{
  "abstract": "Introduction Middle meningeal artery embolization (MMAE) is an effective treatment for chronic subdural hematoma (cSDH); however, optimal post-procedural imaging surveillance remains undefined.Objective To determine whether volumetric reduction on initial follow-up imaging after MMAE can predict subsequent cSDH resolution with high certainty.Methods Single-institution, retrospective cohort review of consecutive patients undergoing MMAE before March 2025. Inclusion required two post-procedural CT scans within 180 days of MMAE: an initial scan within 6 weeks of MMAE and a second scan ≥ 30 days later. Patients undergoing burr hole or craniotomy prior to MMAE were included if an interval CT between surgical evacuation and embolization was available to serve as the pre-embolization baseline. Hematoma volumes were measured using 3D segmentation at baseline and both follow-up time points. Short-term predictors included percent reduction (PR) and percent reduction rate (PRR) from baseline to first follow-up. The primary outcome was near-complete resolution (volume <5 mL) at the second follow-up. Multivariate logistic regression and adjusted receiver operating characteristic (ROC) analyses were performed.Results A total of 112 cSDHs (81 patients) were analyzed; 54 (48.2%) achieved near-complete resolution by second follow-up. Patients achieving resolution were younger (70.5 vs. 77.2 years, p = 0.001) and were less likely to demonstrate membranous hematoma morphology (42.6% vs 70.7%, p = 0.005). In multivariable models, PR remained independently associated with resolution after controlling for baseline hematoma volume (V0), age and morphology (OR 1.07 per 1% increase, 95% CI [1.04-1.10]; p<0.001). A parallel model incorporating PRR similarly demonstrated independent predictive value (OR 2.55 per 1%/day increase, 95% CI [1.69-3.86]; p<0.001).Adjusted ROC analysis demonstrated strong predictive performance in the overall cohort (PR+V0 AUC 0.92; PRR+V0 AUC 0.89; figure 1a). A 55.6% hematoma reduction by first follow-up CT predicted resolution on next CT with 98.3% specificity and 97.1% PPV (34 TP, 1 FP, 57 TN, 20 FN).In patients treated with MMAE alone (n=63), predictive performance improved (AUC 0.95 for both models; figure 1b), with a PR threshold of 45.0% yielding 96.8% specificity and 96.0% PPV (24 TP, 1 FP, 30 TN, 8 FN).In the Surgery+MMAE cohort (n=49), predictive performance remained high (PR+V0 AUC 0.93; PRR+V0 AUC 0.85; figure 1c), with a PR threshold of 48.9% yielding 100% specificity and PPV (14 TP, 0 FP, 27 TN, 8 FN).Conclusions Volumetric reduction of approximately 50% on initial follow-up imaging after MMAE can predict subsequent cSDH resolution with high confidence, potentially obviating additional routine imaging.Disclosures K. Gunn: None. N. Dogra: None. G. Kohli: None. E. Sleasman: None. A. Anandarajah: None. S. Mendoza-Ayus: None. E. Davis: None. D. Jakimovski: None. P. Valdes Barrera: None. V. Nguyen: None. T. Bhalla: None. T. Mattingly: None. M. Bender: None.Abstract E-224 Figure 1Receiver operating characteristic (ROC) curves for prediction of chronic subdural hematoma resolution. (A) Top displays the full cohort. (B) Bottom-left displays the MMAE-only cohort. (C) Bottom-right displays Surgery + MMAE full cohort",
  "authors": [
    {
      "affiliations": [
        "University of Rochester Medical Center, Rochester, NY"
      ],
      "name": "K Gunn"
    },
    {
      "affiliations": [
        "University of Rochester Medical Center, Rochester, NY"
      ],
      "name": "N Dogra"
    },
    {
      "affiliations": [
        "Neurosurgery, University of Rochester Medical Center, Rochester, NY"
      ],
      "name": "G Kohli"
    },
    {
      "affiliations": [
        "University of Rochester Medical Center, Rochester, NY"
      ],
      "name": "E Sleasman"
    },
    {
      "affiliations": [
        "University of Rochester Medical Center, Rochester, NY"
      ],
      "name": "A Anandarajah"
    },
    {
      "affiliations": [
        "Neurosurgery, University of Rochester Medical Center, Rochester, NY"
      ],
      "name": "S Mendoza-Ayus"
    },
    {
      "affiliations": [
        "University of Rochester Medical Center, Rochester, NY"
      ],
      "name": "E Davis"
    },
    {
      "affiliations": [
        "Radiology, University of Rochester Medical Center, Rochester, NY"
      ],
      "name": "D Jakimovski"
    },
    {
      "affiliations": [
        "Neurosurgery, University of Rochester Medical Center, Rochester, NY"
      ],
      "name": "P Valdes Barrera"
    },
    {
      "affiliations": [
        "Neurosurgery, University of Rochester Medical Center, Rochester, NY"
      ],
      "name": "V Nguyen"
    },
    {
      "affiliations": [
        "Neurosurgery, University of Rochester Medical Center, Rochester, NY"
      ],
      "name": "T Bhalla"
    },
    {
      "affiliations": [
        "Neurosurgery, University of Rochester Medical Center, Rochester, NY"
      ],
      "name": "T Mattingly"
    },
    {
      "affiliations": [
        "Neurosurgery, University of Rochester Medical Center, Rochester, NY"
      ],
      "name": "M Bender"
    }
  ],
  "title": "E-224 Short-term radiographic reduction predicts resolution of chronic subdural hematomas after middle meningeal artery embolization",
  "uid": "09ebf54f-6303-5dc8-8028-9f3df8a5243e"
}