Abstract

Aims

This study investigated ventricular electrophysiological characteristics and the correlation between these parameters and heart rate variability (HRV) and baroreflex sensitivity (BRS) in a canine congestive heart failure (CHF) model.

Methods and results

Haemodynamics, HRV, BRS, and ventricular electrophysiological variables were measured 4–5 weeks after sham operation (control dogs) and pacemaker implantation, and rapid right ventricular pacing at 240 bpm (CHF group). In the CHF group, significant differences from the control group in ventricular effective refractory period (VERP), monophasic action potential (MAP) duration (MAPD 90 ), ventricular late repolarization duration (VLRD), the ratio of VERP to MAPD 90 , dispersion of ventricular recovery time (VRT-D), and ventricular fibrillation threshold (VFT) were noted. Both BRS and the time and power domain parameters of HRV were significantly decreased in the CHF group compared with the control group, and a significant, positive correlation between HRV and BRS was identified in the CHF group. Heart rate variability and BRS were negatively and significantly correlated with VLRD and VRT-D, and were positively correlated with VERP/MAPD 90 and VFT in the CHF group.

Conclusion

These results suggest that ventricular electrophysiological characteristics correlated with abnormal autonomic nerve function may have important effects on sudden cardiac death. Further research is warranted.

Introduction

Congestive heart failure (CHF) is a chronic medical condition affecting millions of people worldwide and is associated with a poor prognosis. Complex ventricular arrhythmias are encountered in ∼90% of patients with CHF. Ventricular arrhythmias often result in sudden cardiac death, the leading cause of death in these patients. 1

Changes in ventricular electrophysiological characteristics in CHF can lead to ventricular arrhythmias and cardiac sudden death. 2 , 3 Experimental researches and clinical findings have demonstrated that ventricular monophasic action potential duration (MAPD 90 ) and ventricular effective refractory period (VERP) are increased with CHF. 4 , 5 Dysfunction of the autonomic nerve system has also been closely correlated with the occurrence of CHF and sudden cardiac death. 6 It has been suggested that reductions in heart rate variability (HRV) and baroreflex sensitivity (BRS), which are markers of autonomic nerve function, could be used to predict risk of sudden cardiac death. 7–10 Indeed, decreased HRV has been shown to be associated with an increased risk of cardiac death in a number of studies. 11–14 In particular, BRS and HRV have been shown to be strong predictors of and important contributors to risk stratification for cardiac mortality after myocardial infarction. 8 , 9 As early as 1988, it was suggested that decreased BRS is associated with the development of ventricular fibrillation and, therefore, sudden cardiac death. 15 Further research has demonstrated that the measurement of BRS identified dogs at risk for sudden cardiac death. 8 , 15

Based on the results from these studies, it was hypothesized that autonomic nerve dysfunction may increase the incidence of sudden cardiac death in patients with CHF by altering ventricular electrophysiology. However, in the assessment of risk of sudden cardiac death caused by CHF, clinical information about the association of the variation of cardiac electrophysiological parameters and the changes of HRV and BRS has never been reported. Therefore, the purpose of this study was to determine if HRV and BRS were correlated with various cardiac electrophysiological parameters, particularly those that reflect the electric stability of the ventricles including ventricular recovery time dispersion (VRT-D), ventricular late repolarization duration (VLRD), VERP/MAPD 90 , and ventricular fibrillation threshold (VFT).

Materials and methods

Subjects

Healthy adult specific pathogen-free mongrel dogs ( n = 22) weighing 13.8 ± 1.2 kg were obtained from the Sun Yat-Sen University Animal Laboratory. The dogs were randomly assigned into either a control (sham-operated) group ( n = 10) or a CHF group ( n = 12). Baseline haemodynamic parameters (described below) were obtained in all dogs prior to implantation surgery. This study was approved by the Institutional Review Board of the Sun Yat-Sen University Animal Laboratory.

Induction of congestive heart failure

Congestive heart failure was induced by rapid ventricular pacing as described by Armstrong et al . 16 Briefly, 12 dogs in the CHF group were anaesthetized with an intraperitoneal injection of 3% sodium pentobarbital (30 mg/kg). Five per cent glucose in normal saline (500 mL) with penicillin (4.8 million IU) was administered intravenously. Under fluoroscopy, an endocardial pacemaker electrode (St Jude Medical, St Paul, MN, USA) was inserted into the right ventricular apex via the left external jugular vein. A pacemaker generator (Guangshou Radio Research Institute, People's Republic of China) was implanted in a small subcutaneous pocket created between the scapulas, and the pacemaker lead was connected to the generator through a subcutaneous canal. The pacing threshold was 0.3–1.5 V, the amplitude of the R-wave 4–10 mV, and the impedance 0.3–1.0 kΩ. The pacemaker frequency was set at 240 bpm with an output voltage of 5.0 V and a pulse width of 0.5 ms. Dogs in the control group underwent a sham surgery in which the pacemakers and electrodes were installed, but not connected.

