Internal Atrial Defibrillation in Humans
Improved Efficacy of Biphasic Waveforms and the Importance of Phase Duration
Background The optimal waveform for internal atrial defibrillation (IAD) in humans is unknown. This study tested the effect of waveform duration and phase duration on the efficacy of biphasic waveforms for IAD.
Methods and Results Electrodes were positioned in the right atrial appendage and coronary sinus in 13 patients. In part 1, the atrial defibrillation thresholds (ADFTs) for 5 monophasic waveforms (2, 4, 6, 10, and 20 ms) and 5 symmetrical biphasic waveforms (1/1, 2/2, 3/3, 5/5, and 10/10 ms) were compared in 6 patients. In part 2, the ADFTs for two asymmetrical biphasic waveforms (7.5/2.5 and 2.5/7.5 ms) were compared with those for a symmetrical biphasic waveform (5/5 ms) and a monophasic waveform (10 ms) in 7 patients. In part 1, biphasics with total durations of 4 to 20 ms had significantly lower ADFTs than monophasic waveforms of the same total duration. For a total duration of 2 ms, there was no significant difference in ADFTs between the biphasic and the monophasic waveforms. There was no difference between symmetrical biphasic waveforms of 4 to 20 ms. In part 2, the 7.5/2.5 ms asymmetrical biphasic had significantly lower ADFTs than the three other waveforms tested. Both the 7.5/2.5 ms asymmetrical and the 5/5 ms symmetrical biphasic waveform had significantly lower ADFTs than the 2.5/7.5 ms asymmetrical biphasic and the 10 ms monophasic waveforms.
Conclusions For IAD in humans, biphasic waveforms were more efficacious than monophasic waveforms. This improved efficacy is related to the total duration of the biphasic waveform and each individual phase duration of the biphasic waveform.
Atrial fibrillation is one of the most common arrhythmias encountered in clinical medicine.1 2 3 Therapy includes pharmacological and/or electrical therapy to restore sinus rhythm. Transthoracic electrical cardioversion has been shown to be an effective treatment for atrial fibrillation4 ; however, the relatively high energy shocks (25 to 400 J) that are required can be detrimental to the heart.5 6 7 8 9 10 They also can be painful to the patient, which may result in refusal to undergo repeated cardioversions after unsuccessful attempts to restore sinus rhythm.
Electrical cardioversion of atrial fibrillation has changed little over the past 30 years. However, several groups have shown recently that internal defibrillation of atrial fibrillation in animals and humans is safe as well as effective with the use of transvenous electrodes with certain biphasic waveforms.11 12 13 14 15 However, this method still requires energy levels that are usually painful in nonsedated patients.13 With more efficient defibrillation waveforms, energy requirements would be reduced and discomfort to the patient would be minimized.
For internal ventricular defibrillation in animals and humans, certain biphasic waveforms consisting of two phases of opposite polarity decrease the shock strength required for defibrillation compared with equal duration monophasic waveforms.16 17 18 19 20 21 22 23 24 In animal studies the relative phase durations of these biphasic waveforms have been shown to be an important determinant of ventricular defibrillation efficacy.18 19 25 The effect of total waveform duration and the duration of each phase of a biphasic waveform on internal atrial defibrillation efficacy in humans is unknown. This study investigated the effect of phase duration on the atrial defibrillation efficacy of biphasic waveforms. In part 1, the atrial defibrillation thresholds (ADFT) of multiple monophasic and equal duration biphasic waveforms ranging in total duration of 2 to 20 ms were compared. In part 2, equal duration symmetrical biphasic, asymmetrical biphasic, and monophasic waveforms were compared to assess the importance of phase duration on atrial defibrillation efficacy.
Internal cardioversion was performed in 11 patients undergoing clinically indicated electrophysiological procedures and in 2 patients when external cardioversion was unsuccessful. Informed consent was obtained in all cases, and the study protocol was approved by the Institutional Review Board of Duke University Medical Center. The patients were brought to the clinical electrophysiology laboratory in a fasting, nonsedated state. The patients were then moderately sedated with intravenous midazolam, meperidine, and promethazine. Local anesthesia at the sites used for catheter insertion was achieved by subcutaneous infiltration of a mixture of 0.25% bupivacaine and 2% procaine. A 7F quadripolar catheter was advanced from the right femoral vein to the right ventricular apex for synchronization of the atrial defibrillation shocks to ventricular activation and for postshock pacing if necessary for treatment of bradycardia. A 7F hexapolar catheter was advanced to the low right atrium for induction of atrial fibrillation with rapid atrial burst pacing and bipolar recording of atrial activation.
