Predictors of high-grade atrioventricular block recovery in Takotsubo cardiomyopathy: a systematic review of case reports

Abstract

Background

Atrioventricular block (AVB) in Takotsubo cardiomyopathy (TCM) is a rarely reported complication. Due to its low incidence and lack of randomised trials, there remain significant gaps in the evidence regarding its underlying mechanisms, the probability of conduction recovery and the optimal timing for permanent pacemaker implantation.

Methods

A systematic review was conducted on published case reports on TCM associated with high-grade AVB, diagnosed simultaneously or within the same hospital admission. Cases with low-grade AVB, sole sinus node disease or TCM induced by pacemaker implantation were excluded.

Results

A total of 50 cases were included in the final analysis; stratified into two groups based on AVB recovery. Fifteen patients (30%) developed complete AVB recovery. Compared with persistent AVB, those who experienced recovery were younger (median (IQR): 66 (61–73) vs 77 (65–82), p=0.02), more frequently reported identifiable emotional or physical stressors (93% vs 49%, p=0.004) and demonstrated earlier recovery of left ventricular ejection fraction (LVEF) within 1 week (60% vs 24%, p=0.02). No significant differences were observed in terms of comorbidities (hypertension in 27% vs 47%, p=0.22), presenting symptoms (chest pain in 40% vs 14%, syncope or presyncope in 27% vs 46%, p=0.22), ST elevation (27% vs 14%, p=0.28) or regional wall motion abnormalities (apical akinesia in 47% vs 46%, p=0.93). On multivariate analysis, the presence of an identifiable stressor was an independent predictor of AVB recovery (HR 8.27; 95% CI 1.07 to 63.73, p=0.04).

Conclusion

In patients with TCM and high-grade AVB, the absence of a clear stressor highlights the possibility of baseline conduction disorder, with a higher risk of persistent AVB. Earlier LVEF recovery and younger age may help to predict AVB recovery and guide patient selection for PPM implantation.

PROSPERO registration number

CRD420251173831.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• There is limited knowledge on high-grade atrioventricular block (AVB) in association with Takotsubo cardiomyopathy (TCM), with evidence restricted mostly to isolated case reports. The underlying mechanisms remain uncertain, with proposed explanations including autonomic imbalance, microvascular dysfunction or underlying degenerative conduction disease. The natural history of AVB in TCM is poorly defined, with scarce follow-up data, leading to uncertainty about the decision and optimal timing for permanent device therapy.

WHAT THIS STUDY ADDS

• This study clarifies that AVB in TCM is mechanistically heterogeneous, with reversible stress-mediated conduction impairment in some patients and persistent conduction disease in others, that may itself have triggered TCM. The presence of an identifiable emotional or physical stressor was a strong predictor of AVB recovery, while younger age and early improvement in left ventricular function showed supportive trends.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• This review supports a more individualised, context-specific approach to pacing decisions in patients with TCM complicated by high-grade AVB. A conservative observation period of approximately 10–14 days may be adopted before committing to permanent pacemaker implantation in stable patients with identifiable triggers or early left-ventricular recovery rather than routine early device therapy. The findings underscore the need for prospective dedicated multicentre studies to better define mechanisms, recovery trajectories and optimal pacing strategies.

Introduction

Takotsubo cardiomyopathy (TCM) is a well-recognised cause of hospital admission that accounts for approximately 30% of cases classified as myocardial infarction with non-obstructive coronary arteries.¹ The condition predominantly affects postmenopausal women and is typically triggered by acute emotional or physical stress. Diagnosis is usually confirmed by elevated troponin levels, non-obstructive epicardial coronary arteries and characteristic regional wall motion abnormalities, myocardial oedema and typically the absence of persistent late gadolinium enhancement (LGE) by cardiac magnetic resonance (CMR) imaging.²

Despite left ventricular (LV) function recovery, the short-term prognosis of TCM is comparable to myocardial infarction.³ This is attributed to a higher risk of in-hospital complications, including acute heart failure, cardiogenic shock, intracardiac thrombi or life-threatening arrhythmias. The development of in-hospital arrhythmias is associated with increased short- and long-term mortality.^{4 5} While ventricular arrhythmias have been extensively studied,⁶ complete atrioventricular block (AVB) remains an under-recognised complication.

