Abstract
Middle meningeal artery embolization (MMAE) is emerging as a safe and effective standalone intervention for non-acute subdural hematomas (NASHs); however, the risk of hematoma recurrence after MMAE in coagulopathic patients is unclear. To characterize the impact of coagulopathy on treatment outcomes, we analyzed a multi-institutional database of patients who underwent standalone MMAE as treatment for NASH. We classified 537 patients who underwent MMAE as a standalone intervention between 2019 and 2023 by coagulopathy status. Coagulopathy was defined as use of anticoagulation/antiplatelet agents or pre-operative thrombocytopenia (platelets <100,000/μL). Demographics, pre-procedural characteristics, in-hospital course, and patient outcomes were collected. Thrombocytopenia, aspirin use, antiplatelet agent use, and anticoagulant use were assessed using univariate and multivariate analyses to identify any characteristics associated with the need for rescue surgical intervention, mortality, adverse events, and modified Rankin Scale score at 90-day follow-up. Propensity score-matched cohorts by coagulopathy status with matching covariates adjusting for risk factors implicated in surgical recurrence were evaluated by univariate and multivariate analyses. Minimal differences in pre-operative characteristics between patients with and those without coagulopathy were observed. On unmatched and matched analyses, patients with coagulopathy had higher rates of requiring subsequent surgery than those without (unmatched: 9.9% vs. 4.3%; matched: 12.6% vs. 4.6%; both p < 0.05). On matched multivariable analysis, patients with coagulopathy had an increased odds ratio (OR) of requiring surgical rescue (OR 3.95; 95% confidence interval [CI] 1.68–9.30; p < 0.01). Antiplatelet agent use (ticagrelor, prasugrel, or clopidogrel) was also predictive of surgical rescue (OR 4.38; 95% CI 1.51–12.72; p = 0.01), and patients with thrombocytopenia had significantly increased odds of in-hospital mortality (OR 5.16; 95% CI 2.38–11.20; p < 0.01). There were no differences in follow-up radiographic and other clinical outcomes in patients with and those without coagulopathy. Patients with coagulopathy undergoing standalone MMAE for treatment of NASH may have greater risk of requiring surgical rescue (particularly in patients using antiplatelet agents), and in-hospital mortality (in thrombocytopenic patients).
Introduction
Non-acute subdural hematomas (NASHs) encompass subacute, acute-on-chronic, and chronic subdural hematomas. 1 Typically, NASHs are caused by an antecedent trauma, although risk factors independent of trauma include older age, hematological disease, alcoholism, liver cirrhosis, chronic renal failure, and antiplatelet and anticoagulant use. 2 The incidence of NASH is expected to increase because of aging populations and growing use of anticoagulants and antiplatelet agents. 3,4 In patients >65 years of age, the NASH annual incidence of 48.0–80.1 per 100,000 greatly exceeds the incidence of 1–5.3 per 100,000 seen in patients <65 years. 5 –7 Additionally, the use of anticoagulants is reported to increase the risk of chronic subdural hematomas (cSDH) by 42-fold. 8,9 For these reasons, NASHs are predicted to become the most common cranial neurosurgical pathology by 2030. 2,5
The choice of treatment for NASHs depends on several factors, including hematoma size, presence of midline shift, persistent symptoms or neurological deficits, and individual patient risk factors. 10 For expanding hematomas, or those with worsening symptoms/deficits, surgical evacuation is generally indicated; 4,10 however, as with any surgery, evacuation is not without complication risk, particularly because the pathology is encountered most often in the geriatric population. 11 Further, even after evacuation, the risk of hemorrhage recurrence is not insignificant, with 11.2–33% of cases requiring surgical rescue. 12 –14 To mitigate the need for repetitive evacuation and thereby diminish the opportunity for adverse surgical complications, a variety of strategies that aim to reduce recurrence have been developed. 4 One particular minimally invasive treatment modality with notable momentum is middle meningeal artery embolization (MMAE). 15 –17
Large multi-institutional retrospective studies have demonstrated the efficacy of MMAE at reducing the risk of NASH recurrence when used as an adjunct to surgical evacuation. 16 –18 MMAE has also been used as a standalone treatment modality in the treatment of NASH and may be particularly indicated for patients who bear a high surgical risk, exhibit minimal or no symptoms, and/or possess an inherent or acquired coagulopathy, because these factors increase the likelihood of complications after surgical evacuation. 4,16,19 We analyzed a multi-institutional database of patients who underwent MMAE as a standalone intervention for treatment of NASH to characterize the impact of coagulopathy on patient outcomes.
