Advertisement
Canadian Journal of Cardiology

Implantable Cardioverter Defibrillator Implantation Rates After Out of Hospital Cardiac Arrest: Are the Rates Guideline-Concordant?

  • Edwin C. Ho
    Affiliations
    Department of Medicine, University of Toronto, Toronto, Ontario, Canada

    Division of Cardiology, St Michael's Hospital, Toronto, Ontario, Canada
    Search for articles by this author
  • Sheldon Cheskes
    Affiliations
    Department of Family and Community Medicine, Division of Emergency Medicine, University of Toronto, Toronto, Ontario, Canada

    Rescu, Li Ka Shing Knowledge Institute, St Michael's Hospital, Toronto, Ontario, Canada

    Sunnybrook Centre for Prehopital Medicine, Toronto, Ontario, Canada
    Search for articles by this author
  • Paul Angaran
    Affiliations
    Department of Medicine, University of Toronto, Toronto, Ontario, Canada

    Division of Cardiology, St Michael's Hospital, Toronto, Ontario, Canada
    Search for articles by this author
  • Laurie J. Morrison
    Affiliations
    Department of Medicine, University of Toronto, Toronto, Ontario, Canada

    Rescu, Li Ka Shing Knowledge Institute, St Michael's Hospital, Toronto, Ontario, Canada

    Division of Emergency Medicine, Department of Medicine, University of Toronto, Toronto, Ontario, Canada
    Search for articles by this author
  • Theresa Aves
    Affiliations
    Department of Medicine, University of Toronto, Toronto, Ontario, Canada

    Division of Cardiology, St Michael's Hospital, Toronto, Ontario, Canada
    Search for articles by this author
  • Cathy Zhan
    Affiliations
    Rescu, Li Ka Shing Knowledge Institute, St Michael's Hospital, Toronto, Ontario, Canada
    Search for articles by this author
  • Dennis T. Ko
    Affiliations
    Department of Medicine, University of Toronto, Toronto, Ontario, Canada

    Schulich Heart Centre, Division of Cardiology, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada

    Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada
    Search for articles by this author
  • Paul Dorian
    Correspondence
    Corresponding author: Dr Paul Dorian, St Michael's Hospital, 30 Bond St, 6-050Q, Toronto, Ontario M5B 1W8, Canada. Tel.: +1-416-864-5104; fax: +1-416-864-5849.
    Affiliations
    Department of Medicine, University of Toronto, Toronto, Ontario, Canada

    Division of Cardiology, St Michael's Hospital, Toronto, Ontario, Canada
    Search for articles by this author
  • on behalf of theRescu Epistry Investigators
Published:September 06, 2017DOI:https://doi.org/10.1016/j.cjca.2017.05.013

      Abstract

      Background

      Clinical practice guidelines recommend implantable cardioverter defibrillators (ICDs) for the secondary prevention of sudden death after a cardiac arrest not from a reversible cause, but “real world” implantation rates are not well described.

      Methods

      Adults with out of hospital cardiac arrest attended by Emergency Medical Services are captured in the Toronto Regional RescuNET database. We analyzed those who survived to hospital discharge and collected data on age, sex, initial rhythm, ST-elevation myocardial infarction (STEMI) on presenting electrocardiogram (ECG), in-hospital revascularization, neurologic status (Modified Rankin Scale [MRS]) at discharge, and admission hospital type. To estimate ‘indicated’ ICD implantation rates, “likely ICD-eligible” patients were defined as having an initial shockable rhythm, no STEMI on presenting ECG, no revascularization, and good neurologic status (MRS 0-3). “Not likely ICD-eligible” patients were defined as having a STEMI on presenting ECG, revascularization, or poor neurologic status (MRS 4-5).

      Results

      In the 1238 adults (2011-2014) analyzed, the overall ICD implantation rate was 23.9%. Two hundred fifty-six patients were “likely ICD-eligible,” of whom 146 (57.0%) received an ICD. The implantation rate for “not likely ICD-eligible” patients was 16.7% (112 of 670). ICD eligibility could not be determined for 312 patients, of whom 38 (12.2%) received an ICD. Admission to a hospital with ICD implantation facilities was associated with a higher probability of ICD implantation (odds ratio, 2.85; 95% confidence interval, 1.40-5.82).

      Conclusions

      Postcardiac arrest ICD implantation rates in eligible patients are lower than expected. Implementation strategies to monitor guideline adherence after out of hospital cardiac arrest are warranted.

      Résumé

      Introduction

      Les lignes directrices de pratique clinique recommandent les défibrillateurs cardioverteurs implantables (DCI) pour la prévention secondaire de la mort subite après un arrêt cardiaque non associé à une cause réversible, mais les taux d’implantation réels ne sont pas décrits clairement.

      Méthodes

      Les adultes ayant subi un arrêt cardiaque hors de l’hôpital qui ont été pris en charge par les services médicaux d’urgence sont inscrits dans la base de données RescuNET de la région de Toronto. Nous nous sommes penchés sur ceux qui avaient survécu après la sortie de l’hôpital et avons recueilli les données sur l’âge, le sexe, le rythme initial, l’infarctus du myocarde (IM) avec sus-décalage du segment ST à l’électrocardiogramme (ÉCG) à l’admission, la revascularisation à l’hôpital, l’état neurologique (Échelle de Rankin modifiée [mRS, de l’anglais Modified Rankin Scale]) à la sortie de l’hôpital et le type d’hôpital à l’admission. Pour estimer les taux d’implantation « indiquée » de DCI, nous avons défini les patients « susceptibles d’être admissibles » à l’implantation d’un DCI, soit ceux ayant un rythme initial traitable par choc, n’ayant pas d’IM avec sus-décalage du segment ST à l’ÉCG à l’admission, n’ayant pas de revascularisation et ayant un bon état neurologique (mRS 0-3). Nous avons défini les patients qui « ne sont pas susceptibles d’être admissibles » à l’implantation d’un DCI, soit ceux ayant un IM avec sus-décalage du segment ST à l’ÉCG à l’admission, une revascularisation ou un état neurologique médiocre (mRS 4-5).