Current understanding of AV conduction disturbances in TCM is limited to isolated case reports and small observational series. As a result, major uncertainties persist regarding the underlying mechanisms—whether it represents transient functional stunning, inflammatory reaction, microvascular ischaemia or reflective of pre-existing conduction disease. The natural history of AVB in this setting is similarly unclear, with reported outcomes ranging from complete recovery to permanent conduction system failure. Consequently, clinicians face significant challenges in determining the appropriate timing for permanent pacemaker (PPM) implantation, balancing the risk of unnecessary device placement against the potential risk for recurrent bradycardia or asystole.^{7 8}

To address these uncertainties, we conducted a systematic review of published case reports to examine the baseline characteristics and outcomes of patients with high-grade AVB in acute TCM, with a focus on possible pathophysiological mechanisms and predictors of AVB recovery.

Methodology

This systematic review was registered on International Prospective Register of Systematic Reviews (PROSPERO), registration number (CRD420251173831). A search of this registry revealed no similar studies. The study design was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines.⁹

Search strategy

A structured search was conducted using specific keywords with the following Boolean operations: ((“Takotsubo Cardiomyopathy” OR “Takotsubo Syndrome” OR “stress induced cardiomyopathy” OR “broken heart syndrome”) AND (“Atrioventricular Block” OR “Heart Block” OR “AV block” OR “conduction disorder” OR “conduction disorders”)). The search string was tailored according to four different databases: Medline, Embase, Cochrane and ProQuest. Search included all published articles until 20 August 2025. All retrieved references were exported and uploaded to Rayyan website for further screening and analysis. Duplicate articles were identified automatically, then revised manually before exclusion.

Screening

A combined title and abstract screening was performed independently by two reviewers (HH and PD). This was followed by full-text screening, which was also conducted independently by the same reviewers. Any discrepancies were resolved through discussion until consensus was reached. Inclusion criteria comprised case reports of patients meeting the diagnostic criteria for stress-induced cardiomyopathy, with documented occurrence of high-grade AVB either preceding, concurrent with or following the diagnosis of TCM within the same hospital admission.

Conference abstracts without full texts were excluded. Cases lacking definitive diagnostic confirmation via coronary angiography (CA), left ventriculography or CMR imaging were excluded, as were reports involving myocarditis, pacemaker-induced TCM or incomplete clinical data. Registries were included only if individual-level data were accessible.

Quality assessment

Two independent reviewers evaluated the quality assessment of articles (HH and PD) using the Joanna Briggs Institute (JBI) Critical Appraisal Tools. The assessment uses eight criteria for case reports, covering patient demographics, case presentation, diagnostic tools, interventions and postintervention reporting. Each criterion was answered as ‘yes,’ ‘no,’ ‘unclear’ or ‘not applicable’. An overall quality rating was assigned as follows: ‘Good’ (7–8/8), ‘Moderate’ (5–6/8) or ‘Poor’ (<5/8). Studies rated as poor quality were excluded from analysis due to a high risk of bias.

Data extraction

Two independent reviewers performed manual extraction of the clinical data (HH and HK), including age, gender, comorbidities, specific trigger/stressor, presenting symptom, AVB grade, ST-T changes, LV ejection fraction (EF), regional wall motion abnormality patterns, CA findings, CMR and electrophysiological (EP) study, if performed. A baseline conduction disorder was defined as any form of nodal or infranodal disease that was already known before presentation (left bundle branch block, right bundle branch block, first or second-degree AVB). Data on received medications were retrieved, including pre-presentation therapy with beta-blockers or amiodarone and in-hospital isoprenaline or inotropes. Implantation of a temporary or PPM was documented. Finally, the follow-up data in terms of EF recovery (defined as 10% or more improvement in EF or an absolute EF value >50%) and AVB recovery were recorded, with a focus on event timings. Discrepancies between the two reviewers were revised and resolved by a third reviewer (PD). Finally, the extracted information was tabulated for data synthesis.

Study objectives

Primary objective

1. To identify clinical and contextual predictors of AVB recovery in patients with TCM. AVB recovery was defined as sustained restoration of sinus rhythm or a ventricular pacing (VP) rate less than 1% at follow-up device interrogation.