Methods
This study was a retrospective review of a multi-institutional database maintained by 14 United States academic centers. This database includes prospectively collected data on all patients treated for NASH at these institutions who underwent MMAE. We identified 537 patients treated between 2019 and 2023 who underwent MMAE alone for treatment of NASH. Institutional review board approval to collect and review data has been granted by each institution with a waiver of patient consent.
Demographic information including age, sex, and race was collected for all patients. Clinical data, including use of anticoagulant and/or antiplatelet medications, prior treatment of NASH, perioperative platelet counts and pre-procedural modified Rankin Scale (mRS) scores were also collected. Treatment data collected included whether the procedure was a unilateral or bilateral MMAE, anesthesia type, type of embolic agent used, arterial access location, length of hospital stay, procedural complications, procedure duration, and intraprocedural heparin use. Technical success for MMAE was defined as achieving occlusion of at least one branch of the MMA or of the main trunk of the MMA through proper catheter placement and effective embolization, confirmed by post-procedure imaging showing no blood flow in the treated vessel. Radiographic data included NASH laterality, width at initial presentation, midline shift at initial presentation, and presence of intracranial-extracranial arterial anastomoses from the MMA. 20 NASH width and midline shift were also assessed 90 days after treatment. Clinical outcomes, including need for rescue surgery, mortality, adverse events, and mRS score at 90-day follow-up, were also collected.
Patients were grouped into coagulopathic or non-coagulopathic cohorts. Coagulopathy was defined as one or more of the following: pre-morbid use of antithrombotic and/or antiplatelet medications, or thrombocytopenia (platelet count <100,000/μL) on presentation. Within the coagulopathic group, patients were further stratified based on the presence of thrombocytopenia, use of aspirin (81 or 325 mg), use of P2Y12 inhibitors (prasugrel, ticagrelor, or clopidogrel), or use of anticoagulation medication. Univariate and multivariate analyses were performed. Continuous data were assessed via a Student's t test and categorical data were assessed with the χ 2 test. Missing data were not replaced. Propensity-score matching was then conducted based on the presence of coagulopathy, with matching covariates including age, pre-operative hematoma size, and pre-operative mRS score. These covariates were selected prior to the analyses for their clinical association with reoperation requirement or primary hemorrhage severity. 21 –23 After matching, 239 patients with coagulopathy and 219 patients without coagulopathy were included for analysis. All univariate and multivariate analyses were then repeated. Statistical significance was defined as p < 0.05.
Results
Effect of coagulopathy
Minimal differences were observed in presenting demographic, clinical, and radiographic characteristics between patients with and those without coagulopathy (Table 1), although the coagulopathic patients were older, with a mean age of 72.3 ± 13.3 years (p = 0.01). This difference did not persist after matching (p = 0.32). Pre-procedural NASH thickness, midline shift, and presenting mRS scores were not significantly different between patients with and those without coagulopathy in either unmatched and matched groups. Procedural details were also similar in terms of bilateral procedures, use of radial access, anesthesia type, embolic agent used, MMAE technical success, and complications encountered, regardless of matched status (Table 2). However, patients with coagulopathy who underwent MMAE for primary treatment of NASH were more likely to require surgical rescue than those without coagulopathy, with an absolute difference of 5.6% before matching (p = 0.02). After propensity matching, coagulopathy increased the likelihood of surgical rescue further, with an absolute difference of 8.0% (p < 0.01). Upon multivariate analysis, unmatched patients with coagulopathy demonstrated increased odds of requiring rescue surgery (odds ratio [OR] 3.22; 95% confidence interval [CI] 1.37–7.58; p = 0.01); following propensity matching, the odds ratio increased to 3.95 (95% CI 1.68–9.30; p < 0.01) (Table 3).
