Mechanism of Repetitive Monomorphic Ventricular Tachycardia
Background The most common form of idiopathic ventricular tachycardia (VT) is repetitive monomorphic VT (RMVT), which is characterized by frequent ventricular ectopy and salvos of nonsustained VT with intervening sinus rhythm. Unlike most other forms of idiopathic VT, this tachycardia typically occurs at rest and is nonsustained. The mechanism of RMVT is undefined. Because of a common site of origin, the right ventricular outflow tract (RVOT), we hypothesized that RMVT is mechanistically related to paroxysmal sustained, exercise-induced VT, which has been shown to be consistent with cAMP-mediated triggered activity. Therefore, in this study, we sought to identify (1) the mechanism of RMVT at the cellular level by using electropharmacological probes known to activate either stimulatory or inhibitory G proteins and thereby modify intracellular cAMP levels, (2) potential autonomic triggers of RMVT through analysis of heart rate variability, and (3) whether well-characterized somatic activating mutations in the stimulatory G protein, Gαs, underlie RMVT.
Methods and Results Twelve patients with RMVT underwent electrophysiological study. Sustained monomorphic VT was reproducibly initiated and terminated with programmed stimulation and/or isoproterenol infusion in 11 of the 12 patients (the other patient had incessant RMVT). Induction of VT demonstrated cycle length dependence and was facilitated by rapid atrial or ventricular pacing. Termination of VT occurred in response to interventions that either lowered stimulated levels of intracellular cAMP (and thus decreased intracellular Ca2+)—ie, adenosine (12 of 12), vagal maneuvers or edrophonium (8 of 9), and β-blockade (3 of 5)—or directly decreased the slow-inward calcium current—ie, verapamil (10 of 12). Analysis of heart rate variability during 24-hour ambulatory monitoring in 7 patients showed that the sinus heart rate is increased and accelerates before nonsustained VT (P<.05), whereas high-frequency heart rate variability is unchanged. These findings are consistent with transient increases in sympathetic tone preceding nonsustained VT. Finally, myocardial biopsy samples were obtained from the site of origin of the VT (typically the RVOT) and from the right ventricular apex from 9 patients. Genomic DNA was extracted from each biopsy sample, and three exons of Gαs in which activating mutations have previously been described were amplified by polymerase chain reaction. All sequences from these regions were found to be identical to that of control.
Conclusions Although the arrhythmia occurs at rest, the constellation of findings in idiopathic VT that is characterized by RMVT is consistent with the mechanism of cAMP-mediated triggered activity. Therefore, the spectrum of VT resulting from this mechanism includes not only paroxysmal exercise-induced VT but also RMVT.
Idiopathic ventricular tachycardia (VT) is a generic term that describes various forms of VT that occur in patients without structural heart disease or metabolic or electrolyte abnormalities or in the absence of the long QT syndrome. Several distinct mechanistic entities have been identified in the last decade. These include fascicular tachycardia, which typically originates in the region of the left posterior fascicle, shows verapamil sensitivity, and is dependent on a reentrant mechanism.1 2 Another variety, automatic VT, may originate from either ventricle, can be neither initiated nor terminated with programmed stimulation, is facilitated by catecholamine stimulation, and is responsive to β-blockade.3 A third form, paroxysmal exercise-induced VT, typically originates from the right ventricular outflow tract (RVOT) and presents with a left bundle branch block (LBBB), inferior axis morphology, and a sustained rhythm. This tachycardia is mediated by intracellular calcium overload and terminates with β-blockers, verapamil, vagal maneuvers, and adenosine and is thought to be due to cAMP-mediated triggered activity.4 5
Another type of idiopathic VT, perhaps the most prevalent form, was originally described by Gallavardin6 and has become known by its ECG descriptor, repetitive monomorphic VT (RMVT). It is characterized by frequent ventricular extrasystoles, ventricular couplets, and salvos of nonsustained VT with intervening sinus rhythm. In contrast to most other forms of idiopathic VT, this tachycardia usually occurs at rest and is nonsustained. Despite extensive investigation, the mechanism of RMVT has defied clear characterization. Various mechanisms have been proposed, including enhanced automaticity,7 abnormal automaticity,8 reentry,9 and triggered activity caused by early afterdepolarizations.10 11
The purpose of this study was to characterize the mechanism of RMVT with specific electropharmacological probes that mediate their myocardial effects by altering the levels of intracellular calcium and cAMP. This approach was based on the hypothesis that RMVT is mechanistically related to paroxysmal sustained, exercise-induced idiopathic VT, which is dependent on cAMP-mediated triggered activity.4 5 Furthermore, because of the absence of a structural basis for RMVT and its dependence on cAMP stimulation, we sought to identify a putative molecular defect in the β-adrenergic receptor–cAMP signal transduction pathway. We examined the α-subunit of the stimulatory guanine nucleotide binding protein, Gαs, because it is known to stimulate adenylyl cyclase activity and increase calcium channel permeability12 both in vitro and in vivo and because activating mutations in Gαs have been implicated in a variety of human disease states, including endocrine and nonendocrine tumors.13 14 Biopsy samples were obtained from the region identified by electrophysiological mapping as the site of arrhythmogenic origin and examined for the presence of possible activating mutations in the GTPase regulatory domains contained in exons 2, 8, and 9 of the Gαs gene.15 16 17
