Heart failure currently affects 3–4% of the worldwide population with a prevalence peaking up to 10% after 70 years of age and a constant rise in incidence [1]. Such heart failure epidemic has vigorously stimulated the field of interventional cardiology and boosted the development and widespread use of a number of newer cardiac devices, meant to correct structural or functional cardiac defects. Novel cardiac devices such as MitraClip system (Abbott Vascular, Santa Clara, CA, USA), left atrium appendage occluder devices, septal closure devices, leadless pacemakers (LP), and subcutaneous implantable cardioverter defibrillators (S-ICD), provide safe treatment options for heart failure patients deemed at high-risk for more invasive procedures or open-heart surgery [2, 3, 4, 5, 6]. Unique mechanisms allow for these devices to be less prone to infections, including avoiding of long electronic leads, intravenous access, subcutaneous pockets, or large prostheses and use of novel device materials, as compared to traditional transvenous implantable electronic devices and valve prostheses.

Although supposedly less common, infective endocarditis (IE) involving these newer devices occurs. In light of the steep increase in implantation rates, it appears crucial to properly describe new device-associated infections, and understand the associated burden and whether heightened morbidity and mortality should be expected.

In this review, we aim to summarize the current published evidence on the incidence, clinical characteristics, management, and outcomes of patients who developed IE following implantation of the newer implantable cardiac devices.

MitraClip is a percutaneously implanted cardiac device that allows treatment of severe mitral regurgitation (MR) in patients with annular dilatation and chordal stretching due to dilated cardiomyopathy, and therefore at high risk for open heart surgery [7, 8]. It essentially replicates the edge-to-edge mitral valve repair strategy but with a percutaneous and noninvasive approach (Fig. 1A, Ref. [9, 10, 11, 12, 13]).

Fig. 1.

Novel Cardiac Devices. (A) MitraClip system [9] (via license: Creative Commons Attribution 3.0). (B) Amplatzer® ASD closure device [10]. (C) WATCHMAN LAAO device [11]. (D) Micra® (adapted from [12], Creative Commons Attribution 4.0 International License). (E) Cardiac contractility modulation device (right side) (adapted from [13], Creative Commons Attribution 4.0 International License). ASD, atrial septal defect; LAAO, left atrial appendage occlude.

Incidence of infective endocarditis (IE) following MitraClip was investigated by several studies [6, 14]. In the pivotal EVEREST II trial [14], where patients were followed up for 12 months, 1.1% (2/184 pts) had IE throughout the study period. This was confirmed in subsequent studies which found the infection risk to range between 0 and 1.3% (75% within 1-year after procedure) [6]. However, the exact incidence of MitraClip-associated IE remains unclear and could likely increase in parallel with the rising rates of procedure performance.

The first reported case of MitraClip-associated IE [15], published in 2011, was a 57 year-old man with early onset IE (2 weeks after device implantation) diagnosed by blood cultures positive for Staphylococcus aureus and a mobile lesion on the MitraClip seen on transthoracic (TTE) and transesophageal echocardiography (TEE). Management included targeted antibiotics and, despite the high predicted perioperative mortality (EuroSCORE II 18%), surgical mitral valve replacement, which histologically confirmed IE. Few years later, Frerker et al. [16] described the case of an 88 year-old patient, with high surgical risk (EuroSCORE II 30.4%) and severe MR, presenting with fever and shortness of breath, one month after MitraClip placement. Echocardiography showed large vegetations on the mitral valve within the clip area, complicated by severe MR, and serial blood cultures were positive for S. aureus. Therefore, definite early IE was diagnosed according to the modified Duke criteria [17]. The patient underwent cardiac surgery (logistic EuroSCORE 56.8%) with mitral valve replacement and discharged alive to a rehabilitation facility.

In a review of 12 cases, Asmarats et al. [6] showed that patients with MitraClip-related IE are usually older (median age 76 years) than patients with valvular IE (median age 62 years) [18], and present a higher burden of comorbidities [19]. Indeed, this procedure is indicated in patients at high preoperative surgical risk. Undoubtedly, the mentioned clinical features influence the high mortality associated with MitraClip-related IE, as they are reported to be among the predictors of poor outcome [20]. Upon pooling of available literature data, we were able to appreciate that the clinical presentation of MitraClip-related IE is dominated by classical symptoms of systemic infection (fever in 15/25 pts; 60%) together with signs and symptoms of decompensated heart failure (12/25 pts; 48%) [6, 15, 16, 21, 22, 23, 24, 25, 26], resulting from mitral valve recurrent regurgitation or stenosis. Less common presentations included complete heart block (one case) [27], whilst septic emboli seldom occurred (3/25 pts; 12%) [28, 29].

MitraClip-related IE occurred early (12 months) in most reported cases (18/25 pts, 72%; 75% according to Asmarats et al. [6]). This suggests that the infection could be often acquired perioperatively. Indeed, it has been demonstrated that implanted MitraClips go through a physiological healing process of fibrous encapsulation that completes within around 300 days [30, 31], after which the risk of pathogens’ seeding appears to be significantly reduced.

