Theses- Master of Science in Perfusion Program

Anderson, Mark J. “Further Investigation into the Cavitation Equation’s Ability to Predict Bubble Formation in the Cardiopulmonary Bypass Circuit.”

Jan. 2007, 80pp, 28 references
Available for checkout
Abstract: Cavitation, the formation of Gaseous Microemboli (GME), is a cause for concern in the field of perfusion due to the risk of neurological damage posed when GMEs form while a patient is on Cardiopulmonary Bypass (CPB). Previous studies have looked for ways to predict cavitation with the goal of enabling perfusionists to decrease the risk for impaired neurological function in patients by limiting the formation of the GMEs while on CPB. In his Master’s Thesis, Cameron Bryan developed an equation based on the Bernoulli equation to predict the occurrence of bubble formation in flowing liquid (.=(Po-Pv)/[(1/2)*p*V2]). Thomas Grulkowski tested the equation using porcine blood and was unable to verify that Bryan’s cavitation equation could accurately predict bubble formation in blood. The goal of this study was to determine if Bryan’s cavitation equation could accurately predict the formation of bubbles using sterile water. A normal CPB circuit without an arterial line filter was used to conduct the trials. A Hatteland bubble detector probe was placed at the pump head outlet and set to its most sensitive setting to detect bubble formation. Sterile water was run through the pump at four different flow rates (between 3.5-10.0 L/min) and at two different vacuum pressures (0 and -75 mmHg). Eighty trials were performed. In only one trial did any bubble formation occur, which led to the conclusion that the study could not validate Bryan’s cavitation equation as a means for predicting bubble formation in a CPB circuit with sterile water as the perfusate. Possible reasons for the conclusion include lack of nucleatation sites, circuit setup, bubble counter error, bubble collapse and the assumed vapor pressure of sterile water.

Axarlis, AnnMarie, B.E. “A Retrospective Analysis of the Clinical Benefits of Adenosine Enhanced Cardioplegia”

February 2013, 44pp, 35 references, appendix, figures, tables
Available for checkout
Abstract: The goal of this study was to determine if adding adenosine to the cardioplegia used during cardiopulmonary bypass provided post-surgical benefit to the heart by improving cardiac function. This study was performed retrospectively using patient charts from a single surgeon in which some patients did not receive adenosine in the cardioplegia (control group, n=21) while others did (adenosine group, n=21). The patient charts were examined for the number of inotropes used and their individual infusion doses, as well as the pre-operative, post-operative, and follow up ejection fractions estimated by echocardiography. The results of the inotrope study indicated a non-significant difference in the number of inotrope (p=0.052) and the infusion dose of inotropes (p=0.119) between the adenosine and control group as determined by t-tests and Mann-Whitney tests. Unpaired t-tests were performed for each time point the ejection fraction was estimated from echocardiographys, which was pre-operatively, post-operatively, and a surgical follow up. There was no difference in ejection fractions between the adenosine and control groups as determined in the pre-operative (p=0.421) and post-operative groups (p=0.690). However, the follow up ejection fraction unpaired t-test revealed a significant difference between the control and adenosine group (p=0.045). The results of this study indicated adenosine cardioplegia offers no myocardial protection for this protocol in terms of a decreased need for inotrope support post-operatively, but the study indicated improved ejection fraction in a follow up echocardiography 2-8 months post-operatively. However, the results should be interpreted carefully as the sample sizes in each group were very small, such that this study may not have had the power to determine clinical differences.

Baternik, Aaron “The Effect of Patient Variables on the Rewarming Rate During Cardiopulmonary Bypass”

Aug. 2004, 101pp, 46 references
  • Appendix A: Trend Plots of Data
  • Appendix B: Correlation Matrix
  • Appendix C: Scatter Plots of Data
  • Appendix D: Raw Data
Available for checkout
Abstract: During cardiopulmonary bypass (CPB), hypothermia is commonly used to decrease methbolism and the tissue’s need for oxygen. At the end of the procedure, the patient is rewarmed to their normal body temperature. This author could find no previous studies that aimed to determine how or why patients may rewarm at different rates. This study attempts to determine which patient variables (i.e. age, gender, diabetes, body mass index (BMI), body surface area (BSA)) affect the rewarming rate during CPB). Hypothesis: Patients having the following attributes will rewarm at a slower rate (measures in 0C/min): advanced age, male, non-diabetic, high BMI, and large BSA. Conversely, younger patients, females, diabetics, patients with a low BMI, and patients with a small BSA will rewarm at a faster rate on cardiopulmonary bypass. Sixteen patients participated in the study. The average age of the study participants was 68 ± 9 years. Average BSA and BMI were 2.05 ± 0.30m2 and 33.6 ± 11.1 kg/m2, respectively. Six of the patients were female, ten were male, and five of the sixteen were diabetic. Surgeries were performed using standard protocols. The following data were recorded every two minutes: venous blood temperature, arterial blood temperature, water bath temperature, esophageal temperature, core temperature, mean arterial blood pressure, and pump flow rate. A multiple regression was performed using the rewarming rate as the dependent variable. The rates of esophageal and core rewarming were 0.240 ± 0.1150C/min and 0.0670 ± 0.03270C/min, respectively. All other variables were treated as independent. None of the patient variables originally hypothesized to affect rewarming rate were found to have a significant effect on the rewarming rate. None of the proposed independent variables significantly explained differences in the esophageal rewarming rate; R2 = 0.345, the adjusted R2 = 0.181, and all associated p-values were greater than 0.05. Using a backward elimination stepwise regression, the coldest temperature on bypass (coldest T) was correlated with the highest core rewarming rate; R2 = 0.800, adjusted R2 = 0.786, and p < 0.001. The power of this test was found to be 0.99. All other independent variables were eliminated and not found to be significant.

Bergemann, Michael A., B.S. "Retrospective Analysis Of The Effect Of Intraoperative Oxygen Delivery On Postoperative Renal Complications"

March 2014, 38pp
Available for checkout

Abstract: A common postoperative outcome of patients undergoing cardiac surgery requiring the use of cardiopulmonary bypass (CPB) is acute renal failure (ARF). Acute renal failure is associated with significantly increased mortality, morbidity and healthcare costs; therefore, preventing this condition is of concern. Historically, low intraoperative hematocrit was believed to be one of the most important variables for predicting the postoperative incidence of ARF in CPB patients. Recent studies from a single surgical center, however, have suggested that minimum intraoperative oxygen delivery is a better predictor of ARF in this group. The goal of this study was to further investigate the impact of intraoperative oxygen delivery and intraoperative hematocrit in relation to postoperative ARF and serum creatinine ratios, a traditional measure of the operative insult to the kidneys. A retrospective analysis of 508 coronary artery bypass graft (CABG) patients was conducted using surgical data from Aurora St. Luke’s Medical Center in Milwaukee, Wisconsin. The minimum and average delivery of oxygen (ml O2/min/m2) and lowest and average hematocrit (%) for all patients were compared to postoperative ARF diagnosis and serum creatinine ratio. Twenty patients (3.9%) met the criteria for postoperative ARF with significant associations with elevated weight, diabetes and postoperative blood transfusions. Additionally, elevated serum creatinine ratios associated significantly with elevated weight, diabetes, hypertension, intraoperative blood transfusions, postoperative blood transfusions and average oxygen delivery. The data, however, did not support the hypothesis that minimum delivery of oxygen is a better predictor of ARF compared to lowest hematocrit.


Berkovitz, Danielle “The Effects of Phenylephrine and Vasopressin Use on Renal Outcomes of Open Heart Surgery Patients”

June 2011, 63pp, 31 references, appendices, figures, tables
Available for checkout
Abstract: A subject of debate in the perfusion community is whether the use of the vasoconstrictor vasopressin during cardiopulmonary bypass (CPB) results in better renal outcomes for cardiopulmonary bypass patients than the use of the vasoconstrictor phenylephrine. Studies performed in both animals and humans seem to suggest phenelephrine may decrease renal blood flow. This reduction in renal flow could harm the kidneys, leading to injury or failure. The goal of the present study was to perform a retrospective analysis of renal function of CPB patients and to correlate this with vasoconstrictor use and dose.Data on surgeries performed between January 2007 and February 2011 were obtained from the Society of Thoracic Surgeons adult cardiac database at Aurora St. Luke’s Medical Center, Milwaukee, Wisconsin. These data included information on pre- and post-operative creatinine levels, which were used to classify each patient into a level of renal impairment as outlined by the published RIFLE criteria. Information on phenylephrine usage and dosing was also obtained, as well as whether or not vasopressin was used. Patients were then classified as having been treated with phenylephrine alone (mean post-operative creatinine 1.5 ± 0.87 mg/dL), vasopressin alone (mean post-operative creatinine 1.9 ± 1.8 mg/dL), both drugs (mean post-operative creatinine 1.56 ± 0.91). Additional information included type of surgery, phenylephrine dose, cardiopulmonary bypass time and aortic cross-clamp time. The data were first analyzed using multiple regression models. Follow-up tests included one-way ANOVA and Tukey multiple comparisons tests.

Results demonstrated that neither drug provided solid predictive value with respect to post-operative renal dysfunction. A regression model that took into account a number of variables, including the use of vasoconstrictors, found that CPB time and type of surgery were better predictors of post-operative renal dysfunction than the use of any specific vasoconstrictor. It is recommended that future studies investigate additional predictors of post-operative renal dysfunction.


Brown, Scott M. “A Model of Venous Return While Utilizing Vacuum Assist During Cardiopulmonary Bypass.”

May 17, 2000, 61pp, 38 references
  • Appendix A: Blood Gas Data
  • Appendix B: Random Number Tables
  • Appendix C: Measured Flow and Cannula Pressure Drop Data for the Dual- Stage Cannulas
  • Appendix D: Measured Flow and Cannula Pressure Drop Data for the Single-Stage Cannulas
  • Appendix E: Measured Bubble Counts for the Dual-Stage and Single-Stage Cannulas
  • Appendix F: Predicted Data from the Model for Each Cannula
  • Appendix G: Statistical Powers for the Analysis of the Bubble Count Data of Each Cannula
  • Appendix H: t-Statistic Calculation for the Comparison of the Model and Measured Data
Available for checkout
Abstract: In order for vacuum assisted venous return to be used safely and efficiently, a full understanding of venous return is necessary. Through the use of the concepts of conservation of energy and viscous energy dissipation, a theoretical model of venous return utilizing vacuum assist has been developed in this work. The effectiveness and accuracy of this model has been verified through in vitro laboratory investigations and statistical analysis. In summary, the developed model accurately predicts flow rates given cannula size, amount of vacuum applied, and circuit dimensions. Although vacuum assisted venous return can provide higher flows through smaller cannula, vacuum assist can lead to increased levels of shear stress that the blood is exposed to as well as an increased likelihood of cavitation in the venous line. Both of which may lead to an exacerbation of the detrimental effects of cardiopulmonary bypass.

Bryan, Cameron N. “A Preliminary Investigation into the Occurrence of Cavitation in the Cardiopulmonary Bypass Circuit.”

June 2003, 84pp, 56 references
  • Appendix 1: Additional Equations
  • Appendix 2: Cavitation Number Data
  • Appendix 3: Peak Pressure Data
  • Appendix 4: Bubble Collapse Data
  • Appendix 5: Cavitation Number Dependent upon Tubing Diameter and Flow
  • Appendix 7: Gaussian Distribution Data for Bubble Count
  • Appendix 8: Bubble Collapse Time Dependent upon External Pressure
  • Appendix 9: Pressure Conversion Tables
  • Appendix 10: Bubble Size Dependent upon Tubing Diameter
Available for checkout
Abstract: The use of smaller venous cannulas, increased utilization of peripheral cannulation sites, and the desire to reduce the cardiopulmonary bypass (CPB) circuit, potentially create clinical situations where venous return is reduced. Adequate venous return is crucial for a successful and uneventful surgical outcome. Two different techniques have been used to increase venous return; they are kinetic-assisted (KAVD) and vacuum-assisted venous return (VAVD). In recent years, Kuntz, Jones, and others have reported measuring an increase in the number of air emboli that are being detected (by Doppler ultrasound) distal to the arterial line filter, specifically when VAVD is utilized. One potential cause of these bubbles is cavitation. This thesis will review the science of cavitation and bubble dynamics and will attempt to determine if the mathematical equations used to predict cavitation in water could be used for blood. Additionally, this thesis will identify the characteristics of blood which may increase or decrease the likelihood of cavitation to occur in blood. In conclusion, the cavitation number equation predicts that reference pressures and flow rates, used clinically, is likely to result in cavitation occurring in blood, The bubble growth equation predicts that bubbles are being formed which are small enough to pass through the arterial line filter. The bubble collapse equation predicts a small time interval is necessary to completely collapse a bubble; although during the cycle of growth and collapse a bubble may remain long enough to potentially travel a large distance through the CPB circuit. Blood has many interent properties, which may have a large influence on the occurrence of cavitation. By using the CPB circuit, non-natural conditions such as VAVD and hypothermia, can be applied to blood which increases the likelihood of cavitation occurrence. Any bubble which passes through the arterial line filter and enters into the patients’ circulation may result in a variety of detrimental effects on the patient.

Buckoski, Benjamin "The Effect of Heat on the Generation of Gaseous Microemboli in a Centrifugal Pump Head"

May 2013, 60pp, 45 references
Available for checkout
Abstract: The delivery of microscopic quantities of air is an inevitable consequence of surgery incorporating the use of cardiopulmonary bypass (CPB). Gaseous microemboli (GME) may be room air which has infiltrated the circuit, or it can be generated in the blood through cavitation. GME are believed to be a contributing factor to the cognitive decline seen in patients who have undergone CPB. In order to minimize injury to the patient, it is important for a perfusionist to understand the relationship between their interventions during CPB and the generation of gaseous microemboli. Centrifugal pumps are commonly used to produce the arterial flow in CPB. Centrifugal pumps are afterload dependent, and therefore their outlet may be clamped to cease flow. The purpose of this study is to investigate the generation of heat and GME in an occluded Terumo Delphin centrifugal pump head at various pump speeds, initial fluid temperatures, and occlusion times.

A CPB circuit was created which incorporated EDAC (emboli detection and classification) GME sensors to quantify the amount of air generated by the centrifugal pump head, as well as a temperature probe to monitor the temperature of the fluid inside of the pump head. The circuit was primed with an aqueous glycerol solution to mimic the viscosity of blood at a hematocrit of 25%. The pump head was occluded for periods of 0, 5, 10, and 15 minutes at speeds of 1000, 1800, and 2600 RPM when the initial fluid temperature was 280, and 360C. The temperature of the fluid inside of the pump head was measured throughout the occlusion period, and upon removal of the clamp, the GME exiting the centrifugal pump head was measured for a period of one minute.

The results of this study showed the effect of pump speed and occlusion time on both the volume of GME generation and a change in pump head temperature. Time and the interaction between time and speed had a statistically significant effect on the change in temperature while the pump was occluded. The volume of GME produced at a pump speed of 2600 RPMs was much larger than at the other two pump speeds. Pump speed was found to have a statistically significant effect on GME volume, while occlusion time did not have a significant effect. The initial temperature of the fluid was not found to have a statistically significant effect on GME production. In practice, both centrifugal pump speed and occlusion times are under the control of the perfusionist. A greater understanding of the contribution of these factors to GME and heat generation offers the perfusionist insight into ways in which they may modify their practice to improve patient outcomes.


Buss, Carson “Waste Watchers: An Exploration in Medical Waste Management, Minimization, and Mitigation of Environmental Impact in Perfusion-Based Practice”

February 2017
Available for checkout
Abstract: The cost of healthcare in the United States is projected to continue to rise into 2024, leaving practitioners and healthcare managers looking for ways to reduce operating costs while not sacrificing levels of care. The removal and processing of medical waste significantly contributes to a hospital’s operating budget. In 2016, Aurora St. Luke’s Medical Center spent over a quarter million dollars on removing medical waste from their facility. The operating rooms have been identified as a source of a large amount of this waste. The goal of this inquiry is to identify current disposal techniques utilized by perfusion, state and suggest ways to improve this practice based on the results of a week-long audit of their waste production.

A week-long audit of three waste streams — municipal waste, regulated medical waste, and pharmaceutical waste — was conducted. Based on the results of the week-long audit, current waste segregation practices were evaluated with respect to volume generated and costs.

This report indicates that current practices of waste removal are in compliance with local, state, and federal regulations, but there is widespread uncertainty as to how different types of waste should be disposed of in order to make their removal both financially efficient and environmentally less impactful. To improve current practices, perfusionists and OR personnel need to be educated on the financial and environmental impact of medical waste disposal.


Candek, Christopher J. “A Preliminary Investigation into the Effect of Hematocrit on Bubbles in the Cardiopulmonary Bypass Circuit.”

Aug. 2007, 65pp, 47 references
  • Appendix A: Trial Data
  • Appendix B: Summaries of Data and Johnson Transformations
  • Appendix C: Minitab v. 15 Data (ANOVA and Multiple Comparison tests)
Available for checkout
Abstract: In this preliminary investigation the effect of different hematocrit on the amount of air bubbles observed in the cardiopulmonary bypass (CPB) circuit was studied. Bubbles are eliminated from blood due to many factors. One of these factors is surface tension. Hemodilution is used during CPB for many reasons, and results in the reduction of the surface tension between blood and gas emboli in the blood. Based on this relationship, it was predicted that more air bubbles would be observed, and for longer periods, in lower hematocrit blood than in high hematocrit blood. A standardized model of an adult CPB circuit was used. Three different hematocrit values (32%, 25%, and 18%) and three different monitoring sites (proximal to pump head, distal to oxygenator, and distal to arterial line filter) were tested. The hematocrit values were tested because they represent the normal range of hematocrits seen during CPB. The monitoring sites were selected to evaluate how the air was removed as the blood flowed through the circuit. Twenty trials for each hematocrit were attempted. The trials were divided into groups of ten per hematocrit and randomly assigned the order in which they would be performed, giving a total of 6 groups. Twenty trials were performed for both 32% hematocrit and 25% hematocrit, and ten trials were performed for 18% hematocrit due to deterioration of the blood and circuit over time. Statistical analysis was conducted and showed significant differences between 32% hematocrit blood and 18% hematocrit blood at all three locations, differences between 32% blood and 25% blood at the first two locations, and between 25% and 18% hematocrit blood in one location. The analysis of the results show that there was some effect of hematocrit on bubble elimination between 32% and 18% blood in two of the three locations. The effects between 32% and 25%, and between 25% and 18% were not as large, but still significant. The second location (distal to the oxygenator) displayed results that were not expected, with the higher hematocrit blood having more air and emboli than the lower hematocrit blood. This may have been caused in part by the oxygenation of the blood as it flowed through the membrane oxygenator, or by the oxygenator purge line.

Casey, Trisha L., B.S. “A Mathematical Model of the Effects of Hemoconcentration on Renal Function”

Aug. 9, 2004, 103pp, 26 references
  • Appendix A: Descriptions and Normal Values of Dependent Variables, Constants, and Variable Parameters.
  • Appendix B: Simulink Blocks
  • Appendix C: Complete Model in Simulink Format
  • Appendix D: Equations and Simulink Model Components
  • Appendix E: Line Formatting of Graphical Functions
  • Appendix F: Normal Conditions and Simulation Results
  • Appendix G: Results Tables
Available for checkout
Abstract: Recent research has raised concern over the connection between hemoconcentration and post-operative renal dysfunction. Clinical studies often show conflicting conclusions as to the risk factors and causes of renal dysfunction. These inconsistent results are most likely because the causes are due to many variables. One possible explanation for the association between hemoconcentration and renal dysfunction is the difference in plasma protein concentration (PPC) and hematocrit (HM). It is hypothesized that these variations will adversely effect certain renal functions such as glomerular filtration rate (GFR), urine flow rate (UFR), and renal blood flow (RFB). A mathematical model was utilized from previous research and implemented into Simulink© software. The model contains five sub-systems: renal dynamics, protein and compartment volumes, blood pressure, electrolytes, and hormones. The model was validated by comparing results of a simulation using normal parameters to results from the original author’s work. This model framework was then used to assess the diferences in GFR, URF, and RBF when PPC and HM were varied together and independently at time periods of 12, 24, and 36 hours post-operatively. It was concluded that the model was valid for the purposes of the project. Results are listed according to dependent variable. It was determined that all values in each set of simulations were within normal ranges of GFR. Therefore, changes in PPC and HM similar to those seen after hemoconcentration do not adversely affect GFR. UFR values tended to be lower than normal ranges during each set of simulations. Even though these values were lower, the results are most likely not clinically significant. Finally it was determined that RBF increased when PPC increased but decreased when HM increased. Therefore, there was little change in RBF when PPC and HM were varied together. This also suggests that RBF is not adversely affected when PPC and HM change in a manner similar to the changes occuring during hemoconcentration. Therefore, this research suggests that the changes in PPC and HM that ocur with hemoconcentration do not adversely affect GFR, UFR, and RBF up to 36 hours post-operatively.