      Résultats

      Parmi les 1238 adultes (2011-2014) qui ont fait l’objet de l’analyse, le taux global d’implantation de DCI était de 23,9 %. Parmi les 256 patients « susceptibles d’être admissibles » à l’implantation d’un DCI, 146 (57,0 %) ont reçu un DCI. Le taux d’implantation chez les patients qui « ne sont pas susceptibles d’être admissibles » à l’implantation d’un DCI était de 16,7 % (112 sur 670). Parmi les 312 patients dont l’admissibilité à l’implantation d’un DCI n’a pas pu être déterminée 38 (12,2 %) ont reçu un DCI. L’admission à un hôpital équipé pour l’implantation de DCI a été associée à une plus forte probabilité d’implantation de DCI (ratio d’incidence approché, 2,85 ; intervalle de confiance à 95 %, 1,40-5,82).

      Conclusions

      Les taux d’implantation de DCI après l’arrêt cardiaque chez les patients admissibles sont plus bas que ce à quoi nous nous attendions. Des stratégies de mise en œuvre pour s’assurer du respect des lignes directrices après un arrêt cardiaque hors de l’hôpital sont justifiées.
      Out of hospital cardiac arrest (OHCA) is a major public health issue, with an incidence of Emergency Medical Services (EMS)-attended arrests of approximately 52 per 100,000 people per year in North America.
      • Nichol G.
      • Thomas E.
      • Callaway C.W.
      • et al.
      Regional variation in out-of-hospital cardiac arrest incidence and outcome.
      Among those with ventricular fibrillation (VF) as the initial rhythm, approximately 20%-25% survive to hospital discharge.
      • Kudenchuk P.J.
      • Brown S.P.
      • Daya M.
      • et al.
      Amiodarone, lidocaine, or placebo in out-of-hospital cardiac arrest.
      Guidelines, including those from the American Heart Association, the Canadian Cardiovascular Society, and the European Society of Cardiology, recommend a “secondary prevention” implantable cardioverter defibrillator (ICD) after a cardiac arrest not from a reversible cause if the patient survives with good neurologic function.
      • Tang A.S.
      • Ross H.
      • Simpson C.S.
      • et al.
      Canadian Cardiovascular Society/Canadian Heart Rhythm Society position paper on implantable cardioverter defibrillator use in Canada.
      • Bennett M.
      • Parkash R.
      • Nery P.
      • et al.
      Canadian Cardiovascular Society/Canadian Heart Rhythm Society 2016 implantable cardioverter-defibrillator guidelines.
      • Epstein A.
      • DiMarco J.
      • Ellenbogen K.
      • et al.
      ACC/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the ACC/AHA/NASPE 2002 Guideline).
      • Zipes D.P.
      • Camm A.J.
      • Borggrefe M.
      • et al.
      ACC/AHA/ESC 2006 guidelines for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death: a report of the American College of Cardiology/American Heart Association Task Force and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Develop guidelines for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death) developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society.
      Three large studies and 2 meta-analyses have reported a significant reduction in all-cause mortality and arrhythmic death in such patients who receive ICDs vs medical management with antiarrhythmic medications.
      • Wyse D.G.
      • Friedman P.L.
      • Epstein A.E.
      A comparison of antiarrhythmic-drug therapy with implantable defibrillators in patients resuscitated from near-fatal ventricular arrhythmias. The Antiarrhythmics versus Implantable Defibrillators (AVID) Investigators.
      • Connolly S.J.
      • Gent M.
      • Roberts R.S.
      • et al.
      Canadian implantable defibrillator study (CIDS): a randomized trial of the implantable cardioverter defibrillator against amiodarone.
      • Kuck K.H.
      • Cappato R.
      • Siebels J.
      • Rüppel R.
      Randomized comparison of antiarrhythmic drug therapy with implantable defibrillators in patients resuscitated from cardiac arrest: the Cardiac Arrest Study Hamburg (CASH).
      • Lee D.S.
      • Green L.D.
      • Liu P.P.
      • et al.
      Effectiveness of implantable defibrillators for preventing arrhythmic events and death: a meta-analysis.
      • Connolly S.J.
      • Hallstrom A.P.
      • Cappato R.
      • et al.
      Meta-analysis of the implantable cardioverter defibrillator secondary prevention trials. AVID, CASH and CIDS studies. Antiarrhythmics vs Implantable Defibrillator study. Cardiac Arrest Study Hamburg. Canadian Implantable Defibrillator Study.
      The pooled number needed to treat to prevent 1 death from all causes was found to be 15 on the basis of a follow-up period ranging from 18 to 57 months in these studies.
      • Lee D.S.
      • Green L.D.
      • Liu P.P.
      • et al.
      Effectiveness of implantable defibrillators for preventing arrhythmic events and death: a meta-analysis.
      One cardiac arrest registry and 2 administrative database studies from more than 10 years ago have reported low rates of ICD implantation after OHCA.
      • Birnie D.H.
      • Sambell C.
      • Johansen H.
      • et al.
      Use of implantable cardioverter defibrillators in Canadian and US survivors of out-of-hospital cardiac arrest.
      • Parkash R.
      • Tang A.
      • Wells G.
      • et al.
      Use of implantable cardioverter defibrillators after out-of-hospital cardiac arrest: a prospective follow-up study.
      • Voigt A.
      • Ezzeddine R.
      • Barrington W.
      • et al.
      Utilization of implantable cardioverter-defibrillators in survivors of cardiac arrest in the United States from 1996 to 2001.
      • Saba S.
      • Ravipati L.P.
      • Voigt A.
      Recent trends in utilization of implantable cardioverter-defibrillators in survivors of cardiac arrest in the united states.
      More than 15 years have now passed since the first guideline recommendations for secondary prevention ICDs, but contemporary data on rates of ICD implantation and factors associated with ICD implantation are scarce. Additionally, existing studies have not focused on distinguishing between implantation rates in patients who were likely eligible for a secondary prevention device on the basis of guideline recommendations vs those who were not likely eligible.
      The purpose of this study was to determine recent “real-world” ICD implantation rates after OHCA in a large urban centre, and to provide an estimation of “likely ICD-eligible” implantations on the basis of current guideline recommendations vs “not likely ICD-eligible” implantations using a large prospective registry of OHCAs with in-hospital follow-up.

