Comparison of Chronic Biphasic Pacing versus Cathodal Pacing of the Right Ventricle on Left Ventricular Function in Sheep after Myocardial Infarction

Abstract

Background: The physiological effects of biphasic pacing have not been studied in compromised hearts.

Methods: Myocardial infarction was induced in 12 sheep by high coronary artery ligation. Perioperative mortality was 33%. The surviving eight animals exhibited increased left ventricular volume and reduced percent fractional shortening. Two weeks after the infarction, sheep were implanted with atrial‐triggered, right ventricular pacemaker systems capable of pacing with cathodal (cathodal pulse) and biphasic (anodal pulse followed by cathodal pulse) waveforms, and randomly assigned to an initial mode. At 3‐month intervals, or whenever pacing was lost for any reason, the pacing system was switched to the alternative mode. Cardiac function was assessed at 2‐ to 3‐week intervals through the use of echocardiograms. Successful pacing was confirmed over an average of 8 weeks in each mode.

Results: Cathodal pulsing had neither beneficial nor deleterious effect on the diminished cardiac performance induced by myocardial infarction. When compared to the cathodal mode, biphasic pulsing improved cardiac performance as reflected by decrease of diastolic and systolic ventricular volumes, reduction in left ventricular systolic diameter, and increases in percent fractional shortening. When compared to the unpaced state after the myocardial infarction, the percent fractional shortening was significantly increased by biphasic pacing. Concordant trends in improvement in the other cardiac parameters were also observed for the biphasic mode. No ventricular tachyarrhythmias or mortality was associated with biphasic stimulation.

Conclusion: When compared to cathodal pacing, myocardial biphasic pacing has no safety issues in sheep that have undergone a large myocardial infarction. Importantly, biphasic pulsing elicited significant benefits in cardiac performance.

Ann Noninvasive Electrocardiol 2011;16(2):111–116

Biventricular pacing for treatment of congestive heart failure ¹ was first applied clinically ² in 1991. The initial five implants used a bipolar Y‐adapter, pacing one ventricle with the cathode and the other with the anode. Quite unexpectedly, the pacing thresholds gradually rose at the anode and, in 4–6 weeks, threshold exceeded the output of the generator, at which time pacing at the anode was lost. When this occurred, heart failure returned and the Y‐adapter had to be replaced with a split‐cathode design, thereby allowing pacing of both sites with the cathode. Upon completion of this procedure, pacing of both ventricles resumed and the heart failure again resolved. ³ Patients implanted thereafter used only the split‐cathode design. However, the resolution of the congestive heart failure seemed to occur more slowly than when there had been the presence of the anodal phase, suggesting that there might be some additional beneficial properties of having an anodal component. Experiments in a variety of animal species ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ have indicated that anodal stimulation alone or as part of a biphasic waveform gives rise to driven beats that travel faster over the myocardium and enhance contractility. It was suspected that the rise in threshold associated with anodal pacing could be avoided by adding a cathodal pulse to the end of the anodal one, thereby producing a biphasic pulse. Increasing the speed of myocardial conduction and contractility through the use of a biphasic pulse would be predicted to be beneficial for antitachycardia burst pacing for termination of atrial arrhythmias including fibrillation for enhancing cardiac output in heart failure from a right ventricular pacing site alone, and for improving the performance of pacemakers and defibrillators for conventional indications.

METHODS

This study had two primary objectives and a secondary objective. The first primary objective was to determine whether biphasic pulsing therapy has similar effects as cathodal pulsing on cardiac performance in compromised hearts, specifically in sheep with large myocardial infarcts. Cardiac performance was assessed using echocardiograph‐derived measurements. The second primary objective was to elucidate the safety of biphasic pulsing therapy, particularly in high risk, cardiac compromised sheep. Safety parameters included mortality, morbidity, veterinarian assessment of the sheep, and weight changes. The secondary objective was to examine whether pacing thresholds for biphasic pulsing increased in a manner similar to anodal pacing or remained constant as has been observed for cathodal pulsing. Thresholds were tracked and potential loss of capture was documented over an extended time.

Chronic Sheep Study

This study was initiated at a facility in Wisconsin on October 2, 2007, transferred after a year to a site in Minneapolis, and was finally ended on December 30, 2008. Large myocardial infarctions (MIs) were induced in 12 young female sheep by high coronary ligation. Two weeks thereafter, the eight surviving sheep were surgically implanted with a standard commercially available atrioventricular implantable pacemaker with the ventricular output coupled to an implantable slave stimulator (Rivertek Inc., St. Paul, MN), producing pacing pulses of biphasic or cathodal mode as desired. Both units were placed in a common surgical pocket. The output of the slave stimulator was delivered to a conventional screw‐in endocardial lead inserted in the midportion of the right ventricular septum.

