Since the first implantable pacemaker was introduced in 1958, electronic devices designed to treat cardiac problems have experienced technological leaps. A rapidly expanding number of patients depend on this technology.1 Cardiovascular implantable electronic devices (CIEDs) now include implantable cardioverter defibrillators (ICDs), pacemakers (PMs), cardiac resynchronisation therapy (CRT) devices, implantable loop recorders (ILRs) and implantable haemodynamic monitors (IHMs).2 The indications for CIED implantation have also broadened, and CIEDs are now important therapeutic options for selected patients with bradycardia, tachycardia or heart failure as well as a diagnostic option for patients with syncope.
Regardless of type, once a device has been implanted, its continued monitoring is necessary for evaluating the effects of therapy and has become part of a complete cardiac evaluation, much like an electrocardiogram, echocardiogram or stress test. Data from the device may alert the physician and/or the patient to important events, such as device malfunction, arrhythmia, changes in haemodynamic status or inadvertent changes in programmed parameters. The goals of CIED monitoring include optimising device function, improving patient quality of life, and identifying and correcting device malfunction in a timely fashion.
This article will focus primarily on the remote monitoring of ICDs, PMs and CRT devices (which can be subdivided into devices with additional defibrillator capabilities [CRT-D] or without defibrillator capabilities [CRT-P]). An extensive discussion of ILRs and IHMs is beyond the scope of this article, but these devices will be briefly described in the last section.
The 2008 collaborative Heart Rhythm Society(HRS)/European Heart Rhythm Association (EHRA) expert consensus statement on the monitoring of CIEDs serves as a useful reference point for physicians caring for patients with CIEDs.3 Monitoring of a CIED can be done by scheduled or unscheduled clinic visits (in-person monitoring), by data transmission to the physician initiated by the device or the patient (remote monitoring), or using a combination of both modalities. The recommended frequency for both in-person and remote monitoring is based on patient-specific factors – such as left ventricular function or the possible presence of atrial arrhythmias – and device-specific factors – such as device type (e.g., PM or ICD) and possible hardware issues (e.g., the lead under advisory due to higher than expected failure rates). The data collected at follow-up should be tailored to fit the individual clinical scenario, but often include battery voltage and impedance, magnet rate, sensing and pacing thresholds, review of programmed parameters, pacing requirements, device therapies and detected arrhythmias, haemodynamic measurements, and lead parameters and impedances – plus, for defibrillators, charge time and shock impedance.3
The importance of flexibility in frequency and modality of follow-up is reflected in the HRS/EHRA expert consensus statement – the recommended minimal frequency of CIED evaluation, either in-person or via remote monitoring, ranging from monthly to yearly depending on the type of device and accompanying clinical issues.3 Similarly, another recently published consensus document on CIED remote monitoring, developed collaboratively by the International Society for Holter and Noninvasive Electrocardiology (ISHNE) and the EHRA, provides no specific recommendations for integrating remote monitoring into clinical practice.4 It is not surprising that there is significant variability in clinical practice, illustrated by a recent survey of EHRA members in which 55 % said they follow patients yearly, 35 % every six months and 10 % every three months.5 Remote monitoring was used in approximately 50 % of patients with ICDs or CRT-D devices and in 15–20 % of patients with PMs.5 Device follow-up practices vary dramatically between different countries, in part because of differences in reimbursement policies. For example, reimbursement for remote device checks is currently available in Sweden, Portugal, France and Germany, but not in Belgium, Denmark, Italy and Switzerland.4,6
Transtelephonic monitoring (TTM) was the first method for the remote evaluation of CIEDs. Over 40 years ago, Furman et al. described the use of transmitting data from pacemakers via telephone connections using a special device that can record baseline rhythms and rhythms associated with magnet application from pacemakers from any manufacturer.7 TTM can be used to provide a rudimentary assessment of pacing and sensing function, but has been particularly effective for monitoring battery depletion, with an error rate of <1 %.7 However, this method of monitoring device function does not replace in-clinic visits that provide more comprehensive information from pacemaker interrogation and also allow direct contact with a healthcare professional.8
As communication technology improved, more than a decade ago, devices were developed that allowed the remote transfer, via a telephone connection, of more comprehensive information regarding device function, similar to the information that would be obtained during an in-clinic interrogation of the device. This technology was first applied to ICDs but was later incorporated into most PMs. The next important steps in the evolution of remote monitoring were the application of wireless technology, which meant that no programming head needed to be placed on the implanted device, and the use of the automatic download of information, which essentially eliminated any patient duties for data transfer.
