© 2008 European Society of Cardiology
Changes in myocardial electrical impedance in human heart graft rejection
a Department of Cardiology, Hospital de la Santa Creu i Sant Pau, Autonomous University of Barcelona Barcelona, Spain
b Department of Electronic Engineering, Universitat Politecnica de Catalunya Barcelona, Spain
c Department of Pathology, Hospital de la Santa Creu i Sant Pau, Autonomous University of Barcelona Barcelona, Spain
d Department of Cardiac Surgery, Hospital de la Santa Creu i Sant Pau, Autonomous University of Barcelona Barcelona, Spain
* Corresponding author. Department of Cardiology, Hospital de la Santa Creu i Sant Pau, St. Antoni Ma Claret, 167, 08025 Barcelona, Spain. E-mail address: jcinca{at}santpau.es
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Methods3. Results4. DiscussionFunding sourcesReferences |
|---|
Background: Monitoring of post-transplant heart rejection is currently based on endomyocardial biopsy analysis. This study aimed to assess the effects of heart graft rejection on myocardial electrical impedance.
Methods and results: Twenty-nine cardiac transplant patients and 9 controls underwent measurement of myocardial electrical impedance using a specifically designed amplifying system. The module and phase angle of myocardial impedance were measured. Histopathological rejection grading was performed according to ISHLT classification.
Fifty impedance tests were performed in transplanted patients. Myocardial impedance (Z) was higher in controls than in transplanted patients (p<0.001) and followed a progressive decline at increasing current frequencies (p<0.001). Likewise, the phase angle of impedance in controls ranged from positive values at low frequencies to negative values at higher frequencies (from 2.5±0.9° at 10 kHz to – 3.8±2.1° at 300 kHz, p<0.001). Rejection was associated with a significant decrease in myocardial impedance (Z) (15±6.6 
in grade 0, 13±6.0 in grade 1A, and 3.3±0.9 in grade 3A at 10 kHz, p<0.003).
Conclusions: Mild degrees of cardiac graft rejection are associated with significant changes in myocardial electrical impedance in transplant patients. Further clinical investigation is warranted to assess the potential of cardiac impedance to detect heart graft rejection.
Key Words: Transplantation • Rejection • Catheters
Received December 24, 2007; Revised April 9, 2008; Accepted April 28, 2008
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. DiscussionFunding sourcesReferences |
|---|
Heart graft rejection is an important cause of death in patients with cardiac transplantation. Among 287 heart transplants performed last year in Spain, heart rejection was responsible for about 9% of deaths during the first year of transplantation [1]. Surveillance of graft rejection is nowadays based on histological analysis of serial right ventricular endomyocardial biopsies, since less invasive methods such as echocardiography or magnetic resonance are not sensitive enough to recognize slight to moderate degrees of rejection. During the first two years following transplant, each patient is scheduled for about 14 biopsy sessions in our institution. However, this histologically-guided method is accompanied by the potential risk of repeated right cardiac catheterization and the induction of local fibrosis secondary to repeated myocardial biopsy procedures which can reduce the likelihood of obtaining an appropriate amount of tissue suitable for histopathological analysis. New alternative less invasive diagnostic tools therefore need to be developed.
Experimental models of acute myocardial ischaemia and healed myocardial infarction have shown that the injured myocardial regions can be localized by direct on-line measurement of the electrical impedance of the myocardium using a contact intracavitary electrocatheter [2]. In these studies, the two main components of myocardial electrical impedance (the module and the phase angle) were systematically determined. The module is mostly influenced by alterations occurring in the extracellular space, whereas the phase angle is more affected by alterations in cell membrane integrity and gap junction conductivity. Therefore, it is conceivable that rejection-induced oedema, leukocyte infiltration, haemorrhage and cell necrosis would modify the electrical impedance properties of the myocardium. To date, attempts to evaluate the diagnostic potential of myocardial electrical impedance in heart graft rejection have been limited to in canine models of heterotopic heart transplantation [3], these have shown a high sensitivity and specificity of the impedance method.
