Objective:
To investigate the association between inappropriate use of myocardial perfusion imaging (MPI), specialty training of the ordering provider, and location of the clinical encounter.
Methods:
We conducted a retrospective analysis of MPI performed in a single Veterans Affairs (VA) medical center from August 2010 through November 2011. We classified the indication for each MPI test using the 2009 Appropriate Use Criteria (AUC). We investigated for associations between MPI ordered for inappropriate indications, the subspecialty training of the ordering provider (cardiology vs. other), and the location of the clinical encounter (emergency department [ED], inpatient, or outpatient).
Results:
We analyzed 598 subjects, 95% of whom were male. Diabetes (41.2%), hypertension (82.3%), and hyperlipidemia (76.0%) were common, as were prior myocardial infarction (40.5%) and prior revascularization (34.6%). Overall, 78.4% of MPI were appropriate, 9.5% uncertain, and 10.5% were inappropriate. An indication could not be determined for 3.2% of patients. Distribution of appropriateness by provider type and location of encounter are demonstrated in the Figure. The most common inappropriate indication did differ between groups. For Cardiology providers the most common inappropriate MPI was for asymptomatic patients within 2 years of percutaneous coronary intervention (AUC indication #59) while for noncardiology providers, the most common was asymptomatic patients with intermediate coronary heart disease (CHD) risk and a normal electrocardiogram (AUC indication #13). For inpatient MPI, definite acute coronary syndrome (AUC indication #10) was the most common inappropriate test, while for outpatient MPI, asymptomatic patients with low CHD risk was the most common (AUC indication #12), and for ED MPI syncope with low CHD risk was the most common (AUC indication #20).
Conclusions:
The proportion of MPI ordering for inappropriate indications was similar when comparing the specialty training of the ordering provider and the location of the ordering encounter. The most common inappropriate indications ordered by each group, however, were different. Our findings suggest that initiatives to reduce inappropriate use should be aimed at all providers, but targeted to specific inappropriate indications.
A negative preoperative 99mTechnetium tetrofosmin myocardial perfusion imaging scan with postoperative acute coronary syndrome - A case of balanced ischaemia
