Holter report: what every line on it actually means

A Holter is an ECG that keeps running after you have gone home.

The ECG done in a clinic is a photograph. The machine records about ten seconds of your heart — perhaps ten beats out of the hundred thousand you will produce that day — and if the thing your doctor is hunting for was not happening during those ten seconds, the tracing comes back normal and the question is exactly where it was. A two-day recording, by contrast, captures something like a quarter of a million beats. That difference is the entire reason the test exists.

If you have read how your heart works, this is the inspection of the wiring — the timer, the junction box, and the branches that carry each beat down into the walls — kept running for a day or more, to catch faults that come and go. It looks at nothing else. It cannot see the pipes, and it cannot see the rooms and doors.

This page explains what the machine is, why it was ordered, and what every line on the report means — in the order the report prints them.

What the machine actually is

There are two shapes of device and patients are handed either.

The traditional Holter is a small recorder, roughly the size of a mobile phone, worn on a belt or a neck strap, with five or seven wires running to sticky electrodes on the chest. It has to be kept dry.

The patch is a single adhesive strip stuck to the chest, with no wires and no box to carry. Current patches record several ECG leads at once, are waterproof, and send the recording to a computer for analysis. Most patients now get one of these, and most still call it a Holter, which is why this page does too.

The difference that matters to you is not the shape but the duration, and that is set by the doctor, not the device. One modern patch can be programmed for anything from a single day to fourteen. Your report states the period in its header — study time, 2 days — and the duration was chosen to match how often your symptoms happen. That choice is the single most important decision in the whole test, and this page comes back to it.

Why your doctor asked for one

Six reasons cover almost all of them.

Palpitations. A racing or thumping heartbeat that comes and goes, and is never present in the clinic. The recording is looking for a supraventricular tachycardia — an abnormally fast rhythm arising above the pumping chambers — or for extra beats, which are far commoner and far less serious.

Blackouts. A faint or a collapse, particularly one with no warning. Here the recorder is hunting for two opposite faults: a rhythm too fast to pump properly, or the heart pausing.

Blackouts in an older person. The same event, but the suspicion shifts towards long pauses and heart block — a beat that starts normally but is delayed or lost on the way down. That distinction decides whether a pacemaker is discussed.

A weakened or scarred heart. After a heart attack, or with a cardiomyopathy, the concern is short runs of a fast rhythm arising in the pumping chambers themselves — non-sustained ventricular tachycardia — which often stops on its own and may never be felt.

Atrial fibrillation that comes and goes. In known paroxysmal atrial fibrillation the recording measures the AF burden: what proportion of the days you spent in that rhythm. And after a stroke or a transient ischaemic attack with no explanation found, the monitor is hunting for atrial fibrillation that may be causing no symptoms whatsoever — because if it is there, the treatment that prevents the next stroke is a different one entirely. This is the one setting where a long recording is not a luxury: thirty days of monitoring found five times as much atrial fibrillation as a standard twenty-four-hour Holter.

After a procedure. To watch the conduction system for a few days in someone about to be discharged, where a device can transmit the recording remotely rather than waiting for the patch to be posted back.

The button, and the one way this test fails

Most patches carry a patient trigger — a button you press when you feel something. Your report has a line for it: patient triggered events, count.

Pressing it does not change what is recorded. Every beat is captured whether you press or not. What the button does is put a marker in the recording at the moment you felt the symptom, so the person reading it can go straight to those seconds and see what your heart was doing while you were feeling it.

That link — the symptom matched to the rhythm underneath it — is very often the most valuable thing the whole test produces, and it is the one part that depends entirely on you. A recording with a normal-looking trace and no markers cannot tell anyone whether your palpitations are dangerous. The same recording with a marker at 8.40 pm, and a perfectly normal rhythm running underneath that marker, has answered the question completely: whatever you are feeling, it is not an arrhythmia. That is a real result, and it cannot be obtained any other way.

