Intracardiac impedance monitors stroke volume in resynchronization therapy patients

Transcription

1 Europace (2010) 12, doi: /europace/euq045 CLINICAL RESEARCH Pacing and CRT Intracardiac impedance monitors stroke volume in resynchronization therapy patients Mario Bocchiardo 1, Dorothee Meyer zu Vilsendorf 2, Carmelo Militello 3, Michael Lippert 3 *, Gerald Czygan 3, Fiorenzo Gaita 1, Patrick Schauerte 4, and Christoph Stellbrink 2 1 Cardiology Division, Ospedale Cardinal Massaia, Asti, Italy; 2 Staedt. Kliniken Bielefeld, Klinikum Mitte, Bielefeld, Germany; 3 Biotronik SE & Co. KG, Hartmannstraße 65, Erlangen, Germany; and 4 Universitaetsklinikum Aachen, Aachen, Germany Received 2 December 2009; accepted after revision 26 January 2010; online publish-ahead-of-print 25 February 2010 Aims Monitoring of haemodynamic parameters or surrogate parameters of the left ventricle is especially important for patients under cardiac resynchronization therapy (CRT). Intracardiac impedance reflects left ventricular (LV) volume changes well in animal models. Since it is unknown whether this also holds in humans with heart failure (HF), we examined the correlation of LV intracardiac impedance with haemodynamic parameters in CRT patients for different positions of the LV lead.... Methods In 14 HF patients with non-ischaemic cardiomyopathy (four female, age years, NYHA , EF %), and results one or two suitable implantation sites for the LV lead were selected. Following atrial, right ventricular, and LV catheter positioning, a micro-manometer catheter was placed in the ascending aorta. Surface ECG, impedance, and aortic pressure were recorded during graded overdrive bi-ventricular pacing in DDD mode. The correlation between impedance and stroke volume (SV) or pulse pressure (PP) changes was compared for different LV lead positions. In total, 20 overdrive pacing tests were performed at six different LV lead positions. Strong correlations were found between stroke impedance (SZ) and SV (R ¼ ) as well as between SZ and PP (R ¼ ) without significant influence of LV lead position.... Conclusion In HF patients, a strong correlation between changes in intracardiac impedance and LV SV was found. Typical LV lead implant positions have been tested and all appear to be suitable for this method of LV volume monitoring Keywords Heart failure Pacing Haemodynamics Cardiac volume Pressure Introduction Heart failure (HF) is one of the most frequent diagnoses upon hospitalization and is a major cause of death. Up to one-quarter of patients with HF exhibit a left bundle branch block. 1,2 Cardiac resynchronization therapy (CRT) using an implantable device for synchronous bi-ventricular stimulation leads to acute improvement of haemodynamic parameters 3 6 as well as to long-term improvement of physical capacity and quality of life 7 10 and to a reverse modelling of the ventricle Mortality and morbidity are also reduced. 9,14,15 A reduction of hospitalization rate may be achieved by devicebased monitoring of disease progression, generating a warning before acute decompensation. Thoracic fluid can be detected by implants with thoracic impedance measurement, 16 but fluid formation occurs only secondary to deteriorating haemodynamics. In CRT patients, electrodes implanted in the right ventricle and on the outer side of the left ventricle are available, which allow an intracardiac impedance measurement across the left ventricle, which contains information on cardiac dimensions. It has been shown in acute and chronic animal experiments with healthy as well as with failing hearts 17,18 that an implanted device can detect left ventricular (LV) volume changes via LV intracardiac impedance measurements. One of the questions to be answered before this technique is applied for heart volume monitoring is whether it works in CHF patients with LV leads implanted at positions optimized for resynchronization therapy. * Corresponding author. Tel: þ ; fax: þ , michael.lippert@biotronik.com Published on behalf of the European Society of Cardiology. All rights reserved. & The Author For permissions please journals.permissions@oxfordjournals.org.

