Yield of nuclear scan strategy in chest pain unit evaluation of special populations

Abstract

Patients with chest pain (CP) and nondiagnostic ECG represent heterogeneous population in whom the evaluation of coronary risk factors including metabolic syndrome (MetS) and diabetes mellitus (DM) might improve risk stratification. We enrolled 798 consecutive CP patients; 14% presented with MetS and 10% with DM; the remaining 76% presented with other coronary risk profiles (others). All patients underwent maximal exercise tolerance test (ETT) and myocardial perfusion imaging (exercise-MPI). Those with positive testing underwent angiography, whereas the remaining patients were discharged and later followed up. Primary end-point was a composite of coronary stenoses greater than or equal to 50% documented by angiography or coronary events at follow-up. Patients with MetS or DM had significantly lower survival free from end-point than those patients without (P<0.001). Exercise-MPI showed high negative predictive value in MetS, DM, and others (>96%); however, positive predictive value was 69, 74, and 52%, respectively (P<0.05). ETT alone showed negative predictive value (88%) which was significantly lower than exercise-MPI (98%), (MetS vs. others: P<0.001, and DM vs. others: P=0.05). The area under the receiver-operating characteristic curves obtained from the multivariate model includes clinical data alone, clinical data and ETT results, or clinical data and exercise-MPI results increase progressively. A nuclear scan strategy in special populations, including CP patients with MetS or DM, is a valuable tool for risk stratification and adds incremental prognostic value over clinical and ETT values.