Author + information
- Received May 10, 2001
- Revision received November 21, 2001
- Accepted December 11, 2001
- Published online March 6, 2002.
- ↵*Reprint requests and correspondence:
Dr. Volker Kühlkamp, Medizinische Klinik III, Otfried Müller Str. 10, D-72076 Tübingen, Germany.
Objectives The purpose of this study was to evaluate an implantable cardioverter-defibrillator (ICD) incorporating biventricular pacing.
Background Biventricular pacing improves the symptoms of heart failure, a frequent problem in ICD recipients.
Methods This prospective multicenter study evaluated the safety and efficacy of an ICD with biventricular pacing.
Results A total of 84 patients with a standard ICD indication, symptomatic heart failure, left ventricular (LV) ejection fraction <35% and a QRS duration >130 ms were included in the trial. In 81 of 84 patients the LV lead was successfully implanted. Patients significantly improved in the 6-min hall-walk test (baseline 304 ± 131 m, three months 397 ± 142 m, p < 0.001), quality of life (baseline 38.9 ± 21.2, three months 26.5 ± 21.2, p < 0.001) and the New York Heart Association (NYHA) classification (baseline 2.8 ± 0.6, three months 2.2 ± 0.5). Left ventricular end-diastolic (from 79.6 ± 13.0 mm to 73.6 ± 12.9 mm, p = 0.002) and end-systolic (from 68.3 ± 13.5 mm to 63.9 ± 12.9 mm, p < 0.001) diameter decreased, and fractional shortening increased (from 16 ± 6% to 18 ± 6%, p = 0.018). Of the patients 26 experienced 472 episodes of spontaneous sustained ventricular tachyarrhythmias. All episodes were successfully terminated except for 16 episodes occurring in a patient with incessant ventricular tachycardia. Biventricular antitachycardia pacing was more effective than right ventricular antitachycardia pacing (p < 0.001). During follow-up (median 185 days, range 12 to 344 days) five patients died from progressive heart failure.
Conclusions Incorporating biventricular pacing in an ICD is feasible and leads to an improvement of heart failure symptoms. Therefore, this therapy may become an option for patients who need ICD therapy in the presence of severe heart failure.
Treatment with an implantable cardioverter-defibrillator (ICD) is an accepted mode of treatment in patients with life-threatening ventricular tachyarrhythmias (1). Subgroup analysis in the Antiarrhythmics Versus Implantable Defibrillators (AVID) (2)study has shown that patients with severely depressed left ventricular (LV) function have the largest benefit from ICD treatment. Although ICD treatment in patients at risk prolongs life, it does not improve quality of life or the symptoms of heart failure. In recent years it has been shown in both acute and long-term studies that biventricular pacing improves hemodynamic parameters and symptoms in heart failure patients (3–7). However, the separate implantation of an ICD and a dual-chamber pacemaker capable of biventricular pacing carries additional risks owing to the extracardiac leads and possible interactions between devices (8), similar to what has been described for VVI and DDD pacemakers (9,10). The present study reports the initial experience with an ICD incorporating ventricular resynchronization therapy.
This was a prospective study in patients indicated for ICD therapy because of symptomatic sustained ventricular tachycardia (VT) and/or survival of cardiac arrest. In addition, patients were required to have symptomatic heart failure despite appropriate heart failure therapy, with a LV ejection fraction (LV angiogram or echocardiography) <35% and an LV end-diastolic diameter above 55 mm (echocardiography). The QRS duration had to be >130 ms. All patients participated voluntarily and signed a written, informed consent approved by the study center’s Institutional Review Board.
The device implanted in this study was the InSync 7272 ICD (Medtronic Inc., Minneapolis, Minnesota), which is capable of delivering ICD therapies and cardiac resynchronization therapy. It is based on the Medtronic GEM models 7271 and 7273 ICDs and the InSync model 8040 device. The characteristics of the GEM models 7271 and 7273 have been described elsewhere (11,12). The device has a special five-port header for connection of four leads, which have three pacing electrodes (one atrial lead, one right ventricular [RV] lead and the LV lead) and two defibrillation coils (Fig. 1). Cardiac resynchronization and antitachycardia pacing (ATP) can be programmed to right, left or biventricular stimulation. It is important to note that both ventricular sensing and detection remain on the right side; hence, both sensing and detection in the InSync ICD are unchanged from ICD-only systems.
