Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Managing Advisories of Cardiac Implantable Electronic Devices
Introduction
Cardiac implantable electrical devices (CIEDs) have undergone revolutionary changes in the last decade. These changes have permitted an increase in the use of both pacemakers and implantable cardioverter defibrillators (ICDs), collectively known as CIEDs. Pacemakers were the first generation of CIEDs available for implantation in humans, as early as 1957. Implantable defibrillators in the current form, that is transvenous, have been available since 1997. CIEDs are increasing in use worldwide. Figures 42-1 and 42-2 demonstrate the trends in pacemaker utilization in the United States from 1993 to 2009 and ICD utilization in the U.S. and Europe from 1990 to 2006, respectively. These figures demonstrate trends in the U.S. and Europe, but these trends are reflective of those seen worldwide. These devices are primarily required in older populations, where conduction system disease and significant structural heart disease prevail. As the absolute and relative numbers of the elderly population increase, so do the use of CIEDs. The primary increase in CIED use is due to an increase in the implantation of prophylactic ICDs, as opposed to pacemakers. Increasing use of cardiac interventions, coupled with increasing survival post myocardial infarction (MI), a growing number of heart failure survivors, and the burgeoning elderly population have all combined to increase the population of patients eligible for implantation of prophylactic ICDs. It is projected that the heart failure population will double by the year 2025 such that the absolute number of patients eligible to receive an ICD for primary prevention or cardiac resynchronization therapy will likely rise accordingly. Recent studies have further expanded indications for cardiac resynchronization therapy, which has been found to provide a significant reduction in mortality in mild-moderate heart failure. This will apply further pressure on an already over-burdened health care system, whose budget may not be expected to expand at the same rate. The resource implications are not the only consideration in the broader use for these devices.
ICDs are significantly more complex and fraught with more complications than are pacemakers; however, the absolute numbers of pacemakers are of a large magnitude so that even though the overall risk and complications with these devices are lower, the absolute number of complications and issues that may arise may be equivalent, due to the sheer numbers of pacemakers that are implanted. The introduction of novel algorithms, advancements in battery technology, and new lead construction has resulted in important improvements in the device industry, but has also resulted in increasing complexity and perpetration of device advisories.
Device Performance
It is well known that there is an inherent failure rate with all devices, described in reliability engineering as the “bathtub curve.” Early rates of failure may be seen in devices that fail to function on first implant, known as the early “infant mortality” failure rate. The majority of these causes of failure are addressed before devices are released to the market and are addressed through bench testing, which occurs before in vivo use. Bench testing refers to rigorous experiments replicating how a device may behave in vivo and whether it can withstand such stressors as induced by cardiac motion, shoulder movement, interaction with the bloodstream, and interaction with other intravascular devices. It is well known that the number of cardiac cycles averages 100,000 times per day, up to 3.5 billion times in a lifetime. The degree of stress on an intracardiac device is significant, as exemplified in the video demonstrating the vigor with which cardiac systole can occur ( ); this video is of a patient with an ICD for hypertrophic cardiomyopathy and shows the patient at rest. The cardiac movement would be significantly intensified on exertion. Other interactions, such as metal ion oxidation, were discovered after in vivo use resulting in improvements in bench testing, as has the outcome of the Sprint Fidelis lead, discussed subsequently. Metal ion oxidation was found to occur with leads made with polyurethane as the outer conductor. These fluoropolymers can incur microdamage during the manufacturing process, thereby increasing susceptibility to metal ion oxidation, which is an oxidative degradation resulting from interaction of the microdamaged polyurethane with the bloodstream. Bench testing attempts to address such conditions, but is by no means foolproof, as evidenced by the lead advisories discussed subsequently. A random failure rate of a device is an accepted failure rate, but occurs at a very low rate. Such random malfunctions are reported in the form of product performance reports that are required to be released by manufacturers at regular intervals, where any abnormalities in performance are reported to the manufacturer. These are published as rates, the denominator being the number of products sold worldwide. There are significant limitations to product performance reports, which are discussed subsequently. Once a device reaches its design lifetime, the rate of “wear out” failures is seen to accelerate. Device advisories are declared if and when the known rate of random failures is exceeded, or if a systematic device flaw that requires closer monitoring has been detected.
The Heart Rhythm Society has convened task forces on both device and lead performance in order to standardize the approach to definitions and management. The current recommended terms for description of device performance is presented in Table 42-1 .