Two 6F nonapolar catheters (Electro-Catheter Corp) were used as the defibrillation anode and cathode. Each catheter had nine 5-mm electrode rings with 2-mm interelectrode spacing. All nine electrodes were combined for a total surface area of ≈140 mm2. The anode was advanced from the right femoral vein so that the distal tip was positioned in the anteromedial right atrial appendage with the body of the electrodes positioned along the posterolateral right atrium (Fig 1⇓). The cathode was advanced from the right internal jugular or subclavian vein so that the distal tip was positioned as anteriorly as possible in the coronary sinus, usually with the body of the electrodes positioned in the lateral margin of the heart (Fig 1⇓). Once all catheters were in position, an anticoagulation protocol was followed in the clinical electrophysiology laboratory, consisting of 5000 units of intravenous heparin sulfate as an initial bolus followed by 1000 units intravenously every hour during the entire procedure.
Defibrillator and Waveforms
All defibrillation waveforms were delivered from a Ventritex HVS-02 programmable cardioverter/defibrillator (Ventritex Inc), which has a capacitance of 150 μF. This device has two programmable capacitor outputs, each capable of delivering a truncated exponential monophasic waveform. The pulse widths and polarities of both outputs are programmable. To mimic a biphasic shock from a single capacitor defibrillator, the leading edge voltage of the second phase was set equal to the trailing edge voltage (VT) of the first phase. VT was calculated by using the equation
where VL is the leading edge voltage of the first phase, t is the duration of the first phase, R is the resistance estimated by the resistance from the previous shock, and C is the capacitance (150 μF). The polarity of the second phase (second output) was set opposite to the first phase (first output) with an interphase delay of 0.2 ms.
In part 1, five monophasic waveforms were compared with five biphasic waveforms with equal-duration first and second phases. The biphasic waveforms had phase durations of 1/1, 2/2, 3/3, 5/5, and 10/10 ms (Fig 2A⇓). The five monophasic waveforms had single phase durations of 2, 4, 6, 10, and 20 ms (Fig 2B⇓). Thus, each biphasic waveform was compared with a monophasic waveform with the same total duration to allow comparison of total waveform duration on defibrillation efficacy.
In part 2, all waveforms had the same total duration of 10 ms. One monophasic (10 ms) and one symmetrical biphasic (5/5 ms) waveform were compared with two asymmetrical biphasic waveforms (7.5/2.5 and 2.5/7.5 ms) with unequal first and second phase durations (Fig 2⇑, A, B, C, and D). This allowed determination of the effect of relative phase duration on the efficacy of the biphasic waveform.
If the patient was not already in atrial fibrillation, then atrial fibrillation was induced by rapid atrial burst pacing. Informed consent was obtained after initial screening of 15 patients, of which 13 had at least inducible sustained atrial fibrillation. Atrial fibrillation was considered sustained if it lasted for at least 5 minutes, and the protocol was not performed on patients unless atrial fibrillation was sustained. The 13 patients with sustained atrial fibrillation were divided into two groups. The first group consisted of the 6 patients enrolled in part 1 of the study and the second group consisted of the 7 patients enrolled in part 2 of the study. There was no crossover of patients between parts 1 and 2. After a successful defibrillation, a period of at least 1 minute was used before reinduction of atrial fibrillation. The next test shock was given at least 1 minute after atrial fibrillation was reinduced. The defibrillation shocks were synchronized to right ventricular activation.
In part 1, a step-up protocol was used starting with a shock strength of 0.5 J and increasing to 1, 1.5, 2, 3, 4, 5, 6, 7, 8, 9, up to a maximum of 10 J. In part 2, a step-up protocol was used starting with a shock strength of 0.5 J, increasing to 1, 2, 3, 4, 5, 6, 7, 8, 9, up to a maximum of 10 J. In part 2, the 1.5-J shock strength was not used for two reasons. First, the 1.5-J step in part 1 did not appear to significantly add to the resolution of the ADFT. Second, the number of shocks per patient was reduced and minimized the amount of time added to the clinical electrophysiological procedure time. The ADFT for each waveform was defined as the lowest energy shock that defibrillated the atria.
