International Journal of Cardiology 182 (2015) 440–446
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Defibrillator patients should not be denied a peaceful death Annika Kinch Westerdahl a,⁎,1, Richard Sutton b,1, Viveka Frykman a a b
Department of Clinical Sciences, Karolinska Institutet, Danderyds Hospital, Stockholm, Sweden National Heart & Lung Institute, Imperial College, St Mary's Campus, London, UK
a r t i c l e
i n f o
Article history: Received 16 December 2014 Accepted 2 January 2015 Available online 6 January 2015 Keywords: Deactivation End of life Do-not-resuscitate Implantable cardioverter defibrillator Shock therapy
a b s t r a c t Background: Implantable defibrillators (ICDs) prevent sudden cardiac death. With declining health, ICD therapy may prolong death and expose the patient to unnecessary pain and anxiety. Few studies have addressed end of life care in ICD patients. The objective of this study was to investigate end of life in ICD patients, with respect to location of death; duration between do-not-resuscitate (DNR)-orders and deactivation of ICD therapy or DNR and time of death. Methods and results: A descriptive analysis of 65 deceased ICD patients, all whom had a written DNR-order before death, is presented. The majority (86%) was treated in hospitals, mainly (63%) university hospitals, and many (33%) in cardiology wards. Despite DNR-order, ICD shock therapy was active in 51% of all patients. In those with therapy deactivated at death, therapy deactivation was carried out two days or more after DNR-order in more than a third (38%). The time from DNR decision to death in patients with therapy active had a median of four days (IQR 1–38). During the last 24 h of life, 24% of the patients experienced shock treatment. Conclusions: The majority of ICD patients with a DNR-order were treated in university hospitals. More than half still had shock treatment active at time of death with a median of four days or more between DNR decision and death. Patients with therapy deactivated, two days or more elapsed in more than a third from DNR decision to deactivation of therapy, exposing patients to a high risk of painful shocks before death. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The implantable cardioverter defibrillator (ICD) is highly effective in saving lives [1–5]. Eventually patients may develop terminal illness and a decision of retaining ICD therapy active or not is paramount. The exact number of active implants in the world is unknown, but recent surveys showed over 130 000 active implants/year in the USA and 70–80 000 in Europe [6,7]. Faced with this increase in ICD treatment, end of life related question of discontinuation of ICD treatment is inevitable. When a patient's health status deteriorates, a decision to issue a donot-resuscitate (DNR)-order can be taken. This does not equal ICD deactivation, but for patients who desire comfort care and consent to deactivation, it is not only appropriate, but also ethical and legal to withdraw defibrillator shock therapy [8,9]. Few studies have addressed patients with ICDs at end of life. One revealed an increased risk of shock therapy in the last week [10]. We have previously shown that despite a DNRorder, almost two thirds of ICD patients still had therapy active and a fifth had shocks before death [11]. However, little is known about ⁎ Corresponding author at: Department of Cardiology, Danderyds Hospital, 182 88 Stockholm, Sweden. E-mail address:
[email protected] (A.K. Westerdahl). 1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
http://dx.doi.org/10.1016/j.ijcard.2015.01.012 0167-5273/© 2015 Elsevier Ireland Ltd. All rights reserved.
where these patients die, if different specialities handle dying ICD patients differently and if there is enough time to deactivate therapy in this population. How much does ICD competence influence therapy deactivation? The aim of this study was to investigate end of life in ICD patients, with respect to location of death; duration between DNR-order and therapy deactivation; or DNR and time of death. 2. Patients and methods 2.1. Study group and device collection This is a descriptive analysis of 65 deceased ICD patients all of whom had a written DNR-order before death (Fig. 1). The study population was also included in an earlier study [11]. This is a nationwide study with 26 participating hospitals, which prospectively enrolled 130 ICD devices from deceased patients. In Sweden it is mandatory to explant devices (pacemaker or ICD) after death. The pathology departments of all participating centres were asked and agreed to take part in the present study before device enrolment. All ICDs were sent to the study centre. Of 130 patients, all those (65) with a written DNR-order or in palliative care were included and analysed in this study. The study population's demographic data were obtained from patients' medical records, the Swedish ICD and Pacemaker Register, and
A.K. Westerdahl et al. / International Journal of Cardiology 182 (2015) 440–446
441
Fig. 1. Study flowchart (n = 65). ICD = implantable defibrillator. *One patient's location of death is missing. †Medicine ward including oncology (1). §Medicine ward including infection (1), intensive care unit (2). ‡Medicine ward including intensive care unit (1). $Medicine ward including oncology (1).
the National Board of Health and Welfare. Death certificates were obtained from the Swedish Tax Agency.
