Cerebral performance category at hospital discharge predicts long-term survival of cardiac arrest survivors receiving targeted temperature management*.

OBJECTIVE: Despite recent advancements in post-cardiac arrest resuscitation, the optimal measurement of postarrest outcome remains unclear. We hypothesized that Cerebral Performance Category score can predict the long-term outcome of postarrest survivors who received targeted temperature management during their postarrest hospital care.
DESIGN: Retrospective chart review.
SETTING: Two academic medical centers from May 2005 to December 2012. PATIENTS: The medical records of 2,417 out-of-hospital and in-hospital patients post cardiac arrest were reviewed to identify 140 of 582 survivors who received targeted temperature management.
INTERVENTIONS: None.
MEASUREMENTS AND MAIN RESULTS: The Cerebral Performance Category scores at hospital discharge were determined by three independent abstractors. The 1-month, 6-month, and 12-month survival of these patients was determined by reviewing hospital records and querying the Social Security Death Index and by follow-up telephone calls. The association of unadjusted long-term survival and adjusted survival with Cerebral Performance Category was calculated. Of the 2,417 patients who were identified to have undergone cardiac arrest, 24.1% (582/2,417) were successfully resuscitated, of whom 24.1% (140/582) received postarrest targeted temperature management. Overall, 42.9% of patients (60/140) were discharged with Cerebral Performance Category 1, 27.1% (38/140) with Cerebral Performance Category 2, 18.6% (26/140) with Cerebral Performance Category 3, and 11.4% (16/140) with Cerebral Performance Category 4. Cerebral Performance Category 1 survivors had the highest long-term survival followed by Cerebral Performance Categories 2 and 3, with Cerebral Performance Category 4 having the lowest long-term survival (p < 0.001, log-rank test). We found that Cerebral Performance Category 3 (hazard ratio = 3.62, p < 0.05) and Cerebral Performance Category 4 (hazard ratio = 12.73, p < 0.001) remained associated with worse survival after adjusting for age, gender, race, shockable rhythm, time to targeted temperature management initiation, total duration of resuscitation, withdrawal of care, and location of arrest.
CONCLUSION: Patients with different Cerebral Performance Category scores at discharge have significantly different survival trajectories. Favorable Cerebral Performance Category at hospital discharge predicts better long-term outcomes of survivors of cardiac arrest who received targeted temperature management than those with less favorable Cerebral Performance Category scores.