Infective endocarditis
Guillermo Martinez MD
Kamen Valchanov FRCA
Key points
Infective endocarditis (IE) is
a systemic life-threatening
disease mainly affecting
patients with heart valve
disease, prosthetic valve,
intracardiac devices, and i.v.
drug abusers.
Clinical findings,
echocardiography, and
blood cultures are the
cornerstone of IE
diagnostics, and serological
tests and polymerase chain
reaction may be useful in
culture-negative patients.
Transoesophageal
echocardiography is
recommended in all patients
with suspected or
confirmed IE.
When heart failure or large
vegetations (.10 mm) are
present, early surgery is
recommended and
associated with improved
long-term clinical outcomes.
Antibiotic prophylaxis solely
to prevent IE is no longer
recommended for persons
at risk of IE.
Guillermo Martinez MD
Consultant in Anaesthesia and Intensive
Care
Department of Anaesthesia
and Intensive Care
Papworth Hospital, UK
Kamen Valchanov FRCA
Consultant in Anaesthesia and Intensive
Care
Department of Anaesthesia and
Intensive Care
Papworth Hospital NHS Foundation
Trust Cambridge CB23 3RE, UK
Tel: þ44 1480 830541/364406
Fax: þ44 1480 364936
E-mail: kamen.valchanov@papworth.nhs.
uk
(for correspondence)
134
‘No one can measure his own days, one
must resign oneself, it will be as provi-
dence wills, and so I finish my death-song;
I must not leave it incomplete.’
Mozart (The Mozart Myths, William
Stafford, Stanford University Press, 1991)
There are many theories behind Mozart’s mys-
terious final illness and no reliably confirmed
remains have been found. One of the theories
of his demise is that it was endocarditis that
killed him.1
Introduction
Infective endocarditis (IE) is a microbial infec-
tion of a heart valve (native or prosthetic) or
the mural endocardium, leading to tissue de-
struction and formation of vegetations.
It is primarily a disease of the heart, but by
virtue of its haematogenic spread, it is also a
multisystem disorder. The aim of this article is
to review the epidemiological and microbio-
logical profile of IE, as well as pathophysiology,
clinical presentation, and management of
complications.
Aetiology and epidemiology
The incidence of IE is 1.7–7.2 cases per 100
000 person-years. The female to male ratio
has remained stable over the years at 1:2.2
However, the median age of endocarditis
patients has increased from 30–40 to 47 –69 yr
and rheumatic heart disease is no longer the
main risk factor for IE in Western countries.
Increasing longevity, degenerative valve
disease, and medical treatment, including pros-
thetic heart valves and indwelling devices such
as pacemakers and implanted defibrillators, are
the main factors responsible for these substan-
tial changes in the epidemiological profile over
the last few decades.3
The majority of cases of IE are caused by
gram-positive bacteria. Staphylococcus aureus
is now more common than oral Streptococci
(formerly Streptococcus viridans) and it has
doi:10.1093/bjaceaccp/mks005
Continuing Education in Anaesthesia, Critical Care & Pain | Volume 12 Number 3 2012
& The Author [2012]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia.
All rights reserved. For Permissions, please email: journals.permissions@oup.com
Matrix reference 3G00, 2A12
become the most frequent microorganism
causing IE (31–54%). The clinical profiles of
methicillin-sensitive S. aureus (MSSA) and
methicillin-resistant S. aureus (MRSA) are dif-
ferent. MSSA is more frequently isolated in
community-acquired IE, affects mainly native
valves, and it is associated with bacteraemia of
unknown origin. On the other hand, MRSA IE
is predominantly related to nosocomial infec-
tion, wound infection, permanent i.v. catheters,
or surgical intervention in the previous 6
months.
Viridans group IE is now less common
(17– 26%) than in the past, but these strains
can be challenging to treat due to difficulties in
isolation and partial resistance to antibiotics
(‘penicillin tolerance’).
