considering MMAE used in different treatment stages, Ironside (2021) described that the recurrent rate was 0% after upfront MMAE (n=43, 4 studies) and 3.9% (95% CI 1.4 to 6.4%) after postoperative, adjunctive MMAE (n=256, 15 studies). Authors also reported that the rate of surgical rescue was 0% after upfront MMAE and 2.9% (95% CI 0.8 to 5.0%) after postoperative MMAE. Haldrup (2020) found that the recurrence rate for patients having primary MMAE was 4% (95% CI 1.4 to 11.4%, I2=6%) and for patients having rescue MMAE was 2% (95% CI 0.5 to 11.0%, I2=6%), with OR being 1.7 (95% CI 0.3 to 11.1, p>0.05). Nia (2022) did not report any statistically significant difference in treatment failure or need for surgical rescue between primary, adjunct and rescue MMAEs.
Taking into consideration different embolisates, Scoville (2022) reported that surgical retreatments within 90 days were not statistically significantly different between MMAE using particles and MMAE using liquid embolics in the upfront MMAE (8% [8/100] compared with 0% [0/31]), rescue MMAE (2% [1/41] compared with 6% [1/18]), and prophylactic MMAE (0% [0/11] compared with 0% [0/5]) groups. Khorasanizadeh (2022) found that the rates of rescue surgery were similar between MMAE using coil only (8%, 1/12) and MMAE using coils and particles (9%, 7/82) groups.
Functional outcomes and survival
mRS scores before and after MMAE were reported in 2 studies. Ironside (2021) stated that mRS score of 0 to 2 at the last follow up was presented in 73% (95% CI 46.3 to 99.2%) of patients who had MMAE and 92% (95% CI 10.8% to 100%) of patients who had conservative management. Khorasanizadeh (2022) described that mRS remained stable or improved in 83% of patients at a mean of 114 days after MMAE compared with baseline scores.
The 6-month survival ratio was reported in the study by Nia (2022). Authors found that the survival ratio was estimated to be 1.25% higher in patients who had primary MMAE compared with patients who had primary surgery, but the effect was not statistically significant.
Length of hospital stay
The length of stay in hospital after MMAE compared with surgical evacuation was assessed in 2 studies. Nia (2022) reported that the mean length of hospital stay was statistically significantly shorter in patients who had primary MMAE (7.74 days) than those who had primary surgery (30.20 days). Dicpinigaitis (2021) found that the mean length of hospital stay was statistically significantly longer in patients after MMAE (13.0 days) than those after surgical evacuation (8.3 days). This statistically significant finding was also supported by propensity score-adjusted comparison (13.0 days compared with 7.4 days).
Safety
Overall complications and mortality
The rates of complications were reported in 3 studies. Ironside (2021) found the pooled rate of in-hospital complications was 2% (95% CI 0.8 to 2.6%) after MMAE compared with 5% (95% CI 2.8 to 7.1%) after conventional management. The meta-analysis of 5 double-arm studies showed a lower rate of in-hospital complications after MMAE than conventional management, but the effect was not statistically significant (OR 0.78, 95% CI 0.34 to 1.76, I2=0%; n=902). Haldrup (2020) reported no complications during or after the embolisation procedure (0 complication in 12 studies and not defined in 6 studies). Scoville (2022) considered complications between particles and liquid embolisates, and found no statistically significant differences after upfront (12% compared with 3%), rescue (7% compared with 0%) and prophylactic (0% compared with 0%) MMAEs.
Mortality outcomes were described in 2 studies. Dicpinigaitis (2021) reported that in-hospital mortality was less than 3% in patients who had MMAE. Nia (2022) did not find any statistically significant difference in mortality between the primary MMAE cohort and the primary surgery cohort within 6 months postoperatively.
Neurologic complications
Data for neurologic complications was reported in 4 studies. Dicpinigaitis (2021) found that the rate of haemorrhagic or ischaemic complications was 0% (0/390) after MMAE compared with less than 1% (335/59,655) after surgical evacuation. Scoville (2022) reported neurologic complications within 90 days after MMAE in 9 patients (9/208), including stroke (n=5), worsening headaches (n=2) and new-onset seizures (n=2). Khorasanizadeh (2022) reported global aphasia and right hemiplegia in 1 patient. Nia (2022) did not find any statistically significant difference in the occurrence of headaches between primary MMAE and primary surgery.
Facial droop
Facial droop or weakness was reported in 3 studies. Dicpinigaitis (2021) reported that, of the 390 patients, facial droop was seen in less than 3% of patients after MMAE. Scoville (2022) described 1 case (1/208) of facial droop from Onyx infiltrating the meningeal branches of the skull base and resulting in facial nerve damage. Nia (2022) did not find any statistically significant difference in the occurrence of facial weakness between primary MMAE and primary surgery.