      Methods

      Study population

      This was an observational cohort study using data from the population-based, prospective Rescu Epistry cardiac arrest database. Rescu Epistry is comprised of data points from the Resuscitation Outcomes Consortium Epistry-Cardiac Arrest database and the Strategies for Post Arrest Care database; the methodologies of these 2 databases are described elsewhere.
      • Morrison L.J.
      • Nichol G.
      • Rea T.D.
      • et al.
      Rationale, development and implementation of the Resuscitation Outcomes Consortium Epistry-Cardiac Arrest.
      • Lin S.
      • Morrison L.J.
      • Brooks S.C.
      Development of a data dictionary for the Strategies for Post Arrest Resuscitation Care (SPARC) network for post cardiac arrest research.
      Adults aged 18 years or older who sustained an OHCA of no obvious cause, were treated by EMS, and who survived to hospital discharge, between 2011 and 2014, were included in this study. The Toronto Rescu Epistry database collects prehospital and in-hospital data from 42 hospitals and their catchment areas in the Greater Toronto Area (population 6.6 million) for any cardiac arrest patient.
      Exclusion criteria for this study include a Do Not Resuscitate status at the time of cardiac arrest or during hospital admission and the presence of a preexisting ICD. Approval for Rescu Epistry was provided by all research ethics boards of participating hospitals.

      Data collection and variable definitions

      All data in the Rescu Epistry cardiac arrest database is collected by trained data abstractors with previously documented and verified accuracy and completeness.
      • Morrison L.J.
      • Nichol G.
      • Rea T.D.
      • et al.
      Rationale, development and implementation of the Resuscitation Outcomes Consortium Epistry-Cardiac Arrest.
      • Lin S.
      • Morrison L.J.
      • Brooks S.C.
      Development of a data dictionary for the Strategies for Post Arrest Resuscitation Care (SPARC) network for post cardiac arrest research.
      Baseline characteristics were obtained for the study population including year of cardiac arrest event, age, sex, admission hospital capacity, initial rhythm, witnessed cardiac arrest, arrest location, and neurologic status at discharge according to the Modified Rankin Scale (MRS). Variables related to evidence of ischemia or infarction as a potential cause of the arrest, including ST-elevation myocardial infarction (STEMI) on presenting electrocardiogram (ECG), having a coronary angiogram, and revascularization using percutaneous coronary intervention or coronary artery bypass grafting on the index hospitalization were also collected. Last, data on any hospital transfer during hospitalization were obtained.
      Using each patient's unique provincial health card number, information regarding delayed ICD implantation within 1 year and death within 1 year after discharge from the index hospitalization was obtained by linking with provincial administrative databases (including the Canadian Institute for Health Information Discharge Abstract Database and Same Day Surgery Database as well as the Ontario Registered Persons Database) through the Institute for Clinical Evaluative Sciences.
      For the purpose of this study, a probable ischemic, reversible cause for cardiac arrest was defined as either having: (1) a STEMI on the presenting ECG; or (2) any revascularization procedure, either percutaneous coronary intervention or coronary artery bypass grafting, during the index hospitalization. “No definite ischemia” was defined as either the absence of a STEMI on presenting ECG or a blank STEMI field in the database, combined with no in-hospital revascularization.
      An ICD-capable hospital was defined as any of the 6 hospitals in the greater Toronto area with on-site ICD implantation capability. A non-ICD-capable hospital does not have these facilities and therefore a transfer would need to occur for a patient to receive a device implantation.
      A shockable initial rhythm was defined as the first documented cardiac rhythm from the first prehospital responder as either being VF, pulseless ventricular tachycardia (VT), shock using automatic external defibrillator with no available rhythm strip for documentation, or documentation as “shockable” in the arrest record. A nonshockable initial rhythm includes pulseless electrical activity, asystole, no shock using an automatic external defibrillator with no rhythm strip, undetermined rhythm, no documented rhythm, or documented as “not shockable” in the arrest record.
      We defined good neurologic status at discharge as having an MRS score of 0-3, which indicates no symptoms of neurologic dysfunction to moderate disability requiring some assistance but remaining independently ambulatory. A poor neurologic status was defined as MRS scores of 4-5, which include moderately severe to severe disability.

      Group definitions

      The “likely ICD-eligible” group describes those with a guideline indication for a secondary prevention ICD and who are likely able to benefit from ICD therapy, defined as having all 3 of a shockable initial rhythm, no definite ischemia, and a good neurologic status. The “not likely ICD-eligible” group describes individuals who would not meet a clear guideline indication for a secondary prevention ICD because of a probable reversible cause, or who are unlikely to derive any clinical benefit from ICD therapy. We defined this group as those having either evidence of an ischemic or reversible cause for the cardiac arrest or a poor neurologic status at discharge.