Coagulase-negative Staphylococci were the
main cause of prosthetic valve endocarditis in
the past, particularly within the first 6–12
months after valve surgery. However, current
data show that coagulase-negative Staphylo-
cocci are isolated in only 17% of prosthetic
valve endocarditis, whereas MRSA is isolated
in 23–31% of cases.4 Some gram-positive or-
ganism such as Streptococcus bovis (5– 8%) can
be associated with bowel malignancy or other
mucosal lesions.
Gram-negative microorganisms can also cause
IE. The slow-growing HACEK (Haemophilus
parainfluenzae, Aggregatibacter aphrophilus,
Cardiobacterium hominis, Eikenella corrodens,
and Kingella kingae) group is a well recognized
but unusual cause of IE, responsible for 1.8–
3% of cases. The HACEK group affects mainly
native valves, although up to one-third of the
cases involve prosthetic valves.
Candida and Aspergillus species cause the
majority of fungal IE (1–3% of IE). I.V. drug
abusers, prosthetic-valve recipients, and
patients with long-term central venous catheters
are at highest risk of fungal IE, which should
be suspected in the presence of bulky vegeta-
tions, metastatic infection, perivalvular inva-
sion, or embolization to large blood vessels,
despite negative blood cultures.
Advance Access publication 29 February, 2012

Whenever blood-culture-negative IE is suspected, other organ-
isms such as Coxiella burnetti, Legionella spp., Brucella spp.,
Bartonella spp., and Chlamydiae spp. must be considered.
Pathophysiology
IE originates at sites where the endothelium is damaged by high
blood velocity or mechanical damage and on foreign bodies in the
circulation.
Initially, a sterile thrombotic vegetation (non-bacterial throm-
botic endocarditis) is formed, which facilitates bacterial adherence
during transient bacteraemia. Platelets and fibrin deposits at the
injury site provide an adherent surface for the formation of vegeta-
tions. Finally, the vegetations may produce the secondary effects
of endocarditis such as tissue destruction, generalized and difficult
to eradicate sepsis, and septic emboli and abscesses.
Gram-positive bacteria are particularly resistant to the patient’s
bactericidal activity (i.e. complement), which facilitates the adhe-
sion and formation of vegetations. However, there is no evidence
that bacteraemia associated with invasive or semi-invasive inter-
ventions is more significant than that after teeth brushing, for
example, and other patient risk factors play a greater role in the
physiopathology of IE.5
Diagnosis: clinical features, microbiology,
and echocardiography
Clinical suspicion and prompt investigation of IE is imperative. A
multidisciplinary team involving microbiologists, cardiologists,
neurologists, anaesthetists, surgeons, and intensivists should be
involved in caring for these patients.
The modified Duke criteria (Table 1) are based on clinical,
microbiological, and echocardiographic findings, and provide high
sensitivity and specificity (around 80%) for the diagnosis of IE
when applied to patients with native valve IE with positive blood
cultures.6 The diagnosis of IE is confirmed in the presence of two
major criteria, one major and two minor, or five minor criteria. The
diagnosis of IE is considered possible in the presence of one major
and one minor or three minor criteria. However, the modified Duke
Table 1 Simplified Duke criteria for the diagnosis of IE
Major criteria
Positive blood cultures
Positive echocardiogram for IE defined as
Oscillating intracardiac mass
Intracardiac abscess
New partial dehiscence of prosthetic valve
Minor criteria
Predisposition such as heart condition or i.v. drug use
Fever
Vascular phenomena or immunological phenomena such as major arterial emboli,
septic pulmonary infarcts, mycotic aneurysm, intracranial haemorrhage,
conjunctival haemorrhages, and Janeway lesions
Other microbiological evidence such as PCR, serological tests, or positive blood
culture but does not meet a major criterion
Continuing Education in Anaesthesia, Critical Care & Pain j Volume 12 Number 3 2012
Infective endocarditis
criteria have a low sensitivity and cannot reasonably be applied
when blood cultures are negative, when infection affects a prosthet-
ic valve or a pacing system, and when IE affects the right heart.
