Introduction

Thyroid hormone influences cardiac automaticity, conduction, contractility, and vascular tone at several levels. Two primary rodent studies directly demonstrate thyroid-hormone regulation of the HCN2 pacemaker-channel transcript and the voltage-gated potassium channel Kv1.5 in the rat heart [1, 2], while recent mechanistic and cardiovascular reviews integrate these findings with effects on calcium handling, excitation-contraction coupling, adrenergic responsiveness, and the broader arrhythmic phenotype of hypothyroidism [3, 4]. Severe hormone deficiency can therefore produce bradycardia, low-voltage electrocardiographic complexes, delayed relaxation and impaired cardiovascular reserve; advanced atrioventricular (AV) block remains uncommon but is well described.

The reversibility of AV block in this setting is more difficult to define than the biochemical diagnosis of hypothyroidism. Thyroid-hormone deficiency produces potentially reversible functional electrophysiological changes, including altered pacemaker- and potassium-channel expression, reduced beta-adrenergic responsiveness and autonomic dysregulation; the specific sympathovagal pattern reported in hypothyroidism is heterogeneous [3, 4]. Severe hypothyroidism has also been associated with interstitial oedema and mucinous or mucopolysaccharide myocardial change, while more extensive fibrosis may limit recovery. By contrast, established fibrosis or calcific degeneration of the His-Purkinje system represents a fixed anatomical substrate [5]. Because histological examination of the conduction axis is rarely available during life and was not performed in this patient, these mechanisms provide an explanatory framework rather than patient-specific proof. Degenerative AV block becomes more common with age and with comorbid hypertension and diabetes, and the anatomical level of block cannot always be inferred reliably from the surface electrocardiogram (ECG) alone [5]. A narrow, stable junctional escape may favour a more proximal block, whereas a markedly wide escape complex raises concern for more distal conduction-system disease but is not by itself diagnostic.

Major pacing guidelines agree that potentially reversible causes should be identified and treated, but neither provides a thyroid-specific observation interval before permanent pacemaker implantation [5, 6]. The 2018 ACC/AHA/HRS guideline specifically notes that severe thyroid disease such as myxoedema can rarely be associated with reversible AV block, while also emphasizing that lesser thyroid-function abnormalities often do not eliminate the need for permanent pacing [5].

The clinical literature is similarly heterogeneous. Complete AV conduction recovered after thyroid replacement in individual case reports [7, 8], whereas a retrospective series of 50 patients with thyroid dysfunction and advanced bradyarrhythmia found that 44 received a permanent pacemaker and that thyroid dysfunction was considered unrelated to the AV block in 80% [9]. In that cohort, some conduction recovery occurred only after more than 21 days and after a pacemaker had already been implanted [9]. No evidence-based minimum period of observation on temporary pacing has therefore been defined for thyroid-associated AV block, and the reported time to recovery ranges from a few days to several weeks. No thyroid-specific clinical, biochemical, or electrocardiographic predictor of AV-conduction recovery has been validated. Broader AV-block cohorts have reported heterogeneous associations with recovery or permanent pacing, including younger age, acute coronary syndrome or beta-blocker exposure, absence of aortic-valve calcification, admission blood pressure (BP), sex, and left-ventricular ejection fraction, but these findings were not derived in thyroid-associated AV block and should not be used as a thyroid-specific decision rule [10, 11]. Neither thyroid-stimulating hormone (TSH) magnitude nor escape-QRS width has been prospectively validated as a predictor of recovery in this setting.