Prior to implantation of the device, patient assessment included: New York Heart Association (NYHA) classification and echocardiography for determination of the LV systolic and diastolic diameter, LV ejection fraction and presence and severity of mitral regurgitation. Patients performed a 6-min hall-walk test and completed the Minnesota Living With Heart Failure Questionnaire (13,14).
Optimizing of the atrioventricular (AV) delay.
The AV optimization was obtained using Doppler echocardiography. First, the device was programmed to 40 beats/min to assure an intrinsic sinus rate. The sensed AV delay was programmed to 200 ms. In this setting, mitral valve closure is delayed to the end of the A-wave. The sensed AV interval was decreased in steps of 20 ms until the mitral valve Doppler signal caused truncation of the A-wave. Finally, the sensed AV interval was increased in steps of 10 ms to ensure that the mitral valve closure Doppler signal coincided with or occurred shortly after the end of the A-wave.
Patients were seen when discharged from the hospital and on an outpatient basis at one and three months of follow-up. At the follow-up visits, the 6-min hall-walk test was repeated as was the Minnesota Living With Heart Failure Questionnaire; an echocardiogram was obtained and the severity of heart failure (NYHA classification) was assessed. If necessary, the AV interval was reprogrammed. Arrhythmic episodes were retrieved from the device’s memory and classified by the local investigator. All episodes were reanalyzed independently by investigators V. K. and B. K. (see Appendix), who were both unaware of the patient’s symptoms at the time of the arrhythmic episode and the classification of the episode by the local investigator. The criteria used have been published elsewhere (15). Effective termination of an episode was defined as the detection of eight consecutive paced or sensed cardiac cycles greater than or equal to the ventricular tachycardia detection interval that were not classified as VT or ventricular fibrillation (VF).
The means and standard deviations for hall-walk distance and quality of life were calculated for the baseline, one-month and three-month time points. To test for changes in these parameters with cardiac resynchronization therapy, paired ttests were used to compare the results at one month and three months with baseline values. Two-sided tests with a significance level of 0.05 were used.
Both the number and the percent of patients with each NYHA classification (I to IV) were determined for the baseline, one-month and three-month time points. To test for changes in NYHA functional class with cardiac resynchronization therapy, paired comparisons using the McNemar test for significant change were used to compare the one-month and three-month results with baseline values. Two-sided tests with a significance level of 0.05 were used.
The percent of episodes successfully terminated was estimated using a generalized estimating equations model accounting for correlation within clusters of binary responses. A two-sided 95% confidence interval was also determined. The Kaplan-Meier method was used to estimate overall survival from all causes of death.
Eighty-four patients were enrolled in this study. The InSync 7272 ICD was successfully implanted in 81 patients (96.4%). Cardiac resynchronization therapy was not possible in three patients. These three patients are not included in all data analysis except for the analysis of the ICD and/or lead-associated complications. Patients included in the study are summarized in Table 1. A total of 85% of patients were on angiotensin-converting enzyme inhibitors, and 54% of patients received a beta-blocker at baseline. Left ventricular ejection fraction was 25 ± 7%, mean QRS duration was 170 ± 30 ms and the mean LV end-diastolic diameter was 79.6 ± 13.0 mm.
Paroxysmal atrial fibrillation was documented in 19 patients (23%); 10 patients (12%) had a history of atrial flutter. Symptomatic sinus bradycardia was present in 10 patients (12%), 27 patients (33%) had an AV block of I degree and 7 patients (9%) had an AV block of II or III degree.
Implantation of the ICD and leads.
Standard commercially available leads were used in the right atrium and the RV. Ten patients received the InSync 7272 ICD as a replacement device of an earlier implanted standard ICD. In one patient the implantation was a replacement of a previous cardiac resynchronization system that did not include ICD therapy.
The primary LV leads used in this study were the Medtronic Attain Model 2187 LV lead and the Medtronic Attain CS Model 2188. In five patients, other commercially available leads were used; one patient received an investigational lead and one system was a replacement of a previous cardiac resynchronization system with ICD therapy. A guiding sheath (Medtronic Attain Model 6216) was used to cannulate the coronary sinus, and a balloon catheter (Medtronic Attain Model 6215) was used to obtain an angiogram of the coronary sinus and the ventricular venous structure. Placement of the RV and LV leads was performed according to recently published suggestions with biventricular pacemakers (16). The location of the LV lead was determined from the intraoperative venogram. In the majority of cases the LV lead was forwarded to a posterior-lateral vein (n = 37, 46%) or a lateral vein (n = 26, 32%). After implantation of the leads, mean QRS duration decreased from 170 ± 30 ms to 158 ± 34 ms (p < 0.001).
