REVIEW

C URRENT
OPINION Endocarditis in the intensive care unit: an update
Ines Lakbar a, Louis Delamarre a, Sharon Einav b and Marc Leone a

Purpose of review
The incidence of infective endocarditis (IE) is increasing worldwide, resulting in a higher number of patients
with IE being admitted to intensive care units (ICU). Nearly half of patients with IE develop a complication
during their clinical course. However, few well conducted studies or reviews are devoted to critically ill IE
patients. This review discusses the contemporary perioperative and intensive care literature.
Recent findings
IE epidemiology is changing towards elderly and frail patients. ICU patients are at risk of risk of
developing IE because they are often in a pro-inflammatory state and many also have several indwelling
catheters, which favors infection. Increased performance and recent advances in cardiac imaging allow for
easier diagnosis of EI, but the applicability of these techniques to ICU patients is still relatively limited. New
developments in antibiotic treatment and adjunctive therapies are explored further in this review.
Summary
The lack of evidence on ICU patients with IE highlights the critical importance of multidisciplinary decision-
making and the need for further research.
Keywords
critically ill patient, infective endocarditis, intensive care unit, outcomes

INTRODUCTION Epidemiology and pathophysiology

Infective endocarditis (IE) is a disease with an inci- A large meta-analysis pooling data from nationwide
dence of 3–5 per 100 000 individuals per year population-based registries in Europe reported a
according to European population-based studies doubling of the incidence of IE from 2000 to 2018
&
[1]. Despite recent improvements in treatment, IE [2 ]. This trend was also found among intensive care
remains associated with high morbidity and mortal- unit (ICU) patients [6]. Such an increase may be
ity rates due to a change in the profiles of IE patients. explained by increased performance rates of inva-
Compared to previous cohorts, patients with IE sive procedures in the cardiac patient. Indeed, a
today are older, more likely to have a prosthetic large prospective cohort reporting data from 40
valve or device-related infections and more likely countries showed an increase in the incidence of
&
to develop healthcare associated infections [1,2 ]. In prosthetic and device-related IE from 2016 to 2018,
a US population based-study, complications associ- representing up to 40% of all the diagnosed IE [1].
ated with increased mortality rates were reported in Aortic valve replacement rates have also doubled in
&
nearly 40% of patients [3 ]. Likewise, in a large
European study, the outcome of IE patients with
septic shock was poor, with a 1-year mortality rate a
Department of Anesthesiology and Intensive Care Unit, Aix Marseille
&
approximating two-thirds [4 ]. Despite these data, University, Assistance Publique Hôpitaux Universitaires de Marseille,
few well conducted studies or reviews are dedicated Nord Hospital, Marseille, France and bGeneral Intensive Care Unit of the
Shaare Zedek Medical Centre and the Hebrew University Faculty of
to critically ill IE patients. In addition, nearly half of
&& Medicine, Jerusalem, Israel
IE patients undergo a surgical intervention [5 ],
Correspondence to Ines Lakbar, Department of Anesthesiology and
hence the perioperative management of these Intensive Care Unit, Aix Marseille University, Assistance Publique Hôpi-
patients is of particular interest. This review will taux Universitaires de Marseille, Nord Hospital, 15 chemin des Bourrely,
discuss the contemporary literature and will provide 13015 Marseille, France. Tel: +33 4 91 96 86 55;
an overview of the current state of knowledge e-mail: [email protected]
regarding the perioperative and intensive care of Curr Opin Crit Care 2022, 28:503–512
critically ill IE patients. DOI:10.1097/MCC.0000000000000973

1070-5295 Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved. www.co-criticalcare.com

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.

