Anesthesia and analgesia protocol during therapeutic hypothermia after cardiac arrest: a systematic review.

BACKGROUND Present practice guidelines recommend sedative-analgesic and neuromuscular blocking administration during therapeutic hypothermia in comatose patients after cardiac arrest. However, none suggests the best administration protocol. In this study, we evaluated intensivists' preferences regarding administration. METHODS A systematic literature review was conducted to identify clinical studies published between 1997 and July 2009. Selected articles had to meet the following criteria: use of hypothermia to improve neurologic outcome after cardiac arrest, and specific mention of the sedative protocol used. We checked drugs and dose used, the reason for their administration, and the specific type of neurologic and neuromuscular monitoring used. RESULTS We identified 44 studies reporting protocols used in 68 intensive care units (ICUs) from various countries. Midazolam, the sedative used most often, was used in 39 ICUs at doses between 5 mg/h and 0.3 mg/kg/h. Propofol was used in 13 ICUs at doses up to 6 mg/kg/h. Eighteen ICUs (26%) did not report using any analgesic. Fentanyl was the analgesic used the most, in 33 ICUs, at doses between 0.5 and 10 microg/kg/h, followed by morphine in 4 ICUs. Neuromuscular blocking drugs were routinely used to prevent shivering in 54 ICUs and to treat shivering in 8; in 1 ICU, their use was discouraged. Pancuronium was used the most, in 24 ICUs, followed by cisatracurium in 14. Four ICUs used neuromuscular blocking drug administration guided by train-of-four monitoring and 3 ICUs used continuous monitoring of cerebral activity. CONCLUSIONS There is great variability in the protocols used for anesthesia and analgesia during therapeutic hypothermia. Very often, the drug and the dose used do not seem the most appropriate. Only 3 ICUs routinely used electroencephalographic monitoring during paralysis. It is necessary to reach a consensus on how to treat this critical care population.