2.5. Active therapy All the ICDs were interrogated and data regarding deactivation, incidence of shock treatment for 24 h prior to death were retrieved.
2.2. Time and location of death Location of death, type of hospital, university or non-university, as well as ward specialty, where the patient was receiving care when death occurred, were identified. The definition of a university hospital was an affiliation with a university and provision of clinical education of medical students, in addition to delivering patient care. Type of specialty was categorized as cardiology, medicine (including infection, oncology and intensive care unit), geriatrics, hospice (including nursing home) and home. The time of death for patients dying in a hospital was established from medical and nursing notes and from death certificates, which stated the time, cause, and location of death. For patients dying at home, the time of death was assessed from the death certificate and, if available, notes from the palliative care team. If death was witnessed, the time of death was accepted as correct.
2.3. Do-not-resuscitate The time from a decision to issue a DNR-order and therapy deactivation as well as time to death were analysed.
2.4. Deactivation Therapy deactivation can be performed either permanently by programming the ICD shock therapy off or temporarily by placing a specific magnet over the device. As long as the magnet remains in place, therapies are off. Date of deactivation was collected from the medical notes and in some devices the date and time of deactivation, also with magnet placement, was available at post-mortem interrogation.
2.6. Shock treatment The incidence and number of shocks delivered were recorded. Shock treatment within 24 h of death was classified as shock at end of life. In addition to the occurrence of any shocks, we assessed the number of shocks received during the last 24 h of life.
2.7. Death classification Death was classified using a modified version of the classification scheme of Epstein et al. [12]. Two investigators reviewed all deaths. The primary causes were divided into cardiac and non-cardiac deaths.
2.8. Statistical analysis Data was summarized with descriptive statistics. Continuous variables were presented as mean and standard deviations or as median and interquartile range whereas categorical variables were presented as percentages or as confidence intervals of proportions [13]. The normal approximation to the binomial distribution was used. This is valid because if np = 10 and n (1 − p) = 10, then p can be assumed to have an approximately normal distribution, where n is the sample size and p is the proportion [14]. Statistical analysis was performed using SPSS software version 22.
2.9. Ethics The study was approved by the Regional Ethics Committee (2008/1527-31/4).
442
A.K. Westerdahl et al. / International Journal of Cardiology 182 (2015) 440–446
3. Results 3.1. Study group All patients died in 2003–2010 and mean duration of device implant was 2.8 ± 2.2 years. Secondary prevention was the major indication (78%) for the initial ICD implantation. Baseline characteristics are presented in Table 1. 3.2. Location of death At death the great majority, 86%, died in a hospital, and mainly in university hospitals (Fig. 1). All but 2 patients had access to healthcare assistance during the last days of life. For one patient, who had been deactivated and discharged home a week before, the location of death was unavailable.
Fig. 2. Time from DNR-order to deactivation of ICD therapy (n = 32). *One ICD deactivated 2 days prior DNR.
3.3. Do not resuscitate All patients in this study had a DNR order. Deactivation was performed in 49%, most of them within 24 h from DNR-order, but for more than a third (38%) two days or more passed between decision of DNR and deactivation (Fig. 2). In 3 patients a family member had to call for attention before deactivation was carried out. The time between decision of DNR and patients' death had a median of 5 days (IQR 1–21), nearly half (46%) of the patients died a week or more after DNR decision.
3.6. Death classification
3.4. Active therapy
4. Discussion
In 51%, shock therapies were still active at time of death with the majority (70%) being treated in a hospital (Table 2). Therapy and different locations are shown in Table 3.
