Introduction

Cardiac implantable electronic devices (CIEDs) play an important role in the management of cardiac arrhythmias and the prevention of sudden cardiac death.¹ As the indications for primary prevention and cardiac resynchronization have expanded alongside an aging population, CIED implantation has increased worldwide over recent years.^2,3

Although technological advances have improved both the devices and implantation procedures, cardiac implantable electronic device infections (CIED-Is) remain a significant problem.⁴ These include pocket infection, systemic infection, and infective endocarditis, each of which can be life-threatening.^5-7 The Heart Rhythm Society and the European Heart Rhythm Association (EHRA) have published guidelines for the management of leads and devices, including recommending extraction for CIED-I.^5,8 Despite multiple recommendations on prevention, diagnosis, and management strategies, discussions about barriers to optimal treatment and guideline adherence in clinical practice are limited.

The objective of this review is to summarize the literature to provide a comprehensive overview of CIED-Is and management. A structured literature review was performed in MEDLINE via the PubMed interface on October 21, 2020, for papers published in the last 15 years to inform this review. Titles and abstracts identified by the search were reviewed for eligibility based on inclusion criteria. Publications on CIED-I rates, outcomes of management strategies, associated clinical or economic burden such as health-related quality of life, and cost of infection were included. Furthermore, clinical practice guidelines on CIED-I management were captured in addition to papers discussing trends in clinical practice, adherence to guidelines, and/or barriers to optimal treatment. Search strategy and search results can be found in the Supplemental Appendix.

The review will systematically characterize the following: 1) the epidemiology and burden of CIED-Is; 2) clinical outcomes data on various treatment strategies and health care utilization research CIED-Is; 3) an overview of current guidelines on CIED-I treatment; 4) the adherence to consensus and guideline recommendations; and 5) potential barriers to the recommended treatment. We conclude the review with future directions and possible solutions to address barriers to optimal treatment.

CIED-I Epidemiology and Clinical Impact

CIED-Is can lead to serious systemic complications. They can present either as a pocket infection in the chest, typically diagnosed clinically, or a systemic infection, which is confirmed with positive blood cultures. CIED-I rates across the world range from 0.8% to 4.2%.^4,9-19 Overall, CIED-I rates within a year of implantation range from 1.2% to 3.4%.^11-14 A U.S. cohort study using the National Inpatient Sample database found that 8,060 (4.2%) of 191,610 CIED implantations were admitted for CIED-Is in 2016.¹⁸ Among a Danish CIED population (128,045 devices), the overall incidence of CIED-I was reported to be 1.43%.¹⁶ A complete summary of infection rates can be found in Table 1.

The WRAP-IT (World-wide Randomized Antibiotic Envelope Infection Prevention trial) multicenter RCT evaluated the clinical and economic burden of CIED-I from the hospital, payer, and patient perspectives. Patients were found to have significantly reduced quality of life at the time of infection diagnosis vs baseline (0.83 ± 0.14 vs 0.75 ± 0.19; P = 0.004). Furthermore, patients with infection had a significant increased risk of death compared with the no infection group (risk-adjusted HR: 3.41; 95% CI: 1.81–6.41; P < 0.001).²⁰ Similarly, the Italian POINTED (Impact on Patient Outcome and health care utilization of cardiac ImplaNTable Electronic Device complications) Registry suggested that CIED-I is associated with poor survival compared with patients with other complications or those with none (P < 0.001).¹⁷

CIED-Is may be related to various patient, device and procedural factors (Central Illustration). Patient factors that may increase the risk of CIED-I include renal dysfunction, diabetes, and younger age.^13,14,16,19 Oral anticoagulation use and prior CIED-I have also been reported as risk factors.^14,16

