ICDs Keep Pace With Change

31 August 2006 (Last Updated August 31st, 2006 18:30)

Sudden cardiac death is a leading cause of mortality in the UK, but the regulations governing treatment have changed. Consultant cardiologists Philip Spurrell and Christopher Rinaldi look at the recent advances made in ICDs and their potential for future development.

There are over 300,000 deaths annually in the UK attributable to cardiovascular disease and between 50,000 and 70,000 of these are due to sudden cardiac death (SCD). Of these, 80% are the result of ventricular tachyarrhythmias, and are therefore potentially treatable with an implantable cardioverter defibrillator (ICD).

Since 40% of SCDs are not witnessed or occur during sleep, 80% occur at home and survival rates for out-of-hospital cardiac arrest in Western Europe are less than 5%, ICDs seem to be a logical choice for the primary prevention of SCD. About 15% of survivors will experience another event, usually fatal, within one year, and anti-arrhythmic therapy is ineffective in improving mortality in high-risk patients.

This highlights an additional role for ICDs in secondary prevention. The role of ICDs in both primary and secondary prevention of SCD is supported by a large body of evidence.


The UK National Institute for Clinical Evidence (NICE) issued its first guide-lines following its health technology appraisal on ICDs in September 2000. A revision to these guidelines was released in January 2006 to take account of studies which were not available for consideration at the time of the original appraisal.

The reappraisal resulted in the following changes: 'The recommendation on the use of ICDs for the primary prevention of sudden cardiac death has been expanded to include patients with a left ventricular ejection fraction of less than 30% (no worse than class III on the New York Heart Association functional classification of heart failure) and a QRS duration equal to or greater than 120 milliseconds, without the need for electrophysiological testing. It also includes patients who have undergone surgical repair for congenital heart conditions.'

In addition, the guidelines make the following caveat: 'This appraisal does not cover the use of implantable defibrillators for nonischaemic cardiomyopathy.' The revised guidelines on the use of ICDs will inevitably have a variety of ramifications, partly because of what it says, partly what it doesn’t.

Firstly, the addition of another defined category in the primary prevention criteria has expanded the pool of patients now eligible for consideration for an ICD. This expansion is likely to accelerate the already increasing implant rates seen in the UK. Currently this stands at 50 per million compared with an average of 85 per million in Western Europe and 400 per million in the USA.

Some have estimated that the new guidelines will result in a doubling of implant rates, while others have projected less conservative rises. This will place a greater burden on cardiological services, not just from increased implant rates but also from the associated rise in box changes, ICD-related problems and the ICD follow-up which inevitably follows.

"Concerns over efficacy, acceleration of VT and delay of a definitive shock therapy have discouraged some from using ATP for fast VTs."

As people are living longer with increasingly poor cardiac function, cardiologists are faced with ever more complex patients and problems.

The revised guidelines have also created a group of patients that previously did not fit NICE criteria for ICD implantation but now do. This creates the complex logistical problem of trying to identify who and where these patients are, since they have now disappeared into the healthcare system. Do you review all the left ventriculograms, echo reports, VT stimulation studies and the MINAP database to find these people?

One notable observation is the absence of guidelines for patients with non-ischaemic cardiomyopathy. In fact, this particular patient group has been specifically excluded from the guidelines.

The guidelines also do not look at how ICD implanting centres should determine which of these patients will receive prophylactic ICDs, and presumably this will be left to local policy decision-makers. This is likely to lead to variations in the standard of care provided to such patients in different parts of the country as well as a lack of clarity regarding funding.


One of the potential drawbacks of ICD therapy is the delivery of a painful shock. Many studies have shown that this is deleterious to quality of life. Antitachycardia pacing (ATP), on the other hand, is not painful and, in fact, often goes unnoticed by the patient. Concerns over efficacy, acceleration of VT and delay of a definitive shock therapy have discouraged some from using ATP for fast VTs.

The PainFREE Rx I study addressed this concern, in secondary prevention patients, by observing the outcome of programming two ATP sequences for fast VT (defined as 240ms≤cycle length>320ms) before shock2. Out of a total of 446 fast VT episodes, 89% were terminated by ATP, there was acceleration in 4% and syncope in 2%. Thus, a large proportion of episodes can be terminated painlessly by ATP, which would otherwise have resulted in a painful shock.

Innovations such as delivery of ATP during shock charging will also minimise any potential delay in shock delivery caused by ATP. The PainFREE Rx II study further confirmed, in both secondary and primary prevention patients, that empirical ATP for fast VT is effective and safe and improves quality of life.

The DATAS study recently reported at Cardiostim (Nice) 2006 supports the superiority of dual chamber ICDs over single chamber devices. The study found superior SVT/VT discrimination in dual chamber devices with a 33% reduction in adverse events, largely in the form of inappropriate shocks.


