Jump to comment:

• Response: Middle meningeal artery embolization in the routine care of chronic subdural hematoma?
  David Fiorella, Adam S Artuhur
  Published on: 12 May 2025

• Published on: 12 May 2025

  Response: Middle meningeal artery embolization in the routine care of chronic subdural hematoma?

  • David Fiorella, Physician SUNY Stony Brook
  • Other Contributors:
    • Adam S Artuhur, Physician

  We thank Dr. Heck for his thoughtful commentary on incorporating middle meningeal artery embolization in the routine care of patients with chronic subdural hematoma. This commentary depends on his analysis of the data from the recently published trials of middle meningeal artery embolization – MAGIC-MT, EMBOLISE and STEM. We are not qualified to provide detailed commentary on the data cited from EMBOLISE and MAGIC-MT but can offer a response with reference to the data from STEM.

  Dr. Heck indicates that MMAE did not result in a significant clinical benefit because the STEM trial failed to show a reduction in death and disability based upon comparative mRS scores at 180 days. There are several aspects of this line of reasoning to consider before coming to this conclusion.
  1. Rescue procedures were a necessary part of the trial
  Participants in clinical trials must be provided with standard of care treatment. These patients all presented with symptomatic subdural fluid collections that produced mass effect. They were followed closely throughout the study with both scheduled and unscheduled serial imaging and clinical assessments. Patients who decompensated, with increased mass effect on imaging or worsened clinical symptoms were often taken for a rescue procedure to prevent cSDH-related death or disability. In fact, two thirds of patients taken for rescue had documented worsening on both imaging and clinical examination. So, although these patients were cap...

  Show More

  We thank Dr. Heck for his thoughtful commentary on incorporating middle meningeal artery embolization in the routine care of patients with chronic subdural hematoma. This commentary depends on his analysis of the data from the recently published trials of middle meningeal artery embolization – MAGIC-MT, EMBOLISE and STEM. We are not qualified to provide detailed commentary on the data cited from EMBOLISE and MAGIC-MT but can offer a response with reference to the data from STEM.

  Dr. Heck indicates that MMAE did not result in a significant clinical benefit because the STEM trial failed to show a reduction in death and disability based upon comparative mRS scores at 180 days. There are several aspects of this line of reasoning to consider before coming to this conclusion.
  1. Rescue procedures were a necessary part of the trial
  Participants in clinical trials must be provided with standard of care treatment. These patients all presented with symptomatic subdural fluid collections that produced mass effect. They were followed closely throughout the study with both scheduled and unscheduled serial imaging and clinical assessments. Patients who decompensated, with increased mass effect on imaging or worsened clinical symptoms were often taken for a rescue procedure to prevent cSDH-related death or disability. In fact, two thirds of patients taken for rescue had documented worsening on both imaging and clinical examination. So, although these patients were captured as “treatment failures” (i.e., primary endpoints), in most cases the rescue procedures successfully averted disability or death. While allowing rescue procedures certainly impaired our ability to detect a difference in clinical outcomes, it was the only ethical way in which to conduct this clinical trial.
  2. mRS <2 is not an appropriate measure of effectiveness in the non-emergent cSDH population
  Although many studies in the neurovascular space have focused upon a reduction in overall disability and death as a primary indication of effectiveness, this is not as applicable in a population of non-emergent cSDH patients. Unlike patients with emergent large vessel occlusion (ELVO) or subarachnoid hemorrhage from a ruptured aneurysm, patients with non-emergent cSDH are usually not disabled on presentation. In STEM, 3 in 4 patients had a mRS score of 2 or less at the time of enrollment. This left only 25% of the entire study population available to demonstrate a potential reduction in disability in a binary analysis (i.e. mRS 0-2 vs. mRS 3-6) performed at 180 days. Powering a randomized controlled trial to detect this small potential effect, while also offering rescue treatment to patients who deteriorated, would not have been practical.
  3. Disability and death in the cSDH cohort is typically not directly related to the cSDH diagnosis
  The cSDH population skews older and has a higher burden of medical co-morbidities relative to other neurovascular study populations. Patients were not excluded from STEM because of advanced age and the study population included nonagenarians. As such, the accumulated disability and death throughout the STEM study was more likely to be attributable to unrelated co-morbidities (e.g. cardiac disease, lung disease, cancer) than the index cSDH. This is most clearly reflected in the deaths which occurred during the study. Only 4 of the deaths observed in STEM through 180 days were neurological, while 17, the vast majority (~80%), were non-neurological. Moreover, none of the non-neurological deaths were adjudicated as “related to the cSDH collection”. If an ELVO or SAH study subject experiences death at 90 or 180 days, it is most likely related to the index neurological event. In the cSDH population, death is much more likely to be from unrelated causes. Thus, in cSDH patients, mRS is a more appropriate measure of safety than of effectiveness. The mRS scores, which were equivalent between arms, demonstrate that there was no added disability or death incurred by adding the adjunctive MMAE procedure to the index management strategy.
  4. Avoidance of index treatment failure was beneficial
  STEM demonstrated that adjunctive MMA embolization with SQUID significantly reduced the failure rates of standard management strategies (either surgical or non-surgical) in patients with cSDH. This reduction in index treatment failure is clinically important for a number of reasons. When looking at patients who failed their index treatment (regardless of treatment allocation), the rates of death and disability at 180 days were nearly double those who did not experience index treatment failure. Second, when assessing health care utilization, the number of hospital days and ICU days related to cSDH management, were significantly higher in patients experiencing treatment failure in comparison to those who did not. This provides some evidence that the reduction in treatment failure achieved with MMAE does provide tangible clinical benefits.

