Europace Advance Access originally published online on September 28, 2006
Europace 2006 8(11):988-993; doi:10.1093/europace/eul103
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
PACING/ICD/CRT
The use of unipolar intracardiac impedance for discrimination of haemodynamically stable and unstable arrhythmias in man
1 Department of Cardiology, East Yorkshire Hospital Trust and University of Hull, Hull, East Yorkshire HU7 3AZ, UK; 2 Biotronik UK Ltd, Avonbury Business Park, Bicester, Oxon, OX26 3WX, UK; 3 Biotronik GmbH & Co KG, Erlangen, Germany
Manuscript submitted 26 July 2005. Accepted after revision 9 July 2006.
* Corresponding author: Department of Cardiology, Princess Alexandra Hospital, Ipswich Road, Brisbane Queensland 4102, Australia. Tel: +617 32402537; fax: +617 32407630. E-mail address: gerrykaye100{at}yahoo.co.uk
 |
Abstract |
|---|
 Top Abstract IntroductionRationaleClinical protocolMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
Aims To determine the feasibility of discriminating haemodynamically stable from unstable arrhythmias using right ventricular (RV) unipolar intracardiac impedance (Z).
Methods and results A quadrapolar temporary pacing electrode was positioned at the RV apex and unipolar impedance was measured between the tip electrode and a surface patch electrode. Changes in peak-to-peak Z amplitude were measured simultaneously with surface ECG and blood pressure during induced arrhythmias. Haemodynamic instability was defined as a systolic pressure of <90 mmHg. There were 25 episodes of ventricular fibrillation (VF) induced in 15 patients, 18 episodes of ventricular tachycardia in 16 patients, and 33 episodes of supraventricular tachycardia (SVT) in 16 patients. Compared with the baseline rhythm, mean Z amplitude reduced from 51.3±7.7 to 11.2±7.4 Ohm (P<0.001) during VF, from 52.2±6.3 to 21.7±10.1 Ohm (P<0.01) during haemodynamically unstable VT, from 55.0±6.9 to 39.9±11 Ohm (ns) during stable VT, and from 56.4±8.4 to 36.9±9.3 Ohm during SVT (P<0.001).
Conclusion Right ventricular unipolar impedance is an adequate sensor for determining mechanical ventricular contraction and acts as a surrogate marker for a fall in arterial blood pressure during VF. However, for ventricular and supraventricular tachycardias, variations between patients did not allow adequate discrimination between stable and unstable arrhythmias.
Key Words: Unipolar intracardiac impedance, Ventricular fibrillation, Haemodynamic stability
|
Introduction |
|---|
|
TopAbstractIntroductionRationaleClinical protocolMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
Modern defibrillators (ICD) capable of delivering tiered therapy use heart rate changes to determine arrhythmia onset. Tiered therapy usually involves a sequence of painless antitachycardia pacing, low energy shocks, which may be tolerated by patients, and high energy cardioversion/defibrillation. Shocks may occur while the patient is conscious, and if high energy, are painful and often psychologically debilitating.1
Avoidance of shocks also increases device longevity. An ideal algorithm could include a heart rate detector combined with a haemodynamic sensor. The latter would detect a fall in cardiac output and, by implication, impairment or loss of consciousness and could be used to tailor specific therapy and avoid or delay shocks.
Haemodynamic sensors have been used with variable success in the past. The ideal might include a lead-based pressure sensor in the right ventricle (RV) and current technology is such that long-term pressure sensing may now be feasible.2 Intracardiac impedance has also been used to determine changes in left ventricular performance and there is a good correlation between left ventricular stroke volume and impedance.3 In a previous study by Khoury et al.,4 the impedance amplitude and RV pulse pressure during unstable arrhythmias were demonstrated to be significantly smaller than that during stable arrhythmias. Those with haemodynamic instability had the greatest reduction in impedance, suggesting that this could be used as a sensor to determine haemodynamic stability. Unipolar impedance, recording changes in electrical impedance between the pacing lead tip and the generator, has been shown to determine alterations in RV contractility and is currently used as a rate responsive sensor in commercially available permanent pacemaker systems (Inos, Protos, and Cylos, Biotronik GmbH, Berlin, Germany).5
The aim of this study was to determine the feasibility of discriminating haemodynamically stable from unstable arrhythmias using RV unipolar intracardiac impedance (Z).
