Abstract
Background. Therapeutic hypothermia (TH) improves survival for patients with out-of-hospital (OOH) cardiac arrest. Largely studied in witnessed VT/VF populations, recent data may support broader application for other forms of arrest. Many institutions are electing to expand their inclusion criteria in response. However, no studies have described changes in patient characteristics, resource use, or outcomes that may result with such expansion. Methods. We studied consecutive patients receiving TH at UNC Hospital from Sep 2009 - Jan 2012. Protocol #1 provided TH for witnessed OOH arrest with VT/VF only. In Nov 2010, criteria were expanded to include all cardiac rhythms and un-witnessed events ( protocol #2 ). Baseline demographics, resource use, and survival were compared between protocol groups. Multivariable logistic regression was used to identify key variables associated with in-hospital mortality. Results. The arrest population cooled at our institution had numerous comorbidities. The median age was 57 (IQR 49,70) years and median BMI was 28 (IQR 25,32) kg/m 2 . The majority were male (55%), white (70%), and had history of hypertension (57%). In addition, there was a high prevalence of baseline diabetes (28%), CAD (34%), chronic kidney disease (19%), and heart failure (28%). Despite nominally greater representation of male patients (63 v. 41%) in the expanded protocol population, there were no significant differences in baseline characteristics between protocol #1 and #2. Most arrest events were witnessed, but there were more unwitnessed episodes with the protocol #2 (30 v. 0%, p =0.01). The initial cardiac rhythm was most commonly VT/VF, but there were greater rates of PEA (6 v. 17%, p=0.01) with protocol #2. This was a resource intensive population, with all patients mechanically ventilated, 1/3 receiving cardiac catheterization and anti-arrhythmic drug therapy, and 2/3 treated with vasopressor medications; however, only the rate of aspirin use differed among protocol groups (35 v. 67% for protocol #2, p=0.04). Overall, 68% of all patients died in the hospital. While there was higher mortality with the expanded protocol, this did not reach statistical significance (59 v. 73%, p=0.31). Not a single unwitnessed arrest survived, and age was also independently associated with mortality (OR per 10y = 2.52, 95% CI 1.26-5.05, p=0.009). Conclusions. No matter the inclusion criteria, patients who receive TH for OOH cardiac arrest have multiple comorbidities and consume numerous resources. However, there were no significant differences when comparing our two protocol populations. Mortality does seem to be increasing with broader application of TH, and future attention should focus on the unwitnessed arrests in particular, none of whom survived their hospitalization.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call