Severe infections

In a nutshell, the incidence of IE has increased in

KEY POINTS the last two decades in the overall population and
 The incidence of infective endocarditis increases therefore in the ICU population. Furthermore, ICU
worldwide, including in intensive care unit (ICU), with patients are at risk of developing IE because they are
an increase in patients’ age, frailty and history of often in a pro-inflammatory state and many also have
cardiac devices and prosthetic valves. several indwelling catheters, which favors infection.
Finally, the indications for transcatheter procedures
 Nearly half of infective endocarditis (IE) patients
develop complication(s), which are associated with extend to include the elderly and frail and these
mortality and ICU admission. patients are at greater risk of complications with
increased mortality rates than prior cohorts.
 The diagnosis of IE benefits from the advances in
cardiac imaging (computed tomography and nuclear
medicine), but the applicability of those techniques to Risk factors for a poor outcome
ICU patients is poorly reported. &
In patients with IE, acute respiratory failure [4 ],
 Innovative treatments are described, such as hyperbaric & & &
acute kidney injury [3 ,4 ,15,16], sepsis [3 ,17,18],
oxygen, antiplatelets and anticoagulation medications, & &
acute heart failure [3 ,4 ,18], acute neurological fail-
but none has been thoroughly explored in clinical trials. &
ure [3 ,16,19], multiorgan failure (as illustrated by
high sequential organ failure assessment [SOFA]
&
scores) [19–21], preexisting comorbidities [4 ,15]
and disseminated intravascular coagulopathy are
the last decade with a major increase of transcath- all common causes of ICU admission. All of these
&
eter valve replacement procedures [7,8]. These endo- are also associated with high rates of mortality [3 ].
vascular procedures have the same risk of IE as do Increasing age, which is a trend also observed in
surgical procedures [9]. Both result in some degree of patients admitted to the ICU for IE, seems associated
valvular structural alteration and subsequent endo- with mortality as well [16,17,19,21]. Additional fac-
thelial injury. Structural valve alterations provide a tors which have been associated with complications
breeding ground for circulating micro-organisms as of IE, are congenital heart disease, congestive heart
they expose sub-endothelial molecules (collagen failure, a heavy burden of comorbidities and specific
&
and von Willebrandt factor) which allow bacterial causative pathogens, that is S. aureus and fungi [3 ].
adhesion and also modify the shear stress of the A multicenter prospective observational study
blood flow [10,11]. These predisposing conditions which included 246 ICU patients admitted for com-
have been reported in degenerative heart valve dis- plicated IE, noted that SOFA scores
5 and C-reactive
ease. However, a new animal model revealed that protein (CRP) levels
17.6 mg/l were associated with
generalized inflammation, as it occurs in sepsis, higher in-hospital mortality with an receiver operat-
results in valve inflammation, which, in combina- ing characteristic (ROC) area under the curve (AUC)
tion with bacteremia, leads to IE [12] (Fig. 1). of 0.87 (95% confidence interval [CI] 0.83–0.91),
This animal model was developed using Staph- while SOFA score, but not CRP, was also associated
ylococcus aureus, as its main virulence factors allow it with mortality at 12–36 months [19]. In a meta-anal-
to manipulate the host’s innate and adaptive immun- ysis gathering data from nine studies and 627 non-
ity and coagulation cascades and thus thwart the ICU patients, an increased serum troponin level was
defense mechanisms of the host [10,13]. These semi- associated with increased in-hospital mortality (odds
nal studies are consistent with clinical findings, as S. ratio [OR] 5.7, 95% CI 3.5–10.3) and 1-year mortality
aureus is the most common pathogen in patients with (OR 2.67, 95% CI 1.42–5.02) [22]. Increased D-dimer
IE, followed by Streptococcus and Enterococcus species levels were also associated with in-hospital mortality
&
[2 ]. Furthermore, in a cohort of ICU patients with and 6-month mortality in an observational study of
IE-related septic shock, S. aureus was the main agent 613 non-ICU patients [23]. In a single center retro-
&
responsible for the infection [4 ]. A significant spective study in non-ICU patients, elevated serum
increase in Enterococcus-associated IE has been procalcitonin concentrations both at admission and
reported consistently over the past decade, likely at 48–72 h were associated with in-hospital mortality
related to the increased frequency of transcatheter [16]. Conversely, cardiac surgery is often reported as
&
procedures in the elderly, as the femoral microbiota of a protective factor regarding mortality [4 ,18,19].
the elderly has been documented as displaying a
higher incidence of enterococcal colonization
&
[1,2 ,14]. The same microbiological profile was found Diagnosis
in a cohort of ICU patients [6]. The main etiologies of The diagnosis of IE is challenging in the ICU patient
IE according to valve location are depicted in Fig. 2. [24]. Fever is common in ICU patients and is

504 www.co-criticalcare.com Volume 28  Number 5  October 2022

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.

 Endocarditis in the intensive care unit: an update Lakbar et al.

FIGURE 1. Mechanisms of valve infection.

therefore not specific to IE. Murmur is a classical delayed consultation and testing during the pan-
symptom, but auscultation can be difficult in ICU demic [30] have been shown to affect the diagnosis
patients, particularly in those requiring mechanical and management of IE, as well as other pathologies
ventilation or with tachycardia. Atypical clinical [30–36]. Hence physicians should maintain a high
presentations are therefore frequent in ICU patients. rate of suspicion in ICU patients regarding the pos-
Late diagnosis of IE, based on the occurrence of sibility of IE [30].
complications, is common. Furthermore, in many The cornerstone of diagnosis of IE remains visual-
cases, these complications are often the indication ization of the cardiac valves. Transthoracic echocar-
&
for ICU admission in the first place [3 ]. These diography (TTE) is only moderately sensitive for
complications may be cardiovascular (22.6%), neu- detection of vegetations in both native (sensitivity:
&
rological (16.7%), renal (6.4%) or hematologic [3 ]. 60–70%) and prosthetic (sensitivity: 50%) valves [37].
About 1 in 10 patients with IE will also develop However, in most ICUs, this technique is immedi-
septic shock. Moreover, infection occurs in up to ately available and repeatable at the bedside by the
7.4% of prosthetic valves and 9% of implanted intensivist. Transesophageal echocardiography (TEE)
cardiac devices, but such infections often have mis- exhibits a much better performance for detection of
leading clinical and ultrasound presentations [25]. vegetations, with a sensitivity of 90–100% and 90%
Finally, the diagnosis of other diseases, sharing in native and prosthetic valves, respectively [37].
those nonspecific symptoms, can be falsely priori- The diagnostic criteria of IE have been modified
tized, as it has been the case during the coronavirus in 2015, through the introduction of the modified
disease 2019 (COVID-19) [26–29]. Furthermore, Duke criteria in the North American guidelines [38]

1070-5295 Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved. www.co-criticalcare.com 505

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.

Severe infections

FIGURE 2. The common etiologies of infectious endocarditis according to the location of the involved valve [79].

