Reviewed by Asher Ashfaq, OMPT, PT, CPC, CMP. Last reviewed July 2026.
If you or someone close to you has just had an ICD fitted, the most important thing to understand is this: the device does not shorten life, it protects it. An ICD exists to stop one specific cause of death, a dangerously fast heart rhythm. Survival figures you find online reflect the underlying heart condition that made the device necessary, not harm from the device itself. Those figures also vary enormously depending on age, heart function, kidney function, and other conditions, which is why an average number tells you very little about any individual. Here is an honest account of what is known.
What an ICD Does
A pacemaker treats a heart beating too slowly. An implantable cardioverter defibrillator does something different: it monitors continuously for dangerously fast or chaotic rhythms and stops them, either with rapid painless pacing or with a shock. Nearly all ICDs also include full pacemaker function.
Two reasons people receive one. Secondary prevention, after already surviving a dangerous rhythm or cardiac arrest. Primary prevention, where the heart is weakened enough that the risk is considered high, most often when the ejection fraction is 35% or below.
The Honest Framing
An ICD is a safety net against sudden cardiac death. It does not treat the weakened heart muscle underneath, and that is what sets prognosis.
A large trial called DANISH illustrates this precisely. In patients with non-ischaemic cardiomyopathy, the ICD roughly halved sudden cardiac death, from 8.2% to 4.3%, while overall survival was not significantly different. The device did its job. The underlying disease still determined the outcome.
This is also why every guideline recommendation for an ICD requires an expected meaningful survival of more than a year. The device is offered to people expected to live long enough to benefit.
What the Survival Data Shows
These figures describe groups of people with significant heart disease. They are not a prediction for any individual.
| Population | Finding |
|---|---|
| National cohort, England | 5-year survival 64% after ICD; median survival 6.8 years |
| Secondary prevention registry | Around 9 in 10 alive at 1 year |
| Primary prevention, real world | Mortality 13.9% at 2 years, 17.3% at 3 years |
One figure circulating widely online, a median survival of 8.6 years, traces only to a commercial website and has no identifiable source. Treat it accordingly.
The Trials That Shaped Practice
MADIT-II studied patients with a previous heart attack and an ejection fraction of 30% or below. Mortality was 14.2% with an ICD against 19.8% without, a 31% relative reduction.
SCD-HeFT studied heart failure patients with an ejection fraction of 35% or below and found a 23% relative reduction in mortality, an absolute reduction of 7.2% at five years. Amiodarone, the drug comparator, showed no benefit.
DANISH studied non-ischaemic cardiomyopathy and found no significant difference in overall mortality, though sudden cardiac death halved. Benefit was concentrated in younger patients, roughly under 59.
DANISH changed practice. In non-ischaemic cardiomyopathy an ICD became an individualised decision weighing age and other conditions, rather than an automatic one. If you have that diagnosis and your cardiologist discussed the decision at length rather than simply recommending a device, this is why.
What Affects Prognosis
Independent predictors of higher mortality in the year after implantation include age of 80 or over, significantly reduced kidney function, atrial fibrillation, and more severe heart failure symptoms.
Kidney function matters more than most people expect. In MADIT-II, each modest drop in kidney filtration was associated with a 17% increase in mortality.
There is a genuinely difficult finding worth stating plainly, because it explains why some people are advised against a device. Risk scoring suggests that patients at low risk gain the most survival benefit, while those at the very highest risk gain essentially none, because they are likely to die of progressive heart failure rather than a sudden rhythm. Being advised that an ICD may not help you is not being written off. It is a recognition that this particular device addresses a cause of death that is not the main threat in your case.
The Device Itself
Modern ICDs last a median of around 5.9 years, or about 4.9 years for the type that also resynchronises the heart, since continuous pacing uses more battery.
Replacement is a smaller procedure than the original implant. The generator is exchanged and the leads usually stay in place. Complication rates reflect this: about 4% for a straightforward generator change, rising to over 15% when leads are added or the system upgraded.
Shocks and Quality of Life
Shocks are less common than they used to be. Improved programming means most people never receive one. In one major trial only around 1% of patients received two or more appropriate shocks.
The psychological effect is real and worth naming. Across a large analysis of nearly 40,000 patients, roughly 15% experienced depression, 23% anxiety, and 12% post-traumatic stress symptoms, with around 20% developing PTSD symptoms after receiving a shock.
If you are anxious about your device, or avoiding activity for fear of triggering it, that is common and treatable. Cardiac psychology support and device support groups exist, and raising it with your cardiology team is reasonable rather than trivial.
Driving is usually restricted for about a week after a primary prevention implant, and around six months after a secondary prevention implant or after a shock for a dangerous rhythm. Commercial licences are generally not permitted. Rules vary by country and by US state, so check locally. One reassuring finding: in one study, accident rates among ICD patients were actually lower than in the general driving population.
Turning Off the Shocks at the End of Life
This is under-discussed and matters a great deal, so it is included deliberately.
If someone with an ICD is approaching the end of life from another cause, the device can continue delivering shocks during the dying process. This is distressing for the person and for their family, and it is avoidable.
The shocking function can be deactivated. It is done non-invasively with a programmer held over the device, takes minutes, involves no surgery and no pain. Pacing and comfort functions can be left running.
Professional consensus, endorsed by the major cardiology and palliative care bodies, is that deactivating shock therapy is withdrawal of a life-sustaining treatment. It is legally and ethically permissible, and it is not assisted dying. Patients or their surrogates may request it, and a clinician who is uncomfortable doing it should refer to someone who will.
Note this is different from switching off a pacemaker, where a dependent patient’s heart rate may fall as a result. Turning off ICD shocks prevents painful shocks without causing death.
If you or a family member is in this situation, it is a conversation worth having early rather than during a crisis.
ICD FAQs
Does an ICD shorten your life?
No. It reduces the risk of dying from a dangerous heart rhythm. Survival figures reflect the underlying heart condition that made the device necessary, not harm from the device.
What is the life expectancy after an ICD?
It varies enormously with age, heart function, kidney function, and other conditions. A national cohort found 5-year survival of 64% and median survival of 6.8 years, but averages across people with significant heart disease say little about any individual.
Why was I told an ICD might not help me?
Because the device prevents one specific cause of death. In people whose main risk is progressive heart failure rather than a sudden rhythm, survival benefit can be minimal. It is a judgement about what is most likely to threaten you, not a judgement about your worth.
How long does the battery last?
A median of around 5.9 years, or about 4.9 years for devices that also resynchronise the heart. Replacement exchanges the generator while leaving the leads, and is a smaller procedure than the original implant.
Will I be shocked often?
Most people never receive a shock. Modern programming has made them uncommon. Anxiety about shocks is common and worth raising with your team, as support exists.
Can the device be turned off at the end of life?
Yes. The shocking function can be deactivated non-invasively in minutes, without surgery or pain, while leaving pacing running. Professional consensus supports this as withdrawal of a life-sustaining treatment, and it prevents distressing shocks during dying.
