Neurological prognostication after cardiac arrest and targeted temperature management 33°C versus 36°C: Results from a randomised controlled clinical trial.

BACKGROUND The reliability of some methods of neurological prognostication after out-of-hospital cardiac arrest has been questioned since the introduction of induced hypothermia. The aim of this study was to determine whether different treatment temperatures after resuscitation affected the prognostic accuracy of clinical neurological findings and somatosensory evoked potentials (SSEP) in comatose patients. METHODS We calculated sensitivity and false positive rate for Glasgow Coma Scale motor score (GCS M), pupillary and corneal reflexes and SSEP to predict poor neurological outcome using prospective data from the Target Temperature Management after Out-of-Hospital Cardiac Arrest Trial which randomised 939 comatose survivors to treatment at either 33 °C or 36 °C. Poor outcome was defined as severe disability, vegetative state or death (Cerebral Performance Category scale 3-5) at six months. RESULTS 313 patients (33%) were prognostically assessed; 168 in the 33 °C, and 145 in the 36 °C group. A GCS M ≤ 2 had a false positive rate of 19.1% to predict poor outcome due to nine false predictions. Bilaterally absent pupillary reflexes had a false positive rate of 2.1% and absent corneal reflexes had a false positive rate of 2.2% due to one false prediction in each group. The false positive rate for bilaterally absent SSEP N20-peaks was 2.6%. CONCLUSIONS Bilaterally absent pupillary and corneal reflexes and absent SSEP N20-peaks were reliable markers of a poor prognosis after resuscitation from out-of-hospital cardiac arrest but low GCS M score was not. The reliability of the tests was not altered by the treatment temperature.