Clinical features
In 1885, William Osler presented the first comprehensive descrip-
tion of endocarditis. Thereafter, the descriptions of clinical features
of IE were largely based on data obtained several decades ago.
Nowadays, Oslerian peripheral stigmata of IE such as Osler’s
nodes (3%) or Janeway lesions (5%) are uncommon, and physical
examination is often unremarkable. A history of weight loss and
night sweats is frequently described (up to 96% of cases); a new or
a different heart murmur (48% and 20%, respectively) and
elevated C-reactive protein are also common findings.
When the left heart is affected, vegetations most often develop
on the ventricular aspect of the aortic valve and atrial surface of
the mitral valve, usually along the edges of valve leaflets (Fig. 1).
This explains why peripheral embolism is common. Embolic
events usually occur before clinical recognition of the disease, and
up to 30% of the patients have renal or splenic infarction at the
time of diagnosis. In addition, the heart, brain, intestine, and other
large organs may also be affected by septic emboli.
Rapid evolution of IE is common in S. aureus, with no time for
development of immunological phenomena characteristic of sub-
acute IE. This is also the case in i.v. drug users, where right-sided
IE usually involves the tricuspid valve and occasionally the pul-
monary valve. Therefore, instead of systemic vascular phenomena,
septic pulmonary embolism is the most important complication,
which can evolve to pulmonary infarction, pulmonary abscess,
bilateral pneumothoraces, pleural effusion, and empyema.
The severity of valvular destruction varies with the virulence of
the infecting organism and the duration of the infection, and heart
failure can be the initial presentation of IE.
Fig 1 TOE, mid-oesophageal commissural view of mitral valve vegetation
(arrow). LA, left atrium; LV, left ventricle. For associated video, please see
Supplementary material online.
135
Infective endocarditis
Microbiology
A positive blood culture is still the best method for the identifica-
tion of the microorganisms causing IE and it is considered to be a
major diagnostic criteria. Blood cultures are positive in about 80%
of cases, but may be negative in cases of intracellular or fastidious
pathogens or after previous antibiotic treatment. Therefore, when-
ever IE is suspected (i.e. temperature . 388C, new regurgitant
murmur, and history of valvular disease), it is mandatory to
perform blood cultures before starting antibiotic treatment. When
the antimicrobial agents have been administered before blood
cultures are obtained, the recovery rate of bacteria is reduced by
35–40%. Three sets (including at least one aerobic and one anaer-
obic), obtained from a sterile site, are normally sufficient to
identify the usual microorganisms, but some patients may need
repetitive sampling.
Culture-negative IE often delays diagnosis and the initiation of
treatment, with a profound impact on clinical outcome. Specific
serological data can also be used to identify the organisms of
culture-negative IE. Polymerase chain reaction positivity has been
proposed as a major diagnostic criterion for IE, but the technique
seems unlikely to supersede blood cultures as a prime diagnostic
tool. A polymerase chain reaction of excised valve tissue or
embolic material should be performed in patients with negative
blood cultures who undergo valve surgery or embolectomy.
Echocardiography
Echocardiography is important for the diagnosis and management
of patients with IE. Whatever the level of suspicion, a transthoracic
echocardiogram (TTE) should be performed promptly. It is a non-
invasive technique, providing useful information for both the diag-
nosis and severity of IE. However, the sensitivity of TTE ranges
from 45% to 60%, and the quality of the study is not always
adequate ( particularly in obese patients, obstructive lung disease,
or after thoracic surgery). Transoesophageal echocardiography
(TOE) offers better image quality and the overall sensitivity for IE
is 90–100%.7 TOE is mandatory whenever perivalvular complica-
tions or mitral valve involvement is suspected.8
Three echocardiographic findings are the major criteria for
diagnosis of IE: (i) vegetation, a mobile echodense mass attached
to valvular leaflets or mural endocardium (Figs 2 and 3); (ii) peri-
annular abscess; and (iii) new dehiscence of a valvular prosthesis
(Fig. 4). The vegetation is the hallmark lesion of IE. The sensitiv-
ity of TTE and TOE for vegetations is 75% and 90%, respectively.