Surgery was completed in an average time of 2–3 h, and all experimental procedures were performed in a central laboratory that had a consistent temperature control of 24–26°C. Serum electrolyte concentrations (K + , Na + , Cl , Ca 2+ , and Mg 2+ ) were monitored in all dogs, and appetite, behaviour, activity status, and respiratory rate were recorded post-operatively. Rapid pacing in the experimental group was continued for 4–5 weeks, and during this period dogs in both experimental and control groups were trained to stand quietly in a sling for electrocardiographic (ECG) recording once a week to ensure that the cardiac pacing was stable and sustainable in the CHF group and that normal heart rate and rhythm were present in the control group. 17

Four to five weeks after the initiation of rapid pacing, electrophysiological and haemodynamic measurements were performed in both groups. Continuous 30 min ECG recordings were taken in conscious animals for the analysis of HR variation. The respiratory rate of each dog was also monitored during this period. In the CHF group, as described previously by others, 18 rapid pacing was stopped 30 min prior to the ECG recording. After completion of ECG and respiratory rate recording, a phenylephrine injection protocol was used to measure BRS. Animals were subsequently anaesthetized for the measurement of haemodynamic parameters and cardiac electrophysiological characteristics, and then euthanized, so that morphological and histological examination could be performed.

Measurement of heart rate variability

Heart rate variability was measured from 30 min continuous ECG records (in the case of the CHF group, taken 30 min after discontinuation of pacing) using software from GE Medical Systems (Milwaukee, WI, USA). The time and frequency domain parameters of this variable were analysed using HRV analysis software.

Time domain parameters included the standard deviation (SD) of the RR intervals (SDNN, SD of all normal sinus rhythm RR intervals within a specified time in milliseconds), the root-mean-square of this SD (rMSSD, root-mean-square SD of the continuous normal sinus rhythm RR intervals within a specified time in milliseconds), and the percentage of long RR intervals (PNN50, percentage of the number of two adjacent >50 ms normal sinus rhythm RR intervals in the total number of all normal sinus rhythm RR intervals within a specified time in milliseconds).

Frequency domain parameters (expressed as ms 2 /Hz) included the following frequency bands: high-frequency (HF, >0.15 Hz), a band associated with the modulation of vagal tone, primarily breathing; low-frequency (LF, 0.04–0.15 Hz), a band associated with the modulation of baroreflex activity; very low frequency (VLF, 0.0033–0.04 Hz). In addition, the LF:HF ratio was calculated.

Measurement of baroreflex sensitivity

In all dogs, BRS was measured directly after HR variation data had been collected, using phenylephrine as the challenge substance. 15 The femoral artery was catheterized under local anaesthesia with 1% lidocaine on the day before the baroreflex measurement was taken, and the catheter filled with heparin saline and secured. During BRS testing, the catheter was connected to a pressure transducer attached to a multiple channel physiological recorder (Mingograf 7, Siemens-Elema AB, Solna, Sweden). During the test, arterial blood pressure (ABP) and limb lead ECG were simultaneously recorded at a paper speed of 100 mm/s. An initial dose of phenylephrine (5 µg/kg IV) was rapidly injected. If the increase of the ABP was <15 mmHg (1 mmHg = 0.133 kPa), phenylephrine was re-injected after 15 s. Doses of phenylephrine were then increased by 0.5 µg/kg per injection until a dose was reached that increased the artery systolic pressure 15–40 mmHg. Each dose was administered twice.

The change in pressure (mmHg) from the start of the increase in arterial systolic blood pressure until the maximum pressure was attained was the independent X variable. The RR intervals (milliseconds) seen after this increase were set as the dependent Y variables and at least 15 RR intervals were measured in each test. A linear correlation analysis using these data was then carried out. After plotting each arterial systolic blood pressure change and the RR intervals following this change, a regression coefficient ‘ b ’ and intercept ‘ a ’ were calculated. When the linear correlation coefficient ( r ) was greater than 0.8, a regression equation was calculated: Y = a + bX , the average value of three repetitions of the measurement of b was the BRS, and is shown as ms/mmHg.

Measurement of haemodynamic parameters

In all dogs, haemodynamic parameters were measured first in a closed-chest state before pacemaker implantation and again 4–5 weeks later, at the end of the pacing period. For these measurements, dogs were anaesthetized, and the right femoral vein was catheterized for IV access. A Swan-Ganz catheter was passed into the right atrium, right ventricle, and pulmonary artery from the external jugular vein, and right atrial pressure, right ventricular pressure, pulmonary artery pressure, and pulmonary capillary wedge pressure were each measured using a Spectramed P23XL transducer on a Marquette Transcope 12 monitor (Marquette Co., USA).

Cardiac output (CO) was determined using the temperature dilution curve technique. Arterial blood pressure was measured from the femoral artery using a transducer (Spectramed P23XL). The cardiac index (CI, CI = CO/W), stroke volume (SV, SV = CO/HR), and total peripheral resistance (TPR, TPR = mean ABP/CI × 100) were calculated from these data. 16

Measurement of cardiac electrophysiology parameters

For cardiac electrophysiology measurements, the heart was exposed after anaesthesia through a median sternotomy and cradled in the pericardium. For electrical stimulation of the heart, a pair of stainless steel-wire electrodes (0.125 mm in diameter, 5 mm apart) were inserted into the right atrial appendage, using a 22 gauge injection needle. Additional electrode pairs were inserted into the right ventricular outflow tract, the right ventricular lateral wall, the left ventricular anterior wall, and the apex of the heart. A standard lead II surface electrocardiogram, ventricular bipolar electrograms, and ventricular epicardial monophasic action potentials (MAPs) were recorded synchronously using a multiple channel physiological recorder (Mingograf 7) at a paper speed of 100 mm/s. The ventricular MAP was recorded using unipolar contact electrodes (Guangzhou Radio Research Institute, People's Republic of China). The range of the filter wave of the ventricular bipolar electrograms was 50–500 Hz.

