Abstract
This study was designed to evaluate the effect of variation of atrioventricular (AV) interval (AVI) on left ventricular (LV) diastolic function and ANP and c-GMP levels during DDD pacing in patients with complete AV block and normal systolic function.
The study population comprised 22 patients (mean age 65.2 ± 14.3, 12 males) with complete AV block. All patients underwent complete Doppler echocardiography before implantation of a DDD-pacemaker. Twenty-four hours later, patients were paced for a period of 30 min, at three different AVIs (100 ms, 150 ms and 200 ms), at rest. During each pacing period, Doppler-derived LV diastolic indices were re-evaluated and ANP and c-GMP levels were reassessed.
Overall comparison showed a significant progressive augmentation, from 200 ms to 100 ms AVI, in transmitral E/A wave ratio (from 0.53 ± 0.13 to 0.90 ± 0.25, P = 0.0005) and in LV filling time (from 0.33 ± 0.05 to 0.40 ± 0.06 s, P = 0.0005), followed by a significant progressive reduction in ANP and c-GMP levels. An AVI of 100 ms or 150 ms was associated with improved diastolic indices and lower natriuretic peptides levels, compared with the longer AVI.
Programmed AVI during DDD pacing affects LV diastolic performance and plasma ANP and c-GMP levels. The assessment of these parameters constitutes a useful modality for AVI optimization.
Introduction
Dual chamber (DDD) pacing with a short atrioventricular (AV) interval (AVI) may improve cardiac function in patients with left ventricular (LV) systolic heart failure, as has been suggested by preliminary observations[1–,3]. However, comprehensive data on how permanent DDD pacing affects cardiac function in patients with LV diastolic dysfunction but preserved LV systolic function are currently incomplete.
Atrial natriuretic peptide (ANP) is a potent vasodilator, diuretic and natriuretic substance, secreted by atrial myoendocrine cells in response to atrial distention in several conditions, such as heart failure, mitral stenosis, arrhythmias and cardiac pacing [4–,9]. Additionally, ANP represents an extremely important biochemical marker of global cardiac performance in patients in whom a pacemaker has been implanted[10,,11]. Furthermore, cyclic guanosine 3,5-monophosphate (c-GMP) which is the intracellular second messenger of ANP shares a well documented significant correlation with ANP plasma levels in various types of heart disease[12,,13]. The value of the natriuretic peptide family for the diagnosis and follow-up of heart failure and their predictive power in respect of morbidity and mortality are well established. Recent data suggest that these peptides are strong predictors of the risk of death and cardiovascular events even in the absence of heart failure[14]. Additionally, it has been demonstrated that in patients with normal systolic function N-terminal ANP levels predicted morbidity and mortality even better than ejection fraction[15].
Doppler echocardiography is a well established modality for the evaluation of LV diastolic filling patterns and thus, indirectly, for the noninvasive assessment of LV diastolic function in health and disease[16–,18]. Additionally, it has been established that echo-Doppler measurements have a relatively good correlation with invasive haemodynamic measurements[19,,20]. Consequently, Doppler echocardiography has been widely used for the evaluation of systolic and diastolic cardiac function in various types of pacing and remains, to date, the standard method of assessment of optimal AVI that provides the longest LV filling time or the highest cardiac output in DDD pacing mode[21–,23].
Diastolic performance is critical for normal global myocardial function and is impaired in patients with complete AV block. In this specific cohort of patients with normal systolic function but disturbed diastolic performance needing DDD pacing, AVI optimization has to be focused on improvement in diastolic filling dynamics. The present study was designed to evaluate the different effects of variable AVI on LV diastolic indices and plasma ANP and c-GMP concentrations, assessing indirectly the effectiveness of the combination of these two modalities in optimal AVI selection.
Methods
Study patients
This prospective study included 22 consecutive patients (mean age 65.2 ± 14.3, range: 25–80 years, 12 males) with complete AV block, admitted because of pre-syncopal or syncopal episodes. All patients had a normal chest radiograph and normal basic laboratory biochemical markers. None of the patients had a history or clinical signs or ECG evidence of coronary artery or valvular disease, cardiomyopathy, heart failure, diabetes mellitus, pericardial or lung disease, anaemia or other systemic diseases.
All patients were in NYHA class I and underwent a dual chamber pacemaker implantation following detection of third degree AV block. Atrial and ventricular leads were positioned in the right atrial appendage and in the right ventricular apex, respectively. All procedures were uneventful and patients remained in the Cardiac Care Unit for at least 12 h of surveillance.