The cellular mechanism of cAMP-mediated triggered activity is related to intracellular calcium overload and delayed afterdepolarizations (DADs). Agonist binding (either endogenous or exogenous catecholamines) to the β-receptor activates adenylyl cyclase through the stimulatory G protein, Gs. ATP is thus converted to cAMP, and through subsequent phosphorylation of protein kinase A, conductance of the slow-inward channel to calcium is increased, leading to intracellular calcium overload and an oscillatory release of calcium from the sarcoplasmic reticulum. This in turn gives rise to a transient inward current, ITi, and DADs.18 19 Fig 1⇓ summarizes this cascade of events. This schema also provides the rationale for the receptor-specific effects of the pharmacological probes used in this study. The net effect of these probes is to decrease intracellular calcium overload. β-Blockers will attenuate the facilitatory effects of a β-agonist on intracellular cAMP by competitive receptor antagonism. Adenosine lowers stimulated levels of cAMP through a pertussis toxin–sensitive inhibitory guanine nucleotide binding protein, Gi.20 Potentiation of endogenous acetylcholine through vagal maneuvers (ie, Valsalva, carotid sinus pressure) or edrophonium also lowers cAMP by a similar mechanism. Finally, blockade of the slow-inward calcium current with verapamil directly lowers intracellular calcium.18
The study group was made up of 12 patients with RMVT (Table 1⇓). The 7 men and 5 women had a mean age (±SD) of 52±21 years. All patients had evidence of recurrent nonsustained VT; 4 patients also had one documented episode of sustained VT. VT occurred primarily at rest in all patients, although exercise-induced VT was also demonstrated in 3 patients. Two patients had syncope associated with VT; 10 had palpitations and/or dizziness. The duration of symptoms ranged from 1 month to 10 years. Nine patients underwent cardiac catheterization to evaluate right and left ventricular function and coronary anatomy. The remaining 3 patients were evaluated by two-dimensional echocardiography and exercise stress testing. Nine patients had no evidence of structural heart disease, whereas 1 patient had significant three-vessel coronary artery disease (occlusion >70%) and an ejection fraction of 25% and 2 others had an idiopathic dilated cardiomyopathy with an ejection fraction of 40%. Nine patients had normal ECGs. One patient had evidence of left ventricular hypertrophy; 1 had T-wave inversions in leads II, III, and aVF; and 1(without angiographic or echocardiographic evidence of arrhythmogenic right ventricular dysplasia) demonstrated inverted T waves in leads V1 through V3 (patient 3, Table 1⇓). Response to an exercise stress test was evaluated in 6 patients. Sustained monomorphic VT was induced in 2 patients and nonsustained VT (up to 20 beats) in 1 patient; there was complete suppression of RMVT in 1 patient, and exercise testing had no effect on ventricular ectopy in 2 patients.
Electrophysiological studies were performed with patients in the unsedated, postabsorptive state after informed consent had been obtained. All antiarrhythmic agents were discontinued for at least five half-lives before evaluation. Three quadripolar electrode catheters were inserted percutaneously and advanced under fluoroscopic guidance to the high right atrium, right ventricular apex, and AV junction for recording of the His-bundle electrogram. Bipolar intracardiac recordings were filtered at 40 to 400 Hz and were displayed simultaneously with three surface ECG leads on a multichannel oscilloscope. Data were stored on magnetic tape and later recorded on photographic paper for illustrative purposes. Systemic arterial pressure was monitored continuously (Dinamap, Critikon). Stimulation was performed with a programmable stimulator and an isolated constant current source (Bloom Associates). Stimuli were delivered as rectangular pulses of 2-ms duration at four times the diastolic threshold.
The stimulation protocol included the introduction of single, double, and triple extrastimuli during several paced cycle lengths from the high right atrium, right ventricular apex, and RVOT. In addition, all patients underwent sequential atrial and ventricular burst pacing for 15 to 30 beats, beginning at a cycle length of 500 ms and decreasing in 10-ms steps to 250 ms (or to pacing-induced AV nodal Wenckebach in the case of atrial pacing). The ability of isoproterenol (2 to 10 μg/min) to initiate VT alone, during its washout phase, and during concurrent pacing was also evaluated. In selected patients in whom induction of sustained VT was nonreproducible, facilitation of induction was assessed during concurrent infusion of isoproterenol and bolus doses of aminophylline (2.8 mg/kg IV), an adenosine receptor antagonist, and atropine (0.03 mg/kg IV), a muscarinic cholinergic receptor blocker. The ability to entrain each patient’s tachycardia according to previously described criteria was systematically evaluated from the right ventricular apex and RVOT at multiple paced cycle lengths.21 Attempts at entrainment were initiated at cycle lengths 20 ms shorter than the VT cycle length for a duration of 15 beats. The pacing cycle length was then decreased in 10-ms intervals until either tachycardia terminated or a pacing cycle length of 200 ms was achieved.