The most important finding in this study is that the majority of ICD patients, terminally ill and with a DNR-order, although in a hospital with access to high level care, ICD shock therapy remained active. A DNR-order is not equal to ICD deactivation. Patients may wish for active ICD therapy even though terminally ill. Patients' autonomy must be respected and therapy deactivation should be made only after careful consideration of deactivation consequences. The patient's involvement in the DNR-process, including wish for active treatment or not, was not clearly stated for most cases in this study. However, many decisions regarding therapy are made by physicians, without involvement of patients or family [15]. Some patients favour deactivation [16] and some
3.5. Shock treatment During the last day many patients received shock treatment. Ward specialty and number of shocks are presented in Table 4. The probability of experiencing shock treatment 24 h before death was 24% (95% CI, 11%–37%). Median duration from DNR decision to death was 3 days (IQR 1–21).
Cardiac deaths were the major cause and heart failure the most common specific cause. Ten patients received shock treatment; only one with arrhythmia as the primary cause of death and who received 6 appropriate shocks for the last 24 h (Table 5). ICD therapy was still active in 47% of all cardiac deaths and in 55% of all non-cardiac deaths.
Table 1 Baseline characteristics (n = 65). All n = 65
Medicinea n = 25
Cardiology n = 16
Geriatrics n=4
Surgery n=3
Hospice (4) Nursing home (4) n=8
Home (8) Not known (1) n=9
Male Age at death (years)
59 (91) 74.5 ± 8.3
21 (84) 77.5 ± 10.3
15 (94) 74.3 ± 6.1
4 (100) 74.5 ± 6.7
3 (100) 79.0 ± 7.8
7 (88) 74.4 ± 8.9
9 (100) 73.0 ± 6.5
Morbidityb Hypertension Ischaemic heart disease Acute myocardial infarction Heart failure Diabetes mellitus Chronic pulmonary disease Chronic kidney disease Malignancy
14 (22) 49 (75) 48 (74) 62 (95) 19 (29) 13 (20) 20 (31) 15 (23)
6 (24) 19 (76) 18 (72) 22 (88) 7 (28) 8 (32) 10 (40) 8 (32)
3 (19) 14 (88) 12 (75) 16 (100) 6 (38) 1 (6) 6 (38) 1 (6)
1 (25) 4 (100) 3 (75) 4 (100) 2 (50) 1 (25) 1 (25) 2 (50)
1 (33) 2 (67) 2 (67) 3 (100) 3 (100) 1 (33) 1 (33) 1 (33)
1 (13) 5 (63) 6 (75) 8 (100) 3 (38) 0 2 (25) 1 (13)
2 (22) 5 (56) 7 (78) 9 (100) 1 (11) 2 (22) 9 (100) 2 (22)
Ejection fraction EF ≥55% EF 45–54% EF 30–44% EF b30% Not known
4 (6) 4 (6) 20 (31) 36 (55) 1 (2)
2 (8) 4 (16) 9 (36) 10 (40) 0
0 0 3 (19) 13 (81) 0
0 0 1 (25) 3 (75) 0
1 (33) 0 2 (67) 0 0
1 (13) 0 1 (13) 5 (63) 1 (13)
0 0 4 (44) 5 (56) 0
Values are listed as values n (%). a Medicine ward including infection (1), oncology (2), and intensive care unit (3). b Based on International Classification of Diseases in the medical notes. Not cause of death.
A.K. Westerdahl et al. / International Journal of Cardiology 182 (2015) 440–446
443
Table 2 Time from do-not-resuscitate order to death and location of death for all patients with ICD shocks active at 1 h (n = 33).
Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Patient 7 Patient 8 Patient 9 Patient 10 Patient 11 Patient 12 Patient 13 Patient 14 Patient 15 Patient 16 Patient 17 Patient 18 Patient 19 Patient 20 Patient 21 Patient 22 Patient 23 Patient 24 Patient 25 Patient 26 Patient 27 Patient 28 Patient 29 Patient 30 Patient 31 Patient 32 Patient 33
Location of death
Ward specialty
Time from DNR to death
University hospital University hospital Non-university hospital Hospice University hospital University hospital University hospital Non-university hospital University hospital Nursing home Home (palliative care team) University hospital University hospital Home (palliative care team) Home (palliative care team) University hospital Non-university hospital University hospital Non-university hospital Non-university hospital Non-university hospital Home (palliative care team) Home (palliative care team) Non-university hospital Hospice Non-university hospital Nursing home University hospital University hospital University hospital University hospital Nursing home University hospital
Cardiology Cardiology Geriatrics
1 day 2 days 1 day 4 days 90 days 3 days 9 days 1 day b1 day 38 days 92 days 4 days 14 days 107 days 10 days 3 days 672 days 1 day 1 day Not known 1 day 214 days 104 days 1 day 21 days 1 day 131 days 1 day 1 day b1 day 5 days Not known 8 days
do not [17,18], while many have insufficient knowledge of ICD function rendering them unable to make informed decisions about ICD deactivation [19,20].
Geriatrics Cardiology Medicine Medicine Cardiology
Medicine Cardiology
Surgical Medicine Medicine Medicine/ICU Medicine Medicine/ICU
Cardiology Surgical Medicine Cardiology Medicine Medicine Cardiology
indication for ICD treatment in Europe at this time. Primary prevention was not given a high indication class until 2006 [22]. 4.2. Location of death
4.1. Study group In this study of 65 patients with a DNR order the deactivation process as well as patient location at death was scrutinized. The great majority of the patients were in a hospital or other care facility when they died, confirming the results from the SCD-HeFT-study, where almost two-thirds of ICD patients died in a hospital [21]. All but 2 patients in this study had access to health care professionals, giving almost all patients' physicians the possibility to consider deactivation. Secondary prevention was the indication for ICD treatment in the majority (78%) of the study patients. Device implantations were performed between 1998 and 2010, but many received their first ICD even earlier than this. Secondary prevention was the most common
Being admitted to a university hospital did not affect therapy deactivation per se; more than half of the patients having therapy still active at death were inpatients in a university hospital. Cardiology departments were the most common ward speciality. These findings are striking that even DNR patients under university cardiological care, with high cardiac competence, still have active devices when they die. The persistent availability of ICD shocks is obvious, the patient is close at hand, but still cardiologists fail to perform deactivation in dying patients. 4.3. Do not resuscitate ICD patients with a DNR order left with therapy active are at risk of unnecessary shock therapy for several days. Despite a decision to refrain
Table 3 ICD therapy programmed ON or OFF in different locations (n = 65). ICD therapy
All n = 65
Medicinea n = 25
Cardiology n = 16
Geriatrics n=4
Surgical n=3
Hospice (4) Nursing home (4) n=8
Home (8) Not known (1) n=9
ICD ON 24 h ICD ON 1 h ICD OFF 1 h Magnet Shock b
42 (65) 33 (51) 28 (43) 4 (6) 10 (24)
15 (60) 10 (40) 13 (52) 2 (8) 5 (33)
12 (75) 9 (56) 5 (32) 2 (13) 2 (13)
2 (50) 2 (50) 2 (50) 0 1 (25)
3 (100) 2 (67) 1 (33) 0 1 (33)
5 (63) 5 (63) 3 (38) 0 1 (13)
5 (56) 5 (56) 4 (44) 0 0c
Values are listed as values n (%). a Medicine ward including infection (1), oncology (2), intensive care unit (3). b Shock during the last 24 h for patients with ICD ON 24 h. c 1 aborted shock.
444
A.K. Westerdahl et al. / International Journal of Cardiology 182 (2015) 440–446
Table 4 Hospital, ward specialty and number of shocks (n = 10).
Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Patient 7 Patient 8 Patient 9 Patient 10
University hospital
Ward specialty
Time from DNR to death
Number of shocks
Yes Yes Yes No Yes Yes No Yes No No
Cardiology Geriatrics Cardiology Nursing home Medicine (oncology) Medicine Medicine (oncology) Medicine (neurology) Medicine (ICU) Surgery
2 days 90 days b24 h 38 days 4 days 4 days 15 days 1 day 1 day b24 h
2 1 6 1 2 6 1 42 18 1
from active resuscitation, such as external defibrillation, internal defibrillation remains available. If this is the patient's wish it should be respected. In many patients the DNR decision was taken well before death. In one patient, with a DNR decision due to a cognitive disorder, time between DNR and death was 1.5 years. The ICD was still active and the record did not reveal any information regarding the patient's wish to retain ICD activity. Despite terminal illness only half of our patients had ICD therapy deactivated before death. For many, deactivation was performed 2 days or more after DNR decision, for some the duration until therapy deactivation was as long as 7 days or more, exposing the patients to risk of shock therapy during this time. Physicians have a reluctance to discuss ICD deactivation with patients [23]. The reasons are complex and include personal discomfort, lack of experience and ethical as well as legal issues [24,25]. Furthermore, studies have also shown that physicians are less comfortable with cessation of ICD therapy compared with other life sustaining treatment [26]. Expert consensus statements have been published, in order to improve end of life care for the ICD patient and to highlight both the importance of discussing deactivation with the patient as well as stating the appropriateness of therapy deactivation at end of life [8, 15]. With a steadily increasing patient population, ICD management at
Table 5 Mechanisms of death and shock treatment. Cause of death
4.4. Active therapy Total n = 65
Shock treatment n = 10
I. Primary organ cause A. Cardiac 1. Arrhythmia 2. Heart failure 3. Ischaemic heart disease 4. Other (valvular) 5. Unknown B. Non-cardiac 1. Stroke/cerebrovascular 2. Other vascular a 3. Malignancy 4. Infection 5. Chronic obstructive pulmonary disease 6. Alzheimer 7. Haemorrhage 8. Trauma 9. Other C. Unknown
34 (52) 3 (5) 27 (42) 1 (2) 3 (5) 0 31 (48) 4 (6) 0 15 (23) 8 (12) 3 (5) 0 0 0 1 (2) 0
4 (40) 1 (10) 3 (30) 0 0 0 6 (60) 2 (20) 0 2 (20) 1 (10) 1 (10) 0 0 0 0 0
II. Documentation A. Witnessed 1. Yes 2. No 3. Unknown B. Monitoredb
44 (68) 10 (15) 11 (17) 44 (68)
8 (80) 1 (10) 1 (10) 10 (100)
Values are listed as values n (%). a i.e. embolism, aneurysm. b Detection ON or telemetry at last 24 h.
end of life is something cardiologists and, indeed most clinicians must be able to handle. Even in a hospice facility, experts on palliation, therapy deactivation is not automatically requested when patients have a DNR order. Our results, in relatively few patients, support those of Goldstein et al., where less than half of all ICD patients in hospices had ICD deactivation [27]. Studies have shown that next of kin often have to initiate deactivation [10,28]. In this study it was a family member who asked for deactivation in five cases. In one, the patient's daughter had repeatedly asked for deactivation over 3 days. Therapy was finally deactivated 1 h before the patient died after 6 shocks in the previous 24 h. Permanent deactivation with a programmer is the preferable method of deactivation, but application of a clinical magnet can be used [15]. In this study deactivation with a magnet was used in 4 patients, with a time span ranging from 2 h to 48 h. Magnet application for a short period of time, in the absence of someone capable of programming the ICD, is understandable, but for 3 of 4 patients the time span was more than 12 h, which should be sufficient to arrange for programmer deactivation. Also, we have to consider that ICD behaviour with magnet application can vary. One manufacturer's ICD therapy will automatically resume activity after 8 h of magnet application, even when it remains in place [8]. Collaboration between manufacturers to produce a universal and “smarter” clinical magnet is desirable but this has been raised earlier without decision [29].
In half of the patients with therapy still active, 4 days or more passed between DNR and death and in a third it was more than a week. Ideally, discussion concerning programming of therapy between patient and physician should occur before terminal illness [9], but in the majority no such discussion happens [30]. One previous study has shown difficulties over deactivation due to rapid decline in patients' clinical status [31], but the results from our study suggest the contrary. Time for expert consultation, patient discussion and deactivation should have been more than enough. 4.5. Shock treatment Prior studies have shown that ICD patients risk shocks at end of life and that tachyarrhythmias are frequent at death in this population [10,11]. In this study all patients had an active decision for palliative care, despite this therapy remained active and 10 patients received shocks in the last 24 h. All patients, except one, were in a hospital with access to ICD competence. One patient admitted to a cardiology ward, terrified of shocks after earlier experience, received 5 during the last 24 h. The patient became extremely distressed and called for immediate deactivation, but received an additional shock before a magnet was applied, 1 h before death. Most patients find shocks uncomfortable and distressing [32], emphasizing that the question of deactivation is paramount.
