Page 1


Use and long-term outcomes of implantable cardioverter-defibrillators, 1990 to 2009 Pamela J. Bradshaw, PhD, a Paul Stobie, MBBS, b Tom Briffa, PhD, a and Michael S. T. Hobbs, MBBS, PhD a Perth, Western Australia

Background Automated implantable cardioverter-defibrillators (ICDs) have become standard therapy for patients at high risk for sudden cardiac death. Linked data allow examination of trends in use and long-term survival after ICD implantation in an adult population. Methods Linked state-wide person-based data on hospital admissions and deaths from 1980 to 2009 were used to identify incident cases of ICD implantation. Population rates were calculated using census data. Kaplan-Meier techniques were used to describe cumulative survival. Cox regression models were used to determine the factors associated with the outcomes. Results

Between 1988 and 2009, 1593 devices were implanted in patients in Western Australia, rising from 2 in 1988 to 245 in 2009; standardized population rates rose from 0.8 in 100 000 in 1995 to 14.9 in 100 000 in 2009. Mean age rose from 52.6 (SD 11.6) to 64.1 (11.4) years. Ventricular tachycardia (23%), cardiomyopathy (18%), and heart failure (16%) were the most frequent principal diagnoses. Ischemic heart disease was present in 49% of patients. Five-year cumulative survival was 0.74 (SE 0.01), and at 10 years, 0.53 (SE 0.03); median survival was 11.3 years. Readmission within a year, older age, heart failure, device complications, and chronic ischemic heart disease were associated with poorer survival.


Implantable cardioverter-defibrillator use in adults at risk for sudden cardiac death has grown rapidly. Readmission within 12 months of discharge is associated with worse medium and long-term mortality. Survival for most patients younger than 65 years exceeds 10 years and 5 years for those aged ≼75 years. (Am Heart J 2013;165:816-22.)

Background Use of an implantable cardiac defibrillator (ICD) has been shown in clinical trials to improve overall survival. As secondary prevention in patients who have suffered a life-threatening ventricular arrhythmia, ICD was superior to antiarrhythmic drugs in the AVID trial, 1 and a trend to benefit was observed in 2 smaller studies. 2,3 Survival benefits were also shown for ICD use in primary prevention among patients at high risk for sudden cardiac death (SCD). Patients in both MADIT studies were survivors of acute myocardial infarction (AMI) with additional risk factors for SCD, 4,5 and the MUSTT study included patients with coronary artery disease, reduced

From the aSchool of Population Health, University of Western Australia, Perth, Western Australia, and bDepartment of Cardiovascular Medicine, Sir Charles Gairdner Hospital, Perth, Western Australia. Submitted September 19, 2012; accepted February 13, 2013. Reprint requests: Pamela J. Bradshaw, PhD, School of Population Health, University of Western Australia M431 Stirling Highway, Perth WA 6009 Australia. E-mail: 0002-8703/$ - see front matter Š 2013, Mosby, Inc. All rights reserved.

ejection fraction, and nonsustained ventricular tachycardia (VT). 6 The benefit of ICD therapy persisted during extended follow-up. 7 In a meta-analysis of primary and secondary prevention trials, the pooled reduction of risk for arrhythmic death was 57%, with an absolute reduction of 8% (95% CI 6%-10%), and 30% and 10% (95% CI 4%16%), respectively, for all-cause death. 8 Among patients with symptomatic heart failure (HF), including those with cardiomyopathy, ICD use improved survival either alone 9,10 or when incorporated with cardiac resynchronization therapy. 11 Although evidence for survival benefit among patients with nonischemic dilated cardiomyopathy is less robust, 12,13 the SCD-HeFT trial did demonstrate benefit in patients with nonischemic cardiomyopathy. 9 The positive results from clinical trials have been accompanied by a rapid increase in the use of ICDs in advanced economies. 14,15 Some long-term outcomes after ICD have been established for participants in clinical trials. Mortality to a maximum 11 years for patients randomized to an ICD in the Canadian Implantable Defibrillator Study was 2.8% per year versus 5.5% for those randomized to amiodarone. 16 Among patients from MADIT-II, all-cause mortality to 5 and 8 years was 49% and 62% for patients randomized