Around 10% of IE involves the right side of the heart—most com-
monly, the tricuspid valve alone (98%), although the pulmonary
valve and Eustachian valve IE has been reported. Isolated right-
sided vegetation is well detected by TTE and TOE is not manda-
tory, although 15% of IE in i.v. drug users affects left-side valves
and TOE should be considered.
An abscess typically affects the aortic root and presents as a
perivalvular zone of reduced echo density without blood flow
136 Continuing Education in Anaesthesia, Critical Care & Pain j Volume 12 Number 3 2012
Fig 2 Three-dimensional TOE showing a vegetation (arrow) on the mitral
valve. For associated video, please see Supplementary material online.
detected inside (Fig. 3). The sensitivity of TTE for perivalvular
abscess is low (45–50%) compared with TOE (more than 90%).
The diagnosis of an abscess is an indication for early surgery.
Other echocardiographic findings, which are not the major
criteria but may suggest the presence of IE, include aortic or mitral
valve regurgitation, developing as a consequence of valvular
necrosis, perforation, or prolapse. About 50 –60% of patients with
IE develop heart failure due to valve destruction and early surgery
becomes necessary. The mortality of IE patients in heart failure is
80% with non-surgical therapy.
Vegetation size and mobility is important. Stroke complicates
20 –40% of left-sided IE and it is the second most common cause
of death. A vegetation size of .10 mm or sessile vegetations are
independent predictors of stroke and mortality, and early surgery
(within 1 week of diagnosis) is associated with improved long-term
outcomes through a reduction in systemic embolic events com-
pared with non-surgical therapy.
If the vegetations are small or have already embolized, echocar-
diography can produce false-negative results in about 15% of
cases. When clinical suspicion is high, TOE can be repeated after
7 –10 days.
Echocardiographic diagnosis can be difficult in the early stage
of the disease or in patients with intracardiac devices. TOE allows
the identification of high-risk patients and may identify patients
who need surgery.
Prophylaxis of IE
The role of anaesthetists and intensivists is crucial in the treatment
of IE. They are involved in preventing infective complications
associated with indwelling devices, antibiotic prophylaxis, and
antibiotic treatment when required. Anaesthetists are also involved
in echocardiographic and general assessment for patients for
surgery and perioperative care.
In 2008, the National Institute of Clinical Excellence (NICE)
published guidance for antimicrobial prophylaxis of IE in adults

Fig 3 Mitral and aortic valve endocarditis. TOE, long-axis mid-oesophageal view. LA, left atrium; LV, left ventricle; RV, right ventricle; Ao, ascending aorta.
The arrows point to vegetations on the mitral and aortic valve. The arrows in the colour section point to mitral regurgitation and a perforated anterior
mitral leaflet. For associated video, please see Supplementary material online.
Fig 4 TOE, mid-oesophageal short-axis view showing the prosthetic valve
in the aortic position (dotted arrow) and peri-prosthetic aortic root
abscess (continuous arrow).
and children undergoing interventional procedures.9 The NICE
guidance suggests that there is weak evidence to support routine
preoperative antibiotic prophylaxis for persons at risk of IE. It also
states that there is a risk of allergic reactions related to antibiotics
and there are financial and resistance implications from liberal
overuse of antibiotics.
Antibiotic prophylaxis to prevent IE is therefore no longer rou-
tinely recommended. However, in the case of actual infection at
the operative site, antibiotic prophylaxis is still recommended in
high-risk patients (such as acquired valvular heart disease, previous
Continuing Education in Anaesthesia, Critical Care & Pain j Volume 12 Number 3 2012
Infective endocarditis
valve replacement, or structural congenital heart disease, excluding
repaired atrial or ventricular septal defect or patent ductus arterio-
sus). Suggested antibiotic prophylaxis is listed in Table 2.