The aim of the study was to evaluate the effect of different AVIs during DDD pacing on LV diastolic filling properties of patients with impaired LV diastolic function due to complete heart block and to observe the effect of each individual's AVI on plasma natriuretic peptide levels.
Pacing protocol
All patients were assessed on the day following pacemaker implantation (24 h). They were examined in the supine position and the pacemaker was programmed to a rate of 80 beats/min in order to ensure sequential AV pacing. Patients were then paced for three successive pacing periods of 30 min duration, using three AVIs (100 ms, 150 ms and 200 ms) in a randomized fashion. During each specific pacing period, LV diastolic performance was evaluated by echo-Doppler and in the last 60 s of each period a blood sample was drawn for ANP and c-GMP assays. Blood pressure was evaluated before and at the end of each pacing period in order to secure diastolic function assessment under the same haemodynamic conditions.
The data collected were categorized into three groups according to the specific AVI of DDD pacing during which they were obtained. In order to evaluate the effect of each specific AVI on LV diastolic filling dynamics and natriuretic peptide levels, we performed a series of statistical analyses comparing the data (endpoints) obtained with the different AVIs.
Echocardiography
All patients underwent complete transthoracic echocardiography and Doppler study 2 h before the pacemaker was implanted. The study was repeated 24 h later (before pacing protocol initiation) and at the 1-year follow-up. Additionally, Doppler indices of LV diastolic performance were recorded during the AVIs according to the methodology of the study (as per the pacing protocol).
Transthoracic M-mode, 2-D and spectral Doppler (pulsed and continuous wave) echocardiographic studies were performed with a SIGMA “IRIS” apparatus (Kontron Instruments, France) equipped with 2.8–3.5 MHz transducers. Standard M-mode measurements were obtained from the left parasternal long axis view according to the recommendations of the American Society of Echocardiography[24]. The following parameters were calculated: aortic root dimension, left atrial size, left ventricular end-diastolic diameter, left ventricular end-systolic diameter and left ventricular fractional shortening. LV diastolic indices were accessed from the apical four-chamber view by selecting a 2–4 mm sized sample volume at the tips of mitral leaflets during diastole. The following LV diastolic indices were calculated: peak velocity of E wave, representing early filling; peak velocity of A wave, representing late filling; ratio of peak early to peak late velocity (E/A); deceleration time of E wave; and finally LV filling time. Furthermore, from the apical five-chamber view and by selecting a 2–4 mm sized sample volume in the LV outflow tract, the LV outflow tract velocity time integral was assessed (LVOT VTI). Additionally, from the parasternal short axis view the pulmonary artery velocity time integral (PA VTI) was calculated by selecting the Doppler sample volume in the pulmonary artery close to the pulmonary valve. All Doppler indices were measured from six consecutive beats and their values were averaged.
All echocardiographic studies were recorded on SVHS videotape and were analyzed offline by two experienced operators, blinded to the clinical data. Intraobserver variability was established by having one observer measuring echocardiographic data on at least two occasions in 10 subjects selected at random from the patient population under study (r = 0.94). Interobserver variability was determined by having a second operator independently measuring the same parameters in these subjects (r = 0.89).
ANP and c-GMP assessment
Both natriuretic peptide measurements were performed at the end of each pacing period of AVI according to the study methodology. Patients were resting in the supine position and blood samples via the brachial vein were drawn for the specific assays.
ANP measurement was performed using a radioimmunoassay method, specific for human ANP. Quantification of ANP was achieved using a non-equilibrium radioimmunoassay with delayed addition of the iodinated tracer. The bound and free hormones were separated by the second antibody/polyethylene glycol system. Centrifugation separated the bound and free hormone fractions. The supernatant, which contained the bound fraction, was counted. The final concentration of ANP was calculated and expressed in picograms/ml plasma (pg/ml).
The c-GMP level measurement was accomplished using a radioimmunoassay method, based on the competition between the succinylated c-GMP of the sample and an 125I-labelled tracer for binding with polyclonal antibody coated in the tubes. The final concentration of c-GMP was expressed in picomoles/ml plasma (pmol/ml).
Of the total 22 patients, 20 were discharged on DDD pacing mode with an AVI of 100 ms, while in two an AVI of 150 ms, was used due to a more favourable echocardiographic diastolic filling performance. All patients were followed for 1 year in the outpatient pacemaker clinic of our Department being evaluated for major cardiovascular events and pacemaker function. Patients were re-evaluated by echocardiography in 1 year following implantation according to the study methodology.
