Implantable Loop Recorders: A San Diego Cardiologist's Guide to When, Why, and What to Expect

A few times a month, a patient comes into my clinic in Encinitas with the same story. They have had a fainting spell, or a brief episode of near-fainting, or a run of fluttery palpitations that came and went in a minute. Their primary care doctor ordered an ECG, which was normal. They wore a Holter monitor for a day or two, which caught nothing. They saw me and I ordered a thirty-day event monitor, which also caught nothing. Now they are frustrated, and rightly so, because their symptoms are real and intermittent and keep slipping through the nets. This is exactly the kind of situation an implantable loop recorder (ILR) is designed for.

An ILR is a small subcutaneous heart monitor that stays in place for years. Current devices are about the size of a small paperclip, weigh three to four grams, and continuously watch your heart rhythm. They are not defibrillators and they do not pace. They only record. What they give us that no surface monitor can is time. If your event happens once every six months, a two-week patch will miss it. An ILR will not.

This is a long read because patients deserve to know more than "we're going to implant a chip." I want to walk you through when an ILR is actually indicated, what the evidence shows, what the major device choices look like, what the procedure and follow-up entail, and some of the real limitations of the technology.

The Short Version

An ILR is the right choice when symptoms are infrequent but concerning, and shorter-term monitoring has been unrevealing. The two biggest indications are recurrent unexplained syncope of suspected arrhythmic origin and cryptogenic stroke where we are trying to detect hidden atrial fibrillation. Current ILRs last two to six years depending on the model, are MRI-conditional at 1.5-T or 3-T, and detect AF with 95 to 99 percent sensitivity. In cryptogenic stroke the landmark CRYSTAL-AF trial showed that an ILR detected AF in 30 percent of patients at three years versus 3 percent with conventional monitoring. For unexplained syncope, prospective data show diagnostic yield of 55 percent with an ILR versus 19 percent with conventional testing. False-positive AF alerts are real and common (46 to 86 percent depending on setting) and require a human to review tracings. And the LOOP Study reminds us that detecting AF is not always the same thing as preventing stroke, so selection matters.

What an ILR Actually Is

An ILR, sometimes called an insertable cardiac monitor (ICM), is a thin metal-and-plastic device implanted just under the skin of the left anterior chest, usually between the fourth intercostal space and the sternum. It has two electrodes on its surface that read the heart's electrical signal through the skin. Early generations of the device (Medtronic's original Reveal) were the size of a pack of chewing gum and required a short surgical implant. Current ILRs are an order of magnitude smaller and go in through a tiny incision with a delivery tool, usually in under ten minutes in the office or a short procedure room. No general anesthesia. Most patients drive home the same day.

The device streams data via Bluetooth to a bedside monitor or a smartphone app, and the monitor uploads securely to the manufacturer's cloud. My office reviews alerts continuously; high-priority alerts trigger a same-day call. The device can also be interrogated in clinic, much like a pacemaker.

When an ILR Is Indicated

The 2017 ACC/AHA/HRS Syncope Guidelines give a Class IIa recommendation (reasonable to use) for an ILR in selected ambulatory patients with syncope of suspected arrhythmic origin. Current indications, synthesized across guidelines and the 2020 EHRA/HRS expert consensus, include recurrent unexplained syncope of uncertain origin (especially when ambulatory monitoring has been inconclusive); high-risk syncope after a comprehensive evaluation has found no cause; cryptogenic stroke or TIA for detection of paroxysmal AF; palpitations, presyncope, or dizziness when symptoms are infrequent; suspected AF and post-AF ablation monitoring; and frequent premature ventricular contractions or non-sustained ventricular tachycardia requiring long-term assessment.

If your symptoms happen once or twice a week, you do not need an ILR. A patch or an event monitor will capture them. If your symptoms happen every three to six months, an ILR is often the only way to catch the culprit rhythm in the act.

ILR in Cryptogenic Stroke: The CRYSTAL-AF Story

About a third of ischemic strokes are classified as cryptogenic, meaning no clear cause is found after standard evaluation. Paroxysmal atrial fibrillation is one of the likely culprits in a substantial fraction of these, but AF can be silent, brief, and easy to miss on short-term monitoring.

The CRYSTAL-AF trial, published in the New England Journal of Medicine in 2014, randomized 441 adults with recent cryptogenic stroke or TIA to an ILR versus conventional follow-up. AF detection rates with the ILR were 8.9 percent at six months, 12.4 percent at twelve months, and 30 percent at three years, versus 1.4 percent, 2.0 percent, and 3 percent respectively with conventional follow-up. The hazard ratio at six months was 6.4 (95 percent CI 1.9 to 21.7). The trial was not designed to show that detecting AF reduced recurrent strokes, but it dramatically changed our ability to find the AF in the first place.

