Endocarditis related to cardiovascular implantable devices

Introduction

Cardiac implantable electronic devices (CIEDs), including the permanent pacemaker (PPM), implantable cardiac defibrillator (ICD), and cardiac resynchronization therapy (CRT) devices, provide life-saving intervention for a variety of clinically indicated cardiac conditions. However, their associated short-term and long-term outcomes are not without risk. One serious complication resulting in significant morbidity and mortality is device-related infection (CIED infection). The clinical presentation of CIED infection includes a spectrum of both native and foreign tissue involvement as follows: superficial incisional infection, pocket infection, pocket erosion, bacteremia with or without signs of pocket infection, and/or echocardiographic evidence of lead involvement. Occult bacteremia without an obvious alternative source is also accepted as probable CIED infection.

While positive blood cultures with the identification of lead and/or valvular vegetations by echocardiography establish a diagnosis of CIED-related endocarditis (CIED-RE), additional criteria are often called upon to help define the diagnosis when the presentation is less overt. In a recent guideline document by the European Heart Rhythm Association and endorsed by multiple other scientific bodies, criteria were expanded to include: positive cultures of the extracted lead in case of negative blood cultures, presence of lead vegetations on echocardiography with or without valvular vegetations, abnormal metabolic activity around the CIED generator, and/or leads detected by 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET)/computed tomography (CT) or radiolabeled leukocytes single-photon emission computed tomography/CT ( Table 12.1 ) [ , ].

CIEDs have the capacity to act as a nidus for infection thereby increasing the risk for the development of infective endocarditis. While the rate of device implantation continues to rise as device technology and capabilities advance, a recent study identified cardiac device involvement in 7% of all infective endocarditis cases [ ]. This highlights the value of evidence-based guidelines to underscore best practices for the prevention, identification, and management of this serious complication. In this chapter, we review the various clinically relevant themes concerning CIED-RE recognizing the continuum of presentation of CIED infection as mentioned above. The epidemiology, diagnosis, and microbiology will be explored as well as the current clinical evidence, and its gaps, supporting the most up-to-date guidelines for the management and prevention of such infections. To illustrate the challenges associated with the condition let us consider a hypothetical albeit typical case encountered in modern practice.

Epidemiology

The incidence of infective endocarditis remained relatively constant from 1950 to 2000 at approximately 3.6–7 cases per 100,000 patient-years [ ]. In recent years there has been a change, however, with studies demonstrating an increase in the incidence of infective endocarditis in patients not previously known to have predisposing heart conditions [ ]. This is most likely due to an increase in invasive procedures, prosthetic valves, and CIEDs all recognized as potential predisposing factors by virtue of their violation of normal body barriers and invasion of the intravascular space [ ].

The CIED era began in the 1960s with the development and use of PPMs and subsequently expanded to include ICDs, and CRT devices [ ]. Between 2000 and 2012, the Danish Pacemaker and ICD Register (DPIR) reported an increase of 158% and 535% in the implantation of PPMs and ICDs, respectively [ ]. This evolution is a consequence of growing indications for various cardiac conditions with improved medical technological capability to treat an aging population frequently beset with significant medical comorbidities [ ]. The incidence of CIED-RE has varied across studies [ ]. CIED-RE was first reported in the early 1970s in the context of PPMs and has been growing in incidence ever since. In one study, CIED-RE accounted for approximately 10% of all cases of infective endocarditis [ , ]. A common clinical observation has been borne out in at least one prospective cohort study that found the majority of cardiac device infections involve the subcutaneous generator pocket, of which, 10%–23% resulted in CIED-RE [ ]. These results were in agreement with a retrospective study that reported 20% of cases resulted in CIED-RE [ ]. CIED-RE has risen in recent decades; a longitudinal study conducted over a period of nearly 30 years showed that the number of CIED-RE cases increased significantly in the last decade in parallel with an increase in CIED implantation [ ]. Carrasco et al. reported PPM endocarditis accounted for 6.1% of all cases of infective endocarditis and affected 3.6 per 1000 of all implanted pacemakers [ ]. Ominously, as reported by multiple investigators, the rate of CIED-RE seems to have outstripped implantation rates [ ].

CIED-RE most often occurs early after implant. In a nationwide cohort study of first-time implants including 43,048 patients and 168,343 patient-years follow-up, the incidence of early-onset (presenting within the first year) CIED-RE more than doubled that of late onset CIED-RE across all types of implants [ ]. Early-onset CIED-RE, when defined as presenting during hospitalization or within 3 months following discharge, is attributable to intraoperative contamination during the index device implantation procedure. CIED-RE carries a high mortality of 24% for early-onset endocarditis in one report [ ]. In another, mortality rates ranged from 31% to 66% without device removal, which decreased to 18% with CIED removal and intensive medical care [ ]. The same authors reported that infective endocarditis and ejection fraction were the strongest predictors of in-hospital mortality in patients with CIED infection. Of note, while seasonal bacteremia has been described in the literature, Maille et al. report a seasonal variation with respect to pocket infection (with or without endocarditis), but not endocarditis alone [ ]. Lastly, CIED infections including CIED-RE impose a considerable financial burden on the health-care sector [ , ]. In-hospital costs are estimated to be at least $146,000 per case in the United States [ ].