Temporary pacing and endocrine treatment address different time horizons rather than competing indications. The 2021 ESC pacing guideline recommends temporary transvenous pacing for haemodynamically compromising bradyarrhythmia refractory to intravenous chronotropic therapy and emphasises that its duration should be minimised because of procedure- and lead-related complications, including perforation, bleeding, malfunction, arrhythmia, accidental displacement, infection and thrombotic complications [6]. In parallel, the 2026 joint consensus statement on myxoedema coma recommends early empiric stress-dose glucocorticoids, intensive supportive care and cautious thyroid-hormone replacement without delaying treatment in a clinically suspected endocrine emergency [12]. Temporary pacing provides immediate rate support, whereas thyroid replacement treats the endocrine substrate but cannot be relied upon for immediate restoration of AV conduction. In an unstable older patient with coronary disease, the two strategies should proceed in parallel, with pacing used as a bridge while cautious endocrine correction is undertaken.

Case reports may over-represent recovery, whereas retrospective cohorts include older patients with competing intrinsic conduction disease and variable durations of observation. A practical assessment therefore requires temporal concordance, exclusion of ischaemic, drug-related and electrolyte causes, review of previous ECGs when available, consideration of the escape-QRS morphology and, when survival permits, direct observation of the rhythm response to thyroid replacement. We report a patient in whom complete AV block was initially treated as an intrinsic or ischaemic conduction emergency, but persistent systemic deterioration despite successful pacing led to recognition of severe hypothyroid decompensation.

Timeline

The verified clinical sequence is summarised in Table 1. The timeline distinguishes findings present at admission from abnormalities that developed later during the evolving endocrine and multiorgan crisis.

Narrative

A 75-year-old man presented to the emergency department approximately 30 minutes after the onset of dizziness and general deterioration. His medical history included diabetes mellitus, hypertension, coronary artery disease, and primary hypothyroidism. He had been prescribed levothyroxine 100 µg/day; the fact that he had not taken it for approximately one month was not known at presentation and was established later during the admission. Medication reconciliation identified no beta-blocker, non-dihydropyridine calcium-channel blocker, digoxin, amiodarone, or ivabradine. No previous ECG was available for reliable comparison.

At the initial documented assessment, body temperature was 36.6°C, BP was 100/60 mmHg while receiving norepinephrine at 0.05 µg/kg/min, respiratory rate was 20–25 breaths/min, and peripheral oxygen saturation (SpO₂) was 98% on room air. Glasgow Coma Scale score was 15. Peripheral oedema was absent at that time. Initial venous blood-gas analysis showed mild acidemia without hypercapnia (pH 7.308, partial pressure of carbon dioxide [pCO₂] 43.7 mmHg, bicarbonate 21.4 mmol/L, lactate 0.9 mmol/L).

Bedside monitoring showed profound bradycardia at approximately 25–30 beats/min. The admission 12-lead ECG, recorded at 25 mm/s and 10 mm/mV, captured complete AV block with regular atrial activity at approximately 41–42 beats/min and a ventricular escape rhythm at approximately 13–14 beats/min (Figure 1). There was no fixed AV relationship. The escape QRS was markedly wide (> 200 ms) and limb-lead QRS voltages were low.

Display full size

Figure 1. Admission electrocardiogram. (A) Twelve-lead tracing recorded at 25 mm/s and 10 mm/mV showing complete atrioventricular block. (B) Expanded lead II segment: red arrowheads indicate P waves (dashed where superimposed on the QRS/ST segment) and asterisks indicate ventricular escape complexes. Atrial rate was approximately 41–42 beats/min and the ventricular escape rate approximately 13–14 beats/min.

Because the presentation was dominated by haemodynamically significant complete AV block, intrinsic or ischaemic conduction-system disease was initially considered. A temporary transvenous pacemaker was inserted within the first hour and increased the ventricular rate to approximately 80 beats/min. Coronary angiography was performed during the same early period. Although rate correction was effective, the patient’s systemic and haemodynamic condition did not improve as expected.

During the subsequent hours, the extremities became cold, bilateral 2+ pretibial oedema and pulmonary oedema developed, and renal function deteriorated. Despite a paced ventricular rate of approximately 80 beats/min and the absence of an acute coronary culprit, the progressive haemodynamic and respiratory deterioration could no longer be explained by the presenting AV block alone. Thyroid-function testing was therefore requested approximately 6 hours after presentation as part of a search for an alternative systemic and potentially reversible cause.