Pre Procedural Characteristics
Values reported as number (%) or mean ± SD. Boldface font indicates statistical significance at p < 0.05.
Defined as platelet count <100,000/μL
mRS, modified Rankin Scale; SD, standard deviation.
In-Hospital Characteristics
Values reported as number (%) or mean ± SD or median (IQR). Boldface font indicates statistical significance at p < 0.05.
Technical success was defined as radiographically confirmed occlusion of at least one branch or main trunk of the MMA.
nBCA, n-butyl-cyanoacrylate; SD, standard deviation; IQR, interquartile range; MMA, middle meningeal artery.
Multivariable Analysis for Primary Outcomes
Boldface font indicates statistical significance at p < 0.05.
CI, confidence interval; mRS, modified Rankin Scale; NA, not calculable because of data overfitting.
Effect of coagulopathy subtype
Analysis based on coagulopathy subtype demonstrated no association between surgical rescue and thrombocytopenia, or the use of aspirin or anticoagulation medication (Table 4). However, the use of P2Y12 inhibitors increased the odds of requiring surgical rescue in both the unmatched (OR 4.95; 95% CI 1.76–13.94; p < 0.01) and matched cohorts (OR 4.38; 95% CI 1.51–12.72; p = 0.01). With regards to other outcome measures, patients with thrombocytopenia had significantly increased odds of experiencing in-hospital mortality in both the unmatched (OR 4.92; 95% CI 2.38–10.18; p < 0.01) and matched cohorts (OR 5.16; 95% CI 2.38–11.20; p < 0.01) (Table 4). Increased age was also associated with a slight increase in mortality (unmatched OR 1.04; 95% CI 1.01–1.06; p = 0.01; matched OR 1.03; 95% CI 1.01–1.06; p = 0.03). Functional outcome at 90 days was not affected by coagulopathy subtype in either unmatched or matched analyses.
Multivariate Sub-Analysis of Primary Outcomes per Coagulopathy Subtypes
Boldface font indicates statistical significance at p < 0.05.
Defined as platelet count <100,000/μL
CI, confidence interval; mRS, modified Rankin Scale
Follow-up outcomes
There were no differences in follow-up radiographic outcomes, including mean subdural hematoma thickness at 90 days and at last follow-up (median 74 days) and mean subdural hematoma thickness change from presentation, between patients with and those without coagulopathy in the unmatched or matched cohorts. Similarly, functional outcomes at 90 days and at last follow-up were no different between the cohorts, regardless of matched status (Table 5).
Follow-Up Outcomes
Values reported as number (%) or mean ± SD.
mRS, modified Rankin Scale; SD, standard deviation.
Discussion
Although MMAE may serve as a less-invasive treatment modality than traditional surgical evacuation for patients with NASH, it is imperative to further characterize its safety profile in individuals who are taking antiplatelet and anticoagulant medications, as well as those with thrombocytopenia. This is particularly important given the increased propensity for NASH among geriatric patients, who can be coagulopathic because of intrinsic comorbidities or as a result of medication. 24,25 The present study represents the largest multi-institutional analysis aimed at assessing the impact of coagulopathy in patients who underwent MMAE as a standalone treatment for NASH.
Although several earlier studies have analyzed the effects of antiplatelet and anticoagulant agents on MMAE as a secondary objective, results have been conflicting. One study of 53 patients investigating the timing and influence of various factors on cSDH resolution after MMAE found no significant association between resolution and pre-procedural antiplatelet or anticoagulant agent use. 26 Another study found no impact of antiplatelet use on NASH recurrence after treatment for cSDH. 27 A key limitation to this study was that the analysis included patients who underwent surgery, surgery with concomitant MMAE, and standalone MMAE, making it difficult to draw conclusions on patients undergoing just standalone MMAE. A previously published study from our multi-center data set including 636 total MMAE procedures, of which 318 were standalone MMAE, found that patients on anticoagulants were more likely to experience hematoma expansion or neurological deterioration requiring surgery, but that those taking antiplatelet agents did not experience worsened clinical or radiographic outcomes. 28 That study did not analyze individual antiplatelet agents and did not specifically examine the need for surgical rescue, which may explain the differing results obtained in our study.
