Sequential Biventricular Pacing

Sequential Biventricular Pacing: Evaluation of Safety and Efficacy PETER T. MORTENSEN,* PETER SOGAARD,* HASSAN MANSOUR,† JEAN PONSONAILLE,† DANIEL GRAS,‡ ARNAUD LAZARUS,§ WOLFGANG REISER, CHRISTINE ALONSO,# CECILIA M. LINDE,** MAURIZIO LUNATI,†† BERTHOLD KRAMM,‡‡ and E. MARK HARRISON‡‡ From *Skejby Sygehus, Aarhus, Denmark, †CHU Hospice St Jacques—Hopital G. Montpied, Clermont-Ferrand, ˆ ‡Polyclinique Saint Henri, Nantes, §Clinique Bizet, Paris, France, Cardio Clinic Nurnberg, Nurnberg, Germany, ¨ ¨ #CHU, Rennes, France, **Karolinska Hospital, Stockholm, Sweden, ††Opsdedale Niguarda, Milano, Italy, and ‡‡Medtronic Bakken Research Center, Maastricht, the Netherlands

MORTENSEN, P.T., ET AL.: Sequential Biventricular Pacing: Evaluation of Safety and Efficacy. The study evaluated the clinical safety, performance, and efficacy of sequential biventricular pacing in the InSync III (Model 8042) biventricular stimulator in a multicenter, prospective 3-month study and assessed the proper functioning of features aiming at improving biventricular AV therapy delivery. The system was successfully implanted in 189 (95.9%) of 198 patients with symptomatic systolic heart failure and a prolonged QRS complex duration. Patients significantly improved their 6-minute hall walk distance (baseline 339 ± 92 vs 3-month 422 ± 127 meter, P < 0.001) and NYHA class (baseline 3.1 ± 0.5 vs 3-month 1.9 ± 0.7, P < 0.001). Echocardiographic optimization of sequential biventricular pacing showed an improvement in stroke volume compared to simultaneous stimulation (sequential 68 ± 24 mL vs simultaneous 56 ± 23 mL, P < 0.001) at baseline and at 3 months. In 88% (30/34) of the patients these improvements were seen within a small range of V-V delays of ±20 ms and in 94% (32/34) within V-V delays of ±40 ms. In contrast, programming beyond this range reduced stroke volume below that during simultaneous biventricular pacing. The device functioned as expected. LV lead dislodgement was observed in 12 patients and phrenic nerve stimulation required lead repositioning in 2 patients. Eight patients died during the study. Patient survival at 3 and 6 months was 97 ± 2% and 94 ± 2%, respectively. Cause of death was cardiac (n = 7), heart failure related (n = 3), arrhythmia related (n = 2), and unknown (n = 2). In conclusion, this sequential biventricular pacemaker was safe and efficacious. (PACE 2004; 27:339–345) pacing, heart failure, biventricular pacing, sequential biventricular pacing Introduction Cardiac resynchronization therapy (CRT) with biventricular pacing has been shown to improve left ventricular (LV) function and clinical performance in patients with severe heart failure due to dilated cardiomyopathy (DCM) or ischemic cardiomyopathy and left bundle branch block.1−4 The mechanism of action of CRT is thought to be LV electromechanical resynchronization (EMR)5−7 with reduction of secondary mitral incompetence.8,9 However, in a series of multicenter trials,1−4 around 20% of patients had negligible or even negative effects of CRT as measured by LV ejection fraction (LVEF) and 6-minute hall walk (6MHW). This could be due to nonoptimal patient selection and/or incomplete EMR.10−12 Apart from the etiology, other factors might influence the effects of EMR including right ventricular (RV) and

Supported in part by Medtronic Inc., Minneapolis, MN. Address for reprints: Peter T. Mortensen, M.D., Skejby Sygehus, Aarhus, Denmark Hjertemedicinsk, afd. B, Brendstrupgaardsvej, 8200 Arhus N., Denmark. E-mail: [email protected]

PACE, Vol. 27

LV pacing sites, programmed atrioventricular (AV) delay, and the actual timing of ventricular activation during CRT, due to, for example, areas of delayed conduction.13−15 To date, only biventricular pacemakers with simultaneous RV and LV pacing (i.e., the interventricular [V-V] delay set to 0 ms) have been available. With the introduction of the InSync III (Model 8042, Medtronic, Inc., Minneapolis, MN, USA) device, it has become possible to optimize EMR with programmable V-V timing. This clinical study was conducted to evaluate the efficacy and safety of the InSync III device, which is capable of sequential RV and LV activation.

