ORIGINAL RESEARCH published: 17 April 2020 doi: 10.3389/fphys.2020.00264 Changes in Local Atrial Electrograms and Surface ECG Induced by Acute Atrial Myocardial Infarction Gerard Amorós-Figueras 1 , Elena Roselló-Diez 2 , Damian Sanchez-Quintana 3 , Sergi Casabella-Ramon 1 , Esther Jorge 1 , Jorge Nevado-Medina 3 , Dabit Arzamendi 1 , Xavier Millán 1 , Concepción Alonso-Martin 1 , Jose M. Guerra 1 and Juan Cinca 1* 1 Department of Cardiology, Hospital de la Santa Creu i Sant Pau, Institute of Biomedical Research IIB Sant Pau, CIBERCV, Universitat Autònoma de Barcelona, Barcelona, Spain, 2 Department of Cardiac Surgery, Hospital de la Santa Creu i Sant Pau, Institute of Biomedical Research IIB Sant Pau, Universitat Autònoma de Barcelona, Barcelona, Spain, 3 Department of Anatomy and Cell Biology, Faculty of Medicine, University of Extremadura, Badajoz, Spain Background: Atrial coronary branch occlusion is a hardly recognizable clinical entity that can promote atrial fibrillation. The low diagnostic accuracy of the ECG could deal Edited by: with the characteristics of the ischemia-induced changes in local atrial electrograms, but Veronique Meijborg, these have not been described. Amsterdam University Medical Center (UMC), Netherlands Objectives: We analyzed the effects of selective acute atrial branch occlusion on local Reviewed by: myocardial structure, atrial electrograms, and surface ECG in an experimental model Sander Verheule, close to human cardiac anatomy and electrophysiology. Maastricht University, Netherlands Elena Grossini, Methods: Six anesthetized open-chest anesthetized pigs underwent surgical occlusion University of Eastern Piedmont, Italy of an atrial coronary branch arising from the right coronary artery during 4 h. Atrial *Correspondence: Juan Cinca electrograms and ECG were simultaneously recorded. One additional pig acted as sham jcinca@santpau.cat control. In all cases, the hearts were processed for anatomopathological analysis. Specialty section: Results: Atrial branch occlusion induced patchy atrial necrosis with sharp border zone. This article was submitted to During the first 30 min of occlusion, atrial electrograms showed progressive R wave Cardiac Electrophysiology, enlargement (1.8 ± 0.6 mV vs. 2.5 ± 1.1 mV, p < 0.01), delayed local activation times a section of the journal Frontiers in Physiology (28.5 ± 8.9 ms vs. 36.1 ± 16.4 ms, p < 0.01), ST segment elevation (−0.3 ± 0.3 mV Received: 05 November 2019 vs. 1.0 ± 1.0 mV, p < 0.01), and presence of monophasic potentials. Atrial ST segment Accepted: 09 March 2020 elevation decreased after 2 h of occlusion. The electrical border zone was ∼1 mm and Published: 17 April 2020 expanded over time. After 2 h of occlusion, the ECG showed a decrease in P wave Citation: Amorós-Figueras G, amplitude (from 0.09 ± 0.04 mV to 0.05 ± 0.04 mV after 165 min occlusion, p < 0.05) Roselló-Diez E, Sanchez-Quintana D, and duration (64.4 ± 8.0 ms vs. 80.9 ± 12.6 ms, p < 0.01). Casabella-Ramon S, Jorge E, Nevado-Medina J, Arzamendi D, Conclusion: Selective atrial branch occlusion induces patchy atrial infarction and Millán X, Alonso-Martin C, Guerra JM characteristic changes in atrial activation, R/S wave, and ST segment that are not and Cinca J (2020) Changes in Local Atrial Electrograms and Surface ECG discernible at the ECG. Only indirect changes in P wave amplitude and duration were Induced by Acute Atrial Myocardial appreciated in advanced stages of acute coronary occlusion. Infarction. Front. Physiol. 11:264. doi: 10.3389/fphys.2020.00264 Keywords: atrium, ischemia, ECG, electrophysiology mapping, histopathology Frontiers in Physiology | www.frontiersin.org 1 April 2020 | Volume 11 | Article 264 Amorós-Figueras et al. Atrial Electrograms and ECG in Atrial Infarction INTRODUCTION Experimental Procedures Atrial Coronary Branch Occlusion Atrial myocardial infarction is a potentially harmful disease that A median sternotomy was performed, and the pericardium was can provoke atrial mechanical depression, systemic embolization, opened to expose the heart in a pericardial cradle. Atrial coronary atrial arrhythmias, and cardiac rupture (Lazar et al., 1988). branches arising from the right coronary artery were identified It is caused by occlusion of the atrial coronary branches by fine direct dissection and separated from the surrounding ordinarily emerging from the proximal segments of the right fat tissue. In all cases, the atrial coronary branches originated (RCA) or the left circumflex (LCX) coronary arteries (Baroldi at the proximal segment of the right coronary artery. Then, the et al., 1956; James and Burch, 1958). The diagnosis is atrial coronary branch was selectively occluded with hemostatic often established at autopsy in association with ventricular clips (Peters Surgical, France) (Figure 1) during 4 h in six pigs. myocardial infarction. In a series of 182 necropsy cases of One additional pig was submitted to the same experimental myocardial infarction, the incidence of atrial infarction was 17% procedures, but