Ventricular Tachycardia in Valvular Heart Disease
Facilitation of Sustained Bundle-Branch Reentry by Valve Surgery
Background The clinical characteristics of sustained monomorphic ventricular tachycardia (SMVT), when it develops after valve surgery, have not been described.
Methods and Results Between 1985 and 1996, 31 patients (30 men and 1 woman) who had undergone valve surgery were found to have inducible SMVT. Nine patients (29%) had sustained VT due to bundle-branch reentry (BBR) (group 1). Four of these patients had normal left ventricular function, and VT with a right bundle-branch morphology was inducible in 4 patients. Group 2 included 20 patients with inducible myocardial (ie, non-BBR) VT. Coronary artery disease was present in 15 group 2 patients (75%) due to atherosclerotic (n=12) and nonatherosclerotic (n=3) causes. Two patients had both inducible sustained BBR and myocardial VT (group 3). Sustained BBR VT occurred significantly earlier after valve surgery (median, 10 days) than the onset of postoperative myocardial VT (median, 72 months; P<.005).
Conclusions Myocardial VT was the most common type of inducible SMVT in patients with valvular heart disease. The majority of these patients had underlying coronary artery disease and significant left ventricular dysfunction. However, in almost one third of the patients, sustained BBR VT was the only type of inducible SMVT. This type of VT was facilitated by the valve procedure occurring within 4 weeks after surgery in most patients. In these patients, left ventricular function was relatively well preserved, and the right bundle-branch block type of BBR was frequently induced. Because a curative therapy can be offered to these patients (ie, bundle-branch ablation), BBR should be seriously considered as the mechanism of VT in patients with valvular heart disease, particularly if the arrhythmia occurs soon after valve surgery.
Valve replacement is frequently performed in patients with severe valvular heart disease. However, after surgery, patients continue to be at risk for untoward events from progressive myocardial dysfunction, prosthetic valve–related problems, and sudden cardiac death.1 In a large series of 1533 patients who underwent aortic or mitral valve replacement, 20% of the deaths were sudden, and 6% of the deaths were directly attributed to arrhythmias.2 Among patients with aortic valve disease, VT has been documented as a cause of sudden death.3 Furthermore, ventricular arrhythmias seem to increase in frequency soon after valve surgery.4
Myocardial stretch, ischemia, and excess catecholamines are some of the postulated mechanisms for ventricular arrhythmias in patients with valvular heart disease.5 Few data exist on the clinical characteristics and electrophysiological mechanisms of inducible sustained VT in patients with valvular heart disease. Although VT due to BBR has been reported in patients with valvular heart disease,6 the prevalence and characteristics of this arrhythmia have not been addressed. The purposes of this study therefore were (1) to describe the clinical and electrophysiological characteristics of SMVT in a cohort of patients who had undergone valve surgery and (2) assess the prevalence of BBR as the mechanism of inducible sustained VT. To our knowledge, this is the first study to systematically analyze the characteristics of SMVT in this clinical setting.
We reviewed the clinical, electrophysiological, and surgical records of 96 consecutive patients who had undergone valve surgery and were subsequently referred for electrophysiological evaluation of documented or suspected ventricular arrhythmias between 1985 and 1996. Only patients with inducible SMVT were included in this study. Patients underwent a right and left heart catheterization to assess the hemodynamic severity of their valve disease, presence of coronary artery disease, and left ventricular function. The severity of valvular lesions was also assessed by Doppler echocardiography. Patients with acute mitral regurgitation due to myocardial infarction were excluded.
Electrophysiological evaluation was performed in the postabsorptive state according to previously described methods.7 When possible, the baseline electrophysiological evaluation was performed in the absence of antiarrhythmic drugs. Several multipolar electrode catheters were introduced through central and/or peripheral veins. These were positioned under fluoroscopic guidance in the high right atrium, across the tricuspid valve to record His-bundle and/or right bundle-branch electrograms, and in the right ventricle.
Programmed electrical stimulation was performed from the right ventricular apex or outflow tract using constant or short-to-long basic drives, and up to three extrastimuli were introduced. Intravenous isoproterenol was used if patients were on antiarrhythmic drugs or VT was not inducible at baseline. In patients not taking oral antiarrhythmic drugs, intravenous procainamide8 (up to 10 mg/kg) was administered if VT was not inducible, and the pacing protocols were repeated. Surface ECG leads (I, II, and V1), intracardiac electrograms, and time lines were displayed simultaneously on an oscilloscope and printed on a thermal recorder. Electrical stimulation was performed using a digital stimulator (Bloom Associates).