Secondary objectives

1. To explore potential pathophysiological mechanisms underlying high-grade AVB in the setting of TCM.

2. To assess the rate and timing of AVB recovery among affected patients.

3. To determine the appropriate observation period before considering PPM implantation.

Statistical analysis

Using IBM SPSS Statistics for Windows, V.24, a descriptive statistical analysis was performed. Categorical variables were reported as frequency (percentage), while continuous variables, which did not follow a normal distribution, were presented as median (IQR). χ² test was used to compare categorical variables, whereas the Mann-Whitney U test was used for comparing continuous variables. Univariate followed by multivariate Cox regression analysis was performed to identify the independent predictors of AVB recovery. The HRs and 95% CIs were calculated. For time to AVB recovery analysis, Kaplan-Meier curves were performed with difference between groups compared using the Log rank test. A two-sided p-value of <0.05 was considered significant.

Results

Figure 1 shows the PRISMA flowchart for the current study. Forty-six articles with a total of 50 patients were included in the final analysis (online supplemental file 1). Cases were divided into two groups according to AVB recovery: persistent AVB group (35 patients) and AVB recovery group (15 patients). A direct comparison between both groups was performed for identifying the contrasts in baseline characteristics and clinical outcomes.

Figure 1. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram. AVB, atrioventricular block; JBI, Joanna Briggs Institute; SND, sinus node disease.

Baseline clinical data

Table 1 summarises the baseline clinical characteristics. The median age was 73 (IQR: 62–81) years, with a minority of male patients (6%). A total of 40% had hypertension, while diabetes mellitus was present in 15%. Eight patients (16%) had a previously known conduction disorder prior to presentation.

Table 1. Baseline clinical characteristics.

Variable	Total (n=50)	Persistent AVB (n=35)	AVB recovery (n=15)	P value
Age (years)	73(62–81)	77(65–82)	66(61–73)	0.02
Male gender	3 (6%)	2 (5.7%)	1 (6.7%)	1.00
Hypertension	19 (40.4%)	15 (46.9%)	4 (26.7%)	0.22
Diabetes mellitus	7 (14.9%)	5 (15.6%)	2 (13.3%)	1.00
Hyperlipidaemia	10 (21.3%)	9 (28.1%)	1 (6.7%)	0.13
Smoking	8 (17%)	5 (15.6%)	3 (20%)	0.70
Baseline conduction disorder	8 (16%)	7 (20%)	1 (6.7%)	0.41
Trigger/stressor	31 (62%)	17 (48.6%)	14 (93.3)	0.004
Emotional	18 (36%)	10 (28.6%)	8 (53.3%)
Physical	13 (26%)	7 (20%)	6 (40%)
Presenting symptom				0.22
Mainly chest pain	11 (22%)	5 (14.3%)	6 (40%)
Mainly presyncope/syncope	20 (40%)	16 (45.7%)	4 (26.7%)
Both	4 (8%)	4 (11.4%)	0 (0%)
Mainly shortness of breath	10 (20%)	7 (20%)	3 (20%)
Shock or cardiac arrest	4 (8%)	2 (5.7%)	2 (13.3%)
LVEF				0.31
Normal/mild impairment (EF ≥40%)	14 (37.8)	11 (44%)	3 (25%)
Moderate/severe impairment (EF <40%)	23 (62.2%)	14 (56%)	9 (75%)
RWMA pattern				0.93
Apical akinesia	23 (46%)	16 (45.7%)	7 (46.7%)
Mid segment involvement	26 (52%)	18 (51.4%)	8 (53.4%)
Biventricular	1 (2%)	1 (2.9%)	0 (0%)
ST-T changes				0.28
ST elevation	9 (18%)	5 (14.3%)	4 (26.7%)
Diffuse T-wave inversion	26 (52%)	20 (57.1%)	6 (40%)
Non-specific changes	15 (30%)	10 (28.6%)	5 (33.4%)
Long corrected QT interval	28 (63.6%)	23 (69.7%)	5 (45.5%)	0.17
Corrected QT interval (msec)	594 (547–654)	604 (550–654)	532 (467–646)	0.24
Timing of AVB diagnosis				0.33
Before TCM diagnosis	9 (18%)	7 (20%)	2 (13.3%)
With TCM diagnosis	38 (76%)	27 (77.1%)	11 (73.3%)
After TCM diagnosis	3 (6%)	1 (2.9%)	2 (13.3%)
Grade of AVB				0.46
Second degree AVB, Mobitz II	10 (20%)	6 (17.1%)	4 (26.7%)
Complete AVB	40 (80%)	29 (82.9%)	11 (73.3%)
Receiving beta-blockers before presentation	3 (6%)	2 (5.7%)	1 (6.7%)	0.45