2 Intracardiac impedance monitors SV in CRT patients 703 In this study, we tested acutely whether intracardiac impedance is able to detect provoked changes in stroke volume (SV) in CRT patients, and we investigated the influence of LV lead position on this ability. Methods Patients Fourteen patients with dilated cardiomyopathy of non-ischaemic origin were included in an acute prospective multicentre study. All study participants granted their written informed consent prior to inclusion. The study was approved by the local institutional Ethics Committees and was conducted in compliance with the Declaration of Helsinki. Main inclusion criteria were ejection fraction EF 35%, QRS width 120 ms, left bundle branch block, left ventricular end-diastolic diameter (LVEDD) 55 mm, and age 50 years. Patients with known myocardial infarction, artificial valve, atrial fibrillation, and severe stenosis of the femoral or iliac artery were excluded. The patient data are listed in Table 1. Clinical procedure The measurements for each patient were performed either during an electrophysiological study (EPS) procedure (eight patients) or during the implantation of a CRT defibrillator (four patients) or CRT pacemaker (two patients), respectively. Under local anaesthesia, three electrodes for pacing, sensing, and for impedance measurements were put in place: in the right atrium a 6F Viacath (Biotronik SE & Co. KG, Berlin, Germany) or a bipolar pacemaker lead; in the apex of the right ventricle a 6F Viacath (Biotronik) or a pacemaker/defibrillator lead; and in the LV coronary sinus (CS) a 2.5F Pathfinder (Cardima Inc., Fremont, USA) or a bipolar CS pacemaker lead. An 8F tip manometer catheter for measurement of aortic blood pressure (SPC 780-C, Millar Instruments Inc., Houston, USA) was positioned in the ascending aorta. For the EPS patients, the protocol was performed sequentially at two different LV lead positions that were selected to be suitable for an LV-CS lead implantation. For the measurements, during implantation only one position, the actual CS lead implant position (chosen on the larger delay between the right and left ventricle local electrogram recorded from the tip of the leads), was used. At each LV lead position, a pacing protocol was performed with bi-ventricular pacing in DDD mode, AV delay 100 ms, and VV delay 0 ms. In order to provoke a change in SV, pacing was performed at the pacing rates 80, 100, 120, 140, and 40 ppm (i.e. effectively in VAT mode) in a non-monotonic sequence. Not all rates were available for all patients due to high sinus rate, partial loss of LV capture, technical reasons (caused by high LV capture threshold in one patient), or highest rate not applicable due to blood pressure drop. In average, more than four rates were measured per patient. Each rate was paced for 2 min. Only the last 30 s were used for the analysis in order to allow for a settling phase before measurement. Three-chamber pacing and impedance measurements were performed by an external custom-built device (ETIP, Biotronik). The impedance data were transmitted continuously from the ETIP device to the programmer (PMS1000, Biotronik) and digitally recorded by a laptop computer, together with ECG and pacemaker marker data. The pressure catheter was connected via an isolation amplifier (PCU-2000, Millar Instruments) with a data acquisition system (Powerlab, ADInstruments Ltd, Oxfordshire, UK). A sampling rate of 1000 Hz per channel was applied. Impedance measurement The 4-terminal intracardiac impedance was measured via the RV coil and the RV tip for current injection and LV-CS proximal and distal electrodes for voltage measurement. For the EPS patients, corresponding electrodes on the RV and LV catheters with 30 and 18 mm, respectively, electrode spacing were selected. The RV catheter electrode distance of 30 mm was chosen in order to generate similar current distributions with the EPS catheters and with the ICD electrodes, which was verified by numerical model calculations. This impedance measurement configuration was chosen since it has shown a strong correlation between impedance parameters and SV in experimental settings. 17,18 Constant amplitude (600 ma) biphasic symmetric rectangular current pulses (15 ms for each polarity) were injected every 8 ms. The resulting voltage was sampled and demodulated, amplified, low-pass filtered (40 Hz), and digitized (8 bit) with a smallest possible resolution of 13 mohm. Data evaluation The impedance and pressure data were synchronized and analysed offline using custom software (Matlab TM, The Mathworks Inc., Natick, USA). Extrasystoles and beats with pressure signal artefacts were excluded from analysis. Averaged impedance and pressure curves were calculated over 10 s intervals for reducing the influence of respiration that appeared in both pressure and impedance curves (Figure 1). The impedance curves were additionally smoothed. From the averaged aortic pressure curves, the SV was calculated using a pulse contour method based on the pressure integral between the time of minimum pressure and the dicrotic notch. 19,20 The pulse pressure (PP) was determined as the difference of the maximum and minimum pressure values from the Table 1 Demographic and haemodynamic data for the study patients Patient Sex Age (years) NYHA class EF (%) LVEDD (mm) QRS (ms)... Mean 10/14 male SD EF, LV ejection fraction; LVEDD, LV end-diastolic diameter; QRS, ECG QRS width.

