Article
3D-volumetry of perihematomal edema and the relation to outcome in spontaneous cerebellar hemorrhage
Search Medline for
Authors
-
Leander Matthies
- Justus-Liebig-Universität Gießen, Klinik für Neurochirurgie, Gießen, Deutschland
-
Marco Stein
- Justus-Liebig-Universität Gießen, Klinik für Neurochirurgie, Gießen, Deutschland
-
Malgorzata Kolodziej
- Justus-Liebig-Universität Gießen, Klinik für Neurochirurgie, Gießen, Deutschland
-
Carolin Gramsch
- Justus-Liebig-Universität Gießen, Institut für Neuroradiologie, Gießen, Deutschland
-
Michael Bender
- Justus-Liebig-Universität Gießen, Klinik für Neurochirurgie, Gießen, Deutschland
-
Marcus Reinges
- Klinikum Bremen-Mitte, Klinik für Neurochirurgie, Bremen, Deutschland
-
Eberhard Uhl
- Justus-Liebig-Universität Gießen, Klinik für Neurochirurgie, Gießen, Deutschland
| Published: | June 18, 2018 |
|---|
Outline
Text
Objective: Volume-dependent effects of perihematomal edema (PHE) on patients’ outcome are described for supratentorial intracerebral hemorrhage (ICH). The influence of PHE on outcome in infratentorial hematomas has not been systematically evaluated. This study was conducted to evaluate the reliability of PHE 3D-volumetry and to explore the effect of PHE on outcome after spontaneous cerebellar ICH.
Methods: 104 patients with cerebellar ICH were included in this study. Hematoma volume (HV) and PHE volumes were determined by two independent raters. Intrarater and interrater reliability were calculated with intraclass correlation coefficients (ICC). Bland–Altman plots were calculated to evaluate interrater consistency. Unfavorable outcome was defined as Glasgow outcome scale (GOS) 1-3 and was assessed 3 months after discharge.
Results: 104 patients with cerebellar ICH were included. The median age was 72 years (IQR: 64-79). In 58 patients (55.8%) a craniotomy was performed. The median initial Glagow coma score (GCS) was 12 (IQR: 5-14). Mean HV was 20.8±13.2cm³. Mean PHE volume was 8.5±7.3cm³. Intrarater ICC and interrater ICC for HV were 0.98 (0.97-0.99) and 0.99 (0.99-1.00). Intrarater ICC and interrater ICC for PHE volume were 0.98 (0.97-1.00) and 0.96 (0.90-0.99). The calculated bias for HE volume and PHE volume [-1.16cm³ (SD ±1.32cm³) and 0.69cm³ (SD: ±1.65cm³)] was in acceptable limits.
30-day mortality was 24% (n=25). GOS at 3 months after discharge was available for 63 patients (79.7%). 32 patients (50.1%) had a favorable outcome. Higher hematoma volume was associated with UO in conservatively (15.1±8.5cm³ vs. 7.8±3.5cm³; P=0.006) and surgically treated [30.2±13.7cm³ vs. 18.4±6.8cm³; P=0.003] patients. PHE volume was only predictive for unfavorable outcome in conservatively treated patients (8.9±6.2cm³ vs. 4.7±2.5cm³; P=0.029) but not in surgically treated patients (11.1±5.8cm³ vs. 7.7±4.3cm³; P=0.074).
Conclusion: Reliability of 3D-volumetry for HV and PHE volume are excellent. PHE is associated with unfavorable outcome in non-surgically treated patients. PHE and not only hematoma volume should be evaluated and considered into decision making in patients with spontaneous cerebellar ICH.