After performing propensity matching to adjust for risk factors associated with surgical rescue, our results indicated that patients with coagulopathy do experience a higher rate of surgical rescue following standalone MMAE than do patients without coagulopathy (12.6% vs. 4.6%, p < 0.01). On multivariate analysis, patients with coagulopathy had a 3.95-fold increased OR for requiring rescue surgery. However, upon stratification of subtypes of coagulopathy, only the pre-procedural use of P2Y12 inhibitors was associated with increased odds of NASH recurrence requiring surgical rescue after standalone MMAE (OR 4.38; 95% CI 1.51–12.72; p = 0.01).
The higher rate of surgical rescue after standalone MMAE in patients on P2Y12 inhibitors may be attributed to the pathophysiology of NASHs. NASH formation involves the development of an immature network of blood vessels (i.e., neurovasculature) through angiogenesis within the subdural membranes. This network of blood vessels is thought to originate from MMA arterial branches. In NASH, this network is repeatedly disrupted with cycles of inflammation, leading to intermittent bleeding and hemorrhage growth. 29 Platelets are commonly associated with thrombosis and hemostasis, but they also play a significant role in angiogenesis. 30 Although platelet inhibition can hinder angiogenesis, studies have shown that P2Y12 inhibitors do not effectively inhibit angiogenesis. As a result, these agents have no noticeable impact on circulating levels of angiogenic factors or endothelial activation levels. 31 This is in contrast to aspirin, which is known to restrict angiogenesis. 32 –34 Therefore, it is possible that these pathophysiological relationships explain our findings of an increased likelihood of treatment failure in patients taking P2Y12 inhibitors but not aspirin. Further study is needed to definitively characterize these pharmacological and physiological relationships in the setting of both coagulopathy and MMAE.
Our matched analysis also demonstrates that patients with thrombocytopenia who undergo standalone MMAE have increased odds of in-hospital mortality compared with those with normal platelet counts (OR 5.16; 95% CI 2.38–11.20; p < 0.01). Thrombocytopenia has been widely associated with poor outcomes after surgery, bodily trauma, and in the elderly. 35 –39 However, in the context of endovascular neurosurgical interventions, it remains unclear how pre-operative thrombocytopenia ought to inform pre-procedural decision making. A subanalysis of the Recovery by Endovascular Salvage for Cerebral Ultra-Acute Embolism (RESCUE)-Japan Registry 2 found that thrombocytopenia was associated with increased hemorrhagic complications, disability, and mortality after endovascular thrombectomy. 40 Among patients undergoing endovascular aneurysm coiling, thrombocytopenia has also been associated with increased morbidity and mortality, although the strength of this evidence is limited. 41,42
Limitations
Our database lacks information regarding whether coagulopathy reversal was attempted. Although this was a key limitation of our data set, our goal was to determine the effect of coagulopathy without assessing the influence of reversal agents, which should be the focus of future work. Second, the data set lacked information regarding each patient's platelet inhibitory response to aspirin or P2Y12 inhibitors. Third, although our study is strengthened in size and scope by including patients from multiple institutions, the selection criteria for standalone MMAE embolization was not standardized. This limits our ability to draw conclusions about the characteristics/procedural indications of patients who were selected for standalone MMAE; however, typically, these patients present with minimal neurological symptoms. The real-world nature of this report is also impacted by technical variation in performing MMAE. Finally, the cause of in-hospital mortality and the relation to pre-existing conditions, especially in patients with thrombocytopenia, were not captured in our database. Prospective randomized controlled trials with detailed follow-up are needed to fully define the appropriate use of standalone MMAE in patients with coagulopathy.
Conclusion
Patients with coagulopathy who undergo standalone MMAE for NASH may have an increased risk of requiring surgical rescue, which was associated with P2Y12 inhibitor use. Patients with thrombocytopenia who undergo MMAE for NASH may have increased risks for in-hospital mortality.
