Importance of a reliable admission Glasgow Coma Scale score for determining the need for evacuation of posttraumatic subdural hematomas: a prospective study of 65 patients.

Background: Patients who have an acute subdural hematoma with a thickness of 10 mm or less and with a shift of the midline structures of 5 mm or less often can be treated nonoperatively. We wonder whether the knowledge of the clinical status both in the prehospital determination and on admission to the neurosurgical center can predict the need for evacuation of subdural hematomas as well as the computed tomographic (CT) parameters.
Methods: From January 1, 1994, to May 31, 1996, 65 comatose patients harboring an acute subdural hematoma of 5 mm or more and not brain dead were admitted to our intensive care unit. Of the 65 patients, 15 patients were initially managed conservatively according to a protocol based on clinical, CT, and intracranial pressure parameters. During the study period, the use of long-lasting paralytic agents has been eliminated to allow detection of clinical deterioration in the Glasgow Coma Scale (GCS) score from the prehospital determination to the hospital admission assessment.
Results: Of the 15 patients initially managed conservatively, two were subsequently operated on because of evolving parenchymal hematomas. When comparing demographic, clinical, and CT parameters between the surgical group of patients and the patients initially conservatively treated, hematoma thickness (mean, 17.1 mm vs. 7.5 mm, p < 0.0001) and shift of the midline structures (mean, 12.8 mm vs. 4.7 mm, p < 0.008) were predictive of the need for surgery. A statistically significant change in the GCS score between prehospital determination and admission assessment was shown in the surgical group of patients (mean GCS score, 8.4 vs. 6.7, p < 0.01), and it was not present (mean GCS score, 7.3 vs. 7.2) in the patients initially conservatively treated. Functional outcomes were present in 23 cases (35.4%); functional outcomes in the initially conservatively treated patients were reached by 10 patients (66.7%).
Conclusions: Nonoperative management for selected cases of acute subdural hematomas is at least as safe as surgical management. GCS scoring at the scene and in the emergency room combined with early and subsequent CT scanning is crucial when making the decision for nonoperative management. This strategy requires that administration of long-lasting sedatives and paralytic medications be avoided before the patient arrives at the neurosurgical center.