Four parameters were derived from these records: VRT-D, ventricular late recovery time duration, the proportion of the action potential taken up by the effective refractory period (VERP/MAP 90 ), and the fibrillation threshold (VT).

Ventricular recovery time dispersion was the difference between the shortest and longest VRT recorded from the different ventricular sites (right ventricular outflow tract, right ventricular anterior wall, left ventricular lateral wall, left ventricular anterior wall, and the apex) at the same cardiac cycle length. Ventricular recovery time was the sum of ventricular activation time and VERP in the same cardiac cycle. 18 Ventricular activation time was the interval from the onset of the standard lead II ECG QRS complex to the point where the rapid deflection of the local ventricular bipolar electrograms crossed the baseline. 19 Ventricular effective refractory period was measured by S 1 S 2 programmed stimulation (Medtronic 5323, Medtronic Co., USA). As the S 1 S 2 interval decreased in 10 ms intervals, scanning was carried out until the S 1 S 2 interval was so short that S 2 was no longer able to evoke ventricular depolarization. Stimulation intensity was two times of the pacing threshold in the diastole, and pulse width was 1.8 ms. Ventricular effective refractory period was defined as the longest S 1 S 2 interval that did not evoke ventricular depolarization. 20

Ventricular late repolarization duration, the portion of the action potential subsequent to the effective refractory period, was measured as the difference between the local ventricular MAP (MAPD 90 ) and the VERP at the same cardiac cycle length. Monophasic action potential duration was the interval, along a line horizontal to the diastolic baseline, from the onset of zero phase depolarization to the 90% repolarization level. 21

The ratio of VERP to MAPD 90 at the same cardiac cycle length was calculated (VERP/MAPD 90 ). The above parameters were measured during atrial or ventricular pacing at 375 ms and 400 ms cardiac cycle length.

Finally, the VFT was determined. Ventricular fibrillation threshold was measured by a train of constant current pulses that scanned the T-wave at a stable atrial paced cycle length of 400 ms. 22 Fibrillation threshold, VFT, was defined as the smallest amount of current required to elicit ventricular fibrillation. Within 15 s after ventricular fibrillation occurred, intrathorax non-synchronous DC cardiac defibrillation was performed at the energy of 5–15 J. Ventricular fibrillation threshold was assessed three times in each animal, and the mean value was calculated. The second and third assessments were done after defibrillation had been achieved and a normal sinus rhythm had returned for at least 15 min.

Gross and pathological examination of the heart

Pericardial effusion, pleural effusion, and the gross changes of the lungs were assessed at the time of thoracotomy. Then, after electrophysiological measurement had been taken, the heart was removed and heart weight, heart weight to body weight ratio, left and right ventricular free wall thickness, left ventricular longitudinal diameter, and right ventricular transverse diameter were each measured, and the left ventricular volume was calculated.

The following definitions were employed for these measurements: 16 left and right ventricular free wall thickness was measured at a point on the free ventricular wall half-way from the atrioventricular valvular ring to the apex of heart; left ventricular longitudinal diameter was the length of the chamber from the left ventricular atrioventricular valvular ring to the apex of heart. Right ventricular transverse diameter was the width of the chamber at half-way from the atrioventricular valvular ring to the apex. Left ventricular volume was (π × left ventricular long diameter × right ventricular transverse diameter) 2 /6.

Fresh left and right ventricular tissues were obtained from each heart and fixed for 12–24 h in a 10% formalin solution. The tissues were then embedded in paraffin and sectioned. The sections were observed under optical microscope using haematoxylin and eosin (H&E) staining.

Statistical analysis

Clinical parameters are expressed as mean and SD and compared using non-parametric methods. The differences between haemodynamic parameters before and after implantation surgery for the same dog were examined using the Wilcoxon Sign Rank test, and comparisons between the CHF and control groups were made using the Wilcoxon Rank Sum test. Pearson's correlation was used to examine correlations between ventricular electrophysiological variables and HRV and BRS. The statistical analyses were performed using SAS 9.0 (SAS Institute Inc., Cary, NC, USA) and the significance level was set at 0.05.

Results

Four to five weeks after initiation of rapid right ventricular pacing, all dogs in the CHF group showed signs of congestive failure, such as anorexia, limb oedema, and hypokinetics. Moist rales could be heard bilaterally in their lungs. In addition, their respiratory rate had increased significantly ( P < 0.01), from 18 ± 1 to 40 ± 3 breaths/min. Body weight, however, did not change (14.3 ± 1.9 vs. 13.8 ± 1.2 kg, P > 0.05).

None of the above changes were seen in the control group. The respiratory rates of dogs in the control and CHF groups at the time of ECG recording for HRV analysis were 17.7 ± 2.2 and 40.3 ± 2.5 breaths/min, respectively ( P < 0.01).