All of the major manufacturers have developed proprietary methods for data transfer from a patient’s device to the healthcare professional. Although there are operational differences between manufacturers, the general flow of information is similar with all systems (see Table 1). Data from the CIED are transmitted via radiofrequency waves to a home receiving station. In the past, the information would have been transmitted via a hand-held ‘wand’ but, with the current generation of CIEDs, it is transmitted using wireless technology. The home receiver has indicators that the patient can use to confirm information transfer. The encrypted information is collected and transmitted from the receiver via an analogue or mobile phone line to an Internet-based server that decodes and stores it. The information is then sent to the physician by fax, short message service (SMS) or email. The physician can also access it by logging into a secure database. Scheduled or unscheduled transmissions are available for review. Almost any parameter obtained from a standard device interrogation can be evaluated, including programmed pacing mode, lead parameters, automatic threshold tests, activity logs, automatic alert events, patient-triggered events, all memorised episodes, configurable alerts, heart failure management parameters and more. Although the collected data are basically the same, collection and management require manufacturer-specific equipment. For safety reasons, although the CIED can be interrogated remotely, programming cannot be done without an in-clinic visit using a special programmer.
Potential Clinical Value of Remote Monitoring
Remote monitoring could improve outcomes by more timely identification of new or worsening medical conditions (such as arrhythmia or heart failure) and the detection of device-related problems. Identification of atrial arrhythmias by CIEDs may provide important prognostic information. In the recently published ASSERT trial (Asymptomatic atrial fibrillation and stroke evaluation in pacemaker patients and the atrial fibrillation reduction atrial pacing trial), 2,580 patients with an ICD or a PM and without a history of atrial fibrillation were followed for 2.5 years.9 Approximately 10 % of patients had subclinical atrial fibrillation identified by the device and, at follow-up, these patients had a 6-fold increase in the likelihood of developing symptomatic atrial fibrillation and a 2.5-fold increased risk of stroke or thromboembolism. However, it is not known whether placing patients with atrial fibrillation identified solely by an CIED on anticoagulation therapy is beneficial, although many clinicians will use a risk stratification score such as CHADS2 or CHADS2-VASC to estimate the individual risk of stroke and tailor treatment appropriately.
Other investigators have reported the use of remote monitoring for evaluating efficacy of antiarrhythmic therapy for atrial fibrillation.14,15 Remote monitoring has also been proposed as a method for identifying patients with ventricular arrhythmias. In a prospective cohort study of 200 patients with CRT-P devices, 4 % developed ventricular tachycardia, often identified only by the evaluation of stored electrogram signals from the CRT-P device.10 In patients with ICDs, life-threatening arrhythmia electrograms transmitted remotely are helpful in reaching the correct diagnosis and ensuring appropriate and effective therapy (see Figure 1).
Recent advisories and recalls have made timely access to device data critical to clinical practice. Several years ago, an ICD lead (Medtronic Sprint Fidelis) was noted to have a higher than expected failure rate (5–20 %), and that lead failure was associated with inappropriate therapy or ineffective delivery of therapy. In response, the manufacturer developed specialised algorithms that, when coupled with remote monitoring, led to more timely identification of lead failures. In an early analysis of data from 40 patients who had an implanted Medtronic Sprint Fidelis lead, remote monitoring triggered events notifications and led to unscheduled visits within one to three days of the event in four patients, three of whom had confirmed lead fractures.11 Recently, the St Jude Medical Riata ICD lead has been associated with failures due to the extrusion of one of the conductor wires through the external insulation. In one study of 171 of these leads with electrical abnormalities, there was a reported occurrence of inappropriate therapy in 33 % and failure to deliver high voltage therapy in 6 %.12 In all cases of potential lead malfunction, remote monitoring is an important tool that allows more timely identification of problems and reduces the likelihood of significant clinical consequences. Lead performance can be evaluated by analysing stored data for evidence of noise and over-sensing and by measuring temporal changes in impedance.11–13
Manufacturers have developed algorithms and specialised monitors (both stand-alone and incorporated into CIEDs used for heart rhythm therapy) to facilitate the management of heart failure (see Figure 2). One commercially available system (OptiVol®, Medtronic) measures intrathoracic lead impedance as an estimate of fluid status, with decreasing impedance correlating with increased pulmonary oedema. Several other sensors that have been evaluated include stand-alone or integrated direct left atrial pressure monitoring (HeartPOD® System or Promote LAP® System, St Jude Medical), an intracardiac right ventricular pressure sensor (Chronicle® IHM, Medtronic) or a pulmonary artery pressure transducer (Champion®, CardioMEMS), with many more in development.14
The information gathered from the device or algorithm is incorporated to clinical data to assist the physician in managing complex patients. A meta-analysis of randomised controlled and cohort trials of remote monitoring in heart failure patients showed that remote monitoring was associated with a significantly lower number of deaths and hospitalisations.15 The study population included more than 8,000 patients followed for up to 18 months, with a mean ejection fraction of 35–40 % and New York Heart Association (NYHA) functional class III–IV.