Informed consent was obtained from all patients in the study population. The study complied with the Declaration of Helsinki and was approved by the Institutional Committee on Human Research of our hospital.
Statistics
Results are expressed as mean ± SD. For all variables (except deceleration time of E wave) ANOVA of repeated measurements (multivariate approach) was performed. For paired comparisons “t-test” was used, verified by the “Tukey's honest significant difference”. A linear regression model was used in order to correlate ANP values with echocardiographic data. Values of P < 0.05 were considered to be statistically significant.
Results
The mean heart rate and the baseline echocardiographic data of the study population are presented in Table 1.
Heart rate and echocardiographic data in patients with DDD pacing before, after and 1 year following pacemaker implantation
| Variable | Baseline (N = 22) | 24 h (N = 22) | 1 year (N = 18) |
| Heart rate (bpm) | 41 ± 4 | 78 ± 8a | 73 ± 5b |
| AO (cm) | 3.40 ± 0.28 | 3.41 ± 0.27 | 3.40 ± 0.28 |
| LA (cm) | 3.80 ± 0.30 | 3.72 ± 0.26a | 3.72 ± 0.24b |
| LVEDD (cm) | 5.77 ± 0.45 | 5.64 ± 0.38a | 5.64 ± 0.35b |
| LVESD (cm) | 3.63 ± 0.42 | 3.55 ± 0.39 | 3.57 ± 0.36 |
| LVFS (%) | 37.23 ± 3.75 | 37.27 ± 3.72 | 36.88 ± 3.52 |
| LVOT VTI | NA | 24.28 ± 3.16c | 25.65 ± 2.36 |
| PA VTI | NA | 18.47 ± 2.57c | 19.32 ± 2.12 |
| MV-E/A | NA | 0.57 ± 0.22 | 0.65 ± 0.26 |
| MV-DTE (s) | NA | 0.149 ± 0.03 | 0.150 ± 0.02 |
| Variable | Baseline (N = 22) | 24 h (N = 22) | 1 year (N = 18) |
| Heart rate (bpm) | 41 ± 4 | 78 ± 8a | 73 ± 5b |
| AO (cm) | 3.40 ± 0.28 | 3.41 ± 0.27 | 3.40 ± 0.28 |
| LA (cm) | 3.80 ± 0.30 | 3.72 ± 0.26a | 3.72 ± 0.24b |
| LVEDD (cm) | 5.77 ± 0.45 | 5.64 ± 0.38a | 5.64 ± 0.35b |
| LVESD (cm) | 3.63 ± 0.42 | 3.55 ± 0.39 | 3.57 ± 0.36 |
| LVFS (%) | 37.23 ± 3.75 | 37.27 ± 3.72 | 36.88 ± 3.52 |
| LVOT VTI | NA | 24.28 ± 3.16c | 25.65 ± 2.36 |
| PA VTI | NA | 18.47 ± 2.57c | 19.32 ± 2.12 |
| MV-E/A | NA | 0.57 ± 0.22 | 0.65 ± 0.26 |
| MV-DTE (s) | NA | 0.149 ± 0.03 | 0.150 ± 0.02 |
AO: Aortic root dimension, LA: left atrium size, LVEDD: left ventricular end-diastolic diameter, LVESD: left ventricular end-systolic diameter, LVFS: left ventricular fractional shortening, LVOT VTI: left ventricular outflow tract velocity time integral, PA VTI: pulmonary artery velocity time integral, MV-E/A: E/A ratio, MV-DTE: deceleration time, NA: not applicable, MV = mitral value.
aStatistically significant difference (P < 0.05), between baseline and 24 h post implantation.
bStatistically significant difference (P < 0.05), between baseline and 1 year post implantation.
cStatistically significant difference (P < 0.05), between 24 h and 1 year post implantation.