American Heart Journal Volume 165, Number 5

to medical therapy alone and 33% and 49% for those with an ICD. 7 The few long-term survival outcomes reported for patients not in clinical trials include a cohort of 270 patients with an ICD implanted for life-threatening arrhythmias in the United States in the early 1980s. Allcause mortality at 1 and 5 years was 8% and 26%. 17 Similar mortality was reported from the ALTITUDE project from N2,000 centers across the United States, being 8% at 1 year for ICD patients (12% for cardiac resynchronization therapy-defibrillator patients) and 32% at 5 years. 18 Linked data were used to study trends from 1987 to 1995 among 22,565 US Medicare recipients of ICDs aged 65 years or more. 14 The use of ICDs rose rapidly, whereas complication rates and length of stay fell. The readmission rate within a year remained high with N50% of patients readmitted for ventricular arrhythmias or other cardiac disorders. Reductions in 30-day, 1-year, and 3-year mortality were seen, with 1-year mortality falling from 19.3% in 1987 to 11.4% in 1994. As there are few studies worldwide of long-term outcomes of ICD implantation, we used linked data from Western Australia (WA) to describe their use in a general population and the long-term outcomes to 20 years. Western Australia is a large state, being one-third of the Australian continent and almost 4 times the size of Texas, spanning N1 million square miles (2.5 million square kilometers), but with a relatively small population (2,010,113 in 2005), 19 nearly 75% living in the capital city of Perth.

Methods The Health Department of WA's Data Linkage System 20 is used to link person-based records from several databases, including the state-wide hospital morbidity database and deaths within Australia. The Data Linkage Branch uses World Health Organisation International Classification of Diseases (WHO-ICD) 9/9 CM diagnostic codes 390 to 459 and WHOICD-10 codes I00-I99 to prepare a “vascular” file including cardiac-related admissions, discharges, and deaths. From the most recent iteration of this file, we extracted all adult (≥18 years) cases with a procedure code for the implantation, replacement, adjustment, checking, or removal of an ICD from 1980 to 2009 and all admissions associated with that individual. For all but the few cases with ICDs implanted before 1990, this provided a minimum “look-back” of 10 years. The first hospitalization with a procedure code for the insertion of an ICD was designated the index admission. To describe incidence, an “insertion” was conservatively defined using the first contemporary code for insertion of an ICD “system” or a generator (WHO-ICD-9/9CM 37.94 and 37.96 and WHO-ICD-10 38524-00, 38393-00). Isolated codes for insertion of other elements (leads/electrodes and patches) not accompanied by a code for a system or generator were not included. Vital status and readmissions for ICD-related and cardiac reasons were identified for each patient to December 30, 2009. The cardiac diagnoses and procedures

Bradshaw et al 817

for each admission were identified from the “principal diagnosis,” and up to 19 additional codes and 11 procedure codes assigned at discharge.

Ethics The Human Research Ethics Committees of the University of Western Australia and the Department of Health WA approved the study. The linked file was prepared in accordance with established protocols for the confidentiality and protection of privacy 21 and held at the University under conditions, which meet each jurisdiction's standards for security and confidentiality.

Statistical analysis Age-specific rates of first ICD implantation of rates were calculated using annual age and sex distributions based on 5yearly census data and age groups published by the Australian Bureau of Statistics (ABS). 22 The population incident implantation rates for the years 1995, 2000, 2005, and 2009 were age standardized against the most recent census figures (2009) using the direct method. The mean and median ages of the study population exceeded 60 years, resulting in small numbers in the younger ABS-defined population groups, especially for deaths. Accordingly, age was grouped for survival analyses into b65, 65 to 74, and ≥75 years. The recipients of an ICD are described, with differences between groups defined by sex, age group, era of implantation (half decades from 1990), readmission within 12 months, and diagnoses at admission and readmission tested using Student t test for normally distributed continuous variables, 1-way analysis of variance for between-group differences, and the χ 2 test for categorical variables. All-cause mortality at 30 days and 12 months was calculated from the admission date and from one year to 5 and a maximum of 18 years for those alive at one year as was a cardiac readmission within 12 months. Actuarial survival curves were constructed using the Kaplan-Meier method with the log-rank test used to examine differences in survival between groups. Univariate associations between the outcome variables (12month, 5-year, and long-term mortality and readmission within 12 months) and covariates including age group, era, sex, diagnoses (principal and up to 19 additional diagnoses) and cardiac-related re-admission within one year were identified. After testing to ensure the assumptions of proportionality were met, Cox regression models were used to identify the predictor variables associated with 5-year mortality and long-term survival. P b .05 was considered statistically significant. Analyses were performed using IBM SPSS Statistics for Windows, Version 19.0 (IBM Corp, Armonk, NY). This study was supported by a Grant-in-Aid from the Heart Foundation of Australia, Melbourne made initially to the late Professor Konrad Jamrozik. The authors are solely responsible for the design and conduct, all study analyses, the drafting and editing of the manuscript, and its final contents.