So two things are asked of you, and neither is difficult:

• Press the button whenever you feel the symptom, even if you have felt it twenty times already. If you cannot press it, note the time on a piece of paper.
• Live your ordinary day. Climb the stairs you normally climb, do the housework, go to work, drink your usual coffee, sleep at your usual time. People often rest all week to protect the device, which is the one thing that can waste the test entirely — a heart that is never asked to do anything reveals nothing about what happens when it is.

What a Holter cannot see

Four honest limits, and the first is the one that causes the most confusion.

It cannot see blockages. Some reports carry an ST-segment analysis, the same line a treadmill test reads. On a recording made while you are walking, bending and lying on your side it is affected by posture and movement, and it is not used to rule blockages in or out. A normal Holter says nothing whatever about your arteries.

It cannot see the rooms or the doors. Pumping strength, wall thickness and leaking valves are an echo’s territory.

It only knows what happened while you wore it. This is the real limit. If your palpitations come once a fortnight, a forty-eight-hour recording has roughly a one-in-seven chance of being worn on the right day.

It cannot tell you why. It can show that your heart raced at 4 pm. It cannot say whether that was fever, anxiety, anaemia, a thyroid problem or an arrhythmia. That reading is the cardiologist’s, not the machine’s.

Your report, line by line

Reports differ in layout, but they carry the same information in roughly this order, and most of the lines will be normal and never discussed.

Study time and analysis time

Study time is how long the device was on you. Analysis time is how much of that recording the software could actually read.

The two are usually close, and when they are not it is worth knowing why. A patch that lifted at one corner, or an electrode that dried out, produces a stretch of noise rather than an ECG, and that stretch cannot be analysed. A report analysing 96 per cent of two days is a good recording. One analysing 60 per cent has holes in it, and an event may have happened inside one of them.

Heart rate: maximum, minimum, average

Four numbers, and they are the most useful plain-language lines on the page.

• Maximum, with the time it happened. Usually mid-morning or during exertion.
• Minimum, with the time. Usually in the small hours, during deep sleep.
• Average across the whole recording.
• The proportion of time spent tachycardic and bradycardic — that is, above and below the normal range.

A resting heart rate of 60 to 100 is conventionally normal, so a recording will nearly always contain some minutes above and below that. Brief tachycardia while climbing stairs and a slow rate at 4 am are both entirely expected, and their presence in the summary is not a finding.

What a cardiologist reads here is the pattern, not the numbers. A healthy heart rate falls substantially at night and rises during the day; that daily swing is a sign of a normally functioning nervous system. A rate that stays much the same around the clock — one that never comes down while you sleep — is itself worth a question, even when every other box on the report says zero. It is not a rhythm problem, and it is usually not a heart problem at all: fever, anaemia, an overactive thyroid, pain, dehydration, poor sleep, alcohol and simple deconditioning all raise the resting rate. But it is the kind of thing that gets missed, because the automated summary prints the number and passes no comment on it.

The extra beats — ectopics

This is the section that alarms people most, and it is almost always the least important.

An ectopic is a beat that arrives early, out of turn, from somewhere other than the heart’s own timer. The report divides them by where they came from:

• SVE, or PAC — a supraventricular extra beat, arising in the upper rooms.
• VE, or PVC — a ventricular extra beat, arising in the lower pumping rooms.

and then subdivides each into isolated beats, couplets (two in a row), triplets, and runs.

Almost every heart does this. Extra beats are found in healthy people of every age, and a report that lists none at all is more unusual than one that lists a few hundred.

The reason they are worth explaining is that many people can feel them, and what they feel is frightening out of proportion to what is happening. Patients describe it as a skipped beat — a beat that seems to go missing, followed by a thud in the chest.

The mechanism is worth knowing, because it explains why the sensation is so strong and why it is not a sign of a struggling heart. An extra beat arrives too early, before the pumping room has had time to fill, so it ejects very little blood — which is why you do not feel it at all. The heart then waits slightly longer than usual before the next normal beat, and during that longer pause the room fills more than it usually would. The beat that follows is therefore pumping a fuller chamber, and it lands harder. The thud you feel is not the abnormal beat. It is the perfectly normal beat after it, made forceful by the extra filling time.