  Dr. Heck argues that a better paradigm might be to avoid any “surgical procedures” and to reserve all intervention for patients who fail initial non-surgical management.

  While there are no published treatment guidelines for cSDH patients, STEM patients were required to have a collection measuring greater than 10 mm and to have neurological symptoms related to the collection. Traditionally, these criteria have largely been considered solid indications for surgery. However, for patients with relatively less severe clinical presentations, and particularly for those who are poor surgical candidates, non-surgical management (e.g., steroids, statins, pro-thrombotic medications and/or the reversal or cessation of anti-thrombotic or anti-platelet medications) with close clinical and radiological follow up represents a viable alternative. When we designed STEM, there was a paucity of high-quality data characterizing the feasibility, success rate and safety of this non-surgical approach, despite it being employed regularly in clinical practice. Moreover, there were no prospective data available to demonstrate the effectiveness or safety of adjunctive MMAE as an option for this population. We learned in STEM that some cSDH patients (~45%) managed without surgical drainage were able to recover successfully and resorb their cSDH collections to a thickness of less than 10 mm without any surgical or endovascular intervention. Unfortunately, we cannot determine prospectively which patients will, and which will not, require intervention.
  Given the advanced age and co-morbidities of these patients, watchful waiting is certainly a reasonable strategy for some patients. During this “watchful waiting period”, if the expectation is that there is more than a 50% chance of failure, pragmatic considerations become important. With a real potential for acute neurological decompensation, are physicians, families and patients comfortable with hospital discharge, with the patient being unobserved during the day, or with the patient resuming normal activities like driving a car? Consider that MMAE substantially reduces the risk (from greater than 50% to less than 20%) of deterioration in patients undergoing non-surgical management. The decision to undergo a safe, minimally invasive procedure that substantially reduces the risks associated with cSDH exacerbation is not a difficult decision for many patients with non-emergent cSDH.
  Moreover, the MMAE treatment paradigm itself and the agents and techniques used are actively evolving. More of these procedures are being performed under conscious sedation, either using IA lidocaine as a precursor to DMSO-based liquid embolic infusions or using commercially available agents (e.g. nBCA) which elicit less pain when injected. Moreover, next-generation, purpose-designed liquid embolic agents are currently under clinical investigation. These agents may both allow injection without causing pain and achieve more reproducible distal penetration into the pathological membranes – which may further increase the effectiveness of MMAE. If novel agents perform well in clinical trials, one can easily envision a treatment paradigm in which most patients with non-emergent cSDH can undergo MMAE as an outpatient procedure under conscious sedation with the expectation that the failure rate of that therapy will be even lower than what we have observed thusfar.
  Finally, Dr. Heck suggests that MMAE in STEM was not necessarily as safe as perceived, indicating that there was “one stroke and one facial nerve paralysis in STEM” after discussing the disabling strokes reported in another trial.

  The careful monitoring of every adverse event is an important part of studying any new therapy. In STEM there were no disabling strokes observed after MMAE in either the surgical or non-surgical management strata (although there was one disabling stroke in the control (non-embolization) arm). The site reported “stroke” in the MMAE arm of STEM occurred in a patient who had a positive diffusion MRI after MMAE in the absence of neurological symptoms. The “facial nerve paralysis” reported occurred days after MMAE and resolved with steroids.

  While the MMAE procedure was very safe in STEM, the absence of embolization at times, was not. In fact, of the four neurological deaths in STEM, three occurred in the surgical control (non-embolization) arm. All three of these control arm neurological deaths were related to acute re-bleeding into the index cSDH side after surgery. This never occurred in any patient who underwent pre-surgical MMAE in STEM. Although these are small numbers, it’s an observation worthy of consideration and future study.

  In summary, any treatment can be over-utilized and MMAE is not an exception to this rule. Clearly, the indications for MMAE, and our understanding of the role of MMAE within the global management of cSDH continue to evolve. The data from three FDA IDE studies: EMBOLISE, STEM and MEMBRANE, have established that the technique is useful as an adjunctive procedure to reduce the failure rates of standard (surgical or non-surgical) management strategies. This was the most conservative and important principle to establish at the outset. The literature includes a large number of non-randomized studies that also support this. Further studies of MMAE as an initial therapy for non-emergent symptomatic cSDH patients requiring (or preferring) interventional management will further elucidate how the procedure fits into cSDH management. Progressive device and technical innovation will almost certainly make the procedure safer, less invasive and more effective.

  We anticipate that emerging data from a combined analysis of the existing three US trials, innovative new clinical trials and the development of new agents and techniques will continue to further expand the role of MMAE in cSDH management.

  Show Less

  Conflict of Interest:
  National PI's STEM Trial, Consultant BALT USA (DF)

  • Back to top