3 704 M. Bocchiardo et al. Figure 1 Example of raw data recording (ECG, surface electrocardiogram; AortaP, aortic blood pressure; Z, intracardiac impedance. Vertical lines 100 ms in front of the QRS complex). The respiratory influence is clearly visible in the pressure and impedance traces. averaged aortic pressure curve. From the impedance curves, the stroke impedance SZ ¼ ESZ 2 EDZ was calculated, where the end-systolic impedance (ESZ) is the maximum during systole and the end-diastolic impedance (EDZ) is the minimum during diastole. Owing to the 30 s analysed measurement period and 10 s averaging, three samples are available for each individual heart rate. A linear regression analysis was performed between SZ and SV as well as between SZ and PP for each patient and LV lead position and the correlation coefficient R was determined. The standard deviation s of the distance of data points from the regression line (residuals) is a measure for the scatter of the SZ and SV values obtained after 10 s averaging, assuming a linear relation between SZ and SV. It was calculated as the fraction of the average value at the lowest rate. Classification of left ventricular coronary sinus lead sites The LV lead positions were classified in sections labelled by a combination of anterior, lateral, or posterior and basal, medial or apical. The positions anterior and anterior lateral were pooled under anterior, and the position posterior lateral was labelled posterior in figures and tables. The anterior, lateral, and posterior apical regions were merged to a single apical section. The authors had full access to the data and take responsibility for their integrity. All authors have read and agreed to the manuscript as written. Results Correlation of impedance and haemodynamics Both SV and PP were reduced by increasing the pacing rate for all patients, Figure 2. Between the lowest and highest heart rate, a Figure 2 Haemodynamic effects of overdrive pacing. Stroke volume SV (left panel) and the pulse pressure PP (right panel) behaviour with heart rate (HR) variation for all patients and lead sites. relative reduction of % was found for SV and % for PP. The individual regression results are listed in Table 2. No significant difference was found between the correlation coefficients during EPS (patients 1 8) and during implantation (patients 9 14) (P ¼ 0.48). For the correlation SZ SV, the patients exhibited correlation coefficients of and, besides an outlier (patient 11), were within a range of The measurement variability of SZ measurements was characterized by the standard deviation of residuals s as the fraction of the low rate value (Table 2). The average value for s was %. The results for the correlation SZ PP were very similar (Table 2). Pooling all patients and positions together resulted in R ¼ 0.74, P, 0.001, s ¼ 28% for SZ SV and R ¼ 0.70, P, 0.01, s ¼ 30%

4 Intracardiac impedance monitors SV in CRT patients 705 Table 2 Left ventricular lead positions and correlation results for all patients Patient LV position SZ SV... SZ PP... R s (%) R s (%)... 1 Apical 0.96 * * 17 1 Medial lateral 0.90 * * 17 2 Medial posterior 0.85 * * 14 2 Medial anterior 0.91 * * 21 3 Medial lateral 0.94 * * 22 3 Medial posterior 0.98 * * 13 4 Medial lateral 0.96 * * 4 4 Medial anterior 0.79 * * 17 5 Medial posterior 0.84 * * 32 6 Medial posterior 0.75 * * 24 6 Medial posterior 0.62 * * 28 7 Medial lateral 0.90 * * 13 7 Apical 0.72 * * 66 8 Medial lateral 0.70 * * 24 9 Basal anterior 0.78 n.s n.s Basal lateral 0.85 * * 9 11 Medial lateral 0.26 n.s n.s Basal lateral 0.82 * * 9 13 Medial posterior 0.95 * * Medial posterior 0.92 * * 9 Mean SD SZ, stroke impedance, SV, stroke volume, PP, pulse pressure, R, correlation coefficient, s, normalized standard deviation of residuals, n.s., not significant. *P, for SZ PP. The R-values were smaller than those for the individual correlations since the slope and offset of the regression lines were not the same for all patients. Influence of lead implant site The most common LV lead sites were in the medial sectors (65% of measured