Data are expressed as median (25th–75th percentile) or frequency (percentage) as appropriate.

Bold p-values indicate statistical significance.

^*LVEF data were available in 37 patients.

AVB, atrioventricular block; EF, ejection fraction; LV, left ventricular; RWMA, regional wall motion abnormalities; TCM, Takotsubo cardiomyopathy.

In 31 patients (62%), a clear identifiable stressor was reported before presentation. Regarding the initial echocardiography study, data about LVEF were available for 37 patients, with 62.2% having an LVEF <40%. The LV apical pattern was the most frequently encountered, with 52% of cases having associated mid segment involvement. Diffuse giant T-wave inversion was the most common ECG finding in nearly half of the patients.

On comparing both groups, patients in the recovery group were significantly younger (66 (IQR, 61–73) years vs 77 (IQR, 65–82) years, p=0.02), whereas no significant differences were found in comorbidities. Notably, a clear emotional or physical trigger for TCM was always identified in patients who had recovery (93% vs 49% in the persistent AVB group, p=0.004). The rest of baseline clinical manifestations were similar between both groups.

In-hospital course and clinical outcomes

All patients underwent CA, which revealed non-obstructive epicardial disease (<50% lesions) in 92% (46/50). The remaining four patients (8%) had abnormal findings varying between obstructive CAD (>50% lesions), myocardial bridge or coronary ectasia, all not correlating with the RWMA territory. A minority of patients underwent CMR for confirmation of TCM diagnosis (eight patients, 16%). Similarly, an EP study was performed in only 8/50 (16% of cases) to elucidate the mechanism of AVB.

A temporary pacemaker was inserted in 70% of patients, either aiming for spontaneous AVB recovery or to prevent recurrent long QT-induced ventricular arrhythmias. In nearly one-third of patients, an early recovery of LV function was reported in less than 7 days. The details of in-hospital course and clinical outcomes are listed in table 2.

Table 2. In-hospital clinical course and outcomes.

Variable	Total (n=50)	Persistent AVB (n=35)	AVB recovery (n=15)	P value
Coronary angiography				1.00
Normal	46 (92%)	32 (91.4%)	14 (93.3%)
Abnormal	4 (8%)	3 (8.6%)	1 (6.7%)
EP study	8 (16%)	3 (8.6%)	5 (33.3%)	0.04
CMR imaging	8 (16%)	5 (14.3%)	3 (20%)	0.68
Ventricular arrhythmias	12 (24%)	11 (31.4%)	1 (6.7%)	0.08
TdP	10 (20%)	9 (25.7%)	1 (6.7%)
VF	2 (4%)	2 (5.7%)	0 (0%)
TPM	35 (70%)	26 (74.3%)	9 (60%)	0.33
Timing of EF recovery				0.02
≤1 week	17 (35.4%)	8 (24.2%)	9 (60%)
>1 week	31 (64.6%)	25 (75.8%)	6 (40%)
Device implantation	41 (82%)	35 (100%)	6 (40%)	<0.001
PPM	37 (74%)	31 (88.6%)	6 (40%)
ICD/CRT-D	4 (8%)	4 (11.4%)	0 (0%)
Waiting time before device implantation (days)	9 (6-14)	9 (7-14)	8 (5–14)	0.62
Total follow-up duration after discharge (days)	90 (14–180)	180 (21–300)	42 (7–90)	0.053
Follow-up duration for devices after discharge (days)	121 (36–240)	180 (21–300)	75 (60–90)	0.70

Data are expressed as median (25th–75th percentile) or frequency (percentage) as appropriate.

