Author + information
- David E. Winchester,
- Ryan Meral,
- Ryan Chauffe,
- Scott Ryals,
- Raman Dusaj,
- Rebecca J. Beyth and
- Leslee Shaw
Millions of myocardial perfusion imaging (MPI) tests are performed annually in the United States. Professional societies have developed appropriate use criteria (AUC) to inform clinicians about when patients are most likely to benefit from MPI. Prior reports indicate that 7–24% of MPI are ordered in inappropriate situations, fueling concern about overuse. Profit motives and liability concerns are postulated reasons for overuse. These motives are diminished in the Veteran's Affairs (VA) health system, where appropriate use is not well described. This setting may alter ordering patterns for MPI.
We conducted a retrospective cross–sectional study in a single academic VA center. Outcomes included the distribution of inappropriate MPI ordering and factors associated with inappropriate MPI ordering.
332 consecutive subjects underwent MPI between December 1, 2010 and April 11, 2011. Median body mass index was 32.3. Coronary disease was observed in 44.6% and prevalence of cardiovascular risk factors was high: diabetes 39.2%, hypertension 81.0%, and hyperlipidemia 78%. Chest pain (49.4%) and dyspnea (38.3%) were the most common symptoms. Of these subjects, 259 (78%) of MPI were appropriate, 42 (13%) were inappropriate, 27 (8%) were uncertain. Appropriateness could not be determined for 4 studies (1%), including 2 undergoing non–heart organ transplant evaluation. The most commonly observed inappropriate indications related to pre–operative evaluation and testing of low risk/asymptomatic patients. Absence of symptoms (odds ratio [OR] 4.80, 95% confidence interval [CI] 2.39–9.66, p<0.001) was associated with greater likelihood of an inappropriate test while diabetes (OR 0.37, 95% CI 0.17–0.80, p=0.01) and chest pain (OR 0.07, 95% CI 0.02–0.20, p<0.001) were associated with a lower likelihood of an inappropriate test.
Despite the diminished impact of profit motives and liability risk in the VA setting, we observed inappropriate use of MPI similar in magnitude and pattern to reports from non–VA settings. Diabetes mellitus and chest pain were associated with appropriate MPI while the absence of symptoms was robustly associated with inappropriate MPI.
Special Session North, Room 120
Monday, March 11, 2013, 11:45 a.m.–Noon
Session Title: Young Investigator Awards Competition: Cardiovascular Health Outcomes and Population Genetics
Abstract Category: Cardiovascular Health Outcomes and Population Genetics
Presentation Number: 409–8
- 2013 American College of Cardiology Foundation