Recent guidelines have emphasized the importance of discussing the issue of deactivation near the end of life with patients with an implantable cardioverter-defibrillator (ICD). Few studies have examined the patient perspective and patients’ wishes. We examined patients’ knowledge and wishes for information; and the prevalence and correlates of a favorable attitude toward deactivation. Three cohorts of ICD patients (n = 440) extracted from our institutional database were asked to complete a survey that included a vignette about deactivation near the end of life. Of the 440 patients approached, 294 (67%) completed the survey. Most patients (68%) were aware that it is possible to turn the ICD off, and 95% believed it is important to inform patients about the possibility. Of the patients completing the survey, 84% indicated a choice for or against deactivation. Psychological morbidity was not associated with a response in favor or against deactivation (p >0.05 for all). The wish for a worthy death near the end of life was an independent associate of a favorable attitude toward deactivation (odds ratio 2.14, 95% confidence interval 1.49 to 3.06, p <0.0001), adjusting for the importance of avoiding shock-related pain, anxiety, and poor quality of life and other potential confounders. In conclusion, most ICD patients seemed to favor device deactivation at the end of life, primarily owing to the wish for a worthy death. This finding indicates that patients have thought about the issue of deactivation near the end of life and might welcome the chance to discuss it with their physician.
Little is known about the patient perspective and patients’ attitudes toward deactivation of their implantable cardioverter-defibrillator (ICD) near the end of life, their need for information, and the appropriate timing for discussing deactivation. In addition, to our knowledge, no study has examined the factors that might influence patients’ attitude toward deactivation. Such information is valuable for optimizing the treatment and care of an increasing number of ICD patients and could also be incorporated in the training and education of junior doctors to establish a healthcare system that provides consistent and high-quality care. As advocated by the American Institute of Medicine in their recommendations for improving the healthcare system of the 21st century, all treatment and care should fulfill the 6 criteria of being safe, effective, timely, equitable, efficient, and patient-centered. Hence, in the present study, we examined the patients’ perspective and specifically focused on (1) patients’ knowledge and wishes for information about deactivation of the ICD toward the end of life, and (2) the prevalence of patients in favor of deactivation and the correlates of a favorable attitude.
Methods
Data from 440 patients were extracted from the institutional database of the Thoraxcenter, Erasmus Medical Center (Rotterdam, The Netherlands) and stratified into 3 groups according to the duration since implantation: group 1, recent de novo implanted (implanted from February 1, 2011 to January 31, 2012); group 2, moderate device experience (implanted from February 1, 2008 to January 31, 2009); and group 3, considerable device experience (implanted from February 1, 2005 to January 31, 2006). Patients were approached by mail in May 2012 with an introductory letter explaining the rationale of the study and a survey that included a vignette, purpose-designed questions, standardized and validated questionnaires, and a self-addressed stamped envelope. If patients did not respond within a 4-week period, they were sent a reminder letter, with a new survey. The medical ethics committee of the Erasmus Medical Center approved the study protocol. The study was conducted according to the ethical guidelines of the Helsinki Declaration, as set out by the World Medical Association. All patients provided written informed consent.
For the present study, we designed a vignette to investigate patients’ attitudes toward device activation at the end of life: “Imagine that you are told that you only have 3 to 6 months to live and that no additional treatment options are available. In this situation, it is possible to turn the ICD off. The ICD will then no longer provide therapy to terminate rhythm disturbances. As the patient, you are the only one who can make the decision to have your device turned off. In such a situation, the treating cardiologist will discuss with you (and your family) the pros and cons of turning your ICD off versus keeping it on in the last phase of your life. Below is listed a set of possible pros and cons with respect to turning your device off or keeping it on.” The 4 different statements were listed with respect to device activation and associated pros and cons: (1) ICD shocks can be “painful” (pro: “no painful shocks”; con: “painful shocks”), (2) ICD shocks can lead to more anxiety (pro: “avoid shock-related anxiety”; con: “accept the risk of shock-related anxiety”), (3) ICD shocks can lead to poorer quality of life (pro: “preserve your quality of life”; con: “deterioration in quality of life”), and (4) you could receive ICD shocks during the dying process (pro: “a worthy death”; con: “an unworthy death”). Patients were asked to rate these 4 statements using a 5-point Likert scale ranging from 1 (unimportant) to 5 (decisive) to evaluate to which extent these aspects influenced their decision toward deactivation of the ICD at the end of life. Ultimately, patients were asked to make a decision with respect to device deactivation and whether they were in favor or against (yes or no).