Congestion and oedema in both lungs were seen in all, pericardial effusion in most, and pleural effusion and ascites in some dogs in the CHF group. Gross anatomy of heart in the CHF group showed an enlarged heart: heart weight, heart weight/body weight, left ventricular longitudinal diameter, right ventricular transverse diameter, and left ventricular volume were all significantly increased compared with the control group, and right ventricular free wall thickness was significantly decreased (data not shown). Histological examination of the left and right ventricles in the CHF dogs showed vascular congestion in the endocardium and epicardium and an infiltration of neutrophils and lymphocytes. The myocardial cells showed oedema. Fatty degeneration, increase of cell fusion, and some necrosis and lysis were seen. Interstitial vasculitis with congestion was also observed. Occasionally, haemorrhagic foci could be found with substantial neutrophil and lymphocyte infiltration. Together, these data demonstrate that the model was successful in establishing CHF according to the diagnostic criteria suggested by Armstrong et al . 16

After 4–5 weeks of pacing, mean right atrial and ventricular pressures and pulmonary artery and capillary wedge pressures had significantly increased over pre-pacing values in the CHF group, and CO, CI, and SV were significantly decreased (all P < 0.01, Table  1 ). The control group showed no such changes during this period (data not shown). Total peripheral resistance in the CHF group at the end of pacing was significantly higher than that seen in the control group (78.4 ± 27.8 vs. 36.5 ± 11.4 mmHg min kg/mL, P < 0.01).

Table 1

Impact of 4–5 weeks of right ventricular rapid pacing on various haemodynamic parameters in dogs ( n = 12)

Haemodynamic parameter Pre-ventricular pacing Post-ventricular pacing P -value  
Mean RAP (kPa) 0.189 ± 0.244 1.200 ± 0.461 0.0005 
Mean RVP (kPa) 0.900 ± 0.607 2.000 ± 0.900 0.0015 
Mean PAP (kPa) 1.144 ± 0.782 2.433 ± 1.137 0.0005 
Mean PCWP (kPa) 0.244 ± 0.253 1.188 ± 0.671 0.0005 
CO (L/min) 3.925 ± 0.746 1.425 ± 0.476 0.0005 
CI (L/min/kg) 0.278 ± 0.044 0.120 ± 0.031 0.0005 
SV (mL/beat) 0.025 ± 0.004 0.011 ± 0.002 0.0005 
Haemodynamic parameter Pre-ventricular pacing Post-ventricular pacing P -value  
Mean RAP (kPa) 0.189 ± 0.244 1.200 ± 0.461 0.0005 
Mean RVP (kPa) 0.900 ± 0.607 2.000 ± 0.900 0.0015 
Mean PAP (kPa) 1.144 ± 0.782 2.433 ± 1.137 0.0005 
Mean PCWP (kPa) 0.244 ± 0.253 1.188 ± 0.671 0.0005 
CO (L/min) 3.925 ± 0.746 1.425 ± 0.476 0.0005 
CI (L/min/kg) 0.278 ± 0.044 0.120 ± 0.031 0.0005 
SV (mL/beat) 0.025 ± 0.004 0.011 ± 0.002 0.0005 

RAP, right atrial pressure; RVP, right ventricular pressure; PAP, pulmonary artery pressure; PCWP, pulmonary capillary wedge pressure; CO, cardiac output; CI, cardiac index; SV, stroke volume; TPR, total peripheral resistance.

The CHF group, in addition to showing clear signs of congestive failure after pacing (see above), showed significant decreases in the two markers of autonomic function measured, HRV and BRS. Both the time and frequency domains of HRV were significantly lower, although LF/HF ratio was not ( Table  2 ). Baroreflex sensitivity also decreased to 8.13 ± 2.50 ms/mmHg from a pre-pacing value of 15.85 ± 2.13 ms/mmHg ( P < 0.01). Graphs of BRS are shown in Figure  1 . No such decreases were seen in the control group.

Figure 1

Baroreflex sensitivity graph for congestive heart failure and normal dogs, respectively.

Figure 1

Baroreflex sensitivity graph for congestive heart failure and normal dogs, respectively.

Table 2

Evaluation of heart rate variability time and frequency domain parameters 4–5 weeks following sham operation (control) or rapid ventricular pacing (congestive heart failure)

Time or frequency domain parameters  Control ( n = 10)   CHF ( n = 12)  P -value  
SDNN (ms) 83.000 ± 15.699 52.333 ± 14.549 0.0025 
rMSSD (ms) 41.600 ± 16.290 22.333 ± 7.426 0.0049 
PNN50 (%) 9.330 ± 5.287 5.350 ± 2.616 0.0253 
VLF (ms 2 /Hz)  28.400 ± 9.192 17.250 ± 8.291 0.0231 
LF (ms 2 /Hz)  23.500 ± 5.893 10.750 ± 4.750 0.0016 
HF (ms 2 /Hz)  15.400 ± 4.526 6.250 ± 2.598 0.0009 
LF/HF 1.555 ± 0.199 1.815 ± 0.625 0.2008 
Time or frequency domain parameters  Control ( n = 10)   CHF ( n = 12)  P -value  
SDNN (ms) 83.000 ± 15.699 52.333 ± 14.549 0.0025 
rMSSD (ms) 41.600 ± 16.290 22.333 ± 7.426 0.0049 
PNN50 (%) 9.330 ± 5.287 5.350 ± 2.616 0.0253 
VLF (ms 2 /Hz)  28.400 ± 9.192 17.250 ± 8.291 0.0231 
LF (ms 2 /Hz)  23.500 ± 5.893 10.750 ± 4.750 0.0016 
HF (ms 2 /Hz)  15.400 ± 4.526 6.250 ± 2.598 0.0009 
LF/HF 1.555 ± 0.199 1.815 ± 0.625 0.2008 

CHF, congestive heart failure; SDNN, standard deviation of all normal sinus rhythm RR intervals within the specified time; rMSSD, root-mean-square standard deviation of the continuous normal sinus rhythm RR intervals within the specified time; PNN50, percentage of number of every two adjacent >50 ms normal sinus rhythm RR intervals in the total number of all normal sinus rhythm RR intervals within the specified time; VLF, very low-frequency band; LF, low-frequency band; HF, high-frequency band; LF/HF, ratio of low- to high-frequency bands.

