Optimization of Atrioventricular and Interventricular Delay With Acoustic Cardiography in Biventricular Pacing

Cardiac resynchronization therapy (CRT) improves hemodynamic and echocardiographic parameters, symptoms, quality of life, morbidity, and mortality in patients with medically refractory congestive heart failure associated with a prolonged QRS duration.1-5 To fully exhaust the benefits of CRT, it is important to optimize atrioventricular (AV) and interventricular (W) conduction delays to achieve optimal mechanical synchronization of the heart chambers.6 This has mainly been done using parameters obtained by Doppler echocardiography, but also by measuring left ventricular (LV) dP/dtmax.7-11 However, an experienced echocardiographer and a stable patient are needed to get reproducible results. These examinations are time-consuming, and no standard parameter for optimization is yet available. For these and cost reasons, even today, only a minority of CRT devices are optimized after implantation. Acoustic cardiography (AUDICOR, Inovise Medical Inc., Portland, OR) has been developed to measure time intervals very precisely and with reproducible results. Due to the fact that the strength of the S3 and the electromechanical activation time (EMAT) correlate well with LV function, we used acoustic cardiography for optimizing the W and AV delays in a patient with a biventricular pacemaker.12, 13 Five weeks after this optimization, we also investigated the effects on the Doppler echocardiographic parameters, B-type natriuretic protein (BNP) value, and the functional capacity of this patient. A 67-year-old man presented with New York Heart Association (NYHA) class II dyspnea and a markedly reduced ejection fraction (EF) of 28% following a prior aortic valve replacement in 2001, at which time his EF was normal. The electrocardiogram showed a left bundle branch block morphology and a QRS duration of 180 milliseconds. Doppler echocardiography revealed a VV contraction delay of 40 milliseconds, measured as the difference between the onset of the pulmonary ejection wave and the aortic ejection wave, and an intraventricular septal-posterolateral delay of 200 milliseconds measured with displacement imaging and autotracking (Aplio, Toshiba Medical Systems, New York, NY). There was eccentric LV hypertrophy with an end-diastolic diameter of 67 mm (normal < 60 mm) and an LV mass index of 212 g/m2 (normal <134 g/m2). The patient had known arterial hypertension, which was medically well controlled. No malignant arrhythmias were induced during electrophysiology studies, and a biventricular pacemaker device (Stratos LV, BIOTRONIK, Inc., Berlin, Germany) was implanted. The coronary sinus was cannulated using an electrophysiology catheter, and the optimal posterolateral vein could be identified and cannulated. Postimplantation, medical therapy consisted of an angiotensin-converting enzyme inhibitor, a β blocker, spironolactone, loop diuretics, digoxin, and amiodarone. The patient was orally anticoagulated due to reduced LV function. Medical therapy was not changed during the observation period. There were no clinical signs of congestive heart failure postimplantation. The patient was enrolled in our CRT optimization program 5 weeks after the implantation. The AV delay was set at 120 milliseconds, and simultaneous ventricular pacing was programmed to a baseline W setting. Rate-dependent shortening of the AV interval was turned off because the benefit of this feature has been questioned recently in biventricular pacemakers.14 During optimization and the follow-up period, the patient was in sinus rhythm. Sitting blood pressure was 93/70 mm Hg. Cardiac examination showed no clinical signs of congestive heart failure. After the baseline programming, the patient was sent home for 6 weeks. Eleven weeks postimplantation, the patient's CRT device was optimized using acoustic cardiography (AUDICOR technology). The acoustic cardiography data were obtained using AUDICOR sensors attached to the V3 and V4 positions. For each CRT AV and VV delay combination, a full AUDICOR test (10-second data recording) was recorded and analyzed, and the results were trended for interpretation. The recorded data included the presence and strength of the S3, the EMAT (the interval from the onset of the Q wave of the electrocardiogram to the S1), and the LV systolic time (the interval from S1 to S2). Forty-five different settings were programmed for evaluation using the AUDICOR data (with possible combinations of 100, 125, 150, 175, 200, 225, 250, 275, and 300 milliseconds for AV delays and RV40, RV20, 0, LV20, and LV40 milliseconds for VV delays). Then, Doppler echocardiographic and AUDICOR data