Introduction: Clinicians are urged to decrease radiation exposure from unnecessary medical procedures. Many emergency department (ED) patients placed in an observation unit (EDOU) do not require chest pain evaluation with a nuclear stress test (NucST). We sought to implement a simple ST algorithm that favors non-nuclear stress test (Non-NucST) options to evaluate the effect of the algorithm on the proportion of patients exposed to radiation by comparing use of NucST versus Non-NucST pre- and post-algorithm.

Methods: An ST algorithm was introduced favoring Non-NucST and limiting NucST to a subset of EDOU patients in October 2008. We analyzed aggregate data before (Jan-Sept 2008, period 1) and after (Jan-Sept 2009 and Jan-Sept 2010, periods 2 and 3 respectively) algorithm introduction. A random sample of 240 EDOU patients from each period was used to compare 30-day major adverse cardiac events (MACE). We calculated confidence intervals for proportions or the difference between two proportions.

Results: A total of 5,047 STs were performed from Jan-Sept 2008-2010. NucST in the EDOU decreased after algorithm introduction from period 1 to 2 (40.7%, 95% CI [38.3-43.1] vs. 22.1%, 95% CI [20.1-24.1]), and remained at 22.1%, 95% CI [20.3-24.0] in period 3. There was no difference in 30-day MACE rates before and after algorithm use (0.1% for period 1 and 3, 0% for period 2).

Conclusion: Use of a simple ST algorithm that favors non-NucST options decreases the proportion of EDOU chest pain patients exposed to radiation exposure from ST almost 50% by limiting NucST to a subset of patients, without a change in 30-day MACE.