|
Rationale |
|---|
|
TopAbstractIntroductionRationaleClinical protocolMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
At the outset of the study, we wished to determine whether unipolar intracardiac impedance would be successful in detecting the haemodynamic instability of ventricular fibrillation (VF) before going on to study haemodynamic changes in other arrhythmias. Ventricular fibrillation was chosen as it is a clear and reproducible situation of complete haemodynamic collapse and any failure of impedance, in this instance, would render it useless as a sensor in other arrhythmias. Unipolar impedance was used because this is the mode of measurement in current commercial systems.
|
Clinical protocol |
|---|
|
TopAbstractIntroductionRationaleClinical protocolMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
The study had been approved by the local research Ethics Committee, and all patients gave written informed consent. A 6F intracardiac quadrapolar (5 mm spacing) electrode catheter (Bard Ltd, West Sussex, UK) was positioned at the RV apex from the right femoral vein. A surface patch electrode (surface area of 187 cm2) was placed over the patient's left scapula. Impedance (Z) (discussed subsequently) was measured continuously between the distal electrode of the quadrapolar catheter and the surface patch electrode. Surface ECG lead II, ventricular IEGM, and intracardiac impedance signals were continuously recorded.
For the VF part of the study, patients undergoing implantation of a cardioverter/defibrillator (ICD) were included. Local anaesthesia was used together with light intravenous sedation with propofol. Ventricular fibrillation was induced in a standard manner using either a T-wave shock or high frequency AC fibrillation. In one patient, intra-arterial monitoring was performed to confirm complete loss of blood pressure.
Following analysis of the VF data, a further group of patients was studied: patients undergoing routine clinical electrophysiological studies, including VT stimulation, and supraventricular tachycardia (SVT) studies prior to ablation. Patients were studied in the post-absorptive non-sedated state. Invasive intra-arterial blood pressure was measured in all patients with VT using a 4F desi-valve sheath inserted into the femoral artery and connected to a fluid-filled transducer (Medex Medical, model no. MX 9604; Medex Medical Ltd, Lancashire, UK). For SVT, blood pressure was recorded invasively in 4 patients and non-invasively in 12 patients using a standard sphygmomanometer.
|
Methods |
|---|
|
TopAbstractIntroductionRationaleClinical protocolMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
Impedance
An external Inos pacemaker (Biotronik GmbH) was used for measuring the impedance between the RV catheter tip (distal pole) and the surface patch electrode. The relative size of the distal pole of the catheter and the surface patch electrode was felt to represent an implantable system. The device scans the unipolar intracardiac impedance during the entire heart cycle by delivery of subthreshold biphasic rectangular pulses of 600 µA at a sampling rate of 128 Hz. The device was programmed to VVI mode with a base rate of 40 p.p.m. with nominal voltage, pulse duration, and sensitivity. The alternating component of the impedance signal was measured with a resolution of 0.83 Ohm and a range of ±107 Ohm.
Data acquisition and analysis
Surface ECG lead II and ventricular IEGM were recorded simultaneously via an electrophysiology system to a laptop computer. Femoral artery pressure (FAP) was measured continuously in all VT patients and in only one patient with VF to confirm haemodynamic collapse. Intracardiac impedance was recorded directly to a digital data recorder. The signals were manually synchronized and all data analysed off-line. Each arrhythmia episode was selected manually and the amplitude changes were measured by a specifically designed automatic software package. The peak-to-peak impedance wave amplitude was measured on a beat-to-beat basis. The average signal-to-noise ratio of this signal was 18±6 dB during sinus rhythm.
Ten cardiac cycles of intrinsic rhythm prior to arrhythmia induction were averaged and compared with the first 5 s after arrhythmia onset. Each peak-to-peak amplitude was measured and averaged for every arrhythmia episode. In cases where more than one arrhythmia episode occurred in a patient, an average of all episodes was taken for that patient.
Student's paired t-test was used to compare impedance waveform amplitude between baseline rhythm and the arrhythmia. A one-way analysis of variance of the reduction in impedance amplitude was performed between the four arrhythmia groups using SPSS (version 13.0, SPSS Inc., Chicago, IL, USA). Homogeneity of variances was confirmed by the Levene test and a post hoc multiple comparison was performed according to Bonferroni.
Haemodynamic instability
Haemodynamic instability was arbitrarily defined as a reduction in systolic pressure to <90 mmHg and/or a change in or loss of consciousness.
