CCTA-First ER Strategy Safely Discharges More Patients: ACRIN/PA

Reed Miller

March 26, 2012

Updated March 26, 2012 (Chicago, Illinois) — Coronary computed-tomography angiography (CCTA) screening of patients with chest pain in the emergency department allows more the patients not suffering acute coronary syndrome (ACS) to be discharged safely than standard screening practice, results of a randomized trial show [1].

Today at the American College of Cardiology 2012 Scientific Sessions, Dr Harold Litt (University of Pennsylvania, Philadelphia) presented results of a trial that randomized 1370 low- to intermediate-risk patients presenting to five emergency departments with ACS symptoms to either CCTA screening or traditional chest-pain care. Results of the study, sponsored by the Pennsylvania Department of Health and the American College of Radiology Imaging Network (ACRIN), were simultaneously published online in the New England Journal of Medicine.

The CCTA group had a higher discharge rate than the traditional-care group (49.6% vs 22.7%) and a shorter average hospital stay (18 hours vs 24.8 hours, p<0.001). The coronary disease detection rate was also higher in the CCTA group (9% vs 3.5%). The trial did not strictly dictate how the control-arm patients would be treated, to reflect the diversity of "real-world" practice, Litt said.

Only about 10% to 15% of patients who present to an emergency department with chest pain are ultimately diagnosed with ACS, but most of the patients presenting with suspected ACS are admitted to the hospital, so quickly discharging patients who are not suffering ACS will save scarce hospital resources, Litt explained; however, lower admission rates should not come at the expense of safety. The trial showed no safety disadvantage of the CCTA-based strategy vs the traditional therapy. Each group had one MI within 30 days of presentation, and there were no deaths in the study. So the upper confidence limit of 30-day MI and cardiac death was 0.57%, well below the 1% safety threshold emergency doctors regard as acceptable, Litt said, and he explained why the investigators chose that threshold. "If one were to try to do a randomized controlled trial comparing the safety of coronary CT and traditional care, you'd need over 50 000 subjects, because of the low event rate in the population studied, and this is impractical," Litt said at the ACC meeting. "Therefore, we chose to power our study based on a conservative safety goal, which would allow emergency-department physicians to feel confident about discharging patients with a negative CT."

Study coinvestigator Dr Judd Hollander (University of Pennsylvania), an emergency physician, said, "The reason we do all of this testing is because we want to drive the miss rate below 1%. The reason we want to drive the miss rate below 1% is obvious, but what is not obvious is that there are gobs and gobs and gobs of clinical studies of 10 000 to 20 000 that show that when we use clinical judgment alone, we stink. We can use judgment to drive the miss rate down to about 5%, but going from that 5% to less than 1% cannot possibly be done just with clinical judgment.

"It pains me when I hear people talk too much about clinical judgment. Clinical judgment can say 'high risk' or 'low risk,' but from an emergency-physician point of view, I have to be able to say that the risk is so low that I'm willing to take a chance with that patient's life and send them home," he said. "If you poll emergency physicians, it comes out to be about a half-a-percent miss rate where people can do their jobs and feel okay with it."

CCTA Finds More Disease

In the 908 patients randomized to the CCTA group in the study, the first evaluation was always performed with a 64-slice CCTA scanner. In the control group, the patient's course of care, including diagnostic tests, was determined by the physician. In both groups, the clinical team made the decision to discharge or keep the patient for further treatment. Patients whose chest pain was clearly noncardiac, those who had a coexisting condition necessitating admission, and those at high risk for ACS were excluded from the study.

About 16% of patients randomized to CCTA did not undergo testing because their doctor determined it was unnecessary. The most common reason (27%) for skipping the test was that the patient had a persistently high heart rate. Of the patients who did undergo a CCTA exam, 83% had a maximal coronary-artery stenosis of under 50%, 14% underwent follow-up stress testing, and 12% of the stress-tested patients had reversible ischemia. Only 37 of the CCTA patients (4%) underwent cardiac catheterization, and 28 were found to have coronary stenosis over 50%. By contrast, 64% of the patients in the control group were referred for stress testing, of which 6% were found to have reversible ischemia. Of the 18 control patients who underwent invasive angiography, eight had coronary stenosis over 50%.

Litt said his group will follow outcomes for the patients in this trial for at least a year. The study also found no difference in resource utilization between the two groups.

ACRIN-PA Study Confirms Previous Experience With CCTA

Commenting on the study, Dr Matthew Budoff (Los Angeles Biomedical Research Institute, CA) told heartwire that this study builds on CT STAT, ROMICAT, and other studies by demonstrating that the CCTA-based rule-out strategy is fast, cost-efficient, and safe. "I do not think further trials are warranted, as to get a level of evidence A (highest), multiple randomized trials are necessary, and they are now complete and published," he wrote in an email. "Prior to these latest trials, the European Society of Cardiology already gave CTA a class IIA recommendation, and now this can be advanced to level I evidence--that this strategy, when sufficient expertise is available, is safe, prudent, and cost-efficient."

Senior CT-STAT investigator Dr Gilbert Raff (William Beaumont Hospital, Royal Oak, MI) agrees with Budoff that the study by Litt et al confirms the CT STAT and ROMICAT results, but "the present study is unique in that the primary outcome was safety and that the traditional-care arm included strategies such as no diagnostic testing or accelerated alternative stress or imaging procedures. Furthermore, the results demonstrate an overall decrease in admissions, which was not true of CT-STAT in comparison with rest-stress [myocardial perfusion imaging] MPI, and also demonstrated an overall decrease in emergency-department length of stay."

Raff also noted that while the referral rates to invasive angiography were similar in both arms, CCTA resulted in a slightly higher rate of revascularization. "It is not clear whether this tendency is due to higher diagnostic effectiveness; however, no data from this or other trials support adverse outcomes in the traditional diagnostic arms. Increased revascularization rates could influence the total costs of care for CCTA vs traditional diagnostic evaluation . . . and this is an important area for future research."

Dr Armin Arbab-Zadeh (Johns Hopkins University, Baltimore, MD) told heartwire that the study by Litt et al is in line with all of the previous studies that show that patients who exhibit no coronary disease on CCTA have an extremely low risk of coronary events. "It's all a very consistent theme," he said. "It's easy to know what to do if the CCTA is negative. The patient goes home, and we have good data to show that this is safe. The problem is what to do if the patient has some disease . . . and the most awkward situation is if they have some disease, but it's not obstructive. Then you don't what to do with those patients. So that is the challenge in the upcoming years--what to do with all of the information coming from CT? What is the best strategy to deal with these patients? That for me seems to be the biggest challenge for the next decade or so."

Litt reports receiving consulting fees from Medrad-Bayer, grant funding from Siemens Medical Solutions, and reimbursement for travel expenses from Siemens Medical Solutions. Arbab-Zadeh is funded by a career development grant from the National Institutes of Health.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....