Heart rate and echocardiographic data in patients with DDD pacing before, after and 1 year following pacemaker implantation
| Variable | Baseline (N = 22) | 24 h (N = 22) | 1 year (N = 18) |
| Heart rate (bpm) | 41 ± 4 | 78 ± 8a | 73 ± 5b |
| AO (cm) | 3.40 ± 0.28 | 3.41 ± 0.27 | 3.40 ± 0.28 |
| LA (cm) | 3.80 ± 0.30 | 3.72 ± 0.26a | 3.72 ± 0.24b |
| LVEDD (cm) | 5.77 ± 0.45 | 5.64 ± 0.38a | 5.64 ± 0.35b |
| LVESD (cm) | 3.63 ± 0.42 | 3.55 ± 0.39 | 3.57 ± 0.36 |
| LVFS (%) | 37.23 ± 3.75 | 37.27 ± 3.72 | 36.88 ± 3.52 |
| LVOT VTI | NA | 24.28 ± 3.16c | 25.65 ± 2.36 |
| PA VTI | NA | 18.47 ± 2.57c | 19.32 ± 2.12 |
| MV-E/A | NA | 0.57 ± 0.22 | 0.65 ± 0.26 |
| MV-DTE (s) | NA | 0.149 ± 0.03 | 0.150 ± 0.02 |
| Variable | Baseline (N = 22) | 24 h (N = 22) | 1 year (N = 18) |
| Heart rate (bpm) | 41 ± 4 | 78 ± 8a | 73 ± 5b |
| AO (cm) | 3.40 ± 0.28 | 3.41 ± 0.27 | 3.40 ± 0.28 |
| LA (cm) | 3.80 ± 0.30 | 3.72 ± 0.26a | 3.72 ± 0.24b |
| LVEDD (cm) | 5.77 ± 0.45 | 5.64 ± 0.38a | 5.64 ± 0.35b |
| LVESD (cm) | 3.63 ± 0.42 | 3.55 ± 0.39 | 3.57 ± 0.36 |
| LVFS (%) | 37.23 ± 3.75 | 37.27 ± 3.72 | 36.88 ± 3.52 |
| LVOT VTI | NA | 24.28 ± 3.16c | 25.65 ± 2.36 |
| PA VTI | NA | 18.47 ± 2.57c | 19.32 ± 2.12 |
| MV-E/A | NA | 0.57 ± 0.22 | 0.65 ± 0.26 |
| MV-DTE (s) | NA | 0.149 ± 0.03 | 0.150 ± 0.02 |
AO: Aortic root dimension, LA: left atrium size, LVEDD: left ventricular end-diastolic diameter, LVESD: left ventricular end-systolic diameter, LVFS: left ventricular fractional shortening, LVOT VTI: left ventricular outflow tract velocity time integral, PA VTI: pulmonary artery velocity time integral, MV-E/A: E/A ratio, MV-DTE: deceleration time, NA: not applicable, MV = mitral value.
aStatistically significant difference (P < 0.05), between baseline and 24 h post implantation.
bStatistically significant difference (P < 0.05), between baseline and 1 year post implantation.
cStatistically significant difference (P < 0.05), between 24 h and 1 year post implantation.
Of the total 22 patients three were receiving medication for mild arterial hypertension at the time of randomization, which was withdrawn during hospitalization and before echo evaluation.
Four patients were lost to follow-up: two because of pacemaker pocket erosion, one withdrew from participation in the study and one died of stroke. None of the patients complained of any symptom during the follow-up period. There was no need for AVI change during this period. Additionally, at the time of follow-up, two patients were on antihypertensive medication, which was discontinued before the echo exam.
Effect of variable AVI on Doppler-derived diastolic indices
As shown in Table 2, with regard to overall comparison, there was a progressively significant augmentation, from 200 ms to 100 ms AVI, in peak velocity of E wave (0.46 ± 0.16 → 0.55 ± 0.16 → 0.60 ± 0.18 m/s), in peak E to A velocity ratio (E/A) (0.53 ± 0.13 → 0.72 ± 0.15 → 0.90 ± 0.25) and in LV filling time (0.33 ± 0.05 → 0.37 ± 0.05 → 0.40 ± 0.06 s) followed by a significant progressive reduction in peak velocity of A wave (0.87 ± 0.14 → 0.77 ± 0.15 → 0.69 ± 0.18 m/s). Deceleration time of the E wave could not be evaluated at 200 ms because of E and A wave fusion.