Therefore, this pilot study was designed to explore the possible changes in myocardial electrical impedance induced by cardiac graft rejection in humans. A purposely developed on-line impedance recording system for clinical use was employed and data were correlated with histological rejection grading from simultaneously taken endomyocardial biopsies in cardiac transplant patients. A group of control patients free from structural heart disease, undergoing electrophysiological testing due to paroxysmal supraventricular tachycardia was included to allow estimation of the normal reference values of myocardial electrical impedance in humans.
|
2. Methods |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. DiscussionFunding sourcesReferences |
|---|
2.1. Study population
The study included two groups of patients. The control group was composed of 9 patients with no clinical evidence of structural heart disease who were undergoing electrophysiological testing due to episodes of paroxysmal supraventricular tachycardia. The transplant group was composed of 29 patients with end-stage heart failure who had undergone cardiac transplantation using the bicaval technique. Patients were recruited consecutively from those undergoing the current post-transplant endomyocardial biopsy protocol at our institution. The study was performed in the fasting state under light sedation. All patients gave informed consent and the Ethics Committee of our institution approved the study.
2.2. Myocardial impedance
The magnitude (Z) and phase angle (
) of myocardial tissue electrical impedance were measured at 10, 30, 100 and 300 kHz using a specifically designed amplifying system.
The system uses a custom isolated front end to ensure electrical safety and a commercial impedance analyser (HP4192A). A complete description of the system can be found elsewhere [4]. Briefly, myocardial electrical impedance was measured by applying an alternating current between the distal electrode of the electrocatheter and a skin electrode placed in the dorsal region (Fig. 1). This configuration attenuates the effect of catheter tip position relative to myocardial surface and also the contact impedance [5]. In addition, the large size of the skin reference electrode ensures low impedance in comparison with the myocardial impedance. As shown in Fig. 1, the greatest current density is generated in the endomyocardial regions thus affording the greatest recording sensitivity in these areas [6]. The recording system is capable of acquiring up to 8 impedance measurements per second in each single current frequency with a resolution of 0.1 
. The impedance measurement system has a dynamic range from 10 to 6.7 k.
|
At the start of each test, a 7F electrocatheter (Boston Scientific EPT5031TK2), was introduced percutaneously with local anaesthesia and was directed to the right cardiac cavities under fluoroscopic control. Vascular access was via the jugular vein in the transplanted patients and the femoral vein in the control group. In both groups of patients we used the same type of electrocatheter in order to obtain more uniform impedance measurements.
The surface ECG and the tip electrode endocardial electrogram were recorded. Displacement of the ST segment with respect to the isoelectric line in the endocardial electrogram was considered indicative of a close contact between the electrode tip and the endocardial surface. Only ventricular sites meeting this ST segment pattern were included in the study. The right atrium was reached by withdrawing the ventricular electrocatheter under fluoroscopic guidance and the atrial recording sites were selected from those depicting a large local atrial electrogram. Some transplanted patients were tested more than once.
2.3. Histopathological analysis
Myocardial samples of the right ventricle were obtained under light sedation using a bioptome (Johnson and Johnson; reference 502402BL or 504300L) inserted through the jugular or femoral vein. In each session a mean of about 3 to 4 tissue samples was obtained. The tissue specimens were immediately immersed in buffered formalin and were fixed in paraffin. Sections of each specimen were stained with haematoxylin-eosin. The sections were analyzed using conventional light microscopy and cardiac graft rejection was graded on a seven point scale according to the International Classification [7].
2.4. Study protocol
In the transplanted group of patients, tissue electrical impedance was measured at the septal and apical right ventricular regions and in the right atrium. Immediately thereafter, the electrocatheter was withdrawn and replaced by the bioptome, which was used to obtain myocardial samples from approximately the same endocardial areas.
In the non-transplanted control group of patients, the impedance measurements were performed immediately before the electrophysiological testing protocol.
2.5. Data analysis
The mean values of the module and phase angle of myocardial impedance were calculated at each current frequency. To minimize the influence of skin and extracardiac thoracic tissues on measurements, the ventricular impedance values were referred to the atrial ones and expressed as the difference between the right ventricle (RV) and right atrium (RA). Therefore, for each current frequency and measurement session, two indices were computed as shown in Eqs. (1) and (2):
|
| (1) |
|
| (2) |
Where 
denotes the averaging procedure, |Z| is the magnitude of the impedance and is the phase angle. The frequency dependence of impedance is denoted by the subindex f.