the atrial coronary branches were not occluded (Cushing et al., 1942). (sham experiment). The thorax was maintained opened during Selective occlusion of an atrial coronary branch may the 4 h of atrial branch occlusion. result from a rupture of a local atherosclerotic plaque or, accidentally, during percutaneous coronary interventions Atrial Epicardial Mapping (Álvarez-García et al., 2013, 2016). This condition predisposes The epicardial unipolar atrial electrograms (Figure 1) were to arrhythmia reentry as evidenced in the canine model recorded with 128 electrodes spaced 1 mm mounted on a patch of (Sinno et al., 2003; Nishida et al., 2011) but is hardly 17 mm × 12.5 mm. This probe was gently sutured with Prolene recognized at the surface ECG. The diagnostic paucity of 3/0 snare (Ethicon) to the atrial epicardium in a region close the ECG could be due to intrinsic characteristics of the to the isolated atrial coronary branch. The reference electrode conventional recording systems or to a weak impact of was placed at the upper mediastinal region close to the thymus atrial branch occlusion on local myocardial structure and and to the upper border of the sternotomy. Within this discrete electrophysiology. anatomical region, we chose the site that provided well voltage In the clinical practice, the ECG features that allow suspecting and stable atrial electrograms. Then, the position of the reference atrial myocardial ischemia are changes in P wave morphology electrode remained unaltered during the experiment. The local and PR-segment deviation (Lazar et al., 1988; Lu et al., 2016; electrical potentials were recorded with a BioSemi acquisition Yıldız et al., 2018) although these are inconstant and can require system (BioSemi, Netherlands). Recordings were obtained at a a reference ECG to be reliably discerned. Changes in the QRS sampling rate of 2 kHz and stored continuously during the first complex and ST segment of local electrograms are specific 60 min of atrial coronary occlusion and, thereafter, every 15 min markers of acute myocardial ischemia in the ventricles (Janse until completing the 4-h occlusion. An offline analysis of the et al., 1979), and theoretically, similar electrical changes should atrial electrograms was done with a custom-made Matlab script be found in the atrial electrograms during atrial branch occlusion, (MathWorks, United States), and the following parameters were but it has not yet been proven. measured: R/S wave amplitude, local activation time, measured Therefore, this study aimed to analyze the effects of selective between the start of increment of the R wave and at the occlusion of an atrial coronary branch on atrial electrograms highest downstroke slope after the R wave peak, and ST segment and surface ECG in a model close to human cardiac coronary deviation from the J point. To have a quantitative appreciation distribution and electrophysiology. of the changes in these parameters, we selected a group of six neighboring electrodes with an appropriate signal/noise ratio that have shown monophasic potentials. We defined as MATERIALS AND METHODS monophasic potentials all electrograms showing increased R wave amplitude together with loss of S wave and appearance of Study Population ST segment elevation. This study involved seven anesthetized domestic swine (Landrace-Large White cross) weighing 40 ± 5 kg (six Surface Electrocardiogram successful animals submitted to 4 h ischemia and one sham The ECG was continuously recorded using a CardioSoft system animal). The animals were premedicated with midazolam (GE Healthcare, Germany), as previously reported (Noriega et al., (0.6 mg/kg) and ketamine (12 mg/kg) intramuscularly. General 2013). The ECG was recorded at baseline and continuously anesthesia was induced with intravenous propofol (2–4 mg/kg) during atrial coronary artery occlusion at a sampling rate of and was maintained with a mixture of oxygen and sevoflurane 200 Hz. All ECG recordings analyzed in this study were obtained inhalation (2.5–3.5%). After endotracheal intubation, they were in the open chest condition. Changes in the amplitude and mechanically ventilated. Analgesia was maintained during the duration of the P wave, QRS-complex duration, and ST segment entire procedure with intravenous fentanyl (2.5 µg/kg). The at the J point were measured with Matlab scripts at ECG lead II study protocol was approved by the Animal Care and Use using signal averaging in a window time of 1 min. We selected Committee of our institution and fully conformed the Guide for lead II because, in clinical practice, it is commonly used to the Care and Use of Laboratory Animals, 8th edition (National evaluate the P wave characteristics, and moreover, the signal Research Council, 2010). voltage of the lead was less affected after the median sternotomy. Frontiers in Physiology | www.frontiersin.org 2 April 2020 | Volume 11 | Article 264 Amorós-Figueras et al. Atrial Electrograms and ECG in Atrial Infarction FIGURE 1 | Illustration of the experimental preparation. Upper