      Statistical analysis

      Categorical variables are expressed as the absolute count and percentage of the total, and continuous variables are expressed as the mean ± SD. Statistical analysis was performed using GraphPad Prism version 6.05 (GraphPad Software, Inc, La Jolla, CA) for all parametric and nonparametric analyses of the factors associated with ICD implantation. A 2-tailed χ2 test was used to test for statistically significant differences in categorical variables and a 2-tailed t test was used for continuous variables. A χ2 test for trend was used to evaluate for time trends in ICD implantation rate, analyzing time as an independent variable. Statistical significance was defined as a P value < 0.05.
      Multivariate binary logistic regression was performed to create 2 models of factors associated with ICD implantation using SPSS version 24 (IBM Corp, Armonk, NY). The first model included the entire study cohort. The second model included only patients who met our definition of “likely ICD-eligible” to better understand this subgroup in whom factors associated with ICD implantation is of particular clinical relevance. Independent variables including age, sex, initial presenting rhythm, arrest etiology, location of arrest, hospital implantation capability, transfer, neurological status, and year were first tested in univariate logistic regression models where P < 0.1 or biologic plausibility were criteria for entry to the multivariate model. Odds ratios (ORs) with 95% confidence intervals (CIs) are reported.

      Results

      Baseline characteristics

      One thousand two hundred thirty-eight adults survived to discharge and were included in the analysis. Baseline characteristics for the study cohort are described in Table 1. A flow chart of the study cohort according to neurologic status at discharge, evidence of a STEMI on presenting ECG, revascularization procedures, and ICD implantation is shown in Figure 1. The overall ICD implantation rate was 23.9% (296 implantations), with no statistically significant change over the 4-year data collection period (Fig. 2). The cohort was predominantly male (74.5%) with an average age of 60 years old. Most cardiac arrests were witnessed (84%) by either a bystander (56.3%) or EMS (27.7%) and 64.6% had an initial shockable rhythm; 61.9% occurred in a private location. Approximately half had evidence of ischemia (51.2%). Approximately two-thirds were admitted to non-ICD-capable hospitals (66.8%). Most survivors were discharged with a good neurologic status (96.4% MRS 1-3).
      Table 1Baseline characteristics
      CharacteristicValue
      All surviving to discharge, N1238
      Received an ICD296 (23.9)
      Average age (±SD), years59.9 ± 15.3
      Age category, years
       18-39122 (9.9)
       40-59464 (37.5)
       60-75465 (37.6)
       > 75187 (15.1)
      Sex
       Male922 (74.5)
       Female316 (25.5)
      Initial rhythm
       Shockable800 (64.6)
       Not shockable438 (35.4)
      Evidence of ischemia
       Yes634 (51.2)
       No597 (48.2)
      Witnessed arrest
       No witness198 (16.0)
       Any witness1040 (84.0)
      EMS witness343 (27.7)
      Bystander witness697 (56.3)
      Location of arrest
       Public470 (38.0)
       Private766 (61.9)
      Admission hospital type
       ICD-capable411 (33.2)
       Not ICD-capable827 (66.8)
      Any hospital transfer
       Yes503 (40.6)
       No735 (59.4)
      Neurologic status at discharge
       Good (MRS 0-3)1182 (95.5)
       Poor (MRS 4-5)44 (3.6)
      Values are presented as n (%) except where otherwise noted.
      EMS, Emergency Medical Services; ICD, implantable cardioverter defibrillator; MRS, Modified Rankin Scale.
      Figure thumbnail gr1
      Figure 1Study cohort flow chart. “Revascularized” includes percutaneous coronary intervention or coronary artery bypass grafting during the index hospitalization. ICD, implantable cardioverter defibrillator; MRS, Modified Rankin Scale; STEMI, ST-elevation myocardial infarction.
      Figure thumbnail gr2
      Figure 2Implantable cardioverter defibrillator (ICD) implantation rate over time. “Likely ICD eligible” = all of shockable initial rhythm, no definite ischemia and good neurologic status; “Not likely ICD eligible” = either evidence of ischemia or poor neurologic status. P value for time trend.