Results There were 1,593 index admissions for ICD implantation recorded between 1988 and 2009. Most cases (63%)

American Heart Journal May 2013

818 Bradshaw et al

Figure 1














Non-thoracotomy ICD






















MADIT Multicenter Automatic Defibrillator Implantation Trial 4 AVID Antiarrhythmics versus Implantable Defibrillators 1 MUSTT The Multicenter Unsustained Tachycardia Trial 6 CIDS Canadian Implantable Defibrillator Study 2 CASH Cardiac Arrest Study Hamburg3 MADIT II Multicenter Automatic Defibrillator Implantation Trial II 5

Numbers of incident admissions for implantation of an ICD, 1990 to 2009, showing the year of publication of prominent clinical trials. MADIT, 4 AVID, 1 MUSTT, 6 CIDS, 2 CASH, 3 MADIT II, 5 CAT, 12 and SCD-HeFT. 9

were treated in three public tertiary hospitals, with the remainder treated in the 4 private hospitals with cardiac catheterization laboratories.

Trends in use The annual index admission numbers for ICD implants are shown in Figure 1. The use of ICDs increased rapidly from the year 2000, most noticeably in the 65- to 74-year age group and those 75 years and older (Figure 2). The age-specific rates for the years 1995 to 2009 are shown, as the number of implants was very low before 1995.The age-standardized implantation rate also rose sharply from 0.83 per 100,000 population in 1995, to 2.54/10 5 in 2000, 14.08/10 5 in 2005, and 14.91/10 5 in 2009. Most recipients were male (82%), and this did not change significantly across the study period. The mean age rose from 58.7 years (SD 12.6) in 1995 to 1999 to 60.6 years (SD 12.9) in 2000 to 2004 and 63.5 years (SD 12.6) in 2005 to 2009 (F = 12.6, df = 2, P b .001). Length of hospital stay decreased markedly across the study period being a mean of 24.6 (SD 14.6) days in 1990 to 1994, 12.6 (SD 8.7) days in 1995 to 1999, 6.7 (SD 8.1) days in 2000 to 2004, and 5.1 (SD 8.1) days in 2005 to 2009 (F = 99.6, df = 3, P = .001).

Trends in indications for use Ventricular tachycardia was the most frequent principal diagnosis overall, with “chronic ischaemic heart disease� (IHD), the most frequent condition coded in the principal or one of the other 19 additional diagnostic fields (Table I). Although VT and ventricular fibrillation (VF) and cardiac arrest continued to rise in number across the study period, the proportion of the ventricular arrhythmias fell from N50% to 70% of admissions before 2000 to b20% after 2005 (Figure 3). From the year 2000, the proportion of diagnoses associated with secondary prevention (VT, VF/cardiac arrest) declined relative to the proportions of HF and cardiomyopathy as principal diagnoses rose. The index admission was the first cardiac-related admission (for a minimum of 8 years prior) for 253 patients (16%). For the remainder, AMI was the most common principal diagnosis (23%) at their first cardiac admission; a further 28% had a principal diagnosis of acute coronary syndrome or IHD; 10%, congestive HF (CHF); and 7%, cardiomyopathy, whereas 3.1% and 0.2% had a prior admission for VT or VF/arrest, respectively. A third of patients (34%) had undergone at least 1 revascularization procedure before implantation of an ICD.