To reproducibly assess the effect of pharmacological probes on VT, inclusion in the study required that patients with clinical RMVT have either inducible sustained VT (11 patients) or incessant RMVT (1 patient). Sustained VT was defined as tachycardia lasting >30 seconds. VT sensitivity to the following agents was evaluated: (1) adenosine 100 to 475 μg/kg IV given as a rapid bolus followed by a saline flush, (2) edrophonium 1 mg IV initially given over 15 seconds with an additional 9 mg given 45 seconds later, (3) esmolol 500 μg/kg IV given over 1 minute followed by 50 μg · kg−1 · min−1 for 4 minutes, and (4) verapamil 10 to 15 mg IV infused over 60 seconds.
Ambulatory ECG and Analysis of Heart Rate Variability
In 7 patients, 24-hour ambulatory ECG recordings of bipolar leads CM1 and CM5 were obtained with patients in the drug-free state during unrestricted sedentary physical activity. The recordings were scanned and digitized with a computer-based system (Marquette Laser XP, Marquette Electronics Inc) with visual verification and correction of beat morphology and timing by one of the authors, and the RR interval and beat-type listing were then transferred to an IBM-compatible personal computer for further analysis. Beats were classified as a beat in normal sinus rhythm (N); a ventricular premature contraction (V); or other (O, eg, an atrial premature beat, a junctional or ventricular escape beat, or uninterpretable artifact). By use of the entire recording period, the following measures of mean RR interval and heart rate variability were then computed: (1) the mean sinus RR interval for the two beats (NN) immediately preceding another sinus beat (NN-N), a single premature ventricular contraction (PVC) (NN-V1), a ventricular couplet (NN-V2), or nonsustained or sustained ventricular tachycardia (NN-V3+); (2) the mean sinus RR interval for the two beats (NN) occurring 15 intervals before a sinus beat, a single PVC, a ventricular couplet, or nonsustained or sustained ventricular tachycardia; and (3) the root-mean-square of successive differences between the sinus RR intervals included in the 15 intervals preceding a sinus beat (RMSSD15-N), a single PVC (RMSSD15-V), a ventricular couplet (RMSSD15-V2), or nonsustained or sustained ventricular tachycardia (RMSSD15-V3+). For these computations, only NN intervals were analyzed; intervals before and after PVCs (NV, VN) or other beats (NO, ON) were excluded from analysis. The mean RR interval was determined by the combined influences of the sympathetic and parasympathetic nervous systems on the sinoatrial node. In contrast, RMSSD is a measure of high-frequency heart rate variability,22 23 and therefore is predominantly affected by cardiac parasympathetic activity.24 25
Biopsy Samples and Preparation of Genomic DNA
Percutaneous endocardial biopsies were obtained with a bioptome from the right ventricle in 9 patients after informed consent was obtained for a protocol approved by the Human Investigations Committee (Table 1⇑). In each patient, one to three biopsy samples were obtained from (or near) the site of origin of VT (the RVOT), and one to three samples were obtained from a remote site, usually the right ventricular apex. After excision, one or two samples were processed for routine histological examination, and the remaining biopsy specimens were frozen rapidly and stored for molecular analysis. To prepare genomic DNA, biopsies were thawed in 100-μL lysis buffer (150 mmol/L NaCl, 10 mmol/L Tris · Cl [pH 8.3], 10 mmol/L EDTA, and 0.4% sodium dodecyl sulfate) containing 0.1 mg/mL proteinase K and incubated for 12 to 24 hours at 50°C. Proteinase K was heat-inactivated at 98°C for 3 minutes. DNA was extracted with phenol and chloroform, ethanol-precipitated, and then dissolved in 50 μL of 10 mmol/L Tris · Cl (pH 8.0), 1 mmol/L EDTA. Samples were diluted 1:10 in 1× polymerase chain reaction (PCR) buffer (Perkin-Elmer Cetus).
Polymerase Chain Reaction
Genomic DNA prepared from biopsy samples was used to amplify the coding region of exons 8 and 9 and exon 2 of Gαs by a nested PCR procedure. For amplification of exons 8 and 9, the oligonucleotide primers used in the first reaction were 5′-GCGCTGTGAACACCCCACGTGTCT-3′ (sense) and 5′-CTTGTTACAATTACGTTTCACT-3′ (antisense). The products of the first PCR were then used as a template in the second reaction. The pair of nested primers was 5′-GTGATCAAAGGCTGACTATGTG-3′ (sense) and 5′-GCTGCTGGCCACCACGAAGATGA-3′ (antisense). Similarly, exon 2 of Gαs was amplified with external primers 5′-ACAACAGCAGACCTCCCTGC-3′ (sense) and 5′-CCCACCTATACTTCCTAAAGG-3′ (antisense) and internal primers 5′-GTTAAAATGCCTCCTCATA-3′ (sense) and 5′CTGCACATTTGACA CTTA-3′ (antisense).