Patients
VF group: 15 patients, 13 males, mean age 64±11 years were studied. The mean left ventricular ejection fraction (LVEF) was 40.8±18%.
VT group: there were 16 patients, mean age 64±8 years. The mean LVEF was 43.1±17%.
SVT group: there were 16 patients; 8 with atrioventricular re-entrant tachycardia (AVRT), 5 with junctional re-entrant tachycardia (JRT), and 3 with atrial fibrillation/flutter occurring during the EP study. Their mean age was 44±16 years.
|
Results |
|---|
|
TopAbstractIntroductionRationaleClinical protocolMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
VF group
Impedance: 25 episodes (in 15 patients) of VF were induced and the changes in the amplitude of impedance during VF compared with sinus rhythm are shown in Table 1. For all patients with VF, the mean impedance amplitude reduced by 74±9% (from 51.3±7.7 to 11.2±7.4 Ohm), compared with baseline sinus rhythm (P<0.001).
|
Figure 1 demonstrates a sample of the impedance waveform during sinus rhythm at the induction of VF. There was an immediate and marked reduction in amplitude with a loss of the stable morphology in the Z signal. Blood pressure fell to <30 mmHg. At defibrillation to paced rhythm, there was immediate restoration of the impedance waveform to levels comparable with those pre-VF. In one patient, VF was prolonged. During this episode, the impedance waveform changes were constant and remained low amplitude until defibrillation was successful. In addition, there was no artefact from atrial contraction (during sinus rhythm) on the impedance signal during VF in this case.
|
VT group
Impedance: there were 18 episodes of induced VT of which 13 (12 patients) were haemodynamically unstable (HUSVT) and 5 (4 patients) were haemodynamically stable (HSVT). During HUSVT, Z amplitude reduced significantly by 56±21% (P<0.01) and during HSVT, by 23±15% (ns) when compared with baseline sinus rhythm, as shown in Table 1 and Figure 2. Table 2 shows changes in heart rate and mean arterial pressure during arrhythmias. Figure 3 shows changes in the mean systolic blood pressure and mean impedance amplitude expressed as a percentage of that during baseline rhythm.
|
|
|
SVT group
Impedance: there were 33 episodes of SVT: 21 AVRT (8 patients), 9 JRT (5 patients), and 3 atrial fibrillation/flutter (3 patients). Changes in impedance between baseline rhythm, AVRT, JRT, atrial fibrillation/flutter, and VT are shown in Table 1 and Figure 2. For all episodes of SVT, the mean Z reduced from 56.4±8.4 to 36.9±9.3 Ohm (P<0.001). If SVTs are divided into component arrhythmias, then Z amplitude reduced by 37±39% (ns) during AVRT, by 40±8% (P<0.001) during JRT, and by 17±11% (ns) for atrial fibrillation/flutter compared with intrinsic rhythm (Table 1 and Figure 2).
However, in one patient with AVRT, there was a large fall in both impedance amplitude and blood pressure (from 119 to 70 mmHg systolic). The fall in impedance was of a magnitude which was of a similar level to that seen in unstable VT. The heart rate in this episode of AVRT was 190 b.p.m. The heart rate range for all SVTs was 150–230 b.p.m.
The changes in heart rate and blood pressure are shown in Table 2. There was a significant difference (P<0.05) in the Z amplitude reduction between VF and HSVT and between VF and SVT. There were no significant differences for Z between SVT, HUSVT, and HSVT.
If detection of haemodynamic instability is taken as a reduction of 55% in Z amplitude, then the detection sensitivity is 82% with zero false positive rate. If, however, the threshold reduction is taken as 50%, sensitivity becomes 86% with an 11% false positive rate.
