How Heart Tests Work — and Why One Test Leads to the Next

Written by a practicing, board-certified American cardiac surgeon, grounded in clinical experience and verified primary sources.

Medical Disclaimer: This article is for education only and is not medical advice, diagnosis, or treatment. It explains how care works and how to take part in it, not what your own situation means, whether a symptom is dangerous, or whether your team’s plan is right. Those judgments belong to your clinicians. Always consult qualified providers for medical decisions, never delay care because of something you read here, and in an emergency seek immediate help. Its purpose is to strengthen your partnership with your care team, not to replace it.

In brief: A test rarely hands down a verdict. It shifts a probability, moving your clinician’s estimate up or down, which is why the same result can reassure one person and worry another, why an abnormal screen is not a diagnosis, and why a normal result can still lead to another test. Each cardiac test answers one narrow question: a stress test looks for a flow-limiting narrowing, a calcium scan looks for hardened plaque, a blood troponin looks for injured heart muscle, and none of them answers the others. Knowing what a test is for keeps a single number from frightening you or falsely reassuring you, helps you see why more testing is not always better, and lets you ask the one question that matters most before any test: what would the result change?

Most people picture a heart test as a verdict that tells you whether you have the disease. That is almost never what a test does. A test shifts a probability. Before it runs, your clinician already has an estimate of how likely a condition is, built from your age, history, risk factors, and symptoms. The result moves that estimate up or down, and it rarely moves it all the way to certainty.

That single idea makes the rest of testing make sense. It explains why the same result means different things in different people, why an abnormal screen is not yet a diagnosis, why a normal result sometimes leads to another test, and why two careful doctors can look at the same number and do different things. For heart and metabolic conditions, the diagnosis is defined and tracked by numbers like blood pressure, cholesterol, glucose, and troponin. Misreading one of them is a common way a person either frightens themselves over a result that was never alarming, or relaxes over one that deserved a closer look.

This article is about testing as a process. It covers what a test does and how the common heart tests work, why tests come in a sequence, and why a normal result can still prompt another. It also covers when more testing stops helping, what it costs, and how to prepare so a test is not wasted. What your specific results mean is a conversation for you and your own clinicians, not something to settle from an article or a reference range.

What a heart test does

Picture two people who get the same result on the same test.

The first is thirty-five, with no symptoms, no family history, and no risk factors. The second is sixty-eight, with diabetes, high blood pressure, and chest discomfort walking uphill. A mildly abnormal result means something very different in each. In the first, where the condition was unlikely to begin with, a mild abnormality is more likely a quirk of the test than a real problem. In the second, where it was already probable, the same value is more likely to be true and to matter. The clinician who shrugs at one and acts on the other is not being inconsistent. The result means different things, because the starting point was different.

That starting point has a name, pre-test probability, meaning how likely a condition is before the test, given everything already known about the person. A result is read against that backdrop, not in isolation, which is why you cannot reliably read your own value off a reference range and know what it means. The range tells you where the number sits. It knows nothing about you.

A discipline follows from this, and it is one to borrow. A test earns its place when the result would change what happens next. Before ordering one, a careful clinician is asking, at least silently: what will this tell me, and what would I do differently depending on the answer? If the honest answer is “nothing,” the test often should not be done, which makes “what would this change?” one of the most reasonable questions a patient can ask. It is not a challenge. It is the same question the clinician is already asking.

What a test can and cannot tell you

Many people walk into a test expecting it to deliver certainty: a clear yes or no, a line drawn under the worry. Most tests do not work that way. They reduce uncertainty; they rarely erase it. A good result usually leaves a little room for doubt, an ambiguous one leaves more, and that is the nature of the tool, not a flaw in it. Knowing this in advance changes how a result lands. It is why a normal test is reassurance rather than a guarantee, and why a clinician may still want to watch, repeat, or look again.

Testing is rarely just an intellectual exercise. The test itself is often the easy part; what wears on people is the waiting, and a result that lands in the portal late at night with no one there to explain it. A single number, read alone and out of context, tends to frighten more than it informs, which is why it was not meant to be read that way. The gap between a result and its meaning is normal, and interpreting it is a job for your care team, not one to settle alone.

Two opposite misreadings cause much of the distress. The first is hearing normal as a clean bill of health, so that a follow-up test, or the reminder that cholesterol still matters, feels like a betrayal or a mistake. But normal almost always means normal for the one question that test asked, not all clear for the heart. The second is hearing abnormal as catastrophe, when most abnormal screening results turn out to be false alarms or minor findings that a second look sorts out. Read both more lightly: a normal result as reassurance rather than a guarantee, an abnormal one as a prompt rather than a sentence. That takes much of the fear out of the process.