In part 1, randomization was achieved by drawing chits for each total waveform duration. A coin flip then was performed to determine whether the monophasic or the biphasic shock of the same total duration was delivered first. In part 2, randomization was achieved by drawing chits for each waveform. These randomization procedures were repeated for each patient.
For each delivered shock, the leading edge voltage and the shock waveform phase durations were programmed on the HVS-02 defibrillator. The trailing edge voltage, the delivered energy, and shock impedance were measured by the HVS-02 defibrillator. From the set values of leading edge voltage and pulse width, the leading edge current was calculated using the measured shock impedance. For the initial defibrillation shocks in each patient, the interelectrode impedance was assumed to be ≈50 Ω. Measured impedances from the HVS-02 were then used to determine subsequent shock leading edge voltage requirements.
Results are expressed as mean±SD. Comparisons between the two groups were made by using a two-tailed Student’s t test for paired data and the χ2 test for unpaired data. In part 1, comparisons between monophasic and biphasic waveforms of the same total duration were made by paired t test analysis.26 In parts 1 and 2, ANOVA multivariate analysis with repeated measures was used to compare the mean values of the atrial defibrillation thresholds among waveforms and patients.26 For all statistical tests performed, a value of P≤.05 was considered significant.
The clinical characteristics of the study population for parts 1 and 2 are summarized in Table 1⇓. There was no statistical difference between the two groups of patients in terms of age, height, or weight. The left atrial size was measured by transthoracic echocardiogram. Ejection fraction was determined by either transthoracic echocardiography or angiographic left ventriculography. The reported echocardiographic data were obtained during the admission when the defibrillation protocol was performed. Two patients did not have echocardiograms performed during the admission when the study protocol was performed. However, both patients had undergone echocardiographic evaluation previously at outside hospitals. The study population for parts 1 and 2 are also shown in regard to their electrophysiological diagnosis and therapy (Table 2⇓). Seven of the 13 patients were in atrial fibrillation at the time of admission to the electrophysiology laboratory. In part 1, three patients (50%) had a history of atrial fibrillation, and in part 2, five patients (71%) had a history of atrial fibrillation. Six of the eight patients with a history of atrial fibrillation were on warfarin therapy for at least 3 weeks before the study protocol. Postprocedure warfarin therapy was restarted. Patients No. 6 in part 1 and No. 6 in part 2 were diagnosed with lone atrial fibrillation and were treated with aspirin before the study protocol. Postprocedure aspirin therapy was continued. Although many of the patients were on antiarrhythmic medications before their study, all of the patients except for three in part 2 were in the drug-free state (Table 2⇓), defined as no medication for at least five half-lives of the drug. After ADFT determination, 10 of the 13 patients underwent some type of radiofrequency ablation procedure. Of the three remaining patients, two had failed attempts at external cardioversion and underwent successful internal cardioversion procedures and one had a diagnostic electrophysiological study for wide-complex tachycardia with presyncope and was shown to be noninducible.
Atrial Defibrillation Threshold Part 1
For most waveform durations, the mean and standard deviations in terms of total delivered energy at the ADFT were lower for the symmetric biphasic waveforms compared with the equal-duration monophasic waveforms (Table 3⇓). For total waveform durations of 4 to 20 ms, the symmetrical biphasic waveforms had significantly lower ADFTs than the monophasic waveforms. However, for the total duration of 2 ms, there was no significant difference between the biphasic and monophasic waveforms. The 1/1-ms biphasic waveform had a significantly higher ADFT in terms of delivered energy than the other 4 biphasic waveforms tested, but there was no significant difference between the symmetrical biphasic waveforms of 4- to 20-ms total duration. Similar findings were observed for the mean leading edge voltage and current at the ADFT (Table 3⇓). Thus, prolonging the biphasic waveform duration for longer than 4 ms did not further enhance the defibrillation efficacy, and the strength-duration curves in terms of leading edge voltage as well as total energy were flat for biphasic waveform durations of 4 to 20 ms (Fig 3⇓).
