There is increasing evidence supporting the safety and effectiveness of middle meningeal artery embolization (MMAE) as an adjunct to standard of care for chronic subdural hematomas (cSDH), much of it featured in the pages of JNIS.1–5 At present, there are no fewer than 15 randomized controlled trials (RCTs) in North America and Europe studying MMAE either as a standalone therapy or an adjunct to surgery or medical therapy. While there is considerable enthusiasm for this procedure,4 there are some unique considerations regarding cSDH and MMAE that should be considered going forward.

cSDH behave differently from other disease processes we treat

Unlike many other neurosurgical diseases, cSDH often develop slowly and insidiously weeks after an injury, yet may resolve spontaneously on their own.6–8 Surprisingly, the natural history of cSDH remains poorly defined,9 however neurosurgeons often see these collections shrink spontaneously in practice. We now understand that growth occurs largely secondary to angiogenesis with ongoing oozing from neocapillary beds within cSDH membranes, while resorption largely occurs from slow breakdown of subdural contents via macrophages.10 The formation and/or resolution of cSDH can be philosophically simplified to an equation, with some factors promoting growth and other factors promoting resolution (table 1). The lifecycle of a common cSDH begins with growth factors outpacing resorption factors promoting growth. As the cSDH expands and fills the capacious subdural space, it achieves a size where it exerts mass effect on the brain, increasing intracranial pressure. Now the brain exerts tamponade on the cSDH, restricting further growth, and swinging the pendulum to a homeostasis where growth and resorption have equilibrated. Ongoing resorption factors may begin to now outpace growth factors, the cSDH begins to shrink, and eventually the cSDH resolves spontaneously without intervention. In other patients, growth factors persistently outpace resorption and neurological symptoms occur, necessitating surgical intervention. Studies reporting on progression of cSDH based on these factors and imaging appearance are relatively limited; however, there is evidence that cSDH with acute components seem to be the highest risk for ongoing growth.11

Table 1

The growth or resorption of cSDH can be thought of as a hypothetical equation, with factors promoting growth competing with factors promoting resorption. If growth factors outpace resorption factors, the cSDH will expand; if resorption factors outpace growth factors, shrinkage will occur

Surgical interventions and the equation

Traditional surgical treatments for cSDH are standard-of-care interventions that remove mass effect from the brain and reverse neurological deficits. Burr hole or bedside twist drill evacuation procedures remove cSDH contents and allow for brain re-expansion to swing the equation towards resorption; craniotomy has the added benefits of neomembrane removal and middle meningeal artery (MMA) sacrifice, which dramatically shifts the equation towards resorption. Hypothetically, the less invasive the surgical procedure, the higher the risk of cSDH re-accumulation12 given the less prominent effect on this equation. There is a clear indication for these procedures when there is neurological deficit, and occasionally an indication with large collections in the absence of clear deficit but presence of other symptoms (severe headache, gait imbalance).

MMAE has consistently shown promise in reducing the need for future surgical intervention.1 2 13 Using our growth/resorption model, MMAE achieves this effect by eliminating neomembrane oozing by effectively stripping the cSDH blood supply, resulting in a shift towards resorption. While a number of different authors have suggested that neomembrane penetration of liquid embolic and particles is advantageous, simple MMA coil embolization alone seems to be effective by reducing the arterial pressure head into the friable capillary beds,14 suggesting that any measure to swing the pendulum towards resorption may be beneficial. Agent and localization of embolization are important factors to address in future studies.

cSDH patients are frail

The majority of patients with cSDH are elderly with comorbid conditions, making them high risk for fall injuries at baseline and higher risk surgical candidates. In many of these frail patients, traditional neurointerventional or surgical outcome metrics, such as 1-year mortality or modified Rankin scores, may introduce challenges in proving procedural safety and efficacy compared with medical treatments as life expectancy may be short and functional capacity may be poor. Surgical (or endovascular) procedures pose risks to these patients, especially if general anesthesia is required. Patients with symptomatic cSDH often have worsened gait disturbances, making the risk of fall-related injuries even higher. Bedside burr hole evacuation procedures were designed specifically to eliminate the need for general anesthesia and have become quite popular by neurosurgeons in this population as they eliminate additional anesthetic consequences. Similarly, achieving MMAE as minimally-invasively as possible (eg, radial access) while minimizing anesthetic risks in this frail population is of importance.

