Article
Prediction of cerebral infarction after aneurysmal subarachnoid hemorrhage with a risk score
Search Medline for
Authors
-
Ramazan Jabbarli
- Klinik für Neurochirurgie, Univeristätsklinikum Freiburg; Klinik für Neurochirurgie, Universitätsklinikum Essen
-
Matthias Reinhard
- Klinik für Neurologie, Universitätsklinikum Freiburg
-
Roland Roelz
- Klinik für Neurochirurgie, Univeristätsklinikum Freiburg
-
Mukesch Shah
- Klinik für Neurochirurgie, Univeristätsklinikum Freiburg
-
Klaus Kaier
- Department für Medizinische Biometrie und Medizinische Informatik, Universitätsklinikum Freiburg
-
Christian Taschner
- Klinik für Neuroradiologie, Universitätsklinikum Freiburg
-
Vera Van Velthoven
- Klinik für Neurochirurgie, Univeristätsklinikum Freiburg
| Published: | June 2, 2015 |
|---|
Outline
Text
Objective: Cerebral infarction is a crucial complication of aneurysmal subarachnoid hemorrhage (SAH) associated with poor clinical outcome. We aimed at developing an early risk score for cerebral infarction based on clinical characteristics available at the onset of SAH.
Method: Out of a databank containing 632 consecutive patients with SAH admitted to our institution from January 2005 to December 2012, CT scans up to day 42 after ictus were evaluated for cerebral infarctions. Different demographic, clinical, laboratory and radiological parameters from admission up to aneurysm treatment were collected with subsequent construction of a risk score from independent infarct predictors.
Results: Seven clinical characteristics were independently associated with cerebral infarction and included in the Risk score (BEHAVIOR Score, 0-11 points): Blood on CT scan according to Fisher grade >2 (1 point), Elderly patients (age>54 years, 1 point), Hunt&Hess grade >3 (1 point), Acute hydrocephalus requiring external liquor drainage (1 point), Vasospasm on initial angiogram (3 points), Intracranial pressure elevation >20 mmHg (3 points) and treatment of multiple aneurysms ("Overtreatment", 1 point). The BEHAVIOR score showed high diagnostic accuracy with respect to the absolute risk for cerebral infarction (area under curve=0.806, p<0.0001) and prediction of poor clinical outcome at discharge (p<0.0001) and after 6 months (p=0.0002).
Conclusions: The proposed BEHAVIOR score is an early and reliable predictor of cerebral infarction after SAH in our cohort. This score identifies patients with poor clinical outcome with high accuracy. Further validation in other SAH cohorts is recommended.