Total procedure time was 192 ± 84.1 min (range 69 to 480 min), mean time for the placement of the LV lead was 77 ± 64.5 min (range 2 to 402 min). Mean fluoroscopy time was 40.0 ± 30.7 min, with a range from 1 to 149 min. Of 84 patients, 77 were successfully implanted at first attempted (92%). An additional four patients were successfully implanted following a second attempt.
The three failures to implant an LV lead were due to inability to cannulate the coronary sinus in one patient, LV lead dislodgement while retracting the guiding catheter in one patient and in another patient the procedure was stopped after perforation of the coronary sinus (Table 2).
In four patients a second LV lead implantation attempt was necessary owing to one perforation at the os of the coronary sinus with the LV lead, inability to cannulate the coronary sinus in two cases and no specific explanation in one case. In these four cases the LV lead was successfully implanted in the second attempt. Dissection of the coronary sinus was seen in two patients not requiring further therapy.
A pneumothorax occurred in three cases, requiring drainage of the pleural cavity in two cases; one patient also had a hematothorax, requiring drainage of the pleural cavity.
Effect of biventricular pacing on the symptoms of heart failure.
As outlined in the previous text, patients performed a 6-min hall-walk test and completed a quality of life questionnaire at baseline, one month, three months and six months of follow-up. Severity of heart failure was assessed by the treating physician at each visit using the NYHA classification.
A significant (p < 0.001) improvement in the 6-min hall-walk test was seen at both follow-up visits as compared to baseline (Table 3). Walking distance increased from 304 ± 131 m to 390 ± 137 m (n = 72) at one month and from 319 ± 156 m to 393 ± 143 m (n = 56) at three months of follow-up. Nine patients at baseline, three patients at one month and no patient at three months of follow-up were unable to perform the test due to heart failure. Severity of heart failure was an important predictor of the improvement in the walking distance. The NYHA functional class III/IV heart failure patients had a significant (p < 0.001) improvement in the hall-walk test, whereas NYHA functional class II patients did demonstrate an improvement in hall-walk distance that did not reach statistical significance (Table 3).
Table 3shows the results of the Minnesota Living With Heart Failure Quality of Life Assessment. The paired comparison among baseline, one month and three months of follow-up is demonstrated for all patients, NYHA functional class II patients and NYHA functional class III/IV patients. A significant (p < 0.001) improvement in quality of life was shown for NYHA functional class III/IV patients. Patients with NYHA functional class II heart failure had an almost unchanged quality-of-life score.
Figure 2shows the effect of biventricular pacing on the severity of heart failure symptoms. Mean NYHA classification improved from 2.8 ± 0.6 to 2.3 ± 0.5 at one month and 2.2 ± 0.5 at three months. Dividing the patient population by the severity of heart failure at baseline, it became evident that 67% of the NYHA functional class III/IV patients improved by at least one NYHA class, whereas 85% of the NYHA functional class II patients remained unchanged.
Effects of biventricular pacing of LV function.
With biventricular pacing, LV end-diastolic (p = 0.002) and end-systolic (p < 0.001) diameter decreased significantly (Table 4). Fractional shortening was significantly improved with biventricular pacing (p = 0.018). Again, the effects were more pronounced and reached statistical significance only for patients with NYHA functional class III and class IV heart failure at baseline. However, even in NYHA functional class II patients, a consistent trend toward an improvement with biventricular pacing was seen.
Efficacy of therapy for spontaneous ventricular tachyarrhythmias.
A total of 472 episodes (26 patients) were appropriately detected as VT or VF, 339 episodes in 17 patients were detected in the VT zone, 107 episodes in 8 patients in the fast ventricular tachycardia (FVT) zone and 26 episodes in 8 patients in the VF zone. All ventricular tachyarrhythmias were correctly identified; hence, sensitivity for ventricular tachyarrhythmias was 100%. Double counting of sensed events did not occur; thus, inappropriate therapies due to double counting were not applied.
One patient experienced 27 VT episodes, 16 were not successfully terminated as the patient was experiencing incessant VT. All other episodes were successfully terminated by the device.
