The prognostic value of early transcranial doppler ultrasound following cardiopulmonary resuscitation

The aim of the presented study was to estimate the prognostic value of transcranial Doppler sonography (TCD) for the prediction of clinical outcome of patients after cardiopulmonary resuscitation (CPR) on the basis of the measured flow velocity changes in the basal cerebral arteries. Thirty-nine patients (27 men, 12 women) aged 66 ± 15 y (±SD) who had undergone CPR were involved. Serial TCD examinations of the intracranial arteries were performed 1.5, 4, 8, 16, 24 and 72 h after CPR. Plasma neuron specific enolase (NSE), ventilation parameters (pH, paO2, paCO2) and hemodynamic parameters were registered. Seventeen patients (group 1) survived with moderate or without neurologic deficits. Twenty-two patients (group 2) either died (n = 21) within 9 ± 14 days or remained in vegetative state (n = 1). NSE levels ranged from 9 to 29 μg/L in group 1 and from 22 to 1242 μg/L in group 2 (p < 0.05). Four and 72 h after CPR, peak systolic flow velocities in the middle cerebral artery (MCA) were significantly higher in group 1 than in group 2 (p < 0.05). Twenty-four h after CPR, peak systolic and diastolic flow velocities in the ACA and PCA were also significantly higher in group 1 than in group 2 (p < 0.05). At this time, patients of group 2 showed significantly higher resistance index-values (RI = (sys-dia)/sys) in the anterior cerebral artery (ACA) and the posterior cerebral artery (PCA) (p < 0.05). A high correlation between peak systolic blood flow velocity in the MCA and systemic systolic blood pressure was observed in group 2 early 4 to 16 h after CPR (r = +0.52 to + 0.81, p < 0.05), while there was no such correlation in group 1. Using serial TCD examinations, patients with severely disabling or fatal outcome could be identified within the first 24 h. Besides established clinical and laboratory parameters, postanoxic myoclonus and NSE, serial TCD examinations following CPR may be helpful to predict the clinical outcome, but further studies with a larger number of patients are necessary to approve this hypothesis. (E-mail: tiemo.wessels@uni-duisburg-essen.de)