      Factors associated with ICD implantation

      Table 2 outlines the factors independently associated with ICD implantation after OHCA with survival to discharge. Patients who received an ICD compared with those who did not were more likely to be male (79.4% vs 72.9%; P = 0.026), have a shockable initial rhythm (81.8% vs 59.2%; P < 0.001), an arrest in a public location (47.6% vs 35%; P < 0.001), no definite ischemia (62.2% vs 44.1%; P < 0.001), and good neurologic status (99.3% MRS 0-3 vs 94.3%; P < 0.001). In 37.5% of ICD implantations, there was evidence of ischemia after OHCA.
      Table 2Factors associated with ICD implantation
      Entire cohortReceived an ICDDid not receive ICDP
      All surviving to discharge, n1238296931
      Average age (±SD), years59.9 ± 15.360.0 ± 14.860.0 ± 15.51.00
      Age category0.006
       18-39122 (9.9)29 (9.8)91 (9.8)
       40-59464 (37.5)95 (32.1)364 (39.1)
       60-75465 (37.6)136 (45.9)326 (35.0)
       > 75187 (15.1)36 (12.2)150 (16.1)
      Sex0.026
       Male922 (74.5)235 (79.4)679 (72.9)
       Female316 (25.5)61 (20.6)252 (27.1)
      Initial rhythm< 0.001
       Shockable800 (64.6)242 (81.8)551 (59.2)
       Not shockable438 (35.4)54 (18.2)380 (40.8)
      Evidence of ischemia< 0.001
       Yes634 (51.2)111 (37.5)518 (55.6)
       No597 (48.2)184 (62.2)411 (44.1)
      Witnessed arrest0.79
       No witness198 (16.0)49 (16.6)148 (15.9)
       Witnessed1040 (84.0)247 (83.4)783 (84.1)
      By EMS343 (27.7)39 (13.2)300 (32.2)
      By bystander697 (56.3)208 (70.3)483 (51.9)
      Location of arrest< 0.001
       Public470 (38.0)141 (47.6)326 (35.0)
       Private766 (61.9)154 (52.0)604 (64.9)
      Admission hospital type0.29
       ICD-capable411 (33.2)91 (30.7)317 (34.0)
       Not ICD-capable827 (66.8)205 (69.3)614 (66.0)
      Any hospital transfer< 0.001
       Yes503 (40.6)156 (52.7)344 (36.9)
       No735 (59.4)140 (47.3)587 (63.1)
      Neurologic status< 0.001
       Good (MRS 0-3)1182 (95.5)294 (99.3)878 (94.3)
       Poor (MRS 4-5)44 (3.6)1 (0.3)43 (4.6)
      Values are presented as n (%) except where otherwise noted.
      EMS, Emergency Medical Services; ICD, implantable cardioverter defibrillator; MRS, Modified Rankin Scale.
      A multivariate binary logistic regression analysis performed on the factors associated with ICD implantation and estimated ORs are outlined in Table 3. The factors independently associated with an ICD implantation were good neurologic function (OR, 16.22; 95% CI, 2.13-123.29), evidence of ischemia (OR, 0.14; 95% CI, 0.10-0.20), a shockable initial rhythm (OR, 5.19; 95% CI, 3.56-7.57), any interhospital transfer (OR, 2.83; 95% CI, 2.01-3.97), a public arrest location (OR, 1.99; 95% CI, 1.46-2.70), and admission to an ICD-capable hospital (OR, 1.60; 95% CI, 1.12-2.27). None of the other variables included in the regression model were predictive of ICD implantation, including sex, age, and year of the cardiac arrest when referenced to the first year of the analysis (2011).
      Table 3Factors associated with ICD implantation (multivariate analysis)
      FactorOR (95% CI)
      Year
       2012 vs 20111.46 (0.93-2.29)
       2013 vs 20111.48 (0.96-2.29)
       2014 vs 20111.53 (0.99-2.38)
      Age1.01 (1.00-1.02)
      Sex (female vs male)0.89 (0.61-1.28)
      Shockable initial rhythm5.19 (3.56-7.57)
      Ischemic0.14 (0.10-0.20)
      Public arrest location1.99 (1.46-2.70)
      Admitted to ICD-capable hospital1.60 (1.12-2.27)
      Any transfer2.83 (2.01-3.97)
      Good neurologic function (MRS 0-3)16.22 (2.13-123.29)
      CI, confidence interval; ICD, implantable cardioverter defibrillator; MRS, Modified Rankin Scale; OR, odds ratio.

      Rates of ischemia and revascularization

      Detailed rates of ICD implantation according to evidence of ischemia and revascularization are summarized in Table 4. In the entire cohort, 511 patients (41.3%) were diagnosed with a STEMI on the initial ECG, and 376 of these patients (73.6%) subsequently underwent a revascularization procedure. Most of the patients diagnosed with a STEMI did not receive a secondary prevention ICD (81.8%); of the 93 STEMI patients (18.2%) who did receive an ICD, 30 had been revascularized. An additional 123 patients (9.9%) of the entire study cohort were not diagnosed with a STEMI on the initial ECG but underwent an in-hospital revascularization procedure. Five hundred four patients (40.7%) never had an angiogram during their hospital stay.
      Table 4ICD implantation according to evidence of ischemia
      Entire study cohort (N = 1238), nReceived an ICD, n (%)Did not receive an ICD, n (%)
      All with evidence of ischemia634111 (17.5)518 (81.7)
       STEMI51193 (18.2)418 (81.8)
      With revascularization37630 (8.0)346 (92.0)
      With no revascularization13563 (46.7)72 (53.3)
       No STEMI but revascularized123
      ICD status missing in 5 patients.
      18 (14.6)100 (81.3)
      All with no definite ischemia597184 (30.8)411 (68.8)
       Angiogram with no revascularization9365 (69.9)26 (28.0)
       No angiogram504119 (23.6)385 (76.4)
      ICD, implantable cardioverter defibrillator; STEMI, ST-elevation myocardial infarction.
      ICD status missing in 5 patients.

      Implantation rates in ICD-eligible patients on the basis of guideline recommendations

      Two hundred fifty-six patients were considered “likely ICD-eligible” according to our definition, of whom 146 (57.0%) received an ICD (Table 5). Crude (unadjusted for baseline characteristics) implantation rates increased from 46.3% in 2011 to 63.8% in 2014 (P = 0.05 for time trend; Fig. 2). In the 670 patients who were “not likely ICD-eligible,” 112 (16.7%) received an ICD (Table 5). ICD eligibility could not be defined in 312 patients, most of whom (95.8%) had no definite ischemia and good neurologic status, but did not have a documented shockable initial rhythm. In this group, 38 patients (12.2%) received an ICD (Table 5).
      Table 5ICD implantation rates on the basis of eligibility
      “Likely ICD-eligible”“Not likely ICD-eligible”ICD eligibility unknown
      All surviving to discharge, n256670312
      Received an ICD, n14611238
      ICD implant rate, %57.016.712.2
      ICD, implantable cardioverter defibrillator.
      Factors associated with ICD implantation in the “likely ICD-eligible” group only are described in Table 6. Similar to the whole cohort, admission to an ICD-capable hospital (OR, 2.85; 95% CI, 1.40-5.82) or any interhospital transfer (OR, 3.70; 95% CI, 2.03-6.74) were associated with a higher probability of ICD implantation. Age, sex, public arrest location, and year of the cardiac arrest when referenced to the first year of the analysis (2011) were not associated with receiving an ICD.
      Table 6Factors associated with ICD implantation among “likely ICD-eligible” patients (multivariate analysis)
      OR (95% CI)
      Year
       2012 vs 20111.37 (0.64-2.91)
       2013 vs 20111.48 (0.72-3.03)
       2014 vs 20111.55 (0.72-3.33)
      Age1.00 (0.99-1.02)
      Sex (female vs male)1.06 (0.56-2.02)
      Public arrest location0.94 (0.54-1.65)
      Admitted to ICD-capable hospital2.85 (1.40-5.82)
      Any transfer3.70 (2.03-6.74)
      CI, confidence interval; ICD, implantable cardioverter defibrillator; OR, odds ratio.
      Characteristics of the patients who met criteria for the “likely ICD-eligible” group but did not receive an ICD, are described in additional detail in the Supplemental Table S1. These patients were demographically similar to the overall cohort (75.9% male, average age 58 years, 82.4% witnessed arrests) but a large proportion were from non-ICD-capable hospitals (84.3%) and had no interhospital transfer (75.9%). The relationship between hospital transfer, ICD implantation rate and hospital ICD capability are described in the Supplemental Table S2.