The Dual Chamber and VVI Implantable Defibrillator (DAVID) trial randomised patients with an ICD indication, but no indication for bradycardia pacing, to VVI pacing at 40bpm or DDDR pacing with a lower rate limit of 70bpm. Contrary to expectations, a higher rate of all causes of mortality and heart failure hospitalisation was found in the DDDR patients.

A subsequent sub-study analysis found that frequent RV pacing was an independent predictor of adverse events. However, in DDDR patients where the RV was paced less than 40% of the time, a similar outcome to that seen in the VVI group was observed.

This study has highlighted the potentially deleterious effects of RV pacing and has generated a plethora of AV search hysteresis algorithms incorporated into modern ICDs and designed to minimise RV pacing in DDD mode.

One such algorithm was assessed in the INTRINSIC RV study, which found that DDDR pacing using an AV search hysteresis was equally as effective as backup VVI pacing at 40bpm.


As the number of ICD implants looks set to increase, so too does the number of ICD-related problems. These are likely to include a proportion of ICD systems requiring extraction, whether for infection or generator pocket erosion. Lead design has reflected this concern, with coils designed to reduce tissue ingress or the coating of coils with gortex.

Systems incorporating old chronic coronary sinus pacing leads are likely to be encountered more and more.

Some operators have started to implant coronary sinus defibrillation leads when high defibrillation thresholds are encountered at implant. Some companies are considering developing pacing/ defibrillation leads with improved fixation designs specifically for use within the coronary sinus system. All these developments are likely to create their own set of challenges when removal is required.


Home monitoring systems may provide some relief to the increasing strain that ICD follow-up clinics are likely to encounter. Lead impedances, battery status and tachyarrhythmia Holter information can be automatically beamed from the patient's ICD via a receiver/ transmitter in their home to a central database, which can then send data via email to the patient's cardiology department or physician.

Although not currently licensed to replace standard follow-up in an ICD clinic, this would appear to be a natural progression of the technology.

"Home monitoring systems may provide some relief to the increasing strain that ICD follow-up clinics are likely to encounter."

Decompensation of heart failure is a common reason for admission in the ICD population. If detected earlier, then admission could be avoided. The accumulation of intrathoracic fluid during pulmonary congestion results in improved conductance across the lungs and a corresponding fall in impedance.

Certain ICD devices have exploited this fact by monitoring intrathoracic impedance and thus fluid status. This can be used to provide an early warning system.

Thus, when impedances fall outside a predetermined range a tone is emitted by the device and the patient seeks advice. Cheuk-Man Yu et al have demonstrated that impedances were seen to fall an average of 15 days before the onset of worsening symptoms, confirming its potential as an early warning system.


A flurry of medical device safety advisories for ICDs (and pacemakers) issued throughout 2005 shook the confidence of patients and physicians alike. The number and frequency caught the medical fraternity off-guard; physicians were ill-prepared, with no agreed strategy for the management of these potential device malfunctions.

ICDs by their nature are complex mechanical devices which over time have become more complex and smaller. As with any manufactured item, there is the potential for malfunction.

The management of a patient with a faulty ICD is, in some respects, straightforward and dictated by the fault. The difficulty is when the physician is faced by a patient with a device that might develop a fault (but equally might not).

The physician has to weigh up the relative risks and consequences of not replacing a device which later becomes faulty against the risks of replacing the device (which may not be or ever become faulty).

This decision would be much easier if there were no risk associated with device replacement, but this is not the case. Gould et al, following a survey of 17 centres, reported major complications from advisory-related replacements in 31 patients (5.8%), with death reported in two patients.

Maisel et al analysed annual reports to the FDA between 1990 and 2002 to determine reported rates of pacemaker and ICD malfunction. The annual pacemaker malfunction replacement rate per 1,000 implants fell significantly over the study, from 9.0 in 1993 to 1.4 in 2002.

In contrast, ICD rates fell from 38.6 in 1993 to 7.9 in 1996, but then increased markedly, peaking in 2001 at 36.4 per 1000 implants. 31 deaths (<0.01 per cent) were attributed to ICD malfunction.

In April 2006 the Heart Rhythm Society issued draft recommendations to guide both industry and physicians with regard to the surveillance, reporting and management of device malfunction.


The annual implant rate of ICDs is likely to rise considerably over the next few years following revision of the NICE guidelines in addition to improved patient screening. This will place an ever increasing burden on cardiac departments, necessitating reorganisation of cardiac services.

Technology such as home and heart failure monitoring may facilitate management, but this is likely to be outweighed by the increasing complexity of these patients.