Results of the International Study of the Implantable Pacemaker Cardioverter-Defibrillator
A Comparison of Epicardial and Endocardial Lead Systems
Background The purpose of the present report was to document clinical experience derived from the implantation of 2834 epicardial and endocardial cardioverter-defibrillators (ICDs) in 2807 patients who were followed for almost 1 year and to compare the results obtained with the two systems.
Methods and Results Patients in the two groups had similar clinical characteristics. More than half of the patients had a total of almost 50 000 spontaneous ventricular tachyarrhythmias that were terminated with equal success (≈98%) by epicardial and endocardial ICDs. Lead dislodgement and pocket infection occurred more often with the endocardial than with the epicardial ICD, whereas perioperative mortality was higher with the epicardial ICD than with the endocardial ICD. Mortality from sudden cardiac death was 1.4% in the epicardial ICD group and 0.6% in the endocardial ICD group at 1 year (P=.069). Overall mortality at 1 year was 12.2% and 6.9% for the epicardial and endocardial groups, respectively (P<.001), reflecting the higher surgical mortality for the epicardial system.
Conclusions The endocardial ICD is as effective as the epicardial ICD but incurs lower perioperative mortality.
The efficacy and safety of the implantable cardioverter-defibrillator (ICD) for patients with recurrent ventricular tachycardia (VT) and ventricular fibrillation (VF) have been well established.1 2 3 4 5 6 Whether the ICD reduces mortality compared with drug treatment has been the subject of several prospective, randomized trials.7 8 9 The first generation of these devices provided cardioversion only, which was not optimal for conscious patients during well-tolerated VT. Tiered therapy with antitachycardia pacing reduced the number of high-energy shocks needed for the treatment of VT in this patient group and enhanced patient acceptance of ICD therapy.3 4 5 6 10 The need to incorporate programmability, multiple detection, and therapeutic options has led to the development of third-generation ICDs. Furthermore, this technology has been combined with a nonthoracotomy lead (NTL) system that decreases potential perioperative mortality associated with current epicardial system implantation approaches.11 12
The purpose of the present study was to discuss results of the use of the Pacer-Cardioverter-Defibrillator (PCD) (Medtronic, Inc) device in a worldwide multicenter study of both the epicardial lead and NTL (Transvene) systems and to compare results obtained with the two systems. These systems are identical except for the implanted defibrillation leads and include tiered therapy with bradycardia and antitachycardia pacing capabilities.
Data were obtained from the Medtronic data base, which was used to collect and store information submitted by the individual investigative centers. Observations, complications, and deaths were reviewed by an independent group of four physicians.
Patients were enrolled in the clinical study if they had survived at least one episode of cardiac arrest due to a ventricular tachyarrhythmia not caused by acute myocardial infarction or reversible causes or they had recurrent, sustained VT that remained inducible via programmed electrical stimulation or exercise or occurred spontaneously despite the most efficacious antiarrhythmic drug therapy tolerated by the patient on a long-term basis. The first epicardial implantation was made in May 1989, and the first NTL system was implanted 7 months later. The choice of which system to implant was made by the investigator. In general, permission to implant the NTL system followed a period of use of the epicardial system; once permission was received, the NTL system was the device chosen to avoid a thoracotomy, unless concomitant cardiac surgery was planned. The success rate for implantation of the NTL system was almost 85%. There was no consistent attempt to preselect patients for one type of implantation over the other, and consecutive patients were recruited for implantation first with the epicardial approach and then with the endocardial method.
This analysis includes 2834 clinical implantations in 2807 patients between May 1989 and July 15, 1993. These implantations occurred in 119 investigative centers in Europe, England, and the United States. Of the implants, 1475 patients received a total of 1478 epicardial implants, and 1349 patients received a total of 1356 NTL implants.
Data from a subgroup of 1542 patients who were representative of the overall group were examined in-depth to analyze perioperative mortality. Of this group, 757 patients received an NTL implant and 742 patients received an epicardial system. Forty-three patients received other leads after initially being evaluated for an NTL implant. The 757 patients failed a mean of 2.9 drug trials (range, 1 to 8), and the 742 patients failed a mean of 3.0 drug trials (range, 1 to 9).