Diagnostic criteria for BBR VT have been previously reported9,10 and include (1) the QRS morphology of the tachycardia exhibits a typical RBBB or LBBB pattern; (2) the onset of ventricular depolarization is preceded by His-bundle (H), right bundle-branch, or left bundle-branch potentials with an appropriate sequence of H-RB-LB activation and relatively stable HV (ventricle), RB-V, or LB-V intervals; (3) spontaneous variations in VV intervals are preceded by similar changes in HH intervals; (4) the induction of tachycardia during programmed stimulation is consistently dependent on achieving a critical conduction delay in the His-Purkinje system; and (5) BBR is noninducible after successful RBB ablation. If the VT was induced by programmed stimulation but did not fulfill these criteria, it was considered to originate in the myocardium (ie, non-BBR VT or myocardial VT).
Ablation of the RBB was performed using previously described techniques.9 A 7F deflectable quadripolar catheter with a 4-mm tip was positioned across the tricuspid valve to record the RBB potential. Radiofrequency energy was delivered when a stable RBB potential was obtained from the distal electrodes of the ablation catheter. The end point of the procedure was the appearance of complete RBBB on the surface ECG and noninducibility of BBR tachycardia. In patients treated with an ICD, discharges were considered appropriate if they were preceded by syncope or the electrograms were indicative of a ventricular arrhythmia (ie, the morphology of the electrograms was different from those of sinus rhythm).
Follow-up information was obtained from the hospital records and by contacting the patients or the referring physicians by telephone. We compared the clinical and electrophysiological characteristics of patients with BBR (group 1) and non-BBR VT (group 2).
Data are expressed as the mean±1 SD for all continuous variables. The mean values of the two groups were compared using the Student’s t test (unpaired). Fisher’s exact test was used to compare the categorical variables. A value of P<.05 was considered statistically significant.
Ninety-six patients with valvular heart disease and previous valve surgery underwent an electrophysiological study for a documented or suspected ventricular arrhythmia. Patients were excluded (n=65) if they had polymorphic VT or ventricular fibrillation (n=8), nonsustained VT (n=8), or no inducible sustained VT (n=49) during programmed electrical stimulation. There were 31 patients (30 men and 1 woman; mean age, 64±13 years) with inducible SMVT that developed after valve surgery. Nine of the 31 patients (29%) had sustained BBR VT (group 1), and 20 patients (65%) had non-BBR, myocardial VT (group 2). Two patients (6%) had inducible VT due to both mechanisms (group 3). Cardiac arrest, syncope, or presyncope was the presenting symptom in >70% of these patients (7 patients in group 1 and 15 patients in group 2). The left ventricular ejection fraction of the 31 patients was 36±16% (range, 13% to 68%). Four of the 9 patients (44%) in group 1 had normal left ventricular function (ie, ejection fraction, ≥55%) compared with 1 of the 20 patients (5%) in group 2 (P≤.05). The baseline clinical characteristics of both groups are summarized in Table 1⇓.
Underlying Valvular Disease
Of the 31 patients, 29 (94%) had undergone prosthetic valve replacement. The remaining 2 patients underwent mitral valve repair for mitral regurgitation (n=1) and aortic valvotomy for aortic stenosis (n=1), respectively (Table 2⇓).
All 9 patients (29%) with inducible sustained BBR VT (Figs 1⇓ and 2⇓) had previously undergone valve replacement (metallic prosthesis, n=7; bioprosthetic valve, n=2). The indication for valve replacement was aortic valve disease in 7 patients (23%) (aortic regurgitation, n=5; aortic stenosis, n=2) and mitral regurgitation in 2 patients (6%).
Among the 20 (65%) patients with myocardial VT (Fig 3⇓), 18 patients had undergone valve replacement (metallic prosthesis, n=16; bioprosthetic valve, n=2). The indication for valve replacement was aortic valve disease in 9 patients (29% (aortic stenosis, n=4; aortic regurgitation, n=5), mitral regurgitation in 6 patients (19%), and the involvement of both aortic and mitral valves in 3 patients (10%). One patient with mitral regurgitation had undergone mitral valve repair, and 1 patient with aortic stenosis had undergone aortic valvotomy.
In 2 patients (6%), both BBR VT and myocardial VT were inducible; they developed VT 3 and 8 years after aortic valve replacement for aortic stenosis.
Concomitant Coronary Artery Disease
Two of the 9 patients had three-vessel coronary artery disease (patients 3 and 9; Table 3⇓).
Fifteen of the 20 patients (75%) had significant coronary artery disease: atherosclerotic coronary artery disease in 12 and nonatherosclerotic coronary occlusion in 3 (coronary embolism, n=2; surgical injury to the circumflex artery, n=1). Two of these 3 patients developed a perioperative myocardial infarction. The other patient was suspected of having a coronary embolism 15 years after valve replacement (Table 4⇓).