Left ventricular diastolic indices in patients with DDD pacing at different AV intervals
| LV-diastolic indices | AV delay | Overall significance P-Value | AV delay | ||||
| 100 ms | 150 ms | 200 ms | 100 ms vs 150 ms | 100 ms vs 200 ms | 150 ms vs 200 ms | ||
| MV-PVE (m/s) | 0.60 ± 0.1 | 0.55 ± 0.1 | 0.46 ± 0.1 | 0.009a | NS | 0.002b | 0.007b |
| MV-PVA (m/s) | 0.69 ± 0.1 | 0.77 ± 0.1 | 0.87 ± 0.1 | 0.002a | 0.003b | 0.0005b | 0.002b |
| MV-E/A | 0.90 ± 0.2 | 0.72 ± 0.1 | 0.53 ± 0.1 | 0.0005a | 0.012b | 0.0005b | 0.0005b |
| MV-FT (s) | 0.40 ± 0.0 | 0.37 ± 0.0 | 0.33 ± 0.0 | 0.0005a | 0.003b | 0.0005b | 0.0005b |
| MV-DTE (s) | 0.14 ± 0.0 | 0.14 ± 0.0 | NA | NA | NS | NA | NA |
| LV-diastolic indices | AV delay | Overall significance P-Value | AV delay | ||||
| 100 ms | 150 ms | 200 ms | 100 ms vs 150 ms | 100 ms vs 200 ms | 150 ms vs 200 ms | ||
| MV-PVE (m/s) | 0.60 ± 0.1 | 0.55 ± 0.1 | 0.46 ± 0.1 | 0.009a | NS | 0.002b | 0.007b |
| MV-PVA (m/s) | 0.69 ± 0.1 | 0.77 ± 0.1 | 0.87 ± 0.1 | 0.002a | 0.003b | 0.0005b | 0.002b |
| MV-E/A | 0.90 ± 0.2 | 0.72 ± 0.1 | 0.53 ± 0.1 | 0.0005a | 0.012b | 0.0005b | 0.0005b |
| MV-FT (s) | 0.40 ± 0.0 | 0.37 ± 0.0 | 0.33 ± 0.0 | 0.0005a | 0.003b | 0.0005b | 0.0005b |
| MV-DTE (s) | 0.14 ± 0.0 | 0.14 ± 0.0 | NA | NA | NS | NA | NA |
MV-PVE: Peak E velocity, MV-PVA: peak A velocity, MV-E/A: E/A ratio, MV-FT: filling time, MV-DTE: deceleration time, NA: not applicable, MV = mitral value, NS = not significant.
aStatistically significant difference (P < 0.05).
bStatistically significant difference (P value of paired comparison <0.05).
Left ventricular diastolic indices in patients with DDD pacing at different AV intervals
| LV-diastolic indices | AV delay | Overall significance P-Value | AV delay | ||||
| 100 ms | 150 ms | 200 ms | 100 ms vs 150 ms | 100 ms vs 200 ms | 150 ms vs 200 ms | ||
| MV-PVE (m/s) | 0.60 ± 0.1 | 0.55 ± 0.1 | 0.46 ± 0.1 | 0.009a | NS | 0.002b | 0.007b |
| MV-PVA (m/s) | 0.69 ± 0.1 | 0.77 ± 0.1 | 0.87 ± 0.1 | 0.002a | 0.003b | 0.0005b | 0.002b |
| MV-E/A | 0.90 ± 0.2 | 0.72 ± 0.1 | 0.53 ± 0.1 | 0.0005a | 0.012b | 0.0005b | 0.0005b |
| MV-FT (s) | 0.40 ± 0.0 | 0.37 ± 0.0 | 0.33 ± 0.0 | 0.0005a | 0.003b | 0.0005b | 0.0005b |
| MV-DTE (s) | 0.14 ± 0.0 | 0.14 ± 0.0 | NA | NA | NS | NA | NA |
| LV-diastolic indices | AV delay | Overall significance P-Value | AV delay | ||||
| 100 ms | 150 ms | 200 ms | 100 ms vs 150 ms | 100 ms vs 200 ms | 150 ms vs 200 ms | ||
| MV-PVE (m/s) | 0.60 ± 0.1 | 0.55 ± 0.1 | 0.46 ± 0.1 | 0.009a | NS | 0.002b | 0.007b |
| MV-PVA (m/s) | 0.69 ± 0.1 | 0.77 ± 0.1 | 0.87 ± 0.1 | 0.002a | 0.003b | 0.0005b | 0.002b |
| MV-E/A | 0.90 ± 0.2 | 0.72 ± 0.1 | 0.53 ± 0.1 | 0.0005a | 0.012b | 0.0005b | 0.0005b |
| MV-FT (s) | 0.40 ± 0.0 | 0.37 ± 0.0 | 0.33 ± 0.0 | 0.0005a | 0.003b | 0.0005b | 0.0005b |
| MV-DTE (s) | 0.14 ± 0.0 | 0.14 ± 0.0 | NA | NA | NS | NA | NA |
MV-PVE: Peak E velocity, MV-PVA: peak A velocity, MV-E/A: E/A ratio, MV-FT: filling time, MV-DTE: deceleration time, NA: not applicable, MV = mitral value, NS = not significant.
aStatistically significant difference (P < 0.05).
bStatistically significant difference (P value of paired comparison <0.05).