For PCR, the total reaction volume was 50 μL, containing 5 μL 10× PCR reaction buffer, 2 μL deoxynucleotide triphosphates (20 mmol/L), 3 μL each primer (10 pmol/L), 3 μL genomic DNA (1:10 dilution), 1 U Taq polymerase (Perkin-Elmer Cetus), and sterile water. The PCR amplification was performed for 35 cycles consisting of a 1-minute denaturation step at 94°C, a 1-minute annealing step at 56°C, and a 1-minute extension step at 72°C. A 526–base pair (bp) DNA fragment was amplified by PCR containing exons 8 and 9, while a 200-bp fragment was amplified by PCR containing exon 2. DNA was visualized on 1% agarose gel with ethidium bromide staining, and size-selected PCR products were electroeluted (Micro-Electroeluter model 30, Amicon). DNA was then precipitated with ethanol and used for direct sequencing.
Direct Sequencing of Amplified DNA
Amplified DNA was directly sequenced by an improved method using dimethyl sulfoxide.26 After the annealing reaction, DNA was sequenced by the dideoxy chain termination procedure with a Sequenase version 2.0 DNA Sequencing Kit (United States Biochemical).
Mapping and Ablation
Endocardial ventricular activation mapping during tachycardia and pace mapping during sinus rhythm were performed at ≥15 sites to localize the site of VT origin. Mapping and ablation were performed with a deflectable quadripolar catheter with a 4-mm distal tip electrode (2-mm interelectrode spacing). The location of the catheter was identified with biplane fluoroscopy. Pace mapping was performed during sinus rhythm by use of the distal bipolar pair of electrodes with a stimulus strength of 2 mA and pulse duration of 2 ms. The pacing cycle length was identical to that of the VT cycle length. Ablation was performed at sites that produced 12-lead surface ECG pace maps that showed close concordance with respect to QRS polarity and morphology in all 12 leads. These sites correlated to local activation times, with discrete electrograms occurring 10 to 40 ms before the onset of the surface QRS.
After localization of the ablation site, radiofrequency energy (550 kHz) was delivered during VT from the distal 4-mm electrode of the mapping catheter to a posterior chest wall patch positioned in the left infrascapular region (30 W for 10 to 60 seconds or until an abrupt rise in catheter impedance). After each application of radiofrequency energy, induction of VT was assessed with programmed stimulation and isoproterenol. If VT was inducible, further mapping in a contiguous region was performed and ablation repeated. Successful ablation was defined by the absence of spontaneous or inducible VT 30 minutes after the procedure.
Results are presented as mean±SD where appropriate. Within-group comparisons of variables were performed by ANOVA (Crunch 4.0, Crunch Software Corporation) by use of Dunnett’s correction for multiple statistical comparisons. For all comparisons, a probability value of <.05 was required to reject the null hypothesis.
In general, induction of sustained VT was challenging and required greater effort than for patients with sustained reentrant VT caused by coronary artery disease. A successful induction method often was inconsistent in its reproducibility, mandating multiple different induction methods in the same patient, ie, extrastimuli versus burst pacing versus catecholamine stimulation. Except for patient 10, all patients had sustained VT reproducibly induced with programmed stimulation, particularly with rapid ventricular pacing (Table 2⇓). In each patient (except patient 10), sustained VT was induced at least 10 separate times.
An induction window of 30 to 100 ms was observed during rapid ventricular pacing in 10 of 12 patients (range, 250 to 530 ms). Pacing at a cycle length above or below this window was ineffective in inducing VT. In general, shorter paced cycle lengths (<400 ms) were more effective in inducing VT than longer cycle lengths. Seven patients required concurrent infusion of isoproterenol. Two patients were also induced with rapid atrial pacing. VT was induced in 2 patients with ventricular extrastimuli after an 8-beat priming drive; 1 of these patients also required concurrent infusion of isoproterenol. In 3 patients, VT became noninducible after several relatively facile inductions of VT with either rapid ventricular pacing or isoproterenol infusion. Subsequent induction of VT in these patients was facilitated only by concurrent infusion of isoproterenol and bolus doses of aminophylline and atropine, the net effect of which was to increase intracellular cAMP by attenuating the inhibitory effects of endogenous adenosine and acetylcholine on cAMP while potentiating the facilitatory effects of β-adrenergic stimulation.
The induced tachycardia cycle length was 323±80 ms (range,240 to 510 ms). Once sustained, the tachycardia continued unabated until terminated by programmed stimulation or pharmacological intervention. Although RMVT most often originated from the RVOT based on activation, pace mapping, and the results of radiofrequency catheter ablation, the morphology of induced VT showed considerable variability. In the 9 patients who had a single VT morphology induced, the configuration was identical to that of spontaneous RMVT. In 5 patients, VT configuration and mapping results were consistent with an anterolateral RVOT origin (LBBB, left inferior axis); 3 patients had an LBBB, right inferior axis configuration, and localization to an anteroseptal RVOT site; in 1 patient (patient 5), VT was localized to the left superior interventricular septum. Three patients had two monomorphic configurations with identical frontal plane axes. In each case, the patient had a right bundle branch block (RBBB) and an LBBB VT, suggesting a common septal origin with exit sites to the left and right of the septum, respectively (Fig 2⇓). Alternatively, two separate sites of origin could account for these findings.
