Transcript Slide 1

Computer-based monitoring of global cardiovascular dynamics
during acute pulmonary embolism and septic shock in swine
JA Revie, J Stevenson, JG Chase, BC Lambermont, A Ghuysen, P Kolh, GM Shaw, T Desaive.
Objective
 Retrospectively test a model-based monitoring method on porcine measurements.
Fit subject-specific cardiovascular models from available or inferable ICU data.
Track hemodynamic changes due to acute pulmonary embolism (APE) and septic shock.
(SS).
Retrospectively identify pig-specific cardiovascular models for improved monitoring of
common ICU diseases.
Experimental Protocol
 9 pigs were pre-medicated , ventilated, and anesthetised as explained in Lambermont
et al (2003).
Cardiovascular model
In 5 pigs 3 autologous blood clots were inserted at t0, t120, and t240 minutes into the
trials to simulate APE.
Measurement
ABPsys (mmHg)
ABPdia (mmHg)
PAPsys (mmHg)
PAPdia (mmHg)
SV (ml)
GEDV (ml)
CVPmean (mmHg)
ECGtpeak (msec)
In the other 4 pigs, and endotoxin infusion over t0-t30 was used to induce SS.
6 main cardiac measurement were recorded:
•
•
•
Left and right ventricular volume waveforms (open heart catheterisation)
Left and right ventricular pressure waveforms (open heart catheterisation)
Aortic and pulmonary artery pressure waveforms
Measurements were recorded every 30 minutes for up to 4.5 hours.
.
Model Fitting
80 subject-specific models of the cardiovascular system were identified.
Model identification was achieved using only equivalent measurements of readily
available or inferable ICU data:
•
•
•
•
•
Value
120
80
40
15
70
Aorta
Lung
Capillaries
Pulmonary
artery
Pulmonary
vein
Left
Right
ventricle ventricle
Vena
Cava
Body
Capillaries
Features of the aortic and pulmonary artery pressure waveforms
SV
Global end diastolic volume (GEDV)
CVP
ECG
Aortic pressure
Parameter trends were tracked to monitor the progression of SS and APE.
Left ventricle P-V loop
125
120
Modeled maximum left and right ventricular pressure were identified to median errors
of <7%.
In APE pulmonary afterload tracked experimentally derived afterload to R2 = 081
In SS pulmonary afterload tracked experimentally derived afterload to R2 = 0.95
In both studies RVEDV increased significantly and LVEDV decreased indicating a leftward
shift in the intra-ventricular septum
110
Pmean
105
100
dPmax
This model-based method shows clinical potential. Hemodynamic measurements, from
different monitoring devices, can be aggregated into an easy to understand physiological
form, to assist with diagnosis and real-time monitoring of cardiac and circulatory state.
However, further human clinical trials are required to test the technology, which are
currently underway at the Christchurch Hospital ICU.
Ea
60
Measured 
40
 Model
20
90
85
0
0.1
0.2
0.3
0.4
0.5
0
0.6
0
10
20
30
40
Time (s)
50
60
70
80
Volume (ml)
Systemic vascular resistance
Pulmonary vascular resistance
1.6
3.2
3
Resistance (mmHg.s/ml)
All trends derived from the subject-specific cardiovascular models were consistent with
known trends of septic shock and pulmonary embolism, including an increase in
pulmonary vascular resistance in both diseases, and a decrease in systemic resistance in
septic shock.
Ees
95
Discussions
The model matched independent measures (LVEDV, RVEDV, and max left and right
ventricle pressures), not used during identification, to median errors of less than 10%,
which is smaller than most measurement errors.
PP
80
1.4
APE
Resistance (mmHg.s/ml)
Modeled LVEDV and RVEDV matched measured values to median errors of <7.5%.
Measured 
115
100
 Model
Pressure (mmHg)
Results
Aortic pressure (mmHg)
120
2.8
2.6
2.4
2.2
2
1.8
SS
1.2
1
0.8
0.6
0.4
1.6
1.4
0.2
0
50
100
150
Time (sec)
200
250
0
50
100
150
Time (sec)
200
250
90