and the European Society of Cardiology (ESC) cri- Some authors therefore propose to combine both
teria in the European guidelines [24]. techniques when prosthetic valve endocarditis is sus-
The 2015 ESC diagnostic criteria added three pected. The pharmacological reduction of heart rate
elements to the modified Duke criteria [39]. Two of that is recommended to ensure good-quality images
&
those three, suggested as major criteria, involved the may limit the use of cardiac CT in ICU patients [40 ].
use of advanced cardiac imaging techniques to detect This imaging tool should therefore probably be
cardiac lesions attributable to IE, whereas the detec- chosen only when performing TEE is not feasible.
tion of recent silent embolic events or infectious The comparative performances of TEE and cardiac
aneurysms by imaging is proposed as a new minor CT are presented in Table 1 and Fig. 3.
criterion. The new major criteria based on cardiac The second imaging technique recommended in
imaging techniques are the detection of paravalvular the ESC guidelines, 18F-FDG PET/CT, relies on 18F-
lesions by cardiac multislice computed tomography FDG uptake by white blood cells after an injection 1-h
(CT) and the detection of abnormal activity on pros- prior to imaging. 18F-FDG PET/CT has a sensitivity of
thetic valve by 18F-flurodeoxyglucose (FDG) positron 67–80%, specificity of 75%, positive predictive value
emission tomography/computed tomography (PET/ of 90% and negative predictive value of 40% for
&
CT) or radiolabelled leucocyte single-photo emission the diagnosis of prosthetic valve endocarditis [44 –
computed tomography (SPECT/CT). 46], and a sensitivity and specificity of 85% for the
Cardiac multislice CT is a high-resolution diagnosis of infection in implanted cardiac devices
(<1 mm) imaging technique of the heart after ECG- [47]. In native valve IE, known for inducing a smaller
white blood cell traffic [48], 18F-FDG PET/CT alone
&
gated retrospective reconstruction [40 ,41]. Cardiac
CT has a performance almost similar to TEE for diag- showed a sensitivity within the range of 20–57% and
& & &
nosing IE [40 ,42 ]. However, cardiac CT better a specificity approximating 100% [49 ,50].
detects paravalvular abscesses and pseudo-aneur- The application of ESC criteria, including
ysms, whereas TEE seems to better detect vegetations those observed in 18F-FDG PET/CT, increased the
sized <10 mm and valvular perforations [41]. In sensitivity of modified Duke criteria for native valve
patients with prosthetic valves, the sensitivity of endocarditis from 63.5% to 70% without affecting its
specificity [49 ]. However, imaging with 18F-FDG
&
CT and TEE seem similar, with a trend towards better
performance of cardiac CT for paravalvular abscesses, PET/CT requires fasting for 24 h, hemodynamic
vegetations and inflammatory infiltrations, whereas stability and sometimes heparin (50 UI/kg)
& & & &
TEE better detects paravalvular leakage [40 ,42 ,43 ]. to enhance glucose/FDG uptake [48,51,52 ].

506 www.co-criticalcare.com Volume 28  Number 5  October 2022

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.

 Table 1. Summary of the morphological diagnosis tools in infective endocarditis, with considerations pertaining to their use in ICU
Performance in Availability/
Performance in prosthetic feasibility
Technique Indication native valves valves in ICU Contraindications Limitations in ICU References

TTE Suspected IE Se: 60--70% Se: 50% High None -Lower quality images in Pelletier-Galarneau [37]
ventilated patients
-Operator-dependent ac
curacy
TEE -Suspected IE Se: 90--100% Se: 90% High Esophageal surgery, -Requires sedation Pelletier-Galarneau
&
-Confirmed IE mediastinum - Requires fasting [37], Sifaoui [42 ],
&
-Assessment of radiotherapy, unstable - Requires experienced Koo [40 ],
&
valve damage cervical spine fracture, operator (cardiologist Michalowska [43 ]
and heart upper airway mass or intensivist)
function -More difficult to visualize
-Preoperative prosthetic valves due to
assessment metal artifacts
&
Cardiac multislice -Suspected IE Accuracy: 92.6% Se: 92%, Sp: 86% Scarce data Not specific Heart rate 75 bpm Koo [40 ], Khalique
&
computerized -Confirmed IE available facilitates ECG-gated [41], Sifaoui [42 ],
&
tomography (CT) -Assessment of analysis Michalowska [43 ]
valve damage Can be difficult or poorly
and heart tolerated in ICU
function patients
-Preoperative
assessment

1070-5295 Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.
18
F-FGD PET/CT -Suspected IE Se: 20--57% Se: 67--77% Low, scarce Not specific Requires 24 h fasting, Mahmood [46], Venet
Sp: 100% Sp: 75% data available hemodynamic stability, [45], Abikhzer [50],
&
PPV: 91% avoidance of Philip [44 ], Philip
&
NPV: 42% corticosteroids and [49 ], van Hulst
&
hyperglycemia. [52 ], Schenone [83]
Heparin may enhance
image quality, but
potentially deleterious
in ICU patients
Radiolabelled -Complementary NA NA Low to very low Not specific Similar to 18F-FGD PET/ NA
leucocyte after 18F-FGD CT limitations
SPECT/CT PET/CT Requires
radiopharmaceutical
preparation of
autologous leucocytes.

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.

Time-consuming image
acquisition

www.co-criticalcare.com
18
F-FGD PET/CT, F18-flurodeoxyglucose positron emission tomography and computed tomography; CT, computerized tomography; Se, sensitivity; Sp, specificity; SPECT/CT, single-photo emission computerized
tomography; TEE, transesophageal echocardiography; TTE, transthoracic echocardiography.

507
Endocarditis in the intensive care unit: an update Lakbar et al.
Severe infections

FIGURE 3. Comparison of the performance of transesophageal echocardiography and multislice cardiac computed
tomography for the diagnosis of infective endocarditis. CT, computerized tomography; Se, sensitivity; Sp, specificity; TEE,
transesophageal echocardiography.

Concurrent administration of corticosteroids and the than modified Duke criteria for the diagnosis of
&
presence of hyperglycemia have been associated with prosthetic valve endocarditis [44 ]. The performance
&
false negative results [52 ]. Despite these limitations of those criteria in the ICU setting in not described.
of 18F-FDG PET/CT in ICU patients, the use of this tool An increasing proportion of ICU physicians
in such patients has been described by a few teams, have expertise in point of care ultrasound practice.
with successful diagnosis of IE in several cases The question whether the availability of ultrasonog-
&
[52 ,53]. Unfortunately, there is still very little data raphy-trained physicians in the ICU facilitates the
on the performance and feasibility of nuclear imaging diagnosis of IE remains unresolved in the existing
in ICU patients and no data on the performance of literature. Lau et al. [54] reported similar perform-
nuclear imaging specifically for diagnosis of IE in ICU ance of intensivist-led vs. cardiologists-led TEE in
&
patients [52 ]. ICU patients if the intensivists were board certified
Radiolabelled leucocyte SPECT/CT requires a for ultrasound. The 2016 guidelines for the training
radiolabeling of autologous leucocytes that, once standards in ultrasound in the ICU graded the skills
re-injected, will accumulate in damaged heart tissue of endocarditis diagnosis as a basic skill, whereas TEE
in a time-dependent fashion [24]. The data regard- and prosthetic valve endocarditis diagnosis were
ing its use in critically ill IE patients is insufficient to graded as advanced skills [25].
draw any assessment of its pertinence and feasibility
at the time of this writing. The characteristics and
performance of those diagnosis tools are presented Microbiological innovations
in Table 1. Finally, the 2015 ESC criteria have been Microorganism detection and identification have
shown to be more sensitive but not more specific dramatically improved over the last years, especially

508 www.co-criticalcare.com Volume 28  Number 5  October 2022

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.