Visual loss
Dicpinigaitis (2021) reported the rate of visual deficit or blindness was 0% (0/390) after MMAE compared with less than 1% (140/59,655) after surgical evacuation. Khorasanizadeh (2022) reported 1 patient (1/78) developed left eye pain, vision loss and dilated nonreactive pupil while recovering from the procedure in the postanaesthetic care unit.
Other procedural complications
Scoville (2022) reported MMA rupture (n=1), external carotid artery spasm (n=1), and uncategorised technical complication (n=1) within 90 days after MMAE. Catapano (2021) found a cerebrovascular accidence (n=1), which was likely caused by instability of the access catheter via a transfemoral route.
Piergallini (2019) presented 2 cases of immediate development of a DAVF after prophylactic MMAE using microparticles. Wilseck (2022) reported a case of development of an iatrogenic MMA pseudoaneurysm during endovascular embolisation using a dual-lumen balloon catheter used for injection of a liquid embolic agent.
Infections
Postsurgical infections that necessitated surgical treatments was found in 2 patients after MMAE (Scoville 2022).
Anecdotal and theoretical adverse events
Expert advice was sought from consultants who have been nominated or ratified by their professional Society or Royal College. They were asked if they knew of any other adverse events for this procedure that they had heard about (anecdotal), which were not reported in the literature. They were also asked if they thought there were other adverse events that might possibly happen, even if they have never happened (theoretical).
They listed the following anecdotal and theoretical adverse events not reported in the literature: cerebral infarction, access site bleeding/complications, contrast reactions, radiation related risk, complications of general anaesthesia in a frail and elderly population.
Four professional expert questionnaires for this procedure were submitted. Find full details of what the professional experts said about the procedure in the specialist advice questionnaires for this procedure.
Validity and generalisability
In the main evidence, 9 articles were included but there was no data relevant to the UK context. The 7 primary studies were retrospective, observational in design. The 3 cohort studies used real-world data from multiple sites or national sample which might increase the generalisability and external validity of the results. But there was missing data (such as radiographic data and neurological assessment outcomes) and lack of procedural details and standardised treatment protocols.
Potential conflicts of interest were declared by authors of 2 papers (Catapano 2021; Scoville 2022) and 1 study reported that the data access was supported by NIH grant TR001439 and Clinical and Translational Science Award (UL1).
Patient selection for MMAE was heterogeneous and varied between studies (such as age, severity of the condition with or without midline shift, comorbidity, previous trauma, preoperative anticoagulant or antiplatelet use, unilateral or bilateral haematomas). Many studies did not report long-term outcomes. The definitions of prophylactic and rescue MMAE might have been inconsistent across different studies. There was variation in treatment regimens in both MMAE and conventional management, and different embolisation agents were used across studies. It was noted that a major disadvantage of liquid embolics is their possible penetration into meningeal arteries in the skull base, which can lead to permanent cranial nerve damage resulting in facial droop and blindness. To avoid such complications, the anatomical anastomoses between the extracranial and intracranial arteries should be avoided (Scoville 2022). It might be worth noting that not all embolic materials are CE marked for the procedure and indication, although some are not excluded for use.
Evidence showed reduction in haematoma, and when comparing with standard management, there were significantly lower rates of CSDHs recurrence and surgical rescue. Although Nia (2022) found the 6-month survival ratio was estimated to be 1.25% higher after primary MAAE than primary surgery, the difference was not statistically significant. Using different embolic materials did not lead to statistically significant differences in haematoma reduction, surgical rescue and overall complications. There was lack of data in haematoma resolution, functional outcomes and quality of life. When comparing MMAE used in different stages, the outcomes in haematoma recurrence and surgical rescue were mixed. It would be beneficial for further studies to be done in a manner to elucidate the appropriate patient selection for primary MMAE, recurrent MMAE or in combination with surgical intervention.
There is no data from RCTs comparing MMAE with standard treatment and a lack of long-term outcomes to determine the efficacy of this endovascular approach as an upfront, adjunctive or rescue treatment for CSDH. However, to date, numerous clinical trials are underway, including RCTs (NCT04742920, NCT04372147, NCT04511572, NCT04402632, NCT04272996, NCT04270955, NCT05327933, NCT05267184, NCT05220826, NCT04410146, NCT04750200) and non-randomised controlled trials (NCT04065113, NCT04500795), and single-arm trials (NCT04923984). These RCTs and non-randomised controlled trials compare MMAE with standard (surgical or conservative) management with sample sizes ranging from 40 to 600 patients. These ongoing trials are located in Canada, China, France, Germany, Netherlands, Spain, Sweden and US. These trials would add further data to the existing literature.
How are you taking part in this consultation?
You will not be able to change how you comment later.
You must be signed in to answer questions