Randomised Studies of Remote Monitoring
Several recently published prospective multicentre trials have established the benefits of remote monitoring (see Table 2). In the PREFER (Pacemaker remote follow-up evaluation and review) study, 897 patients with a clinical indication for a permanent pacemaker (33 % for sinus node dysfunction and 67 % for atrioventricular block) were randomised to either a remote monitoring group (remote interrogation every three months and clinic visit at one year) or a conventional follow-up group (clinic visits at six and 12 months combined with traditional TTM every two months).16 The primary endpoint of the study was the identification of ‘clinically actionable events’ (defined as events that might be associated with a change in patient management), which included identification of atrial and ventricular arrhythmias, an increase in ventricular pacing and a possible device malfunction. Remote monitoring was associated with earlier identification of a clinically actionable event (5.7 months) compared with conventional follow-up (7.7 months). Importantly, in the conventional follow-up group, TTM only detected three out of 190 events, while, in the remote monitoring group, 446 out of 676 events were identified.
In the Clinical evaluation of remote notification to reduce time to clinical decision (CONNECT) study, 1,997 patients undergoing ICD or CRT-D device implantation were randomised either to follow-up using wireless automatic remote monitoring or to conventional in-clinic follow-up.17 The interval between a clinical event and a clinical decision was reduced from a median time of 22 days for patients monitored in-clinic to 4.6 days for those in the remote monitoring group (p<0.001). Remote monitoring was not associated with a decrease in healthcare use parameters, such as hospitalisation or accident and emergency department visits, but did increase the number of unscheduled clinic visits.
In the TRUST (Lumos-T safely reduces routine office device follow-up) trial, 1,339 patients undergoing ICD implantation (approximately 75 % for primary prevention) were randomised to standard in-clinic follow-up (every three months) or remote follow-up (initial in-clinic evaluation three months after implant, online evaluations every three months, second planned in-clinic visit 15 months after implant).18 At one year, remote monitoring was associated with a 45 % reduction in unscheduled in-hospital device evaluations without morbidity being affected.
The curves between the two groups began diverging at approximately five months and, importantly, continued to diverge after the initial separation, suggesting that the potential benefits of remote monitoring may be even higher with longer follow-up. Furthermore, median time from onset of clinically significant events to physician evaluation of patients with first episodes of atrial fibrillation, ventricular tachycardia and ventricular fibrillation was lower in remotely monitored patients than in the conventional group (1 day versus 35.5 days). Surprisingly, the promptness of arrhythmia detection did not translate into a reduction in adverse events rate or overall mortality.
Recently, two studies from France were presented at the 2011 European Society of Cardiology meeting that highlighted the potential benefits of remote monitoring. In the EVATEL (Evaluation of the ‘tele-follow-up’ for the follow-up of implantable defibrillators) study, 1,501 patients with newly implanted ICDs were randomised to remote monitoring or clinic visits.19 Importantly, this is the first study on remote monitoring that has not been primarily sponsored by a device manufacturer. At one year after implant, there was no difference between the groups in the rate of major cardiovascular events (a combined endpoint that included death, cardiovascular hospitalisation and ineffective or inappropriate ICD therapy). However, patients who were remotely monitored had fewer inappropriate ICD therapies (4.7 % versus 7.5 %, p=0.03).
The second study, the ECOST (Benefits of implantable cardioverter defibrillator follow-up using remote monitoring) trial, enrolled 433 patients, lasted for 27 months and randomised patients to remote monitoring with daily data transmission or conventional clinic visits.20 The patients who were remotely monitored had fewer inappropriate shocks than the conventionally monitored group (5 % versus 10.4 %, p=0.03) as well as fewer hospitalisations (three versus 11 patients, p=0.02). In the remote monitoring group, fewer inappropriate shocks and a decrease in the number of charged shocks significantly prolonged battery life.
Another benefit of remote monitoring has been the development of large databases containing a significant number of device evaluations.21,22 For example, Boston Scientific Corporation manages data from more than 150,000 patients and over 7 million transmissions through the LATITUDE® system. These large databases have provided important insights into the potential benefits of remote monitoring and the impact of arrhythmias in large numbers of patients, and hold significant promise for optimising the care of patients with CIEDs.