American Heart Journal Volume 165, Number 5

Bradshaw et al 819

Table I. Most frequent diagnoses recorded as “principal diagnosis” and for all diagnoses (principal and up to 19 other fields) at the index hospital discharge

Figure 2 70.0

75+ years 65-74


55-64 rate /100 000



Principal diagnosis All diagnoses n (%) n (%)⁎

45-54 20-44

40.0 30.0 20.0 10.0


















Age-specific incidence rates per 100,000 of population for implantation of an ICD in the Western Australian population1995 to 2009.

Thirty-day case fatality and mortality at 1 year Thirty-day case fatality was 0.6% (10 patients). One year after discharge, crude mortality was 5.5% (n = 74) overall and increased with age, being 3.3, 6.8, and 10.0% for age groups b65 years, 65 to 74 years and N75 years (χ 2 = 15.4, df = 2, P b .001), respectively. It was also greater in 1990 to 1994 (12.5%) than in each subsequent half-decade (3.8%, 6.8%, and 4.7%). Independent predictors of death within 12 months included older age: compared with those b65 years, risk was doubled for those N65 and threefold for those ≥75 years Hazard ratio [HR 3.5, 95% CI 1.9-6.5, P b .001). “Mechanical” or “infective” complications recorded during the index admission were low (1.2%) but associated with increased risk (HR 4.4, 95% CI 1.5-12.4, P = .004), as were CHF recorded in the index admission (HR 1.9, 95% CI 1.2-3.0, P b .01), chronic IHD (HR 1.8, 95% CI 1.1-3.0, P =. 02) and stroke (HR 7.8, 95% CI 1.9-34-1, P b .-1). although the CI was broad. Five-year and long-term survival Cumulative survival to 5 years was 0.74 (SE 0.014). The independent predictors of all-cause mortality within 5 years are shown in Table II. The median survival was 11.3 years (95% CI 9.3-13.4), with cumulative survival to 10 years 0.53 (SE 0.027, 85 patients remaining). Long-term survival was no different between men and women but worsened with age. Survival by age groups is shown in Table III. The factors independently associated with long-term mortality are shown in Table IV, and the adjusted survival curves for patients readmitted or not within a year, in Figure 4.

VT Cardiomyopathy CHF Chronic IHD Acute myocardial infarction VF/cardiac arrest Conduction disorders: CHB AV/LBBB Other conduction disorder Cardiac dysrhythmia † AF/AFl Other heart disease

373 (23.4) 278 (17.5) 256 (16.1) 144 (9.0) 101 (6.3) 90 (5.6) 21 (1.3) 44 (2.8) 18 (1.1) 75 (4.7) 50 (3.1) 55 (3.5)

551 (34.6) 461 (28.9) 494 (31.0) 777 (48.7) 145 (9.1) 205 (12.9) 49 (3.1) 160 (10.0) 57 (3.6) 640 (40.2) 294 (18.4) 456 (28.6)

Abbreviations: CHB, Complete heart block, AV/LBBB, atrioventricular/left bundle branch block. Principal diagnoses not in the table as b2%: stroke, syncope, ACS, noncardiac diagnosis, device programming/management. ⁎ May be in N1 category. † Includes paroxysmal supraventricular and paroxysmal tachycardia (unspecified), other ventricular premature beats, contractions or systoles, “other” rhythm disorders, and unspecified cardiac dysrhythmia.

Readmission There were 837 patients (52.5%) readmitted with a cardiac diagnosis. More than 20% were same-day or 1-day admissions, of which almost one quarter were for cardioversion. The median time to the first cardiac readmission was 6.3 months, ranging from 1 day to 14.4 years. The most frequent principal diagnoses were HF (18%), angina/chest pain (12%), VT (10.5%), and atrial fibrillation/flutter (AF/AFl) (9%). Of 12 patients with ventricular fibrillation/ventricular flutter in any diagnosis field during readmission, 4 died in hospital. The rate of readmission within 12 months for a cardiacrelated event decreased significantly over the study period being 58% of patients in 1990 to 1994 to 43%, 36%, and 30% in subsequent eras (χ 2 = 24.2, df = 3, P b .001). The factors independently associated with readmission within 12 months are shown in Table V.

