-
PDF
- Split View
-
Views
-
Cite
Cite
Masatomo Ozaki, Keisuke Okawa, Akihiro Oka, Ken Okumura, Intra-atrial block after COVID-19 vaccination-induced fulminant myocarditis: a case report, European Heart Journal - Case Reports, Volume 8, Issue 11, November 2024, ytae580, https://doi.org/10.1093/ehjcr/ytae580
- Share Icon Share
Abstract
Myocarditis is associated with an increased risk of conduction disturbances during the acute phase, which recovers in most cases but rarely recurs during the chronic phase.
A 50-year-old man who developed fulminant myocarditis after COVID-19 mRNA vaccination was discharged 24 days after admission. He was readmitted for heart failure associated with two P waves: one P wave (P1) had a normal amplitude and was dissociated from the QRS, and the other (P2) had a very low amplitude and was associated with the QRS. The patient was referred for pacemaker implantation. Before implantation, an electrophysiological study was conducted using electro-anatomical mapping (EAM). During the P1 wave activation, the sinus rhythm spread to the right atrium but was blocked in the atrioventricular (AV) node area. During P2 activation, the activation originated from the right superior pulmonary vein (RSPV), spread to the left atrium and AV node area through the interatrial septum, and was conducted to the right ventricle, exhibiting a complete intra-atrial block with two discrete rhythms. An atrial lead was successfully placed in a narrow area in the right atrial septum where the pacing captured the left atrium and AV node area under the guidance of EAM findings.
The EAM technology has demonstrated intra-atrial block and two types of atrial rhythms, sinus and RSPV. Electro-anatomical mapping findings were also useful for identifying the ideal pacemaker implantation site. Synchrony between the left atrium and ventricle via His-Purkinje conduction, which was achieved with EAM-guided atrial septal pacing, was prioritized.
Myocarditis is associated with an increased risk of conduction disturbances, including intra-atrial block, during the acute phase. Although conduction disturbances recover in most cases, they can recur during the chronic phase and may require pacemaker implantation.
This patient’s electrocardiogram (ECG) findings could have been misinterpreted as complete atrioventricular block if the dissociated P wave had been overlooked due to its low amplitude. Careful ECG diagnosis is necessary, particularly considering the need for an optimal pacing site for a permanent pacemaker.
Complex conduction disturbances, such as intra-atrial block, can be visually and clearly demonstrated using the latest electro-anatomical mapping technology. The findings of electro-anatomical mapping will also be useful for identifying the optimal pacing site for pacemaker implantation.
Introduction
Myocarditis following COVID-19 mRNA vaccination is rare, but it occasionally deteriorates into a fulminant type.1–3 Myocarditis is associated with an increased risk of conduction disturbance during the acute phase,4,5 which recovers in most cases but rarely recurs during the chronic phase.6 Intra-atrial block is a risk factor for atrial arrhythmias, especially atrial fibrillation, and has been frequently observed during COVID-19 infections.7 However, to our knowledge, there are no reports of intra-atrial block after COVID-19 mRNA vaccination-associated fulminant myocarditis.
We recently reported a patient with fulminant myocarditis following COVID-19 vaccination.7 Although he recovered with steroid therapy, an intra-atrial block with two dissociated P waves developed. We obtained the detailed propagation of cardiac activation using electro-anatomical mapping (EAM) during the intra-atrial block and were able to identify the ideal atrial pacing site for a permanent pacemaker under EAM guidance.
Summary figure
Case presentation
A 50-year-old man with no medical history developed fulminant myocarditis after the second dose of COVID-19 mRNA BNT162b2 vaccination.8 Initial bradycardia, which was considered an advanced atrioventricular (AV) block, recovered after steroid therapy. His haemodynamic parameters recovered completely, and he was discharged 24 days after admission. Two weeks later, he was readmitted due to worsening heart failure. Left ventricular (LV) hypertrophy and a reduced LV ejection fraction (LVEF) of 35% were observed in the initial acute phase, but at this point, the LV findings disappeared and the LVEF improved to the lower normal limit of 55%. No suspicious scar lesions were observed in the atria or ventricles on cardiac magnetic resonance imaging (see Supplementary material online, Video S1).
The electrocardiogram (ECG) showed two types of P waves: a P wave with a cycle length (CL) of 780 ms and normal amplitude (red arrowhead, P1) that was dissociated from the QRS, mimicking a complete AV block with an AV junctional escape rhythm. The other type, with a CL of 1140 ms and very low amplitude (blue arrowhead, P2), was associated with QRS (Figure 1A). Pulse steroid therapy was initiated and switched to oral steroid therapy, but dissociated P waves remained. Atrial fibrillation occurred the following day (Figure 1B), and in the fibrillatory waves, a P wave with a CL of 640 ms, mimicking P1 morphology, was recognized, especially in the V1 lead (red arrowhead). Atrial fibrillation ended spontaneously. His heart failure improved with medication, and he was discharged with oral steroid therapy (prednisolone, 5 mg/day); however, the two P waves remained. During follow-up in the outpatient clinic, complete AV block appeared 11 months after the first admission (Figure 1C). Again, two P waves were observed (red and blue arrowheads) with dissociated QRS complexes. The patient had reduced tolerance to exercise and was hospitalized for implantation of a permanent pacemaker.