TABLE 42-1
Definitions of Device Performance Terms
Data from Carlson MD, Wilkoff BL, Maisel WH, et al: Recommendations from the Heart Rhythm Society Task Force on Device Performance Policies and Guidelines Endorsed by the American College of Cardiology Foundation (ACCF) and the American Heart Association (AHA) and the International Coalition of Pacing and Electrophysiology Organizations (COPE). Heart Rhythm 3(10):1250-1273, 2006.
| Device Performance Term | Definition |
|---|---|
| Malfunction | Failure of a device to meet its performance specifications or otherwise perform as intended. Performance specifications include all claims made in the labeling for the generator or lead. The intended performance of a device refers to the intended use for which the generator or lead is labeled or marketed. Whenever possible, device malfunction should be confirmed by laboratory analysis. Malfunctions do not include physician-induced damage during the course of implanting, revising, or removing the generator or lead. Extrinsic malfunctions are those caused by external factors (e.g., therapeutic radiation, excessive physical damage, including subclavian crush and direct trauma to the device pocket) including, but not limited to, hazards that are listed in product labeling. |
| Device performance | A comprehensive assessment of device quality, usability, freedom from failure (malfunction), and conformance to applicable labeling. |
| Device reliability | A measure of a device to be free of specific structural and electrical failures, typically expressed at a given point in time or a failure rate per unit of time (e.g., failure rate per month). |
| Device removed from service unrelated to malfunction | A generator or lead that is removed from service (surgical abandonment, removal, extraction, or programmed off ) for reasons not related to device failure: infection, device upgrade (e.g., pacemaker to implantable cardioverter-defibrillator), pacer/lead incompatibility, cardiac transplantation, mode change not caused by lead failure, patient death unrelated to device failure, etc. |
Types of Advisories
Regulatory authorities often use the term “recall” as dictated by their official language, but the medical community is reticent to use this term as it is suggestive of a device that should be completely removed from use, rather than one that might simply require additional monitoring. Altering this nomenclature to “safety advisory” or “advisory notice” is not easily accomplished by regulatory bodies. A survey of patients suggested a dramatic difference in patient reaction to the term recall versus advisory, where 5% of patients would prefer to have their device removed “no matter what” in the event of a recall, whereas half that number of patients would want their device removed “no matter what” in the event of an advisory. From a regulatory standpoint, device advisories are voluntary on the part of the manufacturer. Recalls must be issued in the event of a product defect resulting in a violation of the law by the regulatory body (e.g., Food and Drug Administration [FDA]) or if the device may present an unreasonable risk of substantial harm. The relative degree of health hazard that is present in the advisory is assigned a numerical classification from I to III, under FDA regulations; similar classifications are used by other regulatory bodies in Europe and Canada ( Table 42-2 ).
TABLE 42-2
Definition of Recalls
Data from U.S. Food and Drug Administration: Recalls background and definitions. 2014. Available at < http://www.fda.gov/safety/recalls/industryguidance/ucm129337.htm >. Accessed May 11, 2015.
| Recall Classification | Definition | Example of Recall |
|---|---|---|
| Class I recall | Dangerous or defective products that predictably could cause serious health problems or death | Sprint Fidelis lead: high risk of lead failure, inappropriate shocks, risk of no defibrillation |
| Class II recall | Products that might cause a temporary health problem, or pose only a slight threat of a serious nature | St. Jude Quickflex cable externalization: no reported failure of pacing |
| Class III recall | Products that are unlikely to cause any adverse health reaction, but that violate FDA labeling or manufacturing laws | Boston Scientific LATITUDE recall: patients unable to perform setup |
| Safety Alerts, Safety Advisories, Field Safety Corrective Actions or Advisory Notice | Linguistically equivalent and are less significant than class III recalls |
Review of Device Advisories
History has proven instructive in providing the basis of response to advisories before the last decade. Over the last two decades, there have been a number of device advisories affecting both pulse generators and leads. The clinical implications of which part of a CIED is affected differ accordingly. The evidence base for management of these devices has also grown over the same time period as clinicians struggle with the optimal management of patients in the context of a CIED advisory. This expanding evidence base has shaped our management of CIED advisories in the current era. A review is provided below of prior device advisories as seen affecting pulse generators and leads, in addition to the available evidence as it pertains to the various scenarios.
Generator-Associated Device Advisories
The advent of complexity with pacemakers and ICDs has been accompanied by a sharp increase in the number of advisories observed ( Figs. 42-3A and B ). These figures represent a meta-analysis performed from three registries documenting pacemaker and ICD malfunctions from 1988 to 2004. The findings demonstrate a decline in ICD malfunction rates from the early years (1988 to 1998) but a rise again between 1998 and 2002. Overall, pacemaker malfunction rates declined significantly since 1988 and have been stable since. Battery abnormalities are the most common type of reported malfunctions ( Table 42-3 ). This may be due to rapid alterations in design to meet increasing demands to improve battery longevity, decrease CIED size, and create MRI-conditional devices. A variety of malfunctions have been observed with battery behavior that may occur from design flaws, firmware issues, and so on. The root cause for these malfunctions may not always be found, but the manifestation of battery malfunctions is often rapid or premature battery depletion, and/or failure to deliver pacing or defibrillation therapy. In a pacemaker-dependent patient, or a patient prone to ventricular arrhythmia, either of these scenarios can prove to be life-threatening. These conditions can result in an overreaction to a device advisory. The threshold for replacement of a battery that is under advisory has changed markedly as evidence has become available regarding the complications associated with this type of response.