Bold p-values indicate statistical significance.

AVB, atrioventricular block; CMR imaging, cardiac MRI; CRT-D, cardiac resynchronisation therapydefibrillator; EF, ejection fraction; EP, electrophysiology; ICD, implantable cardioverter defibrillator; PPM, permanent pacemaker; TCM, Takotsubo cardiomyopathy; TdP, Torsade de Pointes; TPM, temporary pacemaker; VF, ventricular fibrillation.

Among the reported 50 subjects, 41 (82%) underwent device implantation. The median time to PPM insertion was 9 days (IQR 6–14). Follow-up data were available for 35 of these patients, whereas device interrogation data were mentioned in only 15. The median follow-up duration after discharge was 121 days (IQR 36–240). Six patients completely recovered with a VP<1%, while the rest of the patients had persistent AVB with a VP ≥90% in the majority of non-recovered cases.

A total of 15 patients developed sustained AVB recovery. Nine patients had early in-hospital recovery after a median duration of 1 day (IQR 1–11 days); therefore, they were discharged without PPM implantation, whereas six patients developed late recovery after discharge as evidenced by device interrogation. The median time for late recovery was 90 days (IQR 60–90). Notably, an additional three patients exhibited intermittent restoration of sinus rhythm, followed by recurrent AVB before discharge, requiring device implantation (within 10 days for two cases and within 3 months for one) and hence were not classified as having achieved sustained AVB recovery.

Predictors of AVB recovery

Univariate analysis was conducted to identify predictors of time to AVB recovery. Variables with a p value <0.20 were entered into the multivariate model. Given the limited number of events (15 patients), only three variables were retained in the final model: patient age, presence of an identifiable trigger and early recovery of EF within 1 week, as detailed in table 3. In the Cox regression analysis, the presence of an identifiable stressor for TCM (HR 8.27; 95% CI 1.07 to 63.73; p=0.04) emerged as an independent predictor of AVB recovery, while early EF recovery within 1 week showed a borderline statistical significance (HR 2.83; 95% CI 0.96 to 8.35; p=0.06). Figures2 3 present Kaplan-Meier curves for different categorical variables.

Table 3. Univariate and multivariate Cox regression analysis for predictors of AVB recovery.

Variable	Univariate analysis			Multivariate analysis
	HR	95% CI	P value	HR	95% CI	P value
Age	0.97	0.92 to 1.01	0.17	0.98	0.93 to 1.04	0.59
Male gender	1.16	0.15 to 8.90	0.89
Hypertension	0.55	0.17 to 1.76	0.32
Diabetes mellitus	0.86	0.19 to 3.86	0.85
Hyperlipidaemia	0.31	0.04 to 2.41	0.26
Smoking	1.34	0.37 to 4.81	0.65
Baseline conduction disorder	0.48	0.06 to 3.67	0.48
Identifiable trigger	7.74	1.01 to 59.22	0.049	8.27	(1.07 to 63.73)	0.04
Main presentation with syncope/pre-syncope	0.46	0.16 to 1.32	0.15
Main presentation with chest pain	1.23	0.41 to 3.71	0.71
Moderate/severe impairment of systolic function (EF<40%)*	2.67	0.57 to 12.41	0.21
RWMA pattern	0.92	0.32 to 2.67	0.88
Mid-segment involvement
ST-T changes	0.75	0.25 to 2.25	0.61
Complete AVB	0.71	0.20 to 2.57	0.60
Development of VA	0.24	0.03 to 1.82	0.17
EF recovery within 1 week	2.94	1.02 to 8.47	0.046	2.83	(0.96 to 8.35)	0.06

Bold p-values indicate statistical significance.

^*LVEF data were available in 37 patients.

AVB, atrio-ventricular block; EF, ejection fraction; LV, left ventricular; RWMA, regional wall motion abnormalities; TCM, Takotsubo cardiomyopathy; VA, ventricular arrhythmia.

Figure 2. Kaplan-Meier curve for the effect of specific stressor on AVB recovery. AVB, atrioventricular block.

Figure 3. Kaplan-Meier curve for the effect of EF recovery timing on AVB recovery. AVB, atrioventricular block; EF, ejection fraction.