Economic Burden of CIED-I

The economic burden of CIED-I varies depending on the region. A retrospective analysis of a large U.S. health insurer database showed that CIED-I increases the average annual medical costs by 2.4 times, with a 1-year adjusted incremental expenditure of $57,322 per patient (95% CI: $46,572-$70,484; per patient; P < 0.001).¹³ The WRAP-IT trial further reported the mean payer costs per infection was $57,978 ± $29,431 for Medicare Advantage and $26,867 ± $14,893 for Medicare Fee For Service.²⁰ In Germany, a retrospective analysis of health claims data reported €15,822 higher costs in the year after CIED-I compared with the year prior to infection (excluding the cost of index device implantation).¹⁴ The higher costs were approximately 98% driven by inpatient hospital care.¹⁴ A Korean analysis of their National Health Insurance database found that the average cost per person associated with CIED-I was U.S.$17,105,¹⁹ which was similar to the United Kingdom (£14,742).²¹ Regional variations of infection costs per patient is shown in Figure 1.

Clinical Outcomes and Health Care Utilization for CIED Extraction

Optimal CIED-I management includes device extraction and is a highly successful procedure. For safety and effectiveness, there were 16 studies reporting on outcomes for patients where the entire study cohort (100%) underwent extraction for the indication of CIED-I. Procedural success was consistently reported to be very high, with most studies reporting rates of >95% (Table 2). Also, procedural complication rates (range 0.0%-4.0%) and mortality rates (range 0.0%-1.0%) were consistently low across all studies^32-51 (Figure 2).

CIED extraction, when indicated, can be a potentially life-saving procedure. A retrospective study of CIED-I cases from a large academic center between 1991 and 2008 found that antimicrobial therapy without device extraction was associated with nearly a 7-fold higher 30-day mortality (HR: 6.97; 95% CI: 1.36-35.60).³⁸ Additionally, patients with CIED extraction for infection had a 1-year mortality rate of 13.3%, whereas those who were treated with antibiotics only (n = 23) had a 1-year mortality rate of 38.1%.³⁸ Similarly, in an international study examining endocarditis patients, there was an 18.3% difference in 1-year mortality between patients who had their device removed for CIED-I compared with those who did not (19.9% vs 38.2%; P = 0.02) (Figure 3).⁵²

Timing of CIED extraction, early vs delayed extraction, after infection, also reveals significant differences in 1-year mortality (Figure 3). When extraction was conducted at the initial infection presentation (n = 370), 1-year mortality was significantly lower than those who had extraction after failure of antimicrobial therapy (n = 23) (11.4% vs 43.4%; P < 0.001).³⁸ A retrospective analysis of all consecutive CIED-I patients who underwent extraction at a single U.S. university hospital found similar results. For patients who had extraction within 10 days of diagnosis, a 1-year mortality of 16.9% was reported compared with 33.9% in patients who had extraction after 10 or more days (P = 0.028).⁴⁰ A 2020 study examining the impact of early vs delayed extraction with infected CIEDs found similar trends of significantly greater 12-month survival rate in patients with bacteremia (P = 0.022) and isolated pocket infections (P = 0.027) when CIED-I extraction occurred early (within 7 days from hospital admission).⁵³ Last, 4 studies reported on reinfection rates associated with complete vs incomplete extraction. Three large single-center studies (2 retrospective and 1 prospective) reported significantly lower rates of reinfection when complete extraction occurred for CIED-I (1.0% vs 50.0%; P < 0.001; 4.3% vs 20.8%; P = 0.006; 5.3% vs 21.4%; P = 0.007). One (retrospective) study failed to report statistical testing for significance.^39,54-56

A recent analysis of a Medicare population from 2004 through 2019 provided updated information on device infection and all-cause mortality of these patients.⁵⁷ CIED-I incidence was 1.1% and only 25% (n = 2,814) of these patients had extraction.⁵⁷ Cumulative 1-year mortality was 32.4% for patients without extraction at 30 days, compared with 18.5% among patients with extraction within 6 days (P < 0.001).⁵⁷ Furthermore, any extraction was associated with lower mortality when compared with no extraction (adj HR: 0.73; 95% CI: 0.67-0.81; P < 0.001); and extraction within 6 days was associated with even lower risk of mortality (adj HR: 0.59; 95% CI: 0.52-0.67; P < 0.001).⁵⁷ Although impactful, analysis from the Medicare database must be used with caution as there are limitations, including the retrospective nature of the data and underrepresentation of the number procedures.