In the myocardium itself, increase in action potential duration and effective refractory period in the ventricles accompanied the post-pacing signs of congestive failure and decrease in autonomic sensitivity seen in CHF dogs ( Table  3 ). Monophasic action potential duration and VERP were 45 and 26% longer (respectively) in the CHF groups than in the control group ( P < 0.01).

Table 3

Changes in ventricular electrophysiological parameters 4–5 weeks after rapid ventricular pacing (congestive heart failure) or sham operation (control)

Ventricular electrophysiological parameter  Control ( n = 10)   CHF ( n = 12)  P -value  
VERP (ms) 169.00 ± 11.254 212.916 ± 38.106 0.0374 
MAPD 90 (ms)  179.00 ± 18.678 259.166 ± 43.632 0.0039 
VRT-D (ms) 13.800 ± 4.289 37.333 ± 14.163 0.0021 
VLRD (ms) 10.000 ± 14.142 46.250 ± 35.170 0.0247 
VERP/MAPD 90 0.949 ± 0.076 0.831 ± 0.134 0.0357 
VFT (mA) 33.750 ± 6.559 11.541 ± 8.993 0.0011 
Ventricular electrophysiological parameter  Control ( n = 10)   CHF ( n = 12)  P -value  
VERP (ms) 169.00 ± 11.254 212.916 ± 38.106 0.0374 
MAPD 90 (ms)  179.00 ± 18.678 259.166 ± 43.632 0.0039 
VRT-D (ms) 13.800 ± 4.289 37.333 ± 14.163 0.0021 
VLRD (ms) 10.000 ± 14.142 46.250 ± 35.170 0.0247 
VERP/MAPD 90 0.949 ± 0.076 0.831 ± 0.134 0.0357 
VFT (mA) 33.750 ± 6.559 11.541 ± 8.993 0.0011 

Values are expressed as mean ± SD. CHF, congestive heart failure; VERP, ventricular effective refractory period; MAPD 90 , the interval, along a line horizontal to the diastolic baseline, from the onset of zero phase depolarization to the 90% repolarization level; VRT-D, ventricular recovery time dispersion; VLRD, ventricular late repolarization duration; VERP/MAPD 90 , the ratio of ventricular effective refractory period to the interval, along a line horizontal to the diastolic baseline, from the onset of zero phase depolarization to the 90% repolarization level; VFT, ventricular fibrillation threshold.

Action potentials became less uniform as well as longer. Variation in action potential recovery time, VRT-D, increased 164% ( P < 0.01) over control group values. Ventricular late repolarization duration, the portion of the action potential subsequent to the effective refractory period, was 360% longer ( P < 0.05). Because VLRD occupied a longer portion of the total action potential duration, the effective refractory period occupied a shorter fraction, and the VERP/MAPD 90 ratio was smaller in the CHF group than in the control group ( P < 0.05).

Ventricular fibrillation was more easily produced in CHF dogs than in controls, and the threshold for fibrillation, VFT, was significantly lower ( P < 0.01).

In CHF dogs, HRV was negatively correlated with VLRD and VRT-D ( r = −0.861 to −0.579, P < 0.01 and P < 0.04, respectively, Table  4 ), that is, the lower the HRV, the greater the late repolarization duration, and the more variable the action potential recovery time in the ventricles. Heart rate variability was positively correlated with VERP/MAPD90 and VFT ( r = 0.626–0.854, P < 0.05 and P < 0.01, respectively, Table  4 ), that is, the lower the HRV, the smaller the fraction of the action potential occupied by the effective refractory period.

Table 4

Summary of the Pearson's correlations of ventricular electrophysiological variables and heart rate variability and baroreflex sensitivity in the congestive heart failure group of dogs after 4–5 weeks of rapid ventricular pacing

  HRV
 
BRS 
 SDNN rMSSD PNN50 VLF LF HF  
VRT-D −0.61* −0.64* −0.58* −0.59* −0.60* −0.80** −0.86** 
VLRD −0.86** −0.85** −0.73** −0.84** −0.59* −0.75** −0.78** 
VERP/MAPD 90 0.85** 0.80** 0.71* 0.83** 0.59* 0.72** 0.75** 
VFT 0.67* 0.73** 0.66* 0.63* 0.43* 0.69* 0.78** 
BRS 0.63* 0.75** 0.62* 0.73* 0.60* 0.81** — 
  HRV
 