Methods Study Design The study was a multicenter (Europe and Canada), prospective, nonrandomized observational 3-month trial in patients with symptomatic heart failure New York Heart Association class (NYHA) II-IV due to LV systolic dysfunction (LVEF ≤ 0.35; LV end-diastolic dimension [LVEDD]

March 2004

339

MORTENSEN, ET AL.

≥55 mm) and evidence of ventricular dyssynchrony as evidenced by a QRS complex duration ≥130 ms (QRS duration ≥180 ms for pacemaker dependent patients). Patients with an indication for an implantable cardioverter defibrillator (ICD) or with chronic atrial tachyarrhythmias or mechanical right heart valves were excluded from this study. All patients agreed to participate and signed a written, informed consent that was approved by the medical ethics committee, which approved the study.

Device Description The InSync III device (Model 8042) is an atrial biventricular pacing device for cardiac resynchronization. The InSync III provides sequential biventricular pacing for cardiac resynchronization to adjust the timing of inter- and intraventricular activation of patients with heart failure who have ventricular conduction disturbances like interventricular conduction delays that prevent synchronous activation under simultaneous biventricular pacing (Fig. 1A). For sequential biventricular pacing, either ventricle may be paced first with up to an 80-ms delay from the first ventricular pace to the second. V-V programming may thus alter rightand left-sided AV filling and LV intracardiac activation. Conversely, inadequate V-V programming may impair the ventricular filling or in the event of interatrial conduction have detrimental effects (Fig. 1B). The device senses ventricular activity via the RV or LV lead, or a combination of both (RV tip to LV tip). Double sensing of ventricular activity can be avoided by using the interventricular refractory period (IVRP). To provide biventricular stimulation in the presence of frequent ventricular ectopies or short intrinsic AV conduction, the device can be programmed to pace the programmed ventricle(s) immediately after a ventricular-sensed event during the AV interval (or at any time in single chamber ventricular pacing modes), up to a programmable maximum response rate (ventricular sense response [VSR]). The device provides the following diagnostic trend data. The patient activity trend provides the amount of time the patient’s activity, as measured by the activity sensor (accelerometer) of the device, exceeds a programmable threshold. The patient activity is thought to provide information comparable to NYHA classification. The heart rate variability (HRV) trend is the daily standard deviation of all 5-minute averages of normal-to-normal RR intervals (SDANN). 340

Figure 1. (A) Biventricular pacing aims at synchronously activating the LV from the RV and LV pacing site; indicated by the horizontal dashed line. In the presence of DC on the right or left side, delayed activation of the opposite pacing site by use of a V-V delay can reestablished synchronous activation of the LV. (B) In the presence of IAD, RV first pacing and V-V delay of the left LV pacing side can improve left atrioventricular timing. LV first pacing with a V-V delay of the RV pacing side (same A-V delay) shortens the left atrioventricular filling time and may negatively effect LV filling. AVD = atrioventricular delay; DC = delayed conduction; IAD = interatrial conduction delay; IVD = programmable interventricular delay (V-V timing); LA = left atrial; LV = left ventricle; RA = right atrial; RV = right ventricular; empty circle = atrial sensing; filled circle = ventricular pacing.