 Endocarditis in the intensive care unit: an update Lakbar et al.

due to the improvements in 16S-ARN PCR and Next- American guidelines highlight that initiation of
Generation Sequencing (NGS). The impact of these antibiotic treatment is rarely urgent in cases of IE
techniques has been mostly described in cases of [38]. In patients with septic shock, urgent introduc-
suspected IE with negative blood cultures, where tion of empirical antibiotics is recommended, and
NGS or 16S-RNA sequencing were performed post- this recommendation should therefore also apply to
operatively on valvular tissue [55–59], without spe- patients with known or suspected IE and septic
&&
cific application in the ICU setting. shock [61 ]. Of note, preliminary observational
data suggested that ceftaroline and ceftobiprole
could be used when staphylococci and enterococci
MANAGEMENT are the causative agents of IE [62,63], in contrast to
linezolid which may be associated with higher mor-
Antibiotic treatment tality rates in this setting [64].
Randomized controlled trials assessing antibiotic Once the causative microorganism has been
treatment are rare in the general IE population identified, antibiotic treatment should be adapted
[60], and none is available in the ICU patients. to antimicrobial susceptibility, appropriate drug
International guidelines provide recommendations coverage must be chosen with the ‘‘Endocarditis
based mainly on low quality evidence from exper- Team’’, a multidisciplinary group of clinicians, fol-
imental and observational data [24,38]. A summary lowing the principles of antimicrobial stewardship
of empirical treatment as recommended by interna- [65] (Table 2). The choice of treatment should take
tional guidelines is provided in Table 2. Microbio- into account the individual characteristics of the
logical samples should be taken before antibiotic ICU patient: kidney function (from augmented
treatment is started. In case of non-nosocomial IE renal clearance to acute kidney injury), capillary
in native valves, treatment should ensure coverage leakage, hypoalbuminemia, and organ support
for methicillin-susceptible S. aureus, b-hemolytic requirements, such as renal replacement therapy
streptococci according to European experts whereas (RRT) and extracorporeal membrane oxygenation
American experts stated that antibiotic treatment (ECMO) [66]. There is a growing body of evidence
should be tailored to the clinical case [24]. Indeed, supporting the monitoring of therapeutic drug

Table 2. The antibiotic regimens suggested in International guidelines (adapted from reference [60])
American Heart Association European Society of Cardiology (ESC)
(AHA) guidelines, 2015 [38] guidelines, 2015 [24]

Empirical treatment in the ICU To be adjusted based on clinical If community-acquired

course and risk factors Ampicillin þ (cl)oxacillin þ gentamicin
If nosocomial
Vancomycin þ gentamicin þ rifampicina
Staphylococcal endocarditis, Methicillin-susceptible Methicillin-susceptible
native valve(s) Antistaphylococcal penicillin Antistaphylococcal penicillin
Methicillin-resistant Methicillin-resistant
Vancomycin or daptomycin Vancomycin or daptomycin
Staphylococcal endocarditis, Methicillin-susceptible Methicillin-susceptible
prosthetic valve(s) Antistaphylococcal penicillin þ Antistaphylococcal penicillin þ gentamicin þ rifampicin
gentamicin þ rifampicin Methicillin-resistant
Methicillin-resistant Vancomycin or daptomycin þ gentamicin þ rifampin
Vancomycin or daptomycin þ
gentamicin þ rifampin
Streptococcal endocarditis, susceptible Four-week course Four-week course
strains (MIC penicillin < 0.25 mg/l) Penicillin G or ceftriaxone Penicillin G or ceftriaxone or amoxicillin
Enterococcal endocarditis, Ampicillin þ ceftriaxone Ampicillin þ ceftriaxone
penicillin-susceptible strains 6 weeks
6 weeks

Antibiotic treatment should be selected by consensus within the ‘‘Endocarditis Team’’, a team of experts including infectious diseases specialists, microbiologists,
cardiologists and intensivists.
Long-term gentamicin, listed as an option in the guidelines, has been deliberately omitted from this table because the authors believe that long-term
aminoglycosides should be reserved for difficult-to-treat multidrug-resistant infections owing to the high risk of acute kidney injury and muscle weakness in critically
ill patients [80]. This consideration is backed by expert opinion that aminoglycosides could be avoided in 90% of cases [81].
a
Rifampicin is only indicated for prosthetic valve endocarditis and should only be introduced 5--7 days after the start of the pivotal antiinfective regime [60].

1070-5295 Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved. www.co-criticalcare.com 509

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.

