Electrocardiographic characteristics for the prediction of under‐sensing in implantable loop recorders

Abstract Background Under‐sensing (US) in implantable loop recorders (ILRs) interferes with the accurate diagnosis of arrhythmia, but there is little information available on the details of US of ILRs. The aim of this study was to clarify the frequency of US in patients with ILRs and to investigate the predictors of US in ILRs prior to implantation. Methods and Results We studied 46 consecutive patients implanted ILR. During the mean follow‐up period of 499 ± 363 days, 15 events of US were observed in five patients. There were no significant differences in patient characteristics between patients with and without US. In standard 12‐lead electrocardiogram (ECG), QRS complex amplitude in anterolateral chest leads (V2 to V5) were significantly lower in patients with than without US (V2: 0.88 [0.66, 1.22] mV vs. 1.67 [1.23, 2.29] mV, p = .010 V3: 1.25 [1.00, 1.26] mV vs. 1.90 [1.41, 2.29] mV, p = .013; V4: 1.14 [0.96, 1.38] mV vs. 1.93 [1.65, 2.64] mV, p = .023; V5: 0.57 [0.50, 0.75] mV vs. 1.60 [1.20, 1.98] mV, p = .011, respectively). ROC curve analysis showed that cut‐off values of 1.30 mV of QRS complex amplitude in V2, 1.26 mV of that in V3, and 0.75 mV of that in V5 had moderate accuracy for predicting US (V2: sensitivity 68%, specificity 100%, area under the curve [AUC] 0.86; V3: sensitivity 85%, specificity 80%, AUC 0.85; V5: sensitivity 98%, specificity 80%, AUC 0.85, respectively). Conclusions US was observed in 10.9% patients with an ILR. QRS complex amplitude in anterolateral chest leads (V2 to V5) on ECG might be useful for predicting US in patients with ILRs.


| Device implantation and programing
Consecutive patients who were implanted with an ILR (Reveal LINQ®; Medtronic plc) between October 2016 and February 2021 were enrolled in this study. Indications for implantation included recurrent syncope of unknown origin and cryptogenic stroke. All ILRs were implanted in the left sternal border of the fourth or fifth intercostal space by a cardiologist, following standard operating procedures. In all cases, R-wave amplitude of the ILR at implantation was >0.2 mV (as recommended by the manufacturer) and the ILR was programed in the nominal setting: (i) sensitivity, 0.035 mV; (ii) tachycardia detection, >200 beats per minute (BPM) (16 consecutive beats); and (iii) bradycardia detection, <30 BPM (4 consecutive beats), and (iv) pause detection, 3 s.

| Data collection
Patient information was collected from medical records, including age, sex, body mass index (BMI), and medical history. In all patients, echocardiography and standard 12-lead electrocardiogram (ECG) were evaluated before ILR implantation. In echocardiography, left ventricular diastolic diameter, left ventricular ejection fraction, and left atrial diameter were measured by the standard method. Heart angle and ILR angle were manually measured in a chest posteroanterior X-ray image. The heart angle was defined as the angle between the horizontal line and the line dividing the ventricular area symmetrically via the apex. The ILR angle was similarly defined. 11 In standard 12-lead ECG, QRS duration and electrical axis deviation were measured by an electrocardiograph analysis system (MBF-1000; FUKUDA DENSHI, Inc.). Moreover, QRS complex amplitude in standard 12-lead ECG was defined as the absolute value of the potential formed from the Q, R, and S waves ( Figure 2) and measured by an electrocardiograph analysis system (MBF-1000; FUKUDA DENSHI, Inc.). Finally, QRS amplitude of the ILR was measured by a pacing programer (Programmer 2090; Medtronic, Inc.) before the patient was discharged.

| Follow-up
All patients were provided with a remote monitoring system (My Care Link; Medtronic) that automatically downloads key data stored by the ILR and sends them to the website daily. Clinical engineering specialists explained the proper operation of the remote monitoring system to the patients before discharge. In the case of arrhythmia events, the clinical engineer reported the event to the electrophysiologist according to the institution's protocol. For evaluation of the proportion of true and false positive arrhythmia detections, all recordings were reviewed by the same electrophysiologist who was responsible for clinical management. The endpoint of this study was US (defined as failure of the device to sense intrinsic R-wave) of ILRs. No further sensitivity adjustment was made during the follow-up period. The study protocol was approved by the Ethics Committee of Osaka General Medical Center in accordance with the Helsinki Declaration. Since this study was a retrospective study, we used the opt-out system according to the ethics guidelines on human medical research by the Japanese government. The device manufacturer did not sponsor or influence the study in any way.