BRS 
 SDNN rMSSD PNN50 VLF LF HF  
VRT-D −0.61* −0.64* −0.58* −0.59* −0.60* −0.80** −0.86** 
VLRD −0.86** −0.85** −0.73** −0.84** −0.59* −0.75** −0.78** 
VERP/MAPD 90 0.85** 0.80** 0.71* 0.83** 0.59* 0.72** 0.75** 
VFT 0.67* 0.73** 0.66* 0.63* 0.43* 0.69* 0.78** 
BRS 0.63* 0.75** 0.62* 0.73* 0.60* 0.81** — 

HRV, heart rate variability; BRS, baroreflex sensitivity; VRT-D, ventricular recovery time dispersion; VLRD, ventricular late repolarization duration; VERP/MAPD 90 , the ratio of ventricular effective refractory period to the interval, along a line horizontal to the diastolic baseline, from the onset of zero phase depolarization to the 90% repolarization level; VFT, ventricular fibrillation threshold; SDNN, standard deviation of all normal sinus rhythm RR intervals within the specified time; rMSSD, root-mean-square standard deviation of the continuous normal sinus rhythm RR intervals within the specified time; PNN50, percentage of number of every two adjacent >50 ms normal sinus rhythm RR intervals in the total number of all normal sinus rhythm RR intervals within specified time; VLF, very low-frequency band; LF, low-frequency band; HF, high-frequency band.

* P < 0.05, ** P < 0.01.

Like HRV, BRS, the other marker of autonomic function, was negatively correlated with VRT-D and VLRD ( r = −0.861 and 0.781, respectively, P < 0.01) and was positively correlated with VERP/MAPD 90 and VFT ( r = 0.734 and 0.777, respectively, P < 0.01), that is, the less sensitive the autonomic nervous system to blood pressure change, the more variable the recovery time of the action potential, the longer the post-effective refractory period portion of the action potential, and the smaller the electrical stimulus needed to evoke ventricular fibrillation. In the control group, no correlations between either HRV or BRS and these ventricular electrophysiological parameters were found. Representative correlation plots are shown in Figure  2 . Baroreflex sensitivity was also positively correlated to the time and frequency parameters of HRV, that is, SDNN, rMSSD, PNN50, VLF, LF, and HF ( r = 0.603–0.814, P < 0.05 and P < 0.01) ( Table  4 ).

Figure 2

Representative plots of the correlation between baroreflex sensitivity and ventricular fibrillation threshold ( A ), and between SDNN and ventricular effective refractory period/monophasic action potential duration ( B ).

Figure 2

Representative plots of the correlation between baroreflex sensitivity and ventricular fibrillation threshold ( A ), and between SDNN and ventricular effective refractory period/monophasic action potential duration ( B ).

Ventricular fibrillation threshold was negatively correlated to action potential recovery variability (VRT-D, r = −0.579, P < 0.05) and to late repolarization duration (VLRD, r = −0.749, P < 0.01), and positively correlated to VERP/MAPD 90 ( r = 0.674, P < 0.05). In other words, ease of inducing fibrillation was linked to lower HRV, and shorter effective refractory portions and longer late refractory portions of the total action potential.

Discussion

Our pacing protocol was successful in establishing CHF, as has been previously reported by others. 23 Dogs in subjected to 4–5 weeks of rapid right ventricular pacing had decreased activity levels, oedema, increased intracardial pressures, reduced CO, and cardiac dilatation. Changes in clinical manifestations, haemodynamic, and cardiac pathology in these dogs were consistent with those reported for CHF in the literature. 16 , 22–25

The present study showed that both effective refractory period and action potential duration were increased in dogs with pacing-induced CHF compared with control dogs while the VERP/MAP 90 was decreased. This latter finding suggests that the increase in action potential duration was more significantly prolonged than the increase in effective refractory period in CHF.

The VLRD (the portion of the action potential that includes the relatively refractory period, vulnerable period, and super normal period) was significantly prolonged in CHF dogs. The VLRD prolongation can produce temporary conduction block in a portion of the myocardium, and, in this way, enable re-entry to occur. In addition, a prolonged VLRD or incomplete repolarization can result in early after-depolarizations which, if large enough, can become action potentials and cause triggered arrhythmias. Therefore, VLRD prolongation is thought to facilitate the genesis of ventricular arrhythmias. This hypothesis is further supported by the fact that early after-depolarizations have been recorded in both animal models of CHF and in isolated myocytes from human patients with heart failure. 26 , 27

The present study also found increased dispersion in action potential recovery time in the ventricles of CHF dogs. Myocardial cells with long recovery times are thought to be able to form a unidirectional block, and when electrical signals are conducted from myocardium with a short VRT to the myocardium with a long VRT, re-entry may occur and may result in the occurrence of sustained ventricular arrhythmias. 19

Bai et al . 28 have previously reported that the incidence of ventricular fibrillation is increased in dogs with CHF induced by rapid right ventricular pacing. In our study, the fibrillation threshold was lower in CHF than in control dogs. In addition, VRT-D and VLRD were significantly and negatively correlated with VFT, and VERP/MAPD 90 was positively correlated with VFT. Together, our findings suggest that CHF-induced alterations in the electrophysiological properties of the myocytes lead to electrical instability and arrhythmia.

