Background: This study evaluates the appropriate use of myocardial perfusion imaging (MPI) and determines patterns and variables associated with inappropriate testing.
Methods: Over a 10-month period data were collected prospectively from consecutive patients referred for MPI in four academic departments and an appropriate use grade was assigned (appropriate, uncertain, inappropriate, and unclassifiable scans) according to established criteria.
Results: Among 3,032 referrals appropriate MPI had 72.8% of patients and 19.2% of studies were inappropriate, the remaining being uncertain (7.2%) or unclassifiable (0.8%). In multivariate analysis the asymptomatic status (odds ratio 10.7, P < .001), good functional capacity (odds ratio 1.9, P < .001), an interpretable resting electrocardiogram (odds ratio 1.8, P = .004), an age <65 years (odds ratio 1.5, P = .001) and the absence of diabetes (odds ratio 1.7, P < .001) or dyslipidemia (odds ratio 1.3, P = .014) were independent predictors of inappropriate scintigraphy. The most common indication for inappropriate testing was the assessment of asymptomatic patients <2 years after percutaneous coronary intervention (PCI) (38.9%).
Conclusions: The appropriate use of MPI is relatively high, but a considerable proportion of inappropriate scans is noted which is associated with markers of lower risk. The most common source of inappropriate testing is the assessment of asymptomatic patients <2 years after PCI.