After PCI or CABG, and repeat testing

Symptoms first, then time

After a stent (percutaneous coronary intervention, PCI) or bypass surgery (coronary artery bypass grafting, CABG), the 2023 multimodality AUC asks two questions in order: is the patient having symptoms, and if not, how long has it been? It rates symptomatic patients with prior MI or revascularization in Table 1.3, and asymptomatic patients with prior revascularization or MI in Table 2.2 Winchester et al., 2023, Tables 1.3 and 2.2.

If the patient has symptoms

Table 1.3 separates patients whose symptoms are anginal, or similar to a past ischemic episode, from those with nonanginal symptoms Winchester et al., 2023, Table 1.3. The authors explain the intent. With anginal symptoms, invasive testing may be warranted. With symptoms that are clearly not cardiac, ischemia testing can often be deferred. After the ISCHEMIA trial, they also say either testing or deferral may suit a symptomatic patient with prior revascularization, depending on preference and the individual situation Winchester et al., 2023, Section 8, Discussion.

The chest pain guideline adds that in patients with known coronary disease, clinicians should intensify medical therapy first when they can, and defer testing Gulati et al., 2021, Section 5.2.1. See how the criteria treat the two symptom groups after PCI:

If the patient has no symptoms

Table 2.2 splits asymptomatic patients by time since the procedure: under or over 2 years after PCI, and under or over 5 years after CABG Winchester et al., 2023, Table 2.2. It also lists incomplete revascularization, prior high-risk PCI and high risk for silent ischemia as separate scenarios. The AUC notes that routine, surveillance-type imaging means testing only because time has passed, not because anything changed Winchester et al., 2023, Assumption 12.

The PET AUC applies the same time cut-offs. Its stated reasoning is that the yield of testing is low early on, and that later on silent ischemia becomes more plausible, from occlusion of a single graft after CABG or from a new stenosis after PCI Schindler et al., 2025, Scenarios 156, 157, 160 and 161. The 2023 chronic coronary disease guideline is firmer: in patients with no change in clinical or functional status, routine periodic CCTA or stress testing is not recommended (no benefit), and routine periodic invasive angiography should not be done (harm) Virani et al., 2023, Section 7.1, Recs 2 and 4.

Stents, grafts and CT

CCTA can be used after revascularization, but the stent matters. Both the chest pain guideline and the chronic coronary disease guideline say CCTA is reasonable to evaluate graft or stent patency after revascularization (stable chest pain in one, a change in symptoms despite medical therapy in the other), for stents 3 mm or larger Gulati et al., 2021, Section 5.2.1, Rec 4Virani et al., 2023, Section 3.1, Rec 6.

• Why stents are hard. Motion and beam-hardening artifacts, blooming of the stent struts, and calcified plaque in nearby segments can all limit the view of the lumen. The SCCT consensus notes that stable chest pain guidelines have generally preferred functional testing in symptomatic patients with prior stents, partly for this reason Narula et al., 2021, Section 2.8.
• Larger stents. The SCCT consensus calls CCTA appropriate for symptomatic patients with stents 3.0 mm or larger, with measures to help: strict heart-rate control (goal under 60 bpm), iterative and sharp-kernel reconstruction, and mono-energetic reconstruction where available Narula et al., 2021, Table 1.
• Smaller stents. It may be appropriate for stents under 3.0 mm, especially thin-strut stents (under 100 micrometres) that are proximal and not at a bifurcation Narula et al., 2021, Table 1. Studies suggest up to 11% of stents may not be evaluable Narula et al., 2021, Section 2.8.
• Bypass grafts. CCTA is highly accurate for graft patency (sensitivity and specificity of 0.98 each in a meta-analysis of 2,482 grafts). Native vessels in patients with prior CABG are harder, because disease is diffuse and severe Narula et al., 2021, Section 2.7.
• Computed FFR. CT-derived fractional flow reserve has not been validated in stents or bypass grafts. It can still be done in non-stented arteries Weir-McCall et al., 2026, Section 5.7.

Repeat testing and the “warranty period”

The 2023 AUC says the diagnostic and prognostic value of a previous test generally declines over time Winchester et al., 2023, Assumption 3. It also dropped the old 90-day rule for what counts as an earlier test in the same evaluation, calling that cut-off arbitrary Winchester et al., 2023, Section 8, Discussion. If a noninvasive test is equivocal or inconclusive and more testing is warranted, it assumes a different modality will be used Winchester et al., 2023, Assumption 7.

The 2021 chest pain guideline offers the idea of a warranty period for a normal result Gulati et al., 2021, Table 7:

There are conditions. The guideline says that in patients with a recent normal test and no biomarker evidence of acute myocardial injury, further testing is of limited value provided stress was adequate, image quality was sufficient, and symptom frequency and stability have not changed. The intervals differ because few events follow a normal CCTA, whereas people with a normal stress test may still have significant plaque. The warranty for a normal SPECT is highly variable, depending on the type of stress, the patient and the ejection fraction Gulati et al., 2021, Section 4.1, supportive text.

When the tool says “Not specified by the AUCs”

Table 1.3 rates patients with prior MI or revascularization before any testing. Table 1.2 rates patients without known coronary disease after a test Winchester et al., 2023, Tables 1.2 and 1.3. Neither covers a patient with a stent or graft who has already had a stress test in this evaluation. Rather than invent a rating, the tool shows Not specified by the AUCs. The sensible next step is to individualize the decision with the patient’s cardiologist.

Open the scenario builder → to enter prior PCI or CABG, symptom status and any earlier test.