Why tests come in a sequence

Tests are ordered in a sequence because they trade cost, risk, and precision against one another. The sensible path starts with what is cheap, safe, and broadly informative, and escalates to what is expensive, invasive, or precise only when the earlier result raises the odds enough to justify it. A test that carries real risk is where a search ends, once the odds warrant it, not where it begins.

This is also the difference between two kinds of testing that look identical but answer different questions. A screening test looks for a condition in someone without symptoms, where the prior probability is usually low. A diagnostic test answers a specific question raised by a symptom or an earlier result, where the probability is already higher. The same blood test or scan can play either role, and its result means different things depending on the job.

Because screening starts from low probability, an abnormal screen is usually not a diagnosis. It is a reason to confirm. Two everyday examples show it. A single high reading in the office is not hypertension. The diagnosis rests on an average of at least two readings on at least two occasions, ideally confirmed with out-of-office monitoring. One number can be high for ordinary reasons: a stressful commute, a full bladder, or the rise many people show in a clinical setting [1]. Diabetes works the same way: unless sugars are unequivocally high or classic symptoms are present, a single abnormal glucose or A1C does not establish it, and the abnormal test is typically repeated on another day before the diagnosis is made [2]. A value that will shape years of treatment should be confirmed first.

When a result looks wrong: false positives and false negatives

No test is perfect, and the reason is biology, not sloppiness. Whatever a test measures, a level in the blood, a shift on an ECG, a wall moving on ultrasound, that measurement overlaps between people who have the condition and people who do not. The two groups are not cleanly separated; their ranges blur into each other. So wherever the line between “normal” and “abnormal” is drawn, some people will fall on the wrong side of it. That overlap is the source of nearly every false result, and it cannot be engineered away.

The line itself, the cutoff, is the part chosen deliberately, and choosing it is a trade-off. Two properties describe how a test handles the overlap. Sensitivity is how well it catches the people who truly have the condition; a sensitive test misses few of them. Specificity is how well it clears the people who truly do not; a specific test raises few false alarms. The two pull against each other, and the overlap is why. Set the cutoff to catch nearly everyone with the disease, and it will also flag some healthy people whose values sit in the shared range, a false positive. Set it to avoid false alarms, and it will miss some real cases, a false negative.

Screening tests are deliberately built toward sensitivity: catch as many true cases as possible, accepting some false alarms. That is exactly why an abnormal screen leads to a confirmatory test rather than straight to treatment. The false positives are expected, and the next test sorts them out. So when a result turns out not to match reality, “the test was wrong” is usually the wrong frame. More often the test behaved exactly as an imperfect tool behaves across many people. Seeing it that way takes the alarm out of a false positive and the false comfort out of a false negative, and it explains why a surprising result is so often followed by a second, different one.

The common heart tests, and what each one looks for

Most confusion about heart testing comes from a single mistaken assumption: that these tests are interchangeable, that any of them can tell you whether your heart is “fine.” They are not, and none of them can. Each test was built to answer one narrow question, and it is close to blind to the others. The most useful thing you can carry into any of them is a clear idea of what it looks for and what it cannot see.

What follows is a tour of the common ones, roughly in the order of cost and risk, from the simplest blood draw to the monitor on your wrist to the catheter in the artery.

The blood tests

The standard lipid panel reports total cholesterol, LDL, HDL, and triglycerides, and for most people a nonfasting sample is fine, with fasting reserved for particular cases such as high triglycerides [3]. These describe long-term risk, not whether an artery is narrowed today. Newer particle-based markers, apolipoprotein B, lipoprotein(a), and the triglyceride-to-HDL ratio, refine that long-term picture; because they belong to the story of cholesterol and risk rather than to testing, the Cholesterol Series covers them in depth.

Glucose and A1C track blood sugar, glucose as a snapshot and A1C as a roughly three-month average. A fasting glucose requires fasting; an A1C does not.

Troponin is the test most often misunderstood, and the misunderstanding can frighten people badly. It is a protein released when heart muscle is injured, and modern assays detect tiny amounts, which is what makes them so good at catching a heart attack early. But a raised troponin does not equal a heart attack. Many things lift it without any blocked artery, from kidney disease and heart failure to a fast rhythm or severe infection, and some people run a raised level at baseline [4]. This is why a single value tells your team far less than the trend across several hours: a sharp rise and fall points to acute injury, while a mildly raised level that holds steady more often reflects a chronic condition. It is also why an emergency department draws troponin more than once and asks you to wait, because the definition of a heart attack requires a rising or falling pattern alongside other evidence, not one number alone [4]. Reading a single troponin off a portal, out of that context, is one of the most common ways a person frightens themselves over a result their clinician already understands.