Treatment
Antibiotics
Microbiology advice should be sought in all cases. Early anti-
microbial therapy is paramount; empirical treatment (flucloxacillin
and gentamicin) is started in most cases and antibiotics are later
adjusted according to the sensitivity of the microorganism. The
addition of an aminoglycoside is associated with side-effects in-
cluding nephrotoxicity and levels should therefore be measured.
Once-daily aminoglycoside regimens are now widely used for
other infections, but data regarding their efficacy in endocarditis
are limited. For patients with intracardiac prosthetic material or
suspected MRSA vancomycin is recommended (adjusted to renal
function), and its levels should be also monitored.
Benzylpenicillin is the first choice when Streptococcus or
Enterococcus penicillin-susceptible strains are isolated, but some-
times it can be started empirically when the presentation of the IE
is indolent.
For vancomycin-resistant MRSA, the use of teicoplanin, lipo-
peptide daptomycin, or oxazilidones (linezolid) is recommended.
The treatment of fungal endocarditis is currently unsatisfactory
and usually requires surgical intervention. Amphotericin B does
not penetrate well into vegetations, although it has been
137
Infective endocarditis
Table 2 Antibiotic prophylaxis and treatment for the more frequent causes of IE
Clinical situation Agent
A Prophylaxis Amoxicillin
B Prophylaxis (allergic to penicillin) Clindamycin
C Empirical treatment or isolated MSSA Flucloxacillin
Gentamicin
D Empirical treatment when risk of MRSA Vancomycin
infection, confirmed MRSA or Gentamicin
penicillin allergy
E Streptococci and Group D streptococci Benzylpenicillin
(in beta-lactam allergic patients start
treatment D without aminoglycoside)
successfully used in Candida endocarditis. Fluconazole is fungi-
static and is only active against some Candida spp. Caspofungin is
usually fungicidal for Candida spp.; however, the penetration of
caspofungin and other echinocandins into the vegetation is
unknown.
I.V. antibiotic medication is normally continued for 4–6 weeks,
with the aim of sterilizing endocarditic vegetations. Patients with
blood-culture-negative IE should be treated in consultation with an
infectious disease specialist. (Standard antibiotic treatment is listed
in Table 2.)
Surgery
Antimicrobial therapy can offer a curative treatment in only 50%
of patients. The other half requires surgery, and the threshold for
early surgical treatment has been lowered in the last few years.
Whenever possible, the surgical aim is valve repair, but most
patients require valve replacement. Patients with IE and large
vegetations, intracardiac abscess (9–14%), or persisting infection
(9–11%) almost always need surgery.10
Anaesthetic management may be difficult, and patients with
mitral or aortic regurgitation are particularly challenging.
Hypotension despite hyperdynamic left ventricular function and
hypoxaemia due to severe pulmonary oedema can complicate the
anaesthetic induction. Some patients may also develop acute right
ventricular dysfunction and severe tricuspid regurgitation. Invasive
monitoring including arterial pressure and central venous pressure
is necessary. Inotropes and vasopressors should be used to main-
tain haemodynamics.
The use of intraoperative TOE is mandatory in order to confirm
the structural defect, examine the morphology and functionality of
all cardiac structures, provide haemodynamic assessment and guid-
ance for fluid and inotropic support, and assess the results of surgi-
cal intervention. Patients with a peri-annular abscess have a higher
risk of para-valvular regurgitation and valve dehiscence after
operation.
The maintenance of adequate antibiotic plasma levels during
and after cardiopulmonary bypass is essential in order to reduce
the risk of prosthetic valve endocarditis.