Burden — the percentage that actually decides

Beneath the counts, the report gives a burden: the extra beats as a percentage of your total beats. This is the number that matters, and it is the reason the raw count is so misleading.

A heart beating around 90 times a minute produces roughly a quarter of a million beats in two days. Against that, 94 extra beats is well under one per cent. Even a thousand extra beats in a day is about one per cent.

That last row is the one genuinely useful threshold in this part of the report. Above roughly ten per cent, frequent ventricular extra beats can, over time, reduce the pumping strength — and that weakening is reversible if the ectopics are brought under control, which is why the burden is worth taking seriously at that level and not below it.

The five boxes

Most reports print a row of summary boxes: pauses, AV block, atrial fibrillation, supraventricular tachycardia, ventricular tachycardia.

These five are the report. Everything else is counting. Each one names a finding that changes what happens next, which is why they are given their own boxes and why they are worth understanding even when — as in most reports — every one of them reads zero.

A zero in these boxes is the answer you want. It means the recording looked for that finding across every beat and did not find it. There is nothing further to read into it, and nothing to look up.

Pauses

A pause is a gap between beats — the heart simply not producing one for a moment. The report gives a count above a set threshold, often around two seconds, and the length of the longest.

That threshold is a setting on the analysis software, not a medical dividing line, which is worth knowing before you read anything into the exact number printed beside it.

Short pauses are common and often harmless, particularly during sleep in fit people. Long ones are not. If the heart stops producing beats for long enough, blood stops reaching the brain, and what the person experiences is a sudden faint with no warning — sometimes while sitting still, which is what separates it from an ordinary faint. Broadly, a pause beyond three seconds occurring with symptoms, or beyond six seconds even without them, is where a pacemaker enters the conversation.

AV block — the beat that does not get through

Each beat begins at the heart’s own timer in the upper rooms, travels to a junction box between the upper and lower halves, and passes from there down branches into the pumping chambers. AV block is a fault at that junction, and it comes in three degrees according to what happens to the signal when it arrives:

• First degree — every beat gets through, but slowly. It is a delay, not a loss. It is common, often needs nothing at all, and is frequently a normal finding in a fit person.
• Second degree — some beats get through and some are dropped. This one has two forms that look similar on the page and mean very different things.
• Third degree, or complete heart block — the signal reaches the junction and does not pass at all. The lower chambers fall back on a slow escape rhythm of their own. This causes fainting, severe tiredness and breathlessness, and it needs a pacemaker.

The two forms of second-degree block are the distinction the report is really being read for:

Mobitz I, also called Wenckebach, is a junction that tires progressively — each beat takes a little longer to get through than the last until one fails to make it, after which it recovers and the cycle begins again. It is usually a fault at the junction itself, it is often found in healthy young people and during sleep, and it very often needs nothing.

Mobitz II is a junction that conducts normally and then drops a beat without any warning at all. Nothing lengthens, nothing builds up; a beat simply fails. The fault is usually below the junction, in the branches, and it carries a real risk of progressing to complete block. It is treated as a pacemaker indication in a way Mobitz I is not.

So: the gradual failure is usually watched, and the sudden one is usually paced. That is why the two words on your report are worth telling apart.

Atrial fibrillation

Atrial fibrillation is the upper rooms losing their coordinated beat and quivering instead, so the pulse becomes completely irregular. It matters because blood that is not being moved briskly through those upper rooms can clot, and a clot leaving the heart can travel to the brain.

Where atrial fibrillation is already known to come and go, the report gives an AF burden — the proportion of the recording spent in that rhythm — and a chart showing when the episodes fell. That number tracks whether treatment is working.

Where it was not known, finding it is the most consequential thing a Holter can do, because it changes treatment from something that eases symptoms to something that prevents a stroke. This is also why the recording is deliberately run long after an unexplained stroke: brief, symptomless episodes are easy to miss in a single day.