According to the histopathological grading, the transplanted group of patients were divided into 3 subgroups: T0: no rejection; T1A: light rejection; T3A: moderate rejection. Intermediate rejection grades 1B and 2 were not included in this study in order to clearly separate light to moderate rejection groups. None of the patients had rejection levels greater than 3A. The Kolmogorov-Smirnov test (with Lilliefor's correction) was performed to check for normality of the indices. The Levene Median test was used to check the equal variance of the indices. If the indices passed both tests (p>0.05), a One Way Analysis of Variance was performed and if there were significant differences, the Tukey test was used for multiple comparisons between pairs of subgroups. If the normality or the equal variance tests failed (p<0.05), a Kruskal-Wallis One Way Analysis of Variance on Ranks was performed and Dunn's method was used for multiple comparisons whenever significant differences were found.
2.6. Statement of responsibility
The authors had full access to the data and take responsibility for its integrity. All authors have read and agree to the manuscript as written.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. DiscussionFunding sourcesReferences |
|---|
Myocardial electrical impedance was measured in 9 control patients and in 29 heart transplant patients during routine cardiac biopsy sessions. In 15 transplanted patients the impedance test was repeated 2 or more times during different biopsy sessions. Thus a total of 50 measurements were performed in the transplant group. The impedance test was well tolerated and was completed in about 5-9 min.
Clinical characteristics of the studied patients are summarized in Table 1. As shown in the table, the age and body mass index were comparable among the transplant and control groups.
|
The two components of myocardial impedance (module and phase angle) were measured at alternating current frequencies of 10, 30,100, and 300 kHz. Table 2 shows the right ventriculo-atrial impedance difference recorded in the controls and in the transplant patients. The transplant group was divided into three subgroups according to the histopathological degree of graft rejection detected at the moment of the impedance test (grade 0: no rejection, 1A: light rejection, and 3A: moderate rejection).
|
3.1. Myocardial impedance in control patients
As shown in Table 2, the ventriculo-atrial myocardial impedance difference was about 21
at 10 kHz current frequency and it decreased progressively at higher current frequencies (from 21±8.4 at 10 kHz to 14±6.0 at 300 kHz, paired t-test p<0.01). Likewise, the ventriculo-atrial difference in the phase angle of the impedance ranged from positive values at low frequencies to negative values at higher frequencies (from 2.5±0.9° at 10 kHz to –3.8±2.1° at 300 kHz, paired t-test p<0.001).
3.2. Myocardial impedance in transplant patients with no histopathological signs of rejection
Transplant patients with no histopathological signs of rejection showed lower impedance values than non-transplant controls (Two Way ANOVA, p<0.001). As in the control patients, myocardial impedance decreased at higher alternating current frequencies (from 15±6.6 at 10 kHz to 7.3±4.0 at 300 kHz, paired t-test p<0.001). The magnitude and frequency dependence of the phase angle were comparable between controls and transplant patients with no rejection. In 6 patients with two or more repeated impedance measurements taken at stage 0 rejection, the intraindividual variability (absolute differences among individual repeated samples at 10 kHz) was not significantly different from variability among patients (median 6.39 vs 6.46 , p=0.38, respectively).
The earliest post-operative impedance measurements were recorded in 6 patients with no rejection 12 to 20 days after transplantation. At this stage we did not find significant impedance differences with respect to samples taken in 8 patients 100 to 300 days after transplantation (impedance at 10 kHz: 13.8±2.02 vs 16.9±2.8 , p=0.4, respectively).
3.3. Effects of graft rejection on myocardial impedance
Among the 29 transplanted patients, 16 (55%) presented histopathological features of light rejection (grade 1A) in at least one study and four patients showed moderate rejection (grade 3A) in at least one biopsy. As shown in Table 2, moderate degrees of graft rejection were associated with lower impedance values at all applied current frequencies. Fig. 2 illustrates graphically the mean and median values of the impedance module recorded in controls and in all three subgroups of transplanted patients. The figure shows a progressive decline in impedance from the highest values recorded in controls to the lowest values observed in patients with moderate rejection. The trend of these changes was statistically significant (ANOVA, p<0.001). Table 3 shows the median, the 25% and 75% percentile for each transplant subgroup as well as the H statistic and its corresponding p value. The differences between subgroups were most significant (p=0.003) for 
|Z|10 kHz.
|
|
All four patients with grade 3A rejection received intravenous corticosteroid therapy and recovered within one week. In 3 of these patients we repeated the impedance measurement at recovery and all showed a return to the non non-rejection range (case 1: from 3.11
to 11. 3 at 10 kHz; case 2: from 2.13 to 13.6 at 10 kHz; and case 3: from 3.92 to 12.5 at 10 kHz). In our patients we did not observe rejection levels greater than 3A. |
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. DiscussionFunding sourcesReferences |
|---|
This study shows that mild to moderate histopathological degrees of cardiac rejection are associated with changes in myocardial electrical impedance that can be safely detected by on-line catheter-based measurements in post-transplant patients.