With regard to paired comparisons (Table 2), in 100 ms vs 150 ms AVI, the peak velocity of the A wave was significantly reduced (0.69 ± 0.18 vs 0.77 ± 0.15 m/s, P = 0.003), E/A ratio was significantly augmented (0.90 ± 0.25 vs 0.72 ± 0.15, P = 0.012) while LV filling time showed a significant prolongation (0.40 ± 0.06 vs 0.37 ± 0.05 s, P = 0.003).
In paired comparison of 100 ms vs 200 ms (Table 2), there was a statistically significant augmentation in E wave peak velocity (0.60 ± 0.18 vs 0.46 ± 0.16 m/s, P = 0.002), E/A ratio (0.90 ± 0.25 vs 0.53 ± 0.13, P = 0.0005) and in LV filling time (0.40 ± 0.06 vs 0.33 ± 0.05 s, P = 0.0005). Additionally, a significant reduction in A wave peak velocity (0.69 ± 0.18 vs 0.87 ± 0.14 m/s, P = 0.0005) was documented.
In paired comparison of 150 ms vs 200 ms (Table 2), there was a statistically significant augmentation in E wave peak velocity (0.55 ± 0.16 vs 0.46 ± 0.16 m/s, P = 0.007), in E/A ratio (0.72 ± 0.5 vs 0.53 ± 0.13, P = 0.0005) and in LV filling time (0.37 ± 0.05 vs 0.33 ± 0.05 s, P = 0.0005), while a significant limitation in A wave peak velocity (0.77 ± 0.15 vs 0.87 ± 0.14 m/s, P = 0.002) was shown.
Effect of variable AVI on ANP and c-GMP levels
As shown in Table 3, according to overall comparison, a statistically significant progressive reduction in ANP (34.18 ± 11.65 → 27.13 ± 8.51 → 26.30 ± 7.63 pg/ml, P = 0.03) and c-GMP plasma levels (7.86 ± 4.95 → 6.83 ± 4.71 → 6.52 ± 4.53 pmol/ml, P = 0.001) with shortening of the AVI from 200 ms to 100 ms was documented.
ANP and c-GMP levels in patients with DDD pacing at different AV intervals
| Peptides | AV delay | Overall significance P-Value | AV delay | ||||
| 100 ms | 150 ms | 200 ms | 100 ms vs 150 ms | 100 ms vs 200 ms | 150 ms vs 200 ms | ||
| ANP (pg/ml) | 26.30 ± 7.6 | 27.13 ± 8.5 | 34.18 ± 11.6 | 0.03a | NS | 0.008b | 0.016b |
| c-GMP (pmol/ml) | 6.52 ± 4.5 | 6.83 ± 4.7 | 7.86 ± 4.9 | 0.001a | 0.053 | 0.0005b | 0.0005b |
| Peptides | AV delay | Overall significance P-Value | AV delay | ||||
| 100 ms | 150 ms | 200 ms | 100 ms vs 150 ms | 100 ms vs 200 ms | 150 ms vs 200 ms | ||
| ANP (pg/ml) | 26.30 ± 7.6 | 27.13 ± 8.5 | 34.18 ± 11.6 | 0.03a | NS | 0.008b | 0.016b |
| c-GMP (pmol/ml) | 6.52 ± 4.5 | 6.83 ± 4.7 | 7.86 ± 4.9 | 0.001a | 0.053 | 0.0005b | 0.0005b |
For abbreviations see text.
aStatistically significant difference (P < 0.05).
bStatistically significant difference (P value of paired comparison < 0.05).
ANP and c-GMP levels in patients with DDD pacing at different AV intervals
| Peptides | AV delay | Overall significance P-Value | AV delay | ||||
| 100 ms | 150 ms | 200 ms | 100 ms vs 150 ms | 100 ms vs 200 ms | 150 ms vs 200 ms | ||
| ANP (pg/ml) | 26.30 ± 7.6 | 27.13 ± 8.5 | 34.18 ± 11.6 | 0.03a | NS | 0.008b | 0.016b |
| c-GMP (pmol/ml) | 6.52 ± 4.5 | 6.83 ± 4.7 | 7.86 ± 4.9 | 0.001a | 0.053 | 0.0005b | 0.0005b |
| Peptides | AV delay | Overall significance P-Value | AV delay | ||||
| 100 ms | 150 ms | 200 ms | 100 ms vs 150 ms | 100 ms vs 200 ms | 150 ms vs 200 ms | ||
| ANP (pg/ml) | 26.30 ± 7.6 | 27.13 ± 8.5 | 34.18 ± 11.6 | 0.03a | NS | 0.008b | 0.016b |
| c-GMP (pmol/ml) | 6.52 ± 4.5 | 6.83 ± 4.7 | 7.86 ± 4.9 | 0.001a | 0.053 | 0.0005b | 0.0005b |
For abbreviations see text.
