System Dynamics Highlights the Effect of Maintenance on
Hemodialysis Performance
Ahmad Taher Azar’
Assistant Instructor, Systems and Biomedical Engineering Department
Higher Technological Institute, Tenth Of Ramadan City, Egypt
Tel: +2 048 3674557, Mobile: +2 0109418003
Ahmad _t_azar@yahoo.com
D. Khaled M. Wahba*
Assistant Professor, Systems and Biomedical Engineering Department
Faculty of Engineering, Cairo University, Egypt
Academic Advisor, Regional IT Institute, Cairo, Egypt
Tel: +2 02 737 6006, Fax: +2 02 739 1380
Khaled.wahba@riti.org
Prof. Abdalla S. A. Mohamed?
Chairman Of Systems and Biomedical Engineering Department,
Faculty of Engineering, Cairo University, Egypt
profabdo@ gmail.com
Abstract- The Kt/V value demonstrates the dose of hemodialysis (HD). Several studies
suggest an association between hemodialysis machine maintenance and patient outcomes. It
has been suggested that there is a correlation between dose of dialysis and machine
maintenance. However, in spite of the current practice, there are conflicting reports regarding
the relationship between dose of dialysis or patient outcome, and machine maintenance. In
this article, we will discuss the impact of hemodialysis machine maintenance on dialysis
adequacy Kt/V and session performance by building a system dynamics model to evaluate the
effect of machine maintenance on session performance. We will also mention the
interrelationships of dialysis dose and machine maintenance with respect to these patients.
Key words: Urea kinetic modelling, Hemodialysis, Dialysis adequacy, System dynamics,
Hemodialysis machine maintenance.
I. INTRODUCTION
The most widely used definition of the dose of dialysis is fractional clearance of body water
for urea—the product of dialyzer urea clearance (K) and treatment time (t) divided by the urea
distribution volume (V), or Kt/V [1-5]. Not all hemodialysis patients receive their prescribed
dose of hemodialysis [6,7]. Some studies suggested that only 50% of end stage renal disease
(ESRD) patients in the United States actually receive their prescribed hemodialysis dose. To
prevent the Kt/V for any patient from declining to values below the recommended minimum
delivered dose, practitioners should prescribe doses of hemodialysis that are greater than these
minimum values, nephrologists should prescribe doses of hemodialysis that are higher than
the aforementioned minimum delivered levels [8]. Therefore, the HD Adequacy Work Group
suggests that the prescribed minimum Kt/V be 1.3 for patients dialyzing three times per week.
A variety of factors may result in the actual delivered dose of hemodialysis falling below the
prescribed dose [6,9,10,11].
Common factors include reduction in treatment time, ineffective urea clearance due to access
recirculation, inadequate blood flow to the dialyzer, dialyzer clotting, low blood pump and
dialysate flow, or underestimates of flow due to calibration errors and blood pump tubing
collapse that related to hemodialysis machine maintenance. Maintenance must be inclusive of
periodic maintenance, troubleshooting, and problem maintenance. Perfect preventive
maintenance means that the system is restored to good as new condition. Imperfect preventive
maintenance restores the system to a condition that is between "good as new" and "bad as
old". All maintenance must be performed so that equipment and systems operate efficiently
and effectively. Improper maintenance and repairs can lead to unsafe conditions and reduced
system performance. A strong preventive maintenance program can help in reducing the
frequency of emergency and much corrective maintenance and helps utility managers be
aware of, and plan for, capital equipment replacement. With this in mind, the well-run
maintenance system should provide significant benefits in terms of performance, longevity,
and operating cost control. Hemodialysis machine maintenance is extremely important in
evaluation of adequacy of hemodialysis and in assessing dialysis session performance. The
calibration of dialysate pump and blood pump during periodic maintenance is an essential
component to delivering the prescribed hemodialysis treatment. It is important to know the
dialysis machines (i.e., how they work?, are machines truly volumetric?, what is the facility’s
procedure to replace/repair hemodialysis machines?, who does machine maintenance?, how
often is dialysis staff in serviced on machine issues?, and/or what is the facility's procedure
for periodic maintenance?). The dialysate and blood pump must be kept in calibration in order
to deliver the settings on the machine. The clock must be accurate for the dialysate and
ultrafiltration time. Routine preventative and annual maintenance is vital to provide a safe and
adequate dialysis and must be conducted with careful attention and in a timely fashion. Proper
setting of the dialysis machine to achieve the prescribed blood flow rate can also significantly
impact adequacy over time. Table 1 indicates even a 5-ml decrease in the prescribed blood
flow rate will make a significant impact over a week, a month, and a year’s time. Machine
maintenance is extremely important as the machine may indicate the correct blood flow rate
(BFR); but, if not calibrated correctly it may be delivering more or less. Frequent observation
for fluctuating or decreased blood flow rate can also positively impact the delivery of the
prescribed BFR. Frequent interruption of the blood flow rate may cause a loss of blood
volume as well. Needles and bloodlines should be assessed for positioning and corrected as
soon as possible. Needle and bloodline size should be considered if difficulty in achieving
blood flow and Kt/V is a persistent problem. Also, care must be given to ensure that the
machine is set for the prescribed dialysate flow rate. Again, machine maintenance is vital in
the delivery of the prescription. If the dialysate pump is not correctly calibrated the machine
will not deliver the prescribed dialysate flow rate.
