Persistent tachycardia can impair left ventricular function and precipitate heart failure. The ensuing cardiomyopathy is known as tachycardia-induced cardiomyopathy. This form of cardiomyopathy is thought to be a reversible form of heart failure. Left ventricular systolic function, measured by the left ventricular ejection fraction (LVEF) and symptoms, can improve or normalize with rate or rhythm control, or both.^1-6 The associated heart failure generally starts to improve within days of achieving ventricular rhythm control, but clinical recovery may take weeks or months.⁷

Atrial fibrillation,^8-10 atrial flutter,¹¹ supraventricular tachycardia,^3,4,12 ventricular tachycardia,^5,6,13 fascicular tachycardia,² ventricular ectopy,¹⁴ and even persistent rapid DDD pacing¹⁵ can cause tachycardia-induced cardiomyopathy. The effects of recurrent tachycardia after resolution of cardiomyopathy have not been carefully assessed. We evaluated the effects of recurrent tachycardia in patients who had apparently resolved cardiomyopathy due to tachycardia.¹⁶ Our suspicion was that tachycardia-induced cardiomyopathy never really completely disappeared, even though there may have been normalization of the LVEF and improvement in heart failure symptoms. We also suspected that recurrent tachycardia would precipitate rapid development of ventricular dysfunction and heart failure symptoms.

Patients and methods

We identified and followed 24 patients (17 men and 7 women) who had tachycardia-induced cardiomyopathy over a 12-year period. These patients had New York Heart Association (NYHA) functional class III or IV heart failure. Eight patients were heart transplantation candidates when the diagnosis of tachycardia-induced cardiomyopathy was first considered. Tachycardias persisted for a median of 4.2 years, with a mean of 8.0 ± 9.1 (range: 1-30) years before heart failure symptoms and impaired LVEF were noted.

The causes of cardiomyopathy included atrial fibrillation (n = 13), atrial flutter (n = 4), atrial tachycardia (n = 3), permanent junctional reciprocating tachycardia (n = 2), idiopathic ventricular tachycardia (n = 1), and bigeminal premature ventricular contractions (n = 1). Radiofrequency ablation was used to treat patients with atrial flutter, atrial tachycardia, permanent junctional reciprocating tachycardia, and bigeminal premature ventricular contractions. Of those with atrial fibrillation, 2 patients had atrioventricular junctional ablation, and 12 required medication to control the rate or rhythm. One patient with atrial fibrillation additionally had atrial flutter and underwent an atrial flutter ablation procedure.

Results

One patient had an atypical form of idiopathic right ventricular tachycardia. She was nonadherent with her medical regimen and was lost to follow-up for 8 years, after which she developed heart failure. Her LVEF was 15%. Rate control with amiodarone (Cordarone, Pacerone) suppressed nearly incessant ventricular tachycardia, normalized the LVEF, and eliminated heart failure symptoms.

A 33-year-old patient developed tachycardia-induced cardiomyopathy after sustaining uncontrolled atrial fibrillation for > 8 years. Treatment resulted in alleviation of symptoms and normalization of the LVEF. When atrial fibrillation recurred 2 years later, the LVEF dropped precipitously, and heart failure symptoms recurred within 6 months. Rate control relieved the symptoms and normalized the LVEF within 1 to 3 months. Several months later, after problems with nonadherence to medication, heart failure symptoms recurred, and the LVEF deteriorated. After amiodarone treatment, symptoms disappeared, and the LVEF normalized.

For all patients, symptoms resolved and the ejection fraction normalized with tachycardia resolution or with rate control. Before treatment, the mean LVEF was 26% ± 9% (range: 10%-40%), and after rate and or rhythm control, it improved to 51% ± 5% (range: 45%-65%; Wilcoxon signed ranks test, S = 150, n = 24, P < .001). The mean time to improvement of the LVEF was 5.8 ± 4.8 months. Symptom resolution mirrored improvement in the ejection fraction.

Tachycardia recurred in 5 patients. Three patients had more than 1 recurrence. For each of these patients, the ejection fraction decreased and heart failure symptoms developed within 6 months of recurrent tachycardia. This was much faster than the initial decline for each patient: 96 ± 109 months (range: 12-360 months) versus 7.2 ± 6.4 months (Wilcoxon signed ranks test, 2-tailed exact P = .063).

Three patients died suddenly. None had new symptoms and all had NYHA functional class I heart failure immediately before sudden, unexpected death. For those who died, the last measured LVEF approximated 50%. No patient had evidence of or symptoms related to loss of rate control at the last evaluation.

A 41-year-old man with untreated longstanding atrial fibrillation developed NYHA functional class III heart failure and syncope. He had a LVEF of 10%. Treatment with amiodarone did not maintain sinus rhythm but did slow ventricular response. After 4 months of treatment, his LVEF improved to 55%, his symptoms resolved, and amiodarone was discontinued. Ultimately, sotalol (Betapace, Sorine) was started and then stopped. He died suddenly 4 months later without evidence of prior heart failure.

