Can we use Decision Tools to predict Emergent Cardiac Testing and Outcomes among Chest Pain Patients placed in Emergency Department Observation Unit?

Chest pain decision tools (HEART, GRACE, and TIMI) have been used to predict the risk of acute coronary syndrome and their major cardiac event outcomes (MACE) in many studies. However, their use to predict emergent cardiac testing is limited in Emergency Medicine Observation Unit (EDOU) patients. This study aims to 1) identify the role of chest pain decision tools in determining emergent cardiac testing; and 2) further validate their use in the prediction of MACE among EDOU chest pain patients. This is a prospective observational study and included patients placed to EDOU due to chest pain. Cardiac testing included any exercise/dobutamine stress echo, exercise/regadenoson nuclear stress test, treadmill, or invasive coronary angiography. Objective stress testing were ordered by EDOU physician and invasive coronary angiography by cardiologists. All providers were blinded and ordered any cardiac testing at their own discretions. HEART, GRACE, and TIMI scores were calculated and categorized as low or above-low risks. Patients followed up at 6 months upon the index discharge. Results of cardiac testing, EDOU length of stay (LOS), and MACE at 6 months were analyzed and compared with different decision tools. ANOVA was used to compare groups with continuous data and Chi square test was used for categorical data. From 01-2014 until 06-2015, 986 total patients were enrolled. Emergent cardiac testing was performed on 62% of patients. Majority of patients placed to EDOU were deemed low risk by any of the decision tools (85% by HEART, 81% by GRACE, and 80% by TIMI) which also had low abnormal cardiac testing results than patients with above-low risks, especially when HEART tool was used. Average LOS was 33h in patients with emergent cardiac testing versus 26h without. In all low risk patients, the occurrence of MACE in 6 month showed no significant difference regardless of whether cardiac testing was done emergently or not (MACE of 0.31% in low risk HEART patients without versus 0.97% with emergent cardiac testing during index admission, 0.27% versus 0.95% in TIMI, and 0% versus 0.81% in GRACE, p [greater than] 0.05). Using chest pain decision tools may minimize the emergent cardiac testing need among EDOU patients with