Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection or coronavirus disease 2019 (COVID-19) has caused a worldwide pandemic that continues to be relevant globally. Alongside the respiratory system, an arrhythmogenic effect of COVID-19 has been recognized. Tachyarrhythmias such as atrial tachycardia, sinus tachycardia, and ventricular arrhythmias are documented as the most frequent dysrhythmias [1, 2, 3]. Though bradyarrhythmia presenting as atrioventricular block (AVB) and sinus bradycardia are less frequently reported, they are thought to have a worse prognosis [4, 5].

Although increasing amounts of data have reported the incidence of COVID-19-related AVB, however, the clinical characteristics, classification, and outcomes have not been well summarized. Moreover, several hypotheses as cytokine storms, direct viral damage, hypoxia, and immune disorders, have been thought to influence the cardiac conduction system, but the concrete pathogenesis of COVID-19-related AVB remains unknown [6, 7, 8, 9, 10].

In this study, we recruited 17 COVID-19-related AVB patients and summarized their clinical characteristics and outcomes. We hypothesized COVID-19 induces molecular changes that can be detected in the patient’s sera and used mass spectrometry assays for data-independent acquisition to comparatively analyze the proteomics profile of COVID-19-related AVB patients and several control groups.

COVID-19 is a multisystem disease that predominantly influences the respiratory system. However, cardiovascular manifestations including myocardial infarction, myocarditis, and heart failure are not rare [21, 22, 23]. Patients with COVID-19 are at risk for certain arrhythmias, including heart rate–corrected QT interval (QTc) prolongation, supraventricular tachycardia, ventricular tachycardias, and conduction system diseases. Though the incidence of COVID-19-related conduction system diseases is lower than tachycardias, they have been demonstrated to be independent drivers of mortality and adverse events. By retrospectively analyzing 756 patients with COVID-19, McCullough et al. [24] reported atrial fibrillation and flutter occurred in 5.6%, atrioventricular block in 2.6%, and intraventricular conduction block in 11.8%. They concluded that a right bundle branch block or intraventricular block (odds ratio = 2.61, p = 0.002) increased the odds of mortality. Moreover, Pavri et al. [25] revealed that abnormal PR prolongation was associated with an increased risk of endotracheal intubation and death.

The prevalence of COVID-19-related AVB has yet to be well established. In patients hospitalized with COVID-19, studies have reported new-onset AVB occurred in 0.02–11.8% in different study populations [4, 26, 27]. Coromilas et al. [26] retrospectively analyzed patients hospitalized with COVID-19 infection worldwide, and reported the incidence of AVB is 0.02%. Among generalized hospitalized patients with COVID-19, Lao et al. [27] and Antwi-Amoabeng et al. [4] reported that the incidence of AVB was 3.6% and 11.8%. In symptomatic COVID-19 cases, Kunal et al. [23] revealed 5 AVB cases (4.6%) in 109 patients. Our study retrospectively included 72 hospitalized patients with AVB from November 2022 to March 2023 and reported 17 patients with new-onset AVB associated with COVID-19.

The molecular insights for the pathogenesis of COVID-19-related AVB remain to be clarified. Several hypotheses have been proposed, including cytokine storm, direct viral injury, and hypoxia. SARS-CoV-2 infection can induce an increased systemic inflammatory response and cytokine storm manifested by an acute elevation of serum Interleukin-6 (IL-6), Interleukin-1 (IL-1), tumor necrosis factor-$\alpha$ (TNF-$\alpha$), and transforming growth factor- $\beta$1 (TGF-$\beta$1). Accumulating data reveal that inflammatory cytokines can affect the intrinsic conduction system by means of directly affecting the cardiomyocytes, expression, and function of ion channel proteins and connexins [6]. IL-6, IL-1, and TNF-$\alpha$ have been shown to described that modulate the expression of T-type calcium channels (Ca${}_{\text{V}}$3.1 and Ca${}_{\text{V}}$3.2) and L-type calcium channels (Ca${}_{\text{V}}$1.2 and Ca${}_{\text{V}}$1.3) and subsequently slow the atrioventricular conduction [28]. Lazzerini et al. [6] demonstrated that systemic inflammation can cause acute atrioventricular delay, and IL-6 can downregulate the cardiac connexin43 that affects conduction velocity in animal models. Recent research indicates that activation of the transforming growth factor superfamily can cause deleterious effects in multiple organ systems in patients with COVID-19. Wang et al. [29] observed that serum levels of TGF-$\beta$1 were significantly increased in COVID-19 patients. TGF-$\beta$1 is a multifunctional cytokine that plays a crucial role in cardiac arrhythmogenicity. TGF-$\beta$1 plays a pivotal role in regulating T-box-transcription factor T-Box Transcription Factor 3 (TBX3), which controls the embryogenic development of the human cardiac conduction system [30]. Moreover, TGF-$\beta$1 is reported to modulate electrophysiological properties of the atrioventricular junction via canonical Wnt signaling [31].

