Keywords
cardiac implantable electronic devices - left ventricular devices - infection - salvage
therapy - antibiotic beads
Since their earliest application, cardiac implantable electronic devices (CIEDs),
such as pacemakers, implantable defibrillators, and left ventricular assist devices
(LVADs), have proven beneficial in the prevention of fatal cardiac-related disorders.[1] The past 50 years have seen a sharp rise in cardiac implantable device use with
a resultant decline in deaths from ischemic, myocardial, and cardiac rhythm causes.[2] As the indications and guidelines governing CIED use broaden, so too does their
widespread employment.[3] The rate of CIED infections has mirrored their increased use with some studies suggesting
that the rate of device infections has overshadowed their implantation rate.[4] Infections pose a severe burden on patients, lead to significant health care costs
and lengthy hospital stays, and may also lead to mortality.[5] When compared with noninfectious cardiac device complications, pacemaker infections
result in an 8.4- to 11.6-fold increase in mortality rates along with a mean hospitalization
cost ranging from $31,149 to $55,003.[6 ]
Although both pacemakers and LVADs are implantable cardiac devices, their infection
profile and treatment differ significantly due to the size of the device and the need
for an external power source for the LVAD. However, both demonstrate a wide range
of infection rates, with the true incidence of infection remaining elusive.[7] Topkara et al report a pacemaker infection rate of between 13% and 80%; however,
others estimate it to be between 2% and 4% with rates rising 124% between years 1990
to 1999 and a 57% rise from 2004 to 2006, respectively.[8]
[9] Similarly, infection rates related to LVAD placement demonstrate a large range of
between 13% and 80% among recipients.[10]
This wide variability in infection risk is in part due to different types of infections
that have been included under the category of CIED-related infections. Various reports
broadly included patients with surgical site infections, postoperative pneumonia,
central venous catheter–related sepsis, and nosocomial urinary tract infections, in
addition to infection of the CIED.[10] Various comorbidities may contribute to CIED infections. Patients of advanced age,
with congestive heart failure, with a metastatic malignancy, on corticosteroid therapy,
or with renal failure are more likely to develop CIED infections thereby increasing
their mortality.[9 ]
Device Infection Diagnosis
Device Infection Diagnosis
Pacemakers
The diagnosis of pacemaker infections is often challenging. Pocket site infections
are diagnosed clinically, often presenting with inflammatory skin changes including
pain, swelling, and redness. There may be skin and soft tissue ulceration and drainage.
The first sign of infection may be erosion through the skin at the site of the implant
pocket, with external exposure of the device with or without local inflammatory changes
([Fig. 1]).[11] Fever and other signs of systemic toxicity are frequently absent; however, infective
endocarditis may be present. A diagnosis of pacemaker infective endocarditis is based
on clinical parameters, blood cultures, and echocardiographic findings.[12] In cases of noninfective hematoma or seroma, device salvage may be undertaken ([Fig. 2]).
Fig. 1 An 82-year-old man with an exposed pacemaker with necrotic skin.
Fig. 2 (A) A 46-year-old woman with a hematoma. (B) Evacuation of hematoma and changing of generator. (C) Four months postoperative.
One of the major challenges in diagnosis is the determination of the extent of the
infection. On initial assessment, laboratory testing, blood, and exudate cultures
as well as imaging modalities, including chest X-rays, transthoracic and transesophageal
echocardiography, and computed tomography (CT) scanning are used to identify whether
the infection is limited to the device pocket or stems from an endocardial or peripheral
source.[13] However, in addition to such studies, clinical experience helps to guide treatment
decisions.
Identification of the causative agent requires cultures of the pacemaker pocket site
and blood cultures ([Table 1]). Tissue culture sensitivities are higher than swab cultures from the pocket site.
However, up to 30% of patients with clinical signs of pacemaker infections have negative
cultures.[13] Additional gram staining, anaerobic and aerobic bacterial cultures, along with fungal
cultures and staining should be sought as well as mycobacteria cultures, if the initial
gram stain is negative. Usually, bacterial seeding occurs at the time of implantation,
revision, or replacement of the device. Lastly, the pacemaker may become hematogenously
infected in the case of a bacteremia due to another infection.[10]
Table 1
Causative agents behind pacemaker infections
|
Microorganism
|
% Infections
|
|
Coagulase-negative staph
|
42
|
|
S. epidermidis
|
|
|
S. saprophyticus
|
|
|
S. schleiferi
|
|
|
S. lugdunensis
|
|
|
S. haemolyticus
|
|
|
Methicillin-sensitive S. aureus
|
25
|
|
Gram-negative bacilli
|
9
|
|
Enterobacteriaceae
|
|
|
Pseudomonas aeruginosa
|
|
|
Nonfermentative gram-negative bacilli
|
|
|
Polymicrobial
|
7
|
|
Culture negative
|
7
|
|
Methicillin-resistant S. aureus
|
4
|
|
Gram-positive cocci
|
4
|
|
Fungi
|
2
|
Left Ventricular Assist Devices
Similar to pacemakers, determination of the extent of infection when faced with an
LVAD infection is particularly difficult. Left ventricular assist device infections
are classified into three categories: isolated driveline infections, pump pocket infections,
and intravascular device infection or LVAD endocarditis. Patients present with an
array of complaints including cellulitis, drainage from the LVAD driveline and possible
exposure of the device. A CT scan is often employed to determine whether the infection
is limited to the driveline or if it extends to the LVAD pocket, however, the true
extent of infection can only be determined at the time of debridement. Fever, leukocytosis,
and positive blood cultures can herald LVAD device endocarditis, which may ultimately
respond only to device exchange.
