ECG changes after percutaneous edge‐to‐edge mitral valve repair

ABSTRACT Background Mitral regurgitation (MR) has a severe impact on hemodynamics and induces severe structural changes in the left atrium. Atrial remodeling is known to alter excitability and conduction in the atrium facilitating atrial fibrillation and atrial flutter. PMVR is a feasible and highly effective procedure to reduce MR in high‐risk patients, which are likely to suffer from atrial rhythm disturbances. So far, electroanatomical changes after PMVR have not been studied. Hypothesis In the current study, we investigated changes in surface electrocardiograms (ECGs) of patients undergoing PMVR for reduction of MR. Methods We evaluated 104 surface ECGs from patients in sinus rhythm undergoing PMVR. P wave duration, P wave amplitude, PR interval, QRS duration, QRS axis, and QT interval were evaluated before and after PMVR and at follow‐up. Results We found no changes in QRS duration, QRS axis, and QT interval after successful PMVR. However, P wave duration, amplitude, and PR interval were significantly decreased after reduction of MR through PMVR (P < .05, respectively). Conclusion The data we provide offers insight into changes in atrial conduction after reduction of MR using PMVR in patients with sinus rhythm.


| INTRODUCTION
Mitral valve regurgitation (MR) contributes significantly to morbidity and mortality of heart failure and is the second most frequent indication for valvular repair. 1 With advances in catheter-based mitral valve repair using an edge-to-edge repair strategy (PMVR), a nonsurgical Hou Bo and David Heinzmann share first authorship. Johannes Patzelt and Peter Seizer share senior authorship. treatment option has successfully been established in clinical practice (MitraClip). While a reduction of MR has obvious hemodynamical advantages by increasing the effective left ventricular stroke volume and cardiac output, 2 other secondary effects have been described. In a previous study, we have found that improved mitral valve coaptation and improved mitral valve annular size correlates with residual MR after PMVR using the MitraClip system. 3 Furthermore, a recent study has found a significant decrease in left atrial operating chamber stiffness after deployment of PMVR. 4 Increased strain of the left atrium through MR induces a significant remodeling of the affected atrial myocardium, which is an ideal substrate for atrial rhythm disturbances, especially atrial fibrillation (AF). 5 With a reduction of MR, a reverse remodeling of the left atrium could be beneficial to preserve normal electrical conduction and hemodynamical function.
In the current study, we were therefore interested whether a change in hemodynamics through PMVR alters electrical activation of the atria. Thus, we performed an extensive analysis of patients undergoing PMVR to establish whether atrial unloading induces changes in electrocardiogram (ECG) morphology.

| Study cohort
Of 233 patients undergoing PMVR using the MitraClip system (Abbott Vascular, Chicago, Illinois) with MR grade 2+ to 4, patients were included in the study with ECG in sinus rhythm at baseline and within the first 2 days after PMVR and/or at the follow-up visit to evaluate P wave characteristics (n = 104, Figure 1a). Patients with a previous medical history of AF were included in the analysis when ECGs at the specified timepoints showed sinus rhythm.

| Electrocardiogram analyses
Standard 12-lead ECGs with a paper speed of 50 mm/s were obtained before and within 2 days after PMVR, as well as at follow-up. A manual analysis of all ECGs was performed by an investigator blinded to the PMVR results, evaluating mean P wave duration, P wave amplitude, PR interval, QRS duration, and QT interval in three consecutive complexes in lead II. For analyses of P wave duration and amplitude, no fusion with a QRS complex or T wave was allowed and a stable isoelectric line had to be present. P wave duration was measured between the junctions of the P wave with the isoelectric line; P wave amplitude was measured from the peak/nadir of the P wave to the isoelectric line. Length of PR interval was determined from the junction of the P wave with the isoelectric line to the onset of the QRS complex. Length of the QRS interval was measured from its earliest deflection from the isoelectric line to its offset. Length of the QT interval was measured from the beginning of the QRS complex to the end of the following T wave. (Figure 1b) QRS axis was obtained by automated analysis of extremity leads. Evaluation of all ECGs was performed by an experienced investigator, blinded to clinical parameters. follow-up. 7 MR severity after the intervention was assessed using a method described by Foster et al. 8