Patient factors

As device implants have increased in frequency over the past two decades, particularly among an older and sicker population, a variety of patient factors have contributed to the increased incidence of CIED-RE. Comorbidities associated with a higher risk of infection include: diabetes mellitus, renal insufficiency, particularly end-stage renal disease, malignancy, congestive heart failure, chronic obstructive pulmonary disease (COPD), and use of anticoagulants and/or immunosuppressive drugs, especially corticosteroids [ , , ].

A recent prospective study identified the following host factors predisposing patients to device-related endocarditis: malnutrition, malignancy, diabetes mellitus, skin disorders, and the use of steroids and/or anticoagulants [ ]. Additional studies establish diabetes mellitus, dialysis, use of vitamin K antagonists, a history of infective endocarditis, CIED infection, and valve disease as conditions associated with increased risk [ , ]. Concomitant chronic conditions impair the ability to mount an appropriate immune response, thereby setting up a backdrop which predisposes patients to the development of CIED-RE [ ]. The use of vitamin K antagonists increases the incidence of postoperative hematoma, which is recognized as an independent risk factor for infective endocarditis [ ].

CIED-RE was the third most common (16%) presentation of CIED infection in a study conducted by Sohail et al. [ ]. The study reported that chronic dialysis, COPD, and skin condition had statistical significance with ICD infections. However, other conditions including diabetes mellitus, heart failure, malignancy, or immunosuppression therapy were not statistically significant risk factors [ ].

One recent study reported 82% of patients presenting with ICD infection had evidence of pocket infection, while 18% had evidence of systemic infection, only 14% presented with electrode and/or valvular vegetations [ ]. When comparing patients with systemic infections to patients with pocket infections, the presence of a femoral venous catheter was found to increase the risk of systemic infection [ ]. Further, patients with ICD infections were found to be older than those without PPM/ICD infections [ ].

While some studies found elderly men more likely to present with CIED-RE, others found no significant difference [ , , ]. Ozcan et al. in a study of a large national registry reported CABG to be associated with decreased infective endocarditis risk [ ]. This low risk could be explained by a selection bias favoring CABG patients; sicker patients were less likely to undergo surgery. Patients with CIED endocarditis were more likely to be male, of advanced age, and diabetic compared to patients with infective endocarditis without cardiac device [ ]. With increasing longevity of patients with significant comorbidities and their high risk for CIED infection and CIED-RE, the assessment of comorbidities at the time of CIED implant becomes critical.

Presence of skin lesions, poor dentition, or indwelling catheters provides ingress points for organisms as does bowel inflammation or neoplasm. Skin or gut colonization with drug-resistant or particularly virulent organisms pose respective threats as skin microbes are the most commonly implicated offenders [ ]. The latter consideration in turn should serve as a reminder against indiscriminate use of antibiotics to prevent emergence and perpetuation of resistant strains. Screening nasal swabs for methicillin-resistant Staphylococcus aureus (MRSA) and treating carriers with mupirocin ointment are established surgical practices with demonstrated positive effect [ ].

Device factors

Irregular surfaces on CIEDs provide a safe harbor for infecting microorganisms. Infecting microorganisms create a biofilm which effectively shields against antimicrobial agents. While different microorganisms are more or less effective at creating this biofilm, device and lead surfaces prove more or less hospitable to the same. Under electron microscopy, lead insulation material demonstrates an irregular surface of crypts and valleys that provide a natural habitat for microorganisms to multiply and spread. The continuous biofilm provides scaffolding for the microorganisms to migrate along the leads such that a pocket infection routinely extends into the intravascular portions of the leads [ ]. Microorganisms can never be presumed confined to one lead but not others. Gram-positive bacteria are most likely to be isolated from infected CIEDs (70%–90%). While coagulase-negative staphylococci (CoNS) species are normally encountered as harmless skin colonizers, once adherent to devices they can colonize in sufficient numbers to become invasive infectious agents. Staphylococcus species are the most likely causative agents for bacteremia with almost equal distribution among methicillin-resistant and sensitive strains in some series. The increase in methicillin-resistant species noted in recent years is a worrisome trend. Insulation materials surrounding leads are more likely to provide microbes with a hospitable surface than metal (usually titanium) encasing the generator [ , ].

Other device factors impact patient's risk of developing CIED-RE. These include device type, device size, and lead characteristics. The most consistent device factors that increase risk of infection include the implantation of an ICD rather than a PPM, implantation of two or more leads, and the use of intraoperative temporary pacing [ ].

Device factors, such as the type of plastic polymer, irregularity of its surface, and its shape, can affect bacterial adherence to the device [ ]. The hydrophobic nature of the plastic polymers increases bacterial adherence with the degree of hydrophobicity. Polyvinyl chloride favors more adherence than Teflon, polyethylene more than polyurethane, silicone more than polytetrafluoroethylene, and latex more than silicone; some metals (e.g., stainless steel) favor adherence more than others (e.g., titanium) [ ]. In addition to the coating, irregular device surfaces favor microbial adherence more than a smooth surface.