Derek T. Connelly Consultant Cardiologist Glasgow Royal Infirmary - PowerPoint PPT Presentation

Derek T. Connelly Consultant Cardiologist Glasgow Royal Infirmary and West of Scotland Heart & Lung Centre, Golden Jubilee National Hospital Honorary Clinical Associate Professor, University of Glasgow

Meta-analysis - ICD v Amiodarone S J Connolly, Eur Heart J 2000; 21 :2071-8

MADIT II Results Moss et al New Engl J Med 2002; 346 : 877-883

SCD-HeFT: Mortality by Intention-to-Treat 0.4 HR 97.5% Cl P-Value Amiodarone vs. Placebo 1.06 0.86, 1.30 0.529 ICD Therapy vs. Placebo 0.77 0.62, 0.96 0.007 0.3 Mortality 0.2 Amiodarone 0.1 ICD Therapy Placebo 0 0 6 12 18 24 30 36 42 48 54 60 Months of follow-up Bardy GH et al New Engl J Med 2005; 352: 225-37

Swedberg K et al Eur Heart J 1999; 20: 136-9

Emergency: Magnet

Deactivation: challenges  Availability of magnet for inhibition of tachycardia therapies  Availability of programmer for patients being cared for at home or in hospice  Delivery of service: who takes the programmer to the patient’s home?  Limited programming options for patients who might want certain therapies programmed

Functions of the ICD  Bradycardia pacing (including post shock)  (Cardiac resynchronisation pacing)  Detection of arrhythmias  (Detection of worsening heart failure)  Anti-tachycardia pacing to terminate VT  Cardioversion and defibrillation of VT and VF

Palliative Care and ICDs  There is nothing “palliative” about the shock function of an ICD  Shocks are painful  Shocks are generally infrequent  Fear of shocks* may be more of a problem than shocks themselves  *or fear of multiple shocks

Defibrillation and pain  Shock lasts 10 ms  Pain lasts a few seconds  The average ICD patient gets less than one shock per year  1 year = 31.5 million seconds  1 month = approx 2.5 million seconds  If a patient is getting 1 shock per month, pain is present 0.0001% of the time

Sudden Death During Ambulatory ECG Monitoring  157 cases from 10 published studies  98 (62.4%) VT  VF  13 (8.3%) primary VF  20 (12.7%) torsades de pointes  Often in patients without structural heart disease, being treated with antiarrhythmic drugs  26 (16%) bradycardia  asystole  Very few had ischaemic ST-T changes Bayes de Luna, Coumel & Leclerq Am Heart J 1989; 117 : 151-9

Characteristics of Patients Who Die with Heart Failure and Low EF  160 patients with LVEF  35% died in FU at Brigham and Women’s Hospital, Boston, between Jan 1 st 2000 and October 2003  In 6 months prior to death 93% had NYHA III/ IV symptoms  Death was considered sudden in only 21% of cases  Often bradycardia / PEA rather than VT/VF  37% had ICDs (mostly switched off) Teuteberg JJ et al. J Cardiac Failure 2006;12:47-53

If I had an ICD… and a terminal illness  I still wouldn’t want to die suddenly  I wouldn’t want CPR … but still might want my ICD to be active  I would still want antitachycardia pacing for VT  I might still put up with the occasional shock …  But I wouldn’t want to have multiple shocks if the first 1- 2 shocks didn’t work...  And I wouldn’t want to have several shocks in one day

If I had an ICD… and a terminal illness  I still wouldn’t want to die suddenly  I wouldn’t want CPR … but still might want my ICD to be active  I would still want antitachycardia pacing for VT  I might still put up with the occasional shock …  But I wouldn’t want to have multiple shocks if the first 1- 2 shocks didn’t work...  And I wouldn’t want to have several shocks in one day These are not programmable options!