Microbiology of Device Infection
Microbiology of Device Infection
Pacemakers
Cardiac device infections consist of a wide variety of organisms, with reports of
polymicrobial infections between 7 to 15% of the time (see [Fig. 3]).[14]
[15]
[16] The most common organism found across multiple studies is the Staphylococcus species. Methicillin-resistant Staphylococcus aureus, methicillin-sensitive S. aureus, methicillin-resistant S. epidermidis, and methicillin-sensitive S. epidermidis contribute to more than half the pacemaker infection cases reported. Other gram-positives
include Enterococcus faecalis, Enterobacter cloacae, Propionibacterium acnes, and Corynebacterium amycolatum; they make up < 5% of infections. Gram-negative organisms comprise approximately
10% of infections and include Pseudomonas aeruginosa, Klebsiella pneumoniae, Providencia stuartii, Serratia marcescens, Stenotrophomonas
maltophilia, Enterobacter aerogenes, Escherichia coli, Citrobacter koseri, and others. Fungal agents, although rare, include Candida species and Aspergillus fumigatus (see [Fig. 3]).[14]
[15]
[16]
Fig. 3 The microbiology of pacemaker infections.
Staphylococci and certain Candida infections are difficult to treat due to their ability to build biofilms on surfaces
of foreign bodies such as CIEDs. Biofilms are a thick, multilayered film that mechanically
traps bacteria, which when dormant, are highly resistant to bacteriocidal antibiotics
via inhibition of cell wall biosynthesis, such as β-lactam antibiotics.[17]
[18]
[19]
Left Ventricular Assist Devices
Similarly, LVAD infections may be attributed to various microorganisms. Driveline
infections and pocket infections are mostly caused by gram-positive organisms, particularly
the Staphylococcus and Enterococcus species. The most common gram-negative pathogen is Pseudomonas. Candida species may be attributed to driveline, pump-pocket, as well as LVAD-associated endocarditis
in susceptible individuals. In addition, the majority of fungal pathogens may be drug
resistant and be challenging to treat.[5]
Treatment of Device Infections
Treatment of Device Infections
Cardiac implantable electronic device explantation along with culture-driven intravenous
(IV) antibiotics, remain the standard treatment modality in addressing device infections.[14] Device salvage, as discussed below, is reserved for patients that are LVAD dependent
and unable to tolerate explantation.
Pacemakers
The treatment of pacemaker infection consists of complete removal of the infected
hardware and a capsulectomy followed by individualized antimicrobial therapy.[15]
[20]
[21]
[22]
[23]
[24]
[25] If patients are pacemaker dependent, a temporary pacer is placed at or before the
time of exchange.[15]
[23] Reimplantation of devices, if necessary, depends on the location of the infection.
Preferably, the pacemaker should be placed on the contralateral side to the infection,
after cellulitis has resolved and cultures are negative. Another strategy involves
the placement of the device in a novel plane, often subpectoral. This ensures the
device is covered by healthy vascularized muscle. It is particularly useful in emaciated
or thin patients with poor overlying tissues and thin skin. This process is usually
undertaken in a mean of 7 days. When infective endocarditis is present, timing is
guided by negative blood cultures, a clinical assessment of the patient, and improved
vegetation burden.[15]
Time to reimplantation using the traditional approach averages between 7 and 15 days,
depending on the location of the infection and the presence/absence of cardiac device-related
infective endocarditis (CDIE). Patients with CDIE have the longest time to reimplantation.