Transparency, Rigor, and Reproducibility Summary
The analysis plan for this study was not formally pre-registered, but the author with primary responsibility for the statistical analysis certifies that the analysis plan was pre-specified. Statistical analysis was set with an alpha of 0.05. The study sample size was determined by the number of available patients (n = 537) from contributing institutions, and all available data were considered for analysis. To enhance the robustness of our analysis, propensity score matching was performed, which included 458 of the 537 original patients. In our study, 59 patients died as a result of the treatments performed. Patient identifying data were blinded, although data analyses were performed by a statistician who was aware of all participant data. Data were collected from 15 academic centers in the United States in 2019–2023. Data were analyzed using SPSS. The primary surgical technique assessed in this study is well validated and is a standard in the field. Coagulopathy, as defined in our study, is a well-known medical comorbidity. To the best of our knowledge, no replication or external validation of the study results is ongoing. Data and analytical codes used in this analysis are available from the authors upon reasonable request.
Footnotes
Acknowledgments
We thank Kristin Kraus for her editorial assistance.
Authors' Contributions
Walid K. Salah: methodology, writing–original draft, formal analysis, writing – review and editing; Matthew Findlay: methodology, writing–original draft, formal analysis, writing – review and editing; Cordell Baker: investigation, writing – review and editing; Jonathan Scoville: investigation; Michael Bounajem: investigation; Christopher Ogilvy: investigation; Justin Moore: investigation; Howard Riina: investigation; Elad Levy: investigation; Adnan Siddiqui: investigation; Alejandro Spiotta: investigation; C Michael Cawley: investigation; Alexander Khalessi: investigation; Omar Tanweer: investigation; Ricardo Hanel: investigation; Bradley Gross: investigation; Okkes Kuybu: investigation; Brian Howard: investigation; Alex Hoang: investigation; Ammad Baig: investigation; MirHojjat Khorasanizadeh: investigation; Aldo Mendez Ruiz: investigation; Gustavo Cortez: investigation; Jason Davies: investigation; Michael Lang: investigation; Ajith Thomas: investigation; Daniel Tonetti: investigation: Jane Khalife: investigation; Georgios Sioutas: investigation; Kate Carroll: investigation; Zachary Abecassis: investigation; Brian Jankowitz: investigation; Juan Ruiz Rodriguez: investigation; Michael Levitt: investigation; Peter Kan: investigation; Jan-Karl Burkhardt: conceptualization, investigation, resources, project administration; Visish Srinivasan: investigation; Mohamed Salem: project administration, investigation, writing – review and editing; and Ramesh Grandhi: conceptualization, investigation, resources, supervision, project administration, writing – review and editing.
Funding Information
There was no funding provided for this research.
Author Disclosure Statement
E.I.L. received consulting fees from Clarion, GLG Consulting, Guidepoint Global, Imperative Care, Medtronic, StimMed, Misionix, and Mosiac; lecture payments from Clarion, GLG Consulting, Guidepoint Global, Imperative Care, Medtronic, StimMed, Misionix, and Mosiac; expert testimony payment from Renders Medical; has patents planned, issued, or pending for Ultrasonic surgical blade; is on the advisory board for Stryker, NeXtGen Biologics, MEDX, Cognition Medical, Endostream Medical, and IRRAS; is a stockholder of NeXtGen Biologics, RAPID Medical, Claret Medical, Cognition Medical, Imperative Care, Rebound Therapeutics, StimMed, and Three Rivers Medical; and is the chief medical officer for Haniva Technology. A.H.S. received a grant from the National Institutes of Health (NIH); consulting fees from Amnis Therapeutics, Apellis Pharmaceuticals, Boston Scientific, Canon Medical Systems, Cardinal Health 200, Cerebrotech Medical Systems, Cerenovus, Cerevatech Medical, Cordis, Corindus, Endostream Medical, Imperative Care, InspireMD, Integra, IRRAS