4.1. Underlying mechanisms
The mechanism by which cardiac rejection modifies the magnitude and phase angle of myocardial electrical impedance can be inferred from previous studies from our group and others [5,8-10]. Indeed, upon applying an alternating current, the module of tissue impedance is mostly influenced by the extracellular passive electrical properties encountered by this current, whereas the phase angle is predominantly affected by cell membrane and gap junction resistivity. Mild cardiac rejection, as found in our study, causes alterations in extracellular oedema and leukocyte infiltration. Oedema is expected to reduce tissue electrical impedance as it has been demonstrated in arterially perfused rabbit papillary muscle preparations in which increases of interstitial volume induced by lowering the colloid osmotic perfusate pressure caused a drop in extracellular longitudinal resistance [8]. In porcine models of healed myocardial infarction we [2,11] have shown a reduction in the magnitude of tissue impedance in the necrotic scar tissue as it is composed of large extracellular collagen deposition and poor cell population. Moreover, atrial oedema produced by pulmonary vein ablation in patients with atrial fibrillation is also associated with a lowering of atrial impedance assessed by catheter-based atrial mapping [10].
By using an intracardiac electrocatheter configuration comparable to that employed in the present study, we previously reported that on-line measurement of the module and phase angle of myocardial impedance can differentiate areas of normal and necrotic myocardium [2]. Acute myocardial ischaemia elicits a marked increase in the impedance module and a large negative deviation of the phase angle caused by interstitial metabolite accumulation and increased cell gap junction resistivity [9]. In the latter stages of myocardial healing, the occurrence of cell infiltration, oedema, collagen deposition and cell necrosis cause a drop in scar tissue resistance [5,11].
Therefore, the most likely explanation for the decrease in right ventriculo-atrial impedance difference seen here in patients with graft rejection is that the inflammatory process caused a drop in myocardial impedance. The decrease in right ventriculo-atrial impedance difference could be alternatively explained by a rise in atrial myocardial impedance. However, this assumption is very unlikely because, as indicated above, rejection is expected to reduce and not to increase tissue impedance and, moreover, there is no plausible explanation to support the concept that rejection would affect the atrial myocardium more than the ventricular myocardium. The vast majority of our transplant patients underwent the bicaval technique and therefore the atrial and ventricular measurements corresponded to the same donor heart and could not be influenced by remnants of the native atrium.
In contrast to our results, episodes of heart rejection induced by lowering immunosuppressant drug therapy in dogs with heterotopic neck-heart transplantation were reported to be associated with an increase instead of a decrease in intramyocardial tissue impedance [3]. The differences between these two studies could be due to differences in the impedance measurement methodology or to the cardiac transplant model. In the dog model [3] tissue impedance was measured by applying current pulses through intramyocardial screw-in electrodes, whereas in our clinical study we used sinusoidal currents through an electrocatheter. Due to the anisotropy characteristics of the myocardium, tissue impedance is influenced by the direction of the current flow [12]. Moreover, the mechanical and metabolic conditions of the non-working perfused neck-transplanted heart in the dog model are not comparable to those of a working systemically-connected human heart. Indeed, during the rejection episodes the ectopic neck-transplanted heart might undergo certain degrees of coronary underperfusion and if this was the case, an ischaemia-induced rise in myocardial impedance could be expected [9].
Our study was not designed to analyze the immediate post-operative changes in myocardial impedance that might occur in the transplanted heart following surgical reestablishment of coronary perfusion. Indeed, our earliest measurements were taken 10 to 20 days after transplantation and at this stage the impedance values were not statistically different from those encountered 2 to 10 months later.
4.2. Clinical implications
Endocardial biopsy-guided surveillance of cardiac graft rejection is widely used in current clinical practice, but this procedure has important limitations. Repeated biopsy sampling of the relatively thin right ventricular cavity increases the risk for complications related to biopsy tissue removal and to the percutaneous vascular approach. Moreover, local biopsy-induced fibrotic reaction will lessen the probability of obtaining appropriate tissue specimens thus limiting the diagnostic value of the procedure.