Pulse Generator and Lead Systems
The implantable device system included an ICD (models 7216A and 7217B, Medtronic) and an epicardial or endocardial lead system. The 7216A device senses the electrogram registered between the pace/sense lead and the defibrillation common cathode, whereas the 7217B device uses the pace/sense lead(s) for sensing, thus providing standard bipolar sensing. The epicardial system uses two or three epicardial patches, whereas the NTL implant uses a tripolar, right ventricular (RV) lead and either a coronary sinus (CS)–superior vena cava (SVC) lead, a subcutaneous (SQ) lead, or all three of these leads (eg, RV/SVC/SQ or RV/CS/SVC). In these systems, the retractable, helical tip and ring of the RV lead are used for sensing, thus providing standard, bipolar sensing. As many as four separate therapies can be delivered to treat any one arrhythmia episode detected by the device.
Electrophysiological Evaluation and Device Implantation
Preimplantation testing included an electrophysiological evaluation, measurement of ejection fraction, and assessment of the patient’s New York Heart Association (NYHA) classification.
Surgical implantation of an epicardial lead system was accomplished by sternotomy or thoracotomy using standard techniques. The endocardial electrode system was inserted from the cephalic or subclavian vein.
Implantation evaluation was performed to verify that one of the specified lead systems properly detected and treated episodes of VF. This testing included evaluation of pacing and sensing thresholds during sinus rhythm, VF electrogram amplitudes, and defibrillation efficacy. The implantation criteria were (1) successful VF termination in at least three of four attempts with an output energy of ≤18 J, and (2) a 2:1 VF sensing safety margin, ie, adequate VF sensing obtained at half the final programmed sensitivity.
During the implantation evaluation, initial testing was performed using an external tachyarrhythmia control device with an operation similar to the implantable device except that cardioversion and defibrillation therapies were delivered manually. After acceptable values were achieved, the pulse generator to be implanted was also tested to demonstrate proper detection and termination of VF. Pacing and sensing thresholds were assessed to establish that they were acceptable. In general, testing of the efficacy of the device to terminate VT by pacing or cardioversion was performed during a follow-up session.
Regular follow-up testing included an assessment of the relative stability of the energy required to defibrillate VF, of sensing and pacing values, and of the efficacy of the VT therapies. The number of spontaneous VT or VF episodes treated with the PCD device were tabulated, as were complications and mortality.
Patients were evaluated before hospital discharge and/or at 1 and 3 months after implantation and again at 3-month intervals. In some cases, additional testing was performed at the physician’s discretion to evaluate spontaneous events, potential complications, or both.
Reported relevant medical events were divided arbitrarily into complications and observations based on the following definitions. A “complication” was considered to be a symptomatic or an asymptomatic clinical event with potential adverse effects that could not be treated or resolved by reprogramming the device and that required invasive intervention. An “observation” was considered to be a symptomatic or an asymptomatic clinical event with potential adverse effects that either did not require invasive intervention or could be corrected by reprogramming of the device.
Each patient death was categorized with the following definitions. “Noncardiac” was considered to be the primary cause of death when cardiac causes were excluded (eg, stroke, pneumonia, or cancer). “Cardiac” was considered to be the primary cause of death when death was determined to be cardiac in nature. “Nonsudden” was considered to be deaths that occurred more than 1 hour after the onset of symptoms. “Sudden” was considered to be deaths that occurred within 1 hour of the onset of symptoms or the death was unwitnessed. Sudden cardiac deaths (SCDs) also were subdivided into the categories of “witnessed” and “unwitnessed.” “Perioperative” was considered to be deaths that occurred within 30 days of implantation.
For discrete variables, Pearson’s χ2 or Fisher’s exact test was used to test for differences between groups. For continuous variables, a Student’s statistic was used to test for equality of mean values between the groups. A life-table analysis using the Wilcoxon rank sum test was used to determine differences in survival between two groups. Values of P≤.05 were considered statistically significant.