Diagnostic criteria do not differ from those currently employed for valvular IE [17] and are based on echocardiography findings and blood cultures. However, alternative imaging technique as ${}^{18}$F-fluorodeoxyglucose PET/CT could be appropriate in selected cases [29]. S. aureus is the major causative microorganism (11/25 pts; 44%). However, infections of MitraClip caused by Enterococci (3/25 pts, 12%), Streptococci (2/25 pts, 8%) [6, 32], Pseudomonas aeruginosa [33] or atypical microorganisms such as Bartonella haenselae [34] are also reported.

Treatment of MitraClip IE is challenging and still debated. Surgical mitral valve replacement was needed in 56% of patients despite the extremely high calculated preoperative mortality risk (mean logistic EuroSCORE of 41% [6]). A conservative approach, based on antibiotic therapy only, can be considered in selected patients; especially those with coagulase negative staphylococcal etiology [35]. The mortality associated to the IE episode is around 52%, higher than prosthetic valve IE (20–40%) [20].

In conclusion, MitraClip-associated IE is a very uncommon but potentially fatal complication affecting patients with advanced age and multiple comorbidities. It often requires surgical treatment, but the optimal approach is still unclear and needs to be devised on an individual basis.

Closure or Occluder devices are an increasingly implanted class of devices [36]. These include devices implanted to close atrial or ventricular septal defects (Fig. 1B) and devices used to occlude the left atrium appendage in patients with atrial fibrillation (Fig. 1C).

Leadless pacemakers (LPs) are new cardiac implantable electronic devices approved for the treatment of bradyarrhythmia in specific settings such as in patients with risk of transvenous device complications due to old age, significant comorbidities, or high infection risk [5]. Nanostim® LP (St. Jude Medical/Abbott, IL, USA) and Micra® Transcatheter LP (Medtronic, MN, USA) are the two available types of LPs (Fig. 1D), although only the latter is currently used. Both devices are implanted into the right ventricle (RV) using their own delivery system through the femoral vein (Fig. 2, Ref. [84]).

Fig. 2.

Micra delivery system [84].

Two cases of definite LPs infections are reported in the literature. The first case was described by Koay et al. [85], and was an 80-year-old woman admitted with fever, chills, and rigors one month after Micra® implantation due to junctional bradycardia. Diagnosis was done by blood cultures, which grew methicillin resistant S. aureus (MRSA), and TEE, which revealed device vegetations. Treatment initially included targeted antibiotic therapy for one week, but due to persistently positive blood cultures, the device was later removed. Afterwards, the patient received 6 weeks of antibiotic treatment and the infection was resolved. A histopathological analysis of the device vegetation revealed findings consistent with infection (device culture not available) [85].

The second case was an 80-year-old man who developed pocket erosion, without evidence of blood stream infection, nine years after dual-chamber pacemaker (PM) implantation for sick sinus syndrome and atrioventricular block. Subsequently, the generator was removed leaving the leads behind, abandoned under the skin, and a Micra® was implanted into the right ventricle. One month later, the patient had recurrence of the erosion and cultures grew MRSA, with no vegetations visible on TEE. An emergency extraction of Micra® and the leads was done due to failure to respond to antibiotics, and culture of the device and leads grew MRSA. Treatment also included a 6 week course of antibiotics and, after 10 days of documented negative blood cultures, implantation of a cardiac resynchronization therapy device, with no recurrence of infection at nine-month follow up [86].

Initial studies with LPs did not detect any device infections on follow-up [87, 88], which was also the case in subsequent larger observational studies [89, 90, 91, 92]. LPs were safely implanted in high infection risk settings, as in patients with active infections—including CIED pocket infection—and positive blood cultures [93, 94]. Other settings where LP were apparently successfully used were patients who developed a serious infective event after other device implantation [95], patients on chronic hemodialysis [96], nonagenarians [97] or those with a heart transplant [98]. LPs implantation after cardiac implantable electronic device (CIED) infection, was successful and reduced recurrence rates [99, 100, 101, 102, 103]. Consistently, in the few studies comparing infective complications in LPs and PM patients, PM showed higher rates of these complications [104, 105, 106].

Several factors are thought to account for the safety of LPs. Some studies suggest that their small size ($\sim$616 mm${}^2$ and $\sim$846 mm${}^2$, in Micra® and Nanostim®, respectively, compared to $\sim$3500 mm${}^2$ in PM), independence of pockets, and complete and fast encapsulation within an endothelial biomatrix, play an important role in their safety. Other alleged factors include the minimal handling of LPs during implantation, the high velocity and turbulence of blood flow in the RV, and the device constituents themselves (parylene-coated titanium and titanium vs titanium polyurethane and silicone in LPs and PM, respectively) [107]. A very likely additional factor is represented by the absence of any anatomical contiguity with the tricuspid valve leaflets, which generate turbulence and likely heamostasis system activation at the site of CIED lead passage through the annulus.