The reported mortality rates in this population ranges from 13% to 21%.[14]
[15]
[26] Complications that may arise when following the traditional method of device explantation
include cardiac arrest, sepsis, operative cardiac tear, pulmonary embolism, hematoma
of the device pocket site, pericardial effusions, pericarditis, venous or arterial
thrombosis, and pneumothorax. Rodriguez et al reported fatal complications from extraction
procedures in 2 of 506 patients. Tarakji et al reported fatal complications from device
extraction in 2 of 412 patients, and a 1-year mortality rate of 17%; 2.6% of patients
who underwent reimplantation had relapsing infection.[16]
Left Ventricular Assist Device Salvage
Patients presenting with LVAD infections with recovering cardiac tissue are best treated
with removal of the device. This can occur in the setting of cardiac recovery or in
cases where a transplant heart becomes available. However, these procedures are not
without inherent morbidity and mortality. Various techniques have been described to
salvage infected LVADS, including IV antibiotics, wet to dry dressing changes around
the device, and negative pressure wound therapy. Recently, a novel approach for device
salvage of infected LVADs was developed for high-risk patients where the LVAD could
not be removed, patients who were poor candidates for major cardiac surgery, or patients
who had failed previous treatments. Salvage procedures involve surgical incision,
drainage and aggressive debridement, and placement of antibiotic impregnated beads
or slurry ([Fig. 4]).
Fig. 4 Algorithm used for treatment of all patients. Repeated débridement and bead exchange
were typically performed every 1 to 2 weeks until results of surgical-site cultures
were negative or until other definitive endpoints were reached (i.e., device removal,
transplantation, or death).[5]
Using the algorithm above, Kretlow et al successfully cleared infections in 17 of
26 patients with left ventricular assist devices. Cleared infections led to a dramatic
decline in mortality rates. For patients whose infection persisted, mortality rates
were 67% (6 of 9 patients) over the course of the study, whereas that of the cleared
population was 29% (5 of 17 patients). The cause of mortality in all those with persisting
infections was sepsis, whereas the cause in those with cleared infections included
cerebrovascular accident, right heart failure, left ventricular assist device-related
gastrointestinal perforation, and graft rejection.[5]
Antibiotic-impregnated beads have been successfully used in the treatment of prosthetic-related
infections by orthopedic and vascular surgeons with reported clearance rates ranging
from 60% to 100% with recurrence rates ranging from 0 to 20%.[22]
[27]
[28]
[29]
[30]
[31] The beads are capable of delivering high concentrations of necessary antibiotics
directly to a site of infection, thereby reducing the systemic side effects. Beads
are available in resorbable and nonresorbable materials. Nonresorbable beads require
bead exchange or removal, unlike their resorbable counterparts.
Polymethylmethacrylate is a nonresorbable medium that is used across many subspecialties
to deliver high concentrations of antibiotics locally. Its high cure temperature of
93°C requires the use of heat-stable antibiotics such as vancomycin, tobramycin, and
gentamicin, which sometimes limits its use.[31] However, if resistance to these antimicrobials is encountered, fibrin sealant or
calcium sulfate impregnated with the susceptible heat-labile antibiotic of choice
may be used. Kretlow et al encountered difficulty with this alternative regimen. Two
of three patients treated with fibrin sealant or calcium sulfate-based beads had difficulty
clearing infections. This was possibly due to the decreased efficacy of the fibrin
sealant and calcium sulfate compared with polymethylmethacrylate; however, this may
also have been due to the challenge of treating multidrug-resistant organisms.[5]
Repeated debridement and bead exchange typically performed every 1 to 2 weeks until
surgical site cultures were negative or until other definitive endpoints had been
reached (device removal, transplant, or if the patient expired). Subsequently, the
device was permanently covered with a vascularized flap using a myocutaneous, fasciocutaneous,
and/or omental flap.
The location of the infection ultimately played a role in treatment. Once cultures
were negative, patients with infection at the driveline underwent repositioning of
the driveline underneath the anterior rectus sheath ([Fig. 5A–C]). Infection of the device pocket ([Fig. 5D, E]) required local tissue coverage and rectus abdominis muscle flaps where soft tissue
coverage was warranted.[5]
Fig. 5 (A–C) Patient undergoing coverage of salvaged left ventricular assist devices (LVAD) drive
line using the anterior rectus sheath. (D, E) Placement of antibiotic beads for salvage of LVAD.
Conclusion
Although tremendous gains have been made in our understanding of the pathogenesis,
risk factors, and management of CIED infections over the last decade, the burden on
patients and the health care system represents a significant challenge. For these
patients, early diagnosis can make a great difference in terms of survival. Intravenous
antibiotics and nonsurgical approaches may not provide definitive treatment in some
of these conditions, with many recommending extraction and device removal, although
that is not without its attendant risks. Patients who are LVAD dependent and are unable
to undergo major cardiac surgery may benefit from antibiotic bead placement and device
salvage with outcomes comparable with those of the currently recommended treatments
above.