In general, the evidence has demonstrated that prompt CIED extraction is associated with favorable health care resource utilization relative to alternatives. A large U.S. payer database analysis estimated that incremental health care expenditures for patients with an infection managed by inpatient admission with no extraction cost more than double the amount compared with infections managed by extraction in either an inpatient or outpatient setting ($104,077 vs $45,291).⁵⁸ Furthermore, Rungpradubvong et al⁴⁰ found that CIED-I patients with early removal (within 10 days of infection diagnosis) were associated with significantly shorter hospital stays compared with their delayed extraction counterparts (17.59 ± 12.64 days vs 43.77 ± 37.77 days; P < 0.001). In addition to shorter hospital stays, early extraction was associated with lower mean all-cause Medicare costs within 12 months postinfection, with extractions occurring on the same day as diagnosis costing an estimated $34,640 vs $76,836 for extractions occurring ≥180 days from infection diagnosis.⁵⁹

Although the evidence presented supports timely CIED extraction, the context must be considered. Many of these studies are not randomized. Thus, delayed extraction cohorts are more likely to have older, frail patients, including those with end-stage diseases. CIED extraction is not without risk; associated major perioperative adverse events can include cardiac arrest/perforation, dissection, and pericardial tamponade. Also, studies are often limited to specific subgroups and may not be directly comparable. Furthermore, many of these studies may include outcomes from experienced centers with high extraction volumes. Although the positive outcomes reported provide confidence of the procedure, they fail to provide a comprehensive overview of extraction outcomes across all centers and individuals with varied experience.

Guideline Recommendations and Real-World Clinical Practice Patterns

Five current guidelines from professional societies provide recommendations on CIED lead management and extraction: American Heart Association, British Heart Rhythm Society, European Society of Cardiology, Heart Rhythm Society, and EHRA. All report a Class I indication for complete system removal in patients with defined CIED-I (systemic, local, bacteremia, or infection endocarditis). Three of these guidelines explicitly recommended prompt and complete system removal for definite CIED-I.^5,7,8,60,61 A summary of the guideline recommendations for CIED-I can be found in Table 3.

Despite these guideline recommendations, a recent analysis of the U.S. Medicare database demonstrated that >8 of 10 patients were not treated according to Class I guidelines for CIED-I (full system extraction).⁵⁷ Additionally, a global survey of members of 7 arrhythmia societies found regional disparities in the clinical practices for the management of CIED-I. The survey revealed that all regions did not fully comply with current guidelines recommendations.⁶² It found that only 62% of responders would manage CIED pocket infection with complete system removal and 13.7% would complete no action and wait until the next regular visit.⁶² The European adherence rate (68.9%) was similar to a Germany single-center experience over 4 decades in which only 57% of CIED-Is were treated with system extraction.⁵⁶ Furthermore, the physician survey rate in the Asia/Pacific region (41.2%) aligned with 2 other Japanese surveys. One reported a 55.8% annual system removal rate in 2013, and a 2018 web-based survey of 155 Japanese cardiologists reported that 30% of CIED-I cases were treated by entire system removal.^15,63