It has been suggested that significant reductions in HRV, and BRS that reflect autonomic nerve function, could be used to predict sudden cardiac death. 7–10 Prior researches have found that abnormal distribution of autonomic nerves innervating the ventricular muscles, excessive activation of the sympathetic nerve, and heterogeneity of sympathetic innervation of ventricular myocytes may result in uneven ventricular repolarization, an increase in dispersion of repolarization, and an increase in ventricular electric instability so that ventricular fibrillation is easy to induce. 29 , 30 Thus, indirect measurement of autonomic nervous system function via HRV and/or BRS may serve as a means of identifying CHF patients at risk for the development of cardiac arrhythmias and cardiac sudden death. As in previously reported studies, 8 , 9 it was found that HRV and BRS were both significantly lower in the CHF group compared with the control group, and that a correlation between BRS and HRV exists. In the present study, we also demonstrated that the time domain parameters of HRV and BRS were negatively correlated with electrophysiological parameters of VRT-D and VLRD and were positively correlated with VERP/MAPD 90 and VFT. The fact that decreases in the two autonomic markers were associated with increased variability in ventricular action potential recovery times and post-ERP portions of the action potential is certainly interesting and needs to be explored further. The above results together show changes in autonomic control to be associated with changes in the electrophysiology of myocytes that are associated with susceptibility to fibrillation. Whether changes in the autonomic markers used here are sensitive and specific enough to be of clinical use in predicting risk of sudden death in CHF remains to be seen.

In summary, changes in ventricular electrophysiological characteristics in CHF result in an unstable ventricular electric activities, in which variable action potential recovery times and long action potential late repolarization times can cause malignant ventricular arrhythmias. Heart rate variability and BRS are both significantly decreased in dogs with CHF and this decrease is closely correlated to repolarization parameters and with fibrillation threshold. These results support the contention that a relationship exists between autonomic nervous system function and sudden cardiac death, and that alterations in autonomic nervous system function correlate with ventricular electrophysiological characteristics. It is hoped that this and subsequent research will assist in the development of ways to minimize the development of cardiac arrhythmias and sudden cardiac death in patients with CHF.

Conflict of interest: none declared.

Funding

This work was supported by a grant from Sun Yat-Sen University.