Table 1 Blood Volume Not Cleaned due to a 5 ml Decrease in Prescribed Blood Flow Rate [12]
BER 5mi/min (300 || 3 Hour Dialysis: Loss || 4 Hour Dialysis: Loss |] > Hou Dialysis:
ml/hour) less than of Blood Not Dialyzed | of Blood Not Dialyzed Dialyzed as
Prescribed as Prescribed as Prescribed Nze
Prescribed
[ Per Treatment il 900 ml il 1,200 ml il 1,500 ml ]
[Per Week i 7700 mi i 3,600 ml i 4,500ml
[Per Month i 10,800 mi i 14,400 mi | 13,500m |
Ber sear (p2 140,400 ml 187,200 ml 234,000 ml
Weeks)
Il. METHODOLOGY
A. Research Design
The model discussed in [13] was developed using Vensim DSS v 4.0a simulation software for
formulating, analyzing and comparing various policies to determine optimum level of dialysis
parameters for improved session performance. The simulation results with base case values
and with different test scenarios are presented in [13]. The base case values were selected
based on the experts experience in the field of nephrology and the insights from the research
literature. The research analysis started by developing the mental model (Dialysis
performance causal loop diagram explaining and understanding the complex cause and effect
relationships existing between maintenance and dialysis performance.
B. Model Description
Figure 1 shows the overall causal loop diagram of the system. The causal loop diagram shown
below is divided into two models: (1) The intradialytic model (during dialysis session) which
analyzing the dynamic behavior of various factors that characterizes and controlling the
hemodialysis session management process and (2) The interdialytic model (between dialysis
sessions) which identifying the effect of increasing dialysis adequacy on nutritional status of
the patient which in turn reduces the morbidity rate and the intradialytic complications that
lead to session degradation. First we will analyze the behavior of the system for 240 minute
time period. This time period for simulation was decided based on the period of dialysis
treatment sessions and it is the time for intradialytic model. After that the time horizon was
expanded for 10080 minutes to estimate the weekly BUN profile of the patient and it is the
time for interdialytic model. Therefore, the simulation control parameters that were used for
conducting various simulation runs with different scenarios including the base case values are
listed below:
FINAL TIME (The final time for the simulation) = 10080 Minute
INITIAL TIME (The initial time for the simulation) = 0 Minute
TIME STEP (The time step for the simulation) = 1 Minute
A time step of 1 minute was used so as to give smooth time profiles for the different variables
in the model. This time step is used to calculate some parameters in the simulation model. The
change in model time step will affect the accuracy of the results and hence the model has not
been tested for shorter or longer time steps.
Because the causal loop diagrams are excellent for quickly capturing the hypothesis about the
cause of dynamics, eliciting and capturing the mental models and communicating important
feedbacks [14], the following hypotheses are proposed
1. The overall dialysis session performance is not only a function of dialysis adequacy but
also depends on the frequency of intradialytic complications and overall equipment
effectiveness.
2. Session degradation reduction improves session performance by reducing intradialytic
complications episodes and increasing equipment effectiveness over time.
Intradialytic model
Intradialytic
Overall equipment,
effectiveness
Dialysis Session
Performance
Blood Urea A Rt)
Nitrogen (BUN) anes
?
Modelled dialyzer” *
Urea Distribution Clearance i
Volume ge” 42)
ps)
ARs
Ultrailtration
Ultafiltration
Expected Dialyzer
‘Access .
Rectang
Ans)
SS)
Interdialytic Model
ARs)
Probability of
morbidity
PS,
Protein
catabolic rate
Urea generation
rate
Time averaged urea
concentration
Fig.1. The Overall Causal Loop Diagram Of Hemodiadynamics [13]
Figure 2 shows the loops that highlight the effect of dialysis adequacy and session
degradation on session performance. The hemodialysis session degradation depends on the
overall equipment effectiveness and the intradialytic complications. The first positive
feedback loop (R1) showing that as dialysis adequacy increases, it increases the session
performance which in turn increases the delivered dialysis dose. This loop can be used to test
the first hypothesis stated earlier. The second positive feedback loop (R2) showing that the
increase in the intradialytic complications will increase the session degradation which in tum
decreases the session performance. Decrease in session performance increases the probability
of complications during dialysis. This loop can be used to test the second hypothesis.