left: photograph of an extracted porcine heart showing the placement of the 128-electrode patch on the right atrial epicardium. The arrow shows an hemostatic clip occluding an atrial coronary branch from the right coronary artery. Encircled letters A, B, and C indicate atrial regions extending from the closest occlusion site to the remote sites of the atrial appendage. Upper right: local atrial electrograms recorded with a 128-electrode patch 240 min after occlusion of the right atrial coronary branch. Bottom: histological microphotographs of the right atrium showing transmural infarction in region A, subendocardial in region B, and normal myocardium in regions C. RA, right atrium; RV, right ventricle. Anatomopathological Examination direct visualization of the histological preparation by an expert After 4 h of atrial coronary occlusion, pigs were euthanized anatomopathologist at proper magnification. by electrically inducing ventricular fibrillation. The heart was explanted with the 128 electrodes sutured in its initial position. Data Analysis Then, the position of the 128-electrode patch was marked Quantitative data were expressed as the mean ± standard with fine Prolene snares sutured at each electrode corner deviation (SD). The ordinary two-way ANOVA test with (Supplementary Figure S1). This procedure was done to evaluate Dunnett’s multiple comparison correction was used to evaluate a potential correspondence between the electrical electrograms the significance of the changes from baseline to selected occlusion and the underlying anatomical substrate. The heart was then times in the atrial R/S waves, local activation time, and ST cleaned from blood and fixed in buffered formalin. The atria segment. The ANOVA test was also used to assess the significance were separated from the ventricles and embedded in a paraffin of the evolving changes in P wave morphology, QRS duration, block for histological analysis. Sections of 12 µm were mounted and ST segment displacement in the ECG. The paired t-test was and stained with Masson’s trichrome. A morphometry study was used to assess the significance of the changes in the number of performed in six infarcted pigs to evaluate the average extent electrodes showing ischemic changes. A p < 0.05 was considered of myocyte degeneration (myolysis). As described previously significant. All analyses were performed using SPSS v.22.0 (IBM- (Climent et al., 2004), digital images were taken of the sections SPSS, United States). stained with Masson’s using a Nikon SMZ 1500 microscope (Nikon, United States). Morphometry was performed at least in four fields per atrial sample, in the regions of maximum RESULTS extent of infarcted area, using a grid covering an area of ∼4 mm2 of atrial sample. The areas of myolysis were expressed Anatomopathological Findings as a percentage of the entire limits of the grid. Blood vessels, As illustrated in Figure 2A, occlusion of an atrial coronary perivascular interstitial tissue, and intertrabecular spaces between branch induced degeneration and coagulative necrosis of groups the pectinate muscles were excluded from the morphometric of cardiomyocytes. The necrosis depicted a “patchy” distribution analysis. The identification of the areas of myolysis was done by and showed myocardial fiber disarray with cellular infiltration. Frontiers in Physiology | www.frontiersin.org 3 April 2020 | Volume 11 | Article 264 Amorós-Figueras et al. Atrial Electrograms and ECG in Atrial Infarction of the pectinate muscles and encompassed small arterioles (Figure 2C). In some instances, the necrotic foci appeared near local atrial epicardial or intramyocardial neural fibers (Figure 2D). The morphometric analysis in the sections showing the greatest infarct extent revealed that the mean area of myolysis in the six pigs was ∼54% (range, 37.7–77.8%) of the entire examination field. Supplementary Figure S2 illustrates representative examples of the measured areas of myolysis. The atria of the sham-operated pig were free of myocardial necrosis. Atrial Epicardial Mapping At baseline conditions, the atrial electrograms showed an R/S complex (Figure 3) with a duration of 48.6 ± 10.2 ms, closely followed by isoelectric ST segment. Figure 3 also illustrates the time correspondence between the local atrial signals and the surface ECG: the atrial R/S complex coincided with the surface P FIGURE 2 | Histological findings in pigs submitted to 4-h of atrial coronary wave and the atrial ST segment with the PR segment of the ECG. branch occlusion. (A) Microphotograph showing areas of “patchy” necrosis During the first 30 min of atrial branch occlusion with myocardial fiber disarray (arrows) and sharp border zone (yellow marks). (B) Detailed amplification of the ischemic border zone showing a sharp band (Figures 3, 4), the atrial epicardial electrodes overlying the of interdigitating ischemic and normal cells (yellow marks). (C) Subendocardial ischemic area showed the following: (a) progressive increase infarction involving a pectinate muscle and small