A 26-year-old man with uncontrolled atrial fibrillation for 10 years had a heart rate of 195 beats per minute on minimal exertion. He had NYHA functional class III heart failure and an impaired LVEF. Over a 10-year follow-up period, he underwent multiple cardioversions and was treated with amiodarone. His symptoms resolved, and his LVEF improved markedly within a month after each cardioversion. However, within 4 to 8 weeks after recurrence of rapid rates, his condition deteriorated precipitously. He underwent atrioventricular junctional ablation and pacemaker implantation. The symptoms resolved, and his LVEF normalized. Several years later, he died suddenly while gardening.

Discussion

Tachycardia-induced cardiomyopathy was thought to be a completely reversible condition based on resolution of symptoms and normalization of the LVEF after the rhythm was corrected or the rate controlled.^8,12 In this condition, the heart dilates, the LVEF decreases, but hypertrophy does not occur.¹⁷ Our data showed recovery in the LVEF and resolution of symptoms after treatment. Despite apparent resolution of the problem: (1) tachycardia recurrence caused a rapid decline in LVEF with concomitant symptoms, and (2) sudden death occurred in some patients.

The precipitous drop in the LVEF with tachycardia recurrence suggested persistent ultrastructural changes making the left ventricle susceptible to repeated stress.^18-20 This indicates a permanent change in the heart not measurable by the LVEF or by heart failure symptoms.

Animal models support our observations. Rapid pacing can initiate cardiomyopathy in pigs and can lead to ventricular dilatation without hypertrophy.²¹ In a dog model, tachycardia-induced cardiomyopathy may not change heart weights, but myocyte replacement with fibrosis can occur.²²

Myocyte elongation and misalignment and disruption of basement membrane and sarcolemmal interfaces,²³ with developing sarcolemmal festoons²⁴ and changes in myocyte attachment to laminin, fibronectin, and the extracellular collagen matrix, have been described.^23,24 These are thought to be independent of LVEF changes.

Various cytoskeletal changes have been shown to take place in animal models. These include increased betaactin, gamma-actin, and alpha-tubulin; alteration in matrix metalloproteinases (gelatinase, collagenase, and stromolysin);^18-20,25 and depletion of high-energy stores. In a canine model, ventricular dilatation and systolic dysfunction occurred after 1 week of rapid right atrial pacing. Enlarged and disarrayed fibers and mitochondria with disintegrated crystal and an anarchic pattern can occur. Moderate dilation of the rough endoplasmic reticulum and intercalated disk discontinuity can occur after pacing. Matrix metalloproteinase-9 was increased, and tissue inhibitor of matrix metalloproteinase-1 was decreased after the same period.²⁶

While catecholamine levels increase, sympathetic responsiveness as well as beta-1 receptor density is decreased. It is well known that beta-adrenergic blockers enhance reverse remodeling after myocardial infarction.^27-31 It is not clear whether this would apply to tachycardia-induced cardiomyopathy because mechanisms for these conditions may differ. Although LVEF may be a poor surrogate of remodeling, normalization does not necessarily indicate correction of tachycardia-induced cardiomyopathy.³²

The neurohumoral changes seen with tachycardia-induced cardiomyopathy are similar to those seen in other causes of heart failure. There seems to be an activation of the reninangiotensin-aldosterone axis, with abnormalities in sodium handling and elevated levels of atrial natriuretic peptide.³³ Sodium-potassium adenosine triphosphatase (ATPase) activity,³⁴ myofibrillar calcium ATPase activity, and cardiac glycoside binding activity³² are all reduced. This happens in the setting of lower levels of cycling calcium.³⁵

Data in paced animal models are intriguing. When pacing is stopped, hypertrophy continues to develop and diastolic dysfunction persists, although LVEF normalizes.^36,37 This is especially important because high heart rates with recurrent tachycardia, in the setting of reduced myocardial compliance and compromised ventricular filling, may result in shortened filling times and rapidly progressive heart failure. Any combination of these pathophysiological mechanisms could influence the rate at which LVEF declines with recurrent tachycardia. The cause-and-effect relationship is not well established.³⁸ Our report is novel in that it is the first to document the risk of sudden death in this population, even with apparent rate control.

A dog model has been used to determine whether tachycardia-induced heart failure was an effective model for sudden cardiac death.³⁹Six of 25 animals (24%) died suddenly. In 1 dog, a monitor documented polymorphic ventricular tachycardia as the cause of death. Holter recordings showed increasing ventricular tachycardia episodes as heart failure progressed. The corrected QT interval was significantly prolonged (311 ± 25 to 338 ± 25 ms; P < .05), and the monophasic action potential duration increased from 181 ± 19 to 209 ± 28 ms. Dispersion in monophasic action potential duration rose 40%.

Sudden death was reported in a young man with heart failure and supraventricular tachycardia in whom rate control was not achieved.⁴⁰ An echocardiogram showed dilated cardiomyopathy, and an endomyocardial biopsy showed mild interstitial infiltration.

Conclusions

Patients with tachycardia-induced cardiomyopathy may be at long-term risk for sudden death. Surreptitious cardiomyopathy due to occult ultrastructural changes may persist. It has yet to be determined whether rapid and aggressive rate control would prevent structural damage leading to risk of sudden cardiac death.

References