A 2022 meta-analysis in the American Journal of Cardiology pooled randomized data and found an odds ratio of 5.8 (95 percent CI 3.2 to 10.2) for AF detection with an ILR compared to usual care. The 2025 ANTARCTICA pooled analysis in Stroke showed that AF detection was more likely in cryptogenic stroke or TIA patients than in other stroke subtypes (adjusted OR 1.90), with a shorter time to detection (median 65 versus 169 days). About 47 percent of detected AF in that analysis was felt to be pathogenic, meaning likely to have caused the stroke and actionable with anticoagulation.

The 2023 ACC/AHA/ACCP/HRS AF Guidelines and the 2024 ACC Expert Consensus Decision Pathway on arrhythmia monitoring after stroke both support extended monitoring in cryptogenic stroke, with an ILR reasonable when maximum sensitivity is the goal. Patient selection centers on CHA2DS2-VASc score of 2 or greater and whether the patient is a candidate for anticoagulation.

ILR in Unexplained Syncope

Syncope is one of the most frustrating symptoms in cardiology because the diagnostic workup can be long, expensive, and unrevealing. An ILR provides the highest diagnostic yield of any single tool we have for unexplained syncope.

A meta-analysis of 49 studies found a diagnostic yield of 26.5 percent for arrhythmogenic syncope with ILR monitoring. A prospective study comparing ILR with conventional testing (external loop recorder, tilt-table test, electrophysiology study) showed 55 percent diagnostic yield with the ILR versus 19 percent with conventional testing. A 2026 real-world study of 388 patients found that ILRs detected arrhythmias in 62.1 percent of patients overall, with indication-fulfilling diagnoses in 39.9 percent, and therapeutic interventions (pacemaker, anticoagulation, ablation, medication change) prompted in 40.2 percent.

What this means in practice: if you have had two or more unexplained faints, have been through an ECG, a Holter, an echocardiogram, and perhaps a tilt-table test without a clear answer, the odds that an ILR will identify the rhythm causing your events are substantial. For the 60 percent or so in whom an ILR does not find a dangerous rhythm, that information is also useful. It reassures us that we are not missing a life-threatening arrhythmia and redirects our attention to non-cardiac causes of syncope such as orthostatic intolerance, vasovagal reflex, or medication effects.

Current Devices

Six ILRs dominate current U.S. practice. I do not have a strong preference among them and patient-specific factors usually drive the choice. The key differentiators are battery life, MRI compatibility, AF detection sensitivity, and device size.

The Medtronic LINQ II has a 4.5-year battery, is MRI-conditional at 3-T, and is among the smallest devices at 3.4 grams and 8 × 45.1 × 4.2 mm. Its AF sensitivity is reported at 99 percent. The Boston Scientific LUX-Dx II and II+ have a 3-year battery, 3-T MRI compatibility, 97.6 to 100 percent AF sensitivity, and a 3.0-gram footprint. The Abbott Confirm Rx runs for 2 years and is MRI-conditional at 1.5-T only. The Abbott Assert-IQ extends to 3 years at 3-T compatibility, and the Assert-IQ EL+ variant pushes battery life to 6 years. The Biotronik Biomonitor III and IIIm offer 5.5-year battery and 1.5-T or 3-T MRI compatibility, with a slightly larger form factor at 4.0 grams.

Battery life matters most if we anticipate needing long-term monitoring (cryptogenic stroke, rare syncope, post-ablation). MRI compatibility matters for patients who have frequent imaging needs or known future MRI-requiring conditions. Sensitivity differences are small enough in current devices that I rarely let them decide the choice.

MRI Compatibility: The Practical Details

All current ILRs are MRI-conditional, meaning safe in the scanner under specified conditions, typically a 1.5-T or 3-T static field. The 2017 HRS Expert Consensus recommends pre-MRI device interrogation to download recorded events, since the scan can create image artifacts the device may misread as arrhythmias, and post-MRI interrogation to clear those artifacts. Patient safety risk from the scan itself is minimal with current devices. The scanner image may show a small local artifact near the device, which your radiologist will know how to account for.

If you have an ILR and you need an MRI, mention it to the ordering provider and ask them to coordinate with your cardiology office. A 15-minute pre-scan interrogation is usually all it takes.

The Implant Procedure

An ILR implant is a short outpatient procedure. After local anesthesia over the left anterior chest (usually around the fourth intercostal space, one to two centimeters left of the sternum), a small nick is made and the device is inserted through a delivery tool. The incision is closed with a single suture or surgical glue. Most patients are in and out of the procedure room in 10 to 15 minutes.