were collected for the baseline programming and intrinsic conduction without pacing. The Doppler echocardiographic parameters are summarized in the Table. The septal-posterolateral intraventricular delay was measured with new displacement imaging software with autotracking. Displacement curves are used by integrating myocardial Doppler velocities of the medial and lateral mitral annulus, thus forming an apical four-chamber view (Aplio). A full-volume acquisition of the left ventricle (transthoracic three-dimensional echocardiography) was analyzed offline to create global and segmental time-volume curves. A systolic dyssynchrony index was created based on the dispersion of times to minimum volume for each segment (iE 33 system, Philips Medical Systems, Andover, MA). BNP was measured after a resting period of at least 1 hour and, thereafter, symptom-limited spiroergometry was performed. The patient was discharged with the optimized AV and VV delays for another 6 weeks using the lowest EMAT as a surrogate for optimal contractility and timing (right ventricular stimulation 40 milliseconds before LV stimulation and an AV delay of 250 milliseconds). Because the best parameter combination could not be programmed permanently in the Stratos LV, optimization was performed by programming a VV delay of −40 milliseconds and the highest possible AV delay of 160 milliseconds. Seventeen weeks postimplantation, AUDICOR and echocardiographic data were taken again, BNP was measured, and symptom-limited spiroergometry was carried out. Doppler echocardiographic data and results from the other examinations with intrinsic, standard, and optimized programming are shown in the Table. Maximum exercise capacity was 103 W with standard programming and 115 W with optimized programming. Oxygen uptake rose from 22.4 mL/min/kg to 26.9 mL/min/kg. Three-dimensional EF increased from 32% to 45%, and better synchronization was obvious just by looking at the apical four-chamber view. In addition, the BNP value decreased from 99 ng/L to 69 ng/L. Encouraging studies comparing AUDICOR data with invasive measurements obtained during LV catheterization indicate that EMAT correlates with dP/dtmax in subjects with signs of dyssynchrony (i.e., wide QRS complexes) and LV systolic time correlates with EE15 Thus, EMAT was used as a primary parameter to determine the best AV/VV delay combination (Figure). The hypothesis that minimizing EMAT might be beneficial is supported by the work of Jansen and colleagues,16 which showed that the time to onset of systolic velocity measured with tissue Doppler imaging might be a better parameter to predict reverse remodeling than the time to peak velocity. In this case, we were able to achieve a better hemodynamic state through optimized programming, and exercise capacity, maximum oxygen uptake, and ejection fraction were improved by achieving better synchronization. Also, the BNP level decreased, indicating lower filling pressure after optimized delays, although the patients remained in NYHA class II before pacemaker implantation and throughout the entire observation period. The relatively low BNP value (<100 ng/L) before delay optimization documents the optimal medical therapy in these patients.17 Therefore, this case report also underlines that clinical judgment alone is not sufficient for evaluating the effects of synchronization therapy. Schematic of acoustic cardiography and left ventricular pressure tracings for a normal patient and a heart failure patient showing a phonocardiographic S3, an increased electromechanical activation time (EMAT) (Q–S1), reduced left ventricular systolic time (LVST) (S1–S2), and a higher left ventricular diastolic pressure. ECG=electrocardiogram; Ao=aorta; LA=left atrium; LV=left ventricle; PEP=pre-ejection period; LVET=left ventricular ejection time; IVCT=isovolumic contraction time; IVRT=isovolumic relaxation time In comparison with Doppler echocardiography, optimization of biventricular pacing seems to be achievable through the fast and easy use of acoustic cardiography (AUDICOR device), yielding results that are easy to interpret, and importantly, independent of the person operating the device. This case report indicates the potential of this new acoustic cardiography approach to optimizing biventricular pacemakers for heart failure therapy. However, there might be a remodeling process following not just CRT implantation, but also changes to the AV and VV interval programming. We therefore suggest waiting several weeks before carrying out exercise tests after optimizing AV and W delays. Acknowledgment: We acknowledge the assistance of Dr. Peter Bauer, Inovise Medical, Inc., Portland, OR, for data analysis support.