F I G U R E 1
False cardiac arrest events detected by implantable loop recorder. The waves (asterisks) with abrupt decrease of amplitude that occur following a tiny P wave are thought to be under-sensed QRS waves by the implantable loop recorder.

| Statistical analysis
Categorical variables are presented as frequencies and compared using the chi-squared test between groups with or without US. with R Studio (version 3.6.1).

| RE SULTS
Forty-six consecutive patients with ILRs were enrolled in this study.
During the mean follow-up period of 499 ± 363 days, 15 US events were observed in five patients during both the day and night. The time duration from implantation to the first occurrence and the counts of US in five patients were 27 days (five events), 36 days (four events), 44 days (three events), 67 days (two events), and 84 days (one event), respectively. All stored US EGM and standard 12-lead ECG in five patients with US were shown in Figure 3. The five patients with US did not have dilated cardiomyopathy but have 40% (2/5) old myocardial infarction.
The baseline clinical characteristics are shown in Table 1. There were no significant differences in age, sex, BMI, underlying heart disease or echocardiographic parameters between patients with and without US. Peak amplitude of the ILR at discharge was significantly lower in patients with US than in those without US. In standard 12lead ECG, QRS complex amplitude in anterolateral chest leads (V2 to V5) were significant lower in patients with US than in those without US.

| DISCUSS ION
To the best of our knowledge, the present study is the first to clarify in detail the predictors of US in patients with an ILR. Our results showed that US was observed in 10.9% patients with an ILR, and QRS complex amplitude in anterolateral chest leads on standard 12 lead ECG, which suggests that the QRS amplitude of ILRs at discharge might be useful for predicting US in patients with second-generation ILRs. patients with a first-generation ILR. 9 In contrast, in the present patients with a second-generation ILR, US was observed in 5 of 46 (10.9%) patients, and we clarified that QRS complex amplitude in anterolateral chest leads (V2, V3, and V5) on ECG might be a useful predictor of US in patients with ILRs.
In first-generation ILRs, a mapping technique was required to determine the optimal implantation site and signal strength at ILR implantation. 8  Under-sensing interferes with the accurate diagnosis of arrhythmia 10 and causes an increase in inappropriate remote monitoring alerts, which increases the difficulty of providing individual responses. 13  Therefore, the results of our study cannot be generalized to all patients in whom an ILR has been implanted. So, a multi-center observational study that includes a larger number of subjects is required in the future.
QRS complex amplitude in anterolateral chest leads V2 to V5 on standard 12-lead ECG and the peak amplitude of the ILR at discharge might be useful for predicting US in patients with ILRs.

AUTH O R CO NTR I B UTI O N S
Masato Kawasaki, Tetsuya Watanabe and Takahisa Yamada participated in the design of the study and coordination and helped to draft the manuscript. Yoshitaka Kikuchi, Kana Okada carried out data collection. All authors read and approved the final manuscript.

CO N FLI C T O F I NTE R E S T
All authors declare no conflicts of interest associated with this manuscript.

E TH I C S A PPROVA L S TATE M E NT
This study was approved by the Ethics Committee of Osaka General Medical Center in accordance with the Helsinki Declaration.

PATI E NT CO N S E NT S TATE M E NT
Since this study was a retrospective study, we used the opt-out system according to the ethics guidelines on human medical research by the Japanese government.

CLI N I C A L TR I A L R EG I S TR ATI O N
Because of the retrospective study, this study was not registered for clinical trial registration.