A.K. Westerdahl et al. / International Journal of Cardiology 182 (2015) 440–446
The increasing number of ICD patients will make it difficult for device specialists to follow each patient closely, shifting responsibility for end of life questions to primary and palliative care physicians. Our result implies that patients die in a hospital. Interestingly, competence seems not to influence therapy deactivation per se. Deactivation was carried out in 60% of patients treated in a medical ward and only in 28% in patients treated under cardiology. Every patient has the right to the best possible care and it is necessary to increase healthcare professionals' knowledge about implantable devices. Also, communication between specialists must improve. Risk of unnecessary shock therapy due to its remaining active in dying ICD patients must be avoided. When external defibrillation is denied on the basis of the DNR decision, the same should apply to internal defibrillation. 5. Limitations The small number of patients included in this study offers some limitation to widespread adoption of its conclusions despite, to our knowledge, this being the largest study so far on end of life care in ICD patients. Due to its descriptive nature, generalizations may not apply to all ICD patients. It should be noted that the majority of patients reported here received their care before current position statements were published. The timing of the DNR decision was only available by date making the time from decision to death imprecise. If the DNR-order was taken on the same date as death the duration was classified as within 24 h. If the date differed from that of death the duration was 24 h or more. 6. Conclusions The majority of ICD patients with a DNR-order were treated in university hospitals. More than half still had shock treatment active at time of death with a median of 4 days or more between DNR decision and death. Patients with therapy deactivated, 2 days or more elapsed in more than a third from DNR decision to deactivation of therapy, exposing patients to a high risk of painful shocks before death. Funding sources This work was supported by Laerdal Foundation (7500 EUR), Ragnhild and Einar Lundströms Remembrance Foundation (3300 EUR) and Fond 176 Danderyds Hospital (3300 EUR). The researchers were independent from the funders. The funders had no role in conducting the study, writing the paper or the decision to submit the paper for publication. Disclosures Ms Kinch has received lecture-fees from Biotronik, St. Jude Medical and Medtronic and research-grants from Medtronic and Boston Scientific. Prof Sutton has received consultancy fees from Medtronic Inc., research grants from Medtronic Inc., Speakers' Bureau for Medtronic Inc. and St Jude Medical Inc., and is a stockholder in Boston Scientific Inc. and Advanced Circulatory Systems Inc. Dr Frykman has received lecture-fees from AstraZeneca, consultant-fees from Boehringer Ingelheim, and research-grants from Guidant (now Boston Scientific Inc.); conducted studies in collaboration with Cyncron, AstraZeneca, Sanofi Aventis, Guidant, Medtronic and St Jude Medical. Conflict of interest Dr. Frykman reports personal fees from Astra Zeneca, personal fees from Boehringer Ingelheim, and grants from Guidant, during the conduct of the study. Dr. Sutton reports grants and personal fees from Medtronic Inc, and personal fees from St Jude Medical, outside the submitted work; and Stockholder in Boston Scientific Inc. and Advanced Circulatory Systems Inc. Ms. Kinch Westerdahl reports grants from
445