Severe infections

antibiotic levels in critically ill patients [67,68], including surgical, anesthetic and ICU teams in a
&&
which can be applied to the context of IE [69]. recent comprehensive review [5 ].
Oral treatment should be regarded as a step- Specifically, after cerebral embolism, delaying
down treatment and should be reserved only for surgery has been recommended due to the risk of
patients who have cleared their bacteremia [70]. infarct extension, hemorrhagic transformation of
an ischemic infarct and expansion of an intracranial
hemorrhage. Such practice has also not been associ-
Complications ated with higher mortality rates [71]. However recent
Septic shock and sepsis should be managed as recom- data have also shown that early and late surgery have
mended in the Surviving Sepsis Campaign guidelines, the same outcomes in terms of mortality [71,72],
with early optimal management combining fluid which raises ongoing questions regarding the optimal
resuscitation, antibiotic treatment and surgical con- delay for surgery. As for the treatment of the acute
&&
trol of the source of infection if necessary [61 ]. Car- ischemic stroke itself, thrombectomy has been
diac output should be closely monitored as it may be reported but its safety is yet to be proven [73,74].
impaired indirectly by the septic cardiomyopathy
associated with septic shock or directly by valvular
insufficiency. Cardiovascular complications, includ- Future
ing cardiac arrest, cardiogenic shock, acute myocardial As shown in experimental studies, a complex inter-
infarction, acute heart failure and heart block were action between bacteria, coagulation and innate
reported in almost a quarter of patients with IE, regard- immunity is mandatory for valve infection. There-
less of their location of treatment in hospital (i.e. fore, experimental and clinical studies have thus far
&
general ward or ICU) [3 ]. In the ICU, acute heart aimed to target platelets and coagulation factors in
failure (from septic or cardiogenic shock) was reported order to inhibit vegetation growth and embolic
in more than half of the patients with IE [6]. events, using aspirin as prophylactic or adjunctive
The indications for urgent surgery in both Amer- treatment. Clinical findings have already yielded
ican and European international guidelines are conflicting results regarding the use of aspirin as
heart failure, along with uncontrolled infection prophylactic treatment for IE, in patients receiving
&
(ongoing bacteriemia, local abscesses and fistulae) aspirin for long-term treatment [75 ]. In a mouse
and emboli prevention (in cases of recurrent emboli model of bacteremia, the use of ticagrelor enhanced
or large vegetations) (Table 3). However, there are platelet killing of S. aureus [76,77]. It has therefore
discrepancies between American and European been advocated that further studies should evaluate
experts regarding the timing of surgery. American ticagrelor for prophylactic treatment of IE in
experts recommend early surgery, while European patients with prosthetic valves or indwelling devi-
&
experts highlight the need to weigh the benefits of ces [75 ]. Experimental studies also reported
antibiotic completion before surgery against the risk improved outcomes in terms of vegetation size,
of an uncontrolled source of infection. This lack of bacterial load and inflammation reduction with
agreement in the guidelines highlights the paucity the use of dabigatran, a direct thrombin inhibitor
&
of evidence in the literature and has led anesthesi- [78 ]. However, such treatments in are challenging
ologists to advocate for multidisciplinary decisions, in ICU patients, as they were associated with a
&
major risk of bleeding [75 ].
Table 3. Indications for surgery in infective endocarditis Hyperbaric oxygen therapy could also be an
[24,38] adjunctive treatment worth exploring in patients
with IE. In experimental models of IE, hyperbaric
Indication for surgery Symptoms
oxygen therapy has been shown to enhance the
&

Heart failure Severe valve obstruction bactericidal effect of antibiotics [78 ].

Severe valve regurgitation
Uncontrolled Persistent positive blood cultures despite CONCLUSION
infection antibiotic treatment
IE remains a rare and complex disease, and although
Local complications: abscesses, false
aneurysms, fistulas recent research has opened up new avenues for
better management of IE and new treatment
multidrug resistant bacteria or fungi
options, the available evidence is still weak, prompt-
Prevention of Large vegetation with embolic events ing the recommendation for decisions on patient
embolic events
management to be made within a multidisciplinary
Enlarging vegetation despite antibiotic
team, particularly with regard to antibiotic treat-
treatment
ment and timing of surgery. We therefore advocate

510 www.co-criticalcare.com Volume 28  Number 5  October 2022

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.

 Endocarditis in the intensive care unit: an update Lakbar et al.

