CARDIOVASCULAR JOURNAL OF AFRICA • Volume 25, No 2, March/April 2014
50
AFRICA
An echocardiographic study of infective endocarditis,
with special reference to patients with HIV
SH Nel, DP Naidoo
Abstract
Objective:
The aim was to describe the echocardiographic
features of patients with infective endocarditis (IE), and to
compare the manifestations of IE in HIV-positive versus
HIV-negative patients.
Methods:
The study was prospective in nature and screened
patients referred to Inkosi Albert Luthuli Hospital (IALCH)
with suspected IE between 2004 and 2007. Only patients with
a definite diagnosis of IE according to the modified Duke
criteria were enrolled for the purpose of the study. Inkosi
Albert Luthuli hospital is an 842-bed tertiary referral centre,
serving a KwaZulu-Natal population of 10 million people,
who are of various races.
Results:
During this period, 91 patients were screened for
IE. Seventy-seven (HIV infected,
n
=
17) satisfied the criteria
for a definite diagnosis of IE. Blood cultures were positive
in 46% of cases. The commonest organism was
S aureus
.
Most patients had advanced valve disruption with heart
failure and high peri-operative mortality. The clinical profile
in the HIV-infected patients was similar to the that of the
non-infected patients. The prevalence of echocardiographic
complications (abscesses, aneurysms, perforations, fistulae
and chordal ruptures) was 50.6% in the whole group. Except
for the presence of leaflet aneurysms and root abscesses in
four advanced (CD
4
counts
<
250 /mm
3
) HIV-infected cases,
complications were not more frequent in the HIV-infected
group.
Conclusion:
There was a high rate of culture-negative cases
in this study, probably related to prior antibiotic usage; in this
setting the modified Duke criteria have diagnostic limita-
tions. No significant differences in the clinical presentation of
infective endocarditis were noted between HIV-infected and
HIV-negative patients.
Keywords:
infective endocarditis, HIV, rheumatic heart disease
Submitted 8/3/12, accepted 29/11/13
Cardiovasc J Afr
2014;
25
: 50–57
DOI: 10.5830/CVJA-2013-084
Important developments during the last 20 years have facilitated
rapid and accurate diagnosis of infective endocarditis (IE),
and recent guidelines emphasise the role of early surgical
treatment when complications supervene.
1,2
The emergence of
prosthetic valve endocarditis, catheter-related endocarditis, and
an increased incidence of antibiotic resistance has led to new
challenges for the physician.
3
From a microbiological standpoint,
the rise in staphylococcal infections, and the immune paresis
associated with HIV infection pose further diagnostic challenges
that also have important implications for management.
3
Bacteraemia is said to be common in HIV-positive patients,
due to the numerous immunological defects present in this
disease.
4
Furthermore, in the setting of HIV exposure and altered
immunity, infection is not uncommonly caused by unusual
organisms, such as barbonella, salmonella, and listeria.
1
This
raises the question as to whether IE presents a somewhat different
clinical and echocardiographic picture in the HIV-positive
patient.
It is known that the degree of immunosuppression, manifested
by a reduced CD
4
lymphocyte count, strongly correlates with
the presence of echocardiographic abnormalities.
5
Whether
the immunosuppression associated with HIV may alter the
clinical picture of valvular heart disease, particularly IE, is not
clear. Since a decrease in CD
4
count is thought to predispose
to HIV-associated cardiac disease, this study was designed to
determine the pattern of cardiac involvement in the HIV-infected
subjects who develop IE.
5,6
Methods
The study prospectively screened a total of 91 patients with
features of suspected IE between 2004 and 2007. The diagnosis
of IE was made on clinical grounds. The modified Duke criteria
were then used to classify IE as definite or probable.
7
Only
patients with a definite diagnosis of IE according to the modified
Duke criteria were enrolled for the purpose of the study. Inkosi
Albert Luthuli Hospital (IALCH) is an 842-bed tertiary referral
centre, serving a population of 10 million people in whom
rheumatic heart disease is endemic.
The study protocol was approved by the Nelson R Mandela
Research Ethics Committee (H095/04). The study has been
structured in accordance with the Declaration of Helsinski
(2000), which deals with research involving human subjects.
All patients with suspected IE referred from peripheral
hospitals to the Department of Cardiology at IALCH were
assessed by clinicians who documented the clinical features of
IE. Blood sampling was performed for estimation of erythrocyte
sedimentation rate (ESR), C-reactive protein (CRP), serum
complement and blood cultures. Urine tested for microscopic
haematuria and 12-lead electrocardiograms were performed on
all patients.
All study participants were tested for HIV (diagnosis of
HIV was determined by an ELISA test), after adequate pre-test
counselling by a qualified counsellor. If the results were positive,
a CD
4
count was done. The stage of HIV infection was assessed
Department of Cardiology, University of KwaZulu-Natal,
Durban, South Africa
SH Nel, MMed Sc,
DP Naidoo, FRCP,
