Related Experiment Video
Updated: Sep 25, 2025
A Bedside, Single Burr Hole Approach to Multimodality Monitoring in Severe Brain Injury
Published on: March 26, 2019
Midline Shift in Chronic Subdural Hematoma : Interrater Reliability of Different Measuring Methods and Implications
Umberto Zanolini1, Friederike Austein1, Jens Fiehler1
1Department of Diagnostic and Interventional Neuroradiology, University Medical Center Hamburg-Eppendorf, Martinistr. 52, 20246, Hamburg, Germany.
Objective:
Evaluation of chronic subdural hematoma (cSDH) treatment success relies on radiologic measures, in particular hematoma volume, width and midline shift (MLS). Nevertheless, there are no validated standards for MLS measurement in cSDH. Aim of this study was to identify the most reliable measurement location and technique for MLS.
Methods:
Admission CT scans of 57 patients with unilateral cSDH were retrospectively analyzed. Axial slices were evaluated by 4 raters with MLS measurement in 4 locations, foramen of Monro (FM), thalamus (Th), mid-septum pellucidum (SP), maximum overall MLS (max) with 2 different techniques: displacement perpendicular to anatomical (ideal) midline (MLS-M), and displacement relative to the tabula interna in relation to the width of the intracranial space (MLS-T). Intraclass correlation coefficients (ICC) were calculated to assess interrater reliability and agreement of MLS‑M and MLS‑T measurement techniques. Measurements of cSDH volume and width were conducted for further data alignment.
Results:
The ICCs between readers were excellent (> 0.9) for all MLS‑M locations and for MLS-T_Th and ML-T_FM. The ICC was higher for MLS‑M than for MLS‑T in all locations. MLS-M_max showed the highest correlation coefficient of 0.78 with cSDH volume. Variance of MLS-M_max was explained in 64% of cases (adj. R squared) by cSDH volume based on a simple linear regression model. An increase of 10 ml cSDH volume resulted in an average increase of 0.8 mm MLS-M_max.
Conclusion:
The MLS measurement in cSDH patients should be standardized, and due to its high interrater reliability, the MLS‑M technique should be preferred.