TABLE 42-3
Pacemaker and Implantable Cardioverter-Defibrillator (ICD) Malfunction Rates
Data from Maisel WH: Pacemaker and ICD generator reliability: meta-analysis of device registries. JAMA 295(16):1929-1934, 2006, Table 2.
| Malfunction | Pacemaker (%) | ICD (%) |
|---|---|---|
| Battery | 52.8 | 80.1 |
| No or low output | 14.3 | 1.5 |
| Programming malfunction | 3.2 | 0 |
| Miscellaneous | 7.9 | 5.6 |
| Unspecified | 21.6 | 12.9 |
A number of ICD generator advisories were announced in 2005, including the Medtronic Marquis, Ventak Prizm, and others. The risk of generator failure in these advisories was anywhere from 1/1000 to 1/10,000 ( Table 42-4 ) ; yet many centers responded by replacing the pulse generator earlier than would otherwise occur, as dictated by battery replacement indicators. In order to provide evidence-based management in this regard, a group of investigators began a collaboration to collect and report a Canadian perspective on the outcome of generator replacement due to an advisory ICD. The results were remarkable in that the risk of early generator replacement far outweighed any risk that may result from generator failure due to the advisory ( Table 42-5 ). The risk of major complication was 6.3%, and the risk of any complication was 8.6%, far outweighing the risk of failure from the advisory in any of the situations. At least as compelling was the variability in replacement rates across centers, ranging from 0 to 45%. This data resulted in a change in management of ICD generators under advisory. A 1-year follow-up study confirmed the high rate of complications and analyzed the factors associated with complications. These factors were found to be additional procedures on the pocket (odds ratio [OR] 2.53, 95% confidence interval [CI] 1.14-5.62; P = 0.022) and operator experience, where the presence of a consultant and fellow versus a fellow or consultant alone (OR 0.051; 95% CI 0.005-0.51; P = 0.011) was associated with fewer complications.
TABLE 42-4
Current Implantable Cardioverter-Defibrillator (ICD) Advisories Included in the Survey and Associated Risk
From Gould PA, Krahn AD, et al: Complications associated with implantable cardioverter-defibrillator replacement in response to device advisories. JAMA 295(16):1907-1911, 2006.
| Company/Device * | Date of Advisory | Advisory Issue † | Current Risk of Failure, % † |
|---|---|---|---|
| Medtronic Marquis ICD |
February 2005 | Accelerated battery depletion caused by internal battery short | 0.001 |
| Guidant Ventak Prizm 2 DR ICD | June 2005 | Short circuit caused by wire insulation problem within lead connector block | 0.1 |
| Guidant Ventak Prizm AVT, Vitality AVT and Contak Renewal AVT ICDs | June 2005 | Random memory error, limiting delivery of therapies | 0.0095 |
| Guidant Contak Renewal 3, 4 Renewal 3, 4 AVT and Renewal RF ICDs | June 2005 | Magnetic switch faulty, impairing delivery of therapies | 0.009 |
| St. Jude Photon DR, Photon Micro VR/DR and Atlas VR/DR ICDs | October 2005 | Memory chip affected by atmospheric radiation, which can impair pacing and delivery of therapies | 0.167 |
| ELA Alto ICD | August 2001 | Migration of metal which can impair pacing and delivery of therapies | 2.6 ‡ 0.1 § |
* Predominantly subpopulation of listed devices affected by advisory.
† Data obtained from physician communications and public statement releases such as those from Medtronic and Guidant. The current risk of failure represents the number of failures divided by the number of devices implanted at the time of advisory disclosure.
TABLE 42-5
Complications With Generators and Lead Advisories
Adapted from Gould PA, Krahn AD, et al: Complications associated with implantable cardioverter-defibrillator replacement in response to device advisories. JAMA 295(16):1907-1911, 2006.