Electrocardiogram findings. (A) Electrocardiogram findings on the second day of admission. The electrocardiogram showed two types of P waves: a P wave with a cycle length of 780 ms and normal amplitude (P1) that was dissociated from the QRS, mimicking a complete atrioventricular block with an atrioventricular junctional escape rhythm. The other type with a cycle length of 1140 ms and very low amplitude (P2) was associated with QRS. (B) Electrocardiogram findings obtained on the day after the second admission. Atrial fibrillation occurred, and in addition to the fibrillatory waves, a P wave with a cycle length of 640 ms, mimicking P1 morphology, was observed, especially in lead V1. (C) Electrocardiogram findings at discharge from the second admission. Complete atrioventricular block developed 11 months after the first admission. Again, two P waves were observed with dissociated QRS complexes.
Before the pacemaker was implanted, an electrophysiological study and EAM were performed to diagnose the pathophysiology of the conduction disturbance and identify the ideal pacing site. The catheters were placed in the high right atrium, in the area of His bundle, at the apex of the right ventricle, and in the coronary sinus. The ECG was similar to the one shown in Figure 1A, showing two P waves: one (P1) dissociated from the QRS and the other (P2) associated with the QRS. The intracardiac electrograms (Figure 2) showed that the P1 wave, which was dissociated from the QRS, had a CL of 750 ms (red arrowheads) and exhibited the earliest activation in the high right atrium. The P2 wave, which was associated with the QRS with an AH interval of 68 ms, had a CL of 1095 ms (blue arrowheads) and exhibited the earliest activation at a distal site in the coronary sinus. His-ventricular conduction interval following the P2 wave was within the normal range (48 ms).

Intracardiac electrocardiograms. The P1 wave, which was dissociated from the QRS, had a cycle length of 750 ms and showed the earliest activation in the high right atrium. The P2 wave, which was associated with the QRS with an atrio-His interval of 68 ms, had a cycle length of 1095 ms and exhibited the earliest activation at a distal site in the coronary sinus. The His-ventricular conduction interval following the P2 wave was within the normal range (48 ms). AH, atrio-His; CS, coronary sinus; HRA, high right atrium; HV, His-ventricular; LAO, left anterior oblique; RAO, right anterior oblique.
The right and left atria were mapped using a RHYTHMIA system (Boston Scientific, MA, USA). During P1 wave activation (Figure 3), right atrial activation started from the sinus node area, spread to the right atrium, and was blocked around the AV node area, indicating entrance block to the AV node. During P2 wave activation (Figure 4), atrial activation originated from the right superior pulmonary vein (RSPV), spread to the left atrium and AV node area through the interatrial septum, and was conducted to the right ventricle. Thus, the sinus rhythm was confined to the right atrium without reaching the AV node, whereas the rhythm originating from the RSPV supported not only the activation of the left atrium but also of the ventricle (see Supplementary material online, Video S2).

Activation map of the right atrium during sinus rhythm. During P2 wave activation, right atrial activation began in the sinus node area (A and B), spread to the right atrium (C and D), and was blocked around the atrioventricular node area (E and F), indicating entrance block to the atrioventricular node. PA, postero-anterior; SN, sinus node; RA, right atrium; RAO, right anterior oblique.