      Implantation rates according to only STEMI and neurologic status

      When we analyzed patients according to only STEMI and neurologic status, 548 patients (44.3% of the entire cohort) were either diagnosed with a STEMI on presenting ECG or had a poor neurologic status at discharge. Ninety-four (17.2%) of these patients underwent ICD implantation. Among the remaining 690 patients who did not meet either criteria (55.7% of the entire cohort), 202 (29.3%) received an ICD.

      Mortality at 1 year and delayed ICD implantation

      All-cause mortality rates at 1 year after the cardiac arrest are described in Table 7. There were fewer deaths among those who received an ICD (12 deaths; 4.2%) compared with those who did not receive an ICD (86 deaths; 9.5%). The mortality rate was highest in those who were considered “likely ICD-eligible” or “ICD eligibility unknown” according to our study definition and did not receive an ICD (50 deaths; 13.6%).
      Table 7All-cause mortality at 1 year after cardiac arrest
      Received an ICD (n = 288)Did not receive an ICD (all; n = 906)Did not receive an ICD (only if “likely ICD-eligible” or “ICD eligibility unknown”; n = 368)
      Mortality, n (%)12 (4.2)86 (9.5)50 (13.6)
      ICD, implantable cardioverter defibrillator.
      Delayed ICD implantation rates at 1 year after the cardiac arrest are outlined in Table 8. Among those who met our criteria for “likely ICD-eligible,” only 11 (10.6%) had a subsequent ICD implantation after discharge from the index hospitalization. The rate was lower in those who were “not likely ICD-eligible” (17 implants; 3.2%).
      Table 8Delayed ICD implantation rate 1 year after cardiac arrest
      Did not receive an ICD and “likely ICD-eligible” (n = 104)Did not receive an ICD and “not likely ICD-eligible” (n = 538)
      ICD implantation rate11 (10.6%)17 (3.2%)
      ICD, implantable cardioverter defibrillator.