A short episode found only on a monitor is weighed rather than treated automatically. Blood thinners lower stroke risk and raise bleeding risk, and in people whose atrial fibrillation was picked up only by a device the two effects are close enough that the decision turns on your own stroke risk score and your own bleeding risk — which is a conversation, not a calculation.

Supraventricular tachycardia

Supraventricular tachycardia, or SVT, is a fast rhythm arising above the pumping chambers. A true SVT tends to start and stop abruptly, run at 140 to 220 beats a minute, and be felt as a sudden racing that switches on and off like a switch.

And here is the line on the report most likely to frighten you unnecessarily. The box counts episodes, and an episode may be very short indeed. A report headed supraventricular tachycardia — 2 episodes, 129 to 153 bpm can turn out, in the detail printed beneath it, to be four beats lasting 1.4 seconds and five beats lasting 1.8 seconds.

Runs of that length occur in normal hearts, cause nothing, and are not treated. They appear in the box because the software is counting faithfully — three or more consecutive fast beats from above the chambers meets the definition — not because anything is wrong.

So read the detail beneath the box before you read the box. How many beats, how long, how fast, and above all whether you felt anything at the time. A sustained run that coincides with a marker you pressed is a finding. Four beats you never noticed is not.

Ventricular tachycardia

Ventricular tachycardia is a fast rhythm arising in the pumping chambers themselves. Three or more consecutive beats above 100 a minute is the definition; if it stops on its own within thirty seconds it is called non-sustained VT.

This is the box that carries the most weight, and it is read almost entirely in the light of the rest of your heart. Short runs in a completely normal heart are common and are often treated as a variant of frequent ectopics. The same runs in a heart that has been scarred by a previous heart attack or affected by a cardiomyopathy are a different matter, because scarred muscle conducts electricity unevenly and can sustain a dangerous rhythm. In that setting the finding leads to a proper assessment of the heart’s pumping function and a discussion about protection against sudden rhythm disturbances.

The charts

Most reports include graphs across a page: a heart-rate trend showing maximum, average and minimum through each day and night, and separate burden charts for supraventricular and ventricular extra beats.

The trend graph is the one worth looking at. What it shows at a glance is the daily swing — the band riding high through the day and dropping through the night — and any single spike that stands away from the rest. The burden charts show whether extra beats were scattered evenly or clustered into particular hours, which occasionally points at a trigger.

The strips at the back

The last several pages are printed ECG strips: numbered samples, each labelled with a time and often with a tag such as isolated SVE or sinus tachycardia.

These are samples, not incidents. A recording of a quarter of a million beats cannot be printed, so the software prints the maximum, the minimum, anything it flagged, and a scattering of routine hourly samples to show what your ordinary rhythm looked like. Twenty numbered strips does not mean twenty events. Most of them are there precisely because they are normal.

The small print on each — 400 ms, 8 s, 42 mm/mV, Lead II — is the scale and the viewing angle, the settings the tracing was drawn at. It is for the person reading the strip. There is nothing in it for you.

Lines you can safely ignore

Present on most reports, and rarely discussed with anyone:

• Total QRS complexes — the total number of beats counted. Impressive, and of no clinical use to you.
• Longest RR interval — the single longest gap between two beats, reported whether or not it was abnormal.
• Is paced — simply recording whether you already have a pacemaker, so the software knows to look for its signals.
• Bigeminy and trigeminy — patterns in which extra beats alternate regularly with normal ones. They describe the arrangement of ectopics already counted above, and change nothing.
• Heart rate variability, SDNN — a research measure of beat-to-beat variation.

What happens after each finding

The report is a description. The plan comes from the person who ordered it, and it depends on the rest of your heart as much as on the recording. Broadly:

None of these is a decision made from the report alone, and none of them is made in a hurry.

When the report is normal

Most Holter reports are normal, or contain only findings of the kind described above as common and untreated. If that is yours, two things are true at once and both are worth hearing.

