[Validation of the prognostic accuracy of neurosurgical admission scales after rupture of cerebral aneurysms].

The course of 185 patients operated for a ruptured intracranial aneurysm at the University of Saarland between 1991 and 1993 has been followed up. The main emphasis of the investigation was placed on the scrutiny of the coma scales on admission (Hunt & Hess-Scale, Glasgow Coma Scale and WFNS-Scale [= World Federation of Neurological Surgeons]) with regard to the outcome. Outcome was defined as the patients' state six months after aneurysm rupture according to the Glasgow Outcome Scale. The sensitivities, specifities and predictive values of almost all scale grades were poor. The Hunt&Hess-Scale was the one with the best correlation. By half the patients with the worst scale grades on admission had a good outcome. A gradation of the outcome with regard to the middle admission grades has not been identified. These observations have been demonstrated by using ROC (Receiver Operating Characteristic)-curves. The admission scales are not suitable to give a definite prognosis and do not justify any decision neither pro nor contra an operation. A lot of parameters besides the neurological findings have an effect on the prognosis. Additionally, unexpected complications may occur in the pre- and postoperative phase. The score values determined at the day of operation have shown a more precise prognosis than the values determined immediately after hospitalization. Therefore the evaluation of the most relevant phase could improve the prognostic value of the scales.