For example, in the Long-term Outcome After ICD and CRT Implantation and Influence of Remote Device Follow-Up (ALTITUDE) study, almost 200,000 patients with ICDs or CRT-D devices were followed for more than two years, one-third by remote monitoring and two-thirds by conventional in-clinic evaluations.23 Patients who were remotely followed had a 50 % reduction in mortality. Although a non-randomised study, further analysis of the data using scenarios of proportionate risk determined that only if the risk factor burden in the non-remotely monitored population were 5 times that of the monitored patients, would imbalance in these baseline factors reproduce the mortality difference observed. Hence, actual mortality benefit could be gained with addition of remote monitoring to patient care.
An analysis of data from the Discovery Link® database (generated from the Medtronic remote monitoring system) evaluated more than 100,000 patients from almost 3,000 institutions and found that lower thresholds (both rate and duration) for diagnosing ventricular arrhythmias and the presence of atrial fibrillation with rapid ventricular rates were associated with an increased likelihood of shocks.24 A future analysis of the same database will compare remote monitoring of implanted cardiac devices to in-clinic monitoring and, as secondary outcome measures, will evaluate clinician and patient ease of use and satisfaction, cost-effectiveness and work flow issues associated with remote monitoring. Another study based on data from the CareLink® registry (Medtronic) will examine whether the event-triggered heart failure alerts from the Medtronic OptiVol® fluid monitoring system, with automatic notification to physicians, will translate into reduced heart failure-related hospitalisations and death rates in patients monitored remotely compared with those receiving standard clinical assessment.25
A concern arising from the widespread use of remote monitoring is the loss of direct contact and interaction between physician and patient. Although certain clinical decisions cannot be made without a formal clinic evaluation, having prior knowledge of remotely gathered data (or a lack thereof) can facilitate decision-making. At the present time, there is no capability for the remote reprogramming of CIEDs; patients may therefore still require clinic visits. A questionnaire-based study of patients with implanted devices followed via a remote monitoring system as standard clinical practice showed a high level of acceptance and satisfaction of this new technology.26 In another study of Italian patients with different types of CIEDs, 88 % of patients had a positive attitude towards remote monitoring, particularly those with ICDs or CRT-D devices.27
With the accumulation of vast amounts of patient information, data protection and safety becomes even more important. Every manufacturer has incorporated safeguards to protect patient data. However, in 2008, Halperin and colleagues published a report demonstrating that unauthorised access to information on an ICD and actual reprogramming is feasible, although difficult, using radio-based ‘attacks’. Currently, data are stored and managed by the device manufacturers with individual password-protected access given to healthcare providers. In a multispeciality setting, data sharing among physicians involved in the patients’ care is limited to information provided to team members by the physician with primary access. Patient privacy laws must be respected without limiting access to data. The responsibility for data protection falls primarily on the manufacturer and data could be acquired in one country and stored in another with differing data protections laws. With the large amount of data transmitted from a growing number of CIEDs, healthcare providers could be held liable for event notifications and interpretation in the future. If this was to happen, would physicians be as willing to follow patients via remote monitoring? Remote monitoring also brings up patients’ rights issues. Should patients have access to their own data? Should they provide specific consent that information from their device can be used for analyses of large databases?
Initially, data from remote monitoring have focused on arrhythmia management issues. However, since many patients who require ICDs and CRT devices have heart failure, manufacturers have been focusing on developing methods for monitoring heart failure. One manufacturer has developed a special left atrial pressure sensor that can be used to monitor left atrial pressure and fluid status. Beyond this, another manufacturer has developed technology that allows automated assessment of ST segment changes, which could possibly be used for monitoring for ischaemia in patients with coronary artery disease.28 Methods for remotely estimating systolic blood pressure are being developed and hold promise for more detailed monitoring of blood pressure in patients with hypertension but no overt heart disease.
Remote monitoring has emerged as a welcomed tool. Patient satisfaction and compliance increase, while use and cost of healthcare decrease. A clinic visit before which a remote transmission has been evaluated means the clinician has more time to focus on pertinent and pressing medical issues. It is important to remember that careful follow-up once a patient has a device will frequently have as great an impact on their outcomes and quality of life as the initial decision on whether or not to implant the device. Remote monitoring allows more timely identification of problems and this benefit has been noted in numerous studies. In the future, we believe that remote monitoring will become the standard method for follow-up of patients with CIEDs.