Induced hypothermia and determination of neurological outcome after CPR in ICUs in the Netherlands: results of a survey

Abstract

Introduction: Induction of hypothermia is generally accepted to increase survival of out-of-hospital cardiac arrest, but lack of initiation of this treatment has been frequently reported. When patients remain in coma after treatment with hypothermia, determination of prognosis is difficult. Furthermore, little is known about the methods used in clinical practice to predict outcome after cardiopulmonary resuscitation (CPR). The aim of the present survey was to evaluate self-reported implementation of hypothermia after CPR and the methods used to predict neurological outcome at Intensive Care Units (ICUs) in the Netherlands.

Methods: Between April 2008 and July 2008 an e-mail-invitation for an anonymous web-based 22-question survey was sent to one physician of each ICU in the Netherlands.

Results: Of the 97 physicians surveyed, 74 (76%) responded. Thirty-seven (50%) responders always treated patients with hypothermia after CPR, 31 (42%) only when CPR fulfilled several criteria. The most important reason for not using hypothermia (six ICUs) was lack of equipment. Haemodynamic instability was the most cited reason for discontinuing treatment. Neurological outcome was predicted by clinical neurological examination (92%), cortical N20 responses of median nerve somatosensory evoked potentials (SSEP) (94%), an electroencephalogram (56%) or serum levels of neuron-specific proteins (5%).

Conclusions: In the Netherlands, the use of therapeutic hypothermia after CPR is reported by 92% of ICUs which, compared to previous reports, is an exceedingly high percentage. Neurological outcome is reported to be predicted mainly by neurological examination and SSEP or a combination of these and other assessments. The method used varies substantially between ICUs.