VT was terminated in all patients with rapid ventricular pacing except for patient 10, in whom VT termination could not be reliably assessed owing to the incessant nature of the tachycardia. In no patient was overdrive suppression or acceleration observed. Entrainment of tachycardia during incremental ventricular pacing from the right ventricular apex and RVOT could not be demonstrated in any patient.
Pharmacological and Autonomic Assessment
All pharmacological and autonomic evaluations (Table 2⇑) were made during sustained induced VT except for patient 10, in whom the incessant nature of RMVT allowed reliable assessment. Adenosine reproducibly terminated VT in all 12 patients (Fig 3⇓). The mean dose was 177±51 μg/kg, and tachycardia terminated within 8 to 15 seconds. Termination in all cases was abrupt and was not preceded by VT slowing or ventricular extrasystoles. Vagal maneuvers such as carotid sinus pressure, Valsalva, or administration of edrophonium terminated VT in 8 of 9 patients. Similar to the effects seen with adenosine, termination was abrupt in all patients (Fig 4⇓).
VT was sensitive to verapamil in 10 of the 12 patients. Termination typically occurred 60 to 120 seconds after administration of the drug (Fig 5⇓). Verapamil precluded reinduction of VT for at least 1 hour, unlike adenosine, vagal maneuvers, or edrophonium.
Intravenous β-blockade was successful in terminating induced VT in 3 of 5 patients and in preventing reinduction. Particularly instructive were the effects observed in patient 12, in whom VT occurred only at rest. VT was induced with rapid ventricular pacing alone and did not require isoproterenol infusion. Despite the absence of exogenous catecholamine stimulation, esmolol was effective in terminating tachycardia and preventing its reinduction (Fig 6⇓). This particular example suggests a dependence of RMVT on β-receptor activation, despite the absence of a clear β-adrenergic stimulus.
Heart Rate Variability
Among the patients with RMVT, the mean numbers of evaluable sinus beats, single PVCs, ventricular couplets, and runs of nonsustained VT during ambulatory ECG recording were 71 160±33 468, 3211±4193, 314±554, and 535±1182, respectively. The mean RR intervals of the sinus beats immediately preceding other sinus beats and the sinus beats occurring 15 beats before other sinus beats were significantly longer than the corresponding intervals preceding ventricular couplets and runs of nonsustained VT (overall, P=.002 for the interval immediately preceding an event and P=.004 for the interval 15 beats before an event); the RR intervals preceding single PVCs were intermediate (Fig 7A⇓). In contrast, there were no significant alterations in high-frequency heart rate variability (RMSSD, P=.22) preceding ectopic beats (Fig 7B⇓).
There was a significant acceleration in heart rate (decrease in RR interval) when the 15th interval preceding a ventricular couplet was compared with the interval immediately preceding the couplet (697±141 versus 679±144 ms, P=.02) and for the corresponding intervals preceding nonsustained VT (752±131 versus 743±136 ms, P=.04). There was also an acceleration preceding single PVCs that was not statistically significant (793±100 versus 766±99 ms, P=.08).
Polymerase Chain Reaction
Myocardial biopsies were obtained from the RVOT and right ventricular apex in 9 patients. Histologically, all samples were normal. Genomic DNA was extracted from each biopsy sample, and selected regions of the Gαs gene were amplified with PCR. A nested primer strategy was used to amplify a 526-bp fragment for exons 8 and 9 and a 200-bp fragment containing exon 2. These regions of the α-subunit have been shown to maintain an important regulatory function for Gαs. PCR products were purified, fully sequenced, and examined for mutations in each of the three exons. Of the 27 regions analyzed, all sequences for these regions were found to be identical to that of control.
Radiofrequency ablation of RMVT (Table 3⇓) was performed in 9 patients (3 other patients chose medical therapy with verapamil instead). Ablation was performed after myocardial biopsy. Although biopsies were obtained in the region of the site or origin of VT (within approximately 1 to 2 cm), VT remained inducible after biopsy. Eight patients had immediate success as defined by termination of their arrhythmia, an inability to reinduce tachycardia, and abolition of spontaneous ventricular ectopy. The mean number of radiofrequency applications was 9.6±7.8. The site of ablation was the anteroseptal region of the RVOT in 3 patients, the anterolateral aspect of the RVOT in 5 patients, and the superior aspect of the left interventricular septum in 1 patient. Pace mapping proved more precise than activation mapping in identifying an appropriate ablation site, although all successful sites were associated with discrete presystolic activation. In the 1 patient (patient 5) who was ablated from the superior aspect of the left side of the interventricular septum, a pace map from this site was closely concordant with induced sustained VT. Application of radiofrequency energy at this site successfully ablated VT, but the complete therapeutic effect required 27 seconds (Fig 8⇓). In 1 patient (patient 1) in whom both an RBBB and LBBB VT were induced (Fig 2⇑), ablation was unsuccessful from catheter sites along the right and left sides of the septum. However, application of radiofrequency energy between two catheters placed directly across from each other on either side of the septum resulted in success (cathode–RV catheter; anode–LV catheter). The 1 patient who was not successfully ablated had a concordant pace map produced at a site immediately inferior to the pulmonic valve. Because of concern regarding damage to the valve, ablation at this site was not performed, and ablative attempts several millimeters away were unsuccessful.