Barriers to CIED Guideline Adherence

A 2020 physician education needs survey was conducted by the EHRA/European Society of Cardiology (ESC) to help assess guideline adherence and potential barriers in effective CIED-I management in clinical practice.⁶⁴ The 3 main barriers were as follows: 1) identification of the CIED-I; 2) prompt referral; and 3) access to extraction. Identification barriers often reflect gaps in physician knowledge and skills in CIED-I management (25% of extractors and more than 60% of nonextractors reported the need for improvement).⁶⁴ Referral barriers included factors such as proximity to an extraction center (reported by 18% of responders), ease of access to an extraction center (reported by 37% of responders), or fear of losing a patient to an extracting physician (reported by 14% of responders).⁶⁴ Other factors included bed constraints, cost, and perceived lack of a diagnosis. Additional factors associated with extraction barriers included patients’ comorbidities, with 92% of responders reporting this affected their choice; age of the patients (reported by 83% of responders); and age of the lead (reported by 73% of responders). Responders were asked what factors may limit a patient from being considered for lead extraction: 77% reported a high risk of procedural mortality, and 44% reported the perceived difficulty or complexity of the procedure (Table 4).⁶⁴

Barriers to identification may be caused by lack of optimal communication and coordination. These patients are often seen by specialties other than their implanting electrophysiologists (ie, hospitalists, primary care physicians, general cardiologists, and so on). Furthermore, symptoms of CIED-I can mimic other infections and illnesses and can be overlooked or misdiagnosed. Examples of CIED infection are shown in Figure 4. Referral barriers and patient factors seem to largely influence the decision of whether to use total device extraction.

Addressing Barriers to Appropriate and Effective CIED-I Management

Because barriers to optimal treatment for CIED-Is are multifactorial, they may be addressed by various approaches and tools. Identification barriers largely stem from the lack of physician knowledge and awareness that can be mitigated with evidence and education. Another strategy is the incorporation of an electronic medical record alert system in which an automatic alert is sent to responsible physicians when a positive culture is registered in a CIED patient’s medical record. An observational study examining the impact of an electronic medical record alert system in individuals with CIED-I demonstrated positive trends in improvement in the time to evaluation and device removal compared with before implementation of the system.⁶⁵ Many of the cited referral barriers to extraction cannot be easily addressed with one solution, because they involve physicians’ perceptions, preferences, and misconceptions. For example, it is unclear why patients diagnosed with CIED-I would be managed conservatively and why treating physicians are reluctant in seeking expert opinion and a more definite therapy. In many cases, CIED-Is are managed by teams that are not experts in CIED-I management. Creating a team structure that can accurately identify individuals with CIED-Is and efficiently channeling them to appropriate expert care that includes infectious disease specialists and device extraction specialists in a timely fashion could dramatically improve outcomes. For these patients, an explanation about the details of the extraction procedure as a therapy option and potential complications if such a procedure is delayed should be offered and adopted into standard clinical practice. To help address such barriers, a deeper understanding of how these reasons vary across setting, region, and physician type, as well as the relative dominance of such reasons, is key to building on education and resourcing platforms that can help to solve these gaps. It must be considered that untreated, inadequately treated, or delays in optimal treatment for CIED-Is can have consequences not only for the patient but also for the health care system in terms of resource and cost burden. This further supports the need to provide evidence and education to help inform physicians (and health care administrators) to prioritize early complete extraction for CIED-I as indicated by clinical guidelines. A quality initiative with a framework to increase guideline-driven care for patients with suspected CIED-I at hospitals may also address the gaps. Even though there is no existing concept of a center of excellence for lead extraction (ie, how many extractions should be performed), it is important that one must be created with established criteria. As with most implementation challenges, successful interventions need to be multifaceted to address the various barriers to optimal care. These interventions should be directed at the patient, physician, and system level.

Conclusions

CIED-I continues to be a serious and continued problem in clinical practice. Adherence to a guideline-based CIED-I management strategy including appropriate device extraction seems to be low. Barriers related to identification, referral, and perceptions around the extraction procedure are widely prevalent. Addressing knowledge gaps and overcoming the identified barriers in building effective CIED-I management teams and referral networks for timely and appropriate interventions including device extraction could improve the outcomes.

Footnotes