Both of the group 3 patients had prior myocardial infarction due to coronary artery disease.
The HV interval was 82±13 ms in group 1 and 51±15 ms in group 2 (P<.001). During right ventricular programmed stimulation, SMVT was induced in all 31 patients. Five patients were on antiarrhythmic therapy for recurrent VT before electrophysiological testing.
In group 1, the cycle length of the induced tachycardia was 276±46 ms (versus 260±39 ms for the clinical VT in 4 patients). The morphology of the induced BBR tachycardia was LBBB in 5 patients, RBBB in 1 patient, and both RBBB and LBBB in 3 patients (the spontaneous VT was documented by ECG only in 2 patients, both of whom with an LBBB morphology).
Among the group 2 patients, the tachycardia cycle length was 304±47 ms (P=NS compared with group 1). The morphology of the VT was RBBB pattern (n=13), LBBB pattern (n=4), and multiple VT morphologies (n=3).
In the group 3 patients, the BBR VT cycle length was 300±113 ms, and the morphology was RBBB in 1 patient and LBBB in the other. The myocardial VT cycle length was 350±28 ms, and the morphology was RBBB in 1 patient and LBBB in the other.
Facilitation of BBR VT by Valve Replacement Surgery
Among the 9 group 1 patients who had undergone valve replacement surgery, sustained BBR VT occurred in the early postoperative period (median, 10 days) in 7 patients (range, 1 to 27 days). In another patient with an episode of VT a few days before aortic valve replacement, the tachycardia became incessant and refractory 24 hours postoperatively, requiring multiple cardioversions and emergent electrophysiology study and catheter ablation of the RBB. Therefore, in 8 of 9 patients, the tachycardia developed for the first time or worsened in the immediate postoperative period. In the remaining patient, the VT occurred 4 years after the valve surgery. None of the group 1 patients had intramyocardial VT inducible before or after ablation of the RBB.
Myocardial VT and Valve Replacement Surgery
Among the 20 group 2 patients, VT developed after a median duration of 72 months (range, 1 day to 23 years) after valve surgery. Five patients with myocardial VT had no associated coronary artery disease. In this subgroup of patients, the VT cycle length was 295±47 ms; the morphology was LBBB in 2 and RBBB in 3 patients. Of these 5 patients, 3 had severe left ventricular dysfunction with left ventricular ejection fractions of <30%. Eight of 9 patients in group 1 (89%) presented with VT in the first postoperative month versus 3 of 20 patients (15%) in group 2 (P<.01) (Table 5⇓).
Treatment and Follow-up
Eight patients were treated with catheter ablation of the RBB, and 1 patient received an ICD at another institution. The mean follow-up was 30±29 months. Two patients treated with RBB ablation died from progressive heart failure. The other 7 patients are alive and well. Only 1 patient is on antiarrhythmic therapy (amiodarone for atrial fibrillation).
Of the 20 patients, 19 received ICDs (95%) and 1 patient was treated with amiodarone. The mean follow-up of this group was 38±28 months. There were 7 deaths: 6 patients died of cardiac failure, and 1 patient died from recurrent VT (the implanted defibrillator was turned off at the patient’s request). One patient with progressive left ventricular dysfunction is being considered for cardiac transplantation. Among the patients treated with ICDs, 10 received appropriate discharges for recurrent VT. Two patients were lost to follow-up.
Both patients received appropriate ICD discharges for recurrent VT. One patient subsequently underwent cardiac transplantation for progressive heart failure. Both are alive and well.
Two types of SMVT were seen in patients with valvular heart disease: myocardial VT, the most common type, was usually associated with underlying coronary artery disease (75% of the patients), and/or significant left ventricular dysfunction (mean left ventricular ejection fraction, 33%). The other type of SMVT was BBR, which accounted for about one third of all inducible SMVTs. Although the prevalence of BBR VT in this study was similar to that in patients with idiopathic dilated cardiomyopathies, several unique features of BBR VT were seen in those patients with valvular heart disease, as discussed below.