|
Discussion |
|---|
|
TopAbstractIntroductionRationaleClinical protocolMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
This study showed that at the onset of VF, there was a significant reduction in the amplitude of unipolar intracardiac impedance. These changes were immediate, consistent for all episodes, and the loss of amplitude persisted throughout the period of fibrillation. The study confirms that the instability of VF could be reliably detected using unipolar intracardiac impedance. Although there is no current requirement for an additional sensor for VF, any inability of unipolar impedance to fail to detect consistently haemodynamic instability during VF would have rendered any similar impedance sensor non-viable for other arrhythmias. Unipolar intracardiac impedance, as opposed to multipolar, was used because it was simple, was already available in a commercial pacemaker, and relatively non-complex computer analysis could be used to analyse the signal. A further advantage would be that only a single lead without modification would be required (no lead-based sensor is required). Also, changes in impedance provide information on the contracting motion of the heart, which is supplementary to the pure timing information (e.g. heart rate). Ideally, impedance should detect any haemodynamically unstable arrhythmia irrespective of heart rate. The present study would suggest that the faster the heart rate, the greater the fall in impedance amplitude. Mainly, faster arrhythmias, particularly ventricular, are more associated with haemodynamic instability. Although it is possible that the changes in impedance could be purely related to heart rate change, the mechanisms giving rise to the impedance waveform in this setting are complex and heart rate is only one factor directly influencing impedance. Additionally, in the one SVT case where there was a large fall in impedance, there were other cases of SVT with higher heart rates without such a reduction in impedance. It is, therefore, unlikely that heart rate changes alone account for the fall in impedance amplitude.
Although the changes during VF were consistent, impedance during VT and SVT was less predictable. In VT, there was a reduction in impedance in most cases of haemodynamic instability, but there were three cases where the fall in impedance was not particularly large and spanned the impedance range of stable arrhythmias. Although the statistics suggests that there are significant differences between groups, we observed that, in some cases, there was an overlap between the groups making absolute discrimination by impedance alone unreliable.
The changes in unipolar impedance in this clinical setting are also complex. In unipolar impedance, the majority of the current density occurs within 1 cm of the tip of the RV catheter. More than 90% of this is lost when the distance exceeds 1 cm from the catheter. The localized nature of the measurement may not, therefore, reflect the true changes in stroke volume within the left ventricle. In addition, changes in impedance are likely to be sensitive to subtle catheter tip movement.4,5 Even minor catheter tip displacement may cause significant changes in impedance waveform. However, during VF, there is loss of effective ventricular contraction and both LV and RV stroke volumes fall to zero. We believe that a consistent fall in impedance amplitude during VF would, therefore, be expected. Also, during VF, effective movement of the catheter tip is reduced to zero and changes in Z would be expected to remain consistent. During VT and SVT, there may be localized wall motion abnormalities, which not only alter the movement and position of the catheter tip, but may also change the relationship between right and left ventricular function sufficiently to produce unpredictable changes in impedance. Localized changes in blood flow near the catheter tip may also influence impedance. A superposition of these issues may have caused the variability that was observed in the study. The problem of catheter tip movement could have been partly overcome by using leads fixed at the RV apex. It is known that consistent morphologies in impedance waveform are seen in commercial systems that use chronic lead configurations.
In the previous study by Khoury et al.,4 a multipolar (quadrapolar) approach was used. Impedance was measured between poles 2 and 4 while injecting current between pole 1 and an indifferent surface electrode. Multipolar measurements reflect changes in RV volume and output, whereas unipolar impedance does not, as it only looks at localized contraction dynamics. The Khoury results showed a relationship between haemodynamic stability and impedance, irrespective of the underlying arrhythmia. In that study, changes in impedance were consistent, with unstable arrhythmias having the greatest reduction. The impedance signal also correlated well with changes in the RV pulse pressure. Overall, these authors felt that multipolar impedance would be an accurate sensor for haemodynamic monitoring.
Also, in previous studies, persistent atrial systolic activity has been detected with a multipolar catheter in the RV during periods of VF, giving the false impression of continued RV volume changes.5,6 In our study, these changes were not seen during VF with continued sinus rhythm because unipolar impedance changes are so localized to the blood and tissue characteristics within 1 cm radius of the circumference of the distal electrode at the RV apex.
Although RV pulse pressure would also be an ideal sensor,7–10 it is not until recently that the technology has allowed reliable long-term measurements. A specially modified intracardiac lead is required.2 Impedance, however, is commercially viable and long-term signal stability and calibration has not proved to be an issue. In this setting, RV impedance measurements appear to reflect changes in overall cardiac output. Further, there appears to be a consistent reduction in impedance amplitude during haemodynamic instability. Small but sufficient variations occur which do not allow the sensor to be sufficiently discriminatory in the current study.