As respiratory status subsequently worsened, continuous pulse-oximetry monitoring showed that SpO₂ fell into the 30% range, consistent with profound hypoxaemia. At that time, BP was 90/60 mmHg while norepinephrine was continued at 0.2 µg/kg/min, and the temporary pacemaker continued to provide ventricular pacing. Controlled endotracheal intubation was therefore performed. Midazolam and rocuronium bromide were administered as part of the intubation regimen. Venous blood obtained during intubation demonstrated severe acidemia and marked hypercapnia (pH 7.020, pCO₂ 77.4 mmHg, bicarbonate 19.5 mmol/L, base excess –11.4 mmol/L, lactate 2.68 mmol/L). A post-intubation venous sample continued to show severe hypercapnia (pH 7.117, pCO₂ 78.9 mmHg, bicarbonate 24.9 mmol/L, lactate 3.03 mmol/L). Shortly after intubation, cardiac arrest occurred. Return of spontaneous circulation was achieved after approximately 40 minutes of cardiopulmonary resuscitation. A second brief cardiac arrest was subsequently resuscitated.

Following resuscitation, the previously requested thyroid-function results became available and demonstrated severe primary hypothyroidism (TSH 106 mIU/L, free triiodothyronine [fT3] 1.18 pg/mL, free thyroxine [fT4] 0.38 ng/dL). When integrated with the persistent cardiovascular, respiratory, renal and metabolic deterioration that was not explained by the corrected AV block, the overall syndrome was recognised as myxoedema coma. Further history established that the patient had not taken his prescribed levothyroxine for approximately one month. Intravenous hydrocortisone 100 mg every 8 hours was administered. Intravenous levothyroxine was given as a total first-day loading dose of 300 µg, divided into three 100-µg doses at 8-hour intervals as 30-minute infusions under continuous cardiac monitoring, followed by 50 µg intravenously once daily. Temporary pacing was continued.

Despite intensive supportive and endocrine-directed treatment, shock and multiorgan failure progressed. The patient was anuric during the final approximately 6 hours. He died on hospital day 3.

Diagnostics

Coronary angiography showed a patent, dominant right coronary artery. The left anterior descending, left circumflex, and right coronary arteries contained atherosclerotic plaques without significant stenosis or an acute culprit lesion; thrombolysis in myocardial infarction (TIMI) 3 flow was present throughout all major epicardial vessels (Figure 2). In this context, acute coronary occlusion or AV-nodal ischaemia was considered an unlikely explanation for the complete AV block.

Display full size

Figure 2. Coronary angiography. (A) Left coronary system and (B) dominant right coronary artery. All major epicardial vessels showed TIMI 3 flow without a significant stenosis or acute culprit lesion. TIMI: thrombolysis in myocardial infarction.

Transthoracic echocardiography approximately 2 hours after presentation showed a left ventricular ejection fraction of 40–45%, inferior wall akinesia, moderate mitral regurgitation, mild aortic and tricuspid regurgitation, an estimated pulmonary artery systolic pressure of approximately 25 mmHg, and a newly documented small posterior pericardial effusion with a maximum depth of approximately 5 mm, without tamponade physiology. The left-ventricular systolic function and inferior wall-motion abnormality were similar to previous echocardiographic findings; no new major structural explanation for the haemodynamic deterioration was identified.

Thyroid-function testing was requested approximately 6 hours after presentation during otherwise unexplained systemic deterioration. The results became available after resuscitation and showed TSH 106 mIU/L, fT3 1.18 pg/mL and fT4 0.38 ng/dL, establishing severe primary hypothyroidism. Serial measurements are shown in Table 2. During profound critical illness, the short-term fall in TSH while fT3 and fT4 remained markedly low could not distinguish inadequate hormone exposure from critical-illness-related, distributional or assay effects. Administration and sampling times, fT4/fT3 trajectories and the clinical response therefore require integrated interpretation.

