Author + information
- aHeart Research Follow-up Program and the University of Rochester School of Medicine and Dentistry, Rochester, New York
- bSMG Arrhythmia Center, Summit Medical Group, Short Hills, New Jersey
- ↵∗Address for correspondence:
Dr. Jonathan S. Steinberg, SMG Arrhythmia Center, 85 Woodland Road, Short Hills, New Jersey, 07079.
Delayed and asynchronous activation of the left atrium (LA) is a precursor for the development of atrial fibrillation (AF) and can be assessed quantitatively by computing the P-wave duration (PWD) on the 12-lead standard electrocardiogram (ECG). In the past, this simple measurement was successfully used to predict new onset AF (1), post-operative AF (2), recurrent AF after cardioversion (3), AF after isthmus ablation for atrial flutter (4) and ablation of accessory pathways (5), and risk of ischemic stroke (6) and death (7).
Several prior studies have suggested that a prolonged PWD predicts recurrences after catheter ablation when pulmonary vein isolation (PVI) is performed for paroxysmal AF (8). The premise underlying this association is that slow conduction is an important factor in providing a suitable electrophysiological environment for atrial reentry and thus determining the risk of AF, progression of AF burden, and response to interventions, particularly after PV triggers have been eradicated.
Identification of clinical markers of the mechanisms that drive persistent (compared to paroxysmal) AF that is resistant to catheter ablation by PVI alone provides an opportunity to design successful treatment strategies as well as for risk stratification. Structural atrial remodeling that is fixed and related to fibrofatty interruption of atrial tissue architecture is particularly important to identify, as its presence likely represents a low-yield for conventional ablation strategies such as PVI. When atrial remodeling is partially or fully reversible, usually by restoration of sinus rhythm, the ablation strategy can be kept simple with an expectation of good results (9).
In this issue of JACC: Clinical Electrophysiology, Jadidi et al. (10) used the P-wave on the standard ECG to: 1) confirm that mapped LA activation time was accurately reflected on the surface ECG; 2) determine if the degree of low-voltage LA substrate correlated with conduction delay on the surface ECG; and 3) potentially predict post-ablation outcomes and refine ablation tactics.
The study had 2 phases. In the first phase, a series of 72 patients with persistent AF (>7 days and <12 months) who were slated to undergo PVI were studied. Patients underwent pre-ablation cardioversion to restore or maintain sinus rhythm, often facilitated by antiarrhythmic drug therapy (9). The PWD was measured prior to ablation, and a dense voltage map was created during the procedure, and PVI was completed without additional lesion sets.
It is worth focusing on the authors’ technique for P-wave measurement. Historically, analog acquisition of the ECG signal made the measurement of PWD challenging without digital conversion, signal averaging, and filtering (11). Additional indices were sometimes also used including P-wave dispersion, voltage, and morphology, and the ECG presence of interatrial block (12). Jadidi et al. (10) used a simple technique that could be routinely adopted after recording a digital 12-lead ECG in sinus rhythm. The ECG was amplified to 0.2 to 0.25 mV/cm and interpreted during a sweep speed of 100 to 200 mm/s. PWD was measured using digital calipers from “the earliest deflection from the isoelectric line in any lead to the time of latest activation in any lead.” Special care was taken to measure terminal P-wave forces that represented LA activation. Using this methodology, the authors observed excellent interobserver reproducibility.
Despite similar AF phenotype among subjects, there was great variation in the amount of presumed atrial scarring, using various definitions of aggregated low-voltage regions. Invasive estimation of LA activation was sensitive to the amount of low-voltage findings, confirming that slow conduction results from or accompanies this underlying substrate.
The P-wave was analyzed as right atrial (RA) and LA components, principally by dividing it at the maximum negative dV/dt on leads V1 to V2 or the inferior leads. The noninvasive LA P-wave component strongly correlated with the invasive activation time (r = 0.96). Most importantly, the LA component of the P-wave also showed a strong correlation to the amount of low-voltage atrial substrate (r = 0.80), mimicking the invasive data described above. Because the RA component correlated poorly, the entire PWD was also well correlated to the voltage results, allowing for simplified measurement of this marker.
Receiver-operating characteristic curves were constructed for PWD, and the presence of low amplitude substrate with a cutoff of 150 ms, yielding high sensitivity and specificity, 94% and 92%, respectively. The findings were unaffected by use of antiarrhythmic drug.