Figure 4 displays the receiver operating characteristic curve assessing patient age as a predictor of AVB recovery. The area under the curve (AUC) was 0.70 (95% CI 0.56 to 0.85; p = 0.023). An age threshold of <75 years demonstrated a sensitivity of 80% and a specificity of 57% for predicting AVB recovery.

Figure 4. ROC curve for age versus AVB recovery. AUC, area under the curve; AVB, atrioventricular block; ROC, receiver operating characteristic.

Discussion

This systematic review provides a focused analysis of AVB in the setting of acute TCM, examining underlying mechanisms, clinical patterns and predictors of AVB recovery. Recovery of high-grade AVB in TCM was less frequent and heterogenous in timing. In the pooled analysis of 50 patients, only 15 (30%) achieved sustained AVB recovery, with early recovery occurring in nine patients within a median of 1 day. Our findings suggest that patients with an identifiable triggering stressor are more likely to regain sustained normal AV conduction. Patient’s age and the timing of EF recovery seem to play a role in guiding decisions about PPM implantation.

Incidence of AVB and mechanistic insights

TCM is increasingly recognised as a heterogeneous syndrome, representing 5–6% of female patients with an initial diagnosis of ST-elevation myocardial infarction.³ New-onset conduction disorders have been reported in large TCM registries, with incidence ranging from 2.2% to 2.9%.^{4 10} Similarly, the rate of sinus node disease was 1.3%. In a large Asian study on 7795 TCM patients, the incidence of first, second and complete AVB was 0.77%, 0.32% and 0.64%, respectively.¹¹

The mechanisms underlying AVB in TCM are likely multifactorial. Autonomic imbalance is the most frequently suggested mechanism, particularly due to heightened vagal tone following catecholamine excess.¹² Other proposed factors may involve drug-related interactions, such as beta-blockers or antiarrhythmics used for VA, and transient myocardial oedema, particularly in atypical TCM variants involving the basal or mid-ventricular segments, that could be responsible for transient AVB through inflammation of the conduction system and His bundle.

Underlying degenerative conduction disease, especially in older patients, may be an alternative mechanism that induces TCM. In support of this theory, a large Japanese registry reported that TCM patients with AVB were significantly older than patients with VA, and presented more frequently with AVB at admission, rather than after admission.¹³ These findings were also consistent with a recent multicentre European registry.⁵

Stressors and clinical presentations

Although emotional stress was initially believed to be the primary driver, physical stressors are now reported more frequently—especially in men—and may include trauma, severe medical illness requiring intensive care or invasive procedures.¹⁴ Notably, acute coronary syndrome (ACS) itself has also been identified as a potential trigger for TCM.¹⁵

According to the International Takotsubo Diagnostic Criteria (InterTAK criteria),¹⁶ the presence of a stressor is not usually required for diagnosis. Consistent with earlier TCM studies,¹⁷ our cohort showed that 38% of patients had no identifiable trigger. However, the lack of a clear precipitating factor challenges the theory of sympathetic overstimulation as the main pathophysiology of TCM. Given the wide spectrum of physical stressors and medical conditions, it is possible that additional, less apparent factors may act as triggers, such as intermittent high-grade AVB or underlying neurological or psychiatric conditions.¹⁸

Around 24% of patients with TCM may present with atypical symptoms rather than chest pain, making diagnosis more challenging.¹⁹ In our analysis, a substantial proportion of patients (40%) presented with syncope or presyncope rather than chest pain, reflecting the diagnostic complexity of TCM in the presence of bradycardia. The observation that 16% of patients had pre-existing conduction abnormalities and 18% developed AVB before the diagnosis of TCM supports the hypothesis that, in some cases, AVB could have been the precipitating trigger for TCM rather than a consequence of it.

Diagnostic considerations and role of electrophysiology

Diagnosis of TCM is usually challenging due to overlap with ACS and myocarditis. Although CA and echocardiography remain central to TCM diagnosis, CMR remains the gold standard for confirmation of TCM and exclusion of alternative diagnoses. In patients with TCM, a typical apical ballooning pattern is seen, with evidence of transmural myocardial oedema in T2-weighted images, and absence of LGE, negating the presence of necrosis or fibrosis. Oedema usually follows the pattern of regional wall motion abnormalities, and in some cases, it may extend to the entire ventricular myocardium.¹⁴ Recent studies suggested the presence of low-intensity patchy LGE in up to 25% of TCM cases.²⁰ Persistent LGE in follow-up scans may denote a co-existing pathology.