Unlike patients with accessory pathways where a successful site usually results in immediate abolition of pathway conduction, 10 to 15 seconds of radiofrequency energy application was usually required to eliminate VT. The only immediate complication of the procedure was the appearance of a new RBBB pattern in 1 patient. One patient developed a single episode of self-terminating VT 3 days after discharge. This episode spontaneously remitted, and there has been no recurrence in 19 months. During a mean follow-up period of 15.7±7.7 months (range, 4 to 32 months), all patients who were successfully ablated were on no antiarrhythmic therapy and had no clinical recurrence of VT.
The primary finding of this study is that the mechanism of RMVT is consistent with cAMP-mediated triggered activity due to DADs, a mechanism of ventricular arrhythmia previously demonstrated only for paroxysmal idiopathic exercise-induced VT. This conclusion is based on the cycle length–dependent characteristics of VT initiation and the response of VT to electropharmacological probes that decrease intracellular calcium either by direct effects on the L-type calcium current or through inhibitory effects on cAMP.
Although patients with RMVT share some common characteristics with those patients we described previously with paroxysmal idiopathic VT caused by cAMP-mediated triggered activity,4 5 there are important differences. In general, both groups typically have an absence of structural heart disease and have morphologically similar forms of VT that originate from the region of the RVOT. The methods of induction of VT and response to mechanism-specific pharmacological probes are also identical. However, RMVT is an arrhythmia that typically occurs at rest, is nonsustained, and can be incessant. In contrast, paroxysmal idiopathic VT is usually associated with exercise or stress and is sustained. Paroxysmal idiopathic exercise-induced VT is also relatively easier to induce during provocative testing. To increase the yield for induction in patients with RMVT and to convert nonsustained VT to sustained VT, we found that attenuating the inhibitory effects of endogenous adenosine and acetylcholine on cAMP production was useful in some patients. This was achieved by simultaneous infusion of aminophylline, an adenosine receptor antagonist (which at the dose used in this study has minimal effect on phosphodiesterase inhibition),27 and atropine in addition to isoproterenol during rapid burst pacing.
Triggered activity related to catecholamine stimulation is dependent on activation of cAMP. It was shown previously on both cellular and clinical levels that termination of VT in response to adenosine is strongly suggestive of a cAMP-triggered mechanism because adenosine mediates its ventricular effects through activation of Gi and inhibition of adenylyl cyclase, thus lowering stimulated levels of intracellular cAMP.4 5 28 29 30 This conclusion is based on several lines of evidence: (1) the suppressive effects of adenosine on ITi, DADs, and triggered activity are attenuated by pertussis toxin, which inactivates Gi through ADP ribosylation28 ; (2) adenosine abolishes the effects of the adenylyl cyclase activator forskolin on ITi and DADs but not those of dibutyryl cAMP28 ; and (3) the effects of adenosine on triggered activity are mechanism-specific because it fails to abolish DADs and triggered activity caused by non-cAMP mechanisms, ie, inhibition of Na+,K+-ATPase.28 Moreover, adenosine does not terminate triggered activity related to early afterdepolarizations, catecholamine- mediated automatic rhythms, or reentrant ventricular arrhythmias, regardless of whether they are dependent on catecholamine stimulation.4 5 28
The effects of programmed stimulation, although less specific than those of adenosine, also support the diagnosis of triggered activity. In cellular preparations, a window of pacing cycle lengths that induce triggered activity are observed.31 The amplitude of DADs diminishes at cycle lengths above and below the induction window. A window of induction was also observed in our study. Furthermore, in contrast to typical induction of reentry with programmed extrastimuli, induction of VT in the present study was most often initiated with rapid burst pacing from either the atrium or the ventricle. Entrainment (continuous resetting) of VT is considered a specific finding for reentry. Although a negative finding does not rule out reentry, entrainment could not be demonstrated in any patients. An automatic mechanism was ruled out because automatic arrhythmias cannot be initiated with programmed stimulation.