The strength-duration characteristics for atrial defibrillation were different for the monophasic waveform compared with the biphasic waveform. Unlike the biphasic waveform, the shortest-duration monophasic waveform with 2-ms duration had a defibrillation efficacy similar to the longer monophasic waveforms (Table 3⇑). In terms of leading edge voltage at the ADFT, there was not a statistically significant difference for monophasic waveforms throughout the durations tested of 2 to 20 ms. Thus, the strength-duration curve in terms of voltage was relatively flat, although voltage requirements tended to increase for the 20-ms monophasic waveform (Fig 3⇑). Similar findings were seen in terms of leading edge current (Table 3⇑). Reflecting the increase in both voltage and current requirements for the longest monophasic waveform, the 20-ms–duration monophasic waveform had a significantly higher energy ADFT compared with the other monophasic waveforms, suggesting that durations of >10 ms have an adverse effect on the defibrillation efficacy of monophasic waveforms. Thus, unlike the biphasic energy strength-duration curve, the monophasic curve for total energy was flat for durations of 2 to 10 ms but sloped significantly upward at a duration of 20 ms (Fig 3⇑). There were no significant differences in mean impedance for all the waveforms tested (Table 3⇑). The delivered charge for each waveform was calculated from the mean leading edge current, pulse width, and mean impedance. These values are shown in Table 3⇑ for each waveform in part 1. The minimal delivered charge was at a total pulse duration of 2 ms and increased with longer pulse durations up to 20 ms. This occurred for both the monophasic and biphasic charge curves; however, the monophasic curve was steeper than the biphasic curve.
Atrial Defibrillation Threshold Part 2
The 7.5/2.5-ms asymmetrical biphasic waveform had a significantly lower leading edge voltage and total delivered energy at the ADFT than the other three waveforms tested in part 2 (Fig 4⇓). The 5/5-ms symmetrical biphasic waveform had a significantly lower ADFT than the 2.5/7.5-ms asymmetrical biphasic waveform and the 10-ms monophasic waveform. There was no significant difference between the 2.5/7.5-ms asymmetrical biphasic waveform and the 10-ms monophasic waveform. Similar findings were seen with the mean leading current at the ADFT, and there was no significant difference in mean impedance for all waveforms tested (Table 3⇑).
The total number of shocks that had adequate electrogram data for analysis, the longest post–successful shock pause to first P wave, and the longest post–successful shock pause to first R wave are shown in Table 4⇓ for each waveform. The combined mean and standard deviation for the longest post–successful shock pause to first P wave and the longest post–successful shock pause to first R wave for biphasic and monophasic waveforms are also shown in Table 4⇓. The average number of shocks delivered to each patient was 50±18 shocks in part 1 and 26±10 shocks in part 2. There was no significant difference between biphasic and monophasic waveforms in terms of longest pause after a successful shock to the first P or R wave. The duration of the pauses appeared to be patient dependent and shock-strength independent. Several of the patients had postshock pauses that required temporary ventricular pacing for a maximum of 10 seconds. Occasionally, postshock first-degree (38 cases), second-degree (25 cases), or third-degree heart block (8 cases) was observed; however, this was transient and no permanent heart block was induced. The longest episode of any type of atrioventricular conduction block was 6.4 seconds.
No sustained ventricular arrhythmias were induced with shocks appropriately synchronized to the R wave. However, two episodes of ventricular fibrillation were induced when shocks were inadvertently delivered during the ventricular vulnerable period (Fig 5⇓) as the result of sensing of electrical noise in the external sensing circuit. Both of these episodes of ventricular fibrillation were converted to sinus rhythm with a single 200-J transthoracic countershock.
Combined Data From Parts 1 and 2
The patient population consisted of eight patients with a history of clinical atrial fibrillation and five patients without a history of clinical atrial fibrillation. For both groups there was a significant increase in the atrial defibrillation threshold for the 10-ms monophasic waveform compared with the 5/5-ms biphasic waveform. Only one patient had a decrease in threshold with the 10-ms monophasic waveform compared with the 5/5-ms biphasic waveform. The atrial defibrillation threshold with the 5/5-ms biphasic was 2.9±1.3 J for patients with a history of atrial fibrillation and 1.7±1.5 J for patients without a history of atrial fibrillation. This trend toward a higher ADFT for the biphasic waveform in the patients with a history of atrial fibrillation was not statistically significant (P=.08). The atrial defibrillation threshold of the 10-ms monophasic waveform was 5.2±2.5 J for patients with a history of atrial fibrillation and 4.7±2.7 J for patients without a history of atrial fibrillation. There was no significant difference between these two groups (P=.55).