In this review, RWMA analysis did not demonstrate a clear association between mid-basal involvement and AVB development, making direct inflammation and/or fibrosis of the AV node or conduction system less likely, negating the anatomical involvement theory.

EP studies can help differentiate supranodal from infranodal AV conduction disturbances in TCM; however, their role in determining long-term management and prognosis remains unclear. In the included case reports, EP studies, performed in a minority of cases, demonstrated both supra-Hisian and infra-Hisian conduction disturbances. For instance, Prabhu et al²¹ described a patient with supra-Hisian block whose conduction normalised after 3 weeks, suggesting a potential delayed recovery in some TCM-related AV blocks. Similarly, Cativo et al²² reported a high-degree supra-Hisian block that resolved without the need for permanent pacing, reinforcing the reversible nature of those cases. In contrast, Watanabe et al²³ documented persistent infra-Hisian block necessitating permanent pacing. Collectively, these observations suggest that AVB in the setting of acute TCM has different mechanisms. While EP studies are useful for localising conduction defects, their role in guiding long-term pacing in TCM remains limited, underscoring the need for further research.

Repolarisation ECG changes have been linked to outcomes in TCM. A subanalysis of the InterTAK registry showed that T-wave inversion was associated with a higher risk of 60-day mortality.²⁴ However, the magnitude of ST elevation was not consistently associated with adverse events.¹² In our cohort, ST-T change analysis was not helpful for identification of AVB mechanism nor predicting AVB recovery.

Role of beta-blockers and risk of torsade de pointes

There is no consensus to support the routine use of beta-blockers in the acute stage of TCM,¹² although it seems reasonable to counteract the catecholamine effect and reduce the risk of left ventricular outflow tract obstruction. Beta-blocker use is not free of complications, as it may predispose to bradycardia-induced Torsade de Pointes (TdP) or increase the risk of conduction disorders. In patients with hypotension, severe LV dysfunction or acute heart failure, beta-blocker initiation may worsen the condition. On the other hand, long-term beta-blocker use after TCM recovery was associated with a survival benefit, with controversial evidence on reduction of TCM recurrence.²⁵ Similarly, anti-arrhythmic medications, particularly potassium channel blockers, may further prolong the QT interval resulting in a higher risk of TdP. In our analysis, only three patients were mentioned to be on beta-blockers prior to presentation.

The coexistence of QT prolongation and AV block creates a challenging clinical scenario, as the resulting bradycardia can facilitate TdP. In such cases, temporary pacing is commonly used to maintain an adequate heart rate and reduce arrhythmic vulnerability. In our cohort, nearly one quarter of patients experienced VA in association with high-grade AVB. The markedly prolonged median QT interval of 594 milliseconds reflects substantial electrical instability, amplified by catecholamine excess and heightened myocardial automaticity. QT-related VA are well recognised in this setting, and temporary pacing remains the most reliable method to shorten the QT interval and eliminate risk.¹² Beta-blockers are better avoided in this context, although short-term esmolol infusion has been reported.⁵

Early implantable cardioverter defibrillator implantation is not advised given the transient nature of TCM. Current evidence increasingly supports postponing permanent defibrillator placement during the acute phase, as long-term benefits have not been demonstrated.^{26 27} The use of a wearable defibrillator in the acute phase of TCM after life-threatening arrhythmia has been described.^{13 28} Nevertheless, this recommendation is based largely on expert opinion rather than robust evidence.

Predictors and timing of AVB recovery

Recovery of high-grade AVB in TCM remains unpredictable. Prior reports were controversial, with findings ranging from complete recovery within days⁷ to persistent long-term pacing dependence.²⁸ In a case review, Le et al⁸ suggested that patients with early recovery (within 2 days) may not need PPM. On the contrary, more delayed recovery despite LV function improvement tends to persist in long-term device interrogation, advocating the need for PPM while in-hospital.

