Selective dual nuclear scanning in low-risk patients with chest pain to reliably identify and exclude acute coronary syndromes.

STUDY OBJECTIVE: We sought to determine the use in routine clinical practice of selective dual nuclear cardiac scanning (rest and stress) in low-risk patients with chest pain for identifying and excluding acute coronary syndromes (ACSs) during the initial emergency department evaluation.

METHODS: A prospective observational study was conducted over 13 months in 1,775 low-risk patients with chest pain who had intermediate- and high-risk ACSs ruled out by means of our 2-hour protocol, which consists of automated serial 12-lead ECG monitoring in conjunction with baseline and 2-hour creatine kinase (CK) MB and troponin I (cTnI) measurements. At the completion of the 2-hour evaluation period, low-risk patients were stratified by means of physician judgment into 1 of 2 categories: category III, possible ACS; and category IV, probable non-ACS chest pain. Level III patients underwent immediate dual nuclear scanning (rest thallium and stress sestamibi scanning), and level IV patients were discharged directly from the ED unless another serious non-ACS medical condition was thought to exist. Rest and stress scans were interpreted by a board-certified radiologist contemporaneous with patient evaluation. All patients were followed up for 30-day ACS, which was defined as acute myocardial infarction, percutaneous transluminal coronary angioplasty/coronary artery bypass grafting, coronary arteriography revealing stenosis of the major coronary artery of 70% or greater not amenable to percutaneous transluminal coronary angioplasty/coronary artery bypass grafting, life-threatening complication, or cardiac death within 30 days of ED presentation.

RESULTS: A total of 2,206 ED patients with chest pain were evaluated for ACS during the study period. Four hundred thirty-one patients were excluded for having 1 or more of the following findings: initial ECG diagnostic of injury; baseline CK-MB level, cTnI level, or both diagnostic of acute myocardial infarction; 2-hour DeltaCK-MB level of +1.5 ng/mL or greater; 2-hour DeltacTnI level of +0.2 ng/mL or greater; injury or new or evolving ischemia on serial 12-lead ECG monitoring; or clinical diagnosis of ACS. Of the 1,775 study patients, 805 (45.4%) underwent immediate dual nuclear scanning. A positive stress nuclear scan result was more sensitive (97.3% versus 71.2%, P <.0001) and specific (87.7% versus 72.6%, P <.0001) for 30-day ACS than a positive resting nuclear scan result. The protocol of selective dual nuclear scanning (ie, patients who did not undergo dual nuclear scanning were counted as having a negative test result) had a sensitivity and specificity for 30-day ACS of 93.4% and 94.7%, respectively (positive likelihood ratio 17.6; negative likelihood ratio 0.07).

CONCLUSION: Stress nuclear scanning is more sensitive and specific than resting nuclear scanning for identification of ACS in low-risk patients with chest pain. A strategy of using selective dual nuclear scanning once high- and intermediate-risk ACS has been ruled out with our 2-hour evaluation both reliably identifies and reliably excludes 30-day ACS.