SCOT-HEART

Published

Newby DE, et al. "Coronary CT Angiography and 5-Year Risk of Myocardial Infarction". The New England Journal of Medicine. 2018. 379(10):924-933.
PubMed • Full text • PDF

Clinical Question

In patients with stable angina referred to a cardiology clinic for chest pain evaluation, does coronary computed tomographic angiography (CCTA) in addition to standard care compared to standard care alone decrease the composite endpoint of death from coronary heart disease or nonfatal myocardial infarction, after extended follow-up?

Bottom Line

Among patients with stable angina referred to a cardiology clinic for a chest pain evaluation, the addition of coronary computed tomographic angiography (CCTA) to standard care compared to standard care alone were found in this trial to have lower rates of the composite end point of death from coronary heart disease or nonfatal myocardial infarction at 5 years, driven primarily by a lower rate of nonfatal myocardial infarction.

Major Points

Coronary CT angiography (CCTA) is a promising tool in anatomically detecting coronary heart disease underlying stable angina, though whether its use associates with improved hard CVD endpoints was unknown. Previous trials concerning CCTA in the evaluation of angina include the ROMICAT-II (2012) and PROMISE (2015) trials. ROMICAT-II found that patients seen in the emergency department for symptoms suggestive of acute coronary syndrome had reduced length of stay and were admitted less frequently when evaluated with CCTA compared to standard of care. PROMISE found no difference in clinical outcomes in an outpatient population evaluated using CCTA vs functional testing over 2 years.

The Scottish COmputed Tomography of the HEART (SCOT-HEART) trial recruited 4,146 Scottish adults with chest pain thought to be possibly stable angina who were referred to a cardiology clinic. Participants were randomized to CCTA plus usual care or usual care alone. The early findings from the SCOT-HEART trial were published in 2015 and reported improvements in diagnostic classification. The present manuscript was published in 2018 and included hard clinical diagnoses and about 5 years of follow-up. With this follow-up, use of CCTA was associated with a significant reduction in CHD mortality or nonfatal MI (2.3% vs. 3.9%; HR 0.59; 95% CI 0.41-0.84; P=0.004; NNT=62). Ultimately, this trial added additional information on how CCTA can inform clinical decision-making with the goal of better targeting which patients would benefit from preventive care. They did not find that CCTA increased use of angiography or coronary revascularization. It is possible that CCTA detects patients with non-obstructive lesions to target with preventative therapy that would be otherwise missed by standard evaluations. The accompanying editorial notes that the PROMISE trial, which found no difference in clinical outcomes, had an increase in preventive therapy use in both groupings unlike this trial. The authors also note that thresholds for initiating preventative therapies in this population has lowered since the trial completed, so it is possible that the benefit observed would no longer be evident.

Guidelines

The 2012 ACCF/AHA/ACP/AATS/PCNA/SCAI/STS guidelines were written before the results of this trial were published.

Design

• Open-label, parallel group, prospective, randomized trial
• N=4146
  • Standard Care plus CCTA (n=2073)
  • Standard Care (n=2073)
• Setting: 12 centers in Scotland
• Enrollment: November 2010 to September 2014
• Median follow-up: 4.8 years (3-7 years of follow-up)
• Analysis: Intention-to-treat
• Primary outcome: composite of death from coronary heart disease or nonfatal myocardial infarction

Population

Inclusion Criteria

• Aged 18-75 years
• Stable chest pain
• Seen at an outpatient cardiology clinic

Exclusion Criteria

• Can't or won't undergo CT scanning
• CKD with GFR <30 mL/min
• Allergy to contrast
• Unable to give informed consent
• Known pregnancy
• Acute coronary syndrome in prior 3 months

Baseline Characteristics

• Mean age: 57.1 years
• Mean BMI: 29.7 +/- 5.9
• Cardiovascular risk factor
  • Current or former smoker: 53%
  • Hypertension 34%
  • Diabetes mellitus: 11%
  • Hypercholesterolemia: 53%
  • Family history of CHD: 42%
• Atrial Fibrillation: 2%
• History of CHD: 9%
• Relevant medications
  • Antiplatelet agent: 48%
  • Statin: 43%
  • Beta-blocker: 33%
  • ACE inhibitor or ARB: 17%
  • Calcium-channel blocker: 9%
  • Nitrates: 28%
  • Other antianginal agent: 5%
• Anginal symptoms
  • Typical angina: 35%
  • Atypical angina: 24%
  • Nonanginal chest pain: 41%
• Diagnosis at baseline
  • CHD: 47%
  • Angina due to CHD: 36%

Interventions

• Randomized to CCTA plus standard care vs. standard care groups, with management was left to discretion of attending clinician
• CCTA group was prompted to consider the results of the CCTA in management decision
  • When there was evidence of non obstructive (10-70% cross-sectional luminal stenosis) or obstructive coronary artery disease, prescription of preventive therapies was prompted.
• Standard care group was prompted to consider the ASSIGN cardiovascular disease risk score
  • When ASSIGN score ≥20, prompted to prescribe preventive therapies.

Outcomes

Comparisons are CCTA plus standard care vs. standard care alone.

Primary Outcomes

CHD mortality or nonfatal MI

2.3% vs. 3.9% (HR 0.59; 95% CI 0.41-0.84; P=0.004; NNT=62)

Secondary Outcomes

Nonfatal MI

2.1% vs. 3.5% (HR 0.60; 95% CI 0.41-0.87; NNT=71)

CHD mortality

0.2% vs. 0.4% (HR 0.46; 95% CI 0.14-1.48)

All-cause mortality

2.1% vs 2.1% (HR 1.02; 95% CI 0.67-1.55)

Invasive coronary angiography

23.7% vs 24.2% (HR 1.00; 95% CI 0.88-1.13)

Revascularization

13.5% vs 12.9% (HR 1.07; 95% CI 0.91-1.27)

Other Measures

Started on preventive therapies

19.4% vs 14.7% (OR 1.40; 95% CI 1.19-1.65; NNS=21)

Started on antianginal therapies

13.2% vs 10.7% (OR 1.27; 95% CI 1.05-1.54; NNS=40)

Criticisms

• Clinical importance is unclear given that there was no observed mortality difference despite a significant difference in nonfatal MI which drove the difference in primary end point.
• The accompanying editorial brought up the point that the differences in subsequent medical management between groups were modest and would not explain the difference in nonfatal myocardial infarction.
• Endpoints were classified using diagnostic coding, without a formal event adjudication. This could introduce some diagnostic error, and there is general controversy over the clinical significance of the endpoint of nonfatal myocardial infarction given advances in cardiac biomarker sensitivity. If clinically meaningful MIs were being prevented, then there would be an expected mortality benefit. Being non-blinded, there is potential for introduction of biases in the coding of clinical events.

Funding

Study funded by the Scottish Government Chief Scientist Office. Supplementary funding from British Heart Foundation, Edinburgh and Lothians Health Foundation Trust and the Heart Diseases Research Fund.