Similarly, in the current review, early recovery typically occurred within 24 hours, and these patients were safely discharged without PPM implantation. Conversely, delayed recovery was rare and occurred predominantly in patients with a clear stressor and early LV functional improvement. These findings suggest that patients with an identifiable stressor had a substantially higher rate of AVB recovery (HR 8.27), supporting the concept that stress-mediated autonomic mechanisms may play a role in reversible conduction disturbances.

The median waiting period before PPM implantation was 9 days, while recurrent AVB was documented within 10 days in a few patients who initially showed transient recovery. These findings support a cautious observation period of 10–14 days, particularly in younger patients or those demonstrating rapid LV recovery. The lack of correlation between LV function recovery and AVB resolution in some cases suggests distinct pathophysiological pathways for myocardial stunning and conduction impairment.

Prognosis and clinical implications

Although TCM complicated by VA was consistently associated with a high risk of in-hospital mortality,⁴ the acute in-hospital and long-term implications of AVB in TCM remain uncertain due to limited follow-up data.²⁹ In this analysis, no in-hospital deaths were reported among the included cases, suggesting that short-term outcomes may be more favourable than those typically observed in AVB complicating acute myocardial infarction. However, the possibility of publication bias cannot be excluded, and the absence of reported mortality likely reflects selective reporting rather than the true risk profile.

It is important to distinguish AVB-associated TCM from pacemaker-induced TCM,^{30 31} a separate entity that is triggered by the implantation procedure itself, whose invasive nature induces patient’s stress, with a significant rise in serum catecholamines. Failure to differentiate these conditions may lead to misinterpretation of causality and prognosis.

Overall, the findings of this review suggest that in patients without an identifiable stressor, AVB may represent a primary underlying conduction disorder unlikely to recover. Conversely, in patients with a clear emotional or physical trigger, AVB may be a secondary, reversible manifestation of the acute stress response. Given the variability in recovery patterns, a tailored approach to pacing decisions is essential. A watchful waiting strategy of approximately 10–14 days may be appropriate in clinically stable patients, particularly those with identifiable triggers and early improvement in LV function. However, persistent or recurrent AVB, infra-Hisian block on EP study, or advanced age may justify earlier PPM implantation. TPM may be used in intermittent AVB, or more importantly, to prevent TdP in patients with long QT interval, and insertion of an externalised pacemaker seems reasonable in patients with delayed AVB recovery.¹⁰

Routine continuous telemetry of TCM patients during the acute hospitalisation phase is reasonable, particularly for those with prolonged QT interval, conduction disturbance or haemodynamic instability. Nevertheless, prospective studies are required to determine the optimal duration and modality of rhythm monitoring, as well as the role of routine Holter assessment following discharge.

Strengths and limitations

This is the first systematic review specifically dedicated to address the outcomes of high grade AVB in the context of TCM. The study adhered to the PRISMA methodology and incorporated rigorous screening, independent data extraction and structured quality assessment using the JBI tool. A detailed case-level analysis allowed exploration of clinical patterns, timing of events and potential predictors of AVB recovery that were not captured in large retrospective registries.

Limitations of this study are the high risk of publication bias and selective reporting as our findings are based on published case reports. Additionally, large TCM registries could not be included due to the lack of accessible individual-level data on AVB patients. The small sample size further restricts the generalisability of our results and the wide CI for the HR likely reflected the limited number of recovery events. The follow-up pacing data were inconsistent limiting the ability to draw firm mechanistic conclusions. Finally, as all the reported patients survived to discharge, this study could not evaluate the impact of AVB on mortality in TCM patients. Larger prospective studies are needed to validate the observations and hypotheses generated by this analysis.

Conclusion

This systematic review highlights the complex interplay between TCM and high-grade AVB. The presence of an identifiable stressor emerged as the strongest predictor of AVB recovery, suggesting that reversible autonomic or stress-mediated mechanisms may predominate in this subgroup. Early LV functional recovery and younger age also appear to favour conduction restoration although these associations require further validation. Ultimately, larger prospective studies are needed to clarify the mechanisms, natural history and optimal management of AVB in TCM. Until such data are available, individualised clinical judgement remains the cornerstone of care.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.