CARDIOVASCULAR JOURNAL OF AFRICA • Vol 24, No 1, January/February 2013
218
AFRICA
tracking with bedside echocardiography, which can be performed
in much less time and more cost effectively than CMR. We hypoth-
esised that this could be used in the majority of ACHD patients with
average echo windows.
Methods:
We used an industry-developed tracking system and
patented software installed on a standard computer, linked up with
a standard 2D echocardiography machine. Key anatomical points
acquired in expiration were then processed through the software and
sent securely via the internet to the industry server, where the system
generated a three-dimensional model within seconds, providing
volumetric measurements and ejection fraction of the RV, to the user.
Twenty-one (nine male) patients, mean age 27.9 years (range 17–64
years), diagnosis: repaired tetralogy of Fallot 16; pulmonary stenosis
five, underwent standard echocardiography. Key anatomical points
were placed and image reconstruction undertaken. In eight patients,
CMR data were available for comparison.
Results:
Fifteen scans were suitable for reconstruction. Limited
acoustic windows made reconstruction difficult, but poor delinea-
tion of valve planes and RV apex was even more important. In six of
eight patients the echo and CMR-derived RVEF were within 6%. In
two patients with limited echo images, the RVEF differed by
>
10%.
Conclusions:
In the majority of ACHD patients with ToF, this novel
and rapid technique could be performed in out-patients to assess the
RV. Reconstruction can be performed in less than 10 minutes; accu-
racy depends on visualising the right heart valves and the apex. The
use can be extended to other CHD conditions such as systemic RV.
This will greatly reduce costs and waiting times. Further experience
and data are required.
1175: NURSE-LED ECHOCARDIOGRAPHIC SCREENING
FOR RHEUMATIC HEART DISEASE IN FIJI: RESULTS
FROMA PILOT STUDY
Daniel Engelman
1,2
, Joseph Kado
3
, Bo Remenyi
4
, Samantha
Colquhoun
1,4
, Jonathan Carapetis
4
, William May
3
, Nigel Wilson
5
,
Andrew Steer
1,2
1
Centre for International Child Health, University of Melbourne,
Australia
2
Royal Children’s Hospital, Melbourne, Australia
3
Colonial War Memorial Hospital, Suva, Fiji
4
Menzies School of Health Research, Darwin, Australia
5
Starship Children’s Hospital, Auckland, New Zealand
Background:
Rheumatic heart disease causes significant morbidity
and mortality in Fiji, where the prevalence of RHD in children is
approximately 1%. Antibiotic prophylaxis is most effective for mild
cases, many of whom are asymptomatic. Screening with echocardi-
ography is the most sensitive screening approach, but is limited by
lack of resources, particularly experienced cardiologists and techni-
cians. Training nurses in basic echocardiography for screening is a
potential solution.
Hypothesis:
We hypothesised that with simple training and super-
vised field experience in echocardiographic screening for RHD, nurs-
es will have high sensitivity in the identification of children requiring
referral for RHD, when compared with diagnosis by a cardiologist.
Methods:
We designed an eight-week training syllabus for primary
health workers in basic echocardiography for RHD, using a simplified
screening protocol and referral criteria based on significant mitral
or aortic regurgitation. Seven Fijian nurses participated, and were
assessed in a pilot study at the end of the training programme. Nurses
screened 16 children, including selected cases with known RHD. A
technician performed an extended echocardiogram on all children. All
echocardiograms were reported by a paediatric cardiologist.
Results:
Four subjects met the echocardiographic criteria for definite
or borderline RHD, and all were correctly identified for referral by
all nurses (sensitivity 100%). The overall specificity was 45% (range
0–67%); 55 of 56 (98%) nurse-screened echocardiograms were of
adequate quality for diagnosis.
Conclusions:
The high sensitivity observed in this pilot supports
the feasibility of training nurses in echocardiographic screening.
The study has been expanded to a larger fieldwork phase involving
screening 2 000 Fijian school-children, and is powered to formally
assess the sensitivity of nurse-led echocardiographic screening.
These results will assist resource-poor countries to evaluate the utility
of this novel screening strategy.
1177: COMPUTER-BASED SURGICAL PLANNING AND
THE Y-GRAFT: THE NEXT INNOVATION OF FONTAN’S
PROCEDURE
Christopher M Haggerty
1
, Kirk R Kanter
2
, Maria Restrepo
1
, Timothy
C Slesnick
3
, Jarek Rossignac
4
, Thomas L Spray
5
, Mark A Fogel
6
, Ajit
P Yoganathan
1
1
Wallace H Coulter Department of Biomedical Engineering, Georgia
Institute of Technology and Emory University, Atlanta, GA, USA
2
Division of Cardiothoracic Surgery, Emory University, Atlanta,
Georgia, USA
3
Division of Paediatric Cardiology, Department of Paediatrics,
Emory University School of Medicine and Children’s Healthcare of
Atlanta, Atlanta, GA, USA
4
College of Computing, Georgia Institute of Technology, Atlanta,
GA, USA
5
Division of Cardiothoracic Surgery, Children’s Hospital of
Philadelphia, Philadelphia, PA, USA
6
Division of Cardiology, Children’s Hospital of Philadelphia,
Philadelphia, PA, USA
Background:
The total cavopulmonary connection (TCPC) for
single-ventricle lesions creates adverse haemodynamics that are
hypothesised to negatively impact on long-term outcomes. Patient-
specific computational modelling may provide novel means to
improve blood flow characteristics in these patients. This study
reviews our experience with two novel advances: pre-operative surgi-
cal modelling and a bifurcated Fontan baffle (‘Y-graft’).
Methods
: Cardiac magnetic resonance (CMR) images are segmented
to create patient-specific TCPC models and provide flow informa-
tion. A specially designed anatomy editing tool can be used to accu-
rately mimic Fontan surgery by simulating baffle placement. Blood
flow simulations using computational fluid dynamics characterise
haemodynamic metrics (i.e. power loss and hepatic flow distribution)
and compare connection performances.
Results:
Computer-based surgical planning has been prospectively
used for 16 patients. The primary indication has been pulmonary
arterio-venous malformations, which are believed to form in the
absence of hepatic nutrients in the blood reaching pulmonary arterial
segments. Therefore the modelling objective was to optimally distrib-
ute hepatic blood flow to the lungs. In five of 16 patients, a Y-graft
Fontan baffle was the modelling recommendation based on beneficial
flow-distribution characteristics, and was surgically implemented.
Follow up in select patients using arterial oxygen saturations, four-
dimensional phase velocity CMR, and post-operative simulations
have shown favourable clinical outcomes and consistency between
model predictions and operative results. Based on hypothetical ener-
getic improvements, the Y-graft was used for 11 separate patients.
It is too early to conclude on energetic characteristics compared to
traditional extracardiac connections, however operative outcomes
and simulated flow distribution results are both positive.
Conclusions:
Computer-based surgical planning is an exciting new
paradigm for patients with single-ventricle lesions, with the potential
to deliver patient-specific benefit. Findings have motivated the use
of the Y-graft Fontan, which may provide both improved energy effi-
ciency and flow distribution for the TCPC.
1178: CLINICAL CLUES TO IDENTIFY ANOMALOUS
ORIGIN OF LEFT CORONARY ARTERY FROM THE
PULMONARY ARTERY IN PATIENTS DIAGNOSED WITH
CONGENITAL MITRAL REGURGITATION OR DILATED
CARDIOMYOPATHY
Orhan Uzun
1
, Sunitha Sampath
2
, Giridhar Dhandayuthapani
3
, Yusuf