Improved Efficacy of Biphasic Waveforms (Part 1)
Certain biphasic waveforms have been shown to be more effective than certain monophasic waveforms in both animal and human studies involving ventricular defibrillation16 17 18 19 20 21 22 23 24 as well as atrial defibrillation.11 12 13 14 15 For human atrial defibrillation, Johnson and coworkers12 directly compared a biphasic with a monophasic waveform of the same total duration for transcatheter internal cardioversion of atrial fibrillation in humans. They found that for a total duration of 6 ms, a symmetrical biphasic waveform (3/3 ms) had a significantly lower atrial defibrillation threshold than a monophasic waveform (6 ms) in 18 patients undergoing electrophysiological procedures. However, this study did not evaluate the effect of total waveform duration on the efficacy of the biphasic waveform for atrial defibrillation. In the present study, the symmetrical biphasic waveforms with total pulse durations of 4, 6, 10, or 20 ms were more efficacious than monophasic waveforms of the same total duration. However, shorter symmetrical biphasic waveforms of 2-ms total duration were not more efficacious than a 2-ms monophasic waveform. Thus, while symmetrical biphasic waveforms are generally more efficacious than equal-duration monophasic waveforms, this effect is clearly waveform-duration dependent.
Several studies have evaluated the effect of waveform duration on ventricular defibrillation. Schuder and coworkers27 found that the transthoracic defibrillation efficacy of several types of monophasic waveforms in dogs with ventricular fibrillation was critically dependent on pulse duration, with shorter (1 to 4 ms) and longer (64 to 256 ms) pulse durations having lower probabilities of successful defibrillation. Similarly, for ventricular defibrillation in dogs, Chapman et al,28 using nonthoracotomy leads, showed that shorter (2.5 ms) and longer (20 ms) monophasic waveforms had higher defibrillation threshold voltages than waveforms with durations in the middle of this spectrum. Hahn and colleagues29 showed a more complex relationship for ventricular defibrillation in pigs with biphasic waveforms with a total duration of 3 to 20 ms requiring higher leading edge voltage but less total energy compared with a longer biphasic waveform with a total duration of 30 ms. Like monophasic waveforms for ventricular defibrillation, the strength-duration curve in terms of voltage was steep with short biphasic waveform durations. However, unlike monophasic waveforms for ventricular defibrillation, the strength-duration curve in terms of voltage was flat with longer biphasic waveform durations.
Less is known about the effect of waveform duration on atrial defibrillation with either monophasic or biphasic waveforms. Cooper et al11 demonstrated in a sheep model of atrial fibrillation that monophasic waveform durations of 1.5 to 6 ms were not significantly different in terms of threshold voltage; thus, the strength-duration curve was flat for durations of 1.5 to 6 ms. For symmetrical biphasic waveforms, shorter (3 ms) and longer (12 ms) total pulse durations were associated with higher threshold voltages than with a total pulse width in the middle of this spectrum (6 ms). Thus, the strength-duration curve for biphasic waveforms for atrial defibrillation in sheep showed a relatively steep increase at shorter (3 ms) and longer (12 ms) total pulse durations, with a distinct nadir at 6 ms.
In the present study, the strength-duration curve for the monophasic waveform in humans was flat from 2 to 10 ms and thus similar to that described for atrial defibrillation in sheep. However, the 20-ms monophasic waveform had a significantly higher ADFT in terms of total energy compared with the other monophasic waveforms. This suggests that duration ≥20 ms has a negative effect on monophasic waveform defibrillation efficacy and is similar to the adverse effect of excessively long monophasic waveform durations seen for ventricular defibrillation.27 28 29 Thus, the strength-duration curve for symmetrical biphasic waveforms in human atrial defibrillation appears to be steep for total waveform durations <4 ms and flat for total pulse durations of 4 to 20 ms. The strength-duration curve for the symmetrical biphasic waveforms is shifted downward compared with the curve for the monophasic waveforms with 4- to 20-ms total duration. The sharp increase in ADFT at 2 ms for the biphasic waveform nullified any enhancement of efficacy compared with the monophasic waveform. No distinctly unique advantage was seen with the 3/3-ms biphasic waveform as seen in sheep. This study did not demonstrate that the threshold was higher for longer biphasic waveform durations. Whether symmetrical biphasic waveforms with total durations >20 ms would have significantly higher ADFTs is not known. As is shown in this study, waveform duration may have divergent effects on the efficacy of monophasic waveforms and symmetrical biphasic waveforms for atrial defibrillation in humans, especially at relatively shorter and longer durations.
