Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Volume and Attenuation Characteristics of Chronic Subdural Hematoma : An Annotated Patient Cohort of 257 Patients with Interrater Reliability Assessments

Drake, Mattias LU ; Hall, Emma LU orcid ; Ramgren, Birgitta LU ; Hansen, Björn M LU orcid and Wassélius, Johan LU (2025) In Tomography 11(12).
Abstract


Background: Accurate volumetry and imaging characterization of chronic subdural hematoma (cSDH) are essential for prognostication and treatment planning, but manual assessment is time-consuming and therefore underutilized.
Methods: We retrospectively analyzed preoperative non-contrast CT (NCCT) scans of 257 patients undergoing first-time surgery for uni- or bilateral cSDH. Hematoma volumes were measured manually using a semi-automated area-outlining tool on every second axial slice and compared with the volumes estimated through the ABC/2 formula. Hematoma attenuation patterns and components were categorized, and interrater reliability was assessed for volume, maximum diameter, and imaging features using intraclass... (More)


Background: Accurate volumetry and imaging characterization of chronic subdural hematoma (cSDH) are essential for prognostication and treatment planning, but manual assessment is time-consuming and therefore underutilized.
Methods: We retrospectively analyzed preoperative non-contrast CT (NCCT) scans of 257 patients undergoing first-time surgery for uni- or bilateral cSDH. Hematoma volumes were measured manually using a semi-automated area-outlining tool on every second axial slice and compared with the volumes estimated through the ABC/2 formula. Hematoma attenuation patterns and components were categorized, and interrater reliability was assessed for volume, maximum diameter, and imaging features using intraclass correlation coefficients (ICCs) and Cohen's κ.
Results: A total of 339 hematomas were evaluated. Manual and ABC/2 volume measurements correlated strongly (R
2 = 0.83, ICC [3, 1] = 0.90). The interrater agreement for manual volumetry was excellent (ICC [2, 1] = 0.96). Agreement was also excellent for maximum diameter (ICC [2, 1] > 0.9) and good for midline shift assessment (κ = 0.81). Agreement was moderate for the identification of fresh clots, trabeculations, and laminations (κ = 0.62-0.72) but poor for general attenuation patterns (κ = 0.44).
Conclusions: The manual volumetry of cSDH is feasible and highly reproducible between raters of different experience levels. These results provide a robust reference standard for the validation of automated volumetry tools and support the implementation of quantitative hematoma assessment in future clinical trials and routine care.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Humans, Hematoma, Subdural, Chronic/diagnostic imaging, Retrospective Studies, Male, Female, Reproducibility of Results, Aged, Middle Aged, Tomography, X-Ray Computed/methods, Observer Variation, Aged, 80 and over, Adult
in
Tomography
volume
11
issue
12
article number
141
publisher
MDPI AG
external identifiers
  • scopus:105025740974
  • pmid:41441393
ISSN
2379-1381
DOI
10.3390/tomography11120141
language
English
LU publication?
yes
id
f4c12e88-60c6-4a1c-84b2-e96b7042e0a9
date added to LUP
2026-01-07 18:54:08
date last changed
2026-02-05 06:46:30
@article{f4c12e88-60c6-4a1c-84b2-e96b7042e0a9,
  abstract     = {{<p><br>
 Background: Accurate volumetry and imaging characterization of chronic subdural hematoma (cSDH) are essential for prognostication and treatment planning, but manual assessment is time-consuming and therefore underutilized.<br>
 Methods: We retrospectively analyzed preoperative non-contrast CT (NCCT) scans of 257 patients undergoing first-time surgery for uni- or bilateral cSDH. Hematoma volumes were measured manually using a semi-automated area-outlining tool on every second axial slice and compared with the volumes estimated through the ABC/2 formula. Hematoma attenuation patterns and components were categorized, and interrater reliability was assessed for volume, maximum diameter, and imaging features using intraclass correlation coefficients (ICCs) and Cohen's κ.<br>
 Results: A total of 339 hematomas were evaluated. Manual and ABC/2 volume measurements correlated strongly (R<br>
 2 = 0.83, ICC [3, 1] = 0.90). The interrater agreement for manual volumetry was excellent (ICC [2, 1] = 0.96). Agreement was also excellent for maximum diameter (ICC [2, 1] &gt; 0.9) and good for midline shift assessment (κ = 0.81). Agreement was moderate for the identification of fresh clots, trabeculations, and laminations (κ = 0.62-0.72) but poor for general attenuation patterns (κ = 0.44).<br>
 Conclusions: The manual volumetry of cSDH is feasible and highly reproducible between raters of different experience levels. These results provide a robust reference standard for the validation of automated volumetry tools and support the implementation of quantitative hematoma assessment in future clinical trials and routine care.<br>
 </p>}},
  author       = {{Drake, Mattias and Hall, Emma and Ramgren, Birgitta and Hansen, Björn M and Wassélius, Johan}},
  issn         = {{2379-1381}},
  keywords     = {{Humans; Hematoma, Subdural, Chronic/diagnostic imaging; Retrospective Studies; Male; Female; Reproducibility of Results; Aged; Middle Aged; Tomography, X-Ray Computed/methods; Observer Variation; Aged, 80 and over; Adult}},
  language     = {{eng}},
  month        = {{12}},
  number       = {{12}},
  publisher    = {{MDPI AG}},
  series       = {{Tomography}},
  title        = {{Volume and Attenuation Characteristics of Chronic Subdural Hematoma : An Annotated Patient Cohort of 257 Patients with Interrater Reliability Assessments}},
  url          = {{http://dx.doi.org/10.3390/tomography11120141}},
  doi          = {{10.3390/tomography11120141}},
  volume       = {{11}},
  year         = {{2025}},
}