How Doctors Decide Which Test to Order for Chest Pain

For stable chest pain, the most important decision happens before any scanner is switched on. The physician first estimates how likely it is that the pain comes from narrowed coronary arteries, and that estimate, not the symptom alone, decides whether the next step is a CT scan of the heart, a stress test, or no imaging at all. This is why two people who describe the same chest pressure can walk out of the same clinic with different orders. The 2024 European Society of Cardiology (ESC) guideline for chronic coronary syndromes puts this estimate at the center of the whole pathway.

Key points#

• The test follows an estimate of risk, not the symptom by itself.
• That estimate, called the clinical likelihood of obstructive coronary disease, is built from age, sex, symptom pattern, and cardiovascular risk factors.
• Very low likelihood often means the best next step is no cardiac imaging.
• CT angiography and stress imaging answer two different questions: is there plaque, and does a narrowing actually starve the heart of blood.
• A calcium score or resting ECG can move a patient between likelihood bands before any large test is booked.

Why one symptom sends people down different paths#

Chest pressure on exertion is a single complaint, but it is not a single diagnosis. It can come from a tightly narrowed coronary artery, from mild plaque that never limits blood flow, or from something outside the heart entirely. Because the underlying cause varies so widely, ordering the same scan for everyone would waste some patients' time and mislead others. The guideline's answer is to grade the situation first.

The older name for that grading step was pretest probability. The 2024 ESC guideline updates the term to the clinical likelihood of obstructive coronary artery disease. The change of wording is deliberate: it signals that the number is not a symptom checklist but a weighted estimate drawing on several inputs at once.

Building the likelihood number#

The starting inputs are the familiar three: age, sex, and the character of the chest discomfort. Classic angina, meaning central chest discomfort that is brought on by exertion or emotion and eases with rest or nitroglycerin, pushes the estimate up. Discomfort that does not follow that pattern pulls it down.

The 2024 model then layers cardiovascular risk factors onto that base: diabetes, high blood pressure, abnormal cholesterol, smoking, and a family history of early heart disease, along with clues from the resting ECG and echocardiogram. This weighting does real work. According to the ESC guideline and its published summaries, adding risk factors reclassifies a large share of patients downward. The earlier 2019 model placed roughly one in five assessed patients in the very-low-likelihood group (5 percent or less); the risk-factor-weighted model places close to half of them there. In practice that means many people who would once have been sent for a scan are now correctly identified as unlikely to have obstructive disease, and the most evidence-based next step for them is often no test at all.

The four routes#

Once the likelihood is estimated, the guideline sorts patients into bands, and each band points to a different next step. The thresholds below come from the 2024 ESC recommendations and the authors' own summary.

Very low (5 percent or less)#

Deferring cardiac imaging should be considered. The logic is statistical. When disease is this unlikely, a positive result is more often a false alarm than a true finding, and following it up leads to more scans, contrast, and procedures with no net benefit. The recommended plan is to treat risk factors and symptoms and reassess, rather than test until something turns up.

Low to moderate (above 5 up to 50 percent)#

This is the home ground of coronary CT angiography (CCTA). A CT study of the coronary arteries is especially good at ruling disease out, so in a group where blockages are possible but not probable, a clean scan reliably settles the question. It also picks up early, non-obstructive plaque that a stress test would never see.

Moderate to high (above 15 up to 85 percent)#

Here functional imaging takes the lead: stress echocardiography, nuclear perfusion imaging (SPECT or PET), or stress cardiac MRI. These tests do more than map anatomy. They show whether a narrowing actually reduces blood supply to heart muscle when the heart is working hard. When the likelihood is higher, that functional answer is what guides the choice between medication and a procedure.

Very high (above 85 percent)#

When obstructive disease is nearly certain and symptoms justify it, the pathway points toward invasive coronary angiography, the catheter-based study that can both confirm a blockage and, in the same session, treat it.

The bands overlap on purpose. In the middle range, either CCTA or functional imaging can be reasonable, and the guideline says the choice should also weigh patient factors that affect test accuracy and, frankly, the local expertise and equipment available. A test done well by an experienced team beats a theoretically preferable test done poorly.

Two questions: plaque or blood flow#

The split between CT angiography and stress testing is not arbitrary, because the two answer different questions. CCTA asks whether plaque is present and whether a vessel is narrowed. Functional imaging asks whether any narrowing is severe enough to limit blood flow under stress. A vessel can look narrowed on anatomy yet still carry enough blood, and a stress test can be abnormal without showing which vessel is responsible. Matching the test to the question, and to the likelihood band, is the heart of appraising this evidence rather than defaulting to whichever scanner happens to be free.

Small tests that reshuffle the estimate#

Two quick measures can move a patient between bands before any large study is booked. A coronary artery calcium score, a fast CT count of calcified plaque, can shift a low-likelihood patient into the very-low group when the score is zero (or, in selected cases, in the 1 to 9 range), supporting deferral. And an exercise ECG, or the pattern of resting-ECG and echo findings, can nudge the estimate up or down. The guideline's instruction is to start with the risk-factor-weighted estimate and then raise it when abnormal findings appear. This is exactly why two patients with identical symptoms can end up on different paths: their calcium scores, ECGs, and risk profiles carried their estimates into different bands.

Reading your own workup#

Two takeaways help patients make sense of an order. First, being told that chest pain does not need a scan can be a careful, evidence-based decision rather than a brush-off, especially when the calculated likelihood is very low. Second, whether you are sent for a CT or a stress test usually depends on where your likelihood falls and which question the physician needs answered. A useful question to ask is simple: what was my estimated likelihood, and what is this test meant to answer. That turns an opaque order into an understandable one.