P‐Wave Oversensing by the Implantable Cardiac Monitor During Paroxysmal Atrioventricular Block: What Is the Mechanism?

1. Introduction

An implantable cardiac monitor (ICM) is a device that is placed subcutaneously for the continuous monitoring of a patient's cardiac rhythm for up to several years [1]. This device is particularly valuable for identifying the cause of unexplained syncope, and it offers long‐term cardiac monitoring to determine any correlations between symptoms and cardiac rhythm [2, 3].

In this report, we describe a case in which P‐wave oversensing (PWOS) by the ICM prevented episodes of bradycardia or pauses from being detected during episodes of syncope.

2. Case Presentation

A 78‐year‐old male patient experienced multiple episodes of syncope following palpitations.

He underwent an extensive diagnostic evaluation, which included echocardiography, Holter monitoring, and computed tomography of the brain and coronary arteries. However, these examinations did not yield a conclusive diagnosis. After obtaining written informed consent, an ICM (LINQ II; Medtronic, Minneapolis, MN, US) was implanted in the fourth intercostal space at an angle of 45°.

Twenty‐four days after ICM implantation, the patient experienced another episode of syncope and activated the patient assist device function. Upon reviewing the remote monitoring data, it was confirmed that the ICM had not detected any pauses or bradycardia. However, the waveform recorded by the patient assist device indicated paroxysmal atrioventricular block (AVB), leading to a false‐negative diagnosis owing to PWOS by the ICM, and no recordings of pauses or bradycardia were preserved (Figure 1). The R‐wave amplitude at the time of ICM implantation was 1.0 mV. In accordance with the R‐ and P‐wave amplitudes, the minimum sensitivity of the R‐wave was adjusted from the nominal value of 0.035–0.2 mV using the remote programming system, which enabled us to remotely reprogram device alert settings without in‐office patient visits. [4]. Postadjustment, the paroxysmal AVB episodes were accurately captured without false negatives (Figure 2). Subsequently, a leadless pacemaker (Micra AV2; Medtronic) was implanted in the lower portion of the interventricular septum without any complications. Given the patient's active lifestyle (with hobbies including swimming and golf) and strong preference for a leadless pacemaker, the we chose Micra AV2. Although the patient currently presents with paroxysmal AVB, the potential for progression to permanent AVB necessitated choosing a device capable of maintaining AV synchrony. The ICM was then removed with no recurrence of syncope. What is the mechanism behind the occurrence of PWOS?

Figure 1.

Subcutaneous electrocardiogram during PWOS episodes. (A) Subcutaneous electrocardiogram recorded immediately after the patient activated the patient assist device following the syncope episode. (B) In the lower panel of (A), the P‐waves are labeled as A–D (first instance of PWOS) and a–c (second instance of PWOS). The time interval leading up to the first instance of PWOS is defined as < 1 >, and the time between the first and second instances of PWOS is defined as < 2 >. PWOS, P‐wave oversensing.

Figure 2.

Subcutaneous electrocardiogram accurately recording a pause episode without PWOS. By decreasing the minimal sensitivity from the nominal value of 0.035 to 0.2 mV, PWOS was prevented, allowing paroxysmal AVB to be accurately recorded as a pause episode. AVB, atrioventricular block; PWOS, P‐wave oversensing.

3. Discussion

ICMs are valuable tools for patients requiring long‐term cardiac monitoring [5]. This report describes a rare case where PWOS was detected through remote monitoring, which was triggered by the patient's activation of the patient assist device function immediately after experiencing syncope.