Two weeks after implantation, atrial‐triggered pacing was initiated with either the biphasic or cathodal mode, and crossed over to the opposite pacing mode at 3‐month intervals or whenever pacing was lost for any reason. For short times during the study, pacing of either mode was changed to fixed‐rate overdrive mode to keep pacing going when there was difficulty with the atrial‐triggered mode. Programming of rate and sensing of the standard commercial pacemaker were done using the standard pacemaker programmer. The different waveform modalities were implemented by separately programming the implanted slave stimulator. The standard twice‐threshold cathodal pulse used was 0.5 ms at 3.5 V. The biphasic pulse was produced by preceding the cathodal pulse with a subthreshold anodal pulse of 0.5 msec at 1.3 V. Thresholds were measured by reducing the cathodal pulse or cathodal component of the biphasic pulse until capture was lost. These thresholds were consistently 1.7 V. Stimulation with the cathodal and biphasic pulses occur on the leading edge of the cathodal component. The anodal phase was deliberately selected to be subthreshold, but even if higher voltages had been used, stimulation would still only have occurred on the anodal “break” coincident with the “make” stimulation of the cathodal pulse.

Echocardiographic measurements were made by a single operator blinded to the pacing mode. These were made periodically, at 2‐ to 3‐week intervals. Multiple echocardiographic views were carried out to obtain left ventricular diastolic volume (LVDV), left ventricular systolic volume (LVSV), left ventricular diastolic diameter (DD), left ventricular systolic diameter (LVSD), and percent fractional shortening (FS).

Echocardiographic data were not obtained if pacing was lost for programming reasons or battery failure. For most parameters for the eight test animals, there were a total of eight satisfactory measurements for baseline, 36 for the unpaced post‐MI state, 69 for biphasic pacing, and 75 for cathodal pacing. Successful measurements were achieved over an average of 8 weeks in each pacing mode.

Electrocardiograph (EKG) rhythm strips were obtained at roughly weekly intervals to verify pacing. Veterinarian examinations were periodically carried out to evaluate overall condition of the sheep. The changes in pacing were managed by a technician completely independently of the veterinarian. A composite score evaluating a well‐being measure of the animal (WMA) that included parameters such as temperature, appetite, heart and respiratory rates, presence or absence of rales, etc. was computed and tracked. Additional engineering assessment of adequacy of pacing to troubleshoot the pacing systems was accomplished when needed.

Statistical analyses were performed with GraphPad Prism 4 for Windows Ver 4.03 (GraphPad Soiftware Inc., LaJolla, CA). Means are expressed ± standard error of the mean (SEM). Two‐tailed unpaired t‐tests were used to identify significant differences between groups. P < 0.05 was considered significant. Absolute differences of parameters and changes between the modes in percentages were calculated. Box and whisker plots showing median values, 25th and 75th percentiles, and extreme values of the data, and graphs of means were also constructed. Fisher's exact test was used in the case of categorical data.

Study Course

The sheep exhibited a gain in weight from 57.0 ± 2.0 to 74.0 ± 1.8 kg, over an average of 8.6 ± 1.3 weeks. The gain was smooth, the rate was not related to pacing mode, and was considered consistent with normal growth. The first two animals had two replacements each due to battery depletion of the slave stimulator. The first animal also had an episode of eosinophilia and four episodes of fever (three in cathodal mode and one in biphasic), some associated with positive blood cultures. All episodes were treated successfully with antibiotics. Several sheep had loss of appetite for a week initially after implantation. All sheep were agitated when the initial echocardiograms were obtained, but quickly became adapted to the procedure. One animal was found dead in its cage while being paced in the cathodal mode. Partway through the study, the animals were moved from one facility to another, after which another death occurred, this time in the unpaced mode. Relatively little echocardiographic data, and EKG data, and examinations were collected in the second facility. While this had essentially no effect on safety determinations or threshold measurements, the study was terminated shortly thereafter when it was deemed that no further useful data could be obtained.

RESULTS

The infarction induced by high coronary ligation elicited significant reductions in myocardial function with an 80% increase in LVDV (P < 0.001) (Fig. 1), a 50% increase in LVSD (P < 0.001), and a 20% reduction in FS (P = 0.004) (Table 1).

