Fragmented QRS on far‐field intracardiac electrograms as a predictor of arrhythmic events

Abstract Background Studies suggest that fragmented QRS (fQRS) can predict arrhythmic events in various cardiac diseases. However, the association between fQRS recordings on intracardiac electrogram (EGM) and ventricular arrhythmic events remains unknown. Methods We enrolled 51 patients (age, 62 ± 12 years; 40 men) with an implantable cardioverter‐defibrillator (ICD) and structural heart disease and evaluated surface electrocardiogram (ECG) and EGM measurement of fQRS and the association between fQRS and arrhythmic events. Results fQRS was detected on surface ECG and ICD‐EGM in 12 (23.5%) and 15 (29.4%) patients, respectively. fQRS was detected more frequently on ICD‐EGM in patients with fQRS on surface ECG than in patients without fQRS (7/12 [58.3%] vs 8/39 patients [20.5%], P = .01). Appropriate ICD therapies were documented in 16 patients. Among these patients, fQRS was detected more frequently on surface ECG and ICD‐EGM in patients with appropriate ICD therapies (8/16, 50.0%; P = .001 and 11/16, 68.9%; P < .001). Nonsustained ventricular tachycardia was significantly more frequent in patients with appropriate ICD therapies (15/16, 93.8%; P = .04). Multiple logistic regression analysis showed that fQRS on ICD‐EGM was a predictor of arrhythmic events (P = .03). Kaplan‐Meier survival analysis revealed that ICD therapies were significantly more frequent among patients with fQRS on both surface ECG and ICD‐EGM than among those without fQRS (66.7% vs 6.6%, P < .001). Conclusion The presence of fQRS on ICD‐EGM can be a predictor of arrhythmic events in ICD patients. Surface ECG and ICD‐EGM measurement may help predict ventricular arrhythmic events.


| Study population
In this retrospective study, we investigated 51 patients with ICD and structural heart disease, excluding patients with hereditary arrhythmia disease, between January 2003 and May 2016. Patients were followed up every 4 months at the ICD Clinic until January 2020. The outcome of this study was appropriate ICD therapy. We evaluated surface ECG, ICD-EGM, and appropriate ICD therapies for all patients. Surface ECG and ICD-EGM were evaluated without information on ICD treatment and surface ECG or ICD-EGM. This study was approved by our respective institutional review boards. Informed consent was obtained in the form of opt-out on the website.

| ECG criteria for fQRS (RSR pattern and its variants)
fQRS included the presence of an additional R wave (R) or notching in the nadir of the R wave or the S wave, or the presence of more than one R (fragmentation) in two contiguous leads, corresponding to a major coronary artery territory. 1 If the RSR patterns were present in the right precordial leads (leads V1 and V2) with QRS of >100 ms (incomplete right bundle branch block) or QRS of >120 ms (complete right bundle branch block) and in the left precordial lead (RSR in leads I, V5, and V6) with QRS of >120 ms (left bundle branch block), this was defined as a complete or incomplete bundle branch block and was excluded from the definition of fQRS. If the RSR pattern was present in the mid-precordial or inferior leads, this was defined as fQRS. A low-pass filter is frequently used to reduce electrical and muscular noise when recording the 12-lead ECG; however, the cutoff frequency of the low-pass filter influences the detection of fQRS. 20 In our study, ECGs were recorded using a low-pass filter at 150 Hz. Bandpass filters for ICD-EGM vary by manufacturer and their settings are not published.
Near-field (NF) EGM was defined as the difference in potentials between the tip and the ring, or between the tip and the coil of the bipolar ICD lead implanted in the apex of the right ventricle (RV).
Far-field (FF) EGM was defined as the difference in potentials between the ring of the RV lead and the ICD can, or between the coil and can. The fQRS of ICD-EGM was reported if it was recorded by either lead.

| Appropriate ICD therapy
Appropriate ICD therapy was defined as shock and ATP for ventricular arrhythmia. ICD programming for ventricular tachyarrhythmia was based on the judgment of the attending physician.

| Statistical analysis
Continuous variables are expressed as mean ± standard deviation. Continuous and categorical variables were compared using Student's t-test and the chi-square test, respectively. A Cox proportional hazard model was used to assess the association of the baseline variables with the appropriate ICD therapy. Survival curves were calculated using the Kaplan-Meier method. P-values of <.05 were considered statistically significant.
Kaplan-Meier survival analysis revealed that patients with fQRS on both surface ECG and ICD-EGM had a significantly higher rate of appropriate ICD therapies than those without fQRS (66.7% vs 6.6%, P < .001) ( Figure 2). Recording fQRS on both surface ECG and ICD-EGM had a high positive predictive value (PPV; 85.7%) for appropriate ICD therapies. The sensitivity, specificity, and negative predictive value (NPV) for appropriate ICD therapies on recording fQRS using both surface ECG and ICD-EGM were 37.5%, 97.1%, and 77.3%, respectively.

| Main findings
This study is the first to evaluate the association between surface ECG, ICD-EGM, and appropriate ICD therapies. We suggest that evaluation using surface ECG in combination with ICD-EGM is a strong predictor of ventricular arrhythmic events. Appropriate ICD therapies were significantly more common in patients with

| Association between fQRS and arrhythmic events
fQRS has been shown to be associated with increased mortality and arrhythmic events in patients with CAD, 1,2,6,9 Brugada syndrome, and nonischemic cardiomyopathy.   In our study, fQRS on surface ECG was recorded in inferior leads. It has been reported that patients with fQRS in the inferior leads have a poor prognosis. 9 An ICD was implanted in all patients included in our study. These patients might have had more arrhythmogenic substrates, which led to a poor prognosis.
One limitation of our study was that the settings of the bandpass filter vary based on the manufacturers. However, we did not   19 Although our data had some limitations, fQRS is a simple and overt ECG sign, and further research is needed to improve its predictive value of ventricular arrhythmic events. We suggest that evaluations using surface ECG in combination with ICD-EGM as a strong predictor of ventricular arrhythmic events.

| Limitations
First, this study was retrospective in nature, and fQRS on surface ECG and ICD-EGM may not have been recorded in all patients at the same timepoint. Second, this was a small-scale study with a limited number of patients. Third, the changes in fQRS caused by progressing heart failure and/or ischemic events during follow-up were not evaluated. Fourth, the settings of the bandpass filter may vary by company, and the cutoff frequency of the low-pass filter may influence the detection of fQRS on ICD-EGM. Fifth, the decision for ICD management of ventricular tachyarrhythmia was determined by the judgment of the attending physician. ICD programming can affect ICD therapies.

| CON CLUS ION
The presence of fQRS on ICD-EGM can be a predictor of arrhythmic events in patients with ICD. We suggest using surface ECG in combination with ICD-EGM to predict ventricular arrhythmic events.