Because subgroups of patients such as those receiving an epicardial versus an endocardial system represent different distributions of study centers, patient demographics, disease severities, and cardiac histories, adjustments were made for concomitant information that could seriously bias the comparison. For both SCDs and all-cause deaths, adjustment was made with Cox regression analysis for patient age, sex, NYHA class, ejection fraction, and indication for ICD (VF and VT) and for concomitant cardiovascular conditions such as coronary artery disease, previous myocardial infarction, previous coronary artery bypass graft surgery or other cardiovascular surgery, heart failure, and arrhythmias, such as atrial tachyarrhythmias, VT, and VF, with a sufficiently large effect (>20% estimated increase or decrease).
To test whether study center had an effect, an initial analysis was made including all variables and stratified by center. A comparison of this analysis with an unstratified analysis demonstrated no qualitative or statistical difference for any of the variables once patient characteristics were adjusted for, so further regression analysis was done without controlling for study center. Statistical significance was determined with the test score. For success with spontaneous episodes, a logistic regression analysis was made with the same approach to determine which factors to use. To allow computations to converge on an answer, only centers with more than nine patients were used in the regression, which reduced the total number of episodes to only 12 171. Statistical significance was determined with the likelihood ratio test. Estimation of unadjusted cumulative mortality was performed with the Kaplan-Meier method.
Table 1⇓ is a summary and comparison of characteristics of patients with endocardial and epicardial implants. The average patient age was almost 61 years and was comparable in the two populations. Most (>80%) patients were men, although the NTL group had a slightly higher percentage of women than did the epicardial group. Coronary artery disease was the underlying heart disease in ≈75% of patients and was comparable for both populations. Both groups had moderately depressed left ventricular function, although it was slightly better in the NTL group than in the epicardial group. The two populations had comparable NYHA classifications, with >80% in class I or II. The NTL group had more patients with SCD as a primary indication for implantation than did the epicardial group, whereas the number of patients with SCD combined with recurrent, sustained VT as a primary indication for implant was higher in the epicardial population than in the NTL population. Two hundred fifty-five patients (17.3%) receiving the epicardial ICD reported a history of bradycardia compared with 200 patients (13.4%) receiving the NTL ICD.
Table 2⇓ illustrates the lowest energy of defibrillation (LED) tested (without trying to achieve lower levels). The defibrillation threshold (DFT) was defined as the lowest value that successfully terminated fibrillation (with lower values failing to terminate VF). Obtaining a DFT was not a protocol requirement. The mean LED and DFT were both higher for the NTL population than for the epicardial population.
Table 3⇓ shows the follow-up experience for the NTL and epicardial populations. The mean implant duration for the overall population was 11.7 months and ranged from time of implant to 44.6 months. The cumulative implant duration in this analysis was 33 123 months.
Arrhythmic Events After Implantation
Table 4⇓ shows the spontaneous episodes of VT and VF; 1593 patients (56.8%) experienced a spontaneous ventricular tachyarrhythmia episode that was detected and treated with the ICD. There were a total of 49 602 episodes of spontaneous VT or VF detected by the ICD. The overall device success rate in terminating detected spontaneous VT or VF episodes was 98.0%.
Spontaneous VT Episodes
Table 4⇑ provides a summary of efficacy for treating spontaneous VT episodes in epicardial and NTL patients. For the NTL population, 16 649 episodes of VT were detected in 525 patients. The majority of spontaneous VT episodes (87.7%) were successfully treated with the first VT therapy, and 97.5% were successfully treated with one of the four VT therapies. Of the 16 229 spontaneous VT episodes that were terminated successfully, 14 804 (91.2%) were treated with one of the programmed antitachycardia pacing therapies.
For the epicardial population, 25 483 episodes of VT were detected in 588 patients. The majority of spontaneous VT episodes (87.4%) were successfully treated with the first VT therapy, and 98.2% were successfully terminated with one of the four VT therapies. Of the 25 020 spontaneous VT episodes that were terminated successfully, 22 142 (88.5%) were treated with one of the programmed antitachycardia pacing therapies.
Among patients whose indication for implantation was SCD alone, 90 patients (19.2%) in the NTL population and 94 patients (21.6%) in the epicardial population also experienced an episode of VT.
Spontaneous VF Episodes
Table 4⇑ provides a summary of efficacy for treating spontaneous VF episodes in epicardial and NTL patients. For the NTL population, 3490 episodes of VF were detected in 559 patients. The majority of spontaneous VF episodes (89.5%) were successfully treated with the first VF therapy, and 98.8% were successfully treated with one of the four programmed VF therapies.