aStatistically significant difference (P < 0.05).
bStatistically significant difference (P value of paired comparison < 0.05).
With paired comparison (Table 3), there was no statistically significant difference in ANP levels between 100 ms and 150 ms AVI (26.30 ± 7.63 vs 27.13 ± 8.51 pg/ml, P = 0.370). On the contrary, ANP levels were statistically significantly lower with 100 ms vs 200 ms (26.30 ± 7.63 vs 34.18 ± 11.65 pg/ml, P = 0.008) and with 150 ms vs 200 ms AVI comparison (27.13 ± 8.51 vs 34.18 ± 11.65 pg/ml, P = 0.016).
Uniform alterations were seen in c-GMP plasma levels (Table 3), as there was no statistically significant difference between 100 ms and 150 ms (6.52 ± 4.53 vs 6.83 ± 4.71 pmol/ml, P = 0.053), but there was a statistically significant reduction in c-GMP levels with 100 ms vs 200 ms (6.52 ± 4.53 vs 7.86 ± 4.95 pmol/ml, P = 0.0005) and with 150 ms vs 200 ms (6.83 ± 4.71 vs 7.86 ± 4.95 pmol/ml, P = 0.0005) AVI comparisons (Table 3).
Effect of DDD pacing on heart rate and echo parameters throughout the follow-up
DDD pacing was obviously associated with a significant increase in heart rate at 24 h post implantation that remained significant at the 1-year follow-up (Table 1). Simultaneous improvement was documented in left atrial size and LV end-diastolic diameter. There were no significant differences with regard to aortic root dimension, left ventricular end-systolic diameter and fractional shortening throughout the follow-up period. Interestingly, Doppler-derived systolic function parameters (LVOT VTI and PA VTI) showed a statistically significant improvement at follow-up compared with the post implantation values despite the non-significant improvement in fractional shortening (Table 1).
The specific diastolic filling parameters remained unchanged over the follow-up period (Table 1).
ANP correlation with echo data
There was a significant negative (inverse) correlation (r = −0.478, P = 0.05) between ANP levels, with 100 ms AVI DDD pacing at baseline and fractional shortening at 1 year of follow-up (Fig. 1). There were no statistically significant correlations between natriuretic peptide levels and the other systolic or diastolic indices at any time in the study.
Correlation between ANP levels at baseline and LV fractional shortening in 1 year following implantation.
Discussion
Determination of the optimal AV interval during cardiac pacing is crucial especially in patients with heart failure. Many studies have shown haemodynamic improvement in a population with impaired LV systolic function, during AV pacing with a short AVI[1–,3]. In contrast, other studies have not found any beneficial effect of a short AVI on cardiac output[25–,28]. These contradictions may be attributed to different patient selection and also to different methodology used in each study. Interestingly, Leonelli et al.[29] suggested that either shortening or lengthening the AVI from its optimal value in patients with complete heart block and normal systolic function is associated with a decrease in stroke volume. Currently, there is increased evidence that the determination of the optimal AVI for each patient should be individualized, as has been proposed during the last decade [21–,23,,29,,30].
The aim of the present study was to evaluate the effect of variable AVI during DDD pacing on LV diastolic performance and plasma ANP and c-GMP levels, in a cohort of patients with disturbed diastolic filling and preserved systolic function. In this specific population, AVI optimization should obviously be focused mainly on LV diastolic function improvement. We have demonstrated that DDD pacing with an AVI of 100 ms in the majority of the patients tested, produces better diastolic filling dynamics that remain unaffected throughout the follow-up period of 1 year and is associated with lower and thus more favourable plasma natriuretic peptide levels, compared with longer AVIs. Additionally, we have shown that Doppler systolic indices showed significant improvement when they were re-evaluated 1 year after implantation with the same AVI.