| Severity and Complications | Elective Advisory Device Replacements No. (%) |
|---|---|
| Minor | |
| Infection not requiring system removal/incisional infection | 9 (1.7) |
| Significant site pain, medically managed | 1 (0.2) |
| Heart failure requiring admission | 1 (0.2) |
| Major psychological morbidity, medically managed | 1 (0.2) |
| Major | |
| Infection requiring extraction | 10 (1.9) |
| Death (post extraction) | 2 (0.4) |
| Hematoma requiring reoperation | 12 (2.3) |
| System malfunction requiring reoperation | 8 (1.5) |
| Significant site pain requiring reoperation | 1 (0.2) |
Further studies have examined the threshold at which battery replacement should occur, depending on the clinical scenario. As discussed above, a patient with a primary prevention ICD may require a different approach than a patient who recently suffered an electrical storm, particularly if the device malfunction that might occur under the advisory would result in lack of tachy therapy delivery ( Table 42-6 ). The decision to replace a device must take into account a number of important factors including patient, device, and procedural characteristics. Table 42-6 presents a model derived from utilizing decision analytic methods, where various clinical scenarios are modeled with hypothetical rates of failure due to an advisory. In each of these clinical scenarios, the suggested clinical management (immediate replacement vs. continuous monitoring) is provided. The threshold for immediate replacement requires a significant risk of failure, which far outweighs the risk of failure reported by many of the generator advisories, shown in Table 42-4 . Gula et al used similar methods of Markov modeling and sensitivity analysis, demonstrating that alterations in the risk of lethal arrhythmias does not have a differential effect on the decision of device advisory replacement, but that only the risk of annual device failure was impactful. This study suggested that an annual failure rate of 1.8% is required to confer a survival advantage in device replacement at 7 years. Priori et al derived a number needed to replace (NNR), based on the number needed to treat (NNT) formula, where NNR takes into account the annual risk of sudden cardiac death, the remaining device longevity, the difference in failure rate between the implanted and replacement device, and the mortality risk associated with device replacement. As with the two prior models, Figure 42-4 demonstrates that a high device failure rate is required before the NNR suggests replacement. None of these approaches have undergone further validation in practice, but they do serve as a reasonable guide in making clinical decisions ( Figs. 42-4 and 42-5 ).
TABLE 42-6
Optimal Management Decisions and Probabilistic Outcomes
From Amin MS, Matchar DB, Wood MA, et al: Management of recalled pacemakers and implantable cardioverter-defibrillators: a decision analysis model. JAMA 296(4):412-420, 2006.
| Hypothetical Device Advisory and Advisory Failure Rate * | Scenarios (Device; Indication) † | |||||
|---|---|---|---|---|---|---|
| 1(Pacemaker; Pacemaker Dependence) | 2(ICD; Pacemaker Dependence, Prior SCD) | 3(ICD; Primary Prevention) | 4(ICD; Secondary Prevention) | 5(Pacemaker; First- or Second-Degree AVB) | 6(Pacemaker; Sick Sinus Syndrome) | |
| ICD advisory; 0.25% risk of failure/y | NA | Immediate replacement (58) | Continued monitoring (100) | Continued monitoring (95) | NA | NA |
| ICD advisory; 2.5% risk of failure/y | NA | Immediate replacement (100) | Immediate replacement (55) | Immediate replacement (97) | NA | NA |
| Pacemaker advisory; 0.075% risk of failure/y | Immediate replacement (76) | NA | NA | NA | Continued monitoring (100) | Continued monitoring (100) |
| Pacemaker advisory; 0.5% risk of failure/y | Immediate replacement (100) | NA | NA | NA | Continued monitoring (92) | Continued monitoring (99) |
AVB, atrioventricular block; ICD, implantable cardioverter-defibrillator; NA, not applicable; SCD, sudden cardiac death.
*Hypothetical advisory failure rates were used for the analysis.
† Percentages of simulations in which the stated strategy was the preferred decision (associated with a higher life expectancy) are shown in parentheses.
The clinical implications of these registry-based studies and decision analyses suggest very careful consideration to early intervention in patients with generator advisories and that the risk of this is significant. Prevention of repeated pocket interventions is of utmost importance in the management of patients, where it has been well established that this leads to an increase in device-related infection, which may ultimately require extraction using laser or other tools that carry a 0.28% to 1.86% risk of procedural-related/in-hospital mortality, or 4% risk of significant morbidity. These data are from high-volume centers in the United States and Canada and need to be interpreted with caution where laser-assisted lead extraction may be performed less frequently.
There are currently no specific tools or other indicators that signal that a device under advisory is likely to malfunction. In some instances, the manufacturers may provide a software installation that will assist in identifying the impending malfunction or even correct the underlying problem. In the absence of this, however, remote monitoring has permitted close observation of patients with devices under advisory without increasing cost or inconvenience to either patients or follow-up physicians. The risk of harm due to an advisory is dependent on the nature of the advisory, the individual risk profile, but also the timing of follow-up. The hazard of risk is roughly equal to that seen in half the follow-up interval. More frequent follow-ups translate into less harm. The burden of monitoring can be significant in a device follow-up clinic. Remote monitoring has been proven to be both safe and highly effective in patients with CIEDs.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here