arterioles (arrows). in the R wave amplitude (from 1.8 ± 0.6 mV at baseline to (D) Necrotic foci (yellow marks) in the vicinity local atrial epicardial neural fibers 2.5 ± 1.1 mV at 30 min occlusion, p < 0.01), (b) disappearance (arrows). of the S wave (from −2.0 ± 1.1 mV to −0.4 ± 0.4 mV, p < 0.01), (c) prolongation of local activation time (from 28.5 ± 8.9 ms to 36.1 ± 16.4 ms, p < 0.01), and (d) gradual The transitional ischemic border zone was abrupt and was elevation of the ST segment leading to monophasic potentials composed of an interdigitating band of normal and ischemic (from −0.3 ± 0.3 mV to 1.0 ± 1.0 mV, p < 0.01). After 30– cells (Figure 2B). The infarction affected subendocardial regions 40 min of occlusion, the affected atrial electrograms depicted FIGURE 3 | Evolving changes in epicardial atrial electrograms and surface ECG lead II in a pig submitted to 4 h of atrial coronary branch occlusion. Local electrograms in an ischemic (IZ) and normal zone (NZ) recorded at baseline and after 15, 30, 60, 120, 180, and 240 min of right atrial coronary branch occlusion showing increase in R wave amplitude, disappearance of the S wave, ST segment elevation, and monophasic potentials. Frontiers in Physiology | www.frontiersin.org 4 April 2020 | Volume 11 | Article 264 Amorós-Figueras et al. Atrial Electrograms and ECG in Atrial Infarction FIGURE 4 | Evolving changes in atrial epicardial mapping with 128 electrodes in a pig submitted to 4 h of atrial coronary branch occlusion. (A) Mean values of atrialR wave, S wave, ST segment, and activation time in selected electrograms at baseline and at specific times after atrial coronary branch occlusion. (B) Mapping of epicardial atrial unipolar electrograms at baseline, 60, 120, and 240 min after atrial coronary branch occlusion. Red lines delineate the electrical border zone encompassing one or two electrodes. **p < 0.01; from 0 to 165 min of ischemia. a progressive decline in R wave amplitude and ST segment The transition between electrodes with ischemic changes and elevation maximal at 4 h of occlusion (R wave, 0.8 ± 0.4 mV; electrodes with normal potentials encompassed less than two ST segment, 0.4 ± 0.3 mV; p < 0.01, from baseline). Electrodes contiguous electrode sites (Figure 4B). Thus, since the mapping not showing ST segment changes depicted a reduction in R electrodes were spaced 1 mm, the expected width of the electrical wave amplitude of ∼30% after 3 h of occlusion (Figure 4A). border zone would be smaller than 2 mm. The area of the In two cases, we observed a transient electrical recovery phase epicardial mapping encompassing electrodes with monophasic characterized by a reduction in the ST segment elevation and potentials increased during the occlusion period (Figure 4B). The improvement of the local activation delay with reappearance of number of electrodes showing monophasic potentials increased local intrinsic deflections on previously monophasic potentials overtime (28 ± 35 at 60 min, 98 ± 37 at 120 min, and 112 ± 18 at (Supplementary Figure S3). This phase began at about 45– 240 min; p < 0.01). 60 min of occlusion and ended 30 min thereafter. In all animals, we observed a correspondence between the anatomopathological Surface ECG Findings substrate and the electrical mapping. This correspondence was As shown in Figure 5, changes in P wave morphology were only established by identifying first the necrotic regions in the serial appreciable late after 2 h of coronary occlusion. After 165 min section of the atria stained with Masson’s trichrome and then of occlusion P wave amplitude decreased from 0.09 ± 0.04 mV determining their projection into the 128-electrode patch, which at baseline to 0.05 ± 0.04 mV (p < 0.05) and P wave duration was marked with Prolene snares and photographed at the end increased from 64.4 ± 8.0 ms to 80.9 ± 12.6 ms (p < 0.01). We of the experiment. Necrotic regions corresponded to electrode did not observe appreciable deviations of the PR-segment level in areas with monophasic potentials, and healthy regions depicted any ECG lead. The sham pig did not show noticeable changes in normal electrograms (Figure 1, bottom panel). the P wave morphology. Selective occlusion of the atrial branch Frontiers in Physiology | www.frontiersin.org 5 April 2020 | Volume 11 | Article 264 Amorós-Figueras et al. Atrial Electrograms and ECG in Atrial Infarction FIGURE 5 | Changes in P wave morphology induced by 4 h of atrial coronary branch occlusion. (A) ECG lead II for each of the six infarcted pigs at baseline and after 30 and 240 min of atrial coronary branch occlusion. The bottom row shows the absence of P wave changes in a pig not submitted to coronary occlusion (sham experiment). (B) Time course of the P wave mean amplitude and duration in the entire group of the six pigs submitted to 4 h of atrial branch occlusion (red line). The plateau behavior of the P wave amplitude and duration in one sham pig is represented by black line. AB, atrial coronary branch. *p < 0.05, **p < 0.01; from 0 to 165 min of