For the epicardial population, 3 980 episodes of VF were detected in 551 patients. The majority of spontaneous VF episodes (89.1%) were successfully treated with the first VF therapy, and 98.8% were successfully treated with one of the four VF therapies.
The overall success rate of the NTL system in terminating spontaneous episodes of VT or VF was 97.7%. This success rate compares favorably with the overall success rate of the epicardial system (98.3%). There have been no documented reports of patient death due to the ICD not detecting a spontaneous episode of VT or VF.
The logistic regression used to adjust the comparison between epicardial and endocardial systems regarding the efficacy of treating spontaneous arrhythmias combines VT and VF episodes to maximize the power to discriminate between epicardial and endocardial system efficacy. The regression indicates that the rate of failed episodes increases from 1.7% to 2.8% (65.1%; 95% confidence interval [CI], 16.7% to 133.3%) for endocardial systems relative to epicardial systems after adjustment for concomitant variables. This indicates, for example, that a patient whose success rate for an epicardial system is 98.3% could expect to have a 97.2% success rate (95% CI, 96.0% to 98.0%) with an endocardial system. This compares with the observed rate of 97.7% for the endocardial system.
Complications and Observations
Table 5⇓ provides a list of the complications and observations. In the NTL patients, 422 have had either a complication or an observation. For the epicardial patients, 335 have had a complication or an observation.
All patients were included in the survival analysis, based on an intention-to-treat formulation, including those in whom the ICD was programmed in the “off” mode or in whom only leads were implanted. As illustrated in Table 6⇓, 167 patient deaths were reported for the epicardial group, and 67 were reported for the NTL group. Table 7⇓ provides an analysis of patient deaths by follow-up interval.
Projected 1-year survival calculations for the various death classifications are provided in Table 8⇓. The actuarial projections do not include perioperative deaths, except in the calculation of 1-year survival of death due to all causes (shown in “Overall”).
The observed SCD mortality rate at 1 year was 1.4% among those receiving an epicardial system compared with 0.6% among those receiving an endocardial system. The adjusted analysis indicates that had the epicardial group been given an endocardial system, their 1-year SCD mortality rate would have been only 0.3%, or 22.6% of the observed rate (95% CI, 10.2% to 50.1%). All-cause mortality rate at 1 year was 12.2% among epicardial patients compared with 6.9% among endocardial patients. Had the epicardial group received endocardial leads, their 1-year all-cause mortality rate after adjustment would have been 6.2%, (a 50.7% reduction; 95% CI, 37.2% to 69.1%).
In the subgroup of 1542 patients, 897 patients were screened for an NTL system; 757 received it, 97 received an epicardial system, and 43 received other leads. The perioperative mortality rate was 0.7% for the 757 patients, 9.3% for the 97 patients, and 0.2% for the 43 patients. Of 645 patients screened for an epicardial system, 556 received it, and they had a perioperative mortality rate of 4.1%. Eighty-nine patients received an epicardial system and concomitant surgery (usually, coronary artery bypass) and had a perioperative mortality rate of 6.7%. The overall perioperative mortality rate for 742 patients (556 receiving a device alone, 89 receiving concomitant surgery, and 97 failing an initial NTL implant) receiving an epicardial system was 5.3%.
This report provides information from a large database of 2807 patients who received an NTL or a thoracotomy ICD during the 4-year period from 1989 to 1993 and were followed for a mean of almost 1 year. It confirms results from previous reports that the ICD is an extremely effective device that accurately detects and successfully terminates ventricular tachyarrhythmias. Furthermore, it provides information validating the effectiveness of the NTL system as at least comparable to the thoracotomy implant but incurring lower perioperative mortality rates: 10 deaths among 1349 NTL patients (0.74%) compared with 61 deaths among 1475 thoracotomy patients (4.13%). As seen in the subgroup analysis, the fact that some of the patients failed a transvenous implantation attempt probably contributed to some of the excess surgical mortality of the epicardial system. Similarly, concomitant coronary artery bypass graft surgery probably contributed to the perioperative mortality. However, the perioperative mortality rate for the epicardial ICD group still exceeded that of the NTL ICD group without these factors. Naturally, when surgical mortality is included, the overall actuarial survival at 1 year is higher for the NTL group than for the thoracotomy group.