Importance of Phase Duration (Part 2)
Several groups have demonstrated with ventricular defibrillation studies in animals that the efficacy of a biphasic waveform is dependent on the duration as well as the amplitude of both phases of the waveform.18 19 25 Dixon et al18 demonstrated that asymmetrical biphasic waveforms with the second phase shorter than the first phase are more efficacious than biphasic waveforms with a longer second phase than first phase for electrically induced ventricular fibrillation in a canine model. They also found that there was not a significant difference between the 5/5-ms symmetrical (mean voltage at ventricular defibrillation threshold of 116±19.4 V) and 7.5/2.5-ms asymmetrical biphasic (mean voltage at ventricular defibrillation threshold of 114±19.7 V) waveforms with the second phase shorter than the first phase for ventricular defibrillation. The present study demonstrated a similar phase dependence for atrial defibrillation in humans. The 7.5/2.5-ms asymmetrical biphasic waveform with the first phase longer than the second phase was more effective than either the 5/5-ms symmetrical biphasic waveform with equal phase durations or the 2.5/7.5-ms asymmetrical biphasic waveform with a longer second phase than first phase duration. The symmetrical biphasic waveform was significantly better than the unfavorable 2.5/7.5-ms asymmetrical biphasic waveform, but the unfavorable asymmetrical biphasic waveform was similar in defibrillation efficacy to the 10-ms monophasic waveform of the same total duration. Again, these results are very similar to the findings with similar asymmetrical biphasic waveforms for ventricular defibrillation in dogs, suggesting a similar mechanism of enhanced efficacy with biphasic waveforms in atrial and ventricular defibrillation. However, the one major difference is that the 7.5/2.5-ms asymmetrical biphasic waveform with the second phase shorter than the first phase was more effective than the 5/5-ms symmetrical biphasic waveform in this study. This difference between atrial and ventricular defibrillation may be explained in part by differences in species, myocardial properties, electrode configuration, hemodynamic state, autonomic tone, and type of anesthesia/sedation.
This method of atrial defibrillation appears to be safe and effective when appropriate R-wave sensing occurs. There were no long-term complications, and temporary pacing was required in a minority of the patients for postshock conduction delays. No permanent sinus or atrioventricular nodal dysfunction occurred as a complication of the defibrillation protocol. However, since postshock bradycardia does occur, backup bradycardia pacing will be needed with any implantable atrial defibrillator. The only serious complication was induction of ventricular fibrillation due to inappropriate synchronization from equipment error and delivery of the shock during the ventricular vulnerable period. This emphasizes the critical importance of ensuring accurate synchronization of the shock to the R wave to avoid potentially lethal postshock ventricular arrhythmias in future implantable devices; otherwise, backup ventricular defibrillation will be needed.
Study Limitations and Combined Data From Parts 1 and 2
The limitations to this study are the fact that not all of the patients had a history of atrial fibrillation and several of them had relatively normal atria by clinical evaluations. Also, in all cases, subsequent episodes of atrial fibrillation were induced electrically, and this may not be the same as naturally occurring atrial fibrillation. However, all of these atria were able to sustain atrial fibrillation and the majority of the patients did have a clinical history of paroxysmal atrial fibrillation. Another limitation is that the ADFT can vary with time within a patient. The defibrillation threshold determination is not suggesting that a sharp cutoff point exists below which all attempts to defibrillate fail and above which all attempts succeed. A single “threshold” value is used to represent the efficacy of a waveform because it could be measured easily and limited the number of shocks to the patient. It is hoped that by using the defibrillation threshold method, marked differences in the defibrillation requirements would be discovered and the efficacy of multiple waveforms could be compared. Furthermore, the ADFT was determined for each waveform in each patient, which should at least in part control for the variability in ADFT. Support of this method is demonstrated by the finding that all but one of the total 13 patients had at least a small increase in defibrillation requirements with the 10-ms monophasic versus the 5/5-ms symmetrical biphasic waveform. Also, in part 2, the 1.5-J step size in determining the ADFT was not used, and this could have biased part 1 results toward lower defibrillation thresholds. The differences in atrial defibrillation thresholds for the 5/5-ms symmetrical biphasic and the 10-ms monophasic waveforms in parts 1 and 2 can be explained in part by the inherent variability in defibrillation threshold determination, the differences in the way the atrial defibrillation threshold was determined in each part, and the differences in patient populations between the two groups.