positions). In the medial posterior sector, one patient is listed twice, since both positions, despite being different by some centimetres, were still characterized as medial posterior lateral. Grouping the correlation coefficients according to the implant position sector showed no tendency to favour certain regions: all averages of R are between 0.74 and 0.85 (Figure 3A and B). The accuracy and the signal size, however, exhibited better results in the medial sectors (Figure 3C and D). Generally, the differences were too small to become statistically relevant. Complications One patient suffered cerebral embolism after completion of the EPS with partial recovery. Routine laboratory results had shown no coagulation disorders, pre-existing LV thrombi have been excluded by echocardiography, and the coagulation parameters have been closely controlled during the procedure. No other major complications occurred. Discussion Cardiac resynchronization therapy has become an accepted therapy for patients with HF and ventricular conduction delay in the last years. Long-term fluid monitoring of CHF patients by implants using the thoracic impedance measurement is considered clinically useful, but also discussed controversially due to the high false alarm rate. 21,22 Using information on cardiac condition instead of, or additionally to, information on thoracic fluid would probably improve sensitivity and specificity for predicting cardiac decompensation. Thus, a device that measures surrogate haemodynamic parameters with a good correlation to standard haemodynamic parameters would offer an attractive approach for disease monitoring. In fact, SV is one of the haemodynamic parameters that is improved by CRT. 5,6 Cardiac impedance depends on the fluid volume changes in the LV throughout the cardiac cycle. This can be measured between the RV and LV leads implanted. Therefore, in this study, the correlation between SV and intracardiac SZ has been investigated during a bi-ventricular pacing protocol. Acute changes of SV were provoked in each patient by using various pacing rates between sinus rate (VAT pacing) and DDD pacing from 80 to 140 ppm. Different CS lead implantation sites have been used, but RV lead position was always apical. For the correlation SZ SV, an average correlation coefficient of

5 706 M. Bocchiardo et al. Figure 3 Upper panels: linear correlation coefficients R, grouped for left ventricular lead position. Mean values, standard deviations, and number of measurements for the stroke impedance stroke volume (SZ SV) (A) and stroke impedance pulse pressure (SZ PP) (B) correlations. Lower panels: scatter s around regression line. Mean values and standard deviations of the normalized residuals s for the SZ SV (C) and SZ PP (D) correlations. R ¼ was found, close to the correlation coefficient between impedance and volume parameters found in an earlier experimental study (R ¼ , rate test in Zima et al. 17 ). Intracardiac impedance measurement The measured impedance signal in the described 4-terminal configuration depends on the conductivity of the tissue between and in the vicinity of the involved electrodes, on the spacing between the electrodes, and on the tilt angle between the RV and LV leads. The current distribution and the associated electric field generated by current injection over the two RV electrodes resemble a distorted dipole field. The strength of this field is measured between the two LV electrodes. Generally, an increase in the distance between RV and LV leads and an increase in blood volume between them decrease the measured signal. Since the mutual arrangement of the electrodes and its dynamics during contraction or dilatation depend on the anatomical position of the LV lead, the relation between SZ and LV volume will also depend on it. However, after the in-growth phase, this relation is stable. One advantage of this impedance measurement configuration is that local effects (from tissue motion around the current-injecting electrodes) are not dominating the signal, since current-injecting electrodes and voltage measurement