To investigate the underlying mechanism of PWOS, it is essential to understand variations in the R‐wave sensitivity of the ICM (LINQ II; Medtronic). Automatic adjustment of the R‐wave sensitivity of the ICM is illustrated in Figure 3A, and the process can be explained as follows:

• 1.After R‐wave detection, a predefined blanking period is applied, and the sensitivity is set to 65% of the electrocardiogram peak (with a maximum of 0.65 mV), here referred to as the initial sensitivity value. This initial sensitivity value is maintained throughout the delayed decrease in the predefined sensing threshold (decay delay).
• 2.Upon completion of this delay, the sensitivity decreases to 30% of the electrocardiogram peak within 1 s.
• 3.This sensitivity level is sustained until 1.5 s have passed since the R‐wave detection.
• 4.After 1.5 s, the sensitivity further decreases to 20% of the electrocardiogram peak.
• 5.The sensitivity gradually decreases to the pre‐defined minimal sensitivity (0.035 mV in the present case) at a rate of 80% per second. For instance, if the sensitivity is 0.2 mV at 1.5 s after R‐wave detection, it will decrease to 0.04 mV by 2.5 s.
• 6.Upon detection of the next R‐wave, the sensitivity resets to 65% of the electrocardiogram peak, establishing a new initial sensitivity value.

Figure 3.

Graph illustrating the automatic adjustment of R‐wave sensitivity. (A) Graph illustrating the automatic adjustment of the R‐wave sensitivity of the ICM. The sensitivity was adjusted according to the steps outlined in Section 1 through 6 of the text. The adjustable parameters are the decay delay (nominally 150 ms) and the minimal sensitivity (nominally 0.035 mV). (B–D) Graph illustrating the automatic adjustment of R‐wave sensitivity during the PWOS episodes. (B) Graph illustrating the automatic adjustment of R‐wave sensitivity during section < 1 >. The P‐wave located at position C in Figure 1B (occurring approximately 2000 ms after the preceding R‐wave) was not oversensed. In contrast, the P‐wave at position D in Figure 1B (occurring around 2700 ms after the R‐wave) was oversensed, marking the first instance of PWOS. Given that P‐wave sensitivity is assumed to remain stable, it was estimated that the sensitivity in this case was between 0.035 and 0.12 mV. (C) Graph illustrating the automatic adjustment of R‐wave sensitivity during section < 2 >, assuming that the first PWOS was 0.12 mV. The P‐wave at position b in Figure 1B occurs 1.0–1.5 s after the first instance of PWOS. When the R‐wave auto‐adjust sensitivity is applied, the sensitivity is reduced to 0.036 mV (30% of 0.12 mV). Since the P‐wave at position b was not oversensed, its amplitude must have been below 0.036 mV, which differs significantly from the assumed amplitude of 0.12 mV, leading to a discrepancy. (D) Graph illustrating the automatic adjustment of R‐wave sensitivity during section < 2 >, assuming that the first PWOS was 0.035 mV. Since the minimal sensitivity was set to 0.035 mV, the sensitivity in section < 2 > remained constant at this threshold, representing the boundary for oversensing. With slight variations in the P‐wave amplitude, the P‐waves at positions a and b were not oversensed, while the P‐wave at position c may have been oversensed. Therefore, it is estimated that the oversensed P‐wave had an amplitude close to 0.035 mV. ICM, implantable cardiac monitor; PWOS, P‐wave oversensing.

As shown in Figure 1, two instances of PWOS occurred in the present case. The first and second instances of PWOS correspond to the P‐waves at the positions marked as D and C, respectively, in Figure 1B. Herein, we discuss the two instances of PWOS in detail.

Considering the first instance of PWOS, applying the aforementioned R‐wave auto‐adjust sensitivity to section < 1 > results in the scenario depicted in Figure 3B. The P‐wave at the position marked as C in Figure 1B (occurring approximately 2000 ms after the preceding R‐wave) was not oversensed, whereas the P‐wave at the position marked as D in Figure 1B (occurring approximately 2700 ms after the R‐wave) was oversensed (the first instance of PWOS). Assuming that the P‐wave amplitude remained constant, the P‐wave amplitude in this case likely ranged between 0.035 and 0.12 mV (Figure 3B).