Other markers fill in the edges. BNP and NT-proBNP rise when the heart muscle is under strain, useful in heart failure, not in finding a blockage; high-sensitivity CRP reflects inflammation. What any of these means belongs to your clinician and your whole picture, not to a reference range or an article.

The electrocardiogram (ECG or EKG)

An ECG records the heart’s electrical activity over a few seconds. It is fast, cheap, and carries no risk, and it is very good at what it is for: rhythm problems, and the electrical signature of a heart attack happening now or in the past. What it is not is a reliable way to rule out coronary artery disease in someone who feels well. A resting ECG is frequently normal in a person with significant artery narrowing, because at rest the narrowed artery may still supply enough blood to look unremarkable. For that reason, major guidance recommends against using a screening ECG to look for heart disease in adults who have no symptoms and are at low risk [5][6]. In that group it does not improve outcomes and mostly generates false alarms and the procedures that chase them.

Rhythm monitors and the smartwatch

A standard ECG is a snapshot of a few seconds, fine for a problem that is always present but useless for one that comes and goes. Palpitations or a fainting spell often happen nowhere near a clinic, so the rhythm is recorded over time: a Holter for a day or two, a patch for up to about two weeks, an implantable loop recorder for months. The rarer the episode, the longer the recording must run to catch it.

The device on your wrist now plays a part too. A smartwatch is one of the most common ways people first hear the words “atrial fibrillation,” and guidance recognizes a role for wearables in monitoring a known rhythm problem [7]. But an alert is a prompt, not a diagnosis: in the large study that put smartwatch screening on the map, only about a third of those who got an irregular-pulse alert and then wore a medical-grade patch had atrial fibrillation [8]. Save the recording, tell your team, and let a proper recording settle it. It is never a reason to start, stop, or change a medicine, least of all a blood thinner, on your own.

The chest X-ray

A chest X-ray uses a very low dose of radiation, a small fraction of a CT scan, and shows the size and shape of the heart and whether there is fluid backing up into the lungs, which makes it useful in heart failure. It does not show the coronary arteries. A normal chest X-ray says nothing about whether those arteries are narrowed.

The echocardiogram (ultrasound of the heart)

An echocardiogram, or echo, is an ultrasound. It uses sound waves, no radiation and no contrast in its standard form, and it shows the heart in motion: how strongly it pumps, summarized as the ejection fraction, whether the muscle is thickened or weakened, and how the valves open and close. It is one of the most informative and lowest-risk tests in all of cardiology for those questions. What it does not do, in its routine form, is show the coronary arteries directly. A normal echo is reassuring about the pump and the valves and does not by itself rule out a narrowed artery. A transesophageal echo, or TEE, takes the probe down the esophagus to sit just behind the heart for a much closer view of the valves and any clots, and it is more involved, usually done with sedation.

Stress testing

A resting heart with a narrowed artery can look normal because the narrowing only starves the muscle when demand rises. The whole idea of a stress test is to raise the demand, by exercise or by medication, and watch for the signature of a flow-limiting narrowing appearing under load.

The exercise treadmill test is the basic version: you walk on a treadmill while an ECG records the heart. It is widely available and uses no radiation. It is also the least precise of the stress tests, with an average sensitivity around 68 percent and specificity around 77 percent, and it produces more false positives in women than in men [9]. Part of that female “false positive” pattern is not error at all. In a meaningful share of these patients, the reduced blood flow is real but comes from disease of the small vessels that the test and the angiogram cannot see well, rather than from a large-artery blockage. The test is flagging something real that the old reference standard missed [10].

Adding imaging sharpens the picture. A stress echocardiogram takes ultrasound pictures before and right after exercise, with no radiation, looking for a wall of heart muscle that moves poorly when pushed. Nuclear myocardial perfusion imaging injects a small amount of radioactive tracer and images blood flow to the muscle at stress and rest; it is more sensitive than the treadmill ECG, with reported sensitivity varying widely across studies [11]. Current guidance, when nuclear imaging is chosen, prefers PET over the older SPECT for better accuracy and fewer unclear results [11]. Imaging is especially useful to add when the resting ECG is already abnormal in a way that would muddy a plain treadmill test, such as a bundle branch block, a pacemaker, or certain medication effects. In those cases a plain treadmill result is more likely to mislead [11].