electrodes are far apart. Intracardiac impedance is a sufficient surrogate haemodynamic parameter, if it correlates with LV volume for all usual LV lead implant positions. Slope and offset may be different, since each patient serves as his own reference. Therefore, the correlation coefficient and the scatter around the regression line (the accuracy) were investigated as the crucial parameters for application. Correlation between impedance and haemodynamic parameters For healthy subjects, overdrive pacing decreases the SV with increasing rate as the end-diastolic volume is reduced, whereas the end-systolic volume is nearly unchanged. For HF patients, in contrast, end-diastolic volume remains unchanged during overdrive pacing, but end-systolic volume increases. 23 Hasenfuss et al. 23 found a reduction in SV index of 48% between pacing at 84 and 140 ppm in patients with dilated cardiomyopathy, which is nearly the same value as found in this study. The correlation between SZ and SV was compared for LV lead implant positions in the range of the usual implantation sites. All investigated LV-CS electrode implant position sectors appear to be suitable for LV volume monitoring. The values for the scatter s were %. The accuracy was limited by the impedance measurement resolution and the setup with acutely placed LV leads that are prone to small position changes during the measurement. For grown-in electrodes, more stable results are expected. Furthermore, also the reference method of pulse contour analysis has its own limited accuracy, 19

6 Intracardiac impedance monitors SV in CRT patients 707 which represents one major component of the measured standard deviation of residuals. Limitations The study investigated the short-term correlation between SZ and SV in an acute setting. The results are thus relevant for short-term applications such as detection of fast changes like acute decompensation or for optimization of pacing parameters. Long-term stability is to be investigated in the next step. The changes in SV were generated by overdrive pacing, which is an artificial setting predominantly modifying the end-systolic volume in these patients. It remains to be proved whether the method also can detect haemodynamic changes caused by other reasons. One of the patients is a distinct outlier (patient 11) showing a correlation coefficient of R ¼ It needs to be investigated in studies incorporating implantable devices with the capability to measure 4-terminal intracardiac impedance whether this result was due to lead motion or whether this may also occur with grown-in leads in a fraction of patients. Conclusion In HF patients with non-ischaemic aetiology, a strong correlation between changes in intracardiac impedance and LV SV was shown during variation of heart rate with bi-ventricular pacing. Different LV lead implant positions have been tested and all appear to be suitable for the purpose of LV volume monitoring. Funding The study was supported by Biotronik SE & Co. KG. Conflict of interest: C.M., M.L., and G.C. are employees of Biotronik SE & Co. KG; C.S. is a sponsored investigator for Biotronik. The other authors report no conflicts of interest. References 1. Doval HC, Nul DR, Grancelli HO, Perrone SV, Bortman GR, Curiel R. Randomised trial of low-dose amiodarone in severe congestive heart failure. Grupo de Estudio de la Sobrevida en la Insuficiencia Cardiaca en Argentina (GESICA). Lancet 1994;344: Farwell D, Patel NR, Hall A, Ralph S, Sulke AN. How many people with heart failure are appropriate for biventricular resynchronization? Eur Heart J 2000;21: Foster AH, Gold MR, McLaughlin JS. Acute hemodynamic effects of atriobiventricular pacing in humans. Ann Thorac Surg 1995;59: Leclercq C, Cazeau S, Le Breton H, Ritter P, Mabo P, Gras D et al. Acute hemodynamic effects of biventricular DDD pacing in patients with end-stage heart failure. J Am Coll Cardiol 1998;32: Auricchio A, Stellbrink C, Block M, Sack S, Vogt J, Bakker P et al. Effect of pacing chamber and atrioventricular delay on acute systolic function of paced patients with congestive