The admission laboratory profile did not support hyperkalaemia as the cause of the presenting complete AV block: venous-blood-gas potassium was 4.42 mmol/L, while the simultaneous serum potassium of 4.9 mmol/L was flagged as affected by haemolysis. Selected admission laboratory data are summarised in Table 3. A glucose nadir of 45 mg/dL was documented later after intubation, while hyperkalaemia developed only during severe acidosis and multiorgan deterioration.

The admission electrocardiographic measurements are summarised in Table 4. Because atrial and ventricular activity were dissociated, a PR interval could not be assigned. The escape complexes were broad and slurred, with an idioventricular configuration that did not conform to a typical right or left bundle-branch-block pattern. Limb-lead QRS voltages were low. No gross frontal-plane axis deviation was apparent; however, a precise numerical axis was not assigned because of the very low limb-lead voltage and the quality of the archived tracing. The QT interval could not be measured reliably because T-wave termination was indistinct and atrial deflections were superimposed; QTc was therefore not reported. No previous ECG was available for comparison.

Myxoedema coma is a life-threatening decompensated hypothyroid state defined clinically by multisystem dysfunction rather than by a single biochemical threshold [12, 13]. The patient was not hypothermic and was fully conscious at presentation, and serum sodium remained normal; this represents an atypical early phenotype. Applying the 2026 joint-consensus adaptation of the Popoveniuc diagnostic system, the score was 120 when all documented features across the evolving clinical course were incorporated; a score of 60 or above is highly suggestive of myxoedema coma [12, 14]. When restricted to findings documented during the first hours after presentation, with the later-established precipitating event and later abnormalities excluded, the score was 70. Table 5 reports the two calculations separately. Hypotension was scored on the basis of vasopressor dependence rather than the absolute blood-pressure value. Because norepinephrine had already been started before the first documented BP, a sensitivity analysis was also performed: excluding the 20 hypotension points yielded an early-presentation score of 50, within the 45–59 range indicating risk, and an evolving-course score of 100, which remained highly suggestive. The score supports but does not independently establish the diagnosis and must be interpreted with the clinical and biochemical findings, particularly because other critical illnesses can produce overlapping features [12, 14].

Because severe hypothyroidism did not by itself prove that the AV block was endocrine-mediated, competing causes were considered systematically (Table 6). Degenerative conduction-system disease remained the principal competing explanation for the block itself.

Patient perspective

A patient perspective could not be obtained because of the rapid clinical deterioration and subsequent death.

Discussion

Severe primary hypothyroidism was unequivocal in this patient. The clinical question was whether the complete AV block could be attributed to thyroid-hormone deficiency and therefore regarded as reversible. This could not be resolved during the short clinical course.

Conclusions

Complete AV block can be the dominant presenting abnormality in severe hypothyroid decompensation, but correction of the electrical disturbance does not correct the systemic hypometabolic crisis. In this patient, temporary pacing restored ventricular rate and angiography excluded an acute coronary culprit, yet circulatory, respiratory and renal failure continued to evolve until severe primary hypothyroidism and myxoedema coma were recognised.

The case also illustrates why thyroid status should not be used as a surrogate for conduction-system reversibility. Severe hypothyroidism can coexist with intrinsic degenerative AV block. Older age and a markedly wide escape QRS may raise concern for coexisting intrinsic disease, although neither establishes non-reversibility. Immediate pacing and endocrine treatment should proceed in parallel when the patient is unstable, while permanent pacing, when survival permits, should be based on the observed rhythm course rather than on the TSH value alone. The absence of a baseline ECG, electrophysiological localisation and prolonged rhythm follow-up precludes determination of AV-block reversibility in this patient. Prospective studies should evaluate escape-QRS characteristics, anatomical level of block and a standardised observation interval before permanent pacing.