Check, Joseph, B.S. “Recommendations for Improvements in Ancillary Perfusion Services Training in the Milwaukee School of Engineering Master of Science in Perfusion Program”

February 2015, 63pp, 25 references
Available for checkout
Abstract: With the advancement of technologies in cardiovascular surgery, the role of the perfusionist is constantly evolving. Perfusionists historically operated the heart-lung (i.e., bypass) machine, but now may also be responsible for ancillary services such as the operation of intra-aortic balloon pumps (IABP), extra-corporeal membrane oxygenators (ECMO), and ventricular assist devices (VADs). Because these services are becoming more commonplace in perfusion, perfusion education programs are required to survey their graduates, and the employers of their graduates, about the knowledge and ability that the graduates have in these areas. Surveys of Milwaukee School of Engineering (MSOE) Master of Science in Perfusion (MSP) graduates and their employers indicate that some graduates do not seem as comfortable with this equipment as they would like. The goal of this project was to re-examine the current training and assessment of competency of ancillary perfusion services in the MSOE MSP program and make recommendations for improvement where needed.

Graduate and employer comments from the past several years were reviewed in order to identify perceived weaknesses of the program. Then, a thorough review of the curriculum and the assessment tools used was carried out. Curriculum was reviewed for opportunities to add more exposures to topics, while the assessment forms were compared against best practices in medical education. Recommendations were formulated to pass along to the program.

A review of the current MSOE curriculum found that exposure to the ancillary services is limited and that the assessment forms could use some improvement. It was recommended that the MSP program incorporate new objective-based competency assessments and simulation training to improve the ancillary perfusion training and curriculum. Incorporating these new competency assessments and simulation training will allow for improvements and growth of ancillary training and education at MSOE.


Cook, Damien M. “Effects of Viscosity on Cardioplegia Blood Flow as Determined by a Mathematical Model Incorporating Elastic Coronary Arteries”

May 20, 2004, 72pp, 37 references
  • Appendix A: Left Coronary Artery Network of Segments
  • Appendix B: Flow Equations for Conductance Matrix
  • Appendix C: Conductance Matrix
  • Appendix D: Example of MATLAB Program
Available for checkout
Abstract: Cardioplegia is the solution used to arrest the heart during nearly all types of cardiac surgeries. Yet, in spite of this widespread use, the techniques of cardioplegia vary in solution, temperature, and mode of delivery. This paper reviews important issues to consider regarding myocardial protection, including ischemia, reperfusion injury, electrophysiology of the action potential and cardioplegia, as well as the effects of temperature and hematocrit on cardioplegia. One consideration of cardioplegia that has not been well described is how viscosity changes affect the distribution of cardioplegia in the coronary vasculature. Changes in cardioplegia composition and temperature change its viscosity. This in turn would be expected to change the flow (or pressure) and possible the distribution of the cardioplegia throughout the heart. This paper describes a mathematical model developed to predict the distrubition of steady flow through a network of elastic tubes. With constant pressures, it is expected that the flow of cardioplegia would be greater with a less viscous solution. Also, at any given level of stenosis, the relative flow through a coronary stenosis would be greater with the less viscous cardioplegia. The computer software program MATLAB was used to develop and simulate the model. This model was then used to study the relative flows of cardioplegia throughout the left coronary arterial system. The model was run under conditions of set flow as well as with a set source pressure to evaluate the relative flow distribution of an unobstructed artery system. Then, a coronary stenosis was simulated in the left anterior descending artery, and the relative flow was evaluated with respect to the percent of stenosis. These conditions were evaluated for viscosities corresponding to blood, 4:1, and crystalloid cardioplegia. The model results revealed that with a less viscous solution, cardioplegic flows must be increased to maintin a source pressure of 50 mmHg. Also, changes in the viscosity of the medium produced a near linear effect in respect to the relative flow through a stenotic lesion. This was not expected in the nonlinear equation of flow described in the model.

Daley, Kevin “The Effect of Vacuum Assisted Venous Drainage and Venous Reservoir Level on Microbubble Persistence into the Arterial Line”

May 2010, 46pp, 26 references, appendices, figures, tables
Available for checkout
Abstract: Gaseous microemboli (GME) have been shown to be contributing factors of neurological dysfunction following cardiopulmonary bypass (CPB). Air can enter the CPB circuit from a number of sources; the most common site is the venous cannulation site. As air travels down the venous lines and into the venous reservoir, large bubbles are broken up to small GME that have the potential to persist to the arterial line and ultimately to the patient. Two factors that are thought to contribute to the number of GME persisting into the arterial line are the degree of negative pressure used with vacuum assisted venous drainage (VAVD) and the volume of fluid in the venous reservoir. Although both of these have been studied to some extent, there are no published studies that combine the effects of both on persistent GME. This study was designed to test the combined effects of VAVD and venous reservoir volume on the number of bubbles that escape from a hard-shell venous reservoir.

An in vitro CPB circuit was constructed to test the persistence of GME following a 1mL injection of air into the “venous line” of the circuit. The circuit was tested at five levels of vacuum pressure in combination with three levels of venous reservoir volume. The number and volume of GME escaping from the venous reservoir were measured using the Emboli Detection and Classification (EDAC) quantifier (Luna Innovations).

The results of this study show that there are significantly more bubbles that escape from the venous reservoir under conditions of high levels of vacuum pressure (-40mmHg and -80mmHg). They also show that high levels of reservoir volume (2000mL) allow significantly fewer bubbles to persist into the arterial line.


de Leon, Joseph Andrew. “Effects of Various Cardiopulmonary Bypass Components on the Heart- Lung Machine’s Ability to Deliver Pulsatile Power.”

May 23, 1995, 16pp, 20 references
  • Appendix 1: Unit Conversions
  • Appendix 2: Pspice Simulation Program
  • Appendix 3: Compliance Data Linearization
  • Appendix 4: Resistance Data Linearization
Available for checkout
Abstract: The purpose of this study was to determine the effects of 1) substituting different cardiopulmonary bypass (CPB) components and 2) altering component properties on the heart-lung machine’s ability to deliver pulsatile power to a fluid model of a patient. A model of the cardiopulmonary bypass (CPB) circuit connected to a patient was developed. Based on computer simulations of this model, the following observations were made: · Pulsatile power was necessary to qualify pulsatile perfusion. To calculate pulsatile power, it was necessary to measure instantaneous pressure and flow. · Pulse pressure, a common method of clinically evaluating pulsatile perfusion, did not correlate with pulsatile power. A decrease in pulse pressure did not always result in a pulsatile power reduction. Hence, pulse pressure cannot be reliably used to qualify pulsatile perfusion. · The effects of substituting different brands of oxygenators, filters and arterial cannulas on the transmission of pulsatile power was simulated. Based on the computer simulations, no single component substitution significantly improved pulsatile power transmission. · Transmission of pulsatile power in the CPB circuit was most sensitive to changes in oxygenator, filter and arterial cannula resistances. The resistances of these components can be minimized to maximize the delivery of pulsatile power.

Dryden, Kathleen A. “The Effect of Negative Pressures Used During Ultrafiltration on Erythrocytes”

Sept. 14, 2004, 55pp, 36 references
  • Appendix A: Randomized Trials
  • Appendix B: CATECHEM Plasma Free Hemoglobin Manual Procedure
  • Appendix C: Two Way ANOVA SigmaStat Report for PFHb
  • Appendix D: Raw Data
  • Appendix E: Two Way ANOVA SigmaStat Report for Transmembrane Pressure
  • Appendix F: One Way ANOVA SigmaStat Report for Hematocrit
  • Appendix G: Multiple Linear Regression SigmaStat Report
Available for checkout
Abstract: Ultrafiltration is the process of filtering blood through a semi-permeable membrane to selectivly remove water and small molecular weight plasma solutes. Filtration is achieved using the hydrostatic pressure of the inflow solution coupled with a negative pressure created by vacuum suction on the effluent side of the filter. This technique is currently used in cardiothoracic surgery as an effective solution to the continuous challenge of hemodilution and patient volume overload associated with cardiopulmonary bypass and heart surgery. Ultrafiltration, also known as hemoconcentration, has been clinically proven to be a useful tool when incorporated into the circuit of the cardiopulmonary bypass machine to substantially decrease blood volume, increase hematocrit, and decrease the amount of circulating inflammatory mediators. Although ultrafiltration provides several benefits for the patient undergoing CPB, the pressure gradients generated may also be expected to lead to increased stresses on the formed elements of the blood. Therefore, to estaablish the safety of ultrafiltration as an acceptable practice for fluid removal with cardiopulmonary bypass it was necessary to investigate the effects of the pressure gradients used in hemoconcentration on the red blood cells. Based on the prediction that shear stress will increase when negative pressures are used, and because it is known that increased amounts of shear stress may cause damage to erythrocytes, it is reasonable to predict that hemolysis will occur with ultrafiltration. Therefore the purpose of this investigation was to determine if a vacuum suction applied to the effluent side of the hemoconcentrator would cause erythrocyte damage, as indicated by plasma-free hemoglobin levels (PFHb). In order to determine if ultrafiltration damaged the erythrocytes, a standard pump circuit was assembled and filled with porcine blood. Included in this pump circuit was an ultrafiltration device connected to a negative pressure source. Five pump runs were performed at each of five different negative pressure settings; 0, -100, -200, -300, and -400 mmHg. The rate of blood flow throughout the investigation was held constant at 400 mL/min for duration of 40 minutes. 3mL blood samples were obtained at 0, 20, and 40 minutes after the start of the pump run to be used for PFHb measurements. The plasma-free hemoglobin levels and transmembrane pressures at each point were evaluated with a two-way analysis of variance (ANOVA). No significant differences were found to exist between vacuum levels, time, and PFHb levels. A two way ANOVA was used to evaluate the interactions between time and vacuum levels on the transmembrane pressures, and a statistical significance was found to exist among the vacuum levels. Therefore, it is concluded that ultrafiltration at a blood flow rate of 400 mL/min and vacuum up to -400 mmHg applied to the effluent side of the device is a safe practice that does not cause significant hemolysis. Yet, because of the small sample sizes, further studies may be warranted to make sure these results are representative of the population as a whole.

Dunker, Jerry L “Proposed Autoperfusion Technique for Donor Heart Preservation.”

July 24, 1995, 22pp, 23 references
Available for checkout
Abstract: Donor heart availability is the single most limiting factor in heart transplantation today. It is well documented that the safe ischemic time of preserved hearts is currently limited to only 4 to 6 hours. The need for longer periods of preservation is expressed by far more centers than are currently working on this topic. There is also substantial documentation of the problems associated with the current preservation techniques of simple cold storage and microperfusion. These problems, including edema, energy depletion and lack of functional recovery can all be attributed to ischemia. Autoperfusion is a procedure in which the heart perfuses itself and other organs outside the body. Autoperfused heart-lung preparations offer the advantage of increasing the heart preservation time by eliminating organ ischemia, and allowing continuous functional evaluation of the preserved heart. In autoperfusion preparations in which only the heart and lungs are used, preservation times have been limited by early pulmonary edema and subsequent deterioration of myocardial function. This led investigators to include the kidney, and some cases the liver, pancreas and duodenum, into the preparation to eliminate excessive water and metabolic wastes. Although this resulted in increased preservation times of up to 37 hours, the probability of having multiple functioning organs from one donor and the logistics of having recipients for all the organs in one center everely restricts this type of procedure. Therefore, this paper proposed a modified technique in which artificial organs are incorporated into existing autoperfusion circuits to provide a preparation that has the potential of extending heart preservation up to 48 hours.

Dunn, Jeffrey S. “Perfusion Simulator Software Version 2.0.”

Presented at the AmSECT Region VI Meeting under the name Design and Validation of a Perfusion Simulator and Mock Circulation Loop on September 23, 1995 at the Holiday Inn Mart Plaza in Chicago, Illinois with Nicole M. Graff, B.S.
Jan. 22, 1996, 107pp, 23 references
  • Appendix A: Instructions for Authors, The Journal of Extra-Corporeal Technology
  • Appendix B: Hardware Specifications
  • Appendix C: Software Specifications
  • Appendix D: Test Procedure for the Data Acquisition Board
  • Appendix E: Analog Connector Pinout for 37-pin Connector
  • Appendix F: Digital Connector Pinout for 37-pin Connector
  • Appendix G: Mean Arterial Pressure Trend
  • Appendix H: Software Listing
  • Appendix I: References
Available for checkout
Abstract: Simulators have been used for years to train airplane pilots and allow them to develop skill levels prior to flying a particular aircraft. Such a simulator is desired to provide training for perfusion students before operating a cardiopulmonary bypass pump during surgery. The MSOE Perfusion Simulator was a project begun in 1994 consisting of a hardware component, the mock circulatory loop, and a software component. This project deals with the software portion of the simulator. Version 2.0 is a continuation of the original work done in 1994. Changes for version 2.0 include the use of a new loop design, calculated systemic vascular resistance to adjust the loop parameters, a new database design using Microsoft AccessTM 2.0 (1), and increased simulated case evaluation with the use of a mean arterial pressure trend and the addition of a case evaluation form. The code was continued in Microsoft Visual BasicTM 3.0 (2, 3, 4) which is a user-friendly programming language and a powerful tool for creating WindowsTM driven applications. The continuation of this project is essential and will undoubtedly provide a useful training tool for future perfusion students.

Estill, Holly A. “An Investigation into the Production of Air Microemboli During Varying Pulsatile Flow Parameters”

July 2006m 52 references
Available for checkout
Abstract: Pulsatile flow may not be an entirely new development in the arena of cardiopulmonary bypass; however, technology has improved over the years to create more efficient and less rheologically detrimental methods of producing pulsatile blood flow. Many different papers have been written regarding the potential positive and negative effects of pulsatile flow. One area that has yet to be examined with regard to pulsatile flow is the possibility of air microemboli development from the use of varying pulsatile flow techniques. This paper examined the possibility of air microemboli development when a roller head pump was used to create pulsatile flow. Roller head pumps create pulsatile flow by increasing and decreasing the revolutions per minute of the pump head and inherently creating larger pressure drops that could pull air out of solution at the arterial pump head inlet. If large amounts of air are able to be pulled out of solution at the pump head inlet, there would be a threat of air microemboli being passed onto the patient, which could result in catastrophic post operative neurological deficit. To examine the relationship between air microemboli development and pulsatile blood flow, an experiment was developed that included measuring air microemboli at four different circuit locations on a cardiopulmonary bypass circuit (arterial pump inlet, oxygenator inlet, oxygenator outlet, and arterial filter outlet). The circuit that was utilized in this experiment was an SX25 Terumo Oxygenator in conjunction with a standard Terumo X-Coated Tubing Pack. During the experiment the following pulsatile flow parameters were varied to determine if there was a relationship between pulsatile flow parameters and air microemboli production at different locations within the CPB circuit: base flow, continuous flow rate, and frequency. Instantaneous pressure was also measured at the four points within the circuit simultaneously with the bubble measurements. The instataneous pressure measurements were then used to calculate an RMS pressure, average pressure, and minimum pressure for each of the 360 trials. The objective was to find a correlation between pressure and air microemboli production in the cardiopulmonary bypass circuit. The results of the experiment did not show any significant relationship between Base Flow and air microemboli development, Continuous Flow Rate and air microemboli development or Frequency and air microemboli development. The results further did not find a correlation between the minimum, average or RMS pressure and air microemboli production within the circuit. It can therefore be concluded from the results of this experiment that utilizing pulsatile flow settings in the range of base flow 0-50%, continuous flow rate 4.5-6.5 L/min., and frequency of 60-80 BPM will not produce significant amounts of air microemboli at the Arterial Pump Inlet, Oxygenator Inlet, Oxygenator Outlet, or Arterial Filter Outlet.

Fenske, Julie “Development of a Model for Albumin Priming in Pediatric Cardiopulmonary Bypass Circuits”

February 2017
Available for checkout
Available full-text online
Abstract: Albumin is a plasma protein that is present in the native circulation of all humans, and plays a critical role in maintaining blood pressure through the colloid osmotic pressure (COP) it exerts within blood vessels. During cardiothoracic surgery, it is common practice to add albumin to the cardiopulmonary bypass circuit prime in an attempt to reduce the drop in colloid osmotic pressure of the circulation. While this is inarguably important, a universal protocol for finding the appropriate amount of albumin to add does not exist. The purpose of this project was to develop a model for the addition of albumin in the cardiopulmonary bypass circuits of pediatric patients, based on theoretical calculations and current clinical evidence.

This project began with an investigation as to the relationship between outcomes and albumin concentration in a cardiopulmonary bypass prime solution. Literature was reviewed that was both current and relevant to pediatrics, cardiopulmonary bypass circuit priming, albumin, colloid osmotic pressure, and edema. After a thorough exploration of the literature and accompanying data, it was evident that the need for an appropriate albumin concentration was supported, and the development of such a protocol for it was pursued.

The end result of this project produced a set of equations, built into a spreadsheet-based calculator, that could be integrated into the practice of pediatric perfusion to determine the appropriate amount of albumin to be added into the cardiopulmonary bypass prime. The implementation of the equations and calculator is expected to result in better outcomes of the pediatric population after cardiothoracic surgery.


Froebe, Scott, B.A. “Proposal to Automate Clinical Data Input for the MSOE Perfusion Program”

March 2015, 59pp, 11 references
Available for checkout
Abstract: The Milwaukee School of Engineering’s (MSOE) Master of Science in Perfusion (MSP) Program currently uses paper documents to record case participation and to evaluate students during their clinical rotations. When a student participates in a clinical procedure they are tasked with documenting their participation with a case log form while the instructor is tasked with completing a separate evaluation of the student. This method of documentation and evaluation has been a useful tool for MSOE MSP students and staff. But there are certain challenges associated with maintaining the paper forms, mainly the efforts involved in distributing, collecting, and collating the data. This thesis project describes the development of a system whereby the student case log and evaluation forms can be completed electronically and the data automatically collated on a per student basis.

Several methods were investigated for incorporating the case log forms into an electronic format with databasing capabilities. The initial focus was to create a private website that would be completely customizable. This method was abandoned due to the associated costs. Several other options were also considered, including Blackboard, fillable Adobe PDF forms, fillable Adobe LiveCycle, and Google Forms. Each of these four options were developed and tested in order to evaluate their respective strengths and weaknesses.

The final platform implementation includes the use of Adobe PDF forms for the case log forms and Google Forms for the recording of evaluations. This combination of platforms was chosen to maximize their respective benefits. The Adobe PDF form was chosen for the case log form primarily because of its compatibility with all tested operating systems and its familiarity with the end-users. Google Forms was selected for its ease of use and its ability to automatically collate submitted data. This combination of platforms will provide a much more efficient process of distributing, collecting, and collating the student clinical data.


Garofali, Michelle L. “Effect of Depolarizing versus Non-Depolarizing Preservation Solution on the Percentage of Myocardial Water Content in Isolated Rat Hearts.”

70pp, bibliography, references, appendix
Available for checkout
Abstract: Every year approximately 600 patients on the heart transplant list die before a heart becomes available. The significant limitation of heart transplantation is the inability to preserve donor heart beyond 4 hours. In an effort to improve donor preservation, several researchers have focused on the factors that limit this preservation time. One of (the) factors that has been shown to play a role is cardiac edema. This study tested the hypothesis that a depolarizing preservation solution versus non-depolarizing preservation solutions influences myocardial water content formation in isolated rat hearts. Methods: To investigate this hypothesis, isolated rodent hearts were obtained from twenty-four Sprague-Dawley rats. Rat hearts were flushed (15ml/kg) and stored (30 ml) for 4 hours at 0-4 °C in unmodified UW solution (n = 10), or a plasmalyte-A solution containing 0.5mmol/L of Pinacidil (n = 8). Following the 4 hours of hypothermic storage, the hearts were weighed to estimate the wet tissue weight and then dried in a food dehydrator for 6 hours, and weighed to estimate the dry tissue weight. The percentage of myocardial tissue water was a variable calculated to estimate cardiac edema. Results: The percentage of myocardial water content (%MCW) (mean ± standard deviation) was increased in the Pinacidil study group (80.6 ± 1.0) when compared to the UW solution study group (71.8 ± 4.1), (p<0.001). There was a difference in the %MCW in respect to the experimental day as well. The greatest difference in %MCW was between experiment day 1 (66.4 ± 3.4) and experiment day 3 (79.8 ± 1.6), (p<0.001 between all days). Discussion: The experiments indicated that UW solution has less influence on myocardial water content following simple preservation and storage than a plasmalyte solution containing pinacidil. The explanation for this finding results is possibly from the difference in osmotic concentrations of the solutions and the effect UW solution has on the efflux of chloride from the myocyte along with water, to prevent cell swelling during preservation and hypothermic storage. The pharmacologic effects that pinacidil has on the inhibition of calcium from vascular smooth muscle may have also contributed to the increase in %MCW by vasodilating the coronary microvasculature and thus, increasing the hydrostatic pressure during flush administration increasing the interstitial fluid flux. UW solution appears to decrease the myocardial edema associated with methods of preservation.