17. Gatti G, Chocron S, Obadia J-F, et al. Using surgical risk scores in non-
for the active involvement of intensivists in the surgically treated infective endocarditis patients. Hellenic J Cardiol 2020;
Endocarditis team. 61:246–252.
18. Krajinovic V, Ivancic S, Gezman P, et al. Association between cardiac surgery
and mortality among patients with infective endocarditis complicated by
Acknowledgements sepsis and septic shock. Shock 2018; 49:536–542.
19. Lin Y, Dong S, Yuan J, et al. Accuracy and prognosis value of the sequential
All the figures were created with BioRender.com. organ failure assessment score combined with C-reactive protein in patients
with complicated infective endocarditis. Front Med 2021; 8:576970.
20. Hu W, Wang X, Su G. Infective endocarditis complicated by embolic events:
Financial support and sponsorship pathogenesis and predictors. Clin Cardiol 2021; 44:307–315.
21. Lin Y, Liu F, Gong S, et al. Validity of SOFA score as a prognostic tool for
None. critically ill elderly patients with acute infective endocarditis. Rev Cardiovasc
Med 2021; 22:967.
22. Postigo A, Vernooij RWM, Fernández-Avilés F, et al. Cardiac troponin and
Conflicts of interest infective endocarditis prognosis: a systematic review and meta-analysis. Eur
There are no conflicts of interest. Heart J Acute Cardiovasc Care 2021; 10:356–366.
23. Lin Y-W, Jiang M, Wei X, et al. Prognostic value of D-dimer for adverse
outcomes in patients with infective endocarditis: an observational study. BMC
Cardiovasc Disord 2021; 21:279.
REFERENCES AND RECOMMENDED 24. Habib G, Lancellotti P, Antunes MJ, et al. 2015 ESC Guidelines for the
management of infective endocarditis: the Task Force for the Management of
READING Infective Endocarditis of the European Society of Cardiology (ESC) Endorsed
Papers of particular interest, published within the annual period of review, have by: European Association for Cardio-Thoracic Surgery (EACTS), the Eur-
been highlighted as: opean Association of Nuclear Medicine (EANM). Eur Heart J 2015;
& of special interest 36:3075–3128.
&& of outstanding interest
25. Levitov A, Frankel HL, Blaivas M, et al. Guidelines for the appropriate use of bedside
general and cardiac ultrasonography in the evaluation of critically ill patients—Part
1. Habib G, Erba PA, Iung B, et al. Clinical presentation, aetiology and outcome II: Cardiac ultrasonography. Crit Care Med 2016; 44:1206–1227.
of infective 277 endocarditis. Results of the ESC-EORP EURO-ENDO 26. Goswami J, MacArthur TA, Sridharan M, et al. A review of pathophysiology,
(European infective endocarditis) registry: a 278 prospective cohort study. clinical features, and management options of COVID-19 associated coagulo-
Eur Heart J 2019; 40:3222–3232. pathy. Shock 2021; 55:700–716.
2. Talha KM, Baddour LM, Thornhill MH, et al. Escalating incidence of infective 27. Goyal P, Choi JJ, Pinheiro LC, et al. Clinical characteristics of COVID-19 in
& endocarditis in Europe in the 21st century. Open Heart 2021; 8:e001846. New York City. N Engl J Med 2020; 382:2372–2374.
A large meta-analysis of pooled data from nationwide registries in Europe reporting 28. Helms J, Kremer S, Merdji H, et al. Neurologic features in severe SARS-CoV-2
a twofold increase in the incidence of IE. infection. N Engl J Med 2020; 382:2268–2270.
3. Mir T, Uddin M, Qureshi WT, et al. Predictors of complications secondary to 29. Siordia JA. Epidemiology and clinical features of COVID-19: a review of
& infective endocarditis and their associated outcomes: a large cohort study current literature. J Clin Virol 2020; 127:104357.
from the national emergency database. Infect Dis Ther 2022; 11:305–321. 30. Cosyns B, Motoc A, Arregle F, et al. A plea not to forget infective endocarditis
A large nationwide cohort which showed that IE complications are frequent and in COVID-19 era. JACC: Cardiovasc Imaging 2020; 13:2468–2470.
highly associated with increased mortality rates 31. Alizadeh K, Bucke D, Khan S. Complex case of COVID-19 and infective
4. Pericàs JM, Hernández-Meneses M, Muñoz P, et al. Outcomes and risk factors endocarditis. BMJ Case Rep 2021; 14:e242205.
& of septic shock in patients with infective endocarditis: a prospective cohort 32. Amir M, Djaharuddin I, Sudharsono A, et al. COVID-19 concomitant with
study. Open Forum Infect Dis 2021; 8:ofab119. infective endocarditis: a case report and review of management. Int J Infect
A large nationwide cohort whose authors reported that septic shock was frequent Dis 2020; 98:109–112.
in IE, pointing to the role of intensive care medicine in the management of IE and 33. Benmalek R, Mechal H, Choukrallah H, et al. Bacterial co-infections and