AB, Medtronic, MicroVention, Minnetronix Neuro, Peijia Medical, Penumbra, Piraeus Medical, Q'Apel Medical, Rapid Medical, Serenity Medical, Silk Road Medical, StimMed, Stryker Neurovascular, Three Rivers Medical, VasSol, and Viz.ai; is a stockholder of Adona Medical, Amnis Therapeutics, Bend IT Technologies, BlinkTBI, Borvo Medical, Cerebrotech Medical Systems, Cerevatech Medical, Cognition Medical, Collavidence, CVAID, E8, Endostream Medical, Galaxy Therapeutics, Imperative Care, InspireMD, Instylla, International Medical Distribution Partners, Launch NY, Neurolutions, NeuroRadial Technologies, NeuroTechnology Investors, Neurovascular Diagnostics, Peijia Medical, PerFlow Medical, Piraeus Medical, Q'Apel Medical, QAS.ai, Radical Catheter Technologies, Rebound Therapeutics, RIST Neurovascular, Sense Diagnostics, Serenity Medical, Silk Road Medical, Sim & Cure, SongBird Therapy, Spinnaker Medical, StimMed, Synchron, Three Rivers Medical, Truvic Medical, Tulavi Therapeutics, Vastrax, VICIS, Viseon, and Whisper Medical; has received payments related to Cerenovus EXCELLENT and ARISE II Trials; Medtronic SWIFT PRIME, VANTAGE, EMBOLISE, and SWIFT DIRECT Trials; MicroVention FRED Trial & CONFIDENCE Study; MUSC POSITIVE Trial; Penumbra 3D Separator Trial, COMPASS Trial, INVEST Trial, MIVI Neuroscience EVAQ Trial; Rapid Medical SUCCESS Trial; and InspireMD C-GUARDIANS IDE Pivotal Trial. A.M.S. has received grants or contracts from Medtronic, Stryker, Penumbra, Avail, and RapidAI; consulting fees from Stryker, Penumbra, RapidAI, and Terumo; is on the medical advisory board of the Brain Aneurysm Foundation; and is a stockholder of Avail Med. A.A.K. has received consulting fees from Medtronic; payment for expert testimony from Procopio US Attorney; is on the advisory board of Route 92 and Medtronic; has a leadership role in the Congress of Neurological Surgeons and American College of Surgeons; and is a stockholder of Ospitek, Synaptive, and Proximie. R.H. has received grants or contracts from NIH, Interline Endowment, Microvention, Stryker, CNX, and Balt; consulting fees from Medtronic, Balt, Stryker, Q'Apel Medical, Codman Neuro (J&J), Cerenovus, Microvention, Imperative Care, Phenox, and Rapid Medical; is on the advisory board of MiVI, eLum, Three Rivers, Shape Medical, and Corindus; is associate editor of the endovascular section for Neurosurgery Journal; and is a stockholder of InNeuroCo, Cerebrotech, eLum, Endostream, Three Rivers Medical, Scientia, RisT, Blink TBI, and Corindus. B.A.G. has received consulting fees from Medtronic and Microvention. J.M.D. has received institutional grants or contracts from the NIH, NSF SBIR, University at Buffalo Center for Advanced Technology, Buffalo Translational Consortium, Cummings Foundation, NVIDIA, and Google; royalties from RIST Neurovascular; lecture payment from Medtronic; is on the advisory board for NIH Strokenet; has a leadership role in the cerebrovascular section of the Congress of Neurological Surgeons; and is a stockholder of QAS.ai, RIST Neurovascular, Cerebrotech, Synchron, and Hyperion. A.J.T. receives consulting fees from Stryker, Medtronic, and Cerevasc. M.R.L. has received unrestricted educational grants from Medtronic and Stryker; consulting fees from Medtronic, Metis Innovative, Aeaean advisers, and Stereotaxis; is on the data safety monitoring board for Arsenal; has equity interests in Hyperion Surgical, Proprio, Aperture, Cerebrotech, Synchron, Fluid Biomed, and Stereotaxis; and is on the editorial board on the Journal of NeuroInterventional Surgery and Frontiers in Surgery. P.T.K. has received grants from the NIH (1U18EB029353-01), Medtronic (ERP-2019-12070), Siemens (CON30434), and Joe Niekro Foundation (CON30914); consulting fees from Stryker Neurovascular, Imperative Care, Cerenovus, and Microvention; and is on the editorial board of the Journal of Neurointerventional Surgery. J.K.B. receives consulting fees from Q`Apel Medical, Stryker, Medtronic, Cerenovous, and Microvention. R.G. receives consulting fees from Medtronic Neurovascular, Balt Neurovascular, and Cerenovus. The other authors have nothing to disclose.