The current IE perioperative mortality is 5–15%. If sepsis
is under control, however, the surgical mortality is similar to
138 Continuing Education in Anaesthesia, Critical Care & Pain j Volume 12 Number 3 2012
Dosage and route Duration
2 g p.o. or i.v. Single dose
600 mg p.o. or i.v. Single dose
12 g daily divided into 4 –6 doses 4–6 weeks or .6 weeks if prosthetic valve
3 mg kg21 divided into 3 doses. I.V. or i.m 3–5 days or 2 weeks if prosthetic valve
30 mg kg21 24 h21 divided into 2 doses 4–6 weeks or .6 weeks if prosthetic valve
3 mg kg21 divided into 3 doses. I.V. or i.m. 3–5 days or 2 weeks if prosthetic valve
12 – 18 million U day21 i.v., divided into 6 doses 4 weeks
non-infected valve replacement. The most frequent postoperative
complications are persistent septic shock, coagulopathy, acute renal
failure, stroke, refractory heart failure, and conduction abnormalities.
Recurrent IE
The rate of recurrent IE at 5 yr follow-up is around 1.5%
per patient-year. Recurrent IE can be separated into two types:
reinfection and relapse. The term reinfection is primarily used
when a different microorganism produces a new episode of IE in
patients at risk of IE such as previous valve disease or i.v. drug
use. Relapse refers to a repeat episode of IE caused by the same
microorganism as the previous episode and it is normally asso-
ciated with insufficient duration of original treatment, suboptimal
choice of initial antibiotics, and a persistent focus of infection (i.e.
periprosthetic abscess).
Conclusion
IE is an infrequent and dynamic disease. Recent changes in the
epidemiology of IE make the diagnosis a challenge, and traditional
diagnostic criteria are insufficient. Despite modern medical and
surgical therapy, IE is still associated with a high rate of complica-
tions and increased mortality. Early surgery is becoming more
common and TOE should be used for all patients. IE is resource-
consuming, and a multidisciplinary approach is essential to provide
efficient and cost-effective treatment.
Supplementary material
Supplementary material is available at Continuing Education in
Anaesthesia, Critical Care & Pain online. Please note: the supple-
mentary videos are quite short, so readers should play them
continuously to facilitate the understanding of the images.
Declaration of interest
None declared.
References
1. Lee SJK. Infective endocarditis and phlebotomies may have killed Mozart.
Korean Circ J 2010; 40: 611–3

2. Beynon RP, Bahl VK, Prendergast BD. Infective endocarditis. Br Med J
2006; 12: 334–9
3. Prendergast BD. The changing face of infective endocarditis. Heart 2006;
92: 879– 85
4. Fowler VG, Jr, Miro JM, Hoen B et al. Staphylococcus aureus endocarditis:
a consequence of medical progress. J Am Med Assoc 2005; 293: 3012–21
5. Morellion P, Que Ya. Infective endocarditis. Lancet 2004; 10: 139 –49
6. Li JS, Sexton DJ, Mick N et al. Proposed modifications to the Duke
criteria for the diagnosis of infective endocarditis. Clin Infect Dis 2000; 30:
633–8
7. Evangelista A, Gonzalez-Alujas MT. Echocardiography in infective endo-
carditis. Heart 2004; 90: 614– 7
Continuing Education in Anaesthesia, Critical Care & Pain j Volume 12 Number 3 2012
Infective endocarditis
8. The Task Force on the Prevention, Diagnosis, and Treatment of Infective
Endocarditis of the European Society of Cardiology (ESC). Guidelines on
the prevention, diagnosis, and treatment of infective endocarditis. Eur
Heart J 2009; 30: 2369–413
9. National Institute for Health and Clinical Excellence. Antimicrobial
prophylaxis against infective endocarditis. Available from http://guidance.
nice.org.uk/CG64 (accessed 2 February 2011)
10. Remadi JP, Habib G, Nadji G et al. Predictors of death and impact of
surgery in Staphylococcus aureus infective endocarditis. Ann Thorac Surg
2007; 83: 1295– 302
Please see multiple choice questions 21 –24.
139