Doppler-derived LV diastolic indices with different AVIs showed an overall comparison statistically significant differences in almost all parameters tested with a very high significance (most P values < 0.0005). Our findings are in concordance with those observed by Leonelli et al. in a similar study population, where a statistically significant decrease in the total diastolic filling time and a similar progressive decrease in the E wave peak velocity was shown when the AVI increased from 70 to 220 ms. The differences in our study were more significant when shorter AVIs were compared with those of 200 ms, while they were less significant when the shorter AVIs (100 ms vs 150 ms) were compared. Another interesting finding was that the diastolic filling properties of our population remained unchanged when they were re-evaluated 1 year after implantation. From the total of 18 patients that were reassessed, 16 had a programmed AVI of 100 ms while two had an AVI of 150 ms, giving some evidence of the beneficial effect over time, of shorter AVI compared with longer ones.
Previous studies suggested that there is a significant correlation between the changes in ANP values and pulmonary artery wedge pressure during AV synchronized pacing and ventricular pacing with various programmed rates[10]. Furthermore, it has been documented that even after short periods of pacing ANP levels are affected[31]. In the present study, the comparison of plasma ANP and c-GMP values during the pacing protocol at three specific AVIs showed that both natriuretic peptide levels were significantly lower when the 100 ms and 150 ms AVI groups were compared with the longer 200 ms group. Finally, when the two shorter AVIs were compared, no statistical significant difference was found. Interestingly, a statistically significant negative correlation was detected between ANP levels with an AVI of 100 ms and fractional shortening in 1 year following implantation of the pacemaker. We have also documented that the cross correlation of ANP and c-GMP values was adequate, a finding that agrees with the study of Surdacki et al.[11]. These findings reinforced our belief that during DDD pacing for complete heart block, an AVI of 100 ms and/or 150 ms may have similar beneficial effects on LV performance. These results partially agree with Theodorakis et al.[12] observations, who did not find any difference in plasma ANP levels between the AVIs of 100 ms and 150 ms at rest.
Another finding of our study was the beneficial effect of DDD pacing on left atrial and left ventricular geometry in terms of left atrial size and LV end-diastolic diameter reduction attributed to AV sequence optimization. These remained stable for 1 year after implantation and were associated with improvement in Doppler systolic indices. No benefit was shown in fractional shortening which was within normal limits at baseline.
Previous reports have demonstrated a large interindividual variability of the optimal AVI during DDD pacing, necessitating individual testing of every patient to find the best interval. Despite this, the combination of echo-Doppler evaluation of LV diastolic function with ANP and c-GMP levels assessment indicated that an AVI of 100 ms or 150 ms may be the optimal AVI during AV sequential pacing in patients with normal LV systolic function, but disturbed LV diastolic function following the loss of AV sequence due to complete AV block.
Study limitations
In the present study where the patient population had no major cardiac disease, LV systolic function was normal and this may limit the clinical implications of our findings. The different AVIs tested during AV pacing protocol were only three and differed by ±50 ms, so each AVI represents a ±25 ms period. Additionally, each AVI was programmed for the right heart chambers, which may differ from those of the left heart as it is well established that diastolic filling and cardiac output depend dominantly on left AV sequence[20].
In this study, optimal AVI is referred to a programmed heart rate (80 beats/min) at rest and supine position. No attempt was made here to study patients on exercise or with a catecholamine infusion. We believe that the application of measurements during moderate physical exercise probably would improve the reliability of the method. Additionally, diastole is a complex phenomenon and diastolic filling patterns, as currently assessed by Doppler echocardiography, do not necessarily reflect all LV diastolic properties. Doppler-derived indices of the transmitral flow may be influenced by a number of factors such as heart rate, tone of the autonomic nervous system, preload variations and certain drugs. Keeping in mind all these parameters, we tried to evaluate diastolic filling dynamics, under stable haemodynamic conditions, in all patients who also served as their own controls according to the study methodology.
Another minor limitation of the study could be that blood samples for ANP and c-GMP assessment were drawn from a peripheral vein (brachial) and not directly from the atria, which would have been more accurate and sensitive.
Conclusions
Left ventricular diastolic filling dynamics and atrial natriuretic peptide levels in patients with DDD pacing for high degree AV block and normal systolic function are influenced by the programmed AVI. Doppler-derived LV diastolic indices and ANP and c-GMP plasma levels constitute a useful modality for AVI optimization in this specific cohort of patients. An AVI of 100 ms or 150 ms may represent the optimal AVI during AV sequential pacing, producing beneficial and sustained LV diastolic performance and may be associated with a more favourable prognosis.