Discussion Rates per million of new ICD implants estimated in quadrennial world surveys show that, in the calendar year 2005, Australia had the highest rate of ICD implantation in the Asian Pacific region, 142 per million. 23 Although that also exceeded any European country that year, Australia had only 35% of the implants estimated for the United States. In 2009, 5 countries had rates higher than Australia's, which was still only 37% of that in the United States. 24 These are crude estimates, collected from sources such as sales figures, registries, and hospital data so are likely to both underestimate and overestimate the true rates, but they serve as indicators of relative use

American Heart Journal May 2013

820 Bradshaw et al

Figure 3

Table II. Multivariable predictors of all-cause mortality within 5 years of implantation of an ICD between 1988 and 2005 (n=1512) Variable


Hazard ratio (95% CI)

Chronic IHD (index admission) AF/AFl (index admission) Readmission within 1 y CHF (index readmission) Age group compared with b65 y Age group 65-74 y Age group ≥75 y

1.6 1.6 2.4 1.5

(1.2-2.2) (1.1-2.2) (1.8-3.3) (1.1-2.1)

.001 .03 b.001 .01

2.0 (1.4-2.8) 2.8 (1.8-4.3)

b.001 b.001

Adjusted for era, index admission diagnoses of VT, CHF, cardiomyopathy, stroke or AMI.

Table III. Cumulative (Kaplan-Meier) survival by age group 5y Age group

10 y

12 y

Estimate (SE), n of remaining cases

VT=ventricular tachycardia VF/arrest=ventricular fibrillation/cardiac arrest Cardiomyopathy includes dilated cardiomyopathy (46% of total), ischaemic, Type-2 diabetes-related ischaemic, ‘not otherwise specified’ and ‘other hypertrophic’ cardiomyopathies.

Trends in the proportions of principal diagnoses at index admission by era 1990 to 2009. VF/arrest, Ventricular fibrillation/cardiac arrest. Cardiomyopathy includes dilated cardiomyopathy (46% of total), ischemic, type 2 diabetes-related ischemic, “not otherwise specified,” and “other hypertrophic” cardiomyopathies.

b65 y 65-74 y ≥75 y

0.83 (0.02), 259 0.67 (0.03), 111 0.54 (0.05), 28

0.56 (0.04), 39 0.33 (0.08), 5

⁎ At 7.5 years.

Table IV. Multivariable predictors of all-cause mortality to 12 years after implantation of an ICD (n=1512) Variable

of ICDs in countries with roughly similar population demographics. The relative underutilization in the rest of the world, compared with the United States, has been attributed to a paucity of electrophysiologists, ineffective educational programs on guidelines, and inadequate screening and referral pathways to identify and treat eligible patients. 25 Lower levels of service (and fewer implanting centers) in countries with financial constraints on health services may also limit use, even for eligible patients. 26 In particular, elderly patients with a history of ventricular arrhythmias may be less likely to be considered, as the survival benefit is reduced due to the greater risk from nonarrhythmic death. 27 The response to findings from clinical trials is evident in the trends in the use of ICDs in WA, the patients in which they were implanted, and the indications for their use. As the evidence mounted in the late 1990s for survival benefit associated with ICD use for secondary prevention among survivors of life-threatening ventricular arrhythmias or cardiac arrest, 1–3 the numbers of implants in WA rose rapidly, as they did throughout Australia. 15 The use accelerated as benefits was reported from clinical trials of ICD implantation as a primary prevention measure. 4-6,9–13 The expanded indications for ICD use were also evident in the fall in the proportion (not number) of ventricular arrhythmias after 1999, whereas diagnoses associated

0.64 (0.04), 65 0.47 (0.5), 24 0.38 (0.08), 7⁎

CHF (index admission) Chronic IHD (index admission) AF/AFl (index admission) Readmission within 1 y Era (1995-1999 vs 1990-1994) Era 2000-2004 Era 2005-2009 Age group compared with b65 y 65-74 y ≥75 y

Hazard ratio (95% CI)


1.6 (1.2-2.1) 1.5 (1.2-1.8) 1.4 (1.1-1.9) 2.2 (1.7-2.9) 0.5 (0.30-0.83) 0.6 (0.4-0.97) 0.5 (0.3-0.8)