A total of 13 patients were tested with both the 5/5-ms biphasic and the 10-ms monophasic waveforms. However, these patients were from combined data from the two parts of the study, and these two groups were not treated identically in terms of waveforms tested. Patients with a history of atrial fibrillation did not have a significantly higher ADFT than patients without a history of atrial fibrillation for either the biphasic or monophasic waveform; however, there did appear to be a trend with the biphasic waveform. There was no difference in the 10-ms monophasic waveform between the two groups. The reason for this is probably that the sample size was not quite large enough to detect a statistical difference. Johnson et al30 reported on another group of 18 patients in whom a 3/3-ms biphasic waveform and 6-ms monophasic waveform were tested. They found that patients with a history of atrial fibrillation (n=9) had a significantly higher ADFT for both the biphasic and monophasic waveforms. These findings, along with the findings of the present study with a trend toward a lower biphasic threshold in patients without a history of atrial fibrillation, support the findings of Levy et al.31 They reported that the ADFT for a 3/3-ms biphasic waveform was statistically higher in patients with a history of atrial fibrillation (n=120).
Most human internal atrial defibrillation studies have involved heavy sedation of the patients, and the pain associated with the shocks could not be quantified. Murgatroyd et al13 demonstrated that more than ≈1 J of energy was associated with intolerable pain in patients undergoing internal atrial defibrillation shocks without sedation. Although pain to the patients was not measured in this study, the mean ADFT energy for all waveforms tested was >1 J. However, waveforms with total pulse durations >2 ms had significantly lower ADFTs, and it is hoped that future research in this area will be focused toward the development of even more efficient waveforms and lead systems to help lower the threshold and minimize discomfort to the patient.
There are several clinical applications for this type of cardioversion system. A considerable amount of interest has developed over the past 4 years for an implantable atrial defibrillator, and clinical trials assessing this technology are presently in progress. This device will be especially useful in the medically refractory patients with infrequent paroxysms of atrial fibrillation. This type of atrial cardioversion system is useful in the clinical electrophysiological laboratory by providing a means to quickly cardiovert atrial fibrillation induced during an electrophysiological study. Furthermore, this type of system may be useful in patients who have failed external cardioversion.32 33 Last, addition of atrial defibrillation capacity to a ventricular defibrillator would potentially allow for better arrhythmia detection and discrimination as well as provide a more complete arrhythmia treatment system.
Certain biphasic waveforms are more effective than certain monophasic waveforms for internal atrial defibrillation in humans. Biphasic waveforms with the first phase longer than the second phase appear to be more effective than biphasic waveforms with both phases of the same duration as well as biphasic waveforms with the first phase duration shorter than the second phase duration. Although some of the biphasic waveforms evaluated in this study were more efficient for internal atrial defibrillation, the threshold levels were still in the range that most patients would probably feel significant discomfort without sedation. It is hoped that continued research into this area will result in even more efficient defibrillation waveforms and lead systems to help further minimize the discomfort to the patient.
This study was supported in part by National Institutes of Health grant HL-17670.
- Received May 2, 1996.
- Revision received October 24, 1996.
- Accepted November 12, 1996.
- Copyright © 1997 by American Heart Association
Wolf PA, Dawber TR, Thomas E Jr, Kannel WB. Epidemiologic assessment of chronic atrial fibrillation and risk of stroke: the Framingham study. Neurology. 1978;28:973-977.
Pai S, Torres V. Atrial fibrillation: new management strategies. Curr Probl Cardiol. 1993;18:233-300.
Lown B. Electrical reversion of cardiac arrhythmias. Br Heart J. 1967;29:469-489.