Triggered activity is typically induced from Purkinje fibers. More recent data suggest that triggered rhythms may also arise from M cells, cells localized in the deep subepicardial and midmyocardial levels of the ventricle.32 33 These cells have electrophysiological characteristics intermediate between those of myocardial and Purkinje cells. Like myocardial cells, they do not exhibit phase 4 depolarization, and like Purkinje cells, their action potential duration is very sensitive to rate. Our data are consistent with VT possibly arising from such M cells. Three patients had two different morphologies of VT originating from a single site of origin from within the interventricular septum. This finding is compatible with a common midmyocardial origin of VT that can exit from the left or right side of the septum. In 1 of these patients, successful radiofrequency catheter ablation could be achieved only when energy was delivered between catheters on the right and left sides of the interventricular septum, again suggesting a deep subepicardial or midmyocardial site of origin of VT. Our data also show that RMVT typically originates from a single focus, similar to most other forms of idiopathic right ventricular VT.
In the paroxysmal form of VT resulting from cAMP-mediated triggered activity, the provocative stimulus for initiating the rhythm is readily identifiable, ie, adrenergic stimulation in the form of exercise, stress, or exogenous catecholamine infusion. In RMVT, the source of β-adrenergic stimulation is not apparent because the arrhythmia clinically occurs at rest and is often induced in the absence of concurrent catecholamine infusion. Despite these circumstances, however, the arrhythmia appears to be cAMP-mediated because it terminates in response to perturbations that lower stimulated levels of cAMP, ie, adenosine, edrophonium, and β-blockade. Several explanations are possible for this apparent paradox. One contributing factor may be the presence of a constitutively active signal in the β-adrenergic receptor–cAMP signal transduction pathway. Recent evidence indicates that activating mutations in the cardiac β-receptor can result in agonist-independent stimulation of adenylyl cyclase, increased conductance of the L-type calcium current, and increased binding affinity of the agonist for the receptor. Furthermore, these effects can be inhibited by β-blockade34 35 36 (see Fig 6⇑).
In the present study, we focused on possible G-protein activating mutations. Activating mutations in the Gαs subunit inhibit GTPase activity, thus trapping the protein in the active, GTP-bound state.15 16 17 Such mutations have been well characterized in a variety of human endocrine and nonendocrine tumors.13 14
To test our hypothesis, we screened biopsy samples from the RVOT in patients with RMVT for a putative signaling defect in Gαs. Regions of genomic DNA encoding Arg201, Gln227, and Gly49 were amplified by PCR. However, no mutations were found, and a wild-type nucleotide sequence was confirmed for each sample.
In considering the negative results of this G-protein analysis, it should be appreciated that only a small subpopulation of affected individuals may have any given type of mutation. For example, in studies of human tumors, only 18 of 42 patients (43%) with growth hormone–producing pituitary adenomas contained mutations in Gαs, whereas for carcinoma of the thyroid, the incidence was significantly lower (4%).14 Therefore, it is possible that other patients with RMVT may contain these specific mutations. Further, it is also conceivable that we have not sampled from the source of VT in these patients. Unlike a tumor focus, which is morphologically identifiable, cardiac tissue from patients with RMVT is morphologically normal. We did attempt, however, to obtain biopsy material from the site of origin of VT identified during electrophysiological mapping. Nevertheless, a somatic mutation underlying the unique electrophysiological properties in these patients may be contained outside the biopsy sites. In addition, it is important to recognize that other regions of the α-subunit such as the GTP-binding domain, the site of β-γ complex formation, or the region involved in effector activation may be involved in Gαs misregulation.37 38 39 Finally, defects may be localized to other G proteins or perhaps to other proteins in the signal transduction cascade. Further investigation is being directed at these possibilities.
A contributory (although incomplete) explanation for the cAMP-mediated characteristics of RMVT in the absence of an identifiable β-adrenergic stimulus can be inferred from our analysis of heart rate variability. These data suggest the presence of transient increases in subclinical sympathetic tone (unrelated to exertion) that precede ectopic activity in these patients. Heart rate accelerated before ventricular ectopy in patients with RMVT, particularly before ventricular couplets and nonsustained VT,10 but high-frequency heart rate variability (RMSSD), which is a reflection of cardiac parasympathetic tone,24 25 remained unchanged. Because sinus cycle length depends on the combined effects of cardiac sympathetic and parasympathetic tones, RR acceleration in these patients is therefore consistent with an increase in sympathetic tone and not to parasympathetic withdrawal.
In summary, we have presented evidence that the cellular mechanism for the most common form of idiopathic VT, RMVT, is consistent with cAMP-mediated triggered activity. Incorporation of the mechanism-specific electropharmacological approach used in this study into the standard electrophysiological evaluation of patients with idiopathic VT should facilitate identification of this form of VT.
This work was supported in part by grants from the NIH (RO1 44747) and the Michael Wolk Heart Foundation. Dr Lerman is an Established Investigator of the American Heart Association.
- Received September 27, 1994.
- Revision received December 19, 1994.
- Accepted January 9, 1995.
- Copyright © 1995 by American Heart Association
Belhassen B, Rotmensch HH, Laniado S. Response of recurrent sustained ventricular tachycardia to verapamil. Br Heart J. 1981;46:679-682.