When you cannot exercise, the heart is stressed with medication instead. Dobutamine speeds and strengthens the heartbeat to mimic exercise, paired with echo. Vasodilators such as regadenoson, adenosine, or dipyridamole open the arteries to expose a flow difference, paired with nuclear imaging. These pharmacologic stress tests answer the same question for someone whose knees, lungs, or balance will not allow a treadmill.

The single most important thing to understand about every stress test is the question it asks: is there a narrowing severe enough to limit blood flow when the heart works hard? That is a narrower question than “are my arteries healthy.” A great many heart attacks come from plaque that was never narrowing the artery enough to limit flow, plaque that was invisible to a stress test right up until it ruptured. This explains one of the most distressing experiences in cardiology: the person who “passed a stress test” and had a heart attack weeks later. The test was not wrong; it answered its question correctly. That question simply was not the one that predicted the event.

The coronary calcium scan (CAC)

A coronary calcium scan is a quick, non-contrast CT of the heart, about ten to fifteen minutes, no dye, at a radiation dose near that of a mammogram [12]. It looks for calcified, hardened plaque and reports it as an Agatston score, from zero to the hundreds and beyond, also given as a percentile for your age and sex. Its real value is at the low end: an absence of calcium points to a low near-term risk and can reclassify about half of statin-eligible people into a lower-risk group. That is why it helps most for someone at borderline risk who is undecided about a cholesterol medicine [13]. Its blind spot is that it sees hardened plaque, not the soft plaque that often ruptures in younger people, and it never stands in for taking symptoms seriously.

Guidelines treat it as a decision aid for that one question. The bands below, with what the 2018 cholesterol guideline suggested each means for starting a statin, are the categories clinicians have used; what a given score means for you is theirs to read [14].

Agatston score	What it reflects	What the guideline suggested
0	No detectable calcified plaque	Reasonable to defer a statin, unless diabetes, smoking, or a strong family history of early heart disease
1-10	Minimal calcified plaque	Favors a statin, especially after age 55
11-100	Mild calcified plaque	Favors a statin
101-400	Moderate calcified plaque	Supports starting a statin
Over 400	Extensive calcified plaque	Supports starting a statin

A score at or above the 75th percentile for age and sex also supports starting, even when the absolute number is under 100. That is the calcium scan in miniature: it earns its place when its result would change the choice in front of you, and not otherwise. (The 2026 dyslipidemia guideline keeps calcium scoring as a decision aid in selected borderline and intermediate-risk adults; the score bands above are the long-standing reference points clinicians know best.)

Coronary CT angiography (CCTA)

A coronary CT angiogram uses contrast dye and a CT scanner to picture the artery walls directly, at a dose for most people of 3 to 5 millisieverts [11]. Unlike a stress test, which only infers a problem from reduced flow, it looks straight at the arteries and sees both obstructive and non-obstructive plaque. This is the difference between a functional test, which asks whether flow is limited under load, and an anatomic test, which shows what the arteries are [11]. Its particular strength is ruling significant disease out, and for many people with new, stable chest pain current guidance now treats it as a reasonable first imaging test [11]. In a large trial, adding CCTA for stable chest pain lowered heart attacks over five years, mainly by getting more people onto statins rather than more stents, a result that fits the medication-first lesson below [15]. Head to head, an anatomy-first and a function-first approach produced broadly similar outcomes [16].

Invasive coronary angiography (the cardiac catheterization, or angiogram)

This is the most precise look at the coronary arteries and the most involved. A thin catheter is threaded from the wrist or groin to the heart, contrast is injected, and X-ray imaging maps the anatomy directly, at a radiation dose broadly comparable to the other advanced cardiac tests [17]. Its great advantage is that it can both diagnose and treat in the same sitting: if a significant narrowing is found, it can often be measured precisely (with a pressure wire, called FFR) and opened with a stent then and there. Its disadvantage is that it is invasive, with small but real risks. Because of that, it is generally reserved for people at high risk, those with severe or persistent symptoms, or those whose non-invasive tests were positive or inconclusive, rather than used as a first look [11]. When it leads to opening an artery, that step carries its own low-frequency risks, including a small chance of a procedure-related heart attack, bleeding, or, more rarely, stroke [5]. Those numbers are small, but they are the reason the invasive test sits at the end of the sequence and not the beginning.