calling for active involvement of intensivists in this disease. superinfections in COVID-19: a case report of right heart infective endocar-
5. Hermanns H, Eberl S, Terwindt LE, et al. Anesthesia considerations in ditis and literature review. Pan Afr Med J 2020; 35:40. doi: 10.11604/pamj.
&& infective endocarditis. Anesthesiology 2022; 136:633–656. supp.2020.35.2.23577.
Excellent comprehensive review on the perioperative management of IE. 34. Finch LC, Gerdzhikov S, Buttery R. Haemophilus parainfluenzae endocar-
6. Joffre J, Dumas G, Aegerter P, et al. Epidemiology of infective endocarditis in ditis presenting with symptoms of COVID-19. BMJ Case Rep 2021; 14:
French intensive care units over the 2014 period—from CUB-Réa Network. e245210.
Crit Care 2019; 23:143. 35. George A, Venkataramanan SVA, John KJ, et al. Infective endocarditis and
7. Nguyen V, Willner N, Eltchaninoff H, et al. Trends in aortic valve replacement COVID -19 coinfection: an updated review. Acta Biomedica Atenei Parmen-
for aortic stenosis: a French nationwide study. Eur Heart J 2022; 43:666– sis 2022; 93:e2022030–e12022030.
679. 36. Linschoten M, Peters S, van Smeden M, et al. Cardiac complications in
8. Lauck SB, Baron SJ, Irish W, et al. Temporal changes in mortality after patients hospitalised with COVID-19. Eur Heart J Acute Cardiovasc Care
transcatheter and surgical aortic valve replacement: retrospective analysis of 2020; 9:817–823.
US medicare patients (2012–2019). JAHA 2021; 10:e021748. 37. Pelletier-Galarneau M, Abikhzer G, Harel F, et al. Detection of native and
9. Fauchier L, Bisson A, Herbert J, et al. Incidence and outcomes of infective prosthetic valve endocarditis: incremental attributes of functional FDG PET/
endocarditis after transcatheter aortic valve implantation versus surgical aortic CT over morphologic imaging. Curr Cardiol Rep 2020; 22:93.
valve replacement. Clin Microbiol Infect 2020; 26:1368–1374. 38. Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis in adults:
10. Schwarz C, Töre Y, Hoesker V, et al. Host-pathogen interactions of clinical S. diagnosis, antimicrobial therapy, and management of complications: a scien-
aureus isolates to induce infective endocarditis. Virulence 2021; tific statement for healthcare professionals from the American Heart Associa-
12:2073–2087. tion. Circulation 2015; 132:1435–1486.
11. Schwarz C, Hoerr V, Töre Y, et al. Isolating crucial steps in induction of 39. Li JS, Sexton DJ, Mick N, et al. Proposed modifications to the duke criteria
infective endocarditis with preclinical modeling of host pathogen interaction. for the diagnosis of infective endocarditis. Clin Infect Dis 2000;
Front Microbiol 2020; 11:1325. 30:633–638.
12. Liesenborghs L, Meyers S, Lox M, et al. Staphylococcus aureus endocarditis: 40. Koo HJ, Yang DH, Kang J-W, et al. Demonstration of infective endocarditis
distinct mechanisms of bacterial adhesion to damaged and inflamed heart & by cardiac CT and transoesophageal echocardiography: comparison with
valves. Eur Heart J 2019; 40:3248–3259. intra-operative findings. Eur Heart J Cardiovasc Imaging 2018; 19:199–
13. Meyers S, Crescente M, Verhamme P, et al. Staphylococcus aureus and 207.
neutrophil extracellular traps: the master manipulator meets its match in Retrospective study showing that the sensitivities of CT and TEE were not
immunothrombosis. ATVB 2022; 42:261–276. significantly different for both native and prosthetic valve infective endocarditis.
14. Panagides V, Abdel-Wahab M, Mangner N, et al. Very early infective en- 41. Khalique OK, Veillet-Chowdhury M, Choi AD, et al. Cardiac computed
docarditis after transcatheter aortic valve replacement. Clin Res Cardiol tomography in the contemporary evaluation of infective endocarditis. J Car-
2022. doi: 10.1007/s00392-022-01998-0. Epub ahead of print. diovasc Computed Tomogr 2021; 15:304–312.
15. Durante-Mangoni E, Giuffrè G, Ursi MP, et al. Predictors of long-term mortality 42. Sifaoui I, Oliver L, Tacher V, et al. Diagnostic performance of transesophageal
in left-sided infective endocarditis: an historical cohort study in 414 patients. & echocardiography and cardiac computed tomography in infective endocar-
Eur J Intern Med 2021; 94:27–33. ditis. J Am Soc Echocardiogr 2020; 33:1442–1453.
16. Ucciferri C, Auricchio A, Cutone C, et al. Risk factors associated with poor Prospective study in left-sided IE patients showing that TEE detects valvular IE-
outcome in patients with infective endocarditis: an Italian single-center related lesions better than cardiac CT, while those 2 techniques show similar
experience. Infect Dis Rep 2022; 14:213–219. performance for paravalvular lesions.

