Introduction
Lumped-parameter modeling
Results
Conclusions
Characterizing the cardiovascular functions
during atrial fibrillation through
lumped-parameter modeling
Stefania Scarsoglio1
Andrea Guala
Carlo Camporeale2
2
2
Luca Ridolfi2
1
Department of Mechanical and Aerospace Engineering, Politecnico di Torino, Italy
Department of Environment, Land and Infrastructure Engineering, Politecnico di Torino, Italy
19th International Conference on Mechanics in Medicine and
Biology
3-5 September 2014, Bologna, Italy
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
General aspects
State of the art
Present work
What is atrial fibrillation (AF)
AF is the most common arrythmia due to the disorganized electrical activity of the atria, causing irregular and rapid heartbeats;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
General aspects
State of the art
Present work
What is atrial fibrillation (AF)
AF is the most common arrythmia due to the disorganized electrical activity of the atria, causing irregular and rapid heartbeats;
Symptoms: palpitations, chest discomfort, anxiety, fall in blood
pressure, decreased exercise tolerance, pulmonary congestion;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
General aspects
State of the art
Present work
What is atrial fibrillation (AF)
AF is the most common arrythmia due to the disorganized electrical activity of the atria, causing irregular and rapid heartbeats;
Symptoms: palpitations, chest discomfort, anxiety, fall in blood
pressure, decreased exercise tolerance, pulmonary congestion;
Higher incidence with age: 2.3% of people older than 40 years
are affected, up to more than 8% of people older than 80 years;
Prevalence is markedly amplifying in industrialized countries;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
General aspects
State of the art
Present work
What is atrial fibrillation (AF)
AF is the most common arrythmia due to the disorganized electrical activity of the atria, causing irregular and rapid heartbeats;
Symptoms: palpitations, chest discomfort, anxiety, fall in blood
pressure, decreased exercise tolerance, pulmonary congestion;
Higher incidence with age: 2.3% of people older than 40 years
are affected, up to more than 8% of people older than 80 years;
Prevalence is markedly amplifying in industrialized countries;
In the USA and Europe 7 million people are currently affected
by AF ⇒ incidence is expected to double within the next 40 years;
AF is responsible for substantial morbidity and mortality in the
general population;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
General aspects
State of the art
Present work
What is atrial fibrillation (AF)
AF is the most common arrythmia due to the disorganized electrical activity of the atria, causing irregular and rapid heartbeats;
Symptoms: palpitations, chest discomfort, anxiety, fall in blood
pressure, decreased exercise tolerance, pulmonary congestion;
Higher incidence with age: 2.3% of people older than 40 years
are affected, up to more than 8% of people older than 80 years;
Prevalence is markedly amplifying in industrialized countries;
In the USA and Europe 7 million people are currently affected
by AF ⇒ incidence is expected to double within the next 40 years;
AF is responsible for substantial morbidity and mortality in the
general population;
Broad interest: statistical analyses on the heartbeat distributions, risk factors, correlation with other cardiac pathologies.
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
General aspects
State of the art
Present work
Open key aspects
Several key points during AF are still not completely understood from
literature data:
Pulmonary and systemic arterial pressures: hypotension, normotension and hypertension seem to be equally probable;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
General aspects
State of the art
Present work
Open key aspects
Several key points during AF are still not completely understood from
literature data:
Pulmonary and systemic arterial pressures: hypotension, normotension and hypertension seem to be equally probable;
In vivo measures: (i) difficulty due to the heart rate variability, (ii)
necessity of immediate medical treatment;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
General aspects
State of the art
Present work
Open key aspects
Several key points during AF are still not completely understood from
literature data:
Pulmonary and systemic arterial pressures: hypotension, normotension and hypertension seem to be equally probable;
In vivo measures: (i) difficulty due to the heart rate variability, (ii)
necessity of immediate medical treatment;
The anatomical and structural complexity of some regions (e.g.,
right ventricle) makes estimates not always feasible and accurate
⇒ substantial absence of well-established information;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
General aspects
State of the art
Present work
Open key aspects
Several key points during AF are still not completely understood from
literature data:
Pulmonary and systemic arterial pressures: hypotension, normotension and hypertension seem to be equally probable;
In vivo measures: (i) difficulty due to the heart rate variability, (ii)
necessity of immediate medical treatment;
The anatomical and structural complexity of some regions (e.g.,
right ventricle) makes estimates not always feasible and accurate
⇒ substantial absence of well-established information;
Presence of other pathologies (hypertension, atrial dilatation,
mitral stenosis, ...) ⇒ the specific role of AF is not easily detectable and distinguishable. Side pathology is cause or effect?
