Abstract 13846: Utility of Early Empirical Triage Risk Assessment in Acute Heart Failure (HF): Evaluation of the Canadian Triage Acuity Scale (CTAS)
Background: The initial triage and management of acute HF in the emergency department (ED) is often empiric, because there are few validated risk assessment tools. The CTAS is used at virtually all Canadian EDs, and assigned scores that correspond to perceived patient acuity: 1=resuscitation, 2=emergent, 3=urgent, 4=less urgent, and 5=non-urgent.
Methods: Using the National Ambulatory Care Reporting System (NACRS) database, we examined the predictive validity and impact of the CTAS on acute outcomes in 68,380 adults with HF (ICD-10 code I50, mean age 76±12 yrs) who presented to an ED in Ontario (Apr 2003-Mar 2007). A subset of 3303 patients was further examined via detailed chart review.
Results: Patients with greater acuity CTAS levels (score range: 1[highest], 2, 3, 4–5[lowest]) had higher heart rate (106±29, 89±24, 84±20, 82±17 beats/min, p-trend <0.001), higher respiratory rate (29±9, 23±7, 21±5, 20±6 breaths/min, p<0.001), and higher systolic blood pressure (159±36, 148±31, 144±27, 142±25 mmHg, p<0.001). Multivariable predictors of the highest CTAS score included the following: respiratory rate >24 (odds ratio [OR] 1.96, 95%CI; 1.05–3.67), oxygen saturation <90% (OR 5.92, 95%CI; 3.09–11.81), NYHA class IV (OR 5.41, 95%CI; 2.74–11.38), and transport by emergency medical services (OR 3.52, 95%CI; 1.70–8.02) with c-statistic=0.913. Mortality rates were 9.9%, 1.9%, 0.9%, and 0.5% at 1 day and 17.2%, 5.9%, 3.8%, and 2.5%, at 7 days, for CTAS 1, 2, 3, and 4–5, respectively. Although age/sex-adjusted CTAS level had high discrimination for death in the ED (c=0.817), its performance was significantly lower than multivariable-adjusted models that accounted for cardiac/non-cardiac conditions and ED length of stay (c=0.882, p<0.001). Performance of age/sex-adjusted CTAS for mortality decreased with increasing time horizon, with c-statistics for the CTAS vs. multivariable-adjusted models as follows: 1-day (0.724 vs. 0.810, p<0.001), 7-days (0.680 vs. 0.746, p<0.001).
Conclusions: An empiric, triage acuity score assigned in initial ED evaluation of HF was associated with early mortality. Prediction of events occurring after the ED phase was attenuated, emphasizing the need for predictive tools to guide early care of acute HF patients.
- © 2010 by American Heart Association, Inc.