ischemia. did not induce concurrent ventricular ischemia as denoted by the ventricular myocardial infarction caused by occlusion of the absence of significant changes in the QRS-complex duration and RCA or LCX (Cushing et al., 1942; Liu et al., 1961). In these ST segment level (Supplementary Figure S4). Episodes of atrial studies, the atrial infarction was massive, suggesting that the fibrillation were not observed. interruption of coronary flow affected most of the atrial branches that ordinarily arise from the proximal segment of the RCA or LCX (Baroldi et al., 1956; James and Burch, 1958). Similarly, DISCUSSION proximal LCX occlusion induced large atrial infarcts and atrial dilation in pigs (Aguero et al., 2017), which, like humans, have Main Findings a comparable emergence of the atrial branches from the RCA This study is the first describing the changes in local atrial or LCX coronary arteries (Weaver et al., 1986). By contrast, the electrograms induced by acute occlusion of an atrial coronary structural changes induced by selective atrial branch occlusion branch and their temporal relationship with the ECG. Using a have been only described by a single research group in dogs model close to the human cardiac anatomy and electrophysiology and sheep (Sinno et al., 2003; Rivard et al., 2007; Avula et al., (Cinca et al., 1980), we have found that this procedure induces 2018). In dogs, the isolated occlusion of an atrial branch of the patchy atrial infarction with sharp border zone and evolving RCA induced atrial ischemic changes, which were often nearly changes of atrial R/S waves, ST segment, and local activation transmural at the center of the ischemic area but followed a time. The early ischemic changes in atrial electrograms were not patchy distribution at the periphery. Similar to previous studies detectable on the ECG, and only variations in P wave morphology in dogs (Sinno et al., 2003; Rivard et al., 2007), our study in were appreciated in later stages of atrial branch occlusion. pigs showed that a selective occlusion of an atrial branch of the RCA induced a patchy atrial necrosis pattern, perhaps favored Anatomopathological Findings by coronary collaterals that ordinarily interconnect the atrial The structural features of atrial infarction have been analyzed branches in human and porcine hearts (Baroldi et al., 1956; in necropsy studies of patients who presented a concomitant James and Burch, 1958; Weaver et al., 1986). Bragt et al. (2015) Frontiers in Physiology | www.frontiersin.org 6 April 2020 | Volume 11 | Article 264 Amorós-Figueras et al. Atrial Electrograms and ECG in Atrial Infarction measured the coronary blood flow through an atrial branch of showed slowing of atrial conduction and prolongation of the the LCX coronary in pigs and found that it was maximal in the refractory period, leading to a substrate for atrial fibrillation early diastolic phase and ranged 5–10 ml/min. maintenance (Sinno et al., 2003). However, these arrhythmias were absent in our study and also in dogs, in which only increased Epicardial Atrial Mapping stability of electrically induced atrial fibrillation was attained This study reveals that during the first 30 min of selective (Sinno et al., 2003). atrial branch occlusion, the local atrial electrograms show a progressive increase in R wave amplitude associated with ECG Findings parallel prolongation of the local activation time and elevation This study allows to understand the limited diagnostic value of the ST segment leading to monophasic potentials. These of the surface ECG to recognize atrial infarction. The acute changes are fully comparable to those described in acute ischemic changes in the atrial R/S waves and local activation ventricular myocardial ischemia (Janse et al., 1979; Cinca et al., times translate to the surface ECG as unspecific variations of P 1980), and this resemblance suggests that atrial and ventricular wave morphology, but these were only appreciable at advanced myocardial ischemia share similar cellular electrophysiological stages of atrial coronary occlusion. On the other hand, the derangements. Assuming this parallelism, the R/S wave changes characteristic ischemic displacements of the atrial ST segment would indicate the existence of a slow conduction in the ischemic did not induce appreciable shifts of the PR segment at the atrial myocardium and the ST segment elevation from a flow of surface ECG. A relatively low incidence (20%) of PR-segment injury currents created by the loss of resting membrane potential displacement was reported in a series of 39 patients with acute of ischemic cells (Kléber et al., 1978). Likewise, monophasic inferior myocardial infarction submitted to primary coronary potentials will indicate local atrial cellular inexcitability. In two angioplasty and angiographic documentation of atrial branch cases, we observed a transient recovery of local activation time involvement (Yıldız et al., 2018). Moreover, the PR-segment and a reduction in the ST segment elevation similarly found deviation