Graff, Nicole M., B.S. “Design and Evaluation of the MSOE Mock Circulatory Loop: Perfusion Simulator Hardware Version 2.0”

Presented at the AmSECT Region VI Meeting under the name Design and Validation of a Perfusion Simulator and Mock Circulation Loop on September 23, 1995 at the Holiday Inn Mart Plaza in Chicago, Illinois with Jeffrey S. Dunn, B.S., E.I.T.
March 26, 1996, 187pp, 43 references, appendices
Available for checkout
Abstract: Mock circulatory loops have been design for many years; however, computer controlled circulatory loops are not common. The mock circulatory loop described in this paper consists of aortic, arterial, and venous resistance and aortic, arterial, venous, and atrial compliance. The motive for including these elements in separate form instead of lumped form as done in other loops was to produce a loop to function as a perfusion simulator. In doing this, the loop must be able to mimic different situations seen by the perfusionist during a cardiopulmonary case. The ability to mimic aortic stenosis, coarctation of the aorta, aneurysms, and the effects of different pharmaceuticals on the vascular tree must be present. Simulators are not manually controlled; therefore, the addition of a computer to control the resistance and compliance is a required element. The software written to control the MSOE Mock Circulatory Loop via an A/D board was written by Jeffrey S. Dunn and Patrick E. Hoppe. The hardware portion of version 1.0 of the MSOE perfusion simulator was designed by Duane R. Pinter. This paper describes the design and analysis of the new MSOE Mock Circulatory Loop and the hardware used to interface with the computer.

Gruenberger, Jeffrey L. “Examining the Relationship between Cerebral Oxygen Saturation Monitoring and Renal Oxygen Saturation Monitoring using Near-Infrared Spectrometry in Pediatric Cardiac Surgery”

June 21, 2007, 52pp, 27 references
  • Appendix A: Technical Specifications of the INVOS 5100B Cerebral Oximetry System by Somanetics Corporation of Troy, Michigan
  • Appendix B: Individual raw data values and Minitab statistical software analysis output
Available for checkout
Abstract: Since 2001, Children’s Hospital of Wisconsin has taken the technology of NIRS monitoring for cerebral tissue perfusion and expanded its application to include systemic, or specifically renal, perfusion. In addition to placing one of the Somanetics INVOS Cerebral Oximetry Systema SomaSensora pads mid-cranium, a second SomaSensora pad is placed near mid-posterior thoracic, immediately superior to the T12 vertebra. This area provides the closest proximity access to the renal vessels in pediatrics. By placing the second SomaSensora pad in this location, the effectiveness of cardiopulmonary perfusion on the renal and other systemic tissue can be monitored. While clinical trends and observations have been made over the years of this adapted use, no studies have been performed to analyze whether statistical significance exists between the standard cerebral oxygen saturation monitoring using the NIRS system and the modified renal, or systemic, oxygen saturation monitoring using the NIRS system. This study sought to find whether statistical relationships exist. Cardiopulmonary perfusion records from twenty patients having the Norwood surgical procedure from July 7, 2006, through March 19, 2007, were reviewed. Correlation between CsO2 and RsO2 was found to be significant (r = 0.422, p < 0.001). Multiple linear regression analyses were performed to determine any significant relationship between the predictor variables of MAP, Q gas, FiO2, CO2, Q Art, Tr, Hct, and Time, and the response variable of RsO2. RsO2 was found to be significantly correlated to only 4 of the 8 predictor values tested, MAP, Q Art, Tr, and Hct. Multiple linear regression p-values were found to be as follows: MAP p = 0.010; Q Art p < 0.001; Tr p < 0.001; Hct p = 0.023. Although the correlation between x and y was found to be significant, it cannot be stated that CsO2 has a direct correlation to RsO2, or that these values are interchangeable. The finding of significance for the variables MAP, Q Art, Tr and Hct were expected, since, while on-bypass, temperature and perfusion/pump pressure which is related to flow become of greater importance in ensuring proper perfusion, and with the relationship of pressure and flow as defined by Ohm.s Law; and since hematocrit defines the basic carrying capacity of oxygen by the red blood cells. It was unexpected to find no significance for the variables Q Gas, FiO2 and CO2, since these three variables represent the respiratory or gas exchange portion of cardiopulmonary bypass and are as critical to proper perfusion as the actual pumping or flow. Possible explanations for the finding of this lack of significance of these variables for RsO2 are the small study group used and limitation to one surgical procedure, the inequality of the number of data values recorded and analyzed from patient to patient, differences in anatomy and physiology between individual patients, and the use of parametric analysis tests on non-parametric data.

Grulkowski, Thomas J. “A Preliminary Investigation Into The Use Of The Cavitation Number Equation To Predict Emboli Formation In The Cardiopulmonary Bypass Circuit”

Dec. 2004, 95pp, 78 references
  • Appendix A: Trial Order
  • Appendix B: Raw Blood Gas Data
  • Appendix C: Three-Way ANOVA Repeated Measures Results
  • Appendix D: Tukey Multiply Comparison Test
Available for checkout
Abstract: Purpose: In this preliminary investigation the cavitation number equation was used as a model if it could accurately predict bubble formation in a cardiopulmonary bypass (CPB) circuit while using porcine blood. There are many documented sources of gaseous emboli production within the CPB circuit, but this current study focuses on cavitation. Cavitation is the pulling of gas out of solution due to pressure gradients within a liquid. Based on this equation, it was predicted that a larger amount of bubbles would be produced at higher flow rates and as the vacuum pressure becomes more negative. Methods and Materials: A standardized model of an adult CPB circuit was utilized. Two flow rates were tested, 3.5 L/min and 5.5 L/min, along with four vacuum pressures, (0, -25, -50, -75 mmHg). These were chosen due to the wide range of cavitation numbers they represented. The cavitation number is a dimensionless number, the higher the number the less likely bubbles will be produced and the lower the number the more likely bubbles are to be produced. There wee four sites on the circuit where bubble formation was recorded; they were the venous line, pump inlet, pump outlet, and post arterial filter. Ten trials were conducted for each flow rate, vacuum pressure, and circuit location, while recording the bubble formation with a Doppler ultrasound bubble detector. Results: A three way ANOVA repeated measures design was used to evaluate the velocity, vacuum pressure, and circuit locations. When comparing the two flow rates there was a significant increase in bubble formation at 3.5 L/min. When comparing the four vacuum pressures, there was a significant increase in bubble formation at -75 mmHg. When comparing the four circuit locations, there was a significant increase in bubble formation at the pump inlet. There are other significant data, but more importantly there is a trend in the mean bubbles produced to the cavitation number in the 5.5 L/min group at all four circuit locations, while this trend is less clear at the 3.5 L/min group. In conclusion the cavitation number can be used as a general guide in bubble formation but to predict an actual number is too difficult from this data at this time.

Gutowski, Gregory G. “Design and Evaluation of a Hemodynamic Cardioplegia Administration Simulation System”

July, 2012, 89 pp, 16 references, appendices, figures, tables
Available for checkout
Abstract: During cardiopulmonary bypass, clinical training of the perfusion student is limited to developing mastery of the fundamental perfusion skills employed during these procedures during the initial stages of training. The perfusion student should be provided an opportunity to test, refine, and develop these skills in a low risk environment in order to become successful clinically. Perfusion simulation training offers the perfusion student this opportunity.Perfusion simulation training has become recognized as an educational imperative among academic institutions offering perfusion training and education. This training has been shown to accelerate knowledge transfer, competency, and clinical performance. The availability of commercial perfusion simulation systems is limited, and these systems are expensive. The Milwaukee School of Engineering (MSOE) currently has a hemodynamic perfusion simulation system that was custom designed and evaluated to serve the needs of the perfusion student in the absence of a commercial perfusion simulation system.

The current system was designed for the perfusion student to initiate and terminate cardiopulmonary bypass and to adjust for changes in systemic vascular resistance during cardiopulmonary bypass. This paper describes the design and evaluation of a hemodynamic cardioplegia delivery simulation that is to be an adjunct to the current system. The adjunct allows the perfusion student to repetitively practice cardioplegia administration in the form of antegrade and retrograde infusions during cardiopulmonary bypass. It is designed in an open fashion to allow for future system additions. Testing of the adjunct system indicates that it offers the perfusion student usability as well as an accurate representation of conditions observed during cardiopulmonary bypass.


Hanson, Kelsey, MT (ASCP) “An In Vitro, Pilot Study to Determine the Direct Antimicrobial Effects of Platelet Gel.”

June 2007, 50pp, 32 references
Available for checkout
Abstract: Background: Surgical site infections are a common complication of surgery, and their economic impact is quite large. Platelet gel is used as a new medical procedure that shows promise in lessening the number of surgical site infections. In vivo comparisons of wounds tend to indicate that those treated with platelet gel have fewer infections and heal faster than untreated wounds. Because platelet gel can contain some white blood cells, as well as a number of biologically active molecules, it is reasonable to predict that the platelet gel might have some direct antimicrobial properties. But to date, no publications have investigated this. The goal of this thesis was to examine if platelet gel has any direct antimicrobial effect against four bacteria that are commonly known to cause wound infections. Methods: Four strains of bacteria (Staphylococcus aureus, Enterococcus faecalis, Escherichia coli, and Pseudomonas aeruginosa) were plated on a blood agar plate. After 24 hours, platelet gel was prepared and sprayed over the plates of the experimental groups. The plates were then incubated at 37 degrees Celsius and checked at days 1, 2, and 3 to assess growth. Results: At days 1 and 2, there was limited to no growth on the experimental plates of E. faecalis and E. coli. There were some colonies seen on the experimental plates of S. aureus and confluent growth seen on the experimental plates of P. aeruginosa. On day 3, there was confluent growth seen on the experimental plates of E. coli. Conclusion: Platelet gel seems to be an effective antimicrobial agent against certain bacteria. Platelet gel did inhibit, at least to some extent, E. coli, E. faecalis, and S. aureus, but did not seem to retard the growth of P. aeruginosa.

Harmon, Amber, B.S. "A Retrospective Comparison on the Effectiveness of TEG Ratio versus ACT in Providing Adequate Anticoagulation in ECMO Patients"

February 2016, 70pp
Available for checkout

Abstract: Extracorporeal membrane oxygenation (ECMO) is a means of long-term support for patients experiencing acute cardiac and/or respiratory failure. One of the major challenges associated with ECMO treatment is keeping patients appropriately anticoagulated. Proper anticoagulation helps to prevent two major ECMO complications: bleeding and thrombosis. Anticoagulation monitoring can be done via multiple methods including: activated clotting time (ACT), prothrombin time (PT), partial thromboplastin time (PTT), antifactor Xa, Antithrombin III (ATIII), and TEG. To date, little has been published regarding which of these monitoring strategies is the most beneficial for reducing blood product usage and hemorrhagic complications. With no single accepted protocol for managing anticoagulation, clinical sites are left to determine for themselves what might work best for their patients. The goals of this retrospective study were to 1) determine if the use of TEG Ratio of R values results in less chest tube drainage, heparin dosing, and blood product administration than ACT, and 2) determine if a TEG Ratio of R values of 1.5 results in less chest tube drainage, heparin dosing, and blood product administration than a TEG Ratio of R values of 2.0.

Data were collected from patients receiving ECMO treatment at Aurora St. Luke’s Hospital between July 2013 and September 2015, and retrieved from their EPIC database. An analysis of 38 patients was performed that looked at the correlation of age, sex, BSA, type of ECMO, cannulation site, length of ECMO, Hct, Hgb, platelet count, ACT, Kaolin TEG R time, Heparinase TEG R time, and anticoagulation protocol with clinical outcomes. The clinical outcomes included were heparin infusion rate, hourly chest tube output, and PRBC and FFP transfusions. Multiple regressions were performed to identify significant predictors of clinical outcomes within three anticoagulation protocol groups: ACT (n=9), TEG Ratio 2.0 (n=12) and TEG Ratio 1.5 (n=17), with post hoc T-tests being performed to compare clinical outcomes between the protocol groups. One-way ANOVA testing of the protocol groups confirmed the results seen in T-testing.

Results showed that heparin infusion rates decreased significantly with the use of TEG Ratio (5.3 ± 4.0) when compared to ACT (13.1 ± 4.7). Chest tube output was higher in the TEG Ratio group (51.5 ± 86.6) compared to the ACT group (38.9 ± 60.4) Final testing for TEG Ratio versus ACT rendered no significant differences in PRBC and FFP transfusions. Comparison between the TEG Ratio groups showed higher heparin infusion rates in the TEG Ratio 1.5 group (5.7 ± 4.0) than the TEG Ratio 2.0 group (4.9 ± 3.9). Chest tube output, as well as PRBC and FFP transfusions, between the TEG Ratio groups showed no significant differences. Analysis comparing ACT, TEG Ratio 2.0 and TEG Ratio 1.5 individually demonstrated significant differences in heparin infusion rates between all groups, but only a significant difference in chest tube output between the ACT and TEG Ratio 2.0 group. This study shows that use of a TEG Ratio is a safe and effective means of managing anticoagulation in ECMO patients when compared to ACT.


Herrmann, Katelyn “Improving the Viability of Donor Hearts for Transplant: Design and Evaluation of a Circuit to Perfuse an Isolated Rat Heart”

February 2015, 66pp, 30 references
Available for checkout
Abstract: The field of heart transplantation faces many challenges, but few are as substantial as the large discrepancy between the supply and demand of donor hearts. Due to the shortage of available organs, patients in need of a heart transplant face a high risk of waitlist mortality. The research described in this document is intended to address this issue by investigating the use of extended criteria donors.

A proof of concept study was designed to investigate the potential for using a custom designed asanguineous solution and ex-vivo perfusion circuit to preserve and recondition hearts that otherwise do not meet donation criteria to restore them to normal, viable function – thus expanding the donor pool. The purpose of this project specifically was to design and operate, according to given specifications, a circuit to perfuse an isolated rat heart.

To date, the designed circuit has been successfully used to perfuse a total of five isolated rat hearts. While the proof of concept study remains in its early phase, preliminary results are promising as the hearts exposed to six hours of ex-vivo perfusion with the custom solution show minimal signs of irreversible cellular injury.


Hietpas, Todd “The Effect of Aortic Cannula Design on Gaseous Microemboli Distribution into the Head Vessels: an In Vitro Study”

February 2012, 69pp, 53 references
Available for checkout
Abstract: Cerebral microemboli are a common cause of neurological injury. Unfortunately, these microemboli are somewhat common in patients undergoing surgeries that utilize cardiopulmonary bypass (CPB). This is due to gaseous microemboli (GME) forming secondary to various perfusionist interventions. It is the goal of the perfusion community to limit these GME as much as possible, which requires an understanding of how each intervention affects GME production and distribution. The purpose of this study is to investigate the impact of aortic cannula design on the treatment of GME in the aorta and their distribution into the head vessels.An in vitro cardiopulmonary bypass (CPB) circuit that included a cannulated aortic arch model was constructed to test the distribution of GME into the head vessels following injections of 1mL and 5mL of air into the arterial line of the circuit. Seven aortic cannulae of varying outlet shape, length, orifice area, and orifice configuration were tested at each bolus volume. GME volume and count were measured in the model’s innominate, left common carotid, and left subclavian arteries using the Emboli Detection and Classification (EDAC) quantifier from Luna Innovations.

The results of this study demonstrate that aortic cannula design can significantly affect the GME count and total air volume distributed into the innominate and left common carotid arteries during CPB. The top performing cannulae are the Medtronic Select 3D with angled tip, diverted stream, and low pressure drop, and the Sarns D4 with angled tip, single end hole and low pressure drop. The Select 3D passed significantly less microembolic total volume, total count and counts in both the 0-40μm and 40-100μm ranges to the innominate artery, while the D4 passed significantly less microembolic total volume, total count and counts in the 0-40μm range to the left common carotid artery. Diversion of GME away from the head vessels could limit passage into the cerebral circulation and hence reduce the incidence or severity of post-CPB neurological dysfunction. Selection of the aortic cannulae for clinical use that demonstrate the ability to reduce the embolic volume and count distributed to the head vessels is yet another tool available to reduce the incidence of post-CPB morbidity.


Hoppe, Patrick E. “Perfusion Simulator Software.”

April 17, 1995, 52pp, 13 references
  • Appendix A: Hardware Specifications
  • Appendix B: Database Structure
  • Appendix C: Perfusion Software Code
Available for checkout
Abstract: The current educational curriculum for a perfusion student includes both clinical and didactic work. The clinical experience is comprised of observing actual cases and then operating the perfusion equipment under the direct supervision of a certified perfusionist. There is a need for a perfusion simulator which provides the student with all the physical and theoretical requirements of an actual cardiopulmonary bypass case while allowing them to make mistakes without a catastrophic outcome. The Milwaukee School of Engineering’s Master of Science – Perfusion program is currently developing a clinical perfusion simulator. The perfusion simulator is comprised of a modified Donovan Loop, a personal computer, and a cardiopulmonary bypass pump. The personal computer controls the Donovan Loop via a 12 bit Data Acquisition Board (CIO-DAS1601/12). The D/A board converts signals from the computer into voltage levels which control the valves and motors on the modified Donovan Loop. Pressure and flow sensors located within the Donovan Loop provide feed-back loops for the software running on the computer. The software controls the Systolic and Central Venous pressures within the Donovan Loop to provide the student with the same experience as found during an actual cardiopulmonary bypass procedure. The control software, written in Visual BasicTM provides a user friendly interface and allows for future expansion into several areas. Future modifications planned for the software are the inclusion of pharmacology, temperature control, and volume changes.

Howard, Jonathan R. “Design and Evaluation of a Hemodynamic Perfusion Simulator”

April 20, 2010, 92pp, 25 references
Available for checkout
Abstract: Mastering the fundamentals of cardiopulmonary bypass is essential for a perfusion student to be clinically successful. Perfusion simulation offers a student the opportunity to master these skills in a less stressful, low risk environment. Simulation training is quickly becoming an educational standard, and it has been shown to increase clinical performance and competency. However, commercially available perfusion simulators are expensive. This paper describes the design and evaluation of a low cost hemodynamic perfusion simulator. The simulator allows students to repetitively practice initiation and weaning of cardiopulmonary bypass. It also allows students to adjust for changes in systemic vascular resistance while on bypass. The simulator provides the bare essentials for perfusion simulation and is designed in an open fashion to allow for future additions. The simulator is easy to use and accurately reproduces conditions seen in the operating room. Evaluation of the simulator showed that it will help both first and second year MSOE Perfusion students master the fundamentals of cardiopulmonary bypass.

Hummel, Richard T. “In Vitro Comparison of Carmeda and Duraflo II Heparin Bonded Circuits.”

June 15, 1995, 45pp, 46 references
Available for checkout
Abstract: The purpose of this experiment is to compare two different heparin coated cardiopulmonary bypass (CPB) circuits which are both commonly utilized during cardiothoracic operations. Duraflo II (Baxter-Bentley* Healthcare Corp.) is composed of ionically bound heparin, while the Carmeda BioActive Surface (Medtronic Inc.) has heparin covalently bound to a prepared surface resulting in end-point attachment. The combination of elevated activated clotting time (ACT), activated partial thromboplastin time (aPTT), and protamine titration in the Duraflo circuits indicate that heparin is immediately released from the blood contact surface upon initiation of extracorporeal circulation (ECC) and continues to be liberated throughout the circulation time. Increased plasma free hemoglobin levels in the Duraflo coated circuits indicate that they may also be less blood-compatible than the Carmeda circuits. The loss of heparin from the blood contact surface likely has some effect on this difference although it can not be proven from this study whether this was the main determinant. Some of the hemolysis could be due to shear stresses exerted on the blood within the individual components which would be a result of product design. Nevertheless, the covalently bound Carmeda circuits retain their heparin coating more efficiently and thus may offer more benefits in the long run.

Hurd, David K. “A Study to Determine the Adsorption Potential of a Hydroxyapatite Preparation for Removal of Heparin from Circulating Blood”

Nov. 2003, 57pp, 25 references
Available for checkout
Abstract: The purpose of this study was to assess the heparin removal potential of hydroxyapatite as a possible alternative to protamine. For almost 30 years, protamine has been used as a reversal agent for heparin (7). In a recent study, it was reported that almost 2% of heart surgery patients have some degree of protamine sensitivity (13). The same study also reported that 0.1% of heart surgery patients died as a result of a protamine reaction (13). Many alternatives to protamine have been investigated with varied degrees of success (12). This study used hydroxyapatite in a mock bypass circuit to adsorb (neutralize) heparin. Adsorption beds were created with various amounts of hydroxyapatite, in order to assess their adsorption potential. Since hydroxyapatite is a natural component of bone, it is anticipated that the allergic reactions seen with protamine administration could be reduced or avoided. A previous investigation at The Wales College of Medicine by Wassell and Embrey (24) found that heparin and serum bovine albumin competed for the same binding sites on the hydroxyapatite crystals. The groundwork of Wassell and Embrey was the motivation for this study. It was hypothesized that hydroxyapatite’s natural binding properties with negatively charged elements would bind heparin and remove it from blood. The study consisted of 11 experimental trials in which the hematocrit, plasma volume, blood volume and amount of adsorbent were varied. Parameters such as temperature, line pressure, pH, and flow were monitored to assure consistency between trials. The average decrease seen in the activated clotting time (ACT) was 336.3 seconds. Heparin concentration decrease for the eleven trials averaged 2.83 ug/ml. These decreased values for ACT and heparin concentration show hydroxyapatite’s effectiveness as a neutralizing agent. Evaluation of hemolysis of test samples compared favorably to the baseline protamine sample and pH values change at most 0.11 pH units between the beginning and end of the trials.