heart failure. Circulation 1999;99: Kass DA, Chen CH, Curry C, Talbot M, Berger R, Fetics B et al. Improved left ventricular mechanics from acute VDD pacing in patients with dilated cardiomyopathy and ventricular conduction delay. Circulation 1999;99: Cazeau S, Leclercq C, Lavergne T, Walker S, Varma C, Linde C et al., for the Multisite Stimulation in Cardiomyopathies (MUSTIC) Study Investigators. Effects of multisite biventricular pacing in patients with heart failure and intraventricular conduction delay. N Engl J Med 2001;344: Auricchio A, Stellbrink C, Sack S, Block M, Vogt J, Bakker P et al., for the Pacing Therapies in Congestive Heart Failure (PATH-CHF) Study Group. Long-term clinical effect of hemodynamically optimized cardiac resynchronization therapy in patients with heart failure and ventricular conduction delay. J Am Coll Cardiol 2002;39: Abraham WT, Fisher WG, Smith AL, Delurgio DB, Leon AR, Loh E et al., for the MIRACLE Study Group. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002;346: Stellbrink C, Breithardt OA, Franke A, Sack S, Bakker P, Auricchio A et al. Impact of cardiac resynchronisation therapy using hemodynamically optimized pacing on left ventricular remodeling in patients with congestive heart failure and ventricular conduction disturbance. J Am Coll Cardiol 2001;38: Sutton MG, Plappert T, Abraham WT, Smith AL, DeLurgio DB, Leon AR et al., for the Multicenter InSync Randomized Clinical Evaluation (MIRACLE) Study Group. Effect of cardiac resynchronization therapy on left ventricular size and function in chronic heart failure. Circulation 2003;107: Saxon LA, De Marco T, Schafer J, Chatterjee K, Kumar UN, Foster E, for the VIGOR Congestive Heart Failure Investigators. Effects of long-term biventricular stimulation for resynchronization on echocardiographic measures of remodeling. Circulation 2002;105: Yu CM, Chau E, Sanderson JE, Fan K, Tang MO, Fung WH et al. Tissue Doppler echocardiographic evidence of reverse remodeling and improved synchronicity by simultaneously delaying regional contraction after biventricular pacing therapy in heart failure. Circulation 2002;105: Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De Marco T et al., for the Comparison of Medical Therapy, Pacing, Defibrillation in Heart Failure (COMPA- NION) Investigators. Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004; 350: Cleland JGF, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L et al., for the Cardiac Resynchronization Heart Failure (CARE-HF) Study Investigators. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005;352: Yu CM, Wang L, Chau E, Chan RH, Kong SL, Tang MO et al. Intrathoracic impedance monitoring in patients with heart failure: correlation with fluid status and feasibility of early warning preceding hospitalization. Circulation 2005;112: Zima E, Lippert M, Czygan G, Merkely B. Determination of left ventricular volume changes by intracardiac conductance using a biventricular electrode configuration. Europace 2006;8: Stahl C, Beierlein W, Walker T, Straub A, Nagy Z, Knubben K et al. Intracardiac impedance monitors hemodynamic deterioration in a chronic heart failure pig model. J Cardiovasc Electrophysiol 2007;18: Wesseling KH, DeWit B, Weber JAP, Smith NT. A simple device for the continuous measurement of cardiac output. Adv Cardiovas Phys 1983;5: Gratz I, Kraidin J, Afshar M. Continuous cardiac output by pulse waveform. Int Anesthesiol Clin 1993;31: Ypenburg C, Bax JJ, van der Wall EE, Schalij MJ, van Erven L. Intrathoracic impedance monitoring to predict decompensated heart failure. Am J Cardiol 2007;99: Vollmann D, Nagele H, Schauerte P, Wiegand U, Butter C, Zanotto G et al. Clinical utility of intrathoracic impedance monitoring to alert patients with an implanted device of deteriorating chronic heart failure. Eur Heart J 2007;28: Hasenfuss G, Holubarsch C, Hermann HP, Astheimer K, Pieske B, Just H. Influence of the force frequency relationship on haemodynamics and left ventricular function in patients with non-failing hearts and in patients with dilated cardiomyopathy. Eur Heart J 1994;15: