Download this article's Factoid (PDF & PPT for Gold Subscribers)

Even when repeat computed tomographic angiography (CTA) is sanctioned by appropriate use criteria (AUC) for patients with suspected coronary artery disease (CAD), it may not be necessary provided the previous CT scan was relatively recent and showed at most minor stenosis and the patient is free from diabetes.

The findings, published online July 18, 2014, ahead of print in Circulation: Cardiovascular Imaging, make the case for changes in the current guidelines that could conserve resources and spare patients unneeded radiation exposure, the authors say.  

The final number of cases with significant stenosis was 74 (13.3%), 55 confirmed by angiography and 19 detected by CTA alone. Thus, the sensitivity of CTA was 76.4%, the specificity 83.3%, and the overall accuracy 78.1%.

Predictors of Need for Repeat Testing  

Multivariable logistic regression analysis identified 3 predictors of significant stenosis being detected by repeat testing:

• More than 3 years between scans (adjusted OR 2.06; 95% CI 1.22-3.48)
• Diabetes at the time of repeat CTA (adjusted OR 2.42; 95% CI 1.35-4.34)
• At least 1 segment with 26%-50% stenosis on prior CTA (adjusted OR 5.57; 95% CI 3.24-9.58) 

Among patients with none of these risk factors (31.9% of the study cohort), both the calculated probability and angiographic confirmation of significant stenosis were low, while among those with at least 1 predictor the average risk was 17.1% (table 1). 

Table 1. Calculated Probability vs Observed Prevalence of Significant Stenosis

Moreover, among patients with normal coronary arteries on the initial scan, after a median 36.9 months only 1.8% developed a significant stenosis irrespective of the risk factors, and none underwent PCI. 

In a sensitivity analysis in which ‘indeterminate’ stenoses were reclassified as significant, the prevalence of significant stenosis among patients with no predictors remained the same as in the main analysis (4.5%). 

Study Motivated by Imperfect Guidelines  

In an email with TCTMD, Dr. Rybicki said that many clinicians already exercise independent judgment about when to repeat imaging in the absence of data or guideline support. “The AUC [are] not perfect in terms of repeat CTA, and thus we decided to conduct this study to identify the subpopulation for whom we can forgo repeat CTA,” he reported. 

The authors say they “believe that future expansion of scenarios by including prior [CTAs] will improve extant guidelines, contribute to safe and appropriate care of patients, [and] lead to efficient use of resources in patient management.” 

Though the study suggests within 3 years as a safe time frame to forgo repeat CTA in selected patients, the investigators advise that “more data may be needed to identify the most appropriate and clinically relevant cutoff in the general population with indications for repeat CTA.” 

On the cautionary side, the data highlight the importance of taking into account diabetes status when considering repeat CTA for patients with new or worsening symptoms, Dr. Rybicki and colleagues observe. Moreover, they add, “our data support that lesions with mild to moderate stenosis can become more important clinically.”

The authors acknowledge that the study has several limitations, including:

Being performed at a single center and in a Japanese population with low CAD prevalence

Lack of information on the quality of medical therapy following the initial CTA and its preventive impact 

In addition, Dr. Rybicki noted, 50% diameter stenosis is “a conservative outcome but a realistic one to assess a large cohort of patients. This is needed to have the statistical power to make important conclusions as we have done in this study…. Detection of flow-limiting lesions with low flow reserve or [MACE] would serve as better study outcomes. However, we need to update guidelines now.” 

Interests of Patients, Payers Align

In a telephone interview with TCTMD, Matthew J. Budoff, MD, ofHarbor-UCLA Medical Center (Torrance, CA), said, “There are a lot of data that say that if a CT angiography is perfectly normal, the warranty period is at least 5 years,” but this study “adds value” by specifying that when the initial stenosis is greater than 25%, imaging may need repeating, especially beyond 3 years. “These are good numbers for doctors to remember,” he added. 

According to Dr. Budoff, the issue of when CTA should and should not be repeated is one in which the interest of patients in avoiding unnecessary radiation coincides with that of payers in reducing costs. 

Dr. Rybicki agreed, although he called the radiation issue “overcooked” and said the “real benefit is in the financial consequences.” 

Noting the study’s limitations, Dr. Budoff commented: “There is definitely more work to be done, but if we can reproduce [the findings] in a US population, I think it will lead to changes in practice and hopefully in the guidelines as well.” 

Dr. Rybicki confirmed that new criteria for repeat CTA will be incorporated into revised guidelines. “[The study] data are not perfect,” he noted, “but it would be ethically difficult to justify a prospective or randomized control study.”

Source: 
Kumamaru KK, Kondo T, Kumamaru H, et al. Repeat coronary CT angiography in patients with a prior scan excluding significant stenosis. Circ Cardiovasc Imaging. 2014;Epub ahead of print. 

Related Stories:

• CT Angiography Rules Out Stenosis in Patients at Low-Intermediate CAD Risk
• Coronary CTA Bests Exercise Testing as Diagnostic Tool for Suspected CAD
• CTA Reliably Rules Out CAD, Specificity Still an Issue