One thing surprises people here. When the dye reveals disease that could be treated either way, with stents or with bypass surgery, the cardiologist often does not stent on the spot, and patients later ask why they were not simply fixed in the moment. Holding off is usually deliberate: it keeps the surgical option open and lets a cardiologist and a surgeon choose the strategy together first. How that decision gets made is the subject of Article 3.

A word on the dye, and your kidneys

Both CCTA and the invasive angiogram use an iodine-based contrast dye, and two worries about it deserve proportion. Allergic-type reactions are usually mild; a severe one is rare, on the order of a few hundredths of a percent [18]. The effect on the kidneys is usually a temporary dip that recovers on its own, uncommon in people with normal kidneys and more likely with existing kidney disease, diabetes, or dehydration [18]. This is why your team checks kidney function and asks about past reactions beforehand. The practical part is yours: tell them if you have ever reacted to contrast or have kidney trouble, so the plan can be adjusted before the test, not after.

Tests for other blood vessels

A few related tests look at vessels outside the heart, each with its own question. A carotid ultrasound images the neck arteries that feed the brain. An ankle-brachial index compares blood pressure at the ankle and the arm to screen for narrowed leg arteries. And an abdominal ultrasound can screen for an aortic aneurysm, recommended once for men aged 65 to 75 who have ever smoked. A normal result in one vascular bed does not certify another.

The table below is a map of the same tour, not a guide to choosing. Which test fits a given person depends on their pre-test probability, the exact question, and what is available locally, all of which belong to the clinician.

Test	What it answers	Radiation or contrast	A key limit
Lipid panel, A1C	Long-term risk and blood sugar	None	Says nothing about a narrowing today
Troponin (blood)	Whether heart muscle is being injured	None	Raised by many non-cardiac conditions; the trend matters
Resting ECG	Rhythm and the electrical signs of a heart attack	None	Often normal despite significant artery narrowing
Ambulatory monitor or smartwatch	The heart’s rhythm over time, to catch an intermittent arrhythmia	None	A single watch alert is a prompt to confirm, not a diagnosis
Chest X-ray	Heart size and fluid in the lungs	Very low	Does not show the coronary arteries
Echocardiogram	Pumping strength, muscle, and valves	None	Does not show the coronary arteries directly
Exercise treadmill ECG	Whether exertion provokes a flow-limiting narrowing	None	Misses non-limiting plaque; more false positives in women
Stress echo or nuclear (SPECT, PET)	The same question, with imaging for accuracy	Echo none; nuclear moderate	Detects flow-limiting narrowing, not all plaque
Coronary calcium scan	Hardened plaque, to refine long-term risk	Low, no contrast	Misses soft, non-calcified plaque
Coronary CT angiogram (CCTA)	The artery walls: obstructive and non-obstructive plaque	Moderate, contrast	Heavy calcium can blur the picture
Invasive coronary angiogram	A precise anatomic map, with the option to treat	Moderate, contrast	Invasive, with small procedural risks

Radiation, the worry raised most, stays modest. A calcium scan is about the dose of a mammogram, a CT angiogram roughly a year of natural background, with the nuclear and invasive tests somewhat higher. Blood work, the ECG and monitors, echo, and MRI use none at all [17]. A fuller dose-by-test comparison lives in the Resource Center.

Why a normal result can still lead to another test

It is frustrating to be told a test was normal and then sent for another, or told the search is not over. Almost always, the reason is the same: a normal result answers the one question that test asked and leaves other questions open. Three versions of this come up constantly, and each is one to understand before it happens to you.

“My stress test was normal, so why do I need another test?” A stress test asks one thing: is a narrowing severe enough to choke off blood flow when the heart works hard? A normal answer is real reassurance, but it does not rule out plaque that is not yet flow-limiting, the kind that can still matter later, and it does not erase a symptom that is still there. If chest discomfort persists, the symptom, not the test, is the reason to keep looking.

“My calcium score was zero, so why are we still talking about cholesterol?” A calcium scan sees hardened, calcified plaque, the kind that has been building for a while. It does not see soft, early plaque, and it does not measure the LDL and apoB particles that drive new plaque in the years ahead. A zero is good news about today’s calcified burden; it is not a promise about the future, which is why the cholesterol conversation continues alongside it.