In the survey, we also included 6 purpose-designed questions to ascertain patients’ knowledge about deactivation and their wishes with respect to information on the topic: (1) “Did you know that it is possible to turn off the ICD toward the end of life?” (yes or no), (2) “Do you think that ICD patients should be informed about the possibility of turning off the ICD?” (yes or no), (3) “If you think that ICD patients should be informed about the possibility of turning off the ICD, how should this be done?” (response options [1 of 3]: orally; in writing; or both orally and in writing), (4) “When do you think the physician should discuss the issue of deactivation with patients?” (response options [>1 possible]: before implantation; at battery replacement; in the case of life expectancy being significantly reduced; and/or during the dying process), (5) “Do you think that discussing the issue of deactivation leads to increased anxiety in patients?” (yes or no), and (6) “Do you think that discussing the issue of deactivation leads to increased anxiety in patients’ families?” (yes/no).
Information on patients’ demographic and clinical characteristics at the time of implantation was obtained from the patients’ medical records, except for smoking status, partner status, and education, which were obtained by purpose-designed questions in the questionnaire. Information on demographic variables included gender, age, partner status, and education. Information on clinical variables included indication for ICD therapy (primary vs secondary), ICD type (i.e., single chamber, dual chamber, and biventricular), QRS, left ventricular ejection fraction, coronary artery disease (CAD), previous percutaneous coronary intervention, previous coronary artery bypass grafting, heart failure, atrial fibrillation, diabetes mellitus, smoking, and cardiac medication (i.e., amiodarone, β blockers, diuretics, angiotensin-converting enzyme inhibitors, statins, and digoxin). Information on appropriate and inappropriate ICD shocks was obtained by device interrogation during the follow-up period.
Given the possibility that psychological morbidity might influence patients’ attitudes toward deactivation at the end of life, we included questionnaires that assessed anxiety, depression, and Type D personality. Anxiety was assessed using the 7-item Generalized Anxiety Disorder Scale, with responses indicated on a Likert scale from 0 to 3 (score range 0 to 21). A cutoff of ≥10 represented a moderate level of anxiety. We used the 9-item Patient Health Questionnaire to measure depressive symptoms. Items were rated on a Likert scale from 0 to 3 (score range 0 to 27), with a cutoff of ≥10, reflecting a moderate level of depressive symptoms. Type D personality was assessed with the 14-item Type D scale, which assesses the separate personality traits, negative affectivity and social inhibition. A high score (i.e., ≥10) for both negative affectivity and social inhibition defines those with a Type D personality.
Between-group comparisons between study participants and nonparticipants on baseline characteristics were performed using the Chi-square test (or Fisher’s exact test when appropriate) for nominal variables and Student’s t test for independent samples for continuous variables. Between-group comparisons, stratified by time since implantation (recent de novo implanted, moderate device experience, and considerable device experience), on demographic and clinical characteristics at the time of implantation were performed using the Chi-square test (or Fisher’s exact test when appropriate) for nominal variables and one-way analysis of variance for continuous variables. For the analyses of variance, a post hoc Bonferroni correction was applied in the case of statistically significant differences among the 3 groups. Univariable and multivariable logistic regression analyses were used to examine the correlates of a favorable attitude toward deactivation at the end of life. On multivariable analysis, the 4 vignette statements related to a choice for or against device activation toward the end of life, gender, age, partner status, indication for ICD implantation, number of shocks since implantation, CAD, heart failure, diabetes mellitus, and time since implantation were entered as independent variables. For the results of the regression analyses, odds ratios and 95% confidence intervals (CI) are reported. All tests were 2-tailed, and p <0.05 was used to indicate statistical significance. The data were analyzed using PASW Statistics, version 20, statistical software (IBM, Armonk, New York).
Results
Of the 440 patients approached, 294 (67% response rate) completed the survey. No statistically significant differences were found in the response rate (68%, 65%, and 68%) across the 3 groups (recent de novo implanted, moderate device experience, and considerable device experience, respectively; p = 0.79). Given that information on partner status, education, and smoking was captured using purpose-designed questions in the questionnaire, this information was not available for nonresponders. However, for all other demographic and clinical characteristics, including medication, the responders and nonresponders did not differ systematically, except that nonresponders were less likely to be prescribed statins than were responders (49% vs 62%; p = 0.01).
