How Doctors Decide Which Test to Order for Chest Pain

Doctors do not order the same test for everyone with chest pain. Before choosing any scan, the physician estimates how likely it is that the pain comes from narrowed coronary arteries, then routes very-low-risk patients away from testing and sends others to CT angiography or stress imaging based on where that estimate lands. The 2024 European Society of Cardiology (ESC) guideline for chronic coronary syndromes builds its entire diagnostic pathway on that single estimate, which is why two people with identical symptoms can leave a clinic with different orders. Understanding the estimate explains the whole decision.

The estimate that comes before the test

Two people can walk in with the same complaint, chest pressure on exertion, and leave with different plans: one gets a CT scan of the heart, one gets a stress test, and one gets no imaging at all. That is not inconsistency. It reflects a deliberate first step. The older language for that step was "pretest probability." The 2024 ESC guideline reframes it as the clinical likelihood of obstructive coronary artery disease, a change meant to signal that the number is built from more than a symptom checklist. The starting inputs are familiar: age, sex, and the character of the symptom. Classic angina, central chest discomfort brought on by exertion or emotion and relieved by rest or nitroglycerin, raises the estimate. Symptoms that do not fit that pattern lower it.

What the guideline adds is a risk-factor-weighted clinical likelihood model that folds in diabetes, hypertension, abnormal lipids, smoking, and family history of early heart disease, along with findings from the resting ECG and echocardiogram. Layering risk factors onto the symptom-and-demographics estimate does real work. According to the ESC guideline and its published summaries, this weighting reclassifies a large share of patients downward. Where the older 2019 model placed roughly a fifth of assessed patients in the very-low-likelihood group (5 percent or less), the newer risk-factor-weighted model places close to half of them there, sparing many people a test they do not need. That matters because the most evidence-based next step is sometimes no test at all.

Four bands, four routes

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

Very low, 5 percent or less

Deferring further cardiac imaging should be considered. The reasoning is statistical: when disease is this unlikely, a positive test result is more likely to be a false alarm than a true finding, and chasing it exposes the patient to downstream scans, contrast, and procedures without a net benefit. The recommended action is to treat risk factors and symptoms and reassess, not to keep testing until something turns up.

Low to moderate, above 5 up to 50 percent

This is the home territory of coronary CT angiography (CCTA). A CT scan of the coronary arteries is best at ruling out disease. In a group where blockages are possible but not probable, a clean CCTA reliably closes the question, and the scan also detects early non-obstructive plaque that a stress test would miss entirely.

Moderate to high, above 15 up to 85 percent

Here functional imaging moves to the front: stress echocardiography, nuclear perfusion imaging (SPECT or PET), or stress cardiac MRI. These tests go beyond mapping anatomy. They show whether a narrowing actually starves heart muscle of blood under stress. When the pretest likelihood is higher, that functional information, does this lesion cause ischemia, and how severe is it, is what guides decisions about medication versus a procedure.

Very high, above 85 percent

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

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

Why anatomy and function are not interchangeable

The split between CCTA and stress testing is not arbitrary. They answer different questions. CCTA answers "is there plaque, and is a vessel narrowed?" Functional imaging answers "is any narrowing bad enough to reduce blood flow when the heart works hard?" A vessel can look narrowed on anatomy yet not limit flow, and a stress test can be positive without pinpointing which vessel. Matching the test to the question, and to the likelihood band, is the core of appraising this evidence rather than defaulting to whichever scanner is down the hall.

The refinements that move people between bands

Two tools let clinicians adjust the estimate before committing to a big test. A coronary artery calcium score, a quick CT measure of calcified plaque, can reclassify 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 adjust it upward when abnormal findings appear. This is why two patients with identical symptoms can end up on different paths: their calcium scores, ECGs, and risk profiles pulled their estimates into different bands.

What this means for reading your own care

If you are told your chest pain does not need a scan, that can be a considered, evidence-based decision rather than a dismissal, especially when your calculated likelihood is very low. If you are sent for a CT rather than a stress test, or the reverse, the driver is usually where your likelihood falls and which question your physician needs answered. Asking "what was my estimated likelihood, and what question is this test meant to answer?" turns an opaque order into an understandable one. This article is educational and is not medical advice; decisions about your own testing belong with the clinician who can weigh your full history.