{
  "abstract": "Background The EMBOLISE trial showed that middle meningeal artery embolization (MMAE) with Onyx reduces reoperation for chronic subdural hematoma (cSDH) from 11.3% to 4.1% (relative risk 0.36, p=0.008). A numerically higher but non-significant mortality was observed in the treatment group (5.1% vs. 3.0% at 90 days, p=0.32). We investigated whether baseline characteristics explain this difference.Methods We analyzed 400 EMBOLISE surgical patients (197 MMAE + surgery, 203 surgery alone). Charlson Comorbidity Index (CCI) scores were calculated from 934 MedDRA-coded medical history terms by investigators blinded to treatment assignment. Multivariate logistic regression assessed predictors of mortality at 90 and 180 days. An independent Clinical Events Committee (CEC) adjudicated all deaths through 180 days. We also compared mortality across published randomized controlled trials (RCTs) of MMAE for cSDH (EMBOLISE, STEM, MAGIC-MT, EMPROTECT; N=1,774).Results Despite randomization, the MMAE group had worse baseline health: higher CCI scores (5.0 ± 2.45 vs. 4.4 ± 2.20, p=0.017) corresponding to 10-year survival probabilities of 21.4% vs. 40.7%. The MMAE group was also older (73.0 vs. 71.0 years, p=0.074) with worse premorbid function (mRS 0.62 vs. 0.49, p=0.11). Through 180 days, 23 patients died: 14 MMAE (7.1%) and 9 control (4.4%), p=0.29. In multivariate analysis, age (OR 1.24 per year, p=0.0003) and pre-morbid mRS (OR 2.02, p=0.011) predicted 90-day mortality; MMAE did not (all six scenarios p>0.10). The CEC attributed zero deaths to MMAE or Onyx; most deaths occurred beyond postoperative day 30 and reflected failure to recover from the index SDH or medical comorbidities. Across four RCTs, mortality patterns were inconsistent: MAGIC-MT favored MMAE (0.6% vs. 2.2%), EMPROTECT showed similar rates (5.5% vs. 7.9%), and STEM showed higher overall but lower neurological mortality with MMAE.Conclusions The mortality imbalance in EMBOLISE is attributable to baseline patient characteristics, not procedural harm. Comorbidity analysis, multivariate regression, CEC adjudication, and inconsistent mortality patterns across 1,774 patients in four RCTs all argue against a causal relationship between MMAE and death.Disclosures J. Davies: 2; C; Medtronic.",
  "authors": [
    {
      "affiliations": [
        "University at Buffalo, Buffalo, NY"
      ],
      "name": "J Davies"
    },
    {
      "affiliations": [
        "Neurology, UTSCHSA Valley Baptist, Harlingen, TX"
      ],
      "name": "J Knopman"
    },
    {
      "affiliations": [
        "Neurology, UTSCHSA Valley Baptist, Harlingen, TX"
      ],
      "name": "A Hassan"
    },
    {
      "affiliations": [
        "UC San Diego Health System, San Diego, CA"
      ],
      "name": "A Khalessi"
    },
    {
      "affiliations": [
        "University at Buffalo, Buffalo, NY"
      ],
      "name": "M Mokin"
    },
    {
      "affiliations": [
        "University of Buffalo, Buffalo, NY"
      ],
      "name": "A Siddiqui"
    }
  ],
  "title": "E-087 Mortality in the EMBOLISE trial: a post-hoc analysis of baseline comorbidity, multivariate regression, and cross-trial comparison",
  "uid": "9d343b1c-05c5-51f8-ba6e-5d537154f8b2"
}