Study Protocol The study was an open label multicenter study conducted in Europe and Canada. The baseline evaluation was performed before implantation and prehospital discharge visit was scheduled within 1 week after implantation. Follow-up was required at 1, 3, and 6 months (30 ± 21, 90 ± 29, and 182 ± 42 days postimplantation, respectively). At baseline, at prehospital discharge visit, and at the 1-, 3-, and 6-month follow-up visits patients were evaluated by NYHA class, and 6MHW distance. Moreover, pacemaker performance and CRT

March 2004

PACE, Vol. 27

SEQUENTIAL BIVENTRICULAR PACING: SAFETY AND EFFICACY

efficacy were assessed by pacemaker interrogation and in a subset of patients by Holter recordings. Patient medication could be adjusted at the physician’s discretion at any time during the study. Echocardiographic protocol In a subset of patients at preselected centers, V-V timing was optimized at prehospital discharge and at the 3-month follow-up visit by maximizing the stroke volume. Stroke volume was determined as the product of the aortic velocity-time integral (VTI), using the echo Doppler technique, and the aortic cross-sectional area, calculated from the aortic diameter. The AV delay was adjusted prior to V-V timing adjustment optimizing the transmitral filling patterns during an echo Doppler examination using Ritter’s method16 at simultaneous biventricular pacing. The AV delay was kept constant for the LV pacing site (AVL ), while the AV delay for the RV pacing site (AVR ) was modified by a programmable V-V delay between −80 and +80 ms (−80 ms indicating LV first activation and +80 ms RV first activation). For patients not in the subset, V-V delay was set to 4 ms (simultaneous; nominal), or per physician discretion. Diagnostic Trend Data Diagnostic daily trend data were retrieved from the device and collected electronically. Seven-day means for HRV and physical activity data were calculated for the first week after device implantation (days 1–7) and at 3 months (days 88– 94). Statistical Analysis The primary objective of the study was to evaluate safety, with a goal of demonstrating that the 95% lower confidence bound for freedom from device related complication was ≥90%. Sample size and power calculations were based on this goal. Secondary objectives of the trial included evaluation of all adverse events, 24-hour Holter findings, LV lead electrical data, V-V timing, and cardiac resynchronization efficacy. Results are presented as mean ± SD. The difference between baseline and 3-month follow-up for NYHA classification and for 6MHW was analyzed via a Wilcoxon signed-rank test for paired data and a paired t-test, respectively. For V-V timing measurements, a Wilcoxon signed rank test was used to compare the effect on the VTI and stroke volume of nominal and optimized values. Results The study included 198 patients in 29 centers. The InSync III device was successfully implanted in 189 (95.9%) patients; the device could not be PACE, Vol. 27

Table I. Baseline Characteristics of Patients who received the Device Total Population (n = 189) Gender (male) 72.5% Age (years) 66.3 ± 10.6 Ischemic heart disease 41.8% LVEF (%) 24.4 ± 6.9 NYHA Classification II 17.5% III 68.3% IV 14.3% LVEDD (mm) 68.8 ± 9.1 QRS duration (ms) 176.3 ± 27.0 Cardiovascular medications ACE inhibitors 81.3% Antiarrhythmic 25.4% Anticoagulants drug 53.4% β-blockers 50.3% Calcium channel 6.7% blockers Diuretics 94.3% Digoxin 51.3% Nitrates 32.6%

V-V Optimization Population (n = 34) 82.9% 69.6 ± 8.5 54.7% 23.7 ± 6.8 5.7% 77.2% 17.1% 69.0 ± 7.2 182.4 ± 27

91.4% 31.4% 42.9% 60.0% 11.4% 97.1% 60.0% 40.0%

successfully implanted in eight patients and one patient was excluded because he did not meet all inclusion criteria. Results concern the 189-patient cohort, unless otherwise mentioned. Patient Population Baseline characteristics of the study population are summarized in Table I. Most patients were NYHA Class III/IV (83%) and just under half had ischemic heart disease (42%). LV systolic function was impaired. The mean LVEF was 0.24 ± 0.07. The mean LVEDD was 69 ± 9 mm, while QRS durations were long (176 ± 27 ms). Most patients were on angiotensin-converting enzyme (ACE) inhibitors (81%) and/or diuretics (94%) at baseline. Drug use remained stable throughout the trial for all medication types. Thirteen successfully implanted patients exited the study early. In addition, 8 patients died, 3 patients had the InSync III system upgraded to a biventricular ICD because of documented ventricular arrhythmias, 1 patient was lost to follow-up, and 1 patient withdrew from the study.