The slightly higher energy required to defibrillate for the ICD group compared with the thoracotomy group did not adversely affect mortality. In fact, although the two groups were fairly comparable, the noncardiac survival was slightly higher with the NTL system, probably reflecting the facts that this was not a randomized study and the groups differed slightly. Survival for all other categories of death did not differ. Furthermore, in the epicardial group, SCD mortality represents ≈11.5% and cardiac mortality represents 42.2% of all-cause mortality. Even elimination of all cardiac causes does not account for the 50% reduction in all-cause mortality. Differences between the groups not controlled for in the adjustments, and noncardiac mechanisms prevented by using an endocardial approach may be contributing to these findings.
Another important observation is that antitachycardia pacing terminated VT in >90% of episodes. Furthermore, among patients who received an ICD for SCD alone, ≈20% also had VT. The clinical message from this observation appears to be that one should be quite certain before implanting a “shock-only” device that the patient has only VF or VT that cannot be terminated by pacing.
Not surprisingly, lead problems were a much greater issue for the NTL group than for the thoracotomy group. Displacement appeared to be solved by better fixation of the lead using two anchoring sleeves at the implantation site. The increased rate of pocket infection for the NTL group may relate to the extensive tunneling involved during the abdominal placement. This problem should be solved by pectoral implantation of the pulse generator.12
The SCD actuarial survival rates at 1 year of 99.4% for the NTL group and 98.6% for the epicardial group are outstanding for this high-risk population and compare favorably with other reports. Overall actuarial survival rates at 1 year of 93.1% for the endocardial and 87.8% for the epicardial implant recipients are similarly high, especially considering that these patients with VT or VF had a mean ejection fraction of ≈34%. Cardiac survival in the CASCADE study at 1 year was 91% for those receiving amiodarone and 77% for those receiving conventional drugs.13 In ESVEM, the overall survival at 1 year was 88% in the group with efficacy predictions and 71% for all randomized patients.14 What is not answered from the present study is how use of an ICD compares with other forms of therapy. This was not the objective of this multicenter trial and will have to await results from several studies in progress.7 8 9
Data Pool Participants: European
W. Klein, MD, Graz, Austria; H. Schmidinger, MD, Vienna, Austria; K. Steinbach, MD, Vienna, Austria; L. Jordaens, MD, Gent, Belgium; H.E. Kulbertus, MD, Liege, Belgium; E. Aliot, MD, Nancy, France; P. Coumel, MD, Parix Cedex, France; S. Kacet, MD, Lille, France; F. Zacouto, MD, Paris, France; P. Touboul, MD, Lyon, France; D. Andresen, MD, Berlin, Germany; J. Brachmann, MD, Heidelberg, Germany; G. Breithardt, MD, Münster, Germany; K.H. Kuck, MD, Hamburg, Germany; B. Lüderitz, MD, Bonn, Germany; G. Sabin, MD, Essen, Germany; H. Tillmanns, MD, Giessen, Germany; M. Wehr, MD, Essen, Germany; A. Bertulia, Lavagna, Italy; S. Favale, MD, Bari, Italy; F. Furlanello, MD, Trento, Italy; C. Pappone, MD, Napoli, Italy; A.J. Camm, MD, London, UK; M. Clarke, MD, Stoke-on-Trent, UK; W. Davies, MD, London, UK; J.M. McComb, Newcastle-upon-Tyne, UK; A.W. Nathan, MD, London, UK; E.J. Perrins, MD, Leeds, UK; R. Sutton, MD, London, UK; H.J.J. Wellens, MD, Maastricht, Netherlands; J. Amlie, MD, Oslo, Norway; O.J. Ohm, MD, Bergen, Norway; I.W.P. Obel, MD, Johannesburg, South Africa; J.M. Almendral, MD, Madrid, Spain; C. Moro Serrano, MD, Madrid, Spain; D. Blömstrom-Lundqvist, MD, Lund, Sweden; M. Rosenqvist, MD, Stockholm, Sweden; L. Kappenberger, MD, Lausanne, Switzerland; Sam Levy, MD, Marseilles, France.
