A Cohort Study of Myocardial Perfusion Imaging in Veteran Patients Without Symptoms: Contributing Factors and Results of Testing

Abstract

Myocardial perfusion imaging (MPI) is commonly used to detect ischemia. Concerns about silent ischemia may encourage orders for MPI in asymptomatic patients. Factors contributing to this practice are poorly described and the clinical utility is questionable.We conducted a single center retrospective cohort investigation on Veterans who underwent MPI between December 2010 and July 2011. We gathered data on symptoms, baseline characteristics, results of MPI, and cardiovascular events within 1 year. MPI were categorized using 2009 appropriate use criteria (AUC).Of 592 patients, 127 (21.5%) had no symptoms at the time of MPI. Comparing symptomatic and asymptomatic patients, no differences were observed in baseline characteristics except abnormal ECG, more common in asymptomatic patients (n = 86, 67.7% vs. n = 232, 49.9% for symptomatic patients, P < 0.0001). Asymptomatic MPI were more commonly inappropriate (n = 26, 21.5% vs. n = 31, 6.7% for appropriate/uncertain, P < 0.0001). Detection of ischemia between patients with and without symptoms was not different (P = 0.86); however, among asymptomatic MPI that also demonstrated ischemia, none were inappropriate (n = 10 appropriate, n = 7 uncertain). In multivariate regression, 2 factors were associated with asymptomatic status, abnormal ECG (odds ratio [OR] 2.29, 95% confidence interval [CI] 1.5-3.49) and age over the median (OR 0.63, 95% CI: 0.41-0.95).A substantial portion of MPI tests are ordered for patients without symptoms. When compared to symptomatic patients, MPI for asymptomatic patient were more commonly inappropriate; however, the prevalence of ischemia was similar. MPI may be clinically relevant in some asymptomatic patients and decisions to test should be based on the AUC.

FIGURE 1

Distribution of appropriateness between symptomatic and asymptomatic patients. This stacked bar graph displays the proportion of nuclear test appropriateness ratings. Approximately half (52.8%) of nuclear tests for patients without symptoms were rated as appropriate, compared to 86.7% for those with symptoms.