Laerdal Foundation, grants from Ragnhild and Einar Lundströms Remembrance Foundation, personal fees from Medtronic, St Jude Medical & Biotronik, grants from Boston Scientific, and grants from Medtronic, during the conduct of the study. Acknowledgement We thank Jackline Magnsjö, RN, and Dr Johanna Sjöblom, Danderyds Hospital, for their valuable contribution to the study. References [1] Causes of death in the Antiarrhythmics Versus Implantable Defibrillators (AVID) trial, J. Am. Coll. Cardiol. 34 (1999) 1552–1559. [2] G.H. Bardy, K.L. Lee, D.B. Mark, J.E. Poole, D.L. Packer, R. Boineau, et al., Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure, N. Engl. J. Med. 352 (2005) 225–237. [3] A.E. Buxton, K.L. Lee, J.D. Fisher, M.E. Josephson, E.N. Prystowsky, G. Hafley, A randomized study of the prevention of sudden death in patients with coronary artery disease. Multicenter Unsustained Tachycardia Trial Investigators, N. Engl. J. Med. 341 (1999) 1882–1890. [4] S.J. Connolly, A.P. Hallstrom, R. Cappato, E.B. Schron, K.H. Kuck, D.P. Zipes, 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. 21 (2000) 2071–2078. [5] A.J. Moss, W. Zareba, W.J. Hall, H. Klein, D.J. Wilber, D.S. Cannom, et al., Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction, N. Engl. J. Med. 346 (2002) 877–883. [6] A. John Camm, S. Nisam, European utilization of the implantable defibrillator: has 10 years changed the ‘enigma’? Europace 12 (2010) 1063–1069. [7] H.G. Mond, A. Proclemer, The 11th world survey of cardiac pacing and implantable cardioverter-defibrillators: calendar year 2009—a World Society of Arrhythmia's project, Pacing Clin. Electrophysiol. 34 (2011) 1013–1027. [8] R. Lampert, D.L. Hayes, G.J. Annas, M.A. Farley, N.E. Goldstein, R.M. Hamilton, et al., HRS Expert Consensus Statement on the Management of Cardiovascular Implantable Electronic Devices (CIEDs) in patients nearing end of life or requesting withdrawal of therapy, Heart Rhythm. 7 (2010) 1008–1026. [9] B.L. Wilkoff, A. Auricchio, J. Brugada, M. Cowie, K.A. Ellenbogen, A.M. Gillis, et al., HRS/EHRA Expert Consensus on the Monitoring of Cardiovascular Implantable Electronic Devices (CIEDs): description of techniques, indications, personnel, frequency and ethical considerations: developed in partnership with the Heart Rhythm Society (HRS) and the European Heart Rhythm Association (EHRA); and in collaboration with the American College of Cardiology (ACC), the American Heart Association (AHA), the European Society of Cardiology (ESC), the Heart Failure Association of ESC (HFA), and the Heart Failure Society of America (HFSA). Endorsed by the Heart Rhythm Society, the European Heart Rhythm Association (a registered branch of the ESC), the American College of Cardiology, the American Heart Association, Europace 10 (2008) 707–725. [10] S. Sherazi, S. McNitt, M.K. Aktas, B. Polonsky, A.H. Shah, A.J. Moss, et al., End-of-life care in patients with implantable cardioverter defibrillators: a MADIT-II substudy, Pacing Clin. Electrophysiol. 36 (2013) 1273–1279. [11] A. Kinch Westerdahl, J. Sjoblom, A.C. Mattiasson, M. Rosenqvist, V. Frykman, Implantable cardioverter-defibrillator therapy before death: high risk for painful shocks at end of life, Circulation 129 (2014) 422–429. [12] A.E. Epstein, M.D. Carlson, R.N. Fogoros, S.L. Higgins, F.J. Venditti Jr., Classification of death in antiarrhythmia trials, J. Am. Coll. Cardiol. 27 (1996) 433–442. [13] B.R. Kirkwood, J.A.C. Sterne, B.R. Kirkwood, Essential Medical Statistics, 2nd ed. Blackwell Science, Malden, Mass, 2003. [14] R. Conroy, Sample Size. A Rough Guide, in: Committee TBHEMR (Ed.), 2006. [15] L. Padeletti, D.O. Arnar, L. Boncinelli, J. Brachman, J.A. Camm, J.C. Daubert, et al., EHRA Expert Consensus Statement on the management of cardiovascular implantable electronic devices in patients nearing end of life or requesting withdrawal of therapy, Europace 12 (2010) 1480–1489. [16] S.S. Pedersen, R. Chaitsing, T. Szili-Torok, L. Jordaens, D.A. Theuns, Patients' perspective on deactivation of the implantable cardioverter-defibrillator near the end of life, Am. J. Cardiol. 