Figure 1.

Effect of induced myocardial infarction (MI) on left ventricular diastolic volume (LVDV). Two‐chamber LVDV was measured by echocardiography in sheep prior to (baseline) and 2 weeks after (post‐MI) induction of a MI. Individual animals are indicated by dots and the mean for each group by a horizontal line. In the baseline state, a single measurement was made for each animal. Two of the dots are superimposed at 58 cm³. In the post‐MI state, repeated measurements were made for several weeks. The LVDVs post‐MI were 80% greater than at baseline (P < 0.001), Student's unpaired t‐test. The box and whiskers graph shows median, 25th and 75th percentiles, and extreme values.

Table 1.

Modulation of Cardiac Parameters by Myocardial Infarction and Pacing. Left Ventricular Systolic Diameter (LVSD) and Percent Fractional Shortening were Measured by Echocardiography in Sheep prior to (baseline) and after Induced Myocardial Infarction (MI) without Pacing (Post‐MI), and with Cathodal and Biphasic Pacing. Percent Fractional Shortening was Calculated as ([{LV End‐Diastolic Diameter} − {LV End‐Systolic Diameter}]/[LV End‐Diastolic Diameter]) × 100. MI Increases LVSD by 50% (P < 0.001); Fractional Shortening Is Reduced 20% (P = 0.004). Biphasic Pulses Produced Fractional Shortening Increase of 8.7% (P = 0.026). LVSD Showed a Trend with Biphasic Pulses toward a Smaller Ventricle

Condition	Measurements	Cardiac Parametersa
		Fractional Shortening (%)b	LV Systolic Diameter (cm)
Baseline	8	45.8 ± 1.5	20.8 ± 0.6
Post‐MI	36	37.6 ± 1.1*	31.0 ± 0.7*
Cathodal	69	37.5 ± 0.6*	31.8 ± 0.4*
Biphasic	75	40.9 ± 0.9*& #	29.5 ± 0.5*#

^a Data represent the mean ± SEM.

^b Percent fractional shortening was calculated as ([{LV end‐diastolic diameter} – {LV end‐systolic

diameter}]/[LV end‐diastolic diameter]) × 100.

* P < 0.05, unpaired Student's t‐test, compared to measurement under baseline condition.

^& P < 0.05, unpaired Student's t‐test, compared to measurement under post‐MI condition.

^# P < 0.05, unpaired Student's t‐test, compared to measurement under cathodal pulsing condition.

Pacing of the heart with cathodal pulses had no effect (P > 0.05) on any of the parameters monitored, as shown in Table 1 for LVSD (P = 0.28) and FS (P = 0.96). By contrast, biphasic pulses induced improvements in several cardiac parameters compared to the unpaced state, as well as with cathodal pulsing. In the former, biphasic pulsing increased FS by 8.7% (P = 0.026). There was also a trend for a decrease in LVSD (P = 0.096). Figure 2 shows improvement in the FS with biphasic pulses. When compared to the same parameters under cathodal pulsing conditions (Table 1), biphasic pulsing increased FS (P = 0.001) and decreased LVSD (P = 0.0003). Similar improvements were seen in LVDV (Table 2). Of particular note, biphasic pulsing decreased diastolic volumes by 6.4–8.1%. Figure 3 illustrates this reduction (P = 0.019) in LVDV. For this parameter, biphasic pulsing was associated with a lower median as well as the smaller mean.

Figure 2.

Modulation of percent fractional shortening (FS) by myocardial infarction (MI) and pacing. Percent FS was measured by echocardiography in sheep at baseline and after induced MI without pacing (post‐MI) and during cathodal and biphasic pacing. Percent FS was calculated as ([{LV end‐diastolic diameter} − {LV end‐systolic diameter}]/[LV end‐diastolic diameter]) × 100. The box and whiskers graph shows median, 25th and 75th percentiles, and extreme values. Percent FS significantly improved with biphasic pulses compared to both unpaced state and cathodal pulsing. There are two extreme values shown in the plot for the Biphasic waveform, which however do not alter the narrow confidence interval.

Table 2.