Considering the second instance of PWOS, assuming that the P‐wave amplitude at the position marked as D in Figure 3B (the first instance of PWOS) was 0.12 mV, applying the R‐wave auto‐adjust sensitivity to section < 2 > results in the scenario shown in Figure 3C. The P‐wave at the position marked as b in Figure 3C occurred 1.0–1.5 s after the first instance of PWOS. When applying the R‐wave auto‐adjust sensitivity, the value becomes 0.036 mV, which is 30% of 0.12 mV. Since the P‐wave at the position marked as b was not oversensed, the P‐wave amplitude must have been below 0.036 mV, which is a substantial deviation from the assumed value of 0.12 mV, creating a contradiction.

If we assume that the P‐wave amplitude at the position marked as D in Figure 3B (the first instance of PWOS) was 0.035 mV, applying the R‐wave auto‐adjust sensitivity to section < 2 > would result in the scenario shown in Figure 3D. Since the minimal sensitivity is 0.035 mV, the sensitivity in section < 2 > would remain constant at 0.035 mV. With slight fluctuations in the P‐wave amplitude, the P‐waves at the positions marked as a and b would not have been oversensed, but the P‐wave at the position marked as C may have been oversensed. From these observations, it is likely that the P‐wave amplitude in this case was around 0.035 mV (rather than 0.12 mV).

Detecting the R‐wave is crucial for identifying arrhythmias and distinguishing between tachycardia and bradycardia, but P‐wave information is also important for a more detailed diagnosis. A previous report demonstrated that pre‐implant mapping reliably achieved simultaneous sensing of both P‐ and R‐waves in all cases, with the P‐wave amplitude exceeding 0.03 mV in every instance [6].

In cases of paroxysmal AVB, as seen in the present case, the actual R‐wave is not sensed, leading the R‐wave auto‐adjust sensitivity to approach the minimal sensitivity level (nominally set at 0.035 mV). This can result in PWOS, potentially causing failure to detect bradycardia or pause episodes.

If the P‐wave amplitude is sufficiently high, adjusting the minimal sensitivity to a less sensitive threshold, while still reliably sensing the actual R‐wave, can help to prevent false‐negative AVB diagnoses caused by PWOS. In this case, as the actual R‐wave amplitude was around 1.0 mV, increasing the minimal sensitivity to 0.2 mV prevented PWOS, allowing for accurate diagnosis of AVB (Figure 2). Despite advancements in ICM technologies, there are still important false positive and false negatives. Tuning sensitive parameters relative to the sinus R‐ and P‐wave amplitudes for each patient can avoid these potential pitfalls.

To our knowledge, there are reports of T‐wave oversensing (TWOS) in ICM literature [6], but none on PWOS. In the case of TWOS, the R‐ and T‐wave are double‐counted and recorded as a tachycardia episode, making it easier to detect. PWOS, on the other hand, happens when an R‐wave is missing during AVB and sensitivity gradually sharpens. In this case, it is likely that this would not be recorded as a bradycardia or pause episode, making it difficult for us to detect PWOS. Moreover, since the R‐wave is missing, unless the P‐wave rate is significantly fast, it would not be recognized as a tachycardia episode either. In patients who experience syncope following ICM implantation but in whom no pauses or bradycardia episodes are recorded, the possibility of missing bradyarrhythmia due to PWOS must be considered. No pauses or bradycardia episodes were recorded in the present case; however, paroxysmal AVB was successfully detected because the patient activated the patient assist device after experiencing syncope. This strongly suggests that ensuring that patients and their family members understand the importance of activating their assist device function is important.

4. Conclusions

To our best knowledge, this is the first report of the detection of PWOS by an ICM, resulting in a false‐negative diagnosis of AVB. Measuring the P‐ and R‐wave amplitudes on subcutaneous electrocardiogram and adjusting the R‐wave minimum sensitivity on a case‐by‐case basis is crucial. Additionally, the diagnosis of AVB in the present case was possible due to the patient's timely activation of the patient assist device during symptom onset, underscoring the importance of educating patients and their family members on the appropriate use of this function during symptom manifestation.

Ethics Statement

This study was conducted according to the principles of the Declaration of Helsinki. The study was approved by the Institutional Review Board. The patient provided written informed consent.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.