Optimal timing for MMAE and MMAE trials

A generalized traditional neurosurgical management algorithm for cSDH is shown in figure 1. For MMAE to be a useful and effective procedure, not only must we locate an appropriate place for MMAE in this algorithm (yellow circles showing A–D), we must be able to show superiority (or at least non-inferiority) over traditional options. By itself, MMAE does not remove mass effect and therefore will not be effective in immediately improving neurological symptoms (in fact, there are some reports of sudden symptom worsening with MMAE due to suspected expansion related to the procedure). In the presence of deficits, MMAE is probably a poor substitute for evacuation procedures when urgent surgical decompression is necessary (A). Using MMAE for cSDH without deficits during the initial diagnosis phase (B) is one possible point of study, but it may be difficult to show cost-effectiveness and benefit given the potential for spontaneous resorption juxtaposed with the introduction of MMAE procedural risks and cost. There is evidence supporting using MMAE following surgical evacuation (C) to prevent recurrence,15 but this may subject patients to an unnecessary procedure given that many cSDH will resolve after surgical evacuation.13 Finally, introducing MMAE into the management algorithm during the monitoring phase (D) arguably seems most prudent, especially if MMAE is offered when surveillance monitoring shows asymptomatic enlargement as a means of avoiding further surgical evacuation.

Figure 1

Traditional generalized algorithm for management of patients with chronic subdural hematoma before advent of middle meningeal artery embolization (MMAE). Possible points of introduction for MMAE into this traditional management algorithm are shown in yellow as A, B, C, and D.

Natural history of symptomatic convexity cSDH

While cSDH natural history studies are limited, anecdotes from neurosurgical practice and a few small series6–8 seem to suggest that many small cSDH spontaneously regress. However, as the size of the cSDH increases, the rate of spontaneous regression probably decreases. This is supported by recent unpublished data from the control arm of the SQUID Trial for the Embolization of the Middle Meningeal Artery for Treatment of Chronic Subdural Hematoma (STEM), which suggests a roughly 60% failure of medical management alone for patients with convexity symptomatic cSDH measuring at least 1 cm in size.16 This would suggest a wider application for MMAE in the treatment of larger cSDHs.

The best candidates for MMAE are poor trial patients

It is apparent that MMAE is often most useful when traditional surgical intervention is either unsafe or a poor option. Arguably the best candidates for MMAE are those with asymptomatic or minimally symptomatic cSDH and severe thrombocytopenia (acute myelogenous leukemia, liver failure) or there is a need for systemic anticoagulation (mechanical heart valve or pulmonary embolism, for example). Trans-radial MMAE can be performed successfully and safely even in these situations of high bleeding risk.14 Patients with significant bleeding diatheses or that are poor candidates for open surgery will be poorly represented in most trials involving surgery. Nevertheless, trials comparing conservative treatments versus MMAE can study the effect of MMAE in this potentially high-benefit population, such as in STEM, where elderly and coagulopathic patients were included.

Keeping MMAE procedural complications to a minimum is paramount

MMAE does carry certain procedural risks in this frail population, some of which are unique to MMAE (eg, vision loss and facial paralysis). Vessel tortuosity, challenging arches and advanced age increases peri-procedural stroke risk. If a liquid agent is used, general anesthesia may be necessary for adequate and safe embolization, adding anesthesia risk. In addition to the risk of procedural complication, studies have suggested a roughly 5–10% failure rate for MMAE.17 18 We must also keep in mind that the overwhelming majority of series reporting on the topic are retrospective and self-adjudicated, which may underestimate complications and overestimate success compared with rigorous prospective studies.19 Thankfully, the initial RCTs, while unpublished, have reported serious adverse events associated with MMAE to be exceedingly low.20

The cost-effectiveness of MMAE remains a question

A few studies have suggested that MMAE may be more cost-effective than traditional surgical treatment,5 21 but this may not be the case depending on the specific protocol being utilized for MMAE as it pertains to timing within the algorithm (figure 1) and the embolic agent selected (ie, the cost of particles can be 5–10% of the cost of liquid embolics). For instance, whether routine use of MMAE after surgical intervention to prevent the risk for requiring re-operation is cost-effective has not yet been defined. With increasing excitement surrounding trial results, we do run the risk of an expansion of indications beyond what the trial data would support and where MMAE may not be necessary or warranted. Restricting the use of MMAE after evacuation for those with early asymptomatic regrowth or large residual collections would likely enhance cost-effectiveness. Additionally, limiting the use of general anesthesia and overnight hospital admission (except when necessary for patient safety) could further limit procedural costs. As trial data emerge supporting MMAE, cost-effectiveness will likely be an important part of the discussion influencing its widespread acceptance.

Conclusions

The cSDH disease process is complex and specific nuances need to be considered in both interpreting trial results and designing future studies. Thus far, unpublished but presented data from RCTs seem to strongly support a role for adjunctive MMAE in cSDH management, but optimal timing, selection criteria, and technique will need to be further defined. Ultimately the widespread acceptance of this intervention among our non-neurointerventional colleagues will depend on our ability to develop and deliver a cost-effective, low-risk, and minimally-invasive intervention with solid, evidenced-based indications.

Ethics statementsPatient consent for publication

Not applicable.

Ethics approval

Not applicable.