Twenty-three patients had a total of 431 episodes in which ATP therapies were applied (Table 5). Sixteen patients were programmed to RV ATP, and nine patients were programmed to biventricular ATP. Two patients initially programmed to RV ATP were reprogrammed to biventricular ATP, which was significantly more effective as compared to RV ATP for all episodes in the VT zone (p < 0.0001).
In addition, there was a trend for a higher efficacy of biventricular ATP in the FVT zone and fewer episodes that were accelerated by ATP if biventricular ATP was used. Mean cycle length of the ventricular tachyarrhythmias treated by biventricular pacing was longer (389 ± 55 ms) as compared to ventricular tachyarrhythmias treated with RV ATP (367 ± 44 ms, difference NS). However, even after correcting for cycle length of the VT and the type of ATP (ramp/burst), biventricular ATP was significantly (p < 0.01) more effective than RV ATP.
Complications associated with the ICD and the lead system during follow-up.
A total of 10 patients out of 84 experienced 11 complications associated with the ICD and/or the lead system during follow-up (Table 2): The most frequent complication occurring in six patients was dislodgement of the LV lead. This was associated with loss of capture in two cases and without loss of capture in four cases. The lead was repositioned in three patients and removed in three patients as well.
Infection of the ICD and/or lead system was observed in two patients (2.5%), leading to explanation of the system. In one patient infection occurred after an unsuccessful attempt to reposition the LV lead. The second patient with an infection of the ICD and/or lead system had an ICD with dual-chamber pacing implanted, because implantation of the LV lead was unsuccessful.
Survival during follow-up.
During a mean follow-up of 178 days (range 12 days to 344 days, median 185 days) five patients died. In all cases, death was classified as nonsudden cardiac death not associated with the device or the implantation procedure. Survival at three and six months was 96 ± 2%. One patient received a heart transplant after an uneventful follow-up of 67 days.
This study presents the initial experience with an ICD capable of ventricular resynchronization therapy in patients with severe heart failure.
Effect of biventricular pacing on the symptoms of heart failure.
Cardiac resynchronization therapy resulted in a significant improvement in quality of life, symptoms of heart failure and LV dimensions (Tables 3 and 4, Fig. 2). Patients with NYHA functional class III/IV heart failure had the largest benefit from biventricular pacing. Therefore, this study strongly supports current recommendations to implant a system with biventricular pacing in patients with severely depressed LV function and NYHA functional class III or IV heart failure (3). However, even patients in NYHA functional class II heart failure showed an improvement in the 6-min hall-walk test; studies with a longer follow-up are still necessary to define the potential benefit of biventricular pacing in these patients.
Detection of ventricular tachyarrhythmias.
The electrograms recorded via the LV lead are not used for detection, thus effectively preventing the possibility of double counting. Hence, detection in the InSync 7272 ICD is unchanged to detection in standard ICDs. In the Contak ICD, double counting occurs when the two electrograms obtained from the right and left ventricle are fed into a single amplifier. This leads to inappropriate therapy in up to 14% of patients, usually in the situation of sinus tachycardia with spontaneous AV conduction (17). However, double counting in the Contak ICD occurs as well during VT (18).
The site of stimulation may modify the ability of a stimulated impulse to enter the re-entry circuit and terminate VT (19). With biventricular pacing, two distinct pacing sites are used; this might enhance the ability to effectively terminated VT by ATP. This study showed that biventricular ATP was more effective than RV ATP (Table 5), and a trend was seen that biventricular pacing did result in fewer accelerations than RV ATP.
In a recent study with the Contak ICD, biventricular pacing was not significantly more effective than RV ATP. However, the control group in this study consisted of a patient population that received an ICD with VVI or DDD pacing. No data are presented in regard to the patient population included in the control group (20). Thus, a prospective study, including many more patients and episodes, is necessary to answer the question of whether biventricular ATP is more effective than RV ATP.
A further important question is whether biventricular pacing affects the incidence of sustained ventricular tachyarrhythmias. Data from the beta-blocker studies support the concept that improving heart failure reduces sudden death, suggesting that the incidence of sustained ventricular tachyarrhythmias is decreased (21). Two smaller studies using biventricular pacing have shown positive effects on the incidence of ventricular arrhythmias (22,23). In the present study, 26 patients (33%) experienced at least one episode of spontaneous sustained VT or VF, a finding similar to the recently published data with an ICD with dual-chamber pacing, where 96 (32%) of 300 patients experienced at least one episode during a mean follow-up of 1.7 months (24).