| Statistical analysis
Statistical analysis was performed using GraphPad Prism (Ver. 8, Gra-phPad Software, La Jolla, California) and SPSS (Ver. 24, IBM Deutschland GmbH, Ehningen, Germany). Categorical variables are expressed F I G U R E 1 Flow chart of patient cohort and evaluation strategy. Of 233 patients undergoing percutaneous mitral valve repair (PMVR), 104 patients were included in the study with electrocardiogram (ECG) in sinus rhythm at baseline and within the first 2 days after PMVR and/or at the follow-up visit to evaluate P wave characteristics (a). Patients with a previous medical history of atrial fibrillation (AF) were included in the analysis when ECGs at the specified timepoints showed sinus rhythm. (b) illustrates obtained ECG measurements including P wave amplitude (a), P wave duration (b), PR interval (c), QRS duration (d), and QT interval (e) within the time course of PMVR as absolute numbers or as percentage, continuous variables as mean ± SEM, or mean ± SD were specified. Paired t test was used to compare means, and P < .05 was considered statistically significant.

| Baseline characteristics
The baseline characteristics of the patient cohort are summarized in Table 1. The reduction of MR after PMVR is illustrated in Figure 2. 3.2 | P wave duration and P wave amplitude are decreased after PMVR At baseline, patients undergoing PMVR showed a mean P wave duration of 113.8 ± 2 ms before the intervention. In the ECG control shortly after the intervention, a reduction of the P wave duration to 103.3 ± 2 could be observed (P < .0001). At follow-up, no significant change compared to post-PMVR was apparent (106.4 ± 3 ms, P > .05 compared to post-PMVR), as shown in Figure 3. Similar to the shortening of the P wave duration, patients undergoing PMVR showed a decrease of the P wave amplitude compared to the preinterventional results (0.104 ± 0.004 mV vs 0.088 ± 0.004 mV, P < .001). At followup, no significant change of amplitude was found (0.094 ± 0.005 mV, P > 0.05 compared to post-PMVR), as illustrated in Figure 3.

| PR interval shortens after PMVR
Furthermore, we observed a decrease in PR interval from 176.3 ± 4.5 ms to 169.8 ± 4 ms after PMVR (P < .05). Similar to the other atrial parameters, the mean PR interval did not change significantly during follow-up compared to post-PMVR (166 ± 6.7 ms, P > .05, Figure 3).

| Ventricular excitation is unchanged by PMVR
PMVR had no effect on QRS duration, neither after the procedure, nor did it induce changes at follow-up (120.9 ± 3.

| DISCUSSION
PMVR is now a widely used approach to reduce MR in patients unsuitable for surgical repair of the mitral valve. With MR being a well-established risk factor for severe left atrial remodeling, PMVR might be a tool to reduce atrial strain and have a beneficial effect on atrial conduction.
In this study, we describe for the first time distinct changes of atrial conduction found in surface ECGs of patients undergoing PMVR: (a) after PMVR, P wave duration PR interval, and P wave amplitude showed a strong decrease compared to baseline, (b) at follow-up, P wave duration, PR interval, and P wave amplitude remained unchanged compared to post-PMVR values, (c) QRS duration, QRS axis, and QT interval were not affected by PMVR.
In our cohort, patients undergoing PMVR experienced a significant reduction of MR, illustrated in Figure 2.
Atrial enlargement has been found to be an independent predictor for onset of AF after mitral valve repair, with atrial remodeling being the major culprit of electrical instability and susceptibility for AF. 9 We and others have observed that PMVR results in significant geometrical changes, especially of the left atrium. 3 between an increased P wave duration and higher recurrence rates in patients with AF. 18

CONCLUSION
In the current study, we provide data illustrating a decrease of P wave duration and amplitude, as well as a reduction of PR interval through reduction of MR using PMVR.

ACKNOWLEDGMENTS
This study was supported by grants from the Deutsche F I G U R E 4 Ventricular excitation in surface electrocardiogram (ECG) does not change after PMVR. A total of 104 patients undergoing percutaneous mitral valve repair (PMVR) were analyzed regarding changes in QRS duration in ms (a), QT interval in ms (b), and QRS axis in degree (c). Reduction of mitral regurgitation using PMVR showed no significant change in ventricular excitation, neither shortly after the procedure nor at follow-up. All parameters are shown as mean ± SEM, n.s. indicates P > .05