Role of Anatomic Substrate in the Mechanism of VT
Inducible SMVT is unusual in the absence of a prior myocardial infarction or ventricular dysfunction.11 Patients with valvular heart disease but without the above-mentioned substrates are, therefore, less likely to have VT due to reentry in the myocardium. Because these patients frequently have disease in the His-Purkinje system, BBR may emerge as an important mechanism of SMVT in this population, regardless of their underlying ventricular function. The proximity of the bundle of His and proximal bundle branches to the aortic and mitral valve annuli12 makes them vulnerable to the pathological processes involving either of these valve structures and accounts for the frequent occurrence of conduction abnormalities in these patients.13,14
Sudden death due to ventricular arrhythmia has been reported in patients with aortic valve disease,3 and ventricular arrhythmias have been documented before sudden death in patients with mitral valve disease.15 As seen in this and prior studies, surgical correction of the valve lesion does not eliminate the risk of ventricular arrhythmias or sudden death.2 In studies based on Holter monitoring, VT was observed in ≈10% of patients after aortic valvulotomy and in 13% after aortic valve replacement.16
Sustained BBR in the Setting of Valvular Heart Disease
Although sustained BBR VT is known to occur in patients with valvular heart disease, only isolated cases have been previously reported.17,18 In this series of patients with BBR VT due to valvular heart disease, we describe several previously unreported features.
First, sustained BBR VT occurred a median of 10 days after valvular surgery compared with myocardial VT that occurred a median of 72 months after valve surgery. Given the long natural history of valvular heart disease before valve replacement, it is striking that the onset of BBR VT clustered in the immediate postoperative period in most patients (89%). This is in sharp contrast to myocardial VT, which occurred a median of 72 months after surgery. Valve surgery may worsen the His-Purkinje system conduction abnormalities known to be a prerequisite for sustained BBR.10 Conduction abnormalities are common in valvular heart disease13,14 due to the associated ventricular dilatation and the calcification of the valvular annuli. Because the valvular annuli are in close proximity to the specialized conduction system,12 the conduction abnormalities may worsen by the excision of the native valve and its replacement.19,20 In this context, hemorrhagic lesions involving the His bundle and bundle branches among patients who died early after valve replacement have been reported.21,22 These factors, along with the heightened adrenergic state of the postoperative period, could facilitate BBR VT. It is notable that among 13 patients with valvular heart disease and sustained BBR VT,6,17,18,23 BBR VT occurred in 11 of these patients (85%) after valve replacement, a finding almost identical to ours. In a report by Touboul et al,17 a patient developed sustained BBR VT in the immediate postoperative period (ie, within 48 hours) after aortic valve replacement. Mehdirad et al18 reported 8 patients with valvular heart disease and BBR tachycardia. All these 8 patients developed BBR VT after valve replacement surgery, but the interval was not reported.
A high incidence of sudden death in the first 2 years after valve surgery has been reported particularly in those with intraventricular conduction abnormalities.24 Interestingly, the risk for sudden death peaked at ≈3 weeks after mitral or aortic valve replacement surgery.2 The temporal relationship of BBR tachycardia and valve surgery documented in this study and the postoperative sudden deaths reported in the literature are very similar. Our data suggest that sustained BBR VT may be a more common cause of postoperative mortality than is usually suspected in this setting.
(2) Forty percent of patients with sustained BBR VT had normal left ventricular function. In the present study, the ejection fraction was 43±19% and 4 patients had ejection fractions ≥55%. As in other settings where sustained BBR may occur, significant left ventricular dilatation and/or dysfunction is not essential,25 and the HV interval is significantly prolonged.22 The mere presence of conduction abnormalities in the His-Purkinje system may create the appropriate electrophysiological milieu for the development of sustained BBR VT.
(3) Forty percent of the patients had inducible BBR VT with an RBBB configuration. In contrast to patients with dilated cardiomyopathies in whom LBBB-BBR is, by far, the most common type of BBR and RBBB-BBR is rarely seen (ie, 6% of all BBR VTs in the absence of valvular heart disease [unpublished observations]), BBR associated with valvular heart disease frequently manifests an RBBB QRS configuration. BBR VT with RBBB has been previously reported in patients with valvular heart disease.26 In this study, >40% of the patients had RBBB-BBR. The reason for this is not clear. However, the different underlying substrate and trauma associated with valve surgery may result in different types or degrees of His-Purkinje system conduction abnormalities. Nevertheless, this morphology could allow this mechanism of VT to go unrecognized, particularly if the left ventricular function is normal.
Intramyocardial VT in Valvular Heart Disease
Intramyocardial SMVT was the most common type of VT in this study and accounted for 70% of all inducible VTs in this patient population. In contrast to BBR VT, there was no temporal relationship between the onset of myocardial VT and the valve surgery. There were no significant conduction abnormalities in the His-Purkinje system as evidenced by normal HV intervals. The most common underlying substrate for myocardial VT was coronary artery disease and prior myocardial infarction or associated left ventricular dysfunction.