Overall equipment
effectiveness
Session
‘Degradation
DialysisA dequacy Dialysis Session (Rap Intradialy tic
KW 4R1) Performance Re Complications
+ PS
Figure 2. Feedback Structure showing the effect of dialysis adequacy and session degradation on the
overall hemodialysis performance
C. Formulating a Simulation Model (Stock & Flow Diagram)
This next step in modeling involves setting up a formal model complete with equations,
parameters and initial conditions that represent the system. The overall stock and flow
diagram was shown in [13] and is shown also in appendix. For each subsystem the assumed
parameters, initial values, variable were ranged to be entered to the system for the sake of
building the stock & flow diagram. After that the equations and graphs that describe the
relationships between the various variables were entered to the system using the Vensim DSS
software and were elicited from the experts in the field of nephrology. They were asked for
their inputs on the units for measurement of different variables, the functional form of the
various equations between variables, parameters of these equations (elicited through graphical
portrayal of key relationships), and the initial values of all stock variables. To show the effect
of the maintenance on the overall hemodialysis session performance, two structures will be
described from the overall stock & flow diagram. These two structures are the hemodialysis
session degradation and the overall session performance.
a. Hemodialysis Session Degradation Structure
Dialysis session degradation structure is shown in figure 3. Session degradation is caused
because of two factors; complications and equipment deficiency.
Session Degradation = IF THEN ELSE (Session Degradation due to Complications + Effect
of Equipment deficiency on Session Degradation (Session Degradation due to Equipment
deficiency) >1, 1, Session Degradation due to Complications + Session Degradation due to
Equipment deficiency) (1)
<Probability of Effect of Complications Effect of Equipment
lications> on Session De; tion deficiency on Session
Complications: eradai _ —
Session a
ap egcadaiian:
Session Degradation
due to Complications
Session Degradation
due to Equipment
Down time Down time rate deficiency
down time Actual Effectiveness
conversion unit of Equipment
i aes Time ef
breakdowns Breakdown rate erations
breakdown Test times
conversion unit
Mean Time To
Repair
Figure 3 The Dialysis Session Degradation Structure
Session degradation is the addition of session degradation due to complications and session
degradation due to equipment deficiency. Session degradation varies from 0 to 1. It can not
take value greater than 1. If the additive impact of complications and equipment deficiency in
session degradation is more than 1, its value is limited to 1 which indicated that the session is
totally degraded. Figure 4 shows the effect of complications on the session degradation.
Graph Lookup - Effect of Complications on Session Degradation
fi)
mt
z
by
acacia
;
min} _y|x-0.7064 = y=1.031 Xmax:|1 _z| Reset Sealing
OK | ClearPoints | Clear AllPoints | Cur>Ref| ClearReference | Ref>Cur| Cancel |
Figure 4 The Effect Of Complications On The Session Degradation.
There is a non-linear relationship between session degradation and the equipment deficiency.
The rate of session degradation increases with the increase in equipment deficiency. The
lookup table (figure 5) shows the non-linear relationship in graphical format.
Graph Lookup - Effect of Equipment deficiency on Session Degradation
is
2
i=)
is
s
2
eS! o)2
Ni =| =lalo
Sa) a 5
SA) ole
25) 3S
Ss
e/S|/S)5
Rls) Ss) S)0
Blo) s| S
| =| oo
S/N] BB
BS
ARISES
g|5| 8/8
gle] 8/8
g|°| 3/8
ales
"138
a/8
3/3
=}
a
=
a
3
a i Y-rnin:
04067 (0.7325
0.4709 (0.8246 |
New < a |
na a
|
-min:|0 _y|-0.6942 © y=1.031 Xmax|1 _x| Reset Sealing
OK | Clear Points | Clear All Points | Cur>Ref | Clear Reference | Ref>Cur | Cancel _|
Figure 5 The Effect Of Equipment Deficiency On The Session Degradation.
The efficiency of dialysis equipment is extremely important in evaluation of adequacy of
hemodialysis. Effective control of equipment and system status means coordinating
operations and maintenance activities. Equipment deficiencies must be promptly identified for
correction in the work control system. A process for post-maintenance testing should be in
place to ensure that all operation of equipment is controlled by approved operating procedures
and that appropriate maintenance and operations personnel are represented during the testing.
As the equipment deficiency increases from 0 to 1, initially session degradation increases at
higher rate than in the later part. According to the above graph, when 60% of deficiencies are
present in the equipment and are critical to the session, the session has been degraded to the
level of 0.93.
In order to improve system availability and reliability, various maintenance policies have
been proposed based on different assumptions and considerations. System maintenance can be
divided into three main categories preventive maintenance, predictive maintenance and
reactive maintenance.