Yabe S, Smith WM, Daubert JP, Wolf PD, Rollins DL, Ideker RE. Conduction disturbances caused by high current density electric fields. Circ Res. 1990;66:1190-1203.
Jones JL, Lepeschkin E, Jones RE, Rush S. Response of cultured myocardial cells to countershock-type electric field stimulation. Am J Physiol. 1978;235:H214-H222.
Pansegrau DG, Abboud FM. Hemodynamic effects of ventricular defibrillation. J Clin Invest. 1970;49:282-297.
Dahl CF, Ewy GA, Warner ED, Thomas ED. Myocardial necrosis from direct current countershock: effect of paddle size and time interval between discharge. Circulation. 1974;50:956-961.
Cooper RAS, Alferness CA, Smith WM, Ideker RE. Internal cardioversion of atrial fibrillation in sheep. Circulation. 1993;87:1673-1686.
Johnson EE, Yarger MD, Wharton JM. Monophasic and biphasic waveforms for low-energy internal cardioversion of atrial fibrillation in humans. Circulation. 1993;88(suppl I):I-592. Abstract.
Murgatroyd FD, Johnson EE, Cooper RA, Lau C, Alt E, Kappenburger LJ, Smith JM, Camm AJ, Wharton JM. Safety of low-energy transvenous atrial defibrillation: world experience. Circulation. 1994;90(suppl I):I-14. Abstract.
Keane D, Sulke N, Cooke R, Jackson G, Sowton E. Endocardial cardioversion of atrial flutter and fibrillation. PACE. 1993;16:928. Abstract.
Gurvich NL, Yuniev GS. Restoration of regular rhythm in the mammalian fibrillating heart. Bull Exp Biol Med. 1939;8:55-58.
Dixon EG, Tang ASL, Wolf PD, Meador JT, Fine MJ, Calfee RV, Ideker RE. Improved defibrillation thresholds with large contoured epicardial electrodes and biphasic waveforms. Circulation. 1987;76:1176-1184.
Chapman PD, Vetter JW, Souza JJ, Troup PJ, Wetherbee JN. Comparative efficacy of monophasic and biphasic waveforms for non-thoracotomy canine defibrillation. PACE. 1988;11:497. Abstract.
Feeser SA, Tang ASL, Kavanagh KM, Rollins DL, Smith WM, Wolf PD, Ideker RE. Strength-duration and probability of success curves for defibrillation with biphasic waveforms. Circulation. 1990;82:2128-2141.
Probit procedure. SAS User’s Guide. Cary, NC: SAS Institute, Inc; 1982:287.
Schuder JC, Rahmoeller GA, Stoeckle H. Transthoracic ventricular defibrillation with triangular and trapezoidal waveforms. Circ Res. 1966;19:689-694.
Chapman PD, Wetherbee JN, Vetter JW, Troup P, Souza J. Strength-duration curves of fixed pulse width variable tilt truncated exponential waveforms for nonthoracotomy internal defibrillation in dogs. PACE. 1988;11:1045-1050.
Hahn SJ, Heil JE, Lin Y, Lang DJ. Defibrillation strength-duration relationships for fixed tilt biphasic waveforms. Circulation. 1994;90(suppl I):I-1229. Abstract.
Johnson E, Smith W, Yarger M, Cooper R, Broughton A, Wharton JM. Clinical predictors of low energy defibrillation thresholds in patients undergoing internal cardioversion of atrial fibrillation. PACE. 1994;17:742. Abstract.
Levy S, Ricard P, Lau C, Lok N, Camm AJ, Murgatroyd F, Kappenberger LJ, Jordaens LJ, Brugada P. Multicenter Low Energy Transvenous Atrial Defibrillation (XAD) Trial. Circulation. 1995;92(suppl I):I-472. Abstract.
Cooper RAS, Johnson EE, Kanter RJ, Merrill JJ, Sorentino RA, Wharton JM. Internal cardioversion in two patients with atrial fibrillation refractory to external cardioversion. PACE. 1996;19:872-875.
Alt E, Schmitt C, Ammer R, Coenen M, Fotuhi P, Karch M, Blasini R. Initial experience with intracardiac atrial defibrillation in patients with chronic atrial fibrillation. PACE. 1994;17:1067-1078.