Effect of Cathodal and Biphasic Pacing on Cardiac Parameters in Sheep with an Induced Myocardial Infarction. Left Ventricular Diastolic Volumes (LVDV) were Measured by Echocardiography in Two‐Chamber, Four‐Chamber, and Biplane Views during Cathodal and Biphasic Pacing. Reductions in Left Ventricular Size with Biphasic Pulses as Compared to Cathodal Ranged from 7.9 to 10.6 cm³ (6.5–8.1%) (P Values from 0.02–0.04, Student's Unpaired t‐Test)

Cardiac parameter	Volume (cm3)a		Δb	Δ%c	Pd
	Cathodal	Biphasic
Two‐chamber LVDV	121.2 ± 2.5 (74)	113.2 ± 2.8 (69)	−7.9	−6.6	0.04
Four‐chamber LVDV	129.4 ± 3.1 (75)	118.9 ± 3.2 (69)	−10.6	−8.1	0.02
Biplane LVDV	123.9 ± 2.5 (75)	115.9 ± 2.8 (69)	−8	−6.5	0.04

^a Data represent the mean ± SEM from the number of observations noted in parentheses.

^b Difference in volume measured, that is, (biphasic volume) minus (cathodal volume).

^c Difference in volume measured expressed as a percentage of the cathodal volume.

^d P value, unpaired Student's t‐test, compared to cathodal volume.

Figure 3.

Comparison of cathodal and biphasic pacing on left ventricular diastolic volume (LVDV) in sheep with an induced myocardial infarction. Four‐chamber DV was measured by echocardiography during periods of cathodal and biphasic pacing. Left panel: Individual animals are indicated by dots and the mean for each group by a horizontal line. The volumes during biphasic pacing were reduced relative to those during cathodal pacing (P = 0.019, Student's unpaired t‐test). Right panel: The box and whiskers graph shows median, 25th and 75th percentiles, and extreme values. In cathodal pacing, the spread of values is narrower, suggesting some limitation of diastolic stretching at the higher volumes.

Parenthetically, at the higher ventricular volumes associated with cathodal pacing, the spread of values was smaller, perhaps suggesting some limitation of stretching at the higher volumes. Many other parameters only showed trends in favor of the biphasic pacing; however, all changes in direction were concordant.

No safety issues were detected with the biphasic waveform. All safety data, both among animals and between animals, during crossover phases showed no differences between cathodal and biphasic pacing with respect to morbidity and veterinarian examination. There was one death each in the cathodal and in the unpaced modes. The composite WMA score of well being in the animals was equally high (12 of a possible 13 maximum) in both the cathodal and biphasic pacing groups. The Fisher's exact test showed no significant differences in mortality or WMA score.

DISCUSSION

The present study demonstrates that biphasic waveform pacing of a compromised heart improves cardiac performance. Reductions in diastolic and systolic volumes of 6–8% were achieved. In addition, the increase in FS induced by biphasic pulsing was equivalent to about 40% of the total FS loss caused by the infarction. This extent of change is physiologically meaningful. Cathodal waveform pacing is the waveform used in current pacing devices and does not share these beneficial properties. It is logical to think that similar changes would occur in less severely compromised hearts.

Previous experience was that the efficacy of anodal pacing of the heart decreases over time due to increases in the threshold for stimulation. This was not observed for biphasic pulsing in the present study. Thresholds for the cathodal and biphasic waveforms were measured by gradually reducing the amplitude of the cathodal component until loss of capture. The average thresholds for both were 1.7 V. Capture remained essentially constant for each animal and each pacing mode for the duration of the study. The only instances of loss of capture during either biphasic or cathodal pacing were due to programming issues or battery failures. Pacing energies were unchanged during the entire study.

Biphasic pulsing had no deleterious effects on the health of these cardiovascularly compromised animals as evidenced by similar mortality and equal subjective composite assessment scores of well being.

Finally, cardiac tissue retains normal threshold sensitivity to biphasic waveforms, distinguishing biphasic pacing from anodal pacing. While the cause of the threshold rise in anodal pacing remains to be determined, it appears that the cathodal element of the biphasic waveform may have a protective effect. That biphasic waveforms do not exhibit the threshold rise problem is important because it shows that these waveforms can be used in therapeutic settings without being burdened by threshold issues or any negative impact on pacing device longevity.

The beneficial cardiac effects of biphasic pacing found in this study may, in fact, be underestimated because, although the crossover design compensates somewhat for any imbalances between animals at baseline, it also reduces the overall beneficial effects because of the time spent in the inferior cathodal pacing mode. These considerations notwithstanding, the present results provide background rationale for the potential use of biphasic pulses to promote increased contractility in the compromised heart. Use of the biphasic waveform may have many useful applications in device therapy.