In the second phase, PWD was measured in 143 patients of a validation cohort and tested for prediction of recurrent atrial tachyarrhythmia after PVI. The entire group was almost evenly split between subgroups with and without prolonged P waves. Recurrent arrhythmia at 1 year was much greater in the subgroup with P-wave ≥150 ms, 59%, versus 28% for those with PWD that was not prolonged, for a hazard ratio of approximately 2 (p = 0.0003). Only the PWD among all clinical characteristics was significantly associated with recurrent arrhythmia in the multivariate analysis.
These investigators have performed a careful and valuable study of patients with persistent AF who presented serious challenges to achieving high-quality and durable results following catheter ablation. The findings confirm the frequent but not ubiquitous presence of low-voltage substrate (presumably advanced remodeling), which in turn correlated with LA conduction slowing and was quantifiable on the standard surface ECG, using PWD. PWD was a powerful predictor of arrhythmia recurrence following ablation.
Although it appears PWD is an easily measured marker of atrial myocardial substrate, it is important to acknowledge that other attributes of atrial remodeling are not or cannot be assessed by PWD as measured in this study. Patients who cannot maintain sinus rhythm are not eligible for this technique. The study did not assess the possibility of reverse remodeling as a dynamic process with serial measurements of PWD. Ablation itself may create scar and affect the P-wave, which could compromise prediction in patients who previously underwent ablation, or possibly facilitate assessment of proarrhythmic or failed antiarrhythmic interventional strategies. P-wave measurements are global and do not reflect local or site-specific processes, so they cannot be used for targeting atrial regional abnormalities. Although the authors observed low interobserver variability in the measurement of PWD, this will need to be reproduced by other centers to confirm its applicability; the definition of beginning and end of P-wave is likely to be the most challenging aspect of this effort.
It is important to emphasize that PWD is a marker or assessment of the effects of scar and other causes of conduction delay. Nonetheless, compared to other techniques for assessing scar burden such as delayed enhancement cardiac magnetic resonance (13), measurement of PWD is a low-cost, widely available method/marker and investigational tool with comparable strength of prediction that does not require sophisticated equipment or training. We can foresee future investigations that use PWD to: 1) test targeted ablation strategies in high-risk subsets; 2) quantify reverse remodeling for use in planning ablation; 3) assess long-term and progressive stroke risk in patients with history of AF (ablated and nonablated); 4) assess population risk of AF and need for intensified surveillance (e.g., with Smartphone ECGs); and 5) use it in combination with other inexpensive and readily available noninvasive risk factors (e.g., frequency of atrial premature beats) for development of risk scores in at-risk patient populations.
↵∗ Editorials published in JACC: Clinical Electrophysiology reflect the views of the authors and do not necessarily represent the views of JACC: Clinical Electrophysiology or the American College of Cardiology.
Dr. Steinberg is a consultant for and has received research grants from Biosense Webster, Medtronic, Allergan, G Medical, AliveCor, and National Cardiac; has received research support from AliveCor and National Institutes of Health; and has consulted for Boston Scientific and ABIM. Dr. Altman has received research support from Boston Scientific; and has consulted for Abbott Laboratories.
The authors attest they are in compliance with human studies committees and animal welfare regulations of the author’s institution and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- 2018 American College of Cardiology Foundation
- Tse G.,
- Wong C.W.,
- Gong M.,
- et al.
- Steinberg J.S.,
- Zelenkofske S.,
- Wong S.C.,
- Gelernt M.,
- Sciacca R.,
- Menchavez E.
- Enriquez A.,
- Conde D.,
- Hopman W.,
- et al.
- Aytemir K.,
- Amasyali B.,
- Kose S.,
- et al.
- He J.,
- Tse G.,
- Korantzopoulos P.,
- et al.
- Wang Y.S.,
- Chen G.Y.,
- Li X.H.,
- Zhou X.,
- Li Y.G.
- Khan A.,
- Mittal S.,
- Kamath G.S.,
- Garikipati N.V.,
- Marrero D.,
- Steinberg J.S.
- Jadidi A.,
- Müller-Edenborn B.,
- Chen J.,
- et al.
- Guidera S.,
- Steinberg J.S.