Recent Trends Suggest Possible Inappropriate Utilization of Myocardial Perfusion Imaging

The aim of this study was to analyze the utilization of elective stress nuclear myocardial perfusion imaging (MPI) in the Medicare population.

Nationwide Medicare Part B fee-for-service databases for 2004 to 2016 were reviewed. Current Procedural Terminology codes for stress MPI were selected: standard planar and single-photon emission computed tomography (STD) and PET. Utilization rates per 1,000 Medicare beneficiaries were calculated. Elective examinations were identified using place-of-service codes for private offices and hospital outpatient departments (HOPDs). Medicare physician specialty codes identified the performing physician. Because Medicare Part B databases are complete population counts, sample statistics were not required.

Elective STD MPI utilization peaked in 2006 at 74 studies/1,000 and had declined by 36% by 2016. Cardiologists' share of STD MPI grew from 79% to 87% between 2004 and 2016. Cardiologists perform STD MPI primarily in private offices, where utilization peaked in 2008 and then demonstrated an absolute decline of 28 studies/1,000 by 2016. During this same time period, cardiologists' use of STD MPI in HOPDs demonstrated an absolute increase of 8.1 studies/1,000. From 2004 to 2016, STD MPI use by radiologists declined by 58%. Elective PET MPI maintained an upward trend, reflecting increasing use by cardiologists in private offices.

Elective STD MPI use is declining, but cardiologists are performing an increasing share in outpatient settings. The drop in private office STD MPI among cardiologists was far greater than the corresponding increase in its use in HOPDs, suggesting that many studies previously performed in private offices were unindicated. Self-referred PET MPI utilization has rapidly grown in cardiology private offices.