There were limitations of the current experimental setup. All patients were studied supine. The data are acute, and chronic data would be critical for assessing the success of any future sensor. Temporary pacing catheters were used to measure impedance and are notorious for tip displacement. Using catheters fixed to the endocardial surface might overcome this issue, as catheter tip movement is critical to obtain stable impedance curves. Overall patient numbers were small in each group, but clinical practice changed during the project such that less VT studies were being performed in place of direct implantation of defibrillators, thereby reducing the number of patients available for study.
Although impedance looks promising for assessing changes in cardiac function, the optimal signal configuration is not yet defined. Unipolar impedance would have been simple to apply with no lead modification, but has not proved to be an ideal sensor in this experimental setup. However, the advent of biventricular systems allows the possibility of assessing the optimal signal by measuring transventricular impedance between the right and left ventricular leads. This may allow more of the left ventricular dynamics to be studied and may be useful in assessing the benefits of biventricular pacing.
|
Conclusions |
|---|
|
TopAbstractIntroductionRationaleClinical protocolMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
In the current experimental setting, unipolar intracardiac impedance is an adequate sensor for detecting mechanical ventricular contraction and, therefore, is suitable to indicate the onset of VF in man. It has not proved to be a sufficiently discriminatory sensor for assessing haemodynamic changes during VT and SVT in man.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionRationaleClinical protocolMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
We are grateful to Miss Valerie Wilby, senior technician, who helped with obtaining the clinical data. The study was funded by a grant from Biotronik UK, Ltd. G.C.K. has delivered paid lectures on behalf of Biotronik UK. G.C.K. and Biotronik GmbH are joint holders of a patent.
|
References |
|---|
|
TopAbstractIntroductionRationaleClinical protocolMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
[1] Burns JL, Serber ER, Keim S, et al. Measuring patient acceptance of implantable cardiac device therapy. J Cardiovasc Electrophysiol 2005; 16: 384–90.[Web of Science][Medline]
[2] Magalski A, Adamson P, Gadler F, et al. Continuous ambulatory right heart pressure measurements with an implantable haemodynamic monitor: a multicenter 12 month follow-up study of patients with chronic heart failure. J Card Fail 2002; 8: 63–70.[CrossRef][Web of Science][Medline]
[3] Salo RW, Wallner TG, Pedersen BD. Measurement of ventricular volume by intracardiac impedance—theoretical and empirical approaches. IEEE Trans Biomed Eng 1986; 33: 189–95.[Web of Science][Medline]
[4] Khoury D, McAllister H, Wilkoff B, et al. Continuous right ventricular volume assessment by catheter measurement of impedance for antitachycardia system control. Pacing Clin Electrophysiol 1989; 12: 1918–26.[CrossRef][Medline]
[5] Osswald S, Cron T, Gradel C, et al. Closed-loop stimulation using intracardiac impedance as a sensor principle: correlation of right ventricular dP/dtmax and intracardiac impedance during dobutamine stress test S. Pacing Clin Electrophysiol 2000; 23: 1502–8.[CrossRef][Medline]
[6] Woodward JC, Bertram CD, Gow BS. Right ventricular volumetry by catheter measurement of conductance. Pacing Clin Electrophysiol 1987; 10: 863–70.
[7] Cohen TJ, Veltri EP, Lattuca JJ, et al. Hemodynamic responses to rapid pacing: a model for tachycardia differentiation. Pacing Clin Electrophysiol 1988; 11: 1522–8.[CrossRef][Medline]
[8] Beauregard LA, Volosin KJ, Waxman HR. Differentiation of arrhythmias by measurement of intracardiac pressures in man. Pacing Clin Electrophysiol 1991; 14: 161–7.[CrossRef][Medline]
[9] Kapadia KA, Wood MA, Lu B, et al. A prospective study of changes in right ventricular dP/dt during ventricular tachycardia. Pacing Clin Electrophysiol 1991; 14: 1098–104.[CrossRef][Medline]
[10] Wood M, Ellenbogen KA, Lu B, et al. A prospective study of right ventricular pulse pressure, and dP/dt to discriminate induced ventricular tachycardia from supraventricular and sinus tachycardia in man. Pacing Clin Electrophysiol 1990; 13: 1148–57.[CrossRef][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. Kaye, D. Edgar, T. Mudawi, M. Lippert, and G. Czygan Can transventricular intracardiac impedance measurement discriminate haemodynamically unstable ventricular arrhythmias in human? Europace, February 1, 2007; 9(2): 122 - 126. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||