.001 .004 .05 b.001 .001 .04 0.01

2.0 (1.5-2.6) 3.0 (2.0-4.3)

b.001 b.001

Adjusted for stroke (index admission), VT, cardiomyopathy, complications, stroke.

with secondary prevention, HF, and cardiomyopathy rose. Before this, VT and VF/cardiac arrest together had exceeded 70% of all diagnoses for the period 1995 to 1999. Among US Medicare recipients in the period 1987 to 1995, N90% of ICD use was for secondary prevention after a ventricular arrhythmia. 14 More recently in the United States, estimates of ICD use for primary prevention rose from 6.1 per 100,000 in 1993 to 46.2 in 2006. 28 Older age was associated with poorer survival, although for Western Australians 65 years or older (mean 72.3 years), mortality was lower at 1 year (7.3%) than among those of a similar age from the large US Medicare study for the period 1987 to 1995. 14 In the US population, mortality at 1 year declined from 19.3% in 1987 to 11.4% in 1994. 17 Our

American Heart Journal Volume 165, Number 5

Figure 4

Bradshaw et al 821

Table V. Multivariable predictors of readmission within 1 year after implantation of an ICD (n=1584) Variable Era 1995-1999 vs 1990-1994 Era 2000-2004 vs 1990-1994 Era 2005-2009 vs 1990-1994 AF/AFl (index admission) CHF (index admission) Cardiomyopathy (index admission) Chronic IHD (index admission) Cardiac dysrhythmia (index admission) LOS 2-7 d vs 1 d (index admission)

HR (95% CI)


0.49 (0.30-0.79) 0.40 (0.26-0.61) 0.30 (0.16-0.57) 1.8 (1.4-2.1) 1.5 (1.2-1.8) 1.4 (1.1-1.7) 1.3 (1.1-1.6) 1.5 (1.2-2.0) 1.4 (1.1-1.7)

.004 b0.001 b.001 b.001 b.001 .001 .009 .002 .006

Adjusted for sex, age group, VT, VF/arrest, AMI, stroke, complications, and revascularization pre-ICD.

Adjusted survival curves for patients with and without a cardiac readmission within 1 year of the insertion of an ICD.

findings may indicate a continuation of the trend to decreasing mortality, although other factors may account for the different outcomes in the 2 populations. All-cause mortality among a younger cohort (mean 58.2 Âą 11.9, range 12-77 years) of 270 patients from an early US study (19811988) was 8% and 26% at 1 and 5 years, similar to the 5.5% and 26% for WA. In the ALTITUDE survival study (20082009), 1-year mortality was 8% among ICD-only patients with a mean age of 65 years (remotely networked) and 66 years (nonnetworked). 18 In this study, readmission for a cardiac cause within the 12 months more than doubled the hazard for survival to 5 years and in the long term. Heart failure was the most frequent principal diagnosis at readmission within one year and increased the hazard for long-term survival. A similar result was observed in the MADIT-CRT study in which new HF events among mildly symptomatic HF subjects with an ICD greatly increased the risk of death. 29 Mechanical or infective complications arising during the index hospitalization were infrequent but increased the hazard for death within 1 year. The more immediate risk was also seen in the ICD registry from Ontario, where major complications to 45 days postimplant were associated with significantly increased hazard ratios for death at 180 days. 30 Our findings suggest that complications continue to detrimentally influence survival in the longer term.

Strengths The strength of this study lies in 30 years worth of routinely collected state-wide hospital admissions and

separations for all public and private hospitals. The accuracy of the HMDC coding has been confirmed, 21 whereas matching procedures used by the WA Data Linkage Branch have been found to be 99.89% accurate. 22 The WA population is the most representative of the 8 Australia states and territories across national socioeconomic and health economic indicators for 1991 to 2006. Only the uptake of private health insurance fell below other jurisdictions, being ranked sixth. 31

Limitations The study used administrative data, which cannot provide the detailed clinical information available from sources such as registries, limiting information about device types, electrophysiological findings, and the delivery and appropriateness of therapy. Clinical coding of device-related procedures cannot keep pace with rapid changes in technology 14 leading to frequent revisions of codes. In some versions, codes do not distinguish between initial placement, replacement, checking, or adjustment of elements of the ICD system. To validate implantation rates, we compared the number of initial ICD devices identified from the linked data to published sales numbers in 2001, 32 2005, 33 and 2009 15 and found good agreement (N90%) but not for 1997 34 (60%), although total ICD in the linked data for that year (initial and subsequent) were 88% of all sales. As sales figures have been noted to overestimate ICD use, 35 the general agreement with sales suggests that the linked data provide accurate identification of devices for this population study.