“My angiogram was clean, so why do I still have symptoms?” This one surprises people most. Among patients sent for angiography because of angina, a large share, by some estimates between a fifth and two-thirds, turn out to have arteries with no significant blockage [19]. For many of them the symptoms are still cardiac, coming from the small vessels an angiogram cannot show or from arteries that briefly spasm, a pattern called ischemia with non-obstructive coronary arteries. A clean angiogram rules out one important thing. It does not always rule out a cause.

The thread through all three is the same. Each test answered its own question well and was nearly blind to the others, so a normal result narrows the search without ending it. Before any test is simply repeated, the same question applies: would a new result change what happens next, or just swap one snapshot for another?

When more testing isn’t better

None of this is an argument for more testing. More testing finds more things, and not all of them matter. The scale here is not small: more than 10 million stress tests and roughly a million diagnostic catheterizations are performed in the United States each year, and a meaningful share are done in people unlikely to benefit [20].

The clearest example is testing people who feel well and are at low risk. More than 90 percent of low-risk adults will not have a cardiac event in the next decade, so finding one early cannot help them. Meanwhile the false positives that screening produces in this group lead to real downstream harm: more scans, invasive procedures with their own small risks, overtreatment, and the burden of being labeled [6][5]. This is why professional guidance recommends against routine cardiac screening with ECG or stress testing in low-risk people without symptoms [5][6]. It is also useful to know that for a person already at high risk, a cholesterol medicine and lifestyle changes are generally recommended regardless of what a stress test or calcium score shows. A test that will not change that plan adds cost and worry without adding benefit [20].

Modern imaging compounds the problem by turning up findings nobody went looking for. These incidental findings, sometimes called incidentalomas, are common: in pooled data they appear in roughly a third of cardiac MRI scans and close to half of chest CT scans [21][22]. Most are harmless. But each can set off a cascade, another scan, a referral, a stretch of worry, occasionally a biopsy or procedure with its own real risk, often to characterize something that was never going to cause harm.

There is one more place this matters, and it runs against a common assumption. People tend to believe that if a test finds a blockage, the blockage must be fixed, or it will kill them. For a sudden, unstable event that is often true. But for stable disease, a large trial randomized more than 5,000 people with significant narrowing and at least moderate reduced blood flow. One group got an early invasive strategy of catheterization and stenting on top of good medication; the other got good medication first, with catheterization held in reserve. Over several years, the invasive-first approach did not reduce death or heart attack compared with medication first; it did relieve angina better in those who had frequent symptoms [23]. The lesson is not that procedures do not help anyone. It is that “they found a blockage” does not automatically mean “you need a stent today.” For stable disease, starting with medication is an evidence-based path, and whether and when to open an artery is a shared decision that weighs your symptoms and your own goals. Only your own physician can weigh the risks and benefits of a procedure for you.

The same discipline applies. A test earns its place only if the result would change what happens next. Information you do not know what to do with can be worse than none, because it generates action without improving the outcome. The goal of a workup is the right test, not the most tests.

Sometimes the best test is no test

The hardest version of this to hear, because it runs against instinct, is that for some people the right next test is none at all. Three examples patients are rarely told:

A routine yearly ECG or stress test when you feel well and your risk is low. It will not improve your odds, and the false alarms it produces lead to procedures that carry real risk of their own [5][6].

Another scan added only to “be sure,” when the result would not change the plan. If you are already high-risk and already on the right medicines, a stress test or calcium score that cannot change that adds cost and worry, not benefit [20].

An invasive angiogram for stable symptoms before medicine has been tried. For stable disease, starting with medication does as well as opening an artery for preventing death and heart attack, so the catheter can often wait [23].

What a test costs, and the insurance step

In the American system, two practical forces shape testing as much as the medicine does. The first is prior authorization: advanced imaging often needs insurer approval first, and Medicare’s own rule captures the principle, a test is not considered necessary if it would not add information that changes management [24]. That is the “what would this change?” discipline written into a payment rule, and a delayed test is often waiting on that step rather than on a medical doubt. The second is cost, which varies more than people expect: a calcium scan is frequently not covered as screening and is paid out of pocket, often roughly 50 to 150 dollars, with self-pay prices differing widely between centers [25]. Affording care, denials, and confusing bills each get their own article later in this series.

Before you agree to a test

You do not need to second-guess your clinician’s expertise to take part well. You need to understand the plan. Three questions, asked plainly, do most of the work, and a good clinician will welcome them because they are the same questions a careful one is already weighing.