Kramer DB et al N Engl J Med 2012; 366: 291-3

ICD replacement “There are several important opportunities for – and obstacles to – making ICD replacement a more deliberate process”  Clinical data must be re-evaluated  ? deterioration (or improvement) in health  Patients’ experiences of device therapy  Patients’ changes in values / preferences Kramer DB et al N Engl J Med 2012; 366: 291-3

Kramer DB et al N Engl J Med 2012; 366: 291-3

Case Study ♂  CW DoB 23/9/45  ICD implant 2005 for sustained VT  Moderate LVSD; NYHA Class I  Occasional episodes of VT requiring antitachycardia pacing; no shocks  Device now approaching end of battery life  Severe stroke 10 months ago; in nursing home; wheelchair-bound; aphasic  Wife does not accept that his disability is permanent; she wants his device to be replaced

Case study ♂  Rev PJGB DoB 6/2/31  Previous anterior MI, LBBB (QRS width 150ms); previous CABG; LVEF 30%  Primary prevention ICD 2007  No events, no therapies since implant  Recent worsening heart failure  ICD nearing end of battery life  Replace with ICD?  Replace with CRT-defibrillator?  Replace with CRT-pacemaker?

When to deactivate: Literature review Pettit SJ et al BMJ Supportive & Palliative Care 2012; 2: 94-97

 Review of records of 63 patients who died with an ICD in situ  Terminal diagnosis leading to ICD inactivation in 32%  For patients with active ICDs, 21% received a shock within their final month  For patients with deactivated ICDs, 15% had received a shock within their final month (prior to deactivation) Lewis et al Am J Med 2006; 119: 892

 Questionnaire to relatives of 100 patients who died with an ICD in situ  27 patients received ICD shocks in their last month of life  21 of these had “DNR” orders  8 patients received shocks within their last minutes of life Goldstein et al Ann Intern Med 2004; 141: 835

When should ICD deactivation be discussed?  ESC  When it is clinically obvious that a patient is about to die  When a DNR order is in force  When impairment of QOL is such that sudden death might be considered a relief  British Heart Foundation  When markers of poor prognosis in advanced heart failure are present Pettit et al 2012

When should ICD deactivation be discussed? HRS Expert Consensus Statement  Central role of the patient in deciding when their ICD should be deactivated  Counselling should be an ongoing process, starting pre- implant and continuing as patient’s health changes  Deactivation discussion should be prompted by  Multiple ICD shocks  Repeated hospitalisation  DNR orders  “end of life”  Patients decision should be supported, even if it sems illogical to the physician Pettit et al 2012

When is ICD deactivation discussed? EHRA survey (2010):  Only 4% of cardiologists discussed deactivation pre-implant  Onlu 11% were regularly involved in discussions about deactivation  Only 4% gave written info to patients about deactivation Survey of relatives of ICD patients  Deactivation discussed in only 27% of cases  75% in last days of life  22% in last hours of life  4% in last minutes of life Pettit et al 2012

ICD deactivation in hospices  Survey of 900 hospices in USA (2010)  40% of hospices had at least one patient who had experienced multiple shocks in the preceding year  42% of patients had ICD deactivated while receiving hospice care  Only 10% of hospices had policies that addressed deactivation at end of life Goldstein et al Ann Intern Med 2010; 15: 296

Pettit SJ et al BMJ Supportive & Palliative Care 2012; 2: 94-97

ICDs: End-of-life issues  No “evidence base”  Difficult to discuss these issues pre-implant (unless patient / relative initiates discussion)  Should be discussed openly and sensitively when any crisis or deterioration occurs  Multiple shocks  Deteriorating heart failure  Concomitant life-threatening illness, e.g. malignancy, renal failure, stroke  Should be mentioned routinely in booklets / website info that is made available to patients  In the UK, patient info is available from the British Heart Foundation (www.bhf.org.uk ) and the Arrhythmia Alliance (www.Heartrhythmcharity.org.uk)

Why is deactivation so difficult to discuss? No guidelines  No evidence!  Prognostication in heart failure is difficult which makes the  timing of device deactivation uncertain Emotive subject requiring advanced communication skills  Lack of education   Healthcare professionals – considering deactivation at the appropriate time  Patients concern that they will die immediately Ethical issues 

When I die, I want to go quickly in my sleep, like my grandfather …. not screaming in terror like the passengers in his car

Conclusions  All patients want to avoid undue suffering at time of death  BUT many patients want to avoid sudden death!  Even at an advanced stage in their condition  ICD patients may be a “self - selected” group  Death by VT/VF storm is highly unusual  The wishes and preferences of the patient are of the utmost importance

Conclusions  Deactivation should be discussed early  And again at appropriate points in the patient’s illness  Deactivation should be performed late?  At a time when the patient wants it  Deactivation should be an available option in home or hospice  Further research needed regarding therapies that patients might find acceptable close to end of life  Dialogue needed with manufacturers regarding programming options at end of life