Comparative proteomics analysis revealed altered proteins that are highly enriched in the inflammatory response. For instance, plasma levels of serum amyloid A protein, C-reactive protein, Alpha-1-acid glycoprotein 1, mannose-binding protein C, haptoglobin, and attractin were significantly elevated in the COVID-19-related AVB group. Among them, serum amyloid A protein, C-reactive protein, Alpha-1-acid glycoprotein 1, and mannose-binding protein C are acute-phase proteins that are closely related to infection and inflammation. C-reactive protein and serum amyloid A protein were also identified to be significantly elevated in previously reported COVID-19 patients [32]. C-reactive protein and serum amyloid A protein are major acute phase proteins in controlling and propagating the acute phase response. Emerging evidence indicates systemic inflammation can negatively influence autonomic function and atrioventricular conduction. Haarala et al. [33] demonstrated that reduced heart rate variability is associated with elevated C-reactive protein and serum amyloid A protein. Lazzerini et al. [28] found that atrioventricular conduction indices are increased with elevated C-reactive protein levels and are rapidly normalized in association with the reduction of inflammatory markers. Such findings corresponded with the clinical fingdings and lab results, which indicated AVB may be the sequelae of myocarditis. Blood tests from myocarditis patients often show elevated cardiac enzymes as troponin and inflammatory markers as C-reactive protein, lactate, and procalcitonin [2]. The median level of hs-cTnI and NT-pro BNP of COVID-19-relatedAVB is significantly higher than controls, which is supportive for diagnosing COVID-19-related myocarditis. Whether those acute-phase proteins are pivotal to the development of AVB in COVID-19 patient needs to be further elucidated.

SARS-CoV-2 may cause direct injury to the cardiac conduction system via the angiotensin-converting enzyme 2 (ACE2) receptor. Cardiomyocyte entry of SARS-CoV-2 may depend on binding the viral spike proteins to ACE2 receptors and on S protein priming by proteases. ACE2 receptors are abundant in the kidney and heart. Moreover, evidence shows ACE2 receptors be located in the cardiac conduction system [34]. Han et al. [8] revealed SARS-CoV-2 can infect primary pacemaker cells in the heart and induce ferroptosis that lead to arrhythmias. Vaduganathan et al. [9] demonstrated that angiotensin II is accumulated via virus-associated down-regulation of ACE2, resulting in adverse myocardial remodeling and arrhythmogenicity.

The majority of the literature documented that COVID-19-related AVB were transient and self-limiting. Dagher et al. [35] reported 4 patients developing a transient high-degree AVB during hospitalization, with 1 patient being implanted with a temporary pacemaker. After 10–21 days from admission, all patients were discharged in sinus rhythm and a normal PR interval. The spontaneous resolution of COVID-19-related AVB reveals the nature of the viral infection. However, cases of permanent AVB patients requiring paroxysmal or permanent pacing have been well reported. In our study, 8 paroxysmal AVB with symptomatic manifestations of bradycardia had permanent pacemaker implantation. Sensed and paced atrioventricular intervals were programmed as 170 and 200 ms to stimulate intrinsic conduction. After a mean follow-up of 2.9 months, the median VP ratio was 99.8%, which reveals the potential long-term damage to the conduction system. Therefore, the long-term follow-up of symptomatic AVB patients with spontaneous resolution is necessary.

We successfully performed left bundle branch area pacing (LBBAP) and right ventricular septal paing (RVSP) in all COVID-19-related AVB patients. No procedural-related complication occurred. The pacing parameters, including threshold, sense, and impendence remained stable at follow-up. Such finding indicated LBBAP and RVSP are both safe and effective to treat COVID-19-related AVB.

Our study has several limitations inherent to the design. The single-center nature of our study may potentially limit the sample size and introduce selection bias. Our study recruited hospitalized AVB patients associated with SARS-CoV-2 infection which lacks the confidence to demonstrate the causal relationship between AVB and COVID-19. However, based on Hill’s Criteria, we observed the temporal order of COVID-19 and the onset of AVB and carried out lab tests to exclude other potential viral causes besides SARS-CoV-2 in our cohort. Moreover, our study is consistent with other cross-sectional studies [4, 26, 27], case report [35], and review [5], which demonstrate the analogy and plausibility of AVB and COVID-19. Based on Han’s study [8], we believe it is reasonable to assume AVB may be the sequelae of SARS-CoV-2 infection. Additionally, most serum samples were collected from the patients in the late period of the disease, thus limiting the proteomic profiling of the acute phase. Our research also lacked a large independent cohort to validate the potential biomarkers. Despite these limitations, we believe our research uncovers important knowledge pertaining to COVID-19-related AVB. Further studies are necessary for further proteomic validation and the detailed pathogenesis of COVID-19-related AVB.

Our study depicts the clinical characteristics and concludes both LBBAP and RVSP are safe and effective to treat COVID-19-related AVB. We provides a valuable resource of proteomics profile in which several differential expressed proteins (Serum amyloid A protein, Tetranectin, Neutrophil defensin 3) enriched to inflammatory response, complement and coagulation cascades, and immune response. Our results sheds light on identifying the potential biomarkers and pathogenesis of COVID-19-related AVB.

The data regarding personal information is subjected to legal restriction of Chinese legislation. The request to access the data can be granted or rejected by State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College.

All authors have read and approved the final manuscript. YG collected the data and wrote the manuscript. KPC conceptualized the study, verified the data and revise the manuscript critically. ZLC and SJW performed the data analysis and wrote the manuscript. RHC, YD, and SZ did the data validation and revise the manuscript. All authors have participated sufficiently in the study and agreed to be accountable for all aspects of the work.

This study complied with the Declaration of Helsinki and was approved by the ethics committee of Fuwai Hospital Chinese Academy of Medical Sciences (2022-ZX031). We obtained informed consent from all participants.

Not applicable.

This research was funded by CAMS Innovation Fund for Medical Sciences (2022-I2M-C&T-B-049) and Beijing municipal science and technology commission (Z191100006619120).

The authors declare no conflict of interest.