Conclusions In this population study of trends and outcomes in ICD recipients, the last 2 decades have seen rapid growth in the use of ICDs, the expansion of their use among older patients, and for primary prevention in addition to secondary prevention. Survival to 5 years was 74%, and to 10 years, 53%. For those younger than 65 years, more

822 Bradshaw et al

than half survived at least 12 years. For patients aged 75 years or more, a majority survived to 5 years.

References 1. The Antiarrhythmics Versus Implantable Defibrillators (AVID) Investigators. A comparison of antiarrhythmic-drug therapy with implantable defibrillators in patients resuscitated from near-fatal ventricular arrhythmias. N Engl J Med 1997;337: 1576-83. 2. Connolly SJ, Gent M, Roberts RS, et al. Canadian Implantable Defibrillator Study (CIDS): a randomized trial of the implantable cardioverter defibrillator against amiodarone. Circulation 2000;101: 1297-302. 3. Kuck KH, Cappato R, Siebels J, et al. Randomized comparison of antiarrhythmic drug therapy with implantable defibrillators in patients resuscitated from cardiac arrest. The Cardiac Arrest Study Hamburg (CASH). Circulation 2000;102:748-54. 4. Moss AJ, Hall WJ, Cannom DS, et al. Improved survival with an implanted defibrillator in patients with coronary disease at high risk for ventricular arrhythmias. N Engl J Med 1996;329:1933-40. 5. Moss AJ, Zareba W, Hall WJ, et al. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med 2002;346:877-83. 6. Buxton AE, Lee KL, Fisher JD, et al. for the Multicenter Unsustained Tachycardia Trial Investigators. A randomized study of the prevention of sudden death in patients with coronary artery disease. N Engl J Med 1999;341:1882-90. 7. Goldenberg I, Gillespie J, Moss AJ, et al. Long-term benefit of primary prevention with an implantable cardioverter-defibrillator. Circulation 2010;122:1265-71. 8. Lee DS, Green LD, Liu PP, et al. Effectiveness of implantable defibrillators for preventing arrhythmic events and death. J Am Coll Cardiol 2003;41:1573-82. 9. Bardy GH, Lee LL, Mark DB, et al. Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure. N Engl J Med 2005;352:225-37. 10. Hernandez AF, Fonarow GC, Hammill BG, et al. Clinical effectiveness of implantable cardioverter-defibrillators among Medicare beneficiaries with heart failure. Circ Heart Fail 2010;3: 7-13. 11. Anand IS, Carson P, Galle E, et al. Cardiac resynchronization therapy reduces the risk of hospitalizations in patients with advanced heart failure: results from the Comparison of Medical Therapy, Pacing and Defibrillation in Heart Failure (COMPANION) Trial. Circulation 2009;119:969-77. 12. Bӓnsch D, Antz M, Boczor S, et al. Primary prevention of sudden cardiac death in idiopathic dilated cardiomyopathy. The Cardiomyopathy Trial (CAT). Circulation 2002;105:1453-8. 13. Kadish A, Dyer A, Daubert JP, et al. Prophylactic defibrillator implantation in patients with nonischemic dilated cardiomyopathy. N Engl J Med 2004;350:2151-8. 14. Hlatky MA, Saynina O, McDonald KM, et al. Utilization and outcomes of the implantable cardioverter defibrillator, 1987 to 1995. Am Heart J 2002;144:397-403. 15. Mond HG, Whitlock RM. The Australian and New Zealand cardiac pacing and Implantable Cardioverter-Defibrillator Survey: Calendar Year 2009. Heart Lung Circ 2011;20:99-104. 16. Bokhari F, Newman D, Green M, et al. Long-term comparison of a subset of patients in the Canadian Implantable Defibrillator Study (CIDS). Circulation 2004;110:112-6.