The demographic and clinical baseline characteristics for the total sample and stratified by group according to the interval since implantation are listed in Table 1 . The 3 groups differed systematically by the indication for ICD implantation, ICD type, mean QRS, CAD as the underlying etiology, the presence of heart failure, and number of shocks since implantation. Patients with considerable device experience were more likely to have an ICD for a secondary prevention indication (p = 0.03), to have a dual-chamber ICD (p = 0.03), to have CAD as the underlying etiology (p = 0.04), and to have experienced a shock since implantation (p = 0.001). In contrast, the mean QRS (p = 0.04) and the prevalence of heart failure (p = 0.007) were greatest in the recently implanted ICD group. No other systematic differences were found among the 3 groups on the demographic and clinical baseline characteristics.
| Total (n = 294) | De Novo Implanted (n = 110) | Moderate Experience (n = 107) | Considerable Experience (n = 77) | p Value | |
|---|---|---|---|---|---|
| Demographics | |||||
| Mean age (yrs) | 58.6 ± 13.5 | 59.0 ± 14.6 | 59.7 ± 12.0 | 56.5 ± 14.0 | 0.27 |
| Men | 213 (72%) | 75 (68%) | 78 (73%) | 60 (78%) | 0.34 |
| Married/partner | 229 (79%) | 89 (82%) | 82 (77%) | 58 (75%) | 0.47 |
| Lower education ∗ | 36 (13%) | 13 (12%) | 13 (12%) | 10 (13%) | 0.98 |
| Clinical factors | |||||
| Secondary prevention indication | 83 (28%) | 30 (28%) | 23 (22%) | 30 (40%) | 0.03 |
| Implantable cardioverter defibrillator type | |||||
| Single chamber | 157 (53%) | 55 (50%) | 67 (63%) | 35 (46%) | |
| Dual chamber | 66 (22%) | 27 (25%) | 14 (12%) | 25 (33%) | |
| Biventricular | 71 (24%) | 28 (26%) | 26 (24%) | 17 (22%) | 0.03 |
| Shocks since implantation † | 36 (12%) | 3 (2.7%) | 18 (17%) | 15 (20%) | 0.001 |
| Left ventricular ejection fraction ≤35% | 171 (77%) | 65 (77%) | 66 (74%) | 40 (83%) | 0.47 |
| Mean QRS (ms) | 128.5 ± 34.2 | 134.9 ± 36.7 | 123.6 ± 32.4 | 126.4 ± 33.5 | 0.04 |
| New York Heart Association functional class III-IV | 62 (21%) | 21 (19%) | 25 (23%) | 16 (21%) | 0.74 |
| Coronary artery disease | 160 (54%) | 51 (46%) | 59 (55%) | 50 (65%) | 0.04 |
| Previous percutaneous coronary intervention | 85 (30%) | 30 (27%) | 29 (27%) | 26 (38%) | 0.25 |
| Previous coronary artery bypass surgery | 63 (21%) | 21 (19%) | 24 (22%) | 18 (23%) | 0.74 |
| Heart failure | 108 (37%) | 53 (48%) | 32 (30%) | 23 (30%) | 0.007 |
| Atrial fibrillation | 72 (25%) | 23 (21%) | 30 (28%) | 19 (25%) | 0.47 |
| Diabetes mellitus | 50 (17%) | 19 (17%) | 24 (22%) | 7 (9.6%) | 0.08 |
| Smokers | 52 (18%) | 22 (20%) | 21 (20%) | 9 (12%) | 0.25 |
| Medications | |||||
| β Blockers | 232 (79%) | 86 (78%) | 86 (80%) | 60 (79%) | 0.92 |
| Amiodarone | 38 (13%) | 9 (8.2%) | 17 (16%) | 12 (16%) | 0.17 |
| Diuretics | 152 (52%) | 59 (54%) | 56 (52%) | 37 (49%) | 0.80 |
| Angiotensin-converting enzyme inhibitors | 207 (71%) | 79 (72%) | 74 (69%) | 54 (71%) | 0.91 |
| Statins | 181 (62%) | 60 (55%) | 69 (65%) | 52 (68%) | 0.12 |
| Digoxin | 43 (15%) | 14 (13%) | 17 (16%) | 12 (16%) | 0.77 |
∗ Educational level ≤13 years.
The extent of patient knowledge about deactivation toward the end of life and their wishes for information on the topic are listed in Table 2 . Most patients (68%) were aware that it is possible to turn the ICD off; however, remarkably, 1/3 were not aware of this option. As many as 95% of patients believed it is important to inform patients about this option, with most (62%) endorsing that patients should be informed, not only orally, but also in writing. With respect to the timing of discussing the issue of deactivation with patients, before implantation (49%) and in the case of a decreased life expectancy (55%) were most frequently endorsed; fewer patients found that at battery replacement (17%) and during the dying process (26%) were opportune moments. Most patients (82%) found it unlikely that discussing the issue of deactivation would lead to increased anxiety in patients. Similarly, most (68%) found it unlikely that broaching the topic would lead to increased anxiety in the patients’ families.