Data Pool Participants: United States
M. Akhtar, MD, Milwaukee, Wis; G. Bardy, MD, Seattle, Wash; D. Benditt, MD, Minneapolis, Minn; L. Berenbom, MD, Kansas City, Mo; R. Brooks, MD, Boston, Mass; D. Echt, MD, Nashville, Tenn; R. Fletcher, MD, Washington, DC; C. Haffajee, MD, Boston, Mass; S.C. Hammill, MD, Rochester, Minn; M. Josephson, MD, Philadelphia, Pa; J.T. Lee, MD, Nashville, Tenn; W. Miles, MD, Indianapolis, Ind; D. Parker, MD, Rochester, Minn; L. Porterfield, MD, Memphis, Tenn; J. Porterfield, MD, Memphis, Tenn; J. Ruskin, MD, Boston, Mass; S. Saksena, MD, Passaic, NJ; S. Spielman, MD (Greenspan), Philadelphia, Pa; J. Sra, MD, Milwaukee, Wisc; M. Stanton, MD, Rochester, Minn; D. Steinhaus, MD, Kansas City, Mo; J. Swartz, MD, Washington, DC; P. Wells, MD, Dallas, Tex; K. Wheelan, MD, Dallas, Tex; D. Zipes, MD, Indianapolis, Ind.
M. Cain, MD, St. Louis, Mo; D. Cassidy, MD, Tampa, Fla; L. Castel, MD, Cleveland, Ohio; D. Chilson, MD, Spokane, Wash; J. Elson, MD, Pittsburgh, Pa; B. Ferguson, MD, St. Louis, Mo; J.D. Fisher, MD, Bronx, NY; R. Fogoros, MD, Pittsburgh, Pa; S. Furman, MD, Bronx, NY; B. Halperin, MD, Portland, Ore; J. Irwin, MD, Tampa, Fla; J. Kron, MD, Portland, Ore; M.H. Lehman, MD, Detroit, Mich; B. Lindsay, MD, St Louis, Mo; J. Maloney, MD, Cleveland, Ohio; J. McAnulty, MD, Portland, Ore; S. O’Donoghue, MD, Washington, DC; E. Platia, MD, Washington, DC; R. Steinman, MD, Detroit, Mich; C. Swerdlow, MD, Inglewood, Calif; P.C. Thomas, MD, Los Angeles, Calif; D. Wilber, MD, Maywood, Ill.
P. Chapman, MD, Milwaukee, Wis; S. Denker, MD, Milwaukee, Wis; G.S. Greer, MD, Little Rock, Ark; Eleanor Kennedy, MD, Little Rock, Ark; H. Kopelman, MD, Atlanta, Ga; R.A. Krieger, MD, Atlantis, Fla; R. Mahmud, MD, Greenville, NC; H. Mead, MD, Palo Alto, Calif; G. O’Neill, MD, Sacramento, Calif; T. Peter, MD, Los Angeles, Calif; D. Sellers, MD, Danville, Pa; A. Sharma, MD, Sacramento, Calif; J. Trantham, MD, Pensacola, Fla; M. Weinberger, MD, Atlantis, Fla; R. Winkle, MD, Palo Alto, Calif.
K. Beckman, MD, Oklahoma City, Okla; A.K. Bhandari, MD, Los Angeles, Calif; P. Bjerregaard, MD, St. Louis, Mo; L. Chinitz, MD, New York City, NY; A. Del Negro, MD, Fairfax, Va; A. Dougherty, MD, Houston, Tex; T. Friehling, MD, Fairfax, Va; S. Huang, MD, Worcester, Mass; W. Jackman, MD, Oklahoma City, Okla; D. Janosik, MD, St. Louis, Mo; J. Langberg, MD, Ann Arbor, Mich; B. Liem, MD, Stanford, Calif; R. McCowan, MD, Charleston, WV; R. Mittleman, MD, Worcester, Mass; F. Morady, MD, Ann Arbor, Mich; G. Naccarelli, MD, Houston, Tex; S. Pollack, MD, Orlando, Fla; M.R. Pritzker, MD, Minneapolis, Minn; K. Schwartz, MD, Orlando, Fla; R. Sung, MD, Stanford, Calif; D.F. Switzer, MD, Buffalo, NY; Paul Walter, MD, Atlanta, Ga.
