The use of nuclear myocardial perfusion imaging (MPI) dropped by half from 2006 to 2011 within a Kaiser Permanente healthcare system in California, but researchers can't quite point to specific reasons for the decline, according to a study published in the March 26 issue of the Journal of the American Medical Association.
What Kaiser researchers do know is that the decrease in MPI utilization is not due to increased use of other imaging modalities.
"It could've been that an alternative imaging modality like CT angiography became the dominant player, and there was a dramatic shift to stress echocardiography," lead author Dr. Edward McNulty, chief of cardiovascular services at Kaiser Permanente San Francisco Medical Center, told AuntMinnie.com. "But we looked at both of those [modalities] in this population, and we could not explain the decline on that basis."
The abrupt drop in MPI utilization might indicate a change in physicians' attitudes toward the imaging test; however, incident coronary disease, using myocardial infarction as a benchmark, also declined, the authors wrote (JAMA, March 26, 2014, Vol. 311:12, pp. 1248-1249).
End of the boom
The sharp turnaround in 2006 followed a spirited growth cycle during which nuclear MPI accounted for many of the cardiac imaging procedures performed from the 1990s through the middle 2000s.
"Everyone has observed anecdotally that myocardial perfusion imaging use was declining, but most of the data were from industry surveys or Medicare Part B over the age of 65," McNulty said. "That, to me at least, seemed counter to the continuing trend that cardiac imaging use was increasing, because that is not what those of us in the field were observing."
McNulty and colleagues reviewed MPI use from 2000 to 2011 for patients 30 years of age and older in the clinical databases of Kaiser Permanente Northern California. The integrated healthcare delivery system provides inpatient and outpatient care for more than 2.3 million adults.
To assess the potential use of other imaging modalities instead of MPI, the researchers also estimated annual rates of cardiac CT and stress echocardiography from 2007 to 2011. Use of perfusion PET and perfusion MRI were negligible during the study period.
MPI utilization
Overall, MPI was used for 302,506 Kaiser Permanente members (during 23.2 million person-years of follow-up) at 19 facilities. From 2000 through 2006, MPI use increased by 41%. Utilization peaked in 2006 and then decreased by 51% by 2011 -- falling by half in just five years.
Declines in MPI scans were greater among outpatients (58%) than inpatients (31%), as well as for persons younger than 65 years (56%) versus those older than 65 (47%).
"You'll notice that there is less use in younger people and there is less use for outpatients," McNulty said. "Those are persons who have lower risk for coronary disease in general. To us, that is a signal that MPI is being used more selectively for persons who are at greater pretest risk of having coronary disease."
The reduction in MPI did not correspond to an increase in the use of other imaging modalities, the researchers found. Stress echocardiography use was unchanged, with 189 tests per 100,000 person-years in 2007 and 182 per 100,000 person-years in 2011.
Cardiac CT use (tests per 100,000 person-years) increased from 37 in 2007 to 73 in 2011, but even that increase accounted for only 5% of the decline in overall MPI use, the authors noted.
During the period of decreasing MPI utilization, incidents of myocardial infarction also fell by 27% in the study population, from 286 events per 100,000 person-years to 208 events.
"The decline is so abrupt that it seems to argue it was more a change in [physician] behavior than disease prevalence," McNulty said. "Coronary disease, if you use myocardial infarction as the mark, has been declining gradually during the decade, so it is hard for us to attribute the decline to coronary disease."
The utilization decline in MPI since 2011 is beginning to level off, McNulty said.
Why the decline?
So what led to the drop? It could be due to the various forces trying to limit utilization -- or maybe not.
"There are a lot of factors that can potentially limit the use of cardiac imaging that began to emerge in the 2000s, not the least of which was the use of radiology benefits managers and declining reimbursement," McNulty said.
However, Kaiser Permanente is an integrated healthcare delivery network and does not need to obtain third-party approval for tests, nor does it employ radiology benefits managers. So there is no direct financial incentive for Kaiser physicians to perform or not to perform tests, McNulty explained.
"This is one of those studies that describes an interesting behavior phenomenon, but it raises a lot of other important questions," he added.
To answer some of those questions, McNulty and colleagues plan to examine whether less use of MPI actually increases the diagnostic yield of tests that are performed. They also want to more formally assess a patient's pretest risk in conjunction with appropriate use criteria to see if tests are being applied more appropriately.
"All these tests are wonderful tools; I think imaging has been one of the reasons we have started to conquer coronary disease," McNulty said. "But at the same time, every time we perform a test, there is also the potential for more tests to be performed downstream. You have to recognize the positives and the negatives in these tests."
One of the study's limitations is that the findings apply only to the Kaiser environment, but McNulty said that medicine is headed in this direction with the proliferation of accountable care organizations.
"I think these results inform a larger dialogue about how we are going to control use in the future as we have more older people, who potentially will consume more resources," he said. "We may have to look at different delivery systems that are able to manage the growth."