      Discussion

      The results of this study reveal that despite guidelines published in 2005 that recommend ICD implantation in eligible patients after OHCA,
      • Tang A.S.
      • Ross H.
      • Simpson C.S.
      • et al.
      Canadian Cardiovascular Society/Canadian Heart Rhythm Society position paper on implantable cardioverter defibrillator use in Canada.
      the ICD implantation rate remains lower than might be expected, at approximately 25% for all post-OHCA patients who survive to hospital discharge and 57% of all “likely ICD-eligible” patients on the basis of a strict group definition. Although there was a statistically significant increase in the implantation rate over the 4 years of the study in the “likely ICD-eligible” group, there was also a significant increase in the implantation rate among those “not likely ICD-eligible.” Additionally, no significant change over time was found in the multivariate model for the entire cohort as well as the “likely ICD-eligible” group, suggesting that at least some of the increase in implantation rate was because of changes in baseline patient characteristics over time. This highlights that a long time lag between a clinical practice guideline change and real-world implementation might not by itself explain low secondary prevention ICD utilization. The Ontario Prehospital Advanced Life Support (OPALS) prehospital cardiac arrest study database for a similar patient population in Ontario, Canada between 1997 and 2002, recorded an overall ICD implantation rate of 12.8%
      • Parkash R.
      • Tang A.
      • Wells G.
      • et al.
      Use of implantable cardioverter defibrillators after out-of-hospital cardiac arrest: a prospective follow-up study.
      ; an administrative database study of the same time period (2002) by Birnie et al. showed an overall secondary prevention ICD implantation rate in Canada of 26.7%.
      • Birnie D.H.
      • Sambell C.
      • Johansen H.
      • et al.
      Use of implantable cardioverter defibrillators in Canadian and US survivors of out-of-hospital cardiac arrest.
      Because neither evidence of ischemia nor neurologic disability were included in these analyses, the proportion of the overall study cohorts that would represent “likely ICD-eligible” candidates were not reported. Similar to the results of this study, both previous studies reported that individuals with normal neurologic status and VT or VF as their initial rhythm were more likely to receive an ICD.
      • Parkash R.
      • Tang A.
      • Wells G.
      • et al.
      Use of implantable cardioverter defibrillators after out-of-hospital cardiac arrest: a prospective follow-up study.
      Other variables previously associated with increased odds of an ICD were younger age, history of heart failure, male sex, admission to a teaching hospital, and large hospital size.
      • Birnie D.H.
      • Sambell C.
      • Johansen H.
      • et al.
      Use of implantable cardioverter defibrillators in Canadian and US survivors of out-of-hospital cardiac arrest.
      Patients in our “likely ICD-eligible” group were included only if they met specific criteria that should accurately describe patients with an indication for a secondary prevention device. Despite this conservative definition, the ICD implantation rate was only 57%. This rate would not be expected to reach 100% because of additional patient factors, such as personal preference, underlying comorbidities, or clinical circumstances; however, one might expect ICD utilization in a selected population such as our “likely ICD-eligible” group to be higher.
      ICD implantation capability of the admission hospital was a significant predictor of ICD implantation in the multivariate model developed for the entire cohort and in the “likely ICD-eligible” patients alone. There is insufficient information in this database to determine why this association exists, but some possible explanations include the absence of a local electrophysiology consultation service, perceived difficulty in access or transfer to implantation facilities, or differences in the speciality of the most responsible physician at the admission hospitals. This finding might suggest that barriers to implementing ICD guideline recommendations might be more systemic because underutilization of primary prevention ICDs has also been shown. In one primary prevention study, the rate of referral for a device was higher when patients with a guideline indication for an ICD were assessed by an electrophysiologist (83%) compared with a general cardiologist (44%).
      • Sadarmin P.P.
      • Wong K.C.
      • Rajappan K.
      • Bashir Y.
      • Betts T.R.
      Barriers to patients eligible for screening investigations and insertion of primary prevention implantable cardioverter defibrillators.
      Similarly, barriers should be evaluated and addressed at multiple levels including the community, hospital, and individual practitioner level in the secondary prevention population as well.
      In the “not likely ICD-eligible” group, the implantation rate appeared to be appropriately low. We would not expect this rate to be 0% because there are additional factors not available in the database that might have represented an independent indication for ICD implantation, such as very poor left ventricular systolic function that was not expected to recover significantly, recurrent arrhythmias occurring during hospitalization, or ischemia that was not reversible because of a lack of revascularization options. Compared with a study by Al-Khatib et al., who reported that the guideline-discordant ICD implantation rates in cardiology patients with coronary artery disease or heart failure was > 20%, the degree of potentially inappropriate device utilization was lower in this study.
      • Al-Khatib S.M.
      • Hellkamp A.
      • Curtis J.
      • et al.
      Non-evidence-based ICD implantations in the United States.
      ICD eligibility could not be determined in 312 patients in this study cohort. All patients were assessed for “eligibility,” but some were classified as “eligibility unknown” because they did not meet the strict criteria defined in this study for “likely ICD-eligible.” Most patients (240 of 312 patients) had the combination of a nonshockable initial rhythm, no definite evidence of ischemia, and good neurologic status; most of the remainder had absent or insufficient documentation of the initial rhythm. We suspect some in this group without a documented initial shockable rhythm might have been ICD “eligible” (ie, had VT or VF as the cause of their cardiac arrest), but their true initial rhythm was not captured because of the time delay between the onset of the cardiac arrest and rhythm analysis by a first responder. If this group of patients were to be reclassified as “likely ICD-eligible,” the ICD implantation rate would be even lower and further support the suggestion of secondary prevention ICD underutilization. However, it is more conservative not to assume this group of patients were truly ICD-eligible because guidelines recommend an ICD only when VT or VF has been documented.
      In the one subgroup of patients with a documented STEMI on presenting ECG but no subsequent revascularization procedure, the implantation rate was approximately 48%. Possible explanations for this higher than expected rate include database limitations in the way a STEMI diagnosis was defined, or clinical variables such as late ventricular arrhythmias during the index hospitalization or significant ischemia with no revascularization option.
      The major limitation of this study was the data capture that did not include left ventricular ejection fraction or left ventricular systolic function, or a detailed chart evaluation of the rationale for or against an ICD implantation, including accurate data on underlying medical comorbidities and recurrent arrhythmia after hospitalization. There were missing data in the initial rhythm field for 109 (8.8%) because the first responding vehicle was not participating in the Rescu Epistry, undetermined rhythm for 3 (0.2%), and no documented rhythm for 9 (0.7%) cases. STEMI data were missing for 90 (7.3%) patients and ICD status was missing for 11 (0.9%) patients in the study cohort. Additionally, although angiography rates were recorded in this database, the lack of data regarding alternative cardiac testing to assess for ischemia is a limitation of this study. Last, revascularization was included in how evidence of ischemia was defined. Although this characteristic might suggest a reversible cause, one cannot be certain that ischemia was believed to be the cause of the cardiac arrest for that patient. There are instances of ischemia causing a cardiac arrest in the absence of a STEMI,
      • Kuck K.H.
      • Cappato R.
      • Siebels J.
      • Rüppel R.
      Randomized comparison of antiarrhythmic drug therapy with implantable defibrillators in patients resuscitated from cardiac arrest: the Cardiac Arrest Study Hamburg (CASH).
      therefore, these patients were included in the “not likely ICD-eligible” group to maintain a more strictly defined “likely ICD-eligible” group and avoid underestimation of appropriate secondary prevention ICD implantation.
      In conclusion, secondary prevention ICD utilization after OHCA with survival to discharge in this study cohort was frequently not guideline-compliant. Further studies, including quality improvement initiatives such as root cause analysis, will help understand the local factors that contribute to ICD underutilization and potential strategies to improve adherence to guideline recommendations for secondary prevention ICD implantation after OHCA.

      Acknowledgements

      The authors thank the Rescu Epistry investigators and all emergency medical service operators, providers, and medical directors as well as the in-hospital staff in the Strategies for Post Arrest Resuscitation Care network hospitals working together in the front line of emergency patient care for their continued commitment contributions to high-quality care and primary data collection in resuscitation research at Rescu, Li Ka Shing Knowledge Institute, St Michael's Hospital, Toronto Ontario, Canada. We also thank Kim Dawdy, Cathy Chun, and everyone at the Institute for Clinical Evaluative Sciences for their contributions to this study.

