Approximately 15 million patients visit emergency departments (EDs) in the U.S. each year with chest pain or shortness of breath. A convergence of influences compel emergency physicians to over investigate for possible acute coronary syndrome (ACS) and pulmonary embolism (PE), often with imaging tests that impart ionizing radiation. The hypothesis states that computerized, quantitative pretest probability can significantly and safely decrease exposure of low-risk chest pain patients to medical radiation imparted by cardiopulmonary imaging tests that show no clinically significant finding. Quantitative (numeric) pretest probability (PTP) will be assessed for both ACS and PE, using two validated, web-based computer programs (the webtools), which employ the method of attribute matching. Based upon 18 clinical variables, the computer returns patients who share the same in two unique >12,000 patient reference databases (one database each for ACS and PE). Accuracy of the devices is demonstrated in peer review. A pilot randomized trial found that use of the ACS device was associated with an 9.7% absolute reduction in negative cardiopulmonary imaging tests that imparted >5 mSv. The present study will enroll patients with a present or recent history of chest pain and dyspnea from two academic and two community EDs. In year 1, phase I will test the technical reliability of the URL and the webtools, and will test if they produce accurate predictions, particularly at low-range PTP estimates. In years 2-3, phase II will randomize to one of two groups: 1. Intervention group: clinicians and the medical record will receive the printout the ACS and PE programs;or the 2. Standard care group: no output. Pretest probability estimates <2.5% for both ACS and PE result in recommendation for no further testing. For ACS, when the PTP is 2.5-5.5%, the printout suggests exclusion with 2 hour troponin protocol. For PE, when the PTP is 2.5-7.5%, the printout suggests exclusion with a negative high sensitivity D-dimer. All discharged patients will undergo structured telephone and medical record follow up at 14 days using validated methodology to establish presence or absence of ACS, PE and any other significant cardiac or pulmonary diagnosis. Sample size (N=500) will test for a 10% decrease in proportion of patients with no significant cardiopulmonary diagnosis at 14 days who are exposed to >5 mSv within 7 days with a=0.05 and ?= 0.20. Safety will be overseen by an independent DSMB that can stop the study for excessive missed or delayed diagnosis of ACS or PE in either group.
Acute coronary syndrome (ACS) and pulmonary embolism (PE) represent two of the most dangerous causes of chest pain in adult patients in the emergency department (ED). Low-risk ED chest pain patients are often exposed to large doses of radiation imparted by cardiopulmonary imaging that shows no ACS, no PE and no other acute heart or lung disease. This study will test if the presentation of computerized, quantitative pretest probability for ACS and PE is associated with reduced radiation exposure in these patients.