The first is that this is genuine good news. Across the days you wore it, your heart was watched beat by beat — not sampled, not estimated — and none of the five findings that change management was present. That is a far stronger reassurance than a ten-second tracing can give, and it is worth taking.

The second is that a normal recording answers a question only for the days it covered. If the symptom that sent you for the test never happened while you were wearing it, then the test has not yet answered your question, and a clean report is not the same as an explanation. This is not a failure of the test or of you. It is arithmetic: a recording can only catch what occurs inside it, which is why the duration is matched to how often symptoms come, and why guidelines reserve short Holter recordings for symptoms that occur roughly weekly or more.

So the sentence that matters is this: if your symptoms continue, go back and say so. The next step is usually a longer recording rather than a repeat of the same one — the same patch worn for two weeks, a monitor worn for a month, or for genuinely rare and serious events a tiny recorder placed under the skin that watches for years. Symptoms that persist after a normal Holter are a reason to extend the search, not to end it.

Common questions

My Holter report says supraventricular tachycardia. Do I need treatment?

Not on the strength of that line alone. The box counts episodes, but it does not tell you how long they lasted — and a great many reports headed "supraventricular tachycardia" turn out, in the detail beneath, to be runs of four or five beats lasting a second or two. Brief runs like that are found in perfectly normal hearts and are not treated. What decides treatment is whether an arrhythmia is long enough and frequent enough to explain your symptoms, not whether the word appears on the page. Read the detail under the box, and take the report to the doctor who ordered it.

Is it normal to have hundreds of extra beats in a day?

Yes. Almost every heart produces extra beats, and a healthy heart beats roughly 100,000 times a day, so a few hundred extras is a fraction of one per cent. This is why the report gives a burden percentage as well as a count: the count sounds alarming and the percentage almost never is. Ninety-four extra beats out of 268,000 is under 1 per cent. Ventricular extra beats begin to matter to a cardiologist when the burden reaches about 10 per cent, because at that level the extra beats themselves can gradually weaken the pump.

My Holter was normal but I still get palpitations. What happens now?

A normal recording means nothing abnormal happened during the days you wore it. If your symptom did not occur in that window, the test has not answered the question — it has only shown that your heart is normal the rest of the time, which is genuinely worth knowing. The usual next step is a longer recording rather than reassurance, because the same patch can be worn for up to fourteen days, and monitors exist that run for weeks or, implanted, for years. The rule of thumb is that the recording has to be longer than the gap between your symptoms.

Can I bathe and sleep normally with the monitor on?

It depends which device you were given. Modern adhesive patches are waterproof and have no wires, so you shower, sleep and work as usual. Older Holter recorders have wires running to electrodes on the chest and a small box you carry, and those must be kept dry. Either way the instruction that matters is the same: live your ordinary day. The test is only useful if it records you doing the things that normally bring the symptom on, so resting all week to protect the device is the one thing that can waste it.

What happens if I forget to press the button?

The recording still captures every beat, so nothing is lost from the tracing itself. What is lost is the link between what you felt and what your heart was doing at that moment — and that link is often the single most useful thing the test can produce. If you feel something and cannot press the button, write down the time. A note on a piece of paper does the same job.

Does a Holter show blockages in my arteries?

No. A Holter watches the electrical activity of the heart, not the arteries that supply it. Some reports include an ST-segment analysis, which is the same line a treadmill test reads, but on an ambulatory recording it is affected by posture, movement and electrode position, and it is not relied on to rule blockages in or out. If the question is whether your arteries are narrowed, that is what a treadmill test, a CT coronary angiogram or an angiogram is for.

How many days should the monitor be worn?

Long enough to catch the thing being looked for. The same patch can be programmed from one day to fourteen, and the doctor sets the duration to match how often your symptoms occur — daily symptoms need only a day or two, while symptoms that come once a month will not be caught in forty-eight hours. After an unexplained stroke or a transient ischaemic attack the aim is different again: the monitor is hunting for atrial fibrillation that may cause no symptoms at all, and there the recording is deliberately run for around two weeks.