Potential cost effectiveness of initial myocardial perfusion imaging for assessment of emergency department patients with chest pain.

Previous investigations have confirmed the diagnostic and predictive usefulness of initial single-photon emission computed tomography (SPECT) myocardial perfusion imaging using technetium-99m sestamibi in the evaluation of emergency department patients with chest pain. Patients with a normal SPECT perfusion scan performed during chest pain have an excellent short-term prognosis, and may be candidates for expeditious cardiac evaluation or outpatient management. However, there are limited data regarding the cost effectiveness of this technique. This analysis models the potential cost effectiveness of this procedure. In the current investigation we compared 2 model strategies for management of emergency department patients with typical chest pain and a normal or nondiagnostic electrocardiogram (ECG). In 1 model strategy, (the technetium-99m sestamibi SPECT myocardial perfusion imaging [SCAN] strategy), the decision whether to admit or discharge a patient from the emergency department is based on results of initial technetium-99m sestamibi SPECT myocardial imaging. Patients with normal scans are discharged; others are admitted. In the second model strategy, (the NO SCAN strategy), the decision whether or not to admit a patient is based on a combination of clinical and electrocardiographic variables. Patients with > or = 3 cardiac risk factors or an abnormal ECG are admitted; others are discharged. Adverse cardiac events were prospectively defined as cardiac death, nonfatal myocardial infarction, or the need for acute coronary intervention. Costs were assigned using data derived from 102 patients who underwent SPECT myocardial perfusion imaging and an additional 107 emergency department patients with ongoing chest pain who either underwent or were eligible for initial SPECT myocardial perfusion imaging. Mean (+/- SE) costs were highest among hospital admitted patients who experienced an adverse cardiac event ($21,375 +/- $2,733) and lowest in patients discharged from the emergency department ($715 +/- 71). Mean costs per patient of the SCAN strategy and NO SCAN strategy were $5,019 versus $6,051, respectively. These results were stable in a sensitivity analysis across a range of costs and predictive values. Thus, the SCAN model strategy for initial management of emergency department patients with typical ongoing angina and a normal or nondiagnostic ECG using initial myocardial perfusion imaging with technetium-99m sestamibi appears to be safe, accurate, and potentially cost effective. Validation of these preliminary retrospective observations will require further prospective investigation.