111 (2013) 1443–1447. [17] C.E. Raphael, M. Koa-Wing, N. Stain, I. Wright, D.P. Francis, P. Kanagaratnam, Implantable cardioverter-defibrillator recipient attitudes towards device deactivation: how much do patients want to know? Pacing Clin. Electrophysiol. 34 (2011) 1628–1633. [18] G.C. Stewart, J.R. Weintraub, P.P. Pratibhu, M.J. Semigran, J.M. Camuso, K. Brooks, et al., Patient expectations from implantable defibrillators to prevent death in heart failure, J. Card. Fail. 16 (2010) 106–113. [19] A. Stromberg, C. Fluur, J. Miller, M.L. Chung, D.K. Moser, I. Thylen, ICD recipients' understanding of ethical issues, ICD function, and practical consequences of withdrawing the ICD in the end-of-life, Pacing Clin. Electrophysiol. 37 (2014) 834–842. [20] N.E. Goldstein, D. Mehta, S. Siddiqui, E. Teitelbaum, J. Zeidman, M. Singson, et al., “That's like an act of suicide” patients' attitudes toward deactivation of implantable defibrillators, J. Gen. Intern. Med. 23 (Suppl. 1) (2008) 7–12. [21] B. Olshansky, F. Wood, A.S. Hellkamp, J.E. Poole, J. Anderson, G.W. Johnson, et al., Where patients with mild to moderate heart failure die: results from the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT), Am. Heart J. 153 (2007) 1089–1094.
446
A.K. Westerdahl et al. / International Journal of Cardiology 182 (2015) 440–446
[22] D.P. Zipes, A.J. Camm, M. Borggrefe, A.E. Buxton, B. Chaitman, M. Fromer, 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, Circulation 114 (2006) e385–e484. [23] A.S. Kelley, M.C. Reid, D.H. Miller, J.J. Fins, M.S. Lachs, Implantable cardioverterdefibrillator deactivation at the end of life: a physician survey, Am. Heart J. 157 (2009) (702-8 e1). [24] N.E. Goldstein, D. Mehta, E. Teitelbaum, E.H. Bradley, R.S. Morrison, “It's like crossing a bridge” complexities preventing physicians from discussing deactivation of implantable defibrillators at the end of life, J. Gen. Intern. Med. 23 (Suppl. 1) (2008) 2–6. [25] J.E. Russo, Original research: deactivation of ICDs at the end of life: a systematic review of clinical practices and provider and patient attitudes, Am. J. Nurs. 111 (2011) 26–35. [26] D.B. Kramer, A.S. Kesselheim, D.W. Brock, W.H. Maisel, Ethical and legal views of physicians regarding deactivation of cardiac implantable electrical devices: a quantitative assessment, Heart Rhythm. 7 (2010) 1537–1542.
[27] N.E. Goldstein, R. Lampert, E. Bradley, J. Lynn, H.M. Krumholz, Management of implantable cardioverter defibrillators in end-of-life care, Ann. Intern. Med. 141 (2004) 835–838. [28] L.C. Buchhalter, A.L. Ottenberg, T.L. Webster, K.M. Swetz, D.L. Hayes, P.S. Mueller, Features and outcomes of patients who underwent cardiac device deactivation, JAMA Intern. Med. 174 (2014) 80–85. [29] S. Jacob, S.S. Panaich, R. Maheshwari, J.W. Haddad, B.J. Padanilam, S.K. John, Clinical applications of magnets on cardiac rhythm management devices, Europace 13 (2011) 1222–1230. [30] G. Marinskis, L. van Erven, Deactivation of implanted cardioverter-defibrillators at the end of life: results of the EHRA survey, Europace 12 (2010) 1176–1177. [31] W.R. Lewis, D.L. Luebke, N.J. Johnson, M.D. Harrington, O. Costantini, M.P. Aulisio, Withdrawing implantable defibrillator shock therapy in terminally ill patients, Am. J. Med. 119 (2006) 892–896. [32] M. Ahmad, L. Bloomstein, M. Roelke, A.D. Bernstein, V. Parsonnet, Patients' attitudes toward implanted defibrillator shocks, Pacing Clin. Electrophysiol. 23 (2000) 934–938.