Since the first implantable cardioverter–defibrillator (ICD) was used in 1980, device technology and the indications for ICD therapy have changed markedly. The rate of ICD implantation has increased and the ICD has become part of the standard therapy in patients who are at risk of life-threatening ventricular arrhythmias.1,2 The majority of ICD recipients are followed routinely at intervals ranging from three to six months. In addition, a substantial number of patients require additional non-scheduled visits due to arrhythmic events or system-related complications. With the growing number of implantations, ICD follow-up is now demanding huge efforts from follow-up facilities. Routine ICD follow-up visits, including system integrity check and/or confirmation that no clinical events have occurred, are time-consuming and should be avoided. Remote device monitoring offers the possibility of meeting this challenge.3,4 The application of remote monitoring will cause an imminent change in the clinical practice of cardiac rhythm management. The aim of this article is to present the role of remote patient monitoring in ICD follow-up, its potential benefits and its barriers to more general application.
Implantable Cardioverter–Defibrillator Follow-up
Routine technical follow-up visits of patients with ICDs are usually performed at short intervals due to safety concerns. With the advances in device technology, it has become theoretically possible to increase the follow-up intervals. However, the disadvantage of longer intervals is the lack of information on system integrity and the clinical status of the patient. A close follow-up schedule is mandatory over the first three months after implantation as clustering of system- and procedure-related complications occur most commonly during this period.5–8 However, more than half of all system-related complications, e.g. lead failure, may occur at any time during long-term follow-up. These complications are unpredictable and any extension of the follow-up interval increases the risk of delayed detection of late system-related complications, even though these happen infrequently.
Monitoring of System Integrity
The integrity of the implanted system is essential for appropriate device therapy. Unfortunately, a significant proportion of ICD recipients experience system-related complications. The majority of these complications are lead-related and usually occur within three months of implantation.6,9–12 Causes of ICD lead failure are an insulation defect or conductor disruption. These can affect both the high-voltage and the pace-sense circuit of the system.
Potential complications of ICD lead failure include oversensing of noise, undersensing of ventricular arrhythmias, inappropriate therapy and lethal proarrhythmia. However, more than half of all system-related complications may occur at any time during long-term follow-up.11–14 For example, the annual failure rate of defibrillation leads increases progressively over time after implantation and reaches 20% in 10-year-old leads.14
Early detection of these complications is desirable to ensure patient safety. Current ICDs provide lead impedance monitoring for the detection of lead failure, which is based on daily measurement of lead impedance. When the measurement for lead impedance monitoring occurs once daily (this is usually nocturnal), it is very unlikely that this measurement will reveal abnormal impedance values. Alert features producing an acoustic warning signal have been implemented in ICDs for early detection of system-related complications. However, these features are of limited use because of their low sensitivity in detecting lead failures.15,16 In a recently published study, Vollman et al.16 reported on the reliability of acoustic alerts to detect lead failures. These acoustic alerts have a low sensitivity and warn of only 30% of lead failures. Acoustic patient-alert features are a useful additional tool, facilitating early detection of lead-related complications or battery depletion.
Structural lead defects can be discreet at first, and may present with loss of electrical integrity for brief moments (e.g. arm movement may be momentarily affected). To increase the sensitivity to lead failures, the ‘short interval counter’ or ‘sensing integrity counter’ (SIC) has been developed to keep track of the number of short non-paced intervals (<140ms).17 This method allows the detection of noise oversensing that can be caused by clinically silent lead failures. The combination of the abnormal lead impedance trend and an increase in SIC yield a higher sensitivity for the detection of lead failure.16,17 However, the potential limitation of SIC is the likelihood of false-positive detections when applied in patients with integrated bipolar sensing electrodes.16 Integrated bipolar leads are susceptible to oversensing intracardiac signals18 and diaphragmatic myopotentials.19 Remote monitoring has the potential for early detection of lead-related complications, such as an insulation defect, lead fracture and faulty connections with the header, by continuous monitoring of lead and high-voltage impedance, sensing value and detection of episodes caused by oversensing.20–23 In an overall analysis of a worldwide database, the proportion of events related to abnormal device status was 3%. The proportion of events related to system configuration was 11%.24
Recent advisories and recalls concerning the implanted defibrillation system have given a new perspective on remote monitoring.25,26 Potential device or lead failures often occur in unpredictable ways and put patients at risk.27–29 Immediate solutions such as systematic ICD generator replacement are associated with a number of complications, including death.25 ‘Field actions’ and device advisories usually result in a recommendation of close follow-up of the outpatient, with reprogramming and frequent visits at short intervals. These recommendations further increase the workload of ICD clinics. Continuous surveillance of defibrillation systems at risk is the most attractive alternative and provides an immediate detection of device or lead failure, as shown in Figure 1. In addition, remote monitoring can eliminate unnecessary replacements and obviate the increased follow-up of large patient populations affected by these advisories.