Activation map of the right atrial septum and left atrium during right superior pulmonary vein rhythm. During P1 wave activation, the atrial activation originated from the right superior pulmonary vein (A), spread to the left atrium and atrioventricular node area (B, C, and D) via the interatrial septum, and was conducted to the right ventricle (E and F). AV, atrioventricular; LA, left atrium; PA, postero-anterior; RA, right atrium; RSPV, right superior pulmonary vein.
A voltage map revealed large low-voltage areas (defined as ≤0.1 mV) in both atria (Figure 5). Pace mapping in the right atrium was performed to identify the ideal pacing site, but no site where both atria could be captured was identified. By pace mapping along the route of the RSPV rhythm that was conducted to the AV node area, a narrow area was detected in the right atrial septum, where the pacing was captured at a threshold of 116 mA, and the paced atrial activation was conducted to the left atrium and the AV node area (indicated by the yellow arrows in Figure 5). Paced atrial activation at this site was conducted in the ventricles at a ratio of 1:1 up to 120 b.p.m. An atrial lead (Model 3830; Medtronic MN, USA) was placed at this site using a guiding sheath (Model C315S5; Medtronic). A ventricular lead (Model 3830; Medtronic) was placed in the left branch of the right ventricle using a fixed curve sheath (Model C315His; Medtronic). The pacemaker generator (Azure™ XT; Medtronic) was connected to both leads and implanted in the subcutaneous tissue near the left subclavian vein. The pacemaker pacing rate was set at 70–130 b.p.m. in AAIR mode. After pacemaker implantation, there were two P waves consisting of the pacemaker rhythm and the P1 wave (red arrowhead in Supplementary material online, Figure S1). The patient’s exercise tolerance improved, and his condition and pacemaker status showed a stable course with no signs of worsening for more than 12 months. His LVEF improved to 68% 24 months after the first admission (12 months after pacemaker implantation) (see Supplementary material online, Video S1). A weakly positive high-sensitivity cardiac troponin level persisted even with continuous prednisolone treatment (5 mg/day), but the cardiac troponin level eventually became negative.

Voltage map findings of both atria. A voltage map revealed large low-voltage areas (defined as ≤0.1 mV) in both atria. In a narrow area of the right atrial septum, the pacing captured the left atrium and atrioventricular node area at a threshold of 116 mA. AV, atrioventricular; IVC, inferior vena cava; LSPV, left superior pulmonary vein; PA, postero-anterior; RAO, right anterior oblique; RSPV, right superior pulmonary vein; SVC, superior vena cava; TA, tricuspid annulus.
Discussion
There are few reports of COVID-19 vaccination-associated fulminant myocarditis.2,9,10 Furthermore, to the best of our knowledge, there are no reports describing intra-atrial block with P wave dissociation after myocarditis. The ECG findings of the two P waves in this patient could have been misinterpreted as complete AV block if the dissociated P wave (P2 wave in Figure 1) had been overlooked due to its low amplitude. Careful diagnosis of ECG is necessary to ensure appropriate treatment, particularly considering the need for an optimal pacing site for a permanent pacemaker.
To the best of our knowledge, we believe that this is the first case in which an intra-atrial block and two types of atrial rhythms (sinus and RSPV) have been visually and clearly demonstrated using the latest EAM technology. It is likely that atrial inflammation induced by myocarditis and subsequent fibrosis resulted in severe intra-atrial block and P wave dissociation. The EAM findings were also useful for identifying the ideal pacemaker implantation site. In this patient, identifying the location of the atrial lead would have been difficult without the possibility of evaluating the atrial conduction and voltage maps. Pacing from the right atrial appendage and the right ventricle would have caused complete sequential AV pacing, but in this case, the left atrium was dissociated, which may have led to worsening mitral regurgitation. The association between AV conduction abnormalities and diastolic mitral regurgitation has been reported.11 Left atrial and ventricular synchrony via His-Purkinje conduction, which was achieved with EAM-guided atrial septal pacing, was prioritized because none of the current pacing modes could achieve physiologically ideal pacing that synchronized biatrial and ventricular activation.12
Conclusion
We report a patient with two P waves after fulminant myocarditis following COVID-19 mRNA BNT162b2 vaccination. The detailed EAM findings revealed a pathophysiology involving intra-atrial block and RSPV rhythm-producing ventricular activation.
Lead author biography
Dr Masatomo Ozaki, MD, Department of Cardiology of the Kagawa Central Prefectural Hospital. He is a fellow of the Japanese Society of Internal Medicine, board-certified member of the Japanese Circulation Society, and board-certified member of the Japanese Heart Rhythm Society.
Supplementary material
Supplementary material is available at European Heart Journal – Case Reports online.
Acknowledgements
The authors thank the medical staff at Kagawa Prefectural Central Hospital and Mr John Martin for the linguistic assistance.
Consent: Written informed consent was obtained from the patient for the publication of any potentially identifiable images or data included in this article, in accordance with the COPE guidelines.
Funding: None declared.
Data availability
The data relating to this article will be shared upon reasonable request to the corresponding author.
References
Author notes
Conflict of interest: None declared.
Comments