Some limitations of the reported studies with LPs include shorter time of follow-up compared to transvenous CIEDs [108]. In some reports, device culture was not documented after extraction, and device infection was often excluded solely based on visual inspection and/or no vegetation on TEE. Thus, it is not possible to exclude that some LPs infections were missed, implying more evidence is needed to ascertain the safety of LPs on longer term follow-up.

In conclusion, currently available evidence suggests that LPs are cardiac implantable electronic devices with a low risk of infection and can be considered very useful and safe especially in patients at high risk of infections, after CIED extraction for infection in PM-dependent subjects and when limited access to/patency of subclavian veins is an issue.

Cardiac contractility modulation (CCM) is a novel device-based modality developed for chronic heart failure treatment, and mainly functions by delivering non-excitatory signals during the absolute refractory period of the heart [127, 128, 129]. The device includes an impulse generator subcutaneously placed in the left or right pectoral region, connected by two transvenous leads to the right ventricular septum, with or without an additional lead to the right atrium (Fig. 1E) [130, 131, 132]. CCM was shown to increase left ventricular contractility and exercise tolerance in select advanced heart failure with reduced ejection fraction (HFrEF) patients, without increasing myocardial oxygen requirement [130, 133, 134].

In an early pilot study, thirteen patients with advanced HFrEF received a CCM device; after a mean of 7 months, only one patient (8%) experienced a pocket infection and was subsequently treated successfully with antibiotics [135]. A recent systematic review of four clinical trials found 9 cases of device pocket infection out of 469 included participants (2%), after a mean follow-up of six months [136]. One study with longer follow-up retrospectively included 54 HFrEF patients with CCM device and found 6 participants (11%) with device-related infection, over 7 years [137]. Most of these infections started with signs of pocket skin necrosis and advanced to bacterial infection, and caused death in half of affected patients [137]. Furthermore, a recent clinical trial (the FIX-HF-5C2 study) investigated the safety and efficacy of the two-lead system and reported one case of infection among 60 participants over a 6-month period [138]. Other studies with CCM did not show increased adverse events, including infections [134, 139, 140, 141, 142, 143].

Overall, CCM associated infections seem to be uncommon in the short term, and apparently readily treatable with no major serious sequelae. However, further investigation of the devices’ safety is warranted, since currently available studies are still limited, most notably regarding long-term outcomes ($>$6 months) [136]. In addition, differences in the methods of current studies were noted, including lacking blinding and/or sham procedures in some [136, 142]. Thus, we believe future studies must include proper control groups to better define the CCM-related infection risk. Likewise, most available studies do not thoroughly describe the infection, including microbiological culture results, clinical signs and symptoms and management course, which are key for infection risk investigation.

Despite of their favorable cardiovascular safety profile, novel cardiac devices such as MitraClip, occluder devices, LPs, and S-ICD can undergo infection at virtually any time after implantation. As signs of device infection are non-specific and mostly include fever, this diagnosis should be ruled out whenever these patients present with non-organ specific signs of systemic infection. Although most patients did not suffer any long-lasting complication and were discharged alive after treatment, some suffered from complications such as stroke, and others died. Apart from the newer ICDs, cardiac modulation devices and inter-atrial shunts, new cardiac device infection mortality seems to overall overlap with classical prosthetic valve IE [144], and appears to be significantly less than Transcatheter Aortic Valve Implantation (TAVI)-related IE [145]. Further studies should investigate risk factors for the development of infections with these devices and compare medical and surgical management to improve the prognosis. Diffuse reporting of even single case reports appears crucial in light of the overall low rate of new cardiac device infections, so to allow systematic literature review and advancement of knowledge in this emerging field of cardiovascular medicine and clinical infectious disease.

We believe interventional cardiologists should be well aware of the infection risk associated with new cardiac device implant procedures since the initial placement. They should prescribe right away blood cultures whenever lone fever occurs in carriers of these devices. Infectious disease physicians should become acquainted with newer cardiac device structure and function to adequately treat these emerging infections.

Conceptualization—MR, FP, RG, LB and EDM. Original draft writing and data curation—MR, FP, RG and LB. Review and editing—MR and EDM. Figures—MR and EDM. All authors have read and agreed to the published version of the manuscript.

Not applicable.

Not applicable.

This research received no external funding.

Emanuele Durante-Mangoni declares funding via his Institution from MSD, Pfizer, and personal fees or participation in advisory boards or participation to the speaker’s bureau of Roche, Pfizer, MSD, Angelini, Advanz Pharma, Infectopharm, Bio-Merieux, Abbvie, Sanofi-Aventis, Menarini, Medtronic, Tyrx and DiaSorin. Other authors declare no conflict of interest.