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
General aspects
State of the art
Present work
Motivation and Goal
Understand and quantify, through a stochastic modeling approach,
the impact of paroxysmal AF on the cardiovascular system of
a healthy young adult (structural remodeling effects neglected);
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
General aspects
State of the art
Present work
Motivation and Goal
Understand and quantify, through a stochastic modeling approach,
the impact of paroxysmal AF on the cardiovascular system of
a healthy young adult (structural remodeling effects neglected);
AF can be analyzed without other pathologies ⇒ highlight single cause-effect relations, trying to address from a mechanistic point of view the cardiovascular feedbacks which are currently
poorly understood.
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
General aspects
State of the art
Present work
Motivation and Goal
Understand and quantify, through a stochastic modeling approach,
the impact of paroxysmal AF on the cardiovascular system of
a healthy young adult (structural remodeling effects neglected);
AF can be analyzed without other pathologies ⇒ highlight single cause-effect relations, trying to address from a mechanistic point of view the cardiovascular feedbacks which are currently
poorly understood.
The main cardiac parameters can all be obtained at the same
time, while clinical studies usually focus only on a few of them at a
time ⇒ overall good agreement with the clinical state-of-theart measures;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
General aspects
State of the art
Present work
Motivation and Goal
Understand and quantify, through a stochastic modeling approach,
the impact of paroxysmal AF on the cardiovascular system of
a healthy young adult (structural remodeling effects neglected);
AF can be analyzed without other pathologies ⇒ highlight single cause-effect relations, trying to address from a mechanistic point of view the cardiovascular feedbacks which are currently
poorly understood.
The main cardiac parameters can all be obtained at the same
time, while clinical studies usually focus only on a few of them at a
time ⇒ overall good agreement with the clinical state-of-theart measures;
Accurate statistical analysis of the cardiovascular dynamics, which
is not easily accomplished by in vivo measurements.
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
Mathematical framework
Cardiac cycle simulation
Cardiovascular scheme
Pulmonary
Arterial Sinus
R_pas
P: pressure
L_pas
Pulmonary
Artery
Q_pas
R_pat
Pulmonary
Arteriole
L_pat
Q_pat
Pulmonary
Capillary
R_par
Pulmonary
Vein
R_pcp
R_pvn
P_pat
P_pvn
C_pas
C_pat
C_pvn
Pulmonary Circuit
V: volume
P_pas
Q: flow rate
Q_pvn
Right Heart
Left Heart
V_ra P_ra
Q_svn
P_sas
P_la V_la
E_ra
E_la
C_sas
Q_ti
C: compliance
Systemic
Vein
CQ_ti
CQ_mi
L: inductance
R: resistance
P_lv