F. Abi-Samra, MD, New Orleans, La; T. Ahern, MD, Las Vegas, NV; B. Alpert, MD, Pittsburgh, Pa; W.P. Batsford, MD, New Haven, Conn; J. Gallastegui, MD, Clearwater, Fla; D. Gohn, MD, New Orleans, La; C.D. Gottlieb, MD, Philadelphia, Pa; J. Griffin, MD, San Francisco, Calif; J.J. Hayes, MD, Marshfield, Wis; J. Klick, MD, Park Ridge, Ill; M.S. Kremers, MD, Charlotte, NC; M. Lee, MD, Oakland, Calif; R. Luceri, MD, Ft. Lauderdale, Fla; P. Ludmer, MD, Oakland, Calif; D. Martin, MD, Burlington, Mass; T. Mattioni, MD, Phoenix, Ariz; S. Miller, MD, Park Ridge, Ill; S. Nelson, MD, Columbus, Ohio; M.F. O’Toole, MD, Lombard, Ill; J.S. Osborne, MD, Salt Lake City, Utah; J. Seger, MD, Lubbock, Tex; R. Shkiroff, MD, Las Vegas, NV; N. Stamato, MD, Lombard, Ill; F. Venditti, MD, Burlington, Mass; G. Wells, MD, Lubbock, Tex; C. Zaher, MD, Panorama City, Calif.
Data Pool Participants: Canada
J. Boone, MD, Vancouver, BC; S. Connolly, MD, Hamilton, Ontario; P. Dorian, MD, Toronto, Ontario; E. Downar, MD, Toronto, Ontario; M. Dubuc, MD, Montreal, Quebec; M. Gardner, MD, Halifax, Nova Scotia; A. Gilles, MD, Calgary, Alberta; S. Gulamhusein, MD, Edmonton, Alberta; K. Kavanagh, MD, Edmonton, Alberta; C. Kerr, MD, Vancouver, BC; G. Klein, MD, London, Ontario; B. Mitchell, MD, Calgary, Alberta; D. Newman, MD, Toronto, Ontario; M. Sami, MD, Montreal, Quebec; T. Tang, MD, Ottawa, Ontario.
This study was supported in part by the Herman C. Krannert Fund, Indianapolis, Ind; by grants HL-42370 and HL-07182 from the National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md; and by the American Heart Association, Indiana Affiliate, Indianapolis, Ind. We thank Timothy Church and Valerie Sakun, who provided essential help in writing the manuscript.
Dr Zipes is a consultant for Medtronic, Inc.
- Received December 7, 1994.
- Accepted January 3, 1995.
- Copyright © 1995 by American Heart Association
Tchou P, Kadri N, Anderson J, Caceres JA, Jazayeri M, Akhtar M. Automatic implantable cardioverter defibrillators and survival of patients with left ventricular dysfunction and malignant ventricular arrhythmias. Ann Intern Med. 1988;109:529-534.
Bardy GH, Troutman C, Poole JE, Kudenchuk PJ, Dolack GL, Johnson G, Hofer B. Clinical experience with a tiered-therapy multiprogrammable antitachycardia device. Circulation. 1992;86:1689-1698.
Fromer M, Brachmann J, Block M, Siebels J, Hoffmann E, Almendral J, Ohm OJ, den Dulk K, Coumel P, Camm AJ. Efficacy of automatic multimodal device therapy for ventricular tachyarrhythmias as delivered by a new implantable pacing cardioverter-defibrillator: results of an European multicenter study of 102 implants. Circulation. 1992;86:363-374.
Greene HL. Antiarrhythmic drugs versus implantable defibrillators: the need for a randomized controlled study. Am Heart J. 1993;127:1171-1178.
Zipes DP. The implantable cardioverter-defibrillator: lifesaver or a device looking for a disease? Circulation. 1994;89:2934-2936. Editorial.
Bardy GH, Hofer B, Johnson G, Kudenchuk PJ, Poole JE, Dolack GL, Gleva M, Mitchell R, Kelso D. Implantable transvenous cardioverter-defibrillators. Circulation. 1993;87:1152-1168.