In contrast to a prior study of patients with sustained BBR VT and dilated cardiomyopathies,23 in which the cardiovascular mortality was 40% during a 15-month follow-up, the long-term survival in this study was 78% during a follow-up period of >2 years. This may be related to the different degrees of left ventricular dysfunction (mean ejection fraction, 23% versus 40% in this study) and the potential improvement of underlying ventricular function after valve surgery.
Unlike patients with BBR VT, patients with myocardial VT have a high recurrence rate of VT (up to 52% in those with ICDs), because of the palliative nature of the chosen therapy. Although ICDs prevented sudden death in many of these patients with appropriate ICD discharges, the rate of deterioration of left ventricular function ultimately determined longevity in this group.
The study group was a selected population. For example, the number of patients with severe valvular disease with arrhythmic sudden death before valve surgery is unknown. Furthermore, patients in whom clinical VT was not documented or who presented with ventricular fibrillation did not show clear evidence that the induced VT triggered the episode of cardiac arrest or syncope.
Sustained BBR accounts for ≈30% of all SMVT after valve surgery, and it emerges as the most common type of VT in the early postoperative period. In this setting, the underlying ventricular function may be preserved, and the VT may have an RBBB morphology. Because bundle-branch ablation eliminates BBR, this mechanism should be seriously considered when SMVT occurs early after valve surgery. Myocardial VT usually occurs with underlying coronary artery disease (due to atherosclerotic and nonatherosclerotic causes) and/or ventricular dysfunction.
Since the original submission, another patient developed incessant BBR VT (with both RBBB and LBBB QRS morphology) at 240 bpm 3 days after aortic valve replacement for critical aortic stenosis. He had normal left ventricular function and no coronary artery disease. Because of multiple cardioversions, he underwent emergent electrophysiological evaluation and catheter ablation of the RBB, which successfully eliminated the VT.
Selected Abbreviations and Acronyms
|LBBB||=||left bundle-branch block|
|RBBB||=||right bundle-branch block|
|SMVT||=||sustained monomorphic ventricular tachycardia|
We would like to thank Brian Schurer for preparation of the figures.
- Received April 17, 1997.
- Revision received August 25, 1997.
- Accepted September 12, 1997.
- Copyright © 1997 by American Heart Association
Blackstone EH, Kirklin JW. Death- and time-related events after valve replacement. Circulation. 1985;74:753–767.
Janse MJ. Possible mechanisms of arrhythmias in valvular heart disease. Acta Cardiol. 1992;47:93–98.
Toboul P, Kirkorian G, Atallah G, Moleur P. Bundle branch reentry: a possible mechanism of ventricular tachycardia. Circulation. 1983;67:674–680.
Waldo AL, Akhtar M, Brugada P. The minimally appropriate electrophysiologic study for the initial assessment of patients with documented sustained monomorphic ventricular tachycardia. J Am Coll Cardiol. 1985;6:1176–1177.
Tchou P, Jazayeri M, Denker S, Dongas J, Caceres J, Akhtar M. Transcatheter electrical ablation of the right bundle branch: a method of treating macro reentrant ventricular tachycardia due to bundle branch reentry. Circulation. 1988;78:246–257.
Caceres J, Jazayeri M, McKinnie J, Avitall B, Denker S, Tchou P, Akhtar M. Sustained bundle branch reentry as a mechanism of clinical tachycardia. Circulation. 1989;79:256–270.
Josephson ME, Horowitz LN, Farshidi A, Kastor JA. Recurrent sustained ventricular tachycardia, I: mechanisms. Circulation. 1978;57:431–440.
Dhingra RC, Amat-y-Leon F, Pietras RJ, Wyndham C, Deedwania PC, Wu D, Denes P, Rosen KM. Sites of conduction disease in aortic stenosis. Ann Intern Med. 1977;87:275–280.
Wolfe RR, Driscoll DJ, Gersony WM, Hayes CJ, Keane JF, Kidd L, O’Fallon WM, Pieroni DR, Weidman WH. Arrhythmias in patients with valvular aortic stenosis, valvular pulmonary stenosis, and ventricular septal defect: results of 24-hour ECG monitoring. Circulation. 1993;87(suppl I):I-89-I-101.
Kirklin JW, Barratt-Boyes BG. Cardiac Surgery, 2nd ed. New York, NY: Churchill Livingstone; 1993:498–510.
Marchenese K, Schenk EA. Atrioventricular conductive system lesions following cardiac valve replacement. Circulation. 1972;45–46(suppl. II):II-188. Abstract.
Lehman MH, Denker S, Mahmud R, Addas A, Akhtar M. Linking: a dynamic electrophysiologic phenomenon in macro-reentry circuits. Circulation. 1985;71:254–265.