Remote Monitoring of Arrhythmias and Heart Failure
The current generation of devices has the capability to track several biological parameters such as heart rate and rhythm, heart rate variability, patient activity and thoracic or transvalvular impedance. Remote monitoring of these data and interpretations can serve as a triage tool to determine which patients need further medical attention. Atrial fibrillation is the most commonly encountered arrhythmia in clinical practice that requires treatment. Although the majority of patients have episodes of asymptomatic or minimally symptomatic atrial fibrillation, the arrhythmia can have a profound effect on patients with heart failure and can trigger inappropriate device therapy in ICD recipients. Monitoring the atrial rhythm would enable the detection of atrial fibrillation that can be used to guide cardioversion decisions, initiate or continue anticoagulation therapy and optimise pharmacological therapy and device programmation.21,30,31 Studies with remote monitoring demonstrated an accurate and comprehensive diagnosis of arrhythmic events.21–23,30–32 The Home ICD Trial demonstrated that remote monitoring resulted in an immediate detection of arrhythmic events. This facilitated an optimised and individualised cardiac rhythm management.32 Cardiac resynchronisation therapy (CRT) is now considered a class I indication in patients with drug-refractory heart failure and electrical or mechanical dyssynchrony. Despite the significant reduction in heart failure mortality and hospitalisations, it is known that 30% of CRT patients do not respond to this therapy. Assessment with remote monitoring may assist in programming and/or re-intervention.
The Organisation of Remote Monitoring
Remote monitoring is feasible, may facilitate ICD follow-up and may lead to early detection of system-related complications and changes in the patient’s clinical status. Despite these beneficial effects, several questions regarding remote monitoring remain. There are issues related to quality assurance, medico-legal aspects and reimbursement of remote follow-up. Who is qualified to access data obtained from remote monitoring? Should access be given to the electrophysiologist only, the non-electrophysiologist, the technician, the nurse or several of these healthcare workers? Who is responsible? This is particularly relevant in the case of delayed observation of events that may lead to adverse outcomes. Transmitted data should be accessed at regular intervals and responded to if events are observed. This approach is not final, is limited to Home Monitoring® (Biotronik Inc, Berlin, Germany) and remains subject to regulatory interventions. Another issue concerns symptoms or problems reported by the patient that are not mirrored by indicators in transmitted data. This issue can be managed using short interviews by telephone or other remote management systems. The integration of remote monitoring systems in a more comprehensive ‘home’ monitoring approach remains a challenge. A significant problem concerns reimbursement for the virtual outpatient clinic. Medicare in the US has recently approved reimbursement for the analysis of remote data if a full interrogation occurs. Another issue lies in the fact that the available remote monitoring systems are not uniform. How can one code describe the various monitoring systems employed?
The number of ICD implantations is growing due to the increase in indications for defibrillation therapy. As a result, the patient population with ICDs is becoming more heterogeneous, requiring a differential approach to patient management. Some patients need only routine follow-up while others, such as heart failure patients, require more intensive follow-up. In addition, patients affected by device advisories can be closely monitored for early detection of device or lead failure.