can vanish over time as was reported in a prospective in acute ventricular myocardial ischemia, also in swine. This is series of 17 patients with accidental occlusion of atrial branch not due to a concurrent structural improvement of the ischemic during elective coronary angioplasty (Álvarez-García et al., 2016). myocardium but rather to a concurrent transitory decrease in Selective atrial branch occlusion did not elicit ST segment potassium accumulation in the ischemic myocardium (Janse changes in the surface ECG, thus confirming the absence et al., 1979; Coronel et al., 2019). of concurrent ventricular myocardial ischemia in our atrial One hour after atrial branch occlusion, there was a progressive infarction model. decrease in the magnitude of ST segment elevation despite the persistence of the coronary clamping. The decline of Study Limitations the ST segment elevation overtime was constantly observed A median sternotomy was needed to perform the atrial epicardial in ventricular myocardial ischemia and is caused by the mapping and the direct ligature of an atrial coronary branch. progressive increase in the electrical resistivity in the ischemic Since this intervention modifies the spatial relationship of the myocardium, which impedes the propagation of the injury surface ECG leads and the heart, we focused our analysis on currents throughout the ischemic and normal regions (Kléber lead II because it was the less affected by the sternotomy and et al., 1987; Cinca et al., 1997). is currently accepted as an appropriate lead to evaluate the P The electrical border zone of atrial infarction was abrupt. The wave morphology. The relatively small sample size of this study atrial mapping showed that the transition between electrodes only allows a qualitative description of the characteristics and with normal electrograms and electrodes with ST segment time course of the local atrial electrograms induced by selective elevation was about 1–2 mm. Of note, this electrical border acute atrial infarction. This proof-of-concept approach affords shifted during the course of ischemia as denoted by the foundation for future, more highly powered studies aimed at progressive increase in the number of electrodes showing describing ECG markers of the extension and location of the monophasic potentials after 2 h of atrial branch occlusion. These infarction in the right or left atrial chambers. changes have not been appreciated in ventricular myocardial This study does not allow determining the influence of the ischemia in the same experimental model (Janse et al., 1979). extent of atrial infarction on the magnitude of the P wave changes Reasons for the time-dependent instability of the atrial area at of the surface ECG. The median sternotomy needed to perform risk cannot be fully elucidated. However, local edema induced both the atrial mapping and the direct ligature of the coronary in the course of acute myocardial ischemia could interfere atrial atrial branches disrupts the spatial relationship of the heart with coronary perfusion and expand the injured area. In favor of the ECG leads and, consequently will hamper any attempt to the presence of local edema, we found a reduction in the accurately correlate the magnitude of the P wave changes with amplitude of local electrogram that have not shown ischemic ST the size of atrial infarction. However, the finding of significant segment changes. changes in the P wave morphology during atrial ischemia in our Atrial arrhythmias, and particularly atrial fibrillation, have study suggests, at least, that atrial infarction was large enough been considered diagnostic clues to suspect atrial infarction to project into the surface ECG. Future studies in closed chest involvement in patients with acute occlusion of the RCA or models are therefore needed to precisely quantify the effects LCX coronary arteries (Lazar et al., 1988; Lu et al., 2016). Atrial of atrial infarct size on the P wave characteristics. The lack epicardial mapping in dogs with selective atrial branch occlusion of quantification of the atrial infarct size in this study does Frontiers in Physiology | www.frontiersin.org 7 April 2020 | Volume 11 | Article 264 Amorós-Figueras et al. Atrial Electrograms and ECG in Atrial Infarction not invalidate the epicardial mapping data because the changes and interpretation of the data, drafted the manuscript, and in local atrial electrograms are not dependent on the remote given the final approval of the version to be published and extension of infarction but only on local site-to-site transmural the agreement to be accountable of all aspects of the work in spanning of the ischemic injury. ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Clinical Implications ER-D, DS-Q, SC-R, EJ, JN-M, DA, XM, CA-M, and JG made Translation of our findings to the clinical setting is tenable substantial contributions to the design of the work, participated because the distribution of atrial coronary branches and the in the acquisition of the data, revised the draft critically for electrophysiological derangements caused by acute myocardial important intellectual content, and given the final approval of the ischemia in pigs and humans are comparable (Weaver et al., 1986; version to be published and the agreement to be accountable of Cinca et al., 1998). all aspects of the work in ensuring that questions related to the Our study suggests that although acute occlusion of an atrial accuracy or integrity of any part of the work are appropriately branch in patients with coronary atherosclerosis would cause investigated and resolved. atrial infarction, this condition would be hardly recognized by the surface ECG. Indeed, the specific acute local electrical alterations were only reflected by unspecific changes in P wave FUNDING morphology at late stages of occlusion. However, acute atrial myocardial ischemia would be reliably detected clinically by This work was supported by the Spanish Ministerio de Economia direct atrial recordings, and this may open new diagnostic y Competitividad, Instituto de Salud Carlos III (CB16/11/00276 technological developments applicable to patients with implanted and FIS PI17/00069) and Fondo Europeo de Desarrollo Regional. electrostimulation devices. SUPPLEMENTARY MATERIAL DATA AVAILABILITY STATEMENT The Supplementary Material for this article can be found The datasets generated for this study are available on request to online at: https://www.frontiersin.org/articles/10.3389/fphys. the corresponding author. 2020.00264/full#supplementary-material FIGURE S1 | Photographs illustrating the correspondence between the corners of the 128-patch electrode (A) and the fixed black Prolene snare marks (B). ETHICS STATEMENT FIGURE S2 | Representative microphotographs illustrating the morphometric The animal study was reviewed and approved by the Animal analysis of the extent of area of myocardial myolysis. (A–C) show examples of Experimentation Ethics Committee Research Institute of areas of myolysis of 38, 46, and 72%, respectively. Hospital de la Santa Creu i Sant Pau. FIGURE S3 | Evolving changes in epicardial atrial electrograms in a pig submitted to 4-h of atrial coronary branch occlusion showing a transient recovery of local activation between 30–60 min of occlusion. AUTHOR CONTRIBUTIONS FIGURE S4 | Graphs showing the stability of the mean values of ventricular ST-segment and QRS duration at baseline and at specific times of atrial branch GA-F and JC made substantial contributions to the conception occlusion measured in lead II in six pigs submitted to 4-h of atrial coronary and design of the work, participated in the acquisition, analysis, branch occlusion. REFERENCES Baroldi, G., Mantero, O., and Scomazzoni, G. (1956). The collaterals of the coronary arteries in normal and pathologic hearts. Circ. Res. 4, 223–229. doi: Aguero, J., Galan-Arriola, C., Fernandez-Jimenez, R., Sanchez-Gonzalez, J., 10.1161/01.RES.4.2.223 Ajmone, N., Delgado, V., et al. (2017). Atrial infarction and ischemic mitral Bragt, K. A., Van Nasrallah, H. M., Kuiper, M., Van Hunnik, A., Kuijpers, N. H., regurgitation contribute to post-MI remodeling of the left atrium. J. Am. Coll. Schotten, U., et al. (2015). Dynamic regulation of atrial coronary blood fl ow Cardiol. 70, 2878–2889. doi: 10.1016/j.jacc.2017.10.013 in healthy adult pigs. Heart Rhythm 12, 991–1000. doi: 10.1016/j.hrthm.2015. Álvarez-García, J., Vives-Borrás, M., Ferrero, A., Aizpurua, D. A., Peñaranda, 01.016 A. S., and Cinca, J. (2013). Atrial coronary artery occlusion during elective Cinca, J., Janse, M., Moréna, H., Candell, J., Valle, V., and Durrer, D. (1980). percutaneous coronary angioplasty. Cardiovasc. Revasculariz. Med. 14, 270– Mechanism and time course of the early electrical changes during acute 274. doi: 10.1016/j.carrev.2013.07.007 coronary artery occlusion. An attempt to correlate the early ECG changes Álvarez-García, J., Vives-Borras, M., Gomis, P., Ordonez-Llanos, J., Ferrero- in man to the cellular electrophysiology in the pig. Chest 77, 499–505. doi: Gregori, A., Serra-Penaranda, A., et al. (2016). Electrophysiological effects of 10.1378/chest.77.4.499 selective atrial coronary artery occlusion in humans. Circulation 133, 2235– Cinca, J., Warren, M., Carreño, A., Tresànchez, M., Armadans, L., Gómez, P., et al. 2242. doi: 10.1161/CIRCULATIONAHA.116.021700 (1997). Changes in myocardial electrical impedance induced by coronary artery Avula, U. M. R., Hernandez, J. J., Yamazaki, M., Valdivia, C. R., Chu, A., Rojas- occlusion in pigs with and without preconditioning: correlation with local Pena, A., et al. (2018). Atrial infarction-induced spontaneous focal discharges ST-segment potential and ventricular arrhythmias. Circulation 96, 3079–3086. and atrial fibrillation in sheep role of dantrolene-sensitive aberrant ryanodine doi: 10.1161/01.CIR.96.9.3079 receptor calcium release. Circ. Arrhythmia Electrophysiol. 