Should we consider a new gold standard?
Coronary pathology in patients with NSTEMI can range from structurally normal vessels to varying degrees of nonobstructive coronary artery disease (CAD) to extensive obstructive CAD affecting the coronary tree. The current diagnostic pathway to differentiate these different coronary pathologies is invasive coronary angiography (ICA).
A routine invasive strategy, however, can be associated with an increased risk of bleeding, iatrogenic injury from catheterization and prolonged hospital stay.
Coronary computed tomography angiography (CCTA) is a simple, low-risk noninvasive test that can potentially rule out CAD and is gaining more popularity in the evaluation of NSTEMI patients. However, the prognostic information about obstructive CAD and/or the extent of CAD defined by CCTA in patients with NSTEMI compared to ICA is unknown.
This data is from an observational component of the original VERDICT trial where all patients who were deemed eligible for early ICA received a blinded CCTA prior to ICA.
The original VERDICT Trial was an RCT of 2,100 patients with Non-ST Elevation Acute Coronary Syndromes (NSTEACS) from nine hospitals in Denmark that were randomized to acute invasive strategy within 12 hours or deferred invasive strategy within 48 hours to 72 hours. The primary endpoint was a combination of all-cause death, nonfatal recurrent MI, hospital admission for refractory MI or hospital admission for heart failure.
Coronary artery disease was broken down into severity and extent. Severity was defined as obstructive (≥1 coronary with ≥50% stenosis) and non-obstructive (<50% stenosis). Extent was defined as high risk (obstructive left main or proximal LAD stenosis and/or multivessel disease) or non-high risk (all other patients).
In the test, 978 patients had a blinded CCTA prior to heart catheterization. The median time interval between CCTA and ICA was two hours and the median follow-up time was 4.2 years. CCTA identified 73.4% of patients as having obstructive disease while ICA identified 66.9%. CCTA and ICA were concordant in their findings in 88.5% of all patients. As for extent of CAD, CCTA identified 51% of patients as high risk and ICA identified 36.8%. The two modalities were concordant in their findings 76.8% of all patients.
The rate of the primary endpoint was up to 1.7-fold higher in patients with obstructive CAD compared with patients with non-obstructive CAD and in patients with high-risk CAD, the rate of the primary endpoint was 1.5-fold higher compared with the rate in those with non-high-risk CAD.
In this observational trial of CCTA vs. ICA in patients with NSTEACS, CCTA was equivalent, but not identical to ICA for assessment of long-term risk. More obstructive CAD was found with CCTA compared to ICA. This could lead to more downstream testing and interventions. Subsequent ICA in patients with nonobstructive CAD by CCTA or ICA did not provide further risk stratification.
Additionally, CCTA is an anatomic study whereas ICA is not only an anatomic study, but a functional study as well. This is important to understand because there may be some nonobstructive lesions seen on CCTA that are attributable to cardiovascular events from unstable plaques. Advantages of ICA include looking at coronary plaque morphology such as intravascular ultrasound, optical coherence tomography and near-infrared spectroscopy looking at vascular calcification, plaque volume or high-risk plaque features all of which have prognostic implications.
My biggest issue with this trial is, if ICA is considered the gold standard, how many patients were ICA (+) but CCTA (-) and vice versa? Hazard ratios are not as helpful as sensitivity/specificity and LRs. Essentially this is a non-inferiority study, but the stats used are not helpful in advising the individual clinician what to do.
For better or worse CCTA has started to be used in the ED. This multicenter observational sub study from the VERDICT trial shows that CCTA can identify severity and extent of CAD in an equivalent fashion to ICA in assessing long-term risk in patients with NSTEACS.
However, without individual patient data and evaluating false negatives, and false positives, statistically significant hazard ratios do not help guide us on what to do at the bedside. We should not be substituting CCTA for ICA based on this data.