You go home the same day. You can shower the next day. Light activity is fine immediately; avoid heavy lifting or strenuous arm movement for a week. Soreness and minor bruising at the site are common for a few days and respond to acetaminophen. Infection risk is low (well under 1 percent) but not zero, and any persistent redness, warmth, drainage, or fever should be reported.

The patient-facing bedside monitor or smartphone app is usually set up the same day or sent home with you. I tell patients to plug it in, leave it alone, and call us only if the device chirps or the app flags a concern.

False Positives and Why Human Review Still Matters

This is one of the underappreciated realities of ILR monitoring. The automated AF detection algorithms in current devices are good but not perfect. The 2023 HRS/EHRA/APHRS/LAHRS expert consensus on remote device management notes that false-positive AF detection occurs in 46 to 86 percent of episodes flagged by the algorithm, depending on device settings and patient population.

What this means is that a device alert is not the same as a diagnosis. Every alert should be reviewed by a trained human (a device clinic nurse, an electrophysiology nurse practitioner, or a cardiologist) who confirms whether the tracing is actually AF. PVCs, myopotentials from pectoral muscle activity, ectopic beats, undersensing, and oversensing can all trigger false alerts. A clinic that is structured to review tracings promptly and act only on true events is a critical part of the care model.

The consensus document recommends tailored alert programming based on clinical indication, manual electrogram review to confirm events, maximized sensitivity in cryptogenic stroke patients (where missing real AF is high-stakes), and an emphasis on symptom-rhythm correlation in syncope patients (so we can tie a specific symptom to a specific rhythm).

The LOOP Study and the Limits of Screening

A 2025 JAMA review underscored an important caveat. The LOOP Study, published a few years earlier, randomized older adults at elevated stroke risk to ILR implantation for AF screening versus usual care. AF was detected in 31.8 percent of the ILR group versus 12.2 percent of controls, and more patients in the ILR group started anticoagulation. But the stroke rate did not fall significantly (HR 0.80, 95 percent CI 0.61 to 1.05).

This matters for how we think about ILR use in screening populations who do not have a syncope history or a cryptogenic stroke. Detecting brief, subclinical AF episodes in otherwise asymptomatic older adults does not automatically translate into fewer strokes. The benefit-versus-risk calculation shifts. In cryptogenic stroke we are searching for a known culprit in a patient who has already had an event. In pure population screening we may simply be finding brief episodes of AF that do not carry the same stroke risk. Guidelines reflect this: ILR is supported for cryptogenic stroke and for unexplained syncope. Blanket screening of the general older population is not on the ACC/AHA recommended list.

Palpitations and Other Indications

The 2024 American Family Physician review on palpitations recommends an ILR when external recorders cannot capture events or when symptoms are too infrequent to reasonably catch any other way. Average battery life is about 3 years and diagnostic yield for palpitations runs 80 to 90 percent.

For frequent PVCs or non-sustained ventricular tachycardia that we want to quantify over months, an ILR is a reasonable tool, especially in patients who cannot tolerate a long-term patch. For post-AF ablation surveillance, an ILR is used selectively when symptom recurrence is ambiguous and we want continuous monitoring.

What You Will Actually Feel and Notice

Most patients notice the device under the skin for the first couple of weeks. It feels like a small firm object when you press on it. After a few weeks, the tissue heals around it and it becomes much less noticeable. Some patients forget it is there entirely.

You will not feel the device recording. It does not vibrate. It does not buzz. If you have symptoms, you can push a button on the bedside monitor (or tap a button in the smartphone app) to mark that moment, which tells us to look at the tracing from the corresponding time. That symptom-activated marker is one of the most useful tools the device offers and I encourage patients to use it liberally.

How Long It Stays In

An ILR stays in for as long as we need it and as long as its battery lasts. If we find the diagnosis we were looking for in month six, the device can come out at that point. If your syncope was ultimately captured and you had a pacemaker implanted, the ILR is removed at the same setting. If nothing is captured by the end of its useful battery life, the device can either be left in place (it is well tolerated) or removed through a small procedure similar to the implant.

Some patients ask whether leaving an expired device in place is safe. The devices are biocompatible and long-term tolerance is excellent. The main reason to remove one is preference (some patients want it out once it stops functioning) or the rare site issue.