References

1
Deedwania
PC
Ventricular arrhythmias in heart failure: to treat or not to treat
Cardiol Clin
 , 
1994
, vol. 
12
 (pg. 
115
-
32
)
2
Pye
MP
Cobbe
SM
Mechanisms of ventricular arrhythmias in cardiac failure and hypertrophy
Cardiovasc Res
 , 
1992
, vol. 
26
 (pg. 
740
-
50
)
3
Roden
DM
A surprising new arrhythmia mechanism in heart failure
Circ Res
 , 
2003
, vol. 
93
 (pg. 
589
-
91
)
4
Li
HG
Jones
DL
Yee
R
Klein
GJ
Electrophysiologic substrate associated with pacing-induced heart failure in dogs: potential value of programmed stimulation in predicting sudden death
J Am Coll Cardiol
 , 
1992
, vol. 
19
 (pg. 
444
-
9
)
5
Akar
FG
Rosenbaum
DS
Transmural electrophysiological heterogeneities underlying arrhythmogenesis in heart failure
Circ Res
 , 
2003
, vol. 
93
 (pg. 
638
-
45
)
6
Guo
J
Wang
R-Y
Wang
H-Y
Wang
H-X
Change and prognostic value of HRT in CHF patients [Chinese]
Chin J Cardiovasc Med
 , 
2006
, vol. 
11
 (pg. 
425
-
8
)
7
Cygankiewicz
I
Wranicz
JK
Bolinska
H
Zaslonka
J
Zareba
W
Relationship between heart rate turbulence and heart rate, heart rate variability, and number of ventricular premature beats in coronary patients
J Cardiovasc Electrophysiol
 , 
2004
, vol. 
15
 (pg. 
731
-
7
)
8
La Rovere
MT
Pinna
GD
Hohnloser
SH
Marcus
FI
Mortara
A
Nohara
R
, et al.  . 
ATRAMI Investigators
Autonomic tone and reflexes after myocardial infarction. Baroreflex sensitivity and heart rate variability in the identification of patients at risk for life-threatening arrhythmias: implications for clinical trails
Circulation
 , 
2001
, vol. 
103
 (pg. 
2072
-
7
)
9
La Rovere
MT
Bigger
JT
Jr
Marcus
FI
Mortara
A
Schwartz
PJ
Baroreflex sensitivity and heart-rate variability in prediction of total cardiac mortality after myocardial infarction. ATRAMI (Autonomic Tone and Reflexes After Myocardial Infarction) Investigators
Lancet
 , 
1998
, vol. 
351
 (pg. 
478
-
4
)
10
De Ferrari
GM
Sanzo
A
Bertoletti
A
Specchia
G
Vanoli
E
Schwartz
PJ
Baroreflex sensitivity predicts long-term cardiovascular mortality after myocardial infarction even in patients with preserved left ventricular function
Am Col Cardiol
 , 
2007
, vol. 
50
 (pg. 
2285
-
90
)
11
Kleiger
RE
Miller
JP
Bigger
J
Moss
AJ
Decreased heart rate variability and its association with increased mortality after acute myocardial infarction
Am J Cardiol
 , 
1987
, vol. 
59
 (pg. 
256
-
62
)
12
Bigger
JT
Fleiss
J
Steinman
RC
Rolnitzky
JM
Kleiger
RE
Rottman
JN
Frequency domain measures of heart period variability and mortality after myocardial infarction
Circulation
 , 
1992
, vol. 
85
 (pg. 
164
-
71
)
13
Tsuji
H
Larson
MG
Venditti
FJ
Jr
Maders
ES
Evans
JC
Feldman
CL
, et al.  . 
Impact of reduced heart rate variability on risk for cardiac events. The Farmingham Heart Study
Circulation
 , 
1996
, vol. 
94
 (pg. 
2850
-
5
)
14
Jioliro
HV
Makikallio
TH
Peng
K
Goldberger
AL
Hintze
U
Moller
M
Fractal correlation properties of the RR interval dynamics and mortality in patients with depressed left ventricular function after myocardial infarction
Circulation
 , 
2000
, vol. 
101
 (pg. 
47
-
53
)
15
Schwartz
PJ
Vanoli
E
Stramba-Badiale
M
De Ferrari
GM
Billman
GE
Foreman
RD
Autonomic mechanisms and sudden death: new insights from analysis of baroreceptor reflexes in conscious dogs with and without a myocardial infarction
Circulation
 , 
1988
, vol. 
78
 (pg. 
969
-
79
)
16
Armstrong
PW
Stopps
TP
Ford
SE
de Bold
AJ
Rapid ventricular pacing in the dog: pathophysiologic studies of heart failure
Circulation
 , 
1986
, vol. 
74
 (pg. 
1075
-
84
)
17
Warltier
DC
Zyvoloski
MG
Gross
GJ
Brooks
HL
Comparative actions of dihydropyridine slow channel calcium blocking agents in conscious dogs: alterations in baroreflex sensitivity
J Pharmacol Exp Ther
 , 
1984
, vol. 
230
 (pg. 
376
-
82
)
18
Williams
RE
Kass
DA
Kawagee
Y
Pak
P
Tunin
RS
Shah
R
, et al.  . 
Endomyocardial gene expression during development of pacing tachycardia-induced heart failure in the dog
Circ Res
 , 
1994
, vol. 
75
 (pg. 
615
-
23
)
19
Vassallo
JA
Cassidy
DM
Kindwall
KE
Marchlinski
FE
Josephson
ME
Nonuniform recovery of excitability in the left ventricle
Circulation
 , 
1988
, vol. 
78
 (pg. 
1365
-
72
)
20
Fisher
JD
Role of electrophysiologic testing in the diagnosis and treatment of patients with known and suspected bradycardias and tachycardias
Prog Cardiovasc Dis
 , 
1981
, vol. 
24
 (pg. 
25
-
90
)
21
Franz
MR
Method and theory of monophasic action potential recording
Prog Cardiovasc Dis
 , 
1991
, vol. 
33
 (pg. 
347
-
68
)
22
Lu
F
Zhang
XM
Mei
BY
Effects of propafenone, quinidine, and their combination on ventricular fibrillation threshold in dogs
Acta Pharmacol Sin
 , 
1992
, vol. 
13
 (pg. 
364
-
7
)
23
Zhou
S-X
Zhang
X-M
Wu
W
Chen
X-C
Effects of amiodarone on cardiac electrophysiology of right ventricular rapid pacing-induced heart failure dogs
Acta Pharmacol Sin
 , 
1998
, vol. 
19
 (pg. 
363
-
8
)
24
Watanabe
K
Kuroda
H
Sato
E
Makino
H
In vivo evaluation of the improved MCMS-0102 pacemaker with a rapid pacing mode for induction of experimental heart failure in animals
J Artif Organs
 , 
2006
, vol. 
9
 (pg. 
84
-
9
)
25
Nishijima
Y
Feldman
DS
Bonagura
JD
Ozkanlar
Y
Jenkins
PJ
Lacombe
VA
, et al.  . 
Canine nonischemic left ventricular dysfunction: a model of chronic human cardiomyopathy
J Card Fail
 , 
2005
, vol. 
11
 (pg. 
638
-
44
)
26
Riggio
DW
Peters
RW
Feliciano
Z
Gottlieb
SS
Shorofsky
SR
Gold
MR
Acute electrophysiologic effects of amiodarone in patients with congestive heart failure
Am J Cardiol
 , 
1995
, vol. 
75
 (pg. 
1158
-
61
)
27
Janse
MJ
Electrophysiological changes in heart failure and their relationship to arrhythmogenesis
Cardiovasc Res
 , 
2004
, vol. 
61
 (pg. 
208
-
17
)
28
Bai
R
Pu
J
Liu
N
Lu
JG
Zhou
Q
Ruan
YF
, et al.  . 
Influence of pacing site on myocardial transmural dispersion of repolarization in intact normal and dilated cardiomyopathy dogs [Chinese]
Acta Physiol Sin
 , 
2003
, vol. 
55
 (pg. 
722
-
30
)
29
Liu
YB
Wu
CC
Lu
LS
Su
MJ
Lin
CW
Lin
SF
, et al.  . 
Sympathetic nerve sprouting, electrical remodeling, and increased vulnerability to ventricular fibrillation in hypercholesterolemic rabbits
Circ Res
 , 
2003
, vol. 
92
 (pg. 
1145
-
52
)
30
Li
J-J
Qu
X-F
Yue
L
Xi
Y
Gu
H-Y
Wang
G-Z
, et al.  . 
Regional denervation after myocardial infarction and its effect on ventricular repolarization [Chinese]
Natl Med J Chin
 , 
2006
, vol. 
86
 (pg. 
98
-
101
)