In contrast to endocardial biopsies, the present pilot catheter-based measurement of impedance does not cause local tissue damage and therefore, repeated measures and accurate impedance mapping of the right ventricle can be performed safely.
During this investigation we did not have the opportunity to include patients with severe graft rejection. Our data encompass light to moderate degrees of rejection and although it is conceivable that more advanced stages of inflammation and oedema will further decrease myocardial impedance, extrapolation of the present data to cases of extreme rejection should be performed cautiously. Nevertheless, we believe that in light of these findings further clinical studies to validate the clinical applicability of impedance-based methods to detect heart graft rejection are now warranted. Moreover, less invasive impedance methods using for example a transoesophageal or a transthoracic approach can be also envisaged.
|
Funding sources |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. DiscussionFunding sourcesReferences |
|---|
This work has been supported by grants from the Spanish Ministry of Science and Technology and FEDER project SAF2001-1660 and SAF2005-02270 and from the Spanish Ministry of Health, Redes de Investigacion del Instituto de Salud Carlos III, (REDINSCOR).
|
References |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. DiscussionFunding sourcesReferences |
|---|
- Spanish Heart Transplantation Registry. 17th Official Report of the Spanish Society of Cardiology Working Group on Heart Failure, Heart Transplantation, and Associated Therapies (1984-2005). Rev Esp Cardiol (2006) 59:1283–1291.[CrossRef][Web of Science][Medline]
- Warren M., Bragos R., Casas O., et al. Percutaneous electrocatheter technique for on-line detection of healed transmural myocardial infarction. Pacing Clin Electrophysiol (2000) 23:1283–1287.[CrossRef][Medline]
- Pfitzmann R., Muller J., Grauhan O., Hetzer R. Intramyocardial impedance measurements for diagnosis of acute cardiac allograft rejection. Ann Thorac Surg (2000) 70:527–532.
[Abstract/Free Full Text] - Bragós R., Ramos J., Salazar Y., et al. Endocardial impedance spectroscopy system using a transcatheter method. Proc. XII–International Conference on Electrical Bioimpedance, Gdansk (2004) 465–468. Abstract.
- Cinca J., Warren M., Rodriguez-Sinovas A., et al. Passive transmission of ischemic ST segment changes in low electrical resistance myocardial infarct scar in the pig. Cardiovasc Res (1998) 40:103–112.
[Abstract/Free Full Text] - Salazar Y., Cinca J Rosell-Ferrer J. Effect of electrode locations and respiration in the characterization of myocardial tissue using a transcatheter impedance method. Physiol Meas (2004) 25:315–323.[CrossRef][Web of Science][Medline]
- Billingham M.E., Cary N.R.B., Hammond M.E., et al. A working formulation for the standardization of nomenclature in the diagnosis of heart and lung rejection: heart rejection study group. J Heart Transplant (1990) 9:587–593.[Medline]
- Fleischhauer J., Lehman L., Kléber A. Electrical resistances of interstitial and microvascular space as determinants of the extracellular electrical field and velocity of propagation in ventricular myocardium. Circulation (1995) 92:587–594.
[Abstract/Free Full Text] - Cinca J., Warren M., Carreno A., et al. Changes in myocardial electrical impedance induced by coronary artery occlusion in pigs with and without preconditioning: correlation with local ST-segment potential and ventricular arrhythmias. Circulation (1997) 96:3079–3086.
[Abstract/Free Full Text] - Lang C.C.E., Gugliotta F., Santinelli V., et al. Endocardial impedance mapping during circumferential pulmonary vein ablation of atrial fibrillation differentiates between atrial and venous tissue. Heart Rhythm (2006) 3:171–178.[CrossRef][Web of Science][Medline]
- Cinca J., Bardaji A., Carreno A., et al. ST segment elevation at the surface of a healed transmural myocardial infarction in pigs. Conditions for passive transmission from the ischemic peri-infarction zone. Circulation (1995) 91:1552–1559.
[Abstract/Free Full Text] - Salazar Y., Bragos R., Casas O., Cinca J., Rosell J. Transmural versus nontransmural in situ electrical impedance spectrum for healthy, ischemic, and healed myocardium. IEEE Trans Biomed Eng (2004) 51:1421–1428.[CrossRef][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||