What I Actually Tell Patients in Clinic

If you have had unexplained syncope with no answer after a reasonable workup (ECG, Holter, echocardiogram, maybe a tilt-table test) and you have had more than one event, I will usually recommend an ILR rather than another round of short-term monitoring. The yield of a thirty-day event monitor on patients who have already had a clean Holter is low, and continuing to circle that track while you risk another event does not serve you well.

If you have had a cryptogenic stroke and the workup has not found AF after 48 hours of inpatient telemetry plus a 30-day outpatient patch, and your CHA2DS2-VASc score is 2 or higher and you are a reasonable anticoagulation candidate, I will usually recommend an ILR to maximize the chance of catching paroxysmal AF.

If you have infrequent, concerning palpitations that have never been captured but that sound like a real arrhythmia in the history (sudden onset, sudden offset, near-syncope at times), an ILR is often the most efficient next step.

I do not recommend ILRs for patients with vasovagal syncope who have a clear prodrome, warning symptoms, and a typical pattern, because the diagnosis does not hinge on catching an arrhythmia. I also do not recommend ILRs in patients who have frequent palpitations multiple times per week, because a shorter-term patch will almost certainly capture them at a fraction of the cost and effort.

Cost, Insurance, and Access

An ILR implant is covered by Medicare and most private insurance for appropriate indications. Prior authorization is often required for device implantation. Copays vary. In my practice we walk patients through the financial picture before scheduling. I mention this because in rare cases the insurer pushes back on the indication (especially for frequent palpitations without documented symptom-rhythm failure on shorter monitors). Strong documentation of prior negative workups matters.

Bottom Line From My Clinic

An ILR is the right tool when you need time on your side. If your symptoms are infrequent but potentially serious, or if you have had a stroke of unclear cause and we are trying to rule out silent AF, an ILR gives us a monitoring horizon of years instead of days. Current devices are small, safe, and MRI-compatible. The procedure is short. The main costs are vigilance (someone reviewing alerts), patience (the event may still take months), and acceptance that a chunk of alerts will be false positives that your care team will need to sort through.

If you are a patient of mine wrestling with unexplained syncope or a cryptogenic stroke workup, and the shorter-term monitors have not given us an answer, an ILR is often the move that gets us off the diagnostic treadmill. The data are not perfect, the yield is not 100 percent, and the downstream implications of finding brief AF are still being studied. But for the right patient in the right clinical context, an ILR has become a high-leverage diagnostic tool in modern cardiology practice, and I am glad we have it.

References

Sanna, Tommaso, Hans-Christoph Diener, Rod S. Passman, et al. "Cryptogenic Stroke and Underlying Atrial Fibrillation." New England Journal of Medicine 370, no. 26 (2014): 2478–2486.

Joglar, Jose A., Mina K. Chung, Anastasia L. Armbruster, et al. "2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation." Journal of the American College of Cardiology (2024).

Shen, Win-Kuang, Robert S. Sheldon, David G. Benditt, et al. "2017 ACC/AHA/HRS Guideline for the Evaluation and Management of Patients With Syncope." Journal of the American College of Cardiology 70, no. 5 (2017): e39–e110.

Ko, Darae, Qiying Dai, David B. Flynn, et al. "Meta-Analysis of Randomized Clinical Trials Comparing the Impact of Implantable Loop Recorder Versus Usual Care After Ischemic Stroke for Detection of Atrial Fibrillation and Stroke Risk." American Journal of Cardiology (2022).

Yaghi, Shadi, Luciano A. Sposato, Liqi Shu, et al. "Device-Detected Atrial Fibrillation in Patients With and Without Cryptogenic Ischemia: The ANTARCTICA Pooled Analysis." Stroke (2025).

Spooner, Michael T., Steven R. Messé, Seemant Chaturvedi, et al. "2024 ACC Expert Consensus Decision Pathway on Practical Approaches for Arrhythmia Monitoring After Stroke." Journal of the American College of Cardiology (2025).

Ferrick, Aysha M., Satish R. Raj, Thomas Deneke, et al. "2023 HRS/EHRA/APHRS/LAHRS Expert Consensus Statement on Practical Management of the Remote Device Clinic." Heart Rhythm (2023).

Plappert, Christian, Philippe Lacour, Abdul Shokor Parwani, et al. "Expanding the Role of Implantable Loop Recorders: Diagnostic and Therapeutic Yields Across Seven Clinical Indications in 388 Real-World Patients." Journal of Clinical Medicine (2026).

Gauer, Robert L., Melanie F. Thomas, and Robert A. McNutt. "Palpitations: Evaluation, Management, and Wearable Smart Devices." American Family Physician (2024).

Published on damianrasch.com. This article is for educational purposes and is not medical advice. Always discuss your individual situation with your physician.