The three questions before any test:

• What is this test looking for? Naming the one question the test answers tells you what it will and will not settle, and keeps you from reading a normal result as a clean bill of health it was never able to give.
• What would the result change? If a result would not change the plan, the test may not be needed; if it would, you will understand why it matters and read it more calmly when it arrives.
• What are the risks, the radiation, the contrast, and the cost? Knowing these lets you weigh the test fairly, ask whether a lower-risk or lower-cost option answers the same question, and avoid a surprise on the bill.

How to prepare so a test isn’t wasted

A wasted test can cost a week, a co-pay, and an unnecessary follow-up. A few habits prevent most of that.

Ask what to fast for, and don’t assume. A cholesterol panel is usually fine nonfasting; a fasting glucose needs fasting, an A1C does not [3]. The rules are not intuitive, so ask rather than guess.

Never stop a medication on your own to prepare. Some tests do call for holding a specific drug, certain heart-rate medicines before some stress tests, but only on your team’s explicit instruction. Never start, stop, or change a heart medicine on your own.

Follow the specific instructions. Caffeine, for instance, interferes with some stress tests; a test done without the right prep may have to be repeated.

Bring the inputs. Carry prior results, outside scans, your medication list, and home readings, so nothing is repeated only because it could not be found.

The bottom line

A test rarely delivers a verdict. It shifts a probability, and what a result means depends on how likely the condition was beforehand. That is why an abnormal screen is not a diagnosis, why a normal result can still warrant a follow-up, and why one test so often leads to the next. Each heart test answers one narrow question and is nearly blind to the others, so the most useful thing you can know about any of them is what it is for.

The most important question is often not which test to get. It is how the answer would change what happens next. A test whose result would change nothing rarely earns the worry, the cost, or the cascade it can set off, while a test that would change the path ahead is one to do even when the waiting is hard. Ask what a test would look for and what it would change, follow the instructions exactly, bring your prior results, and never stop a medication on your own. Then read each result for what it is: not a final word, but one step that leaves you better informed than the last.

Key Terms

Pre-test probability: how likely a condition is before a test, estimated from age, history, risk factors, and symptoms. A result is read against this, which is why the same result can mean different things in different people.

Screening test: a test used to look for a condition in someone without symptoms, where the prior probability is usually low; an abnormal result generally calls for confirmation.

Diagnostic test: a test used to answer a specific question raised by a symptom or earlier result, where the probability is already higher.

Sensitivity and specificity: how well a test catches people who truly have a condition (sensitivity) and clears people who truly do not (specificity). The two trade off against each other.

False positive and false negative: a result that flags a condition that is not there, or misses one that is. Both are expected behaviors of imperfect tests, not necessarily errors.

Troponin: a protein released when heart muscle is injured. High-sensitivity versions are very good at catching a heart attack, but many other conditions raise the level too, so the trend and the context matter more than a single value.

Anatomic and functional testing: two ways to look for coronary disease: anatomic tests (CCTA, invasive angiography) picture the arteries directly, while functional tests (stress testing) judge whether blood flow is limited.

Coronary calcium score (CAC): a non-contrast CT measure of hardened plaque, reported as an Agatston score, used mainly to refine long-term risk in borderline cases.

Coronary CT angiography (CCTA): a contrast CT that pictures the coronary artery walls directly, showing both obstructive and non-obstructive plaque.

Invasive coronary angiography: a catheter-based map of the coronary arteries that can both diagnose a narrowing and treat it with a stent in the same sitting.

Incidental finding (incidentaloma): an abnormality found by a test that was looking for something else. Most are harmless but can prompt further testing.

Prior authorization: an insurer’s required approval before a test or treatment is covered.