E_rv
R_svn
C_svn
Systemic
Aortic
Sinus
L_sas
V_rv P_rv
E: elastance
R_sas
Q_mi
Q_sas P_sat
E_lv
CQ_po
CQ_ao
Q_po
P_svn
V_lv
C_sat
R_sat
Q_ao
L_sat
Systemic Circuit
R_scp
Systemic
Capillary
S. Scarsoglio, ICMMB 2014
R_sar
Q_sat
Systemic
Arteriole
Lumped-parameter modeling of atrial fibrillation
Systemic
Artery
Introduction
Lumped-parameter modeling
Results
Conclusions
Mathematical framework
Cardiac cycle simulation
Physiologic and fibrillated beating
Normal Sinus Rhythm (NSR)
RR extracted from a correlated pink Gaussian distribution;
Time varying (right and left) atrial elastance;
Full left ventricular contractility;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
Mathematical framework
Cardiac cycle simulation
Physiologic and fibrillated beating
Normal Sinus Rhythm (NSR)
RR extracted from a correlated pink Gaussian distribution;
Time varying (right and left) atrial elastance;
Full left ventricular contractility;
Atrial Fibrillation (AF)
RR extracted from an exponentially modified Gaussian distribution;
Constant (right and left) atrial elastance;
Reduced left ventricular contractility;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
Mathematical framework
Cardiac cycle simulation
Physiologic and fibrillated beating
Normal Sinus Rhythm (NSR)
RR extracted from a correlated pink Gaussian distribution;
Time varying (right and left) atrial elastance;
Full left ventricular contractility;
Atrial Fibrillation (AF)
RR extracted from an exponentially modified Gaussian distribution;
Constant (right and left) atrial elastance;
Reduced left ventricular contractility;
8
2
AF
NSR
NSR
4
2
0
0
1.5
AF
RR [s]
p(RR)
6
1
0.5
0.6
1.2
RR[s]
1.8
S. Scarsoglio, ICMMB 2014
0
0
1000
2000
t [s]
Lumped-parameter modeling of atrial fibrillation
3000
Introduction
Lumped-parameter modeling
Results
Conclusions
Hemodynamic parameters
Systemic arterial pressure
Left atrium
Flow rates
Left ventricle
120
Plv [mmHg]
Plv [mmHg]
120
80
40
0
40
80
120
Vlv [ml]
S. Scarsoglio, ICMMB 2014
80
40
0
40
80
120
Vlv [ml]
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
Hemodynamic parameters
Systemic arterial pressure
Left atrium
Flow rates
Left ventricle
80
40
CO [l/min]
0
40
80
40
0
40
80
120
Vlv [ml]
80
120
Vlv [ml]
8
4
0
EF [%]
120
Plv [mmHg]
Plv [mmHg]
120
Cardiac Cycles
5.000
Cardiac Cycles
5.000
60
40
20
0
S. Scarsoglio, ICMMB 2014
CO [l/min]
SV [ml]
EF [%]
SW [J]
NSR
4.80
63.84
53.27
0.87
AF
4.38
47.21
37.12
0.57
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
Hemodynamic parameters
Systemic arterial pressure
Left atrium
Flow rates
Arterial pressure: time series and statistics
0.3
(a)
(b)
100
p(Psas)
Psas [mmHg]
120
80
0.15
NSR syst
AF syst
NSR dias
AF dias
60
40
589
590
591
t[s]
Psas [mmHg]
NSR
AF
592
Mean
99.52
89.12
0
40
593
Systolic
116.22
103.66
60
Diastolic
83.24
77.24
80
Psas
100
Pulsatile
32.99
26.42
Scarsoglio et al., Med. & Biol. Eng. & Comput., 2014 (in press).