      Funding Sources

      Dr Dorian is a CaNET investigator and is supported by the Cardiac Arrhythmia Network of Canada. The Resuscitation Outcomes Consortium Epistry study is supported by a cooperative agreement (5U01 HL077863) with the National Heart, Lung, and Blood Institute in partnership with the National Institute of Neurological Disorders and Stroke, Canadian Institutes of Health Research–Institute of Circulatory and Respiratory Health, Defense Research and Development Canada, Heart and Stroke Foundation of Canada, and American Heart Association. Rescu Epistry is funded by a centre grant from the Laerdal Foundation, and knowledge translation collaborative grants and operating grants from the Canadian Institutes of Health Research and the Heart and Stroke Foundation of Canada.

      Disclosures

      Dr Cheskes has received funding for educational speaking in the area of CPR quality from Zoll Medical and Physio-Control. The other authors have no conflicts of interest to disclose.

      Supplementary Material

      References

        • Nichol G.
        • Thomas E.
        • Callaway C.W.
        • et al.
        Regional variation in out-of-hospital cardiac arrest incidence and outcome.
        JAMA. 2008; 300: 1423-1431
        • Kudenchuk P.J.
        • Brown S.P.
        • Daya M.
        • et al.
        Amiodarone, lidocaine, or placebo in out-of-hospital cardiac arrest.
        N Engl J Med. 2016; 374: 1711-1722
        • Tang A.S.
        • Ross H.
        • Simpson C.S.
        • et al.
        Canadian Cardiovascular Society/Canadian Heart Rhythm Society position paper on implantable cardioverter defibrillator use in Canada.
        Can J Cardiol. 2005; 21: 11A-18A
        • Bennett M.
        • Parkash R.
        • Nery P.
        • et al.
        Canadian Cardiovascular Society/Canadian Heart Rhythm Society 2016 implantable cardioverter-defibrillator guidelines.
        Can J Cardiol. 2017; 33: 174-188
        • Epstein A.
        • DiMarco J.
        • Ellenbogen K.
        • et al.
        ACC/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the ACC/AHA/NASPE 2002 Guideline).
        Circulation. 2008; 117: e350-e408
        • Zipes D.P.
        • Camm A.J.
        • Borggrefe M.
        • et al.
        ACC/AHA/ESC 2006 guidelines for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death: a report of the American College of Cardiology/American Heart Association Task Force and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Develop guidelines for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death) developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society.
        Europace. 2006; 8: 746-837
        • Wyse D.G.
        • Friedman P.L.
        • Epstein A.E.
        A comparison of antiarrhythmic-drug therapy with implantable defibrillators in patients resuscitated from near-fatal ventricular arrhythmias. The Antiarrhythmics versus Implantable Defibrillators (AVID) Investigators.
        N Engl J Med. 1997; 337: 1576-1583
        • Connolly S.J.
        • Gent M.
        • Roberts R.S.
        • et al.
        Canadian implantable defibrillator study (CIDS): a randomized trial of the implantable cardioverter defibrillator against amiodarone.
        Circulation. 2000; 101: 1297-1302
        • Kuck K.H.
        • Cappato R.
        • Siebels J.
        • Rüppel R.
        Randomized comparison of antiarrhythmic drug therapy with implantable defibrillators in patients resuscitated from cardiac arrest: the Cardiac Arrest Study Hamburg (CASH).
        Circulation. 2000; 102: 748-754
        • Lee D.S.
        • Green L.D.
        • Liu P.P.
        • et al.
        Effectiveness of implantable defibrillators for preventing arrhythmic events and death: a meta-analysis.
        J Am Coll Cardiol. 2003; 41: 1573-1582
        • Connolly S.J.
        • Hallstrom A.P.
        • Cappato R.
        • et al.
        Meta-analysis of the implantable cardioverter defibrillator secondary prevention trials. AVID, CASH and CIDS studies. Antiarrhythmics vs Implantable Defibrillator study. Cardiac Arrest Study Hamburg. Canadian Implantable Defibrillator Study.
        Eur Heart J. 2000; 21: 2071-2078
        • Birnie D.H.
        • Sambell C.
        • Johansen H.
        • et al.
        Use of implantable cardioverter defibrillators in Canadian and US survivors of out-of-hospital cardiac arrest.
        CMAJ. 2007; 177: 41-46
        • Parkash R.
        • Tang A.
        • Wells G.
        • et al.
        Use of implantable cardioverter defibrillators after out-of-hospital cardiac arrest: a prospective follow-up study.
        CMAJ. 2004; 171: 1053-1056
        • Voigt A.
        • Ezzeddine R.
        • Barrington W.
        • et al.
        Utilization of implantable cardioverter-defibrillators in survivors of cardiac arrest in the United States from 1996 to 2001.
        J Am Coll Cardiol. 2004; 44: 855-858
        • Saba S.
        • Ravipati L.P.
        • Voigt A.
        Recent trends in utilization of implantable cardioverter-defibrillators in survivors of cardiac arrest in the united states.
        Pacing Clin Electrophysiol. 2009; 32: 1444-1449
        • Morrison L.J.
        • Nichol G.
        • Rea T.D.
        • et al.
        Rationale, development and implementation of the Resuscitation Outcomes Consortium Epistry-Cardiac Arrest.
        Resuscitation. 2008; 78: 161-169
        • Lin S.
        • Morrison L.J.
        • Brooks S.C.
        Development of a data dictionary for the Strategies for Post Arrest Resuscitation Care (SPARC) network for post cardiac arrest research.
        Resuscitation. 2011; 82: 419-422
        • Sadarmin P.P.
        • Wong K.C.
        • Rajappan K.
        • Bashir Y.
        • Betts T.R.
        Barriers to patients eligible for screening investigations and insertion of primary prevention implantable cardioverter defibrillators.
        Europace. 2014; 16: 1575-1579
        • Al-Khatib S.M.
        • Hellkamp A.
        • Curtis J.
        • et al.
        Non-evidence-based ICD implantations in the United States.
        JAMA. 2011; 305: 43-49