References

1. Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults. Hypertension. 2018;71(6):e13-e115. doi:10.1161/HYP.0000000000000065
2. American Diabetes Association Professional Practice Committee. 2. Diagnosis and classification of diabetes: Standards of Care in Diabetes-2026. Diabetes Care. 2026;49(Suppl 1):S27-S49. doi:10.2337/dc26-S002
3. Writing Committee. 2026 ACC/AHA/AACVPR/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of dyslipidemia. Circulation. Published online March 13, 2026. doi:10.1161/CIR.0000000000001423
4. Thygesen K, Alpert JS, Jaffe AS, et al; Joint ESC/ACC/AHA/WHF Task Force for the Universal Definition of Myocardial Infarction. Fourth universal definition of myocardial infarction (2018). Circulation. 2018;138(20):e618-e651. doi:10.1161/CIR.0000000000000617
5. US Preventive Services Task Force. Screening for cardiovascular disease risk with electrocardiography: US Preventive Services Task Force recommendation statement. JAMA. 2018;319(22):2308-2314. doi:10.1001/jama.2018.6848
6. Chou R; High Value Care Task Force of the American College of Physicians. Cardiac screening with electrocardiography, stress echocardiography, or myocardial perfusion imaging: advice for high-value care from the American College of Physicians. Ann Intern Med. 2015;162(6):438-447. doi:10.7326/M14-1225
7. Joglar JA, Chung MK, Armbruster AL, et al. 2023 ACC/AHA/ACCP/HRS guideline for the diagnosis and management of atrial fibrillation. Circulation. 2024;149(1):e1-e156. doi:10.1161/CIR.0000000000001193
8. Perez MV, Mahaffey KW, Hedlin H, et al; Apple Heart Study Investigators. Large-scale assessment of a smartwatch to identify atrial fibrillation. N Engl J Med. 2019;381(20):1909-1917. doi:10.1056/NEJMoa1901183
9. Gianrossi R, Detrano R, Mulvihill D, et al. Exercise-induced ST depression in the diagnosis of coronary artery disease: a meta-analysis. Circulation. 1989;80(1):87-98. doi:10.1161/01.cir.80.1.87
10. Rethinking false positive exercise electrocardiographic stress tests by assessing coronary microvascular function. J Am Coll Cardiol. 2023. doi:10.1016/j.jacc.2023.10.034
11. Gulati M, Levy PD, Mukherjee D, et al. 2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR guideline for the evaluation and diagnosis of chest pain. Circulation. 2021;144(22):e368-e454. doi:10.1161/CIR.0000000000001029
12. Coronary artery calcium scoring: its practicality and clinical utility in primary care. Cleve Clin J Med. 2018;85(9):707-716. https://www.ccjm.org/content/85/9/707
13. Nasir K, Bittencourt MS, Blaha MJ, et al. Implications of coronary artery calcium testing among statin candidates according to ACC/AHA cholesterol management guidelines: MESA. J Am Coll Cardiol. 2015;66(15):1657-1668. doi:10.1016/j.jacc.2015.07.066
14. Grundy SM, Stone NJ, Bailey AL, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of blood cholesterol. Circulation. 2019;139(25):e1082-e1143. doi:10.1161/CIR.0000000000000625
15. SCOT-HEART Investigators; Newby DE, Adamson PD, Berry C, et al. Coronary CT angiography and 5-year risk of myocardial infarction. N Engl J Med. 2018;379(10):924-933. doi:10.1056/NEJMoa1805971
16. Douglas PS, Hoffmann U, Patel MR, et al; PROMISE Investigators. Outcomes of anatomical versus functional testing for coronary artery disease. N Engl J Med. 2015;372(14):1291-1300. doi:10.1056/NEJMoa1415516
17. Evaluating radiation exposure in patients with stable chest pain in the SCOT-HEART trial. Radiology. 2023. doi:10.1148/radiol.221963
18. Contrast media in patients with kidney disease: an update. Cleve Clin J Med. 2020;87(11):683-692. https://www.ccjm.org/content/87/11/683
19. Clinical relevance of ischemia with nonobstructive coronary arteries according to coronary microvascular dysfunction. J Am Heart Assoc. 2022;11(15):e025171. doi:10.1161/JAHA.121.025171
20. Overuse of cardiac testing. Am Fam Physician. 2018;98(9):561-566. https://www.aafp.org/pubs/afp/issues/2018/1115/p561.html
21. O’Sullivan JW, Muntinga T, Grigg S, Ioannidis JPA. Prevalence and outcomes of incidental imaging findings: umbrella review. BMJ. 2018;361:k2387. doi:10.1136/bmj.k2387
22. The case against coronary artery calcium scoring for cardiovascular disease risk assessment. Am Fam Physician. 2019;100(12):734-735. https://www.aafp.org/pubs/afp/issues/2019/1215/p734.html
23. Maron DJ, Hochman JS, Reynolds HR, et al; ISCHEMIA Research Group. Initial invasive or conservative strategy for stable coronary disease. N Engl J Med. 2020;382(15):1395-1407. doi:10.1056/NEJMoa1915922
24. Centers for Medicare & Medicaid Services. Local coverage determination: coronary computed tomography angiography (CCTA) (L35121). Accessed 2026. https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?LCDId=35121
25. Insurance payers should cover selective coronary artery calcium testing in intermediate-risk primary prevention patients. Circulation. 2023. doi:10.1161/CIRCULATIONAHA.122.061193

HeartBuddi • Your heart. Own it.