ABSTRACT
The American Heart Association recommends therapeutic hypothermia for comatose patients with return of spontaneous circulation after out-of-hospital ventricular fibrillation cardiac arrest. While there is a growing body of evidence for the general efficacy of therapeutic hypothermia, the individualized benefit of therapy is not currently predictable. Ninety-one consecutive patients, from April 2011 to July 2014, were treated at the University of Kentucky Medical Center with the therapeutic hypothermia protocol. Medical records were reviewed retrospectively. Data, such as preexisting comorbidities, cardiac arrest characteristics, and hospital course, were used to compose a multivariate logistic regression with mortality serving as the primary endpoint. The overall in-hospital mortality was 64% (n = 58) in this group. The arrest was considered cardiac etiology in 84% (n = 76) of patients, of which 49% (n = 45) were classed as ventricular fibrillation and 9% (n = 8) as ventricular tachycardia. The presence of a shockable rhythm, as well as shorter duration of cardiac arrest, was associated with increased survival, whereas time to target temperature was not. The presence of a preexisting neurologic disease was associated with a 10-fold increase in estimated odds of mortality. Age, serum lactate, ionized calcium, arterial pH, estimated glomerular filtration rate, and APACHE score were all predictors of mortality. Cardiac arrest is a devastating condition with a high mortality rate. Given the limited resources of the resuscitation community, the ability to predict survivors based on routinely obtained measures upon admission would be of tremendous value. In this study, we show a series of admission parameters that demonstrate predictive ability in identifying patients more likely to survive with therapeutic hypothermia.