Lerman BB, Belardinelli L, West GA, Berne RM, DiMarco JP. Adenosine-sensitive ventricular tachycardia: evidence suggesting cyclic AMP–mediated triggered activity. Circulation. 1986;74:270-280.
Lerman BB. Response of nonreentrant catecholamine-mediated ventricular tachycardia to endogenous adenosine and acetylcholine: evidence for myocardial receptor-mediated effects. Circulation. 1993;87:382-390.
Gallavardin L. Extrasystolie ventriculaire a paroxysmes tachycardiques prolonges. Arch Mal Coeur Vaiss. 1922;15:298-306.
Martins JB, Constanin L, Kienzle MG, Brownstein SL, Hopson JR. Mechanisms of ventricular tachycardia unassociated with coronary artery disease. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. Philadelphia, Pa: WB Saunders; 1990:581-589.
Buxton AE, Waxman HL, Marchlinski FE, Simson MB, Cassidy D, Josephson ME. Right ventricular tachycardia: clinical and electrophysiologic characteristics. Circulation. 1983;68:917-927.
Coumel P. Early afterdepolarizations and triggered activity in clinical arrhythmias. In: Rosen MR, Janse MJ, Wit AL, eds. Cardiac Electrophysiology: A Textbook. Mount Kisco, NY: Futura; 1990:387-411.
Lyons J, Landis CA, Harsh G, Vallar L, Grunewald K, Feichtinger H, Duh QY, Clark OH, Kawasaki E, Bourne HR, McCormick F. Two G protein oncogenes in human endocrine tumors. Science. 1990;249:655-659.
Masters SB, Miller RT, Chi MH, Chang FH, Beiderman B, Lopez NG, Bourne HR. Mutations in the GTP-binding site of Gs alter stimulation of adenylyl cyclase. J Biol Chem. 1989;264:15467-15474.
Graziano MP, Gilman AG. Synthesis in Escherichia coli of GTPase-deficient mutants of Gsα. J Biol Chem. 1989;264:15475-15482.
Wit AL, Rosen MR. Afterdepolarizations and triggered activity. In: Fozzard HA, Haber E, Jenning RB, Katz AM, Morgan E, eds. The Heart and Cardiovascular System. New York, NY: Raven Press Publishers; 1986:1449-1490.
Lerman BB, Belardinelli L. Cardiac electrophysiology of adenosine: basic and clinical concepts. Circulation. 1991;83:1499-1509.
Stein KM, Borer JS, Hochreiter C, Okin PM, Herrold EM, Devereux RB, Kligfield P. Prognostic value and physiologic correlates of heart rate variability in chronic severe mitral regurgitation. Circulation. 1993;88:127-135.
Akselrod S, Gordon D, Ubel FA, Shannon DC, Barger AC, Cohen RJ. Power spectrum analysis of heart rate fluctuation: a quantitative probe of beat-to-beat cardiovascular control. Science. 1981;213:220-223.
Pomerantz B, Macaulay RJB, Caudill MA, Kutz I, Adam D, Gordon D, Kilborn KM, Barger AC, Shannon DC, Cohen RJ, Benson H. Assessment of autonomic function in humans by heart rate spectral analysis. Am J Physiol. 1985;248:H151-H153.
Seto D. An improved method for sequencing double stranded plasmid DNA from minipreps using DMSO and modified template preparation. Nucleic Acids Res. 1990;18:5905-5906.
Fredholm B. Are methylxanthine effects due to antagonism of endogenous adenosine? Trends Pharmacol Sci. 1980;1:129-132.
Song Y, Thedford S, Lerman BB, Belardinelli L. Adenosine-sensitive afterdepolarizations and triggered activity in guinea pig ventricular myocytes. Circ Res. 1992;70:743-753.
Belardinelli L, Isenberg G. Actions of adenosine and isoproterenol on isolated mammalian ventricular myocytes. Circ Res. 1983;53:287-297.
Lerman BB, Wesley RC, DiMarco JP, Haines DE, Belardinelli L. Antiadrenergic effects of adenosine on His-Purkinje automaticity: evidence for accentuated antagonism. J Clin Invest. 1988;82:2127-2135.
Sicouri S, Antzelevitch C. A subpopulation of cells with unique electrophysiological properties in the deep subepicardium of the canine ventricle: the M cell. Circ Res. 1991;68:1729-1741.
Costa A, Ogino Y, Munson PJ, Onaran HO, Rodbard D. Drug efficacy at guanine nucleotide-binding regulatory protein-linked receptors: thermodynamic interpretation of negative antagonism and of receptor activity in the absence of ligand. Mol Pharmacol. 1992;41:549-560.
Mewes T, Dutz S, Ravens U, Jakobs KH. Activation of calcium currents in cardiac myocytes by empty β-adrenoceptors. Circulation. 1993;88:2916-2922.
Samama P, Cotecchia S, Costa T, Lefkowitz RJ. A mutation-inducted activated state of the β2-adrenergic receptor: extending the ternary complex model. J Biol Chem. 1993;268:4625-4636.