Imaging Diabetics in the U.S.: Cost Remains PETs Final Barrier

	Source: Siemens Healthcare

Diagnosing at-risk patients

There are 23.6 million diabetics in the U.S., or 8 percent of the population, with 1.6 million new cases diagnosed in those 20 years and older each year, according to the American Diabetes Association. However, of the 23.6 million cases, 5.7 million are classified as undiagnosed.

“It is particularly important to stratify risk in patients who potentially have diabetes, as they are at higher risk of cardiac events than the general population,” explains Mark Travin, MD, a non-invasive cardiologist at Montefiore Medical Center in the Bronx, N.Y. “Often, their initial presentation of the disease is MI [myocardial infraction] or sudden death.” Studies have shown that the risk of MI or cardiac death is particularly high for female diabetics.  

“In the general population, screening is quite controversial due to its low yield, but due to the high risk among diabetics, screening may be warranted due to the higher yield of substantive findings,” Travin says. However, he acknowledges that screening tends to be costly, and incidental findings could lead physicians to make recommendations that aren’t necessary, due to defensive medicine.

The Detection of Ischemia in Asymptomatic Diabetics (DIAD) trial was designed to preemptively identify diabetics who could potentially develop more complicated cardiac symptoms through nuclear perfusion imaging, says Kim Allan Williams, MD, director of nuclear cardiology at the University of Chicago in Illinois. The trial randomly assigned 561 patients with type 2 diabetes to be screened for coronary artery disease (CAD) with SPECT MPI (myocardial perfusion imaging) and 562 similar patients to not be screened, finding that 20 to 25 percent of the patients had normal SPECT studies (JAMA 2009; 301(15):1547-1555). The cardiac event rates were low and were not significantly reduced by MPI screening for myocardial ischemia over 4.8 years.

“To our simultaneous amazement and disappointment, early screening for diabetics without known CAD did not help based on the results of the DIAD trial,” Williams says. However, he adds that the trial cannot offer “the final word,” as the trial was limited by its own design, including its older patient population and the widespread use of statins.

Which modality is best?

The most recent clinical guidelines specify that SPECT MPI is considered the appropriate test in high-risk diabetics.

“SPECT allows us to assess which of these patients will likely have an adverse event, such as myocardial infarction or cardiac death,” says Regina Druz, MD, director of nuclear cardiology at North Shore University Hospital on Long Island, N.Y. “Usually, more than a 2 percent event rate per year is considered significant. Patients with normal studies have a low event rate, and can continue to be treated with optimal medical therapy and lifestyle modifications.” However, for diabetics with triple vessel disease—which is an extremely high-risk subgroup—the gold-standard evaluation is a coronary angiogram.

For diabetics, Travin says that some of the newer techniques with calcium scoring and CT angiography are promising, but more research is needed to further prove their benefits. Most studies indicate that the physiologic assessment of SPECT MPI is a superior risk stratification tool than anatomic assessment, giving SPECT a “distinct advantage,” Travin says.

“However, the accuracy of PET is consistently higher than SPECT, with PET in the low 90s and SPECT in the mid to high 80s. Yet, due to financial reasons and lack of accessibility to the scanners, PET is used less often,” Travin says. “With the new hybrid systems, like PET/CT and SPECT/CT with new CT attenuation correction, SPECT may be able to close the accuracy gap, but that remains to be seen.”

While North Shore does not currently utilize PET for cardiac patients, Druz says PET is very appropriately suited to this high-risk patient population. “In the setting of advanced coronary disease, with multiple areas of stenoses, the less complex techniques, such as SPECT, are not giving you the full picture,” she says.

One factor that is missing from SPECT evaluation is coronary flow reserve. “A PET exam can provide this evaluation, and provide the amount of coronary flow disparity between rest and stress for every region of the heart. Therefore, if there are several affected areas of the myocardium—some more than others— each of them can be evaluated separately, to assess the absolute and relative coronary flow,” Druz says.

To support the use of PET in this patient population, Druz references a 2009 New England Journal of Medicine interventional cardiology FAME study that found angiography—which just looks at the blockage or the amount plaque itself—is not as good as measuring how much blood is flowing through the coronary artery through fractional flow reserve in patients with multi-vessel CAD (360(3)213-224). “Using PET, we are able to provide a similar, non-invasive measurement,” she says.

Travin concurs with Druz that flow reserve shows promise, and some studies have suggested that PET agents N-13-ammonia and Rubidium-82 are efficacious for quantitative blood flow analysis in diabetics with multi-vessel disease. “With these agents, you are less likely to miss balanced ischemia,” Travin says. “These agents may detect the disease earlier, specifically through the narrowing of the arteries. In fact, older studies using N-13 ammonia show an abnormality of the blood flow reserve predicts cardiac events, even in patients with normal catheterizations.”

Mild uniform impairment of peak stress myocardial blood flow and coronary flow reserve in a patient with hypertension and hypercholesterolemia with normal epicardial coronary arteries is suggestive of microvascular disease. Source: Siemens Healthcare

 

Cost: Final barrier to wider PET adoption

While Druz and her colleagues are “actively pursuing” implementing a PET scanner, they currently refer their cardiac patients to an oncology PET facility on-campus. “PET scanners are quite expensive; plus, they require a lot of expertise and have different radioactive tracers than SPECT. For these reasons, it is more of an undertaking than simply a capital purchase,” she says.

In the hospital setting, PET reimbursement is superior to SPECT. “However, the facility has to take on the initial upfront expense of buying and running a PET system, including the Rubidium supply,” Druz says. “The business side of it may allow a facility to break even, or even be slightly ahead.”

For private practices, CMS on Jan. 1 issued severe cuts to SPECT reimbursements of 36 percent, based on the 2010 Medicare Physician Fee Schedule, which caused many practices to seriously question their viability. Since that time, the agency has issued a reversal of 16 percent, which as of press time, would bring a cut of 25 percent to practices. For private payors, Travin points out that radiology benefit managers (RBMs) “seem to block PET more often than SPECT,” although this varies regionally.

“While RBMs are blocking PET to save money, those savings may only be seen initially,” Travin says. “There is a perception that PET is more expensive, but the end result may be more expensive if the patient gets a false positive, and ends up getting a catheterization.”

The profitability to the provider depends, of course, on how many patients will undergo PET scans. And the name of the game is referrals. “The fiscal viability of using a PET scanner may depend on referrals from oncology or how large the referral base is for cardiology,” Druz explains.

“The adoption of PET for cardiology patients is a challenging proposition because it is a very clinically desirable modality, but the financial considerations are the last hurdle,” Druz says.