March 2004

341

MORTENSEN, ET AL.

System Implantation Standard commercially available leads were used in the right atrium and RV. Medtronic marketreleased LV leads of the Attain family were used in most patients (n = 185; 97.9%). LV lead placement varied with the three most favored locations the anterior cardiac vein (n = 45; 23.8%), the posterior cardiac vein (n = 40; 21.2%), and the posteriorlateral vein (n = 36; 19.0%). Of the 189 patients successfully implanted, 186 were implanted during the initial procedure and 3 were initially unsuccessful but were successfully implanted within 2–7 days of the initial procedure. For the 186 initially successful implants, the total implant time was 126.0 ± 63.4 minutes (range 35–405 minutes); the time spent to place the LV leads was 29.2 ± 30.1 minutes (range 0.5–150 minutes). CRT Clinical Efficacy The effect of cardiac resynchronization on NYHA classification and 6MHW distance was evaluated in all patients with complete baseline and 3-month follow-up data. The results for baseline and 3-month are summarized in Table II. The improvements in NYHA and 6MHW distance seen at 3 months remained stable at 6 months (baseline vs 6 months: 3.0 ± 0.6 vs 1.89 ± 0.75, P < 0.001; 6MHW 376 ± 98 vs 459 ± 115 meter, P < 0.001). To understand if V-V optimization of stroke volume translated into clinical effects, the study compared NYHA and 6MHW in patients who were V-V optimized and those that were not. The clinical improvements from baseline to 3 months did not differ between the two groups (Table II). During a mean follow-up time of 112 days (range 4–246 days, median 102) eight patients died. In seven patients cause of death was cardiac and one patient died due to pulmonary infection. Cardiac deaths were due to heart failure in

three patients, caused by ventricular arrhythmia in two patients, and unknown for two patients. Survival was 97 ± 2% at 3 months and 94 ± 2% at 6 months. V-V Timing Optimization Paired data for V-V timing optimization at prehospital discharge and 3-month follow-up were available in 34 patients. Baseline data for this group are summarized in Table I. The distribution of optimal V-V delay programming at prehospital discharge and 3-month follow-up in these 34 patients is given in Figure 2. The optimization of V-V timing increased the aortic VTI and SV by 23% at prehospital discharge and 18% at 3-month follow-up compared to simultaneous pacing (Table III). These increases were achieved at moderate V-V delays of 27.4 ± 21 ms at prehospital discharge and 22.7 ± 14 ms at 3-month follow-up; the LV was preexcitated in 38% (13/34) of patients at prehospital discharge and 53% (18/34) at 3-month follow-up. However, programming of nonoptimal V-V delays could reduce the aortic VTI and SV up to 25% at prehospital discharge and up to 28% at 3month follow-up compared to simultaneous pacing. These reductions were associated with longer V-V delays of 42.0 ± 27 ms at prehospital discharge and 49.3 ± 23 ms at 3-month follow-up; the LV was preexcitated in most 62% (21/34) patients at prehospital discharge and 50% (17/34) at 3-month follow-up. Stroke volume and VTI were found to be optimal within a small range of V-V delays. In 30 (88%) of 34 patients the optimal V-V delays were within a range of ±20 ms and in 32 (94%) of 34 patients within a range of ±40 ms at prehospital discharge. The optimal V-V delay remained unchanged in 15 of 34 patients between prehospital discharge and 3-month follow-up and with only a

Table II. Cardiac Resynchronization Therapy Efficacy, Baseline/3-Month Visit Variable NYHA all (n = 86) NYHA V-V opt (n = 46) NYHA non V-V opt (n = 40) 6-minute hall walk (m) all (n = 55) 6 MHW V-V opt (n = 32) 6 MHW non V-V opt (n = 23)

Baseline

3-Month

P-Value*

2.97 ± 0.57 3.11 ± 0.48 2.83 ± 0.64 336.5 ± 91 324.6 ± 94 352.9 ± 86

1.88 ± 0.71 1.93 ± 0.74 1.83 ± 0.68 419.1 ± 125 422.2 ± 132 414.9 ± 118