Implantation of the system.
The implantation of the LV lead can be time-consuming. It has been shown with the same LV lead that failure rate drops with the number of leads that a center has implanted. However, tools to implant LV leads need improvement before cardiac resynchronization therapy will be widely applied (25). Initial experience with new “over the wire” leads is promising (26,27).
The long procedural duration may have contributed to the two infections. Predictors of operative complications are the severity of heart failure, the surgical environment, the duration of the intervention and diabetes. Two risk factors were obviously present in our patients: long procedural duration and severe heart failure (28).
Hematoma was another complication seen in three patients (3.7%); all three patients were on anticoagulant therapy at the time of bleeding. In the InSync 7272 ICD study population, 56% of patients were on anticoagulant therapy at baseline as compared to 21% of patients in the AVID study, where a 2.6% risk of bleeding was reported (1).
The major weakness of the study was the lack of a control group of patients with a comparable stage of heart failure receiving ICD therapy. From this study it is unknown to what extent an ICD with biventricular pacing improves symptoms and/or the prognosis in heart failure patients in comparison to a control group. However, the aim of the present study was to gain initial experience with an ICD system capable of ventricular resynchronization therapy. This study offers important information concerning the initial experience with the system and adds important data to the knowledge on ventricular resynchronization therapy.
Ventricular resynchronization therapy and ICD therapy can be safely achieved using the InSync 7272 ICD system. Biventricular ATP was more effective than RV ATP. The NYHA functional class III/IV heart failure patients showed a significant improvement in functional status, LV dimensions and quality of life. The results obtained in NYHA functional class II patients are encouraging. Therefore, an ICD incorporating ventricular resynchronization therapy may be an important option for patients with severe heart failure who need ICD treatment.
The InSync 7272 ICD world wide investigators (and participating centers)
V. Kühlkamp (Tübingen); J. Vogt (Bad Oeynhausen); B. Lemke (Bochum); K. Seidl (Ludwigshafen); L. Jordaens (Rotterdam); S. Cadet (Lille); J. Salerno (Castelllanza); F. Gaita (Asti); G. Kaye (Cottingham); J. Clementy (Pesac); P. Mabo (Rennes); E. Hoffmann (München); T. Lavergene (Paris); W. Jung (Bonn); A. Raviele (Mestre); V. Paul (Middlesex); H. Klein (Graz); G. Baumann (Berlin); H. Klein (Magdeburg); D. Böcker (Münster); W. Kranig (Bad Rothenfelde); G. Moya (Barcelona); S. Osswald (Basel); C. Bucknall (London); B. Kramm (Maastricht, Medtronic Inc.).
- antitachycardia pacing
- Antiarrhythmics Versus Implantable Defibrillations study
- fast ventricular tachycardia
- implantable cardioverter-defibrillator
- left ventricular
- New York Heart Association
- right ventricular
- ventricular fibrillation
- ventricular tachycardia
- Received May 10, 2001.
- Revision received November 21, 2001.
- Accepted December 11, 2001.
- American College of Cardiology Foundation
- The Antiarrhythmics Versus Implantable Defibrillators (AVID) Investigators,
- Domanski M.J,
- Saksena S,
- Epstein A.E,
- et al.
- Cazeau S,
- Ritter P,
- Lazarus A,
- et al.
- Zardini M,
- Tritto M,
- Bargiggia G,
- et al.
- Calkins H,
- Brinker J,
- Veltri E.P,
- Guarnieri T,
- Levine J.H
- Olson W.H
- Wilkoff B.L,
- Kühlkamp V,
- Volosin K,
- et al.
- Pimobendan Multicenter Research Group,
- Rector T.S,
- Cohn J.N
- Blanc J.J,
- Etienne Y,
- Gilard M,
- et al.
- Josephson M.E
- MERIT-HF Study Group
- Ventak CHF investigators,
- Higgins S.L,
- Yong P,
- Scheck D,
- et al.
- Kühlkamp V,
- Volosin K,
- Huegl B.J,
- Stafford W,
- Larsen J.S,
- Cuijpers A
- EASYTRACK Clinical Investigation Study Group and the European EASYTRACK Registry,
- Pürerfellner H,
- Nesser H.J,
- Winter S,
- Schwierz T,
- Hörnell H,
- Maertens S