Preventive and predictive maintenance are the proactive strategies for avoiding equipment
breakdowns. The preventive and predictive maintenance are very similar in concept with
some differences in the criterion for determining the need for specific maintenance activities.
Preventive maintenance represents all the actions performed in order to operate a system at an
acceptable level of performance by providing systematic inspection, detection and prevention
of incipient failures. Corrective maintenance represents all the actions performed as a result of
failure to restore a system to acceptable performance level.
The actual overall equipment effectiveness (OEE) can be calculated from maintenance
software as a function of the equipment breakdowns and the down time rate. The output of the
maintenance software (OEE) can be linked with this model to calculate the overall dialysis
session performance as a function of equipment efficiency (figure 6).
Maintenance Hemodia- Overall dialysis
Software Dynamics Session
Performance
Figure 6 The Link Between Maintenance Software Model
And Hemodiadynamics Model
The estimated average failure rate is the probability of a failure occurring during a stated
period of time or cycle, and can be calculated as follows:
Breakdown rate = Equipment breakdowns/T est times (2)
The reciprocal of the breakdown rate is the average life (0). For repairable items the average
life is called the "mean-time-between failures" (MTBF). The "test times or cycles" in equation
(2) is often a combination of the times that the failed piece of equipment or system was
operational plus the times to repair. The Downtime which is referred to as "maintainability",
can be measured in several ways:
« Active repair time includes only time spent in diagnosis and repair.
* Total downtime is the sum of times spent in active diagnosis and repair, delays
waiting for parts, technical support and administrative work, and preventive
maintenance.
OEE can be viewed as the percent of time that equipment would need to run at its maximum
speed in order to attain the actual output of that tool or machine. Hence, the actual equipment
effectiveness can be calculated as a function of maintainability and the equipment breakdown
rate from the following equation:
Actual Effectiveness of Equipment = (100 — Breakdown rate — Down time rate * down time
conversion unit) /100 (3)
The ratio of the actual equipment effectiveness to its theoretical maximum effectiveness
determines the effect of the equipment effectiveness on dialysis adequacy.
b. Dialysis Session Performance Structure
Dialysis session performance subsystem determined when to take the session down for
corrective actions and when to put it back into operations. There are various factors that
govern this decision. These decision rules are decided based on the expert’s inputs. The
dialysis session performance subsystem is shown in figure 7.
oO. oe - Session
Performance Performance Performance
'\ ees Nae Rate
<Actual
Table of Increasing Treatment Time>
Performance
= Calculated Dialysis
Adequacy (Kt)
<Session
Degradation>
Figure 7 Dialysis Session Performance Subsystem
The session performance stock (Dimensionless) is fed into by session performance increasing
rate (Dimensionless/Minute) and is depleted by session performance decreasing rate
(Dimensionless/Minute). The session performance stock is an integral of the performance
increasing rate less the performance decreasing rate.
Session Performance (t) = Session performance (0) + J [performance increasing rate —
performance decreasing rate] dt (4)
Session Performance (0) = 0
Performance of the session is measured using two variables; dialysis adequacy and session
degradation. The following equations can be used to calculate the increasing and decreasing
rates of session performance.
Performance Increasing Rate = IF THEN ELSE (Session Performance = 1, 0, IF THEN
ELSE (Session Degradation <= 0.3: AND: "Calculated Dialysis Adequacy (Kt/V)" = 1.6, 0,
Table of Increasing Performance/A ctual Treatment Time)) (5)
Performance Decreasing Rate = IF THEN ELSE (Session Performance = 1, 0, IF THEN
ELSE (Session Degradation >= 0.5: OR:" Calculated Dialysis Adequacy (Kt/V)" = 0, Session
Performance/A ctual Treatment Time, 0)) (6)
Session performance varies from 0 to 1. It can not take value greater than 1. If the additive
impact of dialysis adequacy and session degradation in session performance is more than 1,
its value is limited to 1 which indicated that the dialysis session reached the optimum
performance. If the dialysis session degradation is more than 50 % or any interruption in
dialysis adequacy occurs then the performance decreases until session degradation becomes
no more than 30% and the dialysis adequacy increases again. If the dialysis adequacy reaches
the optimum value of 1.6 and session degradation is less than 30 % this means that the session
performance reached the desired performance level.
A linear multiple regression analysis was made through 164 patients to obtain an analytical
expression capturing the effect of dialysis adequacy and session degradation on the overall
dialysis performance. Regression analysis is used when to predict a continuous dependent
variable from a number of independent variables. The curve fitting was done using Data Fit
version 8.0.32. Figure 8 shows the model plot where X, represents the dialysis adequacy, X2
represents the session degradation and Y represents the overall session performance.
Model atb*x1 +07x1*2+d*x2
¥
c000000000=
c000000000=
C-=NWEMONDOO
¥
Figure 8 Effect Of Dialysis Adequacy And Session Degradation
On Session Performance through 164 patients.
The regression analysis revealed that the R-squared, which denotes the percentage of variation
in the dependent variable that can be explained by the independent variables is 0.9317,
meaning that approximately 93% of the variability of effect of dialysis adequacy and session
degradation on session performance is accounted for by the variables in the model. In this
case, the adjusted R-squared indicates that about 93.04% of the variability of effect of dialysis
adequacy and session degradation on session performance is accounted for by the model, even
after taking into account the number of predictor variables in the model. The adjusted R-
squared is a measure of how well the independent, or predictor, variables predict the
dependent, or outcome, variable. The adjusted R-squared adjusts the R-square for the sample
size and the number of variables in the regression model. Therefore, the adjusted R-square is
a better comparison between models with different numbers of variables and different sample
sizes. The adjusted R-squared can be computed as:
AdjustedR? =1-(1-R?) 27!
n-k-1 @
Where, n = sample size and k = number of predictors.
The results of regression analysis are summarized in table 2:
10
Table 2. Regression Analysis Results for the Effect of Adequacy and Session Degradation on the Session
Performance
Regression C oefficient Coefficient Value T-value
a 0.11313 6.10904
b 1.23131 18.49756
c -0.19735 -7.96916
d -0.23921 -12.73172
The regression coefficient of each X variable provides an estimate of its influence on Y,
representing the amount the dependent variable Y changes when the corresponding
independent variables change 1 unit. The variable a is the constant, where the regression line
intercepts the y axis, representing the amount the dependent Y will be when all the
independent variables are 0. T-tests are used to assess the significance of individual X
variable coefficients, specifically testing the null hypothesis that the regression coefficient is
zero. A common rule of thumb is to drop from the equation all variables not significant at the
0.05 level or better. The value of standard error of estimate is 0.07513. The standard error of
estimate indicates the accuracy of a prediction model and can be computed by the equation of
the standard deviation of the error variable. The smaller the standard error of estimate, the
better the prediction. Hence, the overall equation to describe the relationship is:
Effect of adequacy and session degradation on session performance = 0.11313 + 1.23131
* "Calculated Dialysis Adequacy (Kt/V)" — 0.19735 * "Calculated Dialysis Adequacy (Kt/V)"
72. — 0.23921 * Session Degradation (8)
III. Results and Behaviors
Various runs were conducted and the results were relatively compared against each other.
These results were also thoroughly validated by the subject matter experts. The behaviors
observed result from the interactions of numerous feedback loops present in the structure.
Sometimes it might be difficult to attribute the observed behavior to any particular feedback
loop. Partial simulation runs were conducted to ensure the appropriateness of the formulation.
The values of exogenous variables, to certain extent, determine the dominance of feedback
loops which ultimately result into specific system behavior. These values can be changed to
observe their impact on the system's behavior.
Exponential growth arises from positive (self-reinforcing) feedback. Figure 9 shows the
generic structure responsible for exponential growth. Increase in state of the system increases
the net increase rate which in turn increases the state of the system.
11
R
net increase rate 48) State of the system
Figure 9 Positive (Self Reinforcing) Loop
Loops R1 and R2 explain the exponential growth mentioned above exhibit the exponential
increase in session performance. To show the behavior of the intradialytic model, partial
simulation runs were conducted. This partial simulation means that the patient receives the
dialysis treatment. Hence, the performance of the dialysis session is measured using two
variables; dialysis adequacy and session degradation. The resulting behavior is exponential
growth in session performance as shown in figure 10.
Session Performance
0.75
0.5
0.25
0
48 72 96 120 144 «#168 192 216 240
Time (Minute)
Session Performance : High dialysis efficiency Dmunl
Figure 10 Partial Simulations — Session Performance
A. Testing of Dynamic Hypothesis: Dialysis Performance Drivers
The hypotheses were tested at various levels of dialysis adequacy and session degradation.
The first simulation was run at a dialysis adequacy level being less than the recommendation
for a minimum dialysis dose Kt/V of 1.2 and at high level of session degradation (i.e. > 40 %
due to high level of complications and low level of equipment effectiveness).
12
The second simulation was run at a dialysis adequacy level being equal to the minimum
dialysis dose Kt/V of 1.2 and at critical level of session degradation (ie. = 40 % due to
medium level of complications and equipment effectiveness). The third simulation was run at
a dialysis adequacy level being greater than the minimum dialysis dose Kt/V of 1.2 and at and
at low level of session degradation (i.e. < 40 % due to low level of complications and high
level of equipment effectiveness). The results are shown in Figure 11 and 12. It is observed
that the overall dialysis session performance increases as the amount of dialysis dose is
increased. The results also indicate that low session degradation levels increases the dialysis
session performance as the probability of complications decreases and the effectiveness of
equipment increases. The results demonstrate that these hypotheses are shown for the current
structure of the model.
Session Performance
nl
0.75
0.5
0.25
0
0 24 48 72 96 120 144 168 192 216 240
Time (Minute)
Session Performance : high adequacy and low degradation ————————_ Dm
Session Performance : Minimum adequacy and critical degradation Dmnl
Session Performance : low adequacy and high degradation Dmnl
Figure 11 Dialysis Session Performance at various levels Of Kt/V and
Session degradation
Modelled Post BUN
200
150
100
50
ie)
is) 24 48 72 36 120 144 168 192 216 240
Time (Minute)
Modelled Post BUN : high adequacy and low degradation ————————_ mg/dL
Modelled Post BUN : Minimum adequacy and critical degradation mg/dL.
Modelled Post BUN : low adequacy and high degradation mg/dL
Figure 12 Modeled Post BUN at Various Levels of Kt/V And
Session Degradation
13
B. Sensitivity analysis
Sensitivity analysis is used to determine how "sensitive" a model is to changes in the value of
the parameters of the model and to changes in the structure of the model. Parameter
sensitivity is usually performed as a series of tests in which the modeler sets different
parameter values to see how a change in the parameter causes a change in the dynamic
behavior of the stocks. By showing how the model behavior responds to changes in parameter
values, sensitivity analysis is a useful tool in model building as well as in model evaluation.
Sensitivity analysis helps to build confidence in the model by studying the uncertainties that
are often associated with parameters in models. Many parameters in system dynamics models
represent quantities that are very difficult, or even impossible to measure to a great deal of
accuracy in the real world. Also, some parameter values change in the real world. Sensitivity
analysis indicates what level of accuracy is necessary for a parameter to make the model
sufficiently useful and valid.
Five tests were performed as follows where the model yielded an expected behavior in all
tests. (1) Effect of equipment effectiveness on the intradialytic complications, (2) Effect of
equipment effectiveness on session degradation, (3) Effect of equipment effectiveness on
dialysis dose Kt/V, (4) Effect of equipment effectiveness on post-Blood Urea Nitrogen (BUN)
(5) Effect of equipment effectiveness on the overall session performance. The results of these
tests were simulated with three sets of key parameter combinations, namely, (1) New
Equipment that no maintenance procedures were performed, (2) High efficiency equipment
with low number of working hours and regular maintenance procedures, and (3) Low
efficiency equipment with high number of working hours and accrued maintenance
procedures.
1, Effect Of Equipment Effectiveness On The Intradialytic Complications
Probability of Complications
0.6
0.5
oa both
03 Petals eee ery BML, Beso ees 2a
Rwy Ta ey 7 7
0.2
0 24 48 72 96 120 144 168 192 216 240
Time (Minute)
Probability of Complications : New equipment (not maintained) —+——+—+—+— Dmnl
Probability of Complications : high equipment effectiveness (used+calibrated) <2 Dmnl
Probability of Complications : low equipment efficiency (usedthinclaibrated) ——<+—3- Dmnl
Figure 13 Effect Of Equipment Effectiveness On The Intradialytic C omplications
14
It is noted from figure 13 that new equipments without any defects and used equipments with
regular and effective maintenance procedures doesn’t enhance the probability of intradialytic
complications among hemodialysis patients. The experimental study that was applied on 134
hemodialysis patients showed that accrued maintenance programs increase the probability of
complications by about 45 % among hemodialysis patients due to uncalibrated blood and
dialysate pumps so that the proper setting can't be delivered to the patient from the machine.
(2) Effect of equipment effectiveness on session degradation
Session degradation is defined as the session failure due to intradialytic complications and
equipment deficiency. It is a dimensionless variable measured using a relative scale (varying
from 0 to 1; 0 corresponds to total success and 1 corresponds to total failure). Figure 14 shows
the effect of equipment effectiveness on session degradation. The simulation result revealed
that low equipment efficiency due to deferred maintenance procedures increases the
hemodialysis session degradation to about 56 % and may causes severe problems and
complications to the patients. The experimental study revealed that there is no session
degradation was noted due to new hemodialysis equipments but the amount of session
degradation was due to the complications that were happened to patients during session.
Session Degradation
06 P37 4 415
0.45
03
x Tot Z
0 24 48 72 96 120. 144 «#168 192 216 240
Time (Minute)
Session Degradation : New equipment (not maintained) ——+——+——+——+-_ Dmnl
Session Degradation : high equipment effectiveness (usedt+cahbrated) —~2——2 Dmnl
Session Degradation : low equipment efficiency (usedhinclaibrated) —<+—~+— Dmnl
Figure 14 Effect Of Equipment Effectiveness On session degradation
(3) Effect of equipment effectiveness on dialysis dose Kt/V & (4) Effect of equipment
effectiveness on post-blood urea nitrogen (BUN)
To evaluate the effect of equipment effectiveness on dialysis adequacy, the 134 patients were
grouped to:
* 22 patients (16.42%) dialyzed with new hemodialysis equipments
* 102 patients (76.12%) dialyzed with used and calibrated equipments
« 10 patients (7.46%) dialyzed with used and uncalibrated equipments.
15
It is concluded that increasing the equipment effectiveness is associated with a statistically
significant increase in Kt/V and decreasing in post-blood urea nitrogen (BUN). Hemodialysis
with calibrated equipments should be considered in selected patients not achieving adequacy
to optimize blood, dialysate and ultrafiltration flow rates. The statistical analysis revealed also
that there was a statistically significant increase in the dialysis adequacy Kt/V and urea
reduction ration (URR) as the equipment efficiency increases from low efficiency equipment
to high efficiency equipment. It is noted from figure 15 that by using the calibrated
hemodialysis equipments with regular maintenance procedures the dialysis dose Kt/V
increases to the desired value of 1.3. The Kt/V values for those patients dialyzed with low
efficiency and uncalibrated equipments are less than 1 which is inadequate dialysis dose.
Calculated Dialysis Adequacy (Kt/V)
0 24 48 72 96 120 144 168 192 216 240
Time (Minute)
"Calculated Dialysis Adequacy (Kt/V)" : New equipment (not maintained) —t——+—+—+ Dal
"Calculated Dialysis Adequacy (Kt/V)" : high equipment effectiveness (used+calibrated) —— Dmnl
"Calculated Dialysis Adequacy (Kt/V)" : low equipment efficiency (usedtunclaibrated) ——2— Dmal
Figure 15 Effect of equipment effectiveness on dialysis dose Kt/V
The urea reduction ratio (URR) for those patients dialyzed with maintained and calibrated
equipments also increases due to the decrease in the blood urea nitrogen (BUN). It is noted
from figure 16 that use of low-efficiency equipments can result in a low reduction in the post-
BUN.
Modelled Post BUN
200
150
100
50
0
0 24 48 72 96 120 144 168 192 216 240
Time (Minute)
Modelled Post BUN : New equipment (not maintained) ——-t+——+——++—_+—__ mg/dL
Modelled Post BUN : high equipment effectiveness (used+calibrated) ——2——2- mg/dL
Modelled Post BUN : low equipment efficiency (used+unclaibrated) —+——3——3. mg/dL.
Figure 16 Effect of equipment effectiveness on post-Blood Urea Nitrogen (BUN)
16
The statistical analysis demonstrated that there was a statistically significant increase in Kt/V
by about 52.93% (from 0.82 with low efficiency equipment to 1.254 with high efficiency
equipment) The URR increases also by about 23.66% from 56.50% to 69.87% when
switching from low efficiency dialysis equipment to high efficiency equipment.
(5) Effect of equipment effectiveness on the overall session performance
The overall hemodialysis session performance increases by about 34% from 55.94% when the
patients dialyzed with low efficiency and uncalibrated equipments to 74.96% when patients
dialyzed with calibrated and high efficiency equipment. The hemodialysis session
performance increases by about 17.09% when switching dialysis from used and calibrated
equipments to new equipments. The effect of equipment efficiency on session performance is
shown in figure 17.
Session Performance
0 24 48 72 96 120 144 168 192 216 240
Time (Minute)
Session Performance : New equipment (not maintained) ——+——+——+—+_ Dmnl
Session Performance : high equipment effectiveness (usedtcalibrated) ———2 Dmnl
Session Performance : low equipment efficiency (usedtunclaibrated) <——3— Dmnl
Figure 17 Effect of equipment effectiveness on the overall session performance
IV. DISCUSSION
An issue of increasing importance to all nephrologists is the correct assessment of dialysis
performance. Recent studies have shown how efficient the use of urea kinetic modeling
(UKM) is in the quantification and monitoring of dialysis, and also in predicting patient
morbidity and mortality. On this basis, the overall goal of this research was to build a system
dynamics model to quantify the hemodialysis session performance from systems perspective.
No successful results have been reported on this topic to date. We were able to accomplish the
research goal. The system dynamics model was developed using Vensim DSS 4.0a. The
model was structured based on the inputs from the experts in the field of nephrology. The
model was extensively tested and the results were validated by the experts so that we can
conclude that this model has a high degree of statistical significance. It should be noted that
17
the base case values used for simulation are a real data measured from dialysis patients Using
this system dynamics model, a significant improvement in dialysis performance was achieved
by highlighting factors which may alter the delivered dose and may lead to session
degradation. This model represents significant advances over previous urea kinetic models. In
short, the model developed during the course of this research makes possible the accurate
reagentless monitoring of dialysis performance over time where previous models have failed.
The dynamic hypotheses stated in section III.A were tested using the system dynamics model
developed using Vensim DSS 4.0 by varying parameters and observing the changes in the
subsequent results from the simulation. The primary and secondary hypotheses depend on
dialysis session improvement due to the increase in dialysis adequacy and the reduction of
session degradation. These hypotheses were tested at various levels of dialysis adequacy and
session degradation.
The simulation results support the stated dynamic hypothesis and demonstrated that these
hypotheses are shown for the current structure of the model. The simulation shows that the
effective and regular maintenance procedures have a different impact on the behavior of the
dialysis system. By linking the maintenance software with hemodialysis system dynamics
model it was noted that the required preventive maintenance should be completed during the
preventive maintenance cycle to ensure that the dialysis machine is accurate and well
calibrated. Deferred maintenance will leave the machine in partially degraded state which
may further degrade the system at higher rate and hence will decrease the performance of the
system over the entire operational phase. The preventive maintenance interval should be
determined based on the maintainability factors such as mean preventive maintenance time,
mean corrective maintenance time and maintenance man-hours required per operational cycle.
V. CONCLUSION
Dialysis quality is a complex and evolutionary concept that has to be viewed in a quality
assurance process to improve outcomes of end stage renal disease (ESRD) patients. To
simplify this assessment it is very important that dialysis machines have to be maintained on a
regular basis. The necessary amount of regular maintenance can be done by the maintenance
department. The overall goal of the maintenance procedures is to raise the overall equipment
effectiveness. Dialysis machines with a high maintenance standard are able to deliver proper
settings to the patient with less or no failures. Maintenance has become one of the most
expedient approaches to guarantee high machine dependability.
REFERENCES
[1] Frank A. Gotch. Kt/V Is the Best Dialysis Dose Parameter. Blood Purif, vol. 18 pp. 276-
285, 2000.
[2] Daugirdas JT. "Second Generation Logarithmic Estimates of Single-Pool Variable
Volume Kt/V: An Analysis of Error’. J Am Soc Nephrol, vol. 4, pp. 1205-1213, 1993.
[3] De Palma JR, Pittard JD. "Dialysis Dose (Parts I & II)". Dial Transplant, vol. 30 (4), pp.
252- 265 and vol. 30 (5), pp. 315-324, 2001.
18
[4] Kooman JP, van der Sande FM, Leunissen KM. "Kt/V: Finding the Tree Within the
Woods". Nephrology Dialysis Transplantation, vol. 16, pp. 1749-1752, 2001.
[5] Bernard Charra. Is Kt/V Urea a Satisfactory Measure for Dosing the Newer Dialysis
Regimens?. Seminars in Dialysis—V ol 14, No 1 (January-February) 2001 pp. 8-9.
[6] Sehgal AR, Snow RJ, Singer ME, Amini SB, DeOreo PB, Silver MR, Cebul RD. "Barriers
To Adequate Delivery Of Hemodialysis". AmJ Kidney Dis, vol. 31, pp. 593-601, 1998.
[7] Lindsay RM, Heidenheim AP, Spanner E, et al. "Urea monitoring during dialysis: the
wave of the future". A tale of two cities. ASAIO Trans, vol. 37, pp. 49-53, 1991.
[8] National Kidney Foundation. "NKF-DOQI Clinical Practice Guidelines for Hemodialysis
Adequacy". AmJ Kidney Dis, vol. 30 (Suppl), pp. S15-S66, 1997.
[9] Sargent JA. "Shortfalls In The Delivery Of Dialysis". Am J Kidney Dis, vol. 15, pp. 500-
510, 1990.
[10] LeFebvre JMJ, Spanner E, Heidenheim AP, Lindsay RM. "Kt/V: Patients Do Not Get
What The Physician Prescribes". ASAIO Trans, vol. 37, pp. M132-M133, 1991.
[11] Boag JT. "Basic Truths in Optimal Hemodialysis". Dialysis & Transplantation, vol 23,
no11, pp. 636, November 1994.
[12] End Stage Renal Disease (ESRD) Network 13 website. http://www.network13.org/.
[13] Azar, A, Mohamed, Abdalla S.A and Wahba, K., “Analyzing the Dynamic Implications
For Improving Hemodialysis Session Performance By System Dynamics Modeling”, 24”
International Conference of the System Dynamics Society, Nijmegen, The Netherlands, July
23 — 27, 2006.
[14] Sterman, J.D. "Business Dynamics: Systems Thinking and Modeling for a Complex
World". Irwin McGraw-Hill: Boston, MA, 2000.
19
Appendix A The Overall Stock And Flow Diagram Of Hemodiadynamics
20
