SAN DIEGO – The addition of CT angiography to standard care changed the diagnosis and treatment of one in four patients with stable chest pain in the prospective, randomized SCOT-HEART trial.
CT angiography (CTA) also reduced coronary heart disease deaths or myocardial infarctions by 38% after a median follow-up of 1.7 years, although the finding was of borderline significance (hazard ratio, 0.62; P = .053).
A post-hoc, landmark analysis, however, that accounted for the roughly 6-week delay between the clinic visit and implementation or alteration of therapy based on CTA findings, showed a halving of these outcomes (HR, 0.50; P = .015), chief investigator Dr. David Newby reported at the annual meeting of the American College of Cardiology.
SCOT-HEART (Scottish COmputed Tomography of the Heart trial) involved 4,146 patients referred from chest pain clinics across Scotland for assessment of suspected angina due to coronary artery disease, of whom 47% were diagnosed in the clinic with coronary heart disease and 36% with angina due to coronary heart disease. Patients were then evenly randomized to standard care involving cardiovascular risk assessment with the ASSIGN Score alone or with CTA.
When attending clinicians reviewed the cases at 6 weeks, the diagnosis of coronary heart disease (CHD) changed in 25% of patients assigned CTA vs. only 1% assigned standard care alone and the diagnosis of angina due to CHD changed in 23% vs. 1% (P value < .001 for both), Dr. Newby, from the University of Edinburgh, said.
Clinicians reported that CTA significantly increased the certainty (Relative risk, 2.56; P < .0001) and frequency (RR, 1.09; P = .017) of the diagnosis of CHD and increased the certainty of a diagnosis of angina due to CHD (RR, 1.79; P < .0001), but had no effect on its frequency (RR, 0.93; P = .12).
Overall, 63% of patients had evidence of CHD on CTA, with 25% having obstructive disease.
CTA altered subsequent testing in 15% of patients vs. only 1% with standard care (P < .001). CTA use was associated with the cancellation of 121 functional stress tests and 29 invasive coronary angiography exams. CTA also prompted 94 new angiograms vs. just 8 with standard care, but this was mainly the result of the exclusion or discovery of obstructive coronary heart disease, including triple vessel disease, Dr. Newby observed. CT was associated with a nonsignificant increase in coronary revascularizations (11.2% vs. 9.7%; HR, 1.19; P .06).
The changes in diagnosis and testing were associated with changes in subsequent treatment in 23% of CTA patients vs. 5% of standard care patients overall (P < .001), including recommendations and cancellations for preventive and anti-anginal therapies. The results were also simultaneously published online (Lancet 2015 [doi:10.19016/S0140-6736(15)060291-4]).
The most impressive aspect of SCOT-HEART was the strong trend for improved outcomes in patients for which therapeutic alterations were made, Dr. Eric Peterson, with the Duke University in Durham, N.C., commented.
“This sets the standard for how we perform and evaluate whether CT can improve outcomes for patients,” he said.
In an editorial accompanying the report,, Dr. Pamela Douglas, from the Duke Clinical Research Institute in Durham, N.C., called the finding of reduced death and MI “intriguing,” but urged caution in its interpretation because it was one of 22 secondary end points and the absolute difference between groups was only 16 events (Lancet 2015; [doi: 10.1016/S0140-6736(15)60463-9]). Earlier in the meeting, she reported that the PROMISE trial found no difference in its primary composite end point of all-cause death, nonfatal myocardial infarction, unstable angina hospitalization, and major cardiovascular procedural complications among chest pain patients evaluated with CTA or functional testing.
Finally, it was noted that radiation exposure in SCOT-HEART (median 4.1 mSv) was substantially lower than that reported in PROMISE, a finding Dr. Newby said he could not explain.