11, 1–15. doi: 10. Cinca, J., Warren, M., Rodríguez-Sinovas, A., Tresànchez, M., Carreño, A., Bragós, 1161/CIRCEP.117.005659 R., et al. (1998). Passive transmission of ischemic ST segment changes in low Frontiers in Physiology | www.frontiersin.org 8 April 2020 | Volume 11 | Article 264 Amorós-Figueras et al. Atrial Electrograms and ECG in Atrial Infarction electrical resistance myocardial infarct scar in the pig. Cardiovasc. Res. 40, Nishida, K., Qi, X. Y., Wakili, R., Comtois, P., Chartier, D., Harada, M., et al. 103–112. doi: 10.1016/S0008-6363(98)00145-X (2011). Mechanisms of atrial tachyarrhythmias associated with coronary artery Climent, V., Hurlé, A., Ho, S. Y., Sáenz-Santamaría, J., Nogales, A. G., and occlusion in a chronic canine model. Circulation 123, 137–146. doi: 10.1161/ Sánchez-Quintana, D. (2004). Early morphologic changes following microwave CIRCULATIONAHA.110.972778 endocardial ablation for treatment of chronic atrial fibrillation during mitral Noriega, F. J., Jorge, E., Arzamendi, D., and Cinca, J. (2013). Mechanism and valve surgery. J. Cardiovasc. Electrophysiol. 15, 1277–1283. doi: 10.1046/j.1540- diagnostic potential of reciprocal ECG changes induced by acute coronary 8167.2004.04332.x artery occlusion in pigs. Heart Rhythm 10, 883–890. doi: 10.1016/j.hrthm.2013. Coronel, R., de Groot, J. R., and Piek, J. J. (2019). Transient ST-segment elevation 02.022 and coronary flow. Eur. Heart J. 1, 1–2. doi: 10.1093/eurheartj/ehz331 Rivard, L., Sinno, H., Shiroshita-Takeshita, A., Schram, G., Leung, T. K., and Nattel, Cushing, E. H., Feil, H. S., Stanton, E. J., and Wartman, W. B. (1942). Infarction of S. (2007). The pharmacological response of ischemia-related atrial fibrillation the cardiac auricles (atria): clinical, pathological, and experimental studies. Br. in dogs: evidence for substrate-specific efficacy. Cardiovasc. Res. 74, 104–113. Heart J. 4, 17–34. doi: 10.1136/hrt.4.1-2.17 doi: 10.1016/j.cardiores.2007.01.018 James, T. N., and Burch, G. (1958). The atrial coronary arteries in man. Circulation Sinno, H., Derakhchan, K., Libersan, D., Merhi, Y., Leung, T. K., and Nattel, S. 17, 90–98. doi: 10.1161/01.CIR.17.1.90 (2003). Atrial ischemia promotes atrial fibrillation in dogs. Circulation 107, Janse, M. J., Cinca, J., Moréna, H., Fiolet, J. W., Kléber, A. G., de Vries, G. P., et al. 1930–1936. doi: 10.1161/01.CIR.0000058743.15215.03 (1979). The “border zone” in myocardial ischemia. An electrophysiological, Weaver, M. E., Pantely, G. A., Bristow, J. D., and Ladley, H. D. (1986). A metabolic, and histochemical correlation in the pig heart. Circ. Res. 44, 576–588. quantitative study of the anatomy and distribution of coronary arteries in swine doi: 10.1161/01.res.44.4.576 in comparison with other animals and man. Cardiovasc. Res. 20, 907–917. Kléber, A. G., Riegger, C. B., and Janse, M. J. (1987). Electrical uncoupling and doi: 10.1093/cvr/20.12.907 increase of extracellular resistance after induction of ischemia in isolated, Yıldız, S. S., Keskin, K., Avsar, M., Cetinkal, G., Sigirci, S., Aksan, G., et al. (2018). arterially perfused rabbit papillary muscle. Circ. Res. 61, 271–279. doi: 10.1161/ Electrocardiographic diagnosis of atrial infarction in patients with acute inferior 01.RES.61.2.271 ST-segment elevation myocardial infarction. Clin. Cardiol. 41, 972–977. doi: Kléber, G., Janse, M. J., van Capelle, F. J., and Durrer, D. (1978). Mechanism and 10.1002/clc.22987 time course of S-T and T-Q segment changes during acute regional myocardial ischemia in the pig heart determined by extracellular and intracellular Conflict of Interest: The authors declare that the research was conducted in the recordings. Circ. Res. 42, 603–613. doi: 10.1161/01.RES.42.5.603 absence of any commercial or financial relationships that could be construed as a Lazar, E. J., Goldberger, J., Peled, H., Sherman, M., and Frishman, W. H. (1988). potential conflict of interest. Atrial infarction: diagnosis and management. Am. Heart J. 116, 1058–1063. doi: 10.1016/0002-8703(88)90160-3 Copyright © 2020 Amorós-Figueras, Roselló-Diez, Sanchez-Quintana, Casabella- Liu, C. K., Greenspan, G., and Piccrillo, R. T. (1961). Atrial infarction of the heart. Ramon, Jorge, Nevado-Medina, Arzamendi, Millán, Alonso-Martin, Guerra and Circulation 23, 331–338. doi: 10.1161/01.CIR.23.3.331 Cinca. This is an open-access article distributed under the terms of the Creative Lu, M. L. R., De Venecia, T., Patnaik, S., and Figueredo, V. M. (2016). Atrial Commons Attribution License (CC BY). The use, distribution or reproduction in myocardial infarction: a tale of the forgotten chamber. Int. J. Cardiol. 202, other forums is permitted, provided the original author(s) and the copyright owner(s) 904–909. doi: 10.1016/j.ijcard.2015.10.070 are credited and that the original publication in this journal is cited, in accordance National Research Council (2010). Guide for the Care and Use of Laboratory with accepted academic practice. No use, distribution or reproduction is permitted Animals. 8th Edn. Washington, DC: The National Academies Press. which does not comply with these terms. Frontiers in Physiology | www.frontiersin.org 9 April 2020 | Volume 11 | Article 264