1070-5295 Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved. www.co-criticalcare.com 511

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.

Severe infections

43. Michǎˇlowska I, Stoǩˇlosa P, Mǐˇlkowska M, et al. The role of cardiac computed 61. Evans L, Rhodes A, Alhazzani W, et al. Surviving sepsis campaign: interna-
& tomography in the diagnosis of prosthetic valve endocarditis – A comparison && tional guidelines for management of sepsis and septic shock. Intensive Care
with transthoracic and transesophageal echocardiography and intra-operative Med 2021; 47:1181–1247.
findings. Eur J Radiol 2021; 138:109637. Recent international guidelines for management of sepsis and septic shock.
Retrospective study comparing the performance of cardiac computerized tomo- 62. Tascini C, Attanasio V, Ripa M, et al. Ceftobiprole for the treatment of infective
graphy and echocardiography in prosthetic valve endocarditis. Cardiac CT per- endocarditis: a case series. J Global Antimicrob Resist 2020; 20:56–59.
formed better than echocardiography for paravalvular abscesses, vegetations and 63. Rodrı́guez-Garcı́a R, Rodrı́guez-Esteban MÁ, Garcı́a-Carús E, et al. In vitro
inflammatory infiltration, but worse for para valvular leakage. activity of ceftaroline and ceftobiprole against clinical isolates of Gram-
44. Philip M, Tessonier L, Mancini J, et al. Comparison between ESC and Duke positive bacteria from infective endocarditis: are these drugs potential options
& criteria for the diagnosis of prosthetic valve infective endocarditis. JACC: for the initial management of this disease? Diagn Microbiol Infect Dis 2020;
Cardiovasc Imaging 2020; 13:2605–2615. 98:115153.
Prospectively multicentric study demonstrating that 18F-FDG PET/CT leads to 64. Muñoz P, De la Villa S, Martı́nez-Sellés M, et al. Linezolid for infective
better sensitivity of ESC criteria compared to Duke criteria in suspected prosthetic endocarditis: a structured approach based on a national database experience.
valve endocarditis. Medicine 2021; 100:e27597.
45. Venet M, Jalal Z, Ly R, et al. Diagnostic value of 18F-fluorodeoxyglucose 65. Niederman MS, Baron RM, Bouadma L, et al. Initial antimicrobial management
positron emission tomography computed tomography in prosthetic pulmonary of sepsis. Crit Care 2021; 25:307.
valve infective endocarditis. JACC: Cardiovasc Imaging 2022; 15:299–308. 66. Landersdorfer CB, Nation RL. Key challenges in providing effective antibiotic
46. Mahmood M, Kendi AT, Ajmal S, et al. Meta-analysis of 18F-FDG PET/CT in therapy for critically ill patients with bacterial sepsis and septic shock. Clin
the diagnosis of infective endocarditis. J Nucl Cardiol 2019; 26:922–935. Pharmacol Ther 2021; 109:892–904.
47. Mahmood M, Abu Saleh O. The role of 18-F FDG PET/CT in imaging of 67. Dhaese S, Van Vooren S, Boelens J, et al. Therapeutic drug monitoring of b-
endocarditis and cardiac device infections. Semin Nuclear Med 2020; lactam antibiotics in the ICU. Expert Rev Antiinfect Ther 2020; 18:1155–1164.
50:319–330. 68. Abdul-Aziz MH, Alffenaar J-WC, Bassetti M, et al. Antimicrobial therapeutic
48. Scholtens AM. 18F-Fluorodeoxyglucose PET/computed tomography in en- drug monitoring in critically ill adult patients: a position paper. Intensive Care
docarditis. PET Clin 2020; 15:187–195. Med 2020; 46:1127–1153.
49. Philip M, Delcourt S, Mancini J, et al. 18F-fluorodeoxyglucose positron emis- 69. Flateau C, Riazi A, Cassard B, et al. Streptococcal and enterococcal en-
& sion tomography/computed tomography for the diagnosis of native valve docarditis: time for individualized antibiotherapy? J Antimicrob Chemother
infective endocarditis: a prospective study. Arch Cardiovasc Dis 2021; 2021; 76:3073–3076.
114:211–220. 70. Spellberg B, Chambers HF, Musher DM, et al. Evaluation of a paradigm shift
Prospective diagnostic study showing that 18F-FDG PET/CT has low sensitivity from intravenous antibiotics to oral step-down therapy for the treatment of
but high specificity in native valve endocarditis. infective endocarditis: a narrative review. JAMA Intern Med 2020; 180:769.
50. Abikhzer G, Martineau P, Grégoire J, et al. [18F]FDG-PET CT for the 71. Dashkevich A, Bratkov G, Li Y, et al. Impact of operative timing in infective
evaluation of native valve endocarditis. J Nucl Cardiol 2022; 29:158–165. endocarditis with cerebral embolism—the risk of intermediate deterioration.
51. Khalaf S, Al-Mallah MH. Fluorodeoxyglucose applications in cardiac PET: JCM 2021; 10:2136.
viability, inflammation, infection, and beyond. Methodist Debakey Cardiovasc J 72. Zhang LQ, Cho S-M, Rice CJ, et al. Valve surgery for infective endocarditis
2020; 16:122–129. complicated by stroke: surgical timing and perioperative neurological com-
52. van Hulst AM, van Rijk MC, Bavelaar-Croon CDL, et al. The value of F-18- plications. Eur J Neurol 2020; 27:2430–2438.
& fluorodeoxyglucose positron emission tomography (FDG-PET/CT) in the 73. Bettencourt S, Ferro JM. Acute ischemic stroke treatment in infective en-
intensive care unit: a review. Netherlands J Crit Care 2019; 27:7. docarditis: systematic review. J Stroke Cerebrovasc Dis 2020; 29:104598.
Review of the use and limitations of 18F-FDG-PET/CT in ICU patients. 74. D’Anna L. Endovascular treatment of ischemic large-vessel stroke due to
53. Pijl JP, Londema M, Kwee TC, et al. FDG-PET/CT in intensive care patients infective endocarditis: case series and review of the literature. Neurol Sci
with bloodstream infection. Crit Care 2021; 25:133. 2020; 41:3517–3525.
54. Lau V, Priestap F, Landry Y, et al. Diagnostic accuracy of critical care 75. Leeten K, Jacques N, Lancellotti P, et al. Aspirin or ticagrelor in staphylo-
transesophageal echocardiography vs cardiology-led echocardiography in & coccus aureus infective endocarditis: where do we stand? Front Cell Dev Biol
ICU patients. Chest 2019; 155:491–501. 2021; 9:716302.
55. Cai S, Yang Y, Pan J, et al. The clinical value of valve metagenomic next- A review of antiplatelet treatment as adjunctive therapy to IE management.
generation sequencing when applied to the etiological diagnosis of infective 76. Sun J, Uchiyama S, Olson J, et al. Repurposed drugs block toxin-driven
endocarditis. Ann Transl Med 2021; 9:1490. platelet clearance by the hepatic Ashwell-Morell receptor to clear Staphylo-
56. To RK, Ramchandar N, Gupta A, et al. Use of plasma metagenomic next- coccus aureus bacteremia. Sci Transl Med 2021; 13:eabd6737.
generation sequencing for pathogen identification in pediatric endocarditis. 77. Ulloa ER, Uchiyama S, Gillespie R, et al. Ticagrelor increases platelet-
Pediatr Infect Dis J 2021; 40:486–488. mediated Staphylococcus aureus killing, resulting in clearance of bacteremia.
57. Lieberman JA, Naureckas Li C, Lamb GS, et al. Case report: Comparison of J Infect Dis 2021; 224:1566–1569.
plasma metagenomics to bacterial PCR in a case of prosthetic valve en- 78. Lerche CJ, Schwartz F, Pries-Heje MM, et al. Potential advances of adjunctive
docarditis. Front Pediatr 2021; 8:575674. & hyperbaric oxygen therapy in infective endocarditis. Front Cell Infect Microbiol
58. Miller RJH, Chow B, Pillai D, et al. Development and evaluation of a novel fast 2022; 12:805964.
broad-range 16S ribosomal DNA PCR and sequencing assay for diagnosis of A review of hyperbaric oxygen therapy as an adjunctive therapy to IE.
bacterial infective endocarditis: multiyear experience in a large Canadian 79. Shmueli H, Thomas F, Flint N, et al. Right-sided infective endocarditis 2020:
healthcare zone and a literature review. BMC Infect Dis 2016; 16:146. challenges and updates in diagnosis and treatment. JAHA 2020; 9. doi: 502
59. Rodrı́guez-Garcı́a R, Rodrı́guez-Esteban MÁ, Fernández-Suárez J, et al. Eva- 10.1161/JAHA.120.017293.
luation of 16S rDNA heart tissue PCR as a complement to blood cultures for 80. Yang T, Li Z-Q, Li H-L, et al. Aminoglycoside use and intensive care unit-
the routine etiological diagnosis of infective endocarditis. Diagnostics 2021; acquired weakness: a systematic review and meta-analysis. PLoS One 2020;
11:1372. 15:e0230181.
60. Luque Paz D, Lakbar I, Tattevin P. A review of current treatment strategies for 81. Lebeaux D, Fernández-Hidalgo N, Pilmis B, et al. Aminoglycosides for infective
infective endocarditis. Expert Rev Antiinfect Ther 2021; 19:297–307. endocarditis: time to say goodbye? Clin Microbiol Infect 2020; 26:723–728.

512 www.co-criticalcare.com Volume 28  Number 5  October 2022

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.