Inman, Laura “A Model of Acute Renal Failure Risk Following Cardiopulmonary Bypass at St. Luke’s Medical Center, Milwaukee, Wisconsin”

May 25, 2006, 96pp, 240 references
Available for checkout
Abstract: Introduction — Background — Methods — Results — Discussion — Conclusion — Appendix A- List of variables and categories included for analysis B- Fluid balance equation and fluid balance results. Purpose: Acute renal failure is a serious complication of cardiac surgery utilizing cardiopulmonary bypass. This study aims to develop a risk model for post-operative renal failure. Using this model, the clinician can determine risk of renal failure, and therefore take steps to limit any modifiable actions which may increase this risk. Methodology: This retrospective study examined patients undergoing coronary artery bypass grafting with cardiopulmonary bypass during the period of December 2004 to January 2005 at St. Luke’s Medical Center, Milwaukee, Wisconsin. Exclusions were reoperations, concomitant procedures, and preoperative creatinine level greater than 2.0 mg/dl. Forward step logistic regression using the software package SAS was used to develop a model of renal failure risk using pre-determined variables. Major findings: A model was developed to characterize the major risks of postoperative renal failure in the patient population at St. Luke’s. Factors which increased risk, in descending order, are: use of pump suction (3.9 odds ratio), creatinine greater than 1.6 mg/dl (3.1), age 70-79 (2.8), age 80 or older (2.7), pump run greater than three hours (2.5), use of loop diuretic on-pump (2.2), peripheral vascular disease (2.1), body mass index greater than 32 (2.0), aprotinin at more than 600 ml, if creatinine is greater than 1.5 (1.9), congestive heart failure (1.8), diabetic on oral medication or insulin (1.6), and 4 or more distal anastamoses (1.3).

Jackson, Darren E., B.S. “Effect of Simultaneous versus Sequential Antegrade/Retrograde Cardioplegia on the Percentage of Myocardial Water Content in Isolated Fetal Bovine Hearts”

May 22, 2004, 50pp, 42 references
Available for checkout
Abstract: Background: Cardiac surgeries that utilize techniques to stop the heart during surgery have been common since about the 1960s. However, there are still numerous instances of cardiac dysfunction following cardiopulmonary bypass. One of the most important factors in this cardiac dysfunction is myocardial edema. Myocardial edema has been shown to decrease systolic cardiac function, decrease cardiac output and impair diastolic function. During cardiac surgery the technique of cardioplegia delivery can have an effect on the formation of myocardial edema. This study tested the hypothesis that simultaneous antegrade/retrograde cardioplegia (SARC) will increase myocardial edema when compared to sequential antograde/retrograde cardioplegia. Methods: Fetal bovine hearts were obtained from Emmpack Foods, Inc. The hearts were separated into two groups, SARC group (n = 14) and sequential cardioplegia group (n = 14). A Quest MPS? Myocardia Protection System was utilized to perform the cardioplegia delivery. The SARC group received a total of 1000 mL of Plasmalyte A, while the sequential group received 500 mL antegrade and 500 mL retrograde. Following cardioplegia delivery, the hearts were weighed to estimate the wet tissue weight and dried in a food dehydrator for 24 hours, and weighed to estimate the dry tissue weight. From these tissue weights the percentage of myocardial tissue water content was calculated to estimate myocardial edema. Results: The percentage of myocardial water content (%MCW) (mean +/- standard deviation) was increased in the SARC group (78.15 +/- 2.23) when compared to the sequential cardioplegia group (75.20 =/- 3.62. It was also determined that the weight of the heart had an influence on the %MCW. Discussion: This study indicated that SARC has more influence on the formation of myocardial edema than sequential cardioplegia. This may be due to the increase in intramyocardial tissue pressure when antegrade and retrograde cardioplegia are administered simultaneously. There was also an effect of the weight of the heart on %MCW, which may be explained by the difference in cc/gram of cardioplegia administered between small hearts (5.00 =/- 0.96 cc/gram) and large hearts (3.09 +/- 0.48 cc/gram).

Justman, Megan “Recommendations for Improvements in Extracorporeal Membrane Oxygenation (ECMO) Training at Aurora St. Luke’s Medical Center”

February 2017
Available for checkout
Abstract: Aurora St. Luke’s Medical Center (SLMC) is currently not recognized as a Center of Excellence by the governing body of Extracorporeal Membrane Oxygenation (ECMO). In analyzing the criteria that needs to be met in order to apply for the Extracorporeal Life Support Organization’s (ELSO) prestigious award, it was noted that one of the shortcomings was that the facility lacked a training program for those managing the circuit and at the bedside of the ECMO patient. Therefore, the goal of this project was to develop assessment tools and a training manual to aid in the onboarding process for a newly hired perfusionist. These tools can also be utilized as competency assessments and review for established perfusionists as well as by students completing their clinical rotation at this facility.
The assessment and training program was primarily developed utilizing ELSO’s guidelines for educating ECMO specialists and a clinical assessment framework known as “Miller’s Pyramid”. This pyramid features four tiers to assess that an individual “knows”, “knows how”, “shows how”, and “does” the duties to successfully perform their job. To align with this framework, an assessment tool was provided for each of the first three tiers in order to assess the newly hired perfusionist’s skills. In addition, a training manual was created to review the products used for ECMO at SLMC as well as institution policies and procedures. By having a training program in place, Aurora St. Luke’s Medical Center will be one step closer to being eligible to apply for the Center of Excellence Award.

Kallies, Kirsten “An In Vitro Quantitative Analysis of Gaseous Microemboli Production Using a Centrifugal Pump in Pulsatile Mode”

April 2008, 159 pp, 124 references, appendices
Available for checkout
Abstract: A significant amount of research has been dedicated to the debate over the benefits of pulsatile perfusion. Similarly, much research has focused on the debate over which type of arterial pump head should be utilized for cardiopulmonary bypass (CPB) — roller pump or centrifugal pump. However, no experiments have combined these two highly debated areas in order to study the possibility of air production when a centrifugal pump is used to impart pulsatile perfusion. Therefore, the purpose of this experiment was to employ a pulsatile waveform architecture in order to observe any air produced within the experimental CPB circuit.
Consistent with similar studies conducted with roller pump heads, the independent variables chosen to be maniuplated in order to observe their effect on air production, were mean flow and baseline percentage. Additionally, these were the only two variables which could be manipulated on the Sarns Centrifugal System control module to alter the pulsatile waveform generated. The dependent variables, monitored via the embolus detection and classifier (EDAC) QUANTIFIER manufactured by Luna Innovations, Inc., were the volume of air produced and the number of gaseous microemboli (GME) detected. Additionally, pre-pump head pressure was monitored via a BioTek DPM-2 Digital Pressure Meter and recorded onto a laptop computer with Biobench software. The pressure data were then used to calculate average and minimum pressures during each trial in order to conclude whether there was a possibility for cavitation to occur in the system.
For the experimental materials and methods used in this study, the independent variables of mean flow and baseline did not have a statistically significant effect on either the total volume of air or total bubble count. However, mean flow did have a statistically significant effect on both the average and minimum pressures measured at the pre-pump pressure port. Additionally, both the average and minimum pressures were significantly lower (more negative) at the higher mean flow setting of 5.5 L/min. Baseline did not significantly influence either set of pressure data.
In conclusion, for the experimental materials and methods used in this study, the independent variable of baseline had no statistically significant effect on any of the dependent variables observed in this study. These results were consistent with the results of similar studies conducted with roller pump heads. Therefore, clinicians can conclude that no matter the type of arterial pump head, the baseline setting does not affect the number of GME produced in the CPB circuit.
Further, mean flow exhibited an effect on the pressure data, but it did not have an effect on GME production. Therefore, clinicians must keep in mind, when using a centirfugal pump head, that increasing flows will cause a decrease in pressure at the pump inlet, and thus, a potential for cavitation to occur.

Klees, Theresa "A Preliminary Study Comparing the use of Microplegia and Del Nido Cardioplegia Systems and Their Effects on Cardiac Electrophysiology"

May 2015, 46pp
Available for checkout

Abstract: Cardiopulmonary bypass (CPB) has become common practice in most cardiac surgeries today. Often these cases require the cessation of electrical activity accomplished through the use of an aortic cross-clamping and cardioplegia administration. This procedure provides added protection by reducing metabolic activity of the heart while providing a bloodless and motionless field for the surgeon to operate. Currently there are several cardioplegia solutions that can be used during CPB, and there is no consensus as to which is best. The objective of this project was to explore the effects of two cardioplegia solutions, del Nido and microplegia, in their ability to prevent immediate post-surgical arrhythmias.

Data were collected via an online survey completed by the perfusionist at Aurora St. Luke’s Medical Center at the end of each case, and a statistical analysis was performed. A total of 104 surveys were completed, 50 of which used del Nido and 53 used microplegia. The parameters used to determine adequacy of myocardial protection were the time to return to normal sinus rhythm, whether or not external pacing was required, and if defibrillation was attempted to achieve a normal rhythm after the cross clamp was removed.

The findings of this study showed that when del Nido cardioplegia was used, the heart required more time to return to normal rhythm and pacing wires were used more often. It also revealed that when microplegia was used, there was a greater incidence of defibrillation. These results indicate a difference in electrophysiology between the two cardioplegia types with a large incidence of unexplained variability. These results suggest that further study should be completed before adjusting current protocols and practices as this was only a preliminary study to pave the way for a more comprehensive proactive study to determine which type of cardioplegia is superior over the other.


Knechtsberger, Julie Ann “An In-Vitro Study on the Use of an Arterial Bubble Trap Incorporated into the Venous Lines”

June 2006, 83pp, 29 references
Available for checkout
Cardiopulmonary bypass has been linked to neurocognitive dysfunction in numerous studies. Air emboli are one of the main causes of neurocognitive dysfunction. The air emboli can come from the surgical field, the heart lung machine, or from anesthesia. This study investigated the amount of venous air returning to the heart lung machine from the surgical field. Venous air can come from a number of sources, including; loose purse strings, a trans-atrial approach to a mitral valve repair or replacement, or openings in the venous lines of the heart lung machine. Air may also come from an atrial septal defect, a patent foramen ovale, or if the left atrium or left ventricle are opened. Since venous air comes from a variety of sources and is harmful to the patient, a device to remove this air could limit the amount of air emboli and improve patient outcomes. The purpose of this study was to look at the effects of placing an arterial bubble trap in the venous lines to reduce these air emboli. This was done by comparing the use of a bubble trap placed in the venous line to a circuit without a venous bubble trap under conditions of three air injection doses. Additionally the study tested the efficiency of the bubble trap in the venous lines (it is usually placed in the arterial line) by comparing air volume amounts before and after the bubble trap as measured by probes on each side of the trap. For each set of experimental conditions ten trials were performed, and the data were assessed using Minitab (version 14). An arterial bubble trap placed in the venous lines was found to be effective at air removal when the volume amounts before and after the bubble trap were analyzed. There was a statistically significant reduction in air volume amounts when analyzing the injection site volume and after the arterial line filter with a bubble trap in the circuit as well as without a bubble trap in the circuit. There was not a statistically significant difference in the air volume amount removed when comparing a circuit with a bubble trap to one without a bubble trap in the venous lines. One possibility for this is the efficiency of the circuit’s components at air removal. Based on the results of this study, the arterial bubble trap does reduce the amount of air before it enters the cardiotomy and should be considered as another form of prevention for air emboli reaching the patient.

Krueger, Justin D. “The Effects of High Shear Stress on Platelet Function as a Result of Pulsatile Cardiopulmonary Bypass”

Dec. 2005, 62pp, 52 references
  • Appendix A: Derivation of wall shear stress equation under turbulent flows and data for Reynolds number and shear stress calculation
  • Appendix B: Two-way ANOVA SigmaStat report for percent platelet aggregation
  • Appendix C: SigmaStat multiple linear regression report for percent platelet aggregation
  • Appendix D: One-way ANOVA SigmaStat report for baseline platelet counts
  • Appendix E: One-way ANOVA SigmaStat report for baseline HCTs
  • Appendix F: Two-way ANOVA SigmaStat report for change in blood temperature
  • Appendix G: Raw data
Available for checkout
Abstract: Pulsatile cardiopulmonary bypass has been associated with better patient outcomes in respect to postoperative organ function and hemodynamics. In theory, this is because pulsatile flow better imitates the flow characteristics of the native heart, which can be described as having a .pulse pressure. in the arterial pressure waveform. When utilizing a pulsatile pump, an increase in the pulse pressure can be obtained by decreasing the .base flow percent. parameter of the pulse cycle as long as there is a concomitant increase in the high flow rate phase of the pulse cycle. This is necessary to maintain the same average flow rate. However, the higher peak flow rates and increased turbulence that occurs with the high flow rate results in increased shear stresses being applied to the blood and its components. These shear stresses are in the ranges that have been shown to cause shear-induced platelet activation/aggregation (SIPA). It is known that many factors of cardiopulmonary bypass lead to platelet dysfunction; which, consequently result in an increased tendency of uncontrolled bleeding in the post-CPB patient. Therefore, with the use of pulsatile CPB it is possible that the increased shear stresses produced may cause increased platelet dysfunction in comparison to non-pulsatile CPB. The goal of this study was to determine the effect of increased shear stress as a result of pulsatile cardiopulmonary bypass on platelet function, as assessed by the platelets. ability to aggregate in the presence of a platelet agonist. To perform this study, a total of 16 experimental trials were conducted in which bovine blood was perfused through extracorporeal circuits and exposed to increased levels of shear stress produced by varying the base flow percentage setting of the pulsatile cycle. Each trial length was 90 minutes at a base flow percentage of 100% (non-pulsatile), 70%, 50%, or 30%, with all trials utilizing an average flow rate of 5 L/min. The Platelet Works platelet function analysis system (which utilized ADP as the agonist) was used to determine platelet function at the initial, 45 minute, and 90 minute intervals. All statistical analyses performed were considered significant at a p<0.05. No significant differences were found to exist between base flow percentage group (shear stress), time, and percent platelet aggregation. Platelet counts were unexpectedly found not to have decreased as time progressed. Additionally, no significant differences were found between base flow percentage groups and change in blood temperature. However, there was a significant change in blood temperature found to exist in all groups between the initial time point and 45 minutes. The results of this study were in agreement with those that had determined that pulsatile flow does not lead to increased post-surgical blood lost. However, these results differ from those that have found increased platelet dysfunction and increased SIPA with increased shear stress and increased time of exposure to shear stress. There were multiple factors that could have possibly influenced the results of this study. This appears to be the first study to attempt to determine what effect pulsatile CPB has on the ability of platelets to aggregate in the presence of an agonist. Based on the results of this study it is suggested that further research be performed using larger sample sizes and a direct thrombin inhibitor as an anticoagulant. Additionally, clinical patient studies would be beneficial to determine if the platelet dysfunction resulting from the shear stress produced by pulsatile CPB is significantly small in comparison to the global dysfunction that occurs upon exposure to the extracorporeal circuit and heparin.

Larson, Jessica, B.S. “Retrospective Investigation on the Effectiveness of Weight Based Mannitol Administration on Kidney Function during Cardiopulmonary Bypass”

February, 2014, 48pp, 29 references, figures, tables
Available for checkout
Abstract: Mannitol is a diuretic compound that has been proposed to help minimize renal dysfunction during surgeries utilizing cardiopulmonary bypass (CPB). There is, however, debate within the perfusion community over the level of efficiency of mannitol, and how it should be dosed in order to achieve perceived benefits. Literature comparing dosing strategies with renal function outcomes is limited and conclusions are inconsistent in determining effectiveness of varying doses and times of administration. To date, there have been no published studies that have investigated the most effective weight based dosing strategy of mannitol. The goals of this retrospective study were to 1) determine if a standard does of mannitol (25 g) in the pump prime differentially affected renal function in patients of varying weights, and 2) determine if the use of a 25 g pump prime dose plus repeated doses of 12.5 g/hr were more effective in preventing renal dysfunction.

Data were collected from cardiac surgery patients placed on cardiopulmonary bypass between November 30th, 2011 and August 31st, 2013. Data were received from Aurora St. Luke’s Medical Center’s Epic database, the Society of Thoracic Surgeons (STS) Database, and LUMEDX perfusion databases. The patient group was limited to those with normal pre-surgical renal function, common surgeries, and under 80 years of age. The ratio of post-operative to preoperative serum creatinine was used as the indicator of renal dysfunction. Patient groups were also analyzed for confounding variables including gender, age, laboratory values, cardiopulmonary bypass and aortic cross-clamp times, and type of procedure performed.

First, patients receiving only the standard dose of mannitol in the pump prime (25 g) were isolated (N=137 patients) and then divided into two groups, mimicking a weight based dosing strategy, where those patients in the lowest weight group had the highest weight based dose using a standard pump prime of mannitol. Statistical analysis was performed utilizing a one-way ANOVA. Secondly, the effects of repeated administration of 12.5 g/hr of mannitol (N=517) were analyzed utilizing multiple regression, and one-way ANOVA followed by Tukey comparison testing. In order to determine which dosing strategy was more effective, standard pump prime or repeated dosing, a one-way ANOVA was performed across strategies and weight classification. Urine output testing was also performed with correlation and one-way ANOVA testing.

Results showed that when administering 25 g of mannitol in the pump prime only, patients receiving >0.25 g/kg had lower serum creatinine ratio mean (1.03 ± 0.18) when compared to those receiving <0.25 g/kg (1.22 ± 0.44) (p<0.01). Hourly weight based dosing was a significant predictor of serum creatinine ratio (p<0.05) and a preliminary result found >0.45 g/kg/hr resulted in the lowest serum creatinine ration (1.05 ± 0.25). As expected, urine output was significantly correlated to overall doses of mannitol (r=0.410,p<0.001).

This study demonstrated that weight based mannitol may be more effective than standard dosing but further study is required to determine the most effective dosing strategy in order to prevent renal dysfunction postoperatively in cardiac patients.


LaValley, Matt “The Effect of Various Methods of Drug Input into an Open-System Cardiopulmonary Bypass Circuit on Gaseous Microemboli Introduction to the Arterial Line: An In Vitro Analysis”

May 2011, 45pp, 21 references, figures, tables, appendices
Available for checkout
Abstract: Gaseous microemboli (GME), or microscopic air bubbles, have been shown to contribute to post-surgical neurological dysfunction following procedures that use cardiopulmonary bypass (CPB). Although there are many ways for air to enter the CPB circuit and subsequently travel to a patient’s vasculature, most air enters the circuit by way of perfusionist intervention, or the input of fluids into the venous reservoir. It is well established that there are a variety of methods of inputting drugs into a venous reservoir, including both the use of a 60cc syringe atop the cardiotomy as a funnel, and direct injection into the sampling manifold. This study investigates the effects that four different ways to input drugs into the circuit have on gaseous microembolic load within the circuit.An in vitro CPB circuit which included a patient representative reservoir was constructed to test both the total volume and total count of gaseous microemboli in the circuit due to perfusionist intervention. The Emboli Detection and Classification quantifier (EDAC) was used to collect the data.

The results of this experiment show that significantly more air is input into the circuit when a purged sampling manifold is used as the vehicle for drug introduction into the circuit. Using the funnel-syringe atop the cardiotomy with either a pre-opened or closed stopcock provides far less air to the CPB circuit. Also, using a non-purged manifold port yields comparable results to using either funnel-syringe method. In conclusion, it is better to avoid purging the sampling manifold. When the purging sampling manifold is necessary, using an emboli shunt could reduce the embolic load that passes through the extracorporeal circuit.


Legaspi, Harry, B.S. “Does Adding Albumin to the Prime Solution of a Circuit with a Trillium Biopassive Surface-Coated Oxygenator Benefit Platelets?”

May 2011, 65pp, 47 references, figures, tables, appendices
Available for checkout
Abstract: Cardiopulmonary bypass (CPB) activates blood to such an extent that hemostatic derangements are common in the post-operative setting. A significant amount of activation occurs because of blood-biomaterial interactions. In particular, many platelets become activated through interactions with surface-adsorbed fibrinogen. Because platelets are vital contributors to hemostasis, preserving them through mechanisms that decrease their activation may be beneficial to patients. Two common ways of reducing platelet and blood activation are to use circuits with proprietor-developed surface coatings or prime the circuit with an albumin-containing solution. Although many studies assess the individual contributions of albumin priming and surface coatings on platelet preservation, few studies have evaluated whether utilizing both methods results in greater preservation than either method alone. This study seeks to determine whether platelet preservation is enhanced if an in vitro circuit containing a coated oxygenator is albumin primed.

Trillium Biopassive Surface-coated Affinity oxygenators (Medtronic, Minneapolis, MN) (n=10) were separated into three treatment groups based on the level of albumin pre-treatment (albumin priming). Group 1 was the control and received no albumin pre-treatment, group 2 received a low concentration of albumin (0.0375 grams albumin per 100 cm³ prime), and group 3 received a high concentration of albumin (0.125 grams per 100 cm³ prime). A simple circuit consisting of an Affinity oxygenator with an integrated cardiotomy/venous reservoir and pump head was used. These circuits were primed with an isotonic solution and the appropriate amount of albumin. Bovine blood was added to the circuit after this solution had circulated for 20 minutes. Platelet count and function was analyzed using a platelet function analyzer (Plateletworks Whole Blood Aggregation and CBC Testing Device, Helena Laboratories, Beaumont, TX) prior to trial initiation (baseline), then at 5, 15, and 60 minutes trial initiation.

The platelet count of the control group was less than the albumin groups (p<0.001 between control and both pre-treated albumin groups). However, this result does not indicate that the albumin preserved platelet count because all three groups showed a similar platelet count drop for the duration of the experiment (p=0.38). Platelet function, assessed by collagen aggregometry, indicated that the percentage of platelet aggregation of the control group was higher than the albumin groups (p<0.02 between the control group and both albumin pre-treatment groups). The percentage of platelet aggregation did not decrease during the duration of the experiment (p=0.06).

Albumin priming resulted in no difference in platelet count at any time interval measured in this study. The data do not show that albumin-primed circuits containing an oxygenator coated with the Trillium Biopassive Surface preserves platelet function compared to a similar circuit that is not albumin primed. Future studies should use a more precise platelet activation test, such as light transmission aggregometry, or immunoassay testing for markers of platelet activation.


Lindquist, Tiffany K., B.S. “A Risk Assessment Study of Co-Morbidities Associated With The Transfusion of Packed Red Blood Cells During Elective Coronary Artery Bypass Grafting”

Sept. 1, 2006, 106pp, 90 references
Available for checkout
Abstract: Ischemic heart disease has tripled in the last two decades, and in conjunction with the increased incidence of disease, is a concomitant rise in the use of antithrombotic and fibrinolytic drugs used during heart surgery. These drugs, although necessary, increase the potential for bleeding among the patients who need them, and may lead to treatment by blood transfusions. Among the numerous decisions that must be made during the management of the cardiac surgery patient, the decision of when to transfuse is among the most controversial. In theory, transfusing blood increases oxygen delivery and improves outcomes, but there is no definitive evidence to support such a practice. In fact, although transfusions of packed red blood cells (RBCs) have been shown to increase with patient factors such as age, gender, and disease states, such transfusions have been independently linked in some studies to a more negative patient outcome after a variety of surgeries. A few of the adverse effects include increased infection, prolonged mechanical ventilation and mortality rates. In order to try to determine when RBCs should be given, a retrospective study was performed on patients admitted to Aurora Center Hospitals in the Milwaukee area who underwent elective coronary artery bypass grafting (CABG). Patients were grouped by the amount of units of leukocyte-reduced packed red blood cells (pRBCs) they had received during the intraoperative and postoperative periods. This study was carried out to determine which combinations of preoperative demographics and risk factors were the best predictors of pRBC usage during elective CABG. In addition, a multivariate logistic model was used to statistically compare the dose-reponse relationship in infection, prolonged mechanical ventilation and mortality to the amount of pRBC units transfused. Results of this research will be used to predict patient transfusion needs and compare the benefits to the risks in deciding when it is best to transfuse pRBCs using a risk score analysis. The model will then be able to predict pRBC usage and its risks for more complex surgeries, such as redo, valve and aneurysm operations.

Martin, Samantha J. "The Role of Preoperative Hematocrit and Body Surface Area in the Gender Gap Seen in Cardiac Surgery Outcomes"

April 2013, 87pp, 39 references
Available for checkout
Abstract: Since the 1970s, research has shown that men have better outcomes than women after cardiac surgery. This difference between sexes is termed the “gender gap” and determining the source of this gap could lead to changes in treatments that could maximize outcomes for men and women. One possible reason that this gap may occur in cardiac surgery outcomes is the use of cardiopulmonary bypass (CPB). This may be because the hemodilution associated with going “on-pump”, in conjunction with the lower average pre-surgical hematocrit and body size of women, leads to lower intraoperative hematocrit. Decreased hematocrit is itself a risk factor for receiving blood products, which may contribute to the difference in outcomes.

The current study is aimed at determining if a gender gap exists in surgical outcomes from a single cardiac surgeon at Aurora St. Luke’s Hospital, and if so, if the difference would be associated with decreased pre-operative hematocrit and smaller body surface area (BSA) as compared to men. A retrospective analysis of 101 patients was performed that looked at the correlation of pre-operative hematocrit, BSA, age, gender, CPB time, cross-clamp time, incidence of comorbidities, and the use of antifibrinolytics with clinical outcomes. These outcomes included intubation time, length of ICU stay, length of total stay, post-operative mortality and the occurrence of complications including arrhythmias, renal dysfunction, infection, stroke, bleeding, and use of intra-aortic balloon pump. T-tests and Chi-Square tests were performed to compare pre- and intra-operative conditions of males and females, as well as their outcomes. Logistic, general, and multiple regressions were done to compare significant predictors of male, female, and overall outcomes.

Of all the pre- and intra-operative variables recorded only two differed between genders; males had higher pre-operative hematocrit and larger BSA compared to females. Analysis of post-operative data revealed no difference between males and females in any of the outcomes recorded. However, regression analysis did suggest that significant predictors for intubation time, length of ICU stay, total length of stay, and arrhythmias may differ between sexes.


McNair, Trisha M. “The Effects of High Shear Stress on Platelet Function.”

May 6, 2003, 60pp, 56 references
Available for checkout
Abstract: Platelets play a major role in maintaining hemostasis of the body after all types of insults. They are especially necessary after a major surgical procedure that requires cardiopulmonary bypass. Yet cardiopulmonary bypass has itself been shown to cause platelet dysfunction and destruction through intrinsic and extrinsic mechanisms. One of the reasons for platelet damage may be the shear stress produced within an extracorporeal circuit. When vacuum assisted venous return is implemented into the circuit, there is an even greater shear stress produced which could cause greater trauma to the platelets. The goal of the current study was to evaluate the effect that shear stress caused by vacuum assisted venous drainage has on platelet function. To perform this study a total of 24 experimental trials were performed in which bovine blood was perfused through circuits exhibiting different levels of shear stress. Each trial was run for two hours at a pressure of 0, -15, -30, or -45 with a 30 or 36 French venous cannula. A Plateletworks machine was used to determine platelet function at the initial, one hour, and two hour intervals. Within the first hour of CPB there was a decrease in platelet function compared to the initial reading. There was also a correlation between the level of shear stress and the degree of platelet dysfunction. A decrease in white blood cell count was also found between the initial and two hour time points. The changes that were observed during this experimentation are consistent with past studies and with the known effects that artificial surfaces and shear stresses have on platelets. This is the first study that has attempted to evaluate the effect of high shear stress caused by vacuum assisted venous draining on platelet function. Based on this study it is recommended that further research be performed to determine the effects of increased shear stress on platelets over a greater range of shear stress and within larger groups.

Moaref, Kourosh, B.S., E.I.T. “Modeling of Coronary Perfusion in the Presence of a Left Ventricular Assist Device “

June 29, 2000, 146pp, 68 references
  • Appendix A: Complete Model Listing
  • Appendix B: Elastance Functions
  • Appendix C: LVAD Ejection Delay & Coronary Flow
  • Appendix D: LV Elastance & Coronary Flow
  • Appendix E: Heart Rate & Coronary Flow
Available for checkout
Abstract: Left ventricular assist devices have long been utilized in patients with endstage heart failure awaiting cardiac transplantation. Their purpose, in this application, has been to sustain and improve the function of the patient’s other vital organs until such a time when a donor heart becomes available. However, recent advances in LVAD technology and reliability has sparked new interest in the use of these devices as a bridge to myocardial recovery. To this end, a closed loop model of the circulatory system was developed to study the effect of left ventricular assist devices on coronary perfusion. The effect of LVAD ejection delay, heart rate, and LV elastance on coronary blood flow were examined. The results suggest that maximized coronary blood flow and outflow may be achieved if the LVAD ejection delay is synchronized at a 0.3 to 0.5 second delay after cardiac contraction. The simulation results suggest that increasing heart rate reduces coronary blood flow. Finally, the relationship between cardiac elastance and coronary perfusion was examined and the simulation results suggest that coronary blood flow increases with increasing elastance, suggesting that a diseased myocardium may benefit more from an earlier LVAD implantation.

Moe, Beth “The effects of cell salvage on systemic interleukin-6 levels and postoperative blood loss: a meta-analysis study”

August 2009, 72pp, 57 references, figures, tables, nomenclature
  • Appendix A: List of Articles Collected via Literature Search
  • Appendix B: Category One Article Information
  • Appendix C: Spreadsheet Used to Organize and Calculate Data from IL-6 Articles
  • Appendix D: Spreadsheet Used to Organize and Calculate Data Pertaining to Postoperative Blood Loss Articles
Available for checkout
Abstract: The use of cardiotomy suction has long been a part of cardiac surgery. Unfortunately, research has shown that shed pericardial blood returned through pump suction to the cardiotomy reservoir is often filled with emboli, activated platelets, and increased levels of inflammatory cytokines. In an effort to counteract this, the use of cell salvage to process shed blood during cardiac surgery has increased. Research does support that cell salvage will wash away most of the emboli and activated platelets in the shed blood, Yet it is unclear if the use of cell salvage affects certain surgical outcomes such as postoperative bleeding and transfusion requirements.

In past studies, researchers have attempted to determine the benefits of cell salvage through animal studies, clinical trials, and in vitro experiments. Overall, the results remain mixed. Most of the studies are plagued with type II errors (false negative) and low powers, caused by small sample sizes. A meta-analysis was performed with the hopes of increasing the power and reducing type II error to provide more concrete answers.

For this meta-analysis study, two dependent variables were chosen: interleukin-6 levels and postoperative bleeding. These two variables had common sample times and sampling techniques among multiple articles. They also represent potential benefits to the patient. Interleukin-6 levels are used as a marker for the extent of the inflammatory response occurring during cardiopulmonary bypass. Postoperative bleeding is an important indicator of patient hemostasis. An inadequate hemostatic state can result in re-exploratory surgery and autologous blood product transfusions.

The null hypotheses were stated as follows: H0: There is no difference in systemic interleukin-6 levels between patients in which cell salvage is used to process the shed pericardial blood and when pump suction is used to process the shed pericardial blood. H0: There is no difference in postoperative bleeding between patients in which cell salvage is used to process the shed pericardial blood and when pump suction is used to process the shed pericardial blood.

A meta-analysis was created to compare all discovered research containing adult clinical trials, written in English, of cell salvage versus cardiotomy suction (control). It was determined that the use of cell salvage reduced the interleukin-6 levels after surgery. At the end of cardiopulmonary bypass interleukin-6 levels were found to have a mean effect size of 1.91± 10.85 pg/ml. However, the mean effect sizes were not found to be statistically significant. The results for postoperative blood loss showed that the mean effect size of the postoperative bleeding at 8 hours was found to be 110.21± 18.09 ml while at 24 hours it was found to be 24.04± 113.63 ml. Significance was found to occur at the eight hour postoperative measurement.


Mullins, Victor M., B.S. “Design and Application of a Pharmacological Control Module for a Perfusion Simulator.”

June 27, 1996, 86pp, 73 references
  • Appendix A: “Instructions for Authors” for The Journal of Extra-Corporeal Technology
  • Appendix B: List of Changes to Software Code
  • Appendix C: Software Code Added of Changed
Available for checkout
Abstract: The Perfusion Simulator being developed at the Milwaukee School of Engineering is comprised of computer software as well as hydraulic and electronic hardware. A cardiopulmonary bypass simulator can be useful to train a perfusion student without an element of risk to a patient. Cardiopulmonary bypass alters the pharmacology of drugs. This is caused by such factors as hypothermia, hemodilution, and contact with the cardiopulmonary bypass apparatus. Much of the changes in pharmacology of drugs while on bypass has not been quantitatively studied. There is a large degree of variability in drug pharmacokinetics and pharmacodynamics among patients during cardiopulmonary bypass as well as among patients in general. Mathematical models were developed to simulate the actions of three drugs which affect the systemic vascular resistance of a patient. The drugs modeled are isoflurane, nitroglycerin, and phenylephrine. For isoflurane, a pharmacodynamic model was used. Both nitroglycerin and phenylephrine were modeled with pharmacokinetic-pharmacodynamic models. After a dose of a specific drug is selected, the resulting effect occurs which in turn interacts with the main module of perfusion software to control a fluid valve which decreases or increases the resistance in the patient model.

Murad, Tarek, B.S. “Evaluation of the Safety of Chest Tube Drainage Reinfusion.”

June 1998, 45pp, 28 references
  • Appendix A: “Instructions for Authors” for The Journal of Extra-Corporeal Technology
  • Appendix B: Data Collection Protocols
  • Appendix C: Table for Pulmonary Shunt Fraction Data
  • Appendix D: Table for Red Blood Cell Data
  • Appendix E: Table for Pulmonary Vascular Resistence Data
  • Appendix F: Table for Oxygen Index Data
  • Appendix G: Table for White Blood Cells Data
  • Appendix H: Table for Arterial-Venous Difference Data
  • Appendix I: Raw Data
Available for checkout
Abstract: The reinfusion of patients’ chest tube drainage (CTD) is a technique advocated for decreasing the use of donor blood in cardiac surgery. Contraindications for this procedure include coagulopathy, systemic infection, contamination of shed blood, renal failure, and malignancy. Shed mediastinal blood is generally neither concentrated nor washed and is characterized by having high granulocytes, low hematocrit (20-25%), dysfunctional platelets, high concentrations of anaphylatoxin (C3a) and high fibrinolytic activity. This prospective study was designed to compare the respective effects of lipid/leukocyte filtered and unfiltered shed mediastinal blood transfusion on pulmonary shunt fraction and the inflammatory response. The collected blood was reinfused to the patient within four hours of initiating collection. Thirty patients undergoing elective cardiac surgery were randomly assigned to a lipid/leuko-reducing group or a control group. In the lipid/leuko-reduced group (n=15) all collected blood was filtered by a lipid globule, C3a, and leukocyte-reducing filters. In the control group (n=15) all collected blood was filtered by a 20 um microaggregate screen filter. The lipid/leuko-reduced group showed a significant decrease in pulmonary shunt fraction (p<0.05) following reinfusion, where the control group did not (p<0.05). In addition, the lipid/leuko-reduced group showed a trend towards decreased pulmonary vascular resistance, but this trend was not statistically significant. The level of red blood cells did not show a measurable increase after reinfusion of the CTD; the suspected reason for this was the low volume of CTD returned to the patients in this study. The study shows that the negative effects of lipids and leukocyte activation in the chest tube blood can be reduced using filtration. These results suggest that lipid/leuko-reducing chest tube blood improves pulmonary function, and is safe when reinfusing chest tube blood.

Pinter, Duane R. “Hardware Components for an In Vitro Perfusion Simulator.”

Oct. 20, 1995, 43pp, 26 references
  • Appendix A: Observed data for tests conducted on the Omega PX136 pressure transducers
  • Appendix B: Perfusion Simulator
  • Appendix C: Numerical values
  • Appendix D: Perfusion Simulator program which calculates the Reynolds number for varying flow rates and verying radius in the circulation loop
  • Appendix E: Transfer Functions
  • Appendix F: Unit Conversions
  • Appendix G: The Journal of Extra-Corporeal Technology Instructions for Authors
Available for checkout
Abstract: In order to enhance the quality of a perfusionists training, a perfusion simulator is being developed which consists of a PC based system using Visual BasicTM as the programming language, a positive displacement (roller) or vortex (centrifugal) pump, and a mock circulation loop. The perfusion simulator will mimic various physiological responses of the human body during cardiopulmonary bypass and also allows the person controlling the computer to input abnormal values (i.e. hypertension, hypotension, hyperkalemia, anemia, etc.). This paper will concern itself primarily with the hardware aspect of the simulator. The simulator circuit consists of a modified Donovan circulation loop for arterial and venous pressures and systemic resistances. Either a roller or centrifugal pump can be connected to the circulation loop. Gas flow rate and the O2 mole fraction (sweep and FiO2) is modeled by an electric circuit. The computer has the ability to control the flow, pressure, and resistance of the loop by opening and closing valves. As the characteristics of the circulation loop changes, adjustments made by the perfusionists are detected by the feedback components in the loop. These feedback signals are sent to the computer via a data acquisition board. The signals are analyzed and an algorithm determines the appropriate response to each signal. In order to elicit an appropriate response, data was collected from actual patients to construct a “model” from which the responses can be generated. After the response signal is generated, a control signal is sent to the various mechanical actuators in the circuit. Future improvements to the simulator will consist of adding a total artificial heart or ventricular assist device to the loop, gas transfer equations, an additional tank to simulate metabolism, and the introduction of pharmacological agents.

Posch, Benjamin, B.S. “Development of Simulation Curriculum for the Education of Cardiopulmonary Perfusionists”

May 2014, 49pp, 24 references, appendices
Available for checkout
Abstract: The goal of this project was to explain and develop a general curriculum that encompassed the main components of cardiovascular perfusion. The thought process behind the curriculum was to produce a general concept for a curriculum that entails the necessary aspects of perfusion, but leaves room for manipulation and additional detail. Ideally, the simulation curriculum would be modeled in such a fashion that it would suit the Master of Science in Perfusion program at the Milwaukee School of Engineering, so that the program could incorporate simulation into the program curriculum and begin teaching with their simulator.

The end result of the project was a general curriculum that the Milwaukee School of Engineering can utilize to begin incorporating simulation into their perfusion program. The curriculum was developed primarily through personal experience and knowledge, which was supported by literature. The benefits of simulation in a training curriculum can already be observed. As more perfusion programs continue to adopt simulation into their curriculum, further literature and collaboration of ideas will help to produce more skilled perfusionists. Without a sound curriculum, the potential of simulation training in perfusion will not be fully realized.


Powell, Jayson “An Assessment of the Air Handling Capabilities of Currently Available Adult Oxygenators That Contain Integrated Arterial Line Filters”

March 2012, 82pp, 60 references, appendices
Available for checkout
Abstract: There is a trend among manufacturers of perfusion equipment to produce oxygenators with integrated arterial line filters, such that perfusionists only need to add a single component to their circuit, instead of the traditional two. Any time new components are brought to market, perfusionists are interested in how they handle air emboli that may enter the circuit because of the risk that emboli pose to patients. The goal of this study was to determine the air handling capabilities and characteristics of the three currently available oxygenators with integrated arterial line filters. Each of the three companies producing these products takes a different approach to eliminating emboli from the circulation. The Terumo FX utilizes a fiber bundle with a filter membrane wrapped around it in a single compartment. The Maquet Quadrox-i uses a set of filter cassettes and a hydrophobic membrane to stop and remove particulate and gaseous emboli. The Sorin Synthesis utilizes a more traditional design with the oxygenator and the arterial line filters residing in separate compartments within the same assembly.The oxygenators were tested in a circuit built to isolate the oxygenator from other sources of emboli removal. Three channels of data collected volume of air and number of emboli at three locations around the circuit. Each oxygenator was challenged with 10 mL of air in each of 20 trials at two flow and pressure combinations. The trials were divided into groups of 10 and randomly assigned the order in which they would be performed.

The results of the study revealed that the Sorin Synthesis allowed the passage of the most emboli overall, while the Maquet Quadrox-i and the Terumo FX performed similarly. The Maquet Quadrox-i passed a significantly higher volume of air at the higher flow rate and pressure, but not a different number of emboli. Both the Terumo FX and the Sorin Synthesis passed the same volume of air but a decreased number of emboli.


Prossen, Erik F., B.S. “An Investigation into the Dampening of Pulsatile Energy in a Cardiopulmonary Bypass Circuit”

April 2012, 75pp, 54 references, appendices
Available for checkout
Abstract: Pulsatile flow may not be an entirely new development in the arena of cardiopulmonary bypass; however, pump technology has developed to the point that pulsatile flow during cardiopulmonary bypass is safe, cost-effective, and relatively efficient. A plethora of research has been conducted into the potential positive and negative effects of employing pulsatile flow during cardiopulmonary bypass. One area in which considerable research exists in the pediatric cardiopulmonary bypass arena involves the dampening of Energy Equivalent Pressure (EEP) in the bypass circuit. Little research has been conducted into the dampening of EEP in adult-sized bypass circuits.This paper examines the dampening of EEP in an adult-sized bypass circuit. Specifically, a comparison between the dampening of EEP through two commercially pulsatile pump settings on EP production. An experiment was developed to test the dampening of EEP by measuring flow and pressure at the pump outlet, oxygenator outlet, and arterial filter outlet. Two circuits were built for this experiment: one circuit used the Terumo RX25 reservoir with a separate Medtronic Affinity Arterial Line Filter. The other circuit used a Terumo FX25 oxygenator with integrated arterial filter. Independent variables included two levels of oxygenators or circuits (RX25 and ALF, FX25), three flow rates (3.4 L/min, 4.3L/min, and 4.9L/min) and four different pulsatile pump settings (base percentages varying between 100%-30%, and width percentages varying between 100%-60%). Flow was measured with a Transonic ME-PXL clamp-on flow sensor and data recorded with Windaq software. Pressure was measured with a BioBench analog-digital converter and BioBench software. The degree of EEP dampening across each circuit element, overall circuit dampening of EEP, and EEP production as a result of pulse settings were compared for analysis.

The results of this experiment show that the FX25 circuit dampened half as much EEP as the RX25 and arterial line filter circuit. Additionally, the amount of EEP going to the “patient” in the FX25 circuit was statistically significantly higher than in the RX25 circuit. The results of the pulse setting analysis show that EEP production increases significantly as base percentages and width percentages decrease from 100%. Additionally, a base setting of 30% and width setting of 60% significantly increased EEP production over all other pulse settings. It can therefore be concluded that to maximize EEP production in a cardiopulmonary bypass circuit, a perfusionist should employ a FX25 oxygenator using pulse pump settings of 30% base and 60% width.


Resnik, Josh “Persistent venous line air at the initiation of cardiopulmonary bypass”

March 2009, 54pp, 31 references, tables, figures, nomenclature, glossary, appendix
Available for checkout
Abstract: Gaseous microemboli (GME) are potential contributors to significant postoperative morbidity following surgery utilizing cardiopulmonary bypass (CPB). Air present in the venous lines of the extracorporeal circuit prior to the initiation of CPB is one of several possible sources of air entry and GME production. There is currently disagreement among perfusion and surgical communities as to the consequence of venous air prior to the initiation of CPB. This study aimed to investigate whether or not venous line air persists through the venous reservoir, leading to the passage of GME into arterial circulation.

In order to measure the ability of a venous reservoir to eliminate air and identify related factors, a circuit was designed to test the effects of air volume, venous reservoir volume, and speed of initiation on embolism passage into the arterial side of the circuit. A range between full liquid and full air was used in the venous lines to study air emboli generation, while increasing reservoir volume and prolonged time of initiation were tested to determine efficacy in reducing embolic load. Linear regression and three-way ANOVA were used to determine statistical significance of effects.

Air volume present in the venous line was found to be the most important factor influencing GME generation, showing a linear tread in emboli production with increasing volume. Prolonging time of initiation and increasing reservoir volume significantly reduced embolic counts in a linear fashion across the experimental range, but time had a more powerful effect that occurred quicker.


Richman, Adam “The Effects of Flow Rate on Mixed Venous Oxygen Content and Saturation in Veno-Venous Extracorporeal Life Support: A Comparison of a Computer Simulation and an in vitro Analysis”

May 2012, 72 pp, 66 references, appendices, figures, tables
Available for checkout
Abstract: Extracorporeal life support (ECLS) is the extended use of cardiopulmonary bypass (CPB) to support patients suffering from reversible cardiac and/or respiratory failure. Veno-venous ECLS (VV ECLS) is used to support patients suffering from only respiratory failure. Because the drainage and reinfusion cannulae for VV ECLS are both located in the venous circulation, a phenomenon known as recirculation occurs where oxygenated blood will be drained out again and recirculated through the ECLS circuit. This study compared a MATLAB simulation of recirculation with an experiment to characterize the relationship between ECLS drainage blood oxygen content and recirculation.A MATLAB graphical user interface (GUI) was created which accepts user-specified values for saturation (SO2) and partial pressure of oxygen (PO2) for oxygenated and deoxygenated blood, as well as hemoglobin concentration, recirculation fraction, and the pH of the drainage blood. This model then predicts, based on conservation equations, the oxygen content and saturation of ECLS drainage blood. Two in vitro CPB circuits were also constructed, one for oxygenation and one for deoxygenation. The blood was then mixed in different proportions to simulate different levels of recirculation, and the oxygen content and saturation were measured and compared to model predictions.

The results illustrate that a linear relationship exists between recirculation fraction and ECLS drainage blood oxygen content (p < 0.001), and that the MATLAB model is a very accurate predictor of ECLS drainage blood oxygen content and saturation, based on comparison to experimental data. It was also determined that hemoglobin concentration and pH of the drainage blood can have a dramatic effect on the ECLS drainage blood oxygen content and saturation, respectively.


Robertson, Michael, B.S. "The Effect of Magnesium on Anticoagulation for Cardiopulmonary Bypass"

May 2015, 30pp
Available for checkout

Abstract: During the use of cardiopulmonary bypass (CPB), the use of ultrafiltration is often necessary. Ultrafiltration is a process that filters the blood via a hemoconcentrator for a variety of reasons. During this process, there is a risk of blood components being removed from the blood which can become deficient if not accounted for. These components include drugs, electrolytes, and proteins. Schilling, in a thesis completed at the Milwaukee School of Engineering (MSOE), proposed that the removal of magnesium during ultrafiltration might interfere with blood clotting. The goal of this study was to further investigate whether magnesium concentration of highly filtered blood would affect coagulation.

Bovine blood was used to prime a CPB circuit that was used to filter the blood as a means of removing magnesium. As the blood was filtered, 0.9% saline was added back to maintain the overall volume. Baseline activated clotting times (ACTs) were run prior to the filtration at multiple heparin doses in an attempt to construct a heparin dose response curve. After the filtration, magnesium was then added back into the circuit in small amounts, and ACT values were recorded after each additional amount of magnesium was added.

The ACT after the filtration was nearly equal to the baseline ACT, suggesting that large volumes of ultrafiltration had no effect on the ACT. However, when magnesium was added back into the circuit, the ACTs then dropped. Unfortunately, due to the unexpected heparin response of the bovine blood, adequate anticoagulation was not achieved and the circuit clotted off before all data points were reached. After significant clot was formed, ACTs then rose likely due to the consumption of clotting factors in a given volume of blood. This study demonstrated that excessive ultrafiltration had little effect on the ACT relative to baseline. However, magnesium was found to facilitate hemostasis when added back into the circuit in physiological amounts.


Rogers, K. Todd “The Effects of Shear Stress on Neutrophil Activation”

May 1, 2004, 55pp, 50 references
  • Appendix A: Institutional Review and Patient Informed Consent
  • Appendix B: Raw Data
Available for checkout
Abstract: Neutrophils play a significant role in the mediation of the systemic inflammatory response as a result of cardiac surgery involving the use of cardiopulmonary bypass (CPB). Yet CPB itself has been shown to cause neutrophil activation. One of the reasons for the neutrophil activation may be the shear stress produced within an extracorporeal circuit. The use of vacuum assisted venous drainage (VAVD) can double shear stress levels during CPB. When VAVD is used in a CPB circuit, the greater shear stress produced could cause greater trauma to the neutrophils resulting in increased levels of myeloperoxidase (MPO), a marker of neutrophil activation. The goal of the current study was to evaluate the effects of clinically significant shear stress levels that occur with the use of VAVD on the levels of MPO, a marker of neutrophil activation. This study consisted of 12 experimental trials in which human blood from three different donors was perfused through circuits under four different experimental conditions. The four treatment conditions were high hematocrit/high flow (HH), high hematocrit/low flow (HL), low hematocrit/high flow (LH), and low hematocrit/low flow (LL)). Each trial was run for 90 minutes. MPO levels were assessed initially and at the 45 minute, and 90 minute intervals. In all the trials, there was an increase n MPO level compared to the initial reading within the first 45 minutes. Time was the only independent variable found to have any statistical significant effect on the MPO level. Wall shear stress and wall shear rate did not correlate with MPO levels in this investigation. This is the first study that has attempted to examine neutrophil activation as measured by MPO levels at the clinically relevant shear stress levels observed during CPB with VAVD. Based on the results of this investigation, it is recommended that other alternatives be investigated that may play a role in the activation of neutrophils, as assessed by MPO levels, other than shear stress and shear rate.

Schaller, Suzanne Lyn, B.S. “Comparison of Cardioplegia Methods in Patients Undergoing Cardiopulmonary Bypass Surgery”

Oct. 21, 1999, 96pp, 47 references
  • Appendix A: Instructions For Authors, The Journal of Extra- Corporeal Technology
  • Appendix B: Large-Sample Significance Test
  • Appendix C: Institutional Review and Patient Informed Consent Documents
  • Appendix D: Raw Data Tables
Available for checkout
Abstract: Two cardioplegia methods currently utilized at St. Luke’s Medical Center, Milwaukee, Wisconsin are the BCD Vanguardä to deliver Buckberg solutions (Buckberg) and the Quest Myocardial Protection Systemä to deliver potassium chloride and magnesium sulfate (Quest). While the Buckberg is widely used and its performance well documented, concerns have arisen regarding high serum potassium levels- perhaps due to its design limitations- and the cost of its customized crystalloid solutions. In addition to being less expensive to operate, the recently introduced Quest offers more flexible and precise potassium delivery, but little published data is available on its impact on serum potassium levels and myocardial protection. The study was undertaken to compare the effect of the Buckberg and Quest on serum potassium levels, myocardial protection, hemodilution, and cost. The study compared potassium levels at the time the aortic cross-clamp was removed, potassium balance at termination of cardiopulmonary bypass by subtracting urinary excretion and hemofiltrate from the sum of contributions from cardioplegia, blood transfusion, fluid administration, and hemolysis. Myocardial protection was analyzed by comparing cardiac index; cardiac output; time to arrest; activity between cardioplegia doses; number of patients who required cardiac pacing, defibrillation, inotropes, and intra-aortic balloon pump; and intensive care unit and hospital stay. The effect on hemodilution was compared by measuring hematocrit levels at the time the aortic cross-clamp was removed and administration and excretion of fluid during CPB. A cost evaluation was performed on the cardioplegia disposable delivery sets, equipment, and the solutions utilized with each method. Statistical analysis was performed using MicrosoftÒ Excel 97 (Microsoft Corporation, USA) and SystatÒ 6.0 for WindowsÒ Student Version (SPSS Inc., Chicago, IL) (significant differences were defined as probabilities for each test of p < 0.05). Two of the 40 patients were excluded from the electrolyte statistical analysis due to specimen containing hemolyzed blood from incorrect collection. The two-sample student t-test (or independent t-test) was used on continuous data to compare means from the Buckberg and Quest groups. A binomial large-sample significance test was used to compare count-data from the Buckberg and Quest groups. One-way analysis-of-variance (ANOVA) and Bonferroni pairwise procedures were used to compare means of continuous data from three groups based on serum potassium levels. The study found no significant difference between the Buckberg and Quest in regards to their effect on serum potassium levels or myocardial protection. The two methods did not differ significantly on hematocrit values, but the Quest patients did receive significantly more packed red blood cells than the Buckberg group. It was found that the Quest is less expensive with a total cost saving of $247.50 per case. It was concluded that both the Quest and Buckberg provide adequate myocardial protection and that the occurrence of high serum potassium levels appears to be due to potassium homeostasis rather than the amount of potassium delivered by either method. All three groups (normal, mild, and moderate to severe hyperkalemia group) in the ANOVA analysis had approximately the same potassium intake from cardioplegia and other sources, however, the preoperative creatinine was higher in the moderate to severe hyperkalemia group and the normal group had a significantly higher urine output. This data suggest the importance of renal function in regards to potassium homeostasis rather than the actual amount administered.

Scherkenbach, Peter S., B.S. “The Use of Continuous Cerebral Oximetry as a Diagnostic Tool to Reduce the Incidence of Stroke in a High Risk Population.”

July 1999, 83pp, 91 references
  • Appendix A: Instructions for Authors for The Journal of Extra-Corporeal Technology
  • Appendix B: Saint Luke’s Medical Center Database Information
  • Appendix C: Institutional Review and Patient Informed Consent Documents
  • Appendix D: Physiologic Correlation Data
Available for checkout
Abstract: The occurrence of neurologic deficits of patients undergoing cardiac procedures is an area of much discussion. Since neurologic dysfunction remains one of the major contributors to patient morbidity following cardiac procedures, there is a necessity to alleviate the brain injury incurred. This study focused on reducing the stroke rate in the cardiac surgery patient requiring cardiopulmonary bypass (CPB). The hypothesis tested was aimed at evaluating whether the stroke rate at St. Luke’s Medical Center (Milwaukee, Wisconsin) could be reduced using the Somaneticsâ INVOSâ 4100 Cerebral Oximeter (Troy, MI) to evaluate intervals of cerebral ischemia and subsequently modify the perfusion and operative technique to adjust for the malady. The study was also intended to evaluate whether hospital and intensive care unit (ICU) stay was lengthened by the occurrence of neurologic dysfunction. Secondary explorations into collected cerebral oximeter data were investigated to ascertain possible correlation between the cerebral saturation waveforms and the patients’ physiologic parameters. This was done to identify possible trends that were useful in predicting cerebral saturation and to evaluate the success of the perfusion interventions. Risk scores and risk factors were obtained from a cardiac patient database at St. Luke’s Medical Center using a stepwise multiple logistic regression. Patients were included into the study based upon a risk score that was used to determine whether to place those patients into a category termed as high risk for stroke. Seven independent risk factors were discovered and represented by odds ratios. The odds ratios demonstrated that a patient who exhibited that particular risk was more likely to have a stroke by a number of times greater than the average. Patients who were included into the study exhibited risk factors that set the sum of the odds ratio greater than or equal to 9.0. This value was selected because the sum of the odds ratios above 9.0 placed those patients in a category where the risk of stroke was approximately 11.4%. The patients were all monitored using the Somanetics INVOSâ 4100 cerebral oximeter (Troy, MI) and subsequently managed with the implemented perfusion protocol. A patient population from June 1996 to July 1998 served as a historic control. The procedures included in this study were limited to coronary artery bypass grafts and valvular procedures. Redo operations were not excluded, however, concurrent carotid endarterectomy and aortic aneurysm repairs were excluded from this study. Thirty subjects were enrolled in the study. Two patients were excluded from the statistical analysis: one was for a carotid endarterectomy added late into the scheduling, and the second was for a concomitant ascending aortic aneurysm repair on an aortic valve replacement. The 28 patients that remained in the study for postoperative evaluation exhibited no signs of a cerebral vascular accident. This data when compared to the historic control showed moderate statistical significance (x2=3.555,p=0.059). Six patients that had been noted for exhibiting signs of transient neurologic dysfunction (e.g. cognitive deficits, confusion, or dementia), showed a statistically significant increases in hospital stay (p=0.042) and ICU stay (p=0.012). The results of this clinical evaluation show some evidence that neurologic injury can be reduced using the Somanetics INVOSâ 4100 Cerebral Oximeter as a neuromonitoring technique. Although the tested hypothesis was not statistically significant at the p=0.05 level, the trend may indicate a reduction in stroke rate. Valuable information was gathered correlating the patient’s physiologic parameters and cerebral saturation as indicated by the Somanetics INVOSâ 4100 cerebral oximeter. This information could be used to validate the diagnostic device and give credence to the interventions that were implemented to alleviate the ischemic intervals. Further research is necessary to support these results.

Schilling, Kyle, B.A.S. “Analysis of the Effect of Zero-Balance Ultrafiltration on Circulating Levels of Unfractionated Heparin”

May 2014, 43pp, 19 references
Available for checkout
Abstract:The process of cardiopulmonary bypass (CPB) cannot be performed without proper interruption of hemostasis. Unfractionated heparin anti-coagulates the blood by preventing activation of thrombin, and helps to prevent clot formation that can be detrimental to the patient. Therefore, proper circulating heparin levels must be maintained by the perfusionist throughout CPB. Various types of ultrafiltration are commonly used as an adjunct to CPB, and each type can potentially alter circulating heparin levels. No previous studies have analyzed the effects of zero-balance ultrafiltration (ZBUF) on heparin concentration. The goal of this experiment was to establish whether significant amounts of heparin could be removed through ZBUF, and whether patient anti-coagulation status would be affected.

A pediatric circuit was constructed to minimize prime volume and resources needed for the experiment. Each circuit was primed with 300 mL of packed red blood cells (PRBC) and 300 mL of expired, compatible fresh frozen plasma (FFP). Two separate trials were conducted analyzing changes in heparin concentration as compared with the volume of ZBUF utilized, and varying transmembrane pressures (TMP). Trial 1 consisted of 9000 mL of ZBUF with three predetermined TMP. Trial 2 consisted of 7500 mL of ZBUF while TMP was held constant. Heparin concentrations were measured with a Hepcon HMS PLUS system while utilizing gold heparin-protamine titration assay cartridges. Regression analyses were performed on data from both trials.

Results showed that no significant amounts of heparin were removed from the circulation during ZBUF. In Trial 1, volume of ZBUF proved to be a significant predictor of heparin concentration (p=0.001), which decreased from 4.0 mg/kg to 3.5 mg/kg. In Trial 2, volume of ZBUF was not found to be a significant predictor (p>0.05) as there was no decrease in heparin concentration. Transmembrane pressure was not a significant predictor (p>0.05) of heparin concentration in either trial. Although minute filtration of heparin may have occurred, patient anticoagulation status was not significantly altered in either trial. Excessive amounts of ZBUF with normal saline were found to significantly prolong the clotting time of the blood. Depleting significant amounts of magnesium through ultrafiltration was thought to be the cause of this clotting prolongation.

This study demonstrated that ZBUF does not result in significant heparin removal, and therefore, patient anti-coagulation status remains clinically stable. However, it has been shown that excessive removal of the electrolyte magnesium has potential to cause hemostatic issues if not addressed.


Schmidt, Jennifer L., B.S. “The Determination of the Effects of Intra-Operative Hemoconcentration on Renal Function.”

July 18, 2001, 55pp, 54 references
  • Appendix A: Instructions for Authors for The Journal of Extra-Corporeal Technology
  • Appendix B: Institutional Review and Patient Informed Consent Documents
  • Appendix C: Raw Data Tables
Available for checkout
Abstract: Background. Hemoconcentration is a technique that has been used during cardiac surgery in conjunction with cardiopulmonary bypass (CPB) since the mid-1970’s. It is commonly used as a means of reducing the hemodilution that occurs as a result of CPB priming solutions and other intra-operative drug and fluid administration. While there are few stated risks or contraindications associated with its use, there has been recent concern that hemoconcentration may contribute to peri- and post-operative renal dysfunction. Purpose. This study was undertaken to determine whether or not hemoconcentration is a risk factor for the development of renal dysfunction. Methods. A prospective study on 40 patients undergoing cardiac surgery with the use of CPB was performed. Patients were assigned to either the hemoconcentration or the control group based on their individual intra-operative needs. CPB and anesthetic technique were standardized for all patients. Renal function was determined by intra-operative analysis of serum creatinine concentrations and urinary creatinine clearances; and post-operative analysis of serum creatinine and BUN concentrations. The incidences of acute renal insufficiency (ARI), characterized by serum creatinine concentrations of 1.6-2.4 mg/dL, and acute renal failure (ARF) characterized by serum creatinine concentrations >2.5 mg/dL were recorded. The endpoints for data collection were death, the need for dialysis, and length of hospital stay. Results. Patients in the hemoconcentration group experienced significantly elevated serum creatinine concentrations at all time intervals post-operatively, as well as significantly elevated BUN concentrations on post-operative day five in comparison to the control group. This finding was, however, complicated by the fact that the two groups differed significantly with regard to serum creatinine prior to treatment. However, the hemoconcentration group did experience a greater percentage increase in both serum creatinine and BUN concentrations post-operatively and was the only group to experience serum creatinine concentrations outside of the normal range. The hemoconcentration group also experienced a significantly more rapid increase in serum creatinine concentrations post-operatively. Furthermore, the hemoconcentration group had a higher incidence of both ARI and ARF post-operatively, with two of the patients requiring dialysis. Renal dysfunction was more common in patients with pre-operative renal dysfunction and those undergoing a surgery involving a redo sternotomy. Conclusions. The data confirms that the hemoconcentration group had a greater tendency to develop renal dysfunction post-operatively. However, this study cannot conclusively confirm that hemoconcentration was the sole contributor to the renal dysfunction seen in the treatment group because the two groups differed pre-operatively with regard to serum creatinine, thereby calling into question their renal condition prior to treatment. Further investigation with a larger patient sample size is warranted for conclusive determination of the effects of hemoconcentration on renal function.

Schrand, Kelsey, B.S. "The Personality of Perfusion: Determining the Big Five Personality Traits as a Student Success Predictor"

February 2016, 52pp
Available for checkout

Abstract: Mannerisms, lifestyles, sociability and values are unique to each individual, and it has been established that this combination of traits can influence one’s occupational choice as well as the level of satisfaction they obtain from that occupation. Perfusion is a highly specialized field where practitioners function within the cardiac operating room to maintain the life of the patient during surgery. Incoming students often have little exposure to the setting, which makes predicting their success in the profession a difficult task. This project seeks to identify personality characteristics of successful perfusionists in an effort to better review applicants for entry into perfusion programs. If successful, this could reduce the number of students that leave perfusion education programs by not admitting those with a personality trait profile that predicts non-completion.

This pilot study surveyed a group of 20 perfusionists, 9 perfusion students, and 2 former perfusion students. Using the International Personality Item Pool (IPIP) public domain, a Google form was created. There were 120 questions, 24 questions regarding each of the “big five” traits: neuroticism, extroversion, openness to experience, agreeableness, and conscientiousness.

Results of this pilot study found that there were no significant personality trait distinctions between the former students and the rest of those surveyed. As a group, perfusionists scored lowest on neuroticism and highest on conscientiousness, both significantly different from the normative population mean. Results indicate that perfusionists as a group share personality traits, which were also shared by the two former students. Predicting success in perfusion school may not be possible with a personality test; however, a shared perfusion profile is potentially present.


Simmons, Mark. “An Evaluation of Performance Factor Testing for Extra-Corporeal Heat Exchangers.”

April 4, 1995, 14pp, 15 references
  • Appendix A: Diagrams of Laboratory Circuits
  • Appendix B: Laboratory Results
Available for checkout
Abstract: The standard method of evaluating heat exchangers incorporated into an oxygenator is using an equation that determines a value called performance factor. The standard testing is performed under specific conditions and all temperatures are at steady state. The belief is that the higher the performance factor, the better the heat exchange, and hence, the better the heat exchanger. Laboratory testing was done to examine the validity of the performance factor testing. A patient model circuit was used to evaluate the heat exchangers during simulated cardiopulmonary bypass. The patient model was cooled to 18°C and then rewarmed to 35°C. The results were compared to the heat exchangers performance factor which was measured using the standard methods. Data obtained demonstrated that the performance factor was not an accurate representation of the performance of the heat exchanger in the dynamic environment, which exists during cardiopulmonary bypass.

Staffaroni, Jenna "Clinical Considerations Associated with Ventricular Assist Device Patients Receiving Hyperbaric Oxygen Therapy"

March 2016, 52pp
Available for checkout

Abstract: Hyperbaric Oxygen Therapy (HBOT) is a widely accepted treatment for patients suffering from conditions associated with low oxygen content or delivery, and utilizes a pressurized environment and high fraction of inspired oxygen to drive oxygen deep within the body’s tissues to facilitate healing. Although a variety of patients may benefit from this therapy, many medical devices have not been tested in the hyperbaric environment, including ventricular assist devices. Such testing needs to consider both the environmental changes as well as hemodynamic changes that have been shown to occur with HBOT. These hemodynamic changes include increases in systemic vascular resistance that occur during the descent to therapeutic depth and the return to surface that occur with HBOT. The goal of this study was to identify and assess the clinical concerns that accompany a VAD patient receiving HBOT.

The main goal of this project was to determine if two common VADS (HeartWare and HeartMate II) would maintain their flow rates within 1 liter of baseline when faced with increasing afterloads. These afterloads mimicked the range of increases reported in the HBOT literature. This specification was assessed by incrementally increasing the afterload pressure for each VAD and measuring the corresponding flow rates.

Results showed that a SVR increase of 150% was needed in order to decrease the Thoratec Heart Mate II flow by roughly 1 L/min, while the HeartWare required an increase in SVR by roughly 180% for the RPM levels tested. Each of these increases is greater than the maximum increase in SVR of 130% reported in the literature. These data support that the increase in SVR that occurs during HBOT should not pose a significant threat to the patient’s circulatory support. Although this study concluded that the VADS can meet the clinical needs of patients undergoing HBOT, mechanical operation of the VAD at high pressures must still be verified prior to implementing HBOT with a VAD patient.


Starkey, Nick, B.A. "Cardiac Surgery Mission Trips: Improving the Perception and Preparation of Perfusionists"

May 2015, 46pp
Available for checkout

Abstract: The prevalence of cardiovascular diseases (CVDs) is increasing at an astonishing rate. It is the number one cause of death in the world and every facet of society is feeling its repercussions. Particularly affected are low- and middle-income (LIMI) countries that are struggling to develop. But most importantly is the effect it has on the poor in these developing countries, where families struggle to survive as is.

A number of initiatives have been introduced to address the prevalence of CVDs, but these initiatives are aimed solely at reducing risk factors and not increasing access to health care. Yet cardiac surgery is often the only option for individuals who have diagnosed CVD. It is the aim of non-profit organizations to address this demand and provide these individuals with the life saving surgery they need. To provide cardiac surgery in LIMI countries, these organizations establish relationships with local hospitals. When a relationship has been developed between the non-profit organization and the local hospital, the goal for organizations is to provide cardiac surgery to individuals that would have otherwise not been able to have such an operation. Perfusionists are critical members of the cardiac surgery team because they operate the heart-lung machine during such surgeries. But most perfusionists who volunteer to work on these teams are ill prepared for the trip, which can impact their ability to perform their job well. The goal of this project was to provide a description (based on personal experience) of cardiac surgery mission trips and to generate a series of protocols and guidelines for perfusionists planning on participating in such trips.

After attending the 2014 CardioStart International mission trip to India, the author recognized that guidelines and protocols unique to mission trips should be developed and utilized. Using information acquired from perfusionists with mission trip experience, guidelines were created guide a perfusionist through all aspects of a mission trip. In addition, checklists and protocols are provided that will aid the mission trip team in preparing for their trip and ensuring they are as productive as possible in their unique situation.


Starrett, Chelsea, B.S. “Analysis of the Efficacy of Leukocyte Depleting Filters During Cardiopulmonary Bypass”

May 2011, 100pp, 149 references, appendices, figures, tables
Available for checkout
Abstract: The systemic inflammatory response (SIR) is activated through the usage of cardiopulmonary bypass (CPB) during cardiovascular surgery. This response is mediated by leukocyte activation, mainly neutrophils. Recently, leukocyte depletion filters (LDFs) have been designed to be incorporated into the bypass circuit to reduce the effect of inflammatory activation. Although some studies have shown this method to be effective, there are several other studies that do not show any benefit. The aim of this paper is to evaluate the existing literature in order to determine the efficacy of LDF utilization during CPB surgery.This paper reviews available information on LDF usage. This includes the placement of filters in the arterial, venous, and/or cardioplegia systems, as well as the timing of such usages. The focus of this literature review was on the modulatory effects of LDFs in regards to markers of systemic inflammation and its influence on patient outcomes. The hypothesis tested in this study was that the use of LDFs correlates with politive clinical outcomes by attenuating the SIR.

A careful review of the existing research, evidence shows that the markers of inflammation decrease and clinical outcomes improve. This review found a majority of studies showed that LDFs improve patient outcomes and decrease the markers of SIRS especially when LDFs are used strategically in the arterial line. Studies that demonstrated no improvements simply showed no effect versus negative outcomes. However, published studies are inconsistent in filter usage/timing and tracked clinical outcomes. Therefore, these factors add complixity to evaluate.This situation is compounded by the individual pathologies of the patient being treated, surgical techniques, and perfusion management. Therefore, maximal effectiveness must be determined by the protocols of the surgical team, the amount and length of depletion necessary to have effect, and the placement of the filter in the CPB circuit. Therefore, filtration may need to be individualized to the institution and the surgical team.


Sun, Han-Sen. “The Validity of Utilizing Swine Blood in Cardiopulmonary Models: A Literature Review”

July 2005, 89pp, 171 references
Available for checkout
Abstract: The swine is one of the models that has been used for biomedical research. It is used in place of humans to test physiological responses to challenges, including the responses of blood to the challenges of cardiopulmonary bypass. Whether these studies accurately reflect the human condition is dependent upon how closely swine blood resembles and reacts to challenges in the same way that human blood would. It is important to understand the similarities and differences between swine and human blood when interpreting studies performed using the swine as a model. The goal of this thesis is to investigate the similarities and differences between human and swine blood. Through a literature review, this study focuses on comparing blood between the two species in the areas of: red blood cells, coagulation/fibrinolytic systems, and immune response. This is done as an aid for interpreting CPB studies that utilize swine blood. Although smaller, the porcine red blood cell displays similar behaviors to human red blood cells. Erythrocytes from both species have biconcave shapes and form roleaux complexes. The porcine red blood cell is more prone to hemolysis from osmotic pressure than the human red blood cell. Isotonicity in NaCl is 0.85% in the porcine RBC compared to 0.90% in humans. The components of the coagulation/fibrinolytic pathways are highly conserved between humans and swine. Yet the swine has a more active intrinsic clotting pathway and less responsive extrinsic clotting pathway. Platelet behaviors are comparable under various conditions including cardiopulmonary bypass. Coagulation disorders in humans can be modeled in the swine. The similarities in the coagulation factor VIII permits clinical use in treating hemophilic patients. The immune response in swine is comparable to humans. The functions and expression of different immune system components are alike between the two species. Transfusion reactions in the swine are witnessed when non cross-matched blood is given to the recipient. This phenomenon can result from red blood cell antigen incompatibility from one of the sixteen porcine red blood cell blood groups. Two distinctions in the swine immune systems are the lack of native antibodies necessary for an immune response to blood type incompatibility and the inability of swine complement complex to attach to homologous red blood cells. These findings indicate that the possibilities of erythrocyte agglutination and lysis when pooling of swine blood is used may be neglected.

Terreau, Brian, B.S. “Making and Utilizing Instructional Videos to Enhance the Learning of Perfusion Students”

February 2015, 36pp, 22 references
Available for checkout
Abstract: New perfusion students commonly learn complex tasks by watching a certified clinical perfusionist preforming a live demonstration. While this method works, it leaves something to be desired. Students can often be overwhelmed by the vast amount of information, and instructors can be frustrated with multiple explanations of the same material. Fortunately, technological advancements permit the use of more modern teaching techniques. One of these techniques is to augment the learning of complex tasks through the use of instructional video.

Instructional video has been around since at least the 1950’s, and may be a useful tool in teaching new perfusion students. Because video can be edited, it can provide a near perfect demonstration for students to learn from. In addition, students can refer back to it whenever a question arises. For these reasons, the goal of this project was to create a series of instructional videos for new perfusion students. The videos were recorded, edited, and posted to a popular online video site for easy access and distribution to students and perfusionists. A survey was created, and feedback about the quality and usefulness of the videos was generated.

The survey data and responses from perfusionists and perfusion students showed that the instructional videos would be very useful learning tools for new perfusion students. The use of these videos should result in a faster and less frustrating learning process for both students and their instructors. Furthermore, existing perfusion students can improve upon the current set of videos, benefitting themselves and future students.


Thompson, Bradley Allen “The Effects of Temperature Extremes on Hemolysis in an Extracorporeal Circuit.”

April 24, 1996
  • Appendix A:
    • Figure 1. In vitro circuit with pump inlet and outlet pressure monitoring. All circuits contain a 12 inch segment of 1/8 inch internal diameter tubing.
    • Figure 2. The temperature of the respective circuits over time.
    • Figure 3. Plasma hemoglobin concentration over time.
  • Appendix B:
    • Table 1. Hemodynamics of the respective circuits.
    • Table 2. Mean plasma hemoglobin concentration (mg/dl).
  • Appendix C. Instructions for Authors, Journal of Extra-Corporeal Technology (JECT).
Available for checkout
Abstract: Hemolysis is a serious problem that must be improved upon in order to advance cardiopulmonary bypass (CPB) technology. This study was conducted to determine whether temperature extremes could increase or decrease the amount of hemolysis observed in extracorporeal circulation. The temperature extremes were designed to simulate situations that may occur in CPB. The blood of thirteen New Zealand White rabbits was subjected to either profound hypothermia, normothermia, or hyperthermia. The hyperthermia group underwent moderate hypothermia with hyperthermic rewarming. The hypothermia group underwent profound hypothermia and was gradually rewarmed to normothermia at the end of the experiment. The normothermia group remained at normal rabbit core temperature for the entire experiment. Hemolysis was measured by analysis of plasma hemoglobin concentrations over eight hours of circulation. The degree of hemolysis was significantly less in the profound hypothermia group compared to the normothermia and hyperthermia groups. Hemolysis in the hyperthermic group was not statistically different from the normothermic group. Variation of hemolysis between individual rabbits in the hyperthermic group was greater than either the profound hypothermia or normothermia groups. This data supports the concept that profound hypothermia (<20°C) protects against RBC trauma whereas hyperthermic rewarming increases the potential for hemolysis in extracorporeal circulation.

Tisdell, John H., B.S. “Analysis and Evaluation of the MSOE Perfusion Simulator.”

April 15, 1997, 54pp, 16 references
  • Appendix A: “Instructions for Authors,” ASAIO Journal.
  • Appendix B: Patient #2 Perfusion Log Data and Simulation Results.
  • Appendix C: Linearity Data and Linear Regression Results
  • Appendix D: Patient #15 Data and Simulation Results with a Venous Reservoir.
Available for checkout
Abstract: The Milwaukee School of Engineering (MSOE) has developed a perfusion training simulator to help students become familiar with the hemodynamics of cardiopulmonary bypass (CPB). The purpose of this master’s essay is to analyze and evaluate the present condition of the perfusion simulator to determine how close the simulations are to real CPB and what direction the project should take. The most important CPB experience was determined to be the initiation/termination of bypass, which was made possible in the original design with the addition of a venous reservoir in the system. The MSOE mock circulatory loop attempts to simulate both the resistive and compliant properties of the vascular system. In reality, the loop predominantly displays only a resistive element. The compliance remains constant throughout each simulation. The systemic vascular resistance (SVR) of the mock circulatory loop is an independent variable controlled by the software from the SVR of patient data logs. The simulated SVR was found to be greater than the data log SVR which causes the user to pump at low flow rates. The proper operation of the loop’s flow meter and pressure values were verified. Initiation/termination of CPB was found to be possible when a venous reservoir was added to the present MSOE perfusion simulator. The project’s next recommended step is to correct the excessive resistance found in the loop.

Tittl, Matthew P., B.S. “A Comparative Study of Adult Arterial Line Filters”

Feb. 5, 2002, 65pp, references, appendices
Available for checkout
Abstract: This experiment is titled “A Comparative Study of Adult Arterial Line Filters”. The arterial line filter is considered one of the most important devices in the cardiopulmonary bypass (CPB) circuit since it represents the final filtering mechanism for the bloodstream before it returns to the patient. Considering both the significance and the limited amount of literature or standards available concerning this subject, a definite need existed for an independent, unbiased, and updated comparison of arterial line filter performance. This project examined the performance characteristics of some prominent adult arterial line filters currently available on the market. Areas of evaluation included particle filtration efficiency, pressure drop (resistance), air filtration efficiency, platelet depletion, plasma-free hemoglobin generation (hemolysis), and priming ease. These are the most critical factors for positive patient outcomes when CPB is employed. Presently, the primary adult arterial line filter manufacturers include C.R. Bard, Baxter (Bentley), Cobe, Gish, Jostra, Medtronic (Acevor), Pall, Sorin (Dideco), and Terumo. To request comparable product samples and explain the objectives and testing protocol of the study, representatives from each of these corporations were contacted. Eight samples of eleven filter brands and six samples of two filter brands were then obtained. Miscellaneous tubing, connectors, and supplies were also acquired. The testing protocol was divided into three distinct phases. Phase I investigated particle filtration using a normal saline perfusate. Phase II studied air filtration efficiency using a normal saline perfusate. Finally, Phase III examined air filtration efficiency, platelet depletion, and plasma-free hemoglobin generation using a fresh bovine blood perfusate. In all phases, priming ease and pressure drop characteristics were recorded. To accurately determine particulate filtration efficiency, the Beckman-Coulter Z2 particle counting device (Beckman-Coulter Incorporated, Hialeah, Florida) was utilized. Flows and pressures were maintained at constant levels. A mixture of polymer spheres was used to represent foreign particulate in the CPB circuit. These contaminants were injected into the circuit prior to each filter, and the resulting number of spheres remaining downstream were measured and recorded. Pressure drop was measured by attaching a DLP-60000 digital pressure monitor (DLP Incorporated, Grand Rapids, Michigan) near the inlet and outlet of each filter. Flow was then varied from 1 LPM to 6 LPM, with the purge line opened for one measurement set and closed for the other. The difference between the two pressure monitor readings represented the pressure drop. Air removal efficiency was measured with the Hatteland Instrumentering CMD-20 microbubble detection device (Hatteland Instrumentering, Reyken, Norway). Flows and pressures were maintained at constant levels. Small and large air boluses were injected into the circuit prior to each filter, with the purge line open for one measurement set and closed for the other. The number of all downstream bubbles, with their respective sizes, was then measured and recorded. Platelet depletion and plasma-free hemoglobin generation were measured over a two-hour CPB duration by drawing periodic blood samples from the CPB circuit. Flows, pressures, and hematocrits were maintained at constant levels. Platlets were isolated from other blood cells using a prepared chemical solution and counted manually under a microscope. Plasma-free hemoglobin levels were found using a specific chemical assay available from SIGMA Diagnostics (SIGMA-Aldrich Corporation, St. Louis, Missouri). Priming ease was measured in all three phases by observing the time taken from beginning of filter priming to the total absence of air bubbles downstream, as verified by the Hatteland Instrumentering CMD-20 microbubble detection device. An identical flow rate and priming procedure was utilized for each filter. Average priming times were then calculated from the results of the three phases. Whenever appropriate, control trials were performed to determine what influence the arterial line filter actually had on the performance variables tested. In addition, some controls were necessary to validate the Phase III testing protocol. Using these controls, various measurements of arterial line filter performance were adjusted accordingly. The arterial line filters removed particles greater than their respective pore sizes (20 Sm to 43m) in an 86.6% to 100% range. At a maximum flow rate, arterial line filter resistances (which were calculated using pressure drop data) ranged betwen 5.7 PRU and 8.1 PRU using a normal saline perfusate and 5.7 PRU and 8.2 PRU using a bovine blood perfusate. The arterial line filters passed air bubbles greater than their respective pore sizes in a 0 to 5 count range. After 120 minutes of operation with the bovine blood perfusate, platelets were depleted in a 5.20% to 17.85% range with respect to the initial counts, and a plasma-free hemoglobin (a blood cell hemolysis indicator) was generated in an 85.0% ro 177.2% range with respect to the initial levels. Finally the arterial line filters were completely primed and debubbled in a 57 sec to 120 sec range. Control values are not included in these results, but they are included with the adjusted performance values in the report.R Statistical analyses were performed using the one-way and two-way analysis of variance, pairwise comparisons, and correlations. The tables and graphs necessary for effectively presenting the information were also developed. Finally, three overall rating systems were devised to serve as ultimate comparisons of these arterial line filters. Significant differences in performance, for both better and worse, were identified among arterial line filters in most of the evaluation areas (p <0.05). Significances were also observed in specific performance variables, regardless of arterial line filter brand, as they were adjusted according to protocol (p <0.05). Finally, excluding platelet depletion and plasma-free hemoglobin generation, significantly different results were evident when comparing control circuits to those including arterial line filters (p <0.05). This supports the initial premise that the use of arterial line filters in CPB circuits provides satisfactory protection against particulate contaminants and air emboli. Altogether, 532 statistically significant correlations were identified and quantified (p <0.05). However, only twenty-three of these correlations were considered applicable using a +/-0.7 correlation coefficient (r) standard. The number of significant relationships was lower than anticiplated, and those identified oftentimes did not occur in the expected areas of performance. This may be the result of the particular data recorded during this experiment, which was subject to fluctuations considering the possible circumstances for relativity. Therefore, the relationships identified and unidentified in this experiment might not be indicative of the arterial line filter population as a whole. Each filter was ranked in six categories according to the results, and three overall performance indices were developed. A variety of best performers was observed, which suggested an effort by individual manufacturers to produce arterial line filters excelling in specific performance areas. While designing and fabricating an arterial filter to perform the best in all of these categories may be an impossible assignment, the Baxter-Bentley AF1025D Duraflo II appeared to achieve the superior balance throughout the six primary performance areas. Emphasized throughout the experiment, the arterial line filter is regarded as the most important safety device of the CPB circuit and the last line of defense for removing dangerous microemboli from the blood before it returns to the patient. Considering this significance, frequent and through arterial line filter assessments are essential now and in the future. The ultimate values of this project, therefore, are ease of repeatability and flexibility for future evaluation of new and additional arterial line filters, including pediatric varieties. A continual pursuit toward filtration perfection will enhance CPB outcomes and ultimately improve the entire profession of cardiac surgery.

Tomich, Angela N., B.A. “In Vitro Quantitative Analysis of Air Production With Varying Base Flows.”

April 27, 1998, 88pp, 29 references, appendices
Available for checkout
Abstract: The main idea of this project was to determine how the production of gaseous microemboli is affected by base flow, and how that relationship is affected by other variables such as hematocrit and membrane oxygenator type. Microemboli can be defined as extremely small air bubbles which can be produced by various methods. If these small air bubbles enter a patient’s blood stream, the effects can range from stroke to death, depending upon the quantity and size of the bubbles. The analysis of microbubble population in response to variations in base flow is an idea which is pertinent to perfusion. This thesis begins with background in cavitation, microair, roller pumps and air production, as well as the processes used to reduce the incidence of air. The experiment, methods and materials used, as well as the ideas behind the idea are then described and illustrated. Statistical analysis, linear regression and second order polynomial evaluation are then discussed in detail with graphs illustrating the results. All raw data and calculations are included in the appendices.

Tounshendeaux, Jennifer M. "The Combined Effect Of Cell Salvage Reinfusion Volume And Hemodilution On Thromboelastogram Values During Coronary Artery Bypass Grafting"

March 2014, 33pp, references, appendices, figures, tables
Available for checkout

Abstract: The re-use of shed blood, called cell salvage, has become common during cardiac surgeries requiring the use of cardiopulmonary bypass (CPB). Cell salvage allows for preservation of hematocrit caused by hemodilution while lessening the use of homologous blood products. Although cell salvage decreases the necessity for blood transfusion, it is unclear whether or not the reinfusion of cell saver blood causes a disruption of hemostasis, or if the combination of cell salvage and overall hemodilution alters coagulation. Literature using traditional coagulation panels and indirect measure of hemostasis suggest that there may be an increase in fibrinolysis and postoperative bleeding in cardiac patients when cell salvage is used during surgery. However, these methods of quantifying the disruption of hemostasis are complicated by various factors, including Coumadin therapy or residual heparin present after bypass, which makes isolating the effect of cell salvage difficult. Thromboelastogram testing allows for the complete depiction of a coagulation profile, and because it corrects for anticoagulant use, it can better identify how other interventions affect hemostasis. The goal of this study was to compare presurgical and postsurgical thromboelastogram values from patients who underwent CPB with the use of cell salvage to determine the impact of cell salvage on clot profiles. Data were collected from the records of patients who underwent coronary artery bypass grafting using cardiopulmonary bypass (including the use of cell salvage) at Aurora St. Luke’s Medical Center from December 2011 through August 2013. The population sample was limited to first time surgical candidates who also underwent harvesting of a single mammary artery. Patients with preoperative thromboelastogram values outside of the normal range and/or received intraoperative or postoperative blood product transfusions were excluded.

Changes in pre- to postoperative thromboelastogram were evaluated to determine if hemodilution and/or cell saver reinfusion volumes were significant predictors within the sample (N=196). Student’s t-tests found that there was a significant decrease in the TEG maximum amplitude value (from 68.1 ± 8.5 mm preoperative to 63.1 ± 6.7 mm postoperative, p<0.05), which signifies a decrease in clot strength, during surgery. Multiple regression analysis was performed to determine significant predictors of this change. Hemodilution, which was represented by change in hematocrit, was a significant predictor (p<0.05) of change in maximum amplitude.

It can be concluded that cell saver reinfusion volumes within the represented range (676.2 ± 305.4 mL) can be utilized without concern for the development of coagulopathies in this population. However, precautions should be taken in order to reduce hemodilution as much as possible in order to avoid resulting coagulopathies.


Tsarovsky, Gary, B.S. “Effect of Cardiopulmonary Bypass Duration on Platelet Function and Procoagulant Inhibitor Systems.”

June 1998, 56pp, 34 references
  • Appendix A: Instructions for Authors, Journal of Extra-Corporeal Technology (JECT).
  • Appendix B: Raw Data Tables
  • Appendix C: Normal Hematologic Values
Available for checkout
Abstract: The purpose of this study was to examine the effect of cardiopulmonary bypass (CPB) duration on platelet function and procoagulant inhibitor systems. The hypothesis tested in this study was that the duration of CPB has a direct correlation with the severity of platelet dysfunction and the activity of the fibrinolytic system of hemostasis. Demographic and hematologic data were recorded for 21 patients undergoing elective cardiac surgery. Serial measurements of platelet function, using HemoSTATUSÒ (HemoTec, Medtronic, Inc., Parker, CO) platelet function test, and activity of procoagulant inhibitor systems were obtained, starting preoperatively and terminating at 24 hours after the operation. A significant decrease in platelet function following termination of CPB and neutralization of heparin (17% decrease from baseline, P <0.05) was observed, and persisted through the 24-hour postoperative period (11% decrease from baseline). Correlation between CPB time and 24-hour chest tube drainage (CTD) was present (r = 0.402, P = NS). Correlation between procoagulant inhibitor systems and CTD existed and was significant for protein C (r = 0.525, P = 0.0277) and protein S (r = 0.486, P = 0.409) at the time of heparin reversal. Correlations between CPB duration and activity of procoagulant inhibitor systems, CPB duration and platelet function, as well as platelet function and activity of procoagulant inhibitor systems existed, but were not statistically significant. The data obtained in this study does not demonstrate a statistically significant correlation between duration of CPB and platelet function. Platelet function, as assessed by HemoSTATUS platelet function test, and activity of procoagulant inhibitor systems may be useful in the observing trends in patient’s perioperative hemostasis, but, based on the data collected in this study, do not definitively identify patients at risk of excessive postoperative blood loss.

Tucker, Sherrelle “The Effect of Clamping the Outlet of a Centrifugal Pump on Gaseous Microemboli Production”

April 2012, 53pp, 37 references, appendices
Available for checkout
Abstract: Clamping the outlet of a running centrifugal pump is common practice, and generally considered safe by perfusionists. Gaseous microemboli (GME) production caused by clamping the inlet of a centrifugal pump is well documented. Yet no experiments studying the effect of clamping the outlet of a centrifugal pump on GME production have been carried out. The purpose of this experiment was to investigate if outlet clamp time or speed of the running centrifugal pump during clamping affects GME production.Total GME count and air volume were measured while varying clamp time and pump speed. an EDAC™ QUANTIFIER sensor manufactured by Luna Innovations, Inc. was placed in the outlet of the centrifugal pump to measure both the number of GME and their combined total volume of aid (μL).

Results indicate that both clamp time and pump speed had a statistically significant effect on total GME count and air volume (P<0.05). In this experiment, longer clamp time produced more total GME and overall air volume. Additionally, as pump speed increased while the outlet was clamped, more total GME and air volume were measured.

The results of this study suggest that clamping the outlet of the pump for at least 5 minutes will have an effect on air production in a cardiopulmonary bypass (CPB) circuit. Internal recirculation cavitation caused by increased pump head turbulence and heat generation in the pump head may be the cause of the air production.


Van Hill, Chad “The Effects of Heparin Loading Dose on Postoperative Hemostatic Transfusions: A Retrospective Review.”

April 2008, 54pp, 69 references, appendices
Available for checkout
Abstract: Objective: The purpose of this study was to determine if the amount of heparin given as a loading dose, prior to cardiopulmonary bypass (CPB), correlated to the transfusion of blood products following CPB. Methods: The objective was accomplished by performing a retrospective review of 1944 primary CABG-only patients who had surgery at Aurora St. Luke’s Hospital (Milwaukee, WI) from 2004 to 2007. Conclusions: Findings from this study were not significant and indicate that the pre-CPB heparin loading dose does not affect the transfusion of blood products following CPB.

VanderWal, Patrick L., B.S. “Modification of a Proposed Autoperfusion Technique.”

Oct. 24, 1996, 36pp, 52 references, appendices
Available for checkout
Abstract: The aim of this investigation was to develop an autoperfusion circuit which incorporates artificial organ components with the ability to maintain myocardial viability for an extended period of time. These artificial organ components were investigated for their ability to accomplish their respective tasks. An autoperfusion circuit and fully transportable cart are described as well as a surgical technique for explantation and mounting of a beating, isolated, donor heart.

Varner, Caleb S. “Addition of Pulmonary Circulation and Variable Venous Compliance to the Howard Perfusion Simulator”

February 2013, 111pp, 25 references, appendices, figures, tables
Available for checkout
Abstract: Simulation is becoming an important tool in the training of perfusionists. The goal of this project was to add both a pulmonary circulation and a variable venous compliance to the current Howard simulator that was developed, and is used, at the Milwaukee School of Engineering. This was achieved with the development of a pulmonary Windkessel chamber to simulate the pulmonary artery compliance, as well as through the addition of a variable venous compliance chamber. These components were tested individually, matching the specifications to their physiological counterparts, and then combined as a whole. The testing determined that the simulator will accurately represent the hemodynamic responses seen during cardiopulmonary bypass. These additions have expanded the use of the simulator such that students can now use it to practice initiation of bypass, and/or weaning from bypass with or without an aortic root vent, as well as the practice of cardioplegia administration and catastrophic event management.

Varso, Samantha, B.S. "A Prospective Analysis of Aurora St. Luke’s Medical Center Transfusion Practices and a Proposal for Quality Improvement of Blood Product Administration"

March 2016, 43pp
Available for checkout

Abstract: Although red blood cell transfusions are common treatment for postoperative anemia in cardiac surgery patients, homologous blood products can initiate various morbidities or mortality. The decision to transfuse must be based on supported clinical evidence, and clinical sites are expected to maintain a transfusion-tracking database to provide this clinical evidence for future decision making. But it is often a challenge to maintain such a database in a manner that provides reliable data. This was exemplified when the Aurora St. Luke’s Medical Center (ASLMC) cardiothoracic surgery department was flagged with inappropriate blood usage for January through July 2015 by the reporting company Accumen. The department believed that the report may have not been reflective of the actual transfusion practices in the department. In order to better understand the blood product usages of the ASLMC cardiothoracic (CT) surgery department, a Plan-Do-Study-Act (PDSA) concept was applied to the blood usage practices of Society of Thoracic Surgery (STS) procedure patients in the cardiovascular intensive care unit (CVICU). The goals of the PDSA cycle were to accurately characterize how transfusions are given by collecting data that characterize and provide a more thorough context of the transfusion process.

A prospective analysis was performed on data from cardiac surgical patients in the ASLMC CVICU between January 1, 2016 and January 31, 2016. All data were obtained manually from the ASLMC electronic database EPIC. The electronic charts for 72 STS procedure patients were analyzed for packed red blood cell (PRBC) transfusion records, resulting in data for 51 transfusions for 27 patients. Data collected included transfusion indications, ordering hemoglobin (hgb) concentrations, and other hemodynamic variables. The percentage of transfusions ordered as single units, ordered with a hgb < 7 g/dL, and ordered with a hgb > 8 g/dL were calculated and compared to the minimum goal value as well as the transfusion data for all departments and CT surgery from July through December 2015. Additionally, admit hgb was also reviewed to determine the average degree of anemia of patients upon admittance to the CVICU.

The results of the PDSA cycle showed that 46% of transfusions were given when hgb < 7 gm/dL, 19% of transfusions were given when hgb > 8 gm/dL, and 92% of PRBC transfusions were ordered as single units, compared to the goal minimums of at least 50%, less than 15%, and at least 50% respectively. Analysis of other recorded hemodynamic corresponding to the time of recorded transfusions shows an inconsistency of data recording, causing difficulty for accurate transfusion characterization. This project demonstrated that for STS procedure patients housed in the CVICU in January 2016, the hgb trigger-based goals were not reached, but the single unit per order goal was achieved. The removal of non-STS procedure patients improved the transfusion outcomes for the CVICU, indicating that the patient blood management of non-STS patients should be handled with different blood management algorithms.


Voborsky, Shannon R., B.S. “Assessment of Right Heart Perfusion in the Presence of A Non-Pulsatile Left Ventricular Assist Device.”

June 9, 2003, 126pp, 40 references
  • Appendix A: SIMULINK Blocks Used to Build the Functional Model Components
  • Appendix B: The Variables Used in the Complete Model: Descriptions, Initial Settings, and Units
  • Appendix C: SIMULINK Functional Model Components
  • Appendix D: Complete SIMULINK Cardiovascular Model with Incorporated Non-Pulsatile Left Ventricular Assist Device System
  • Appendix E: MATLAB Elastance Functions
  • Appendix F: The Adjusted Variable Settings for Each of the Experimental Simulations
Available for checkout
Abstract: Recent focus has been on utilizing ventricular assist devices (VAD’s) as destination therapies. To address this possibility, many new centrifugal VAD’s have been designed which are less complex mechanically, have fewer moving parts, and are much smaller allowing for easier implantation in comparison to the conventional pulsatile VAD’s. This research attempted to determine whether a centrifugal pump, when placed in parallel with a compromised left ventricle, allows for adequate perfusion of the myocardium. Perfusion of the right side of the heart was of particular concern because right heart failure is too often a devastating complication of left ventricular assist device (LVAD) insertion. A mathematical model of the cardiovascular and coronary system from previous research was utilized and revised to accommodate the application of a centrifugal LVAD and to examine the effects of left ventricular contribution and systemic vascular resistance on right coronary artery perfusion. The results suggested that while the LVAD does improve right coronary blood flow, when the left ventricular ejection fraction is 22% or less, the coronary flow becomes substantially less than under normal physiological conditions. In the presence of low systemic vascular resistance (SVR) a centrifugal pump may not be able to produce enough flow to provide adequate aortic pressures, however, in the presence of high SVR, the right coronary artery blood flow increased in comparison to baseline. These results imply that centrifugal LVAD may have a place as an interventionary tool (such as a bridge to recovery or transplant), which, when applied early on in the disease process may avoid right heart failure all together. Its application as a destination therapy may or may not be realized in the future.

Warhoover, Matthew T., B.S. “Determination of the Effects of Two Intra-Operative Blood Conservation Techniques (Cardiotomy Reservoir Versus Cell-Saver) On Systemic Inflammatory Response When Augmented by Leukocyte Filtration: A Pilot Study.”

Aug. 29, 2002, 79pp, 38 references
  • Appendix A: Institutional Review and Patient Informed Consent Documents
  • Appendix B: Raw Data Tables
  • Appendix C: Supplemental Graphs
  • Appendix D: Statistical Analysis Worksheets
Available for checkout
Abstract: Background: In cardiac surgery utilizing cardiopulmonary bypass (CPB) there is significant blood loss due to surgical incision and the use of anticoagulants. This volume is returned systemically to the patient by means of cardiotomy suction. However, returning this shed blood, which contains numerous inflammatory mediators, results in a “whole body inflammatory response” that contributes to multi organ failure, diffuse tissue edema, and pathological hypotension. By utilizing a cell-saver, which filters the inflammatory mediators from the shed blood, as the only means for autologous blood salvaging, this “whole body inflammatory response” may be minimized improving patient outcomes. Purpose: This pilot study was undertaken to examine the effect of different intraoperative blood salvage techniques on peri- and post-operative inflammatory response, and to determine if utilizing a blood processing device (cell-saver) for retrieval of shed blood is an effective tool to minimize overall inflammatory response during cardiac surgery. Methods: Using the same surgeon, a blinded study on 12 patients undergoing first time coronary artery bypass grafting with the use of CPB was performed. Patients were randomly assigned pre-operatively to either a control group, utilizing traditional cardiotomy suction or a study group, utilizing only a cell-saver for salvaging shed blood. CPB, except for the use of cardiotomy suction, and anesthetic techniques were standardized for all patients. Inflammation was determined by pre-, peri-, and post-operative levels of terminal complement complex (TCC) and white blood cell counts. In addition, cardiac indexes were also calculated pre-, peri- and post-operatively as a clinical parameter to measure inflammation. Results: Patients in the control group experienced higher TCC levels at all time intervals than the study group, and at the 15 minute post-CPB time interval, patients in the control group experienced significantly higher TCC levels than the study group. Furthermore, 33% of the patients in the control group needed inotropic support immediately following the termination of CPB, whereas no patients in the study group required pharmacological intervention. In addition 8 hour post-operative cardiac indexes increased 31% from pre-operative calculations in the study group, while the cardiac indexes of the control group increased only 16% without inotropic support and 24% with inotropic support within the same time periods. Conclusion: These data suggest that utilizing a cell-saver decreases the degree of “whole body inflammation.” However, only at the 15 minute post-CPB time interval did statistical significance differences occur in TCC levels. Thus, this study does not conclusively show that traditional cardiotomy suction generated a higher inflammatory response than the use of a cell-saver for salvaging shed blood. However, with the trending data of this study suggesting greater inflammatory with the use of traditional pump suction, further investigation with a larger patient sample size is warranted for conclusive determination of the difference, if any between the use of traditional cardiotomy suction and a cell-saver for salvaging intra-operative shed blood.

Watts, George G., B.S. “The Effects of Clamping the Outlet of Two Different Centrifugal Pumps on Platelet Function”

May 2011, 49pp, 39 references, appendices, figures, tables
Available for checkout
Abstract: One of the most important components of hemostasis is the function and efficacy of platelets. They are responsible for being one of the first response mechanisms to tissue injury and they also activate the coagulation and wound healing responses. Cardiopulmonary bypass (CPB) is well known to have an adverse effect on platelet function and number, yet is a necessary component for most cardiac surgical procedures. One possible cause of platelet dysfunction during CPB is platelet damage caused by the shear stress generated by centrifugal pump heads. Shear stress has a notable effect on platelet function both in terms of irreversible aggregation and complete lysis. Shear stress is generated at different locations within centrifugal pumps, depending on the pump design.
Every surgery involving CPB eventually involves weaning of the patient off of the extracorporeal pump. In a circuit that uses a centrifugal pump, this requires eventually shutting off forward flow (and also backflow) to the patient by placing a clamp distal to the pump head. The non-occlusive nature of the pumps require a clamp be in place, and from there, the perfusionist will fill the patient with whatever volume remains in their circuit depending on the current patient hemodynamics. The practice of many perfusionists is to leave the centrifugal pump running at a speed high enough to still overcome the resistance posted by the patient. With the clamp in place, this will result in high pressures and shear forces on the blood within the pump head. The goal of this novel study was to determine a correlation between pump speed, the time that the outflow of the pump is clamped, the pump design and the resultant level of platelet activation due to each factor.
A total of 24 trials were performed in which bovine blood was circulated through miniature mock CPB circuits. A Bio-Medicus Biopump BP-80 and a Sarns 7850 centrifugal pump were used with two different levels of clamp time and pump speed. The levels of time were chosen to be 30 seconds and 5 minutes, and the speeds were correlated for each pump to a flow rate of 6.15 LPM and 12.9 LPM in this minimal resistance circuit. The Plateletworks platelet function assay, with collagen as an agonist, was used to determine levels of platelet activation. All statistical analyses were performed with a level of significance at p<0.05.
There was a positive correlation found with the Biopump, increased clamp time, and increased pump speed with an increase in platelet inhibition. However, the findings were not statistically significant and had a low statistical power associated with the analyses. Considering that this was a novel study, it was encouraging to see the expected correlation despite a lack of statistical significance. Further study should be performed to determine a statistically significant association between the variables.

Webb, David P., B.A. “Pulmonary Implications of Filtering Various Mediators of Morbidity Found In Salvaged Blood.”

Feb. 20, 1997, 42pp, 17 references, appendices
Available for checkout
Abstract: Leukocyte reducing residual circuit blood after cardiopulmonary bypass has been demonstrated to improve lung function and reduce the inflammatory response after surgery. In this study, we examined whether lipid/leuko-reducing cell salvaged blood would improve pulmonary function and reduce the inflammatory response that can develop following the reinfusion of salvaged blood. In addition, all residual circuit blood was transferred to the cell washer at the end of bypass to contribute a significant portion of the final washed product. Fifteen patients undergoing elective cardiac surgery were randomly allocated to a lipid/leuko-reduced group or a control group In the lipid/leuko-reduced group (n = 10) all processed blood was filtered by a lipid globule, C3a, and microaggregate pre-filter (lipiguard), followed by a leukocyte reducing filter (RC 400). In the control group (n = 5) all processed blood was filtered by a 40/150 um dual screen transfusion filter. The lipid/leuko-reduced group showed a significant decrease in pulmonary shunt fraction and an increase in the arterial-venous oxygen difference (p < 0.05) following reinfustion, whereas the control group did not (p > 0.05). The lipid/leuko-reduced group also showed a trend towards decreased pulmonary vascular resistance and a blunting of the leukocytosis that develops following the reinfusion of salvaged blood, although these trends were not statistically significant. The magnitude of rise in circulating red blood cells was equivalent in both groups following reinfusion. These results suggest that lipid/leuko-reducing salvaged blood improves postoperative lung function, is efficacious, and safe for the patient population studied.

Wellstein, Marguerite “Design and Evaluation of a MATLAB® Simulated Blood Gas Monitor”

February 2017
Available for checkout
Available full-text online
Abstract: Immersive perfusion simulation is becoming more prevalent in perfusion education and training. This is best thought to be achieved through high fidelity simulations that mimic as much of the operating room environment as possible. Thus, this type of simulation would include all equipment, both application and monitoring. MSOE currently has an in-house developed hemodynamic perfusion simulator. It allows students to practice tasks such as initiating bypass, maintaining arterial pressure, and weaning from bypass. Currently this simulator is not paired with blood gas monitoring equipment. The goal of this project was to add blood gas monitoring capabilities to the simulator.
The blood gas monitor simulator was developed by modeling the