American Heart Journal May 2013

17. Winkle RA, Mead RH, Ruder MA, et al. Long-term outcome with the automatic implantable cardioverter-defibrillator. J Am Coll Cardiol 1989;13:1353-61. 18. Saxon LA, Hayes DL, Gilliam FR, et al. Long-term outcomes after ICD and CRT implantation and influence of remote device follow-up. Circulation 2010;122:2359-67. 19. Australian Bureau of Statistics. Population by age and sex, Western Australia. 0~2011~Main+Features~Western+Australia?OpenDocument. 20. Holman CDJ, Bass AJ, Rouse IL, et al. Population-based linkage of health records in Western Australia: development of a health services research linked database. Aust NZ J Public Health 1999; 23:453-9. 21. Kelman CW, Bass AJ, Holman CDJ. Research use of linked health data—a best practice protocol. Aust NZ J Public Health 2002;26: 251-5. 22. Australian Bureau of Statistics 2010, Australian Demographic Statistics, Jun 2010, cat no. 3101.0, ABS, Canberra. www.ausstats.…nsf/0/…/31010_jun%202010.pdf. 23. Mond HG, Irwin M, Ector H, et al. The World Survey of Cardiac Pacing and Implantable Cardioverter-Defibrillators: Calendar year 2005. PACE 2008;31:1202-12. 24. Mond HG, Proclemer A. The 11th World Survey of Cardiac Pacing and Implantable Cardioverter-Defibrillators: Calendar year 2009. PACE 2011;34:1013-27. 25. Camm AJ, Nisam S. European utilization of the implantable defibrillator: has 10 years changed the “enigma”? Europace 2010; 12:1063-9. 26. Birnie DH, Sambell C, Johansen H, et al. The use of implantable defibrillators in Canadian and US survivors of out-of-hospital cardiac arrest. CMAJ 2007;177:141-6. 27. Healey JS, Hallstrom AP, Kuck K-H, et al. Role of the implantable defibrillator among elderly patients with a history of life-threatening ventricular arrhythmias. Eur Heart J 2007;28: 1746-9. 28. Kurtz SM, Ochoa JA, Lau E, et al. Implantation trends and patient profiles for pacemakers and implantable cardioverter defibrillators in the United States: 1993-2006. PACE 2010;33:705-11. 29. Goldenberg I, Hall WJ, Beck CA, et al. Reduction of the risk for recurring heart failure events with cardiac resynchronisation therapy; MADIT-CRT (Multicenter Automatic Implantation Trial with Cardiac Resynchronisation Therapy). J Am Coll Cardiol 2011;58:729-37. 30. Lee DS, Krahn AD, Healey JS, et al. Evaluation of early complications related to de novo cardioverter defibrillator implantation insights from the Ontario ICD database. J Am Coll Cardiol 2010;55:774-82. 31. Clark A, Preen DB, Ng JQ, et al. Is Western Australia representative of other Australian States and Territories in terms of key socio-economic and health economic indicators? Aust Health Rev 2010;34:210-5. 32. Mond HG, Whitlock RM. The Australian and New Zealand cardiac pacing and Implantable Cardioverter-Defibrillator Survey: Calendar Year 2001. Heart Lung Circ 2004;13:145-9. 33. Mond HG, Whitlock RM. The Australian and New Zealand cardiac pacing and Implantable Cardioverter-Defibrillator Survey: Calendar Year 2005. Heart Lung Circ 2008;17:85-9. 34. Mond HG, Whitlock RM. The Australian and New Zealand cardiac pacing and Implantable Cardioverter-Defibrillator Survey: Calendar Year 1997. Intern Med J 2001;31:267-71. 35. Peinado R, Torrecilla EG, Ormaetxe J, et al. Spanish Implantable Cardioverter-Defibrillator Registry. 5th Official Report of the Spanish Society of Cardiology Working Group on Implantable Cardioverter-Defibrillators (2008). Rev Esp Cardiol 2009;62: 1435-49.