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
120
Introduction
Lumped-parameter modeling
Results
Conclusions
Hemodynamic parameters
Systemic arterial pressure
Left atrium
Flow rates
Pressure and volume behaviour
80
plateau
Vla [ml]
Pla [mmHg]
12 (a)
10
8
rapid
grow
atrial kick
589
590
591
t[s]
Vla [ml]
NSR
AF
592
Mean
56.53
65.95
(b)
60
40
593
589
End-Systolic
64.41
71.41
S. Scarsoglio, ICMMB 2014
plateau
atrial kick
590
591
t[s]
rapid grow
592
End-Diastolic
55.37
68.84
Lumped-parameter modeling of atrial fibrillation
593
Introduction
Lumped-parameter modeling
Results
Conclusions
Hemodynamic parameters
Systemic arterial pressure
Left atrium
Flow rates
Left heart: mitral and aortic flows
1500 (b) 47% reduction
atrial
kick
Qao [ml/s]
Qmi [ml/s]
1000 (a)
500
0
589
1000
500
0
590
591
t[s]
592
593
589
590
591
t[s]
592
593
Qmi and Qao : the increased portion of regurgitant flow during short
beats is not systematically accompanied by a higher contribute
of direct flow ⇒ possible functional mitral regurgitation and
aortic valve insufficiency;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
Hemodynamic parameters
Systemic arterial pressure
Left atrium
Flow rates
Right heart: tricuspid and pulmonary flows
atrial
kick
1000 (b)
Qpo [ml/s]
Qti [ml/s]
800 (a)
400
0
500
0
131% increase
589
590
591
t[s]
592
593
589
590
591
t[s]
592
593
Qti and Qpo : the greater amount of regurgitant flow due to a rapid
beat is in large part compensated by a greater amount of direct
flow ⇒ right valves insufficiency is less likely to occur.
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
Conclusions
Discussion and Conclusive Remarks
First attempt to quantify, through a stochastic modeling, the role
of acute AF on the whole cardiovascular system;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
Conclusions
Discussion and Conclusive Remarks
First attempt to quantify, through a stochastic modeling, the role
of acute AF on the whole cardiovascular system;
Anatomical remodeling due to long-term effects and short-term regulation effects of the baroreceptor mechanism are absent;
Reduced contractility of the right ventricle and the ventricular interaction should be properly accounted for;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
Conclusions
Discussion and Conclusive Remarks
First attempt to quantify, through a stochastic modeling, the role
of acute AF on the whole cardiovascular system;
Anatomical remodeling due to long-term effects and short-term regulation effects of the baroreceptor mechanism are absent;
Reduced contractility of the right ventricle and the ventricular interaction should be properly accounted for;
Isolate single cause-effect relations, a thing which is not possible in real medical monitoring:
the drops of systemic arterial pressure and cardiac output are entirely induced by the reduced ventricular contractility during AF;
the decrease of the ejection fraction and the LA enlargement are
primarily caused by the irregular heart rate;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
Conclusions
Discussion and Conclusive Remarks
Moderate systemic hypotension and left atrial enlargement
should be interpreted as pure consequences of AF alone and
not induced by other pathologies;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
Conclusions
Discussion and Conclusive Remarks
Moderate systemic hypotension and left atrial enlargement
should be interpreted as pure consequences of AF alone and
not induced by other pathologies;
Accurate statistical description of the cardiovascular dynamics,
a task which is rarely accomplished by in vivo measurements;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
Conclusions
Discussion and Conclusive Remarks
Moderate systemic hypotension and left atrial enlargement
should be interpreted as pure consequences of AF alone and
not induced by other pathologies;
Accurate statistical description of the cardiovascular dynamics,
a task which is rarely accomplished by in vivo measurements;
New information on hemodynamic parameters (e.g., flow rates),
difficult to measure and almost never treated in literature;
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation
Introduction
Lumped-parameter modeling
Results
Conclusions
Conclusions
Discussion and Conclusive Remarks
Moderate systemic hypotension and left atrial enlargement
should be interpreted as pure consequences of AF alone and
not induced by other pathologies;
Accurate statistical description of the cardiovascular dynamics,
a task which is rarely accomplished by in vivo measurements;
New information on hemodynamic parameters (e.g., flow rates),
difficult to measure and almost never treated in literature;
Future work:
Response to AF with the combined presence of altered cardiac conditions (e.g., left atrial appendage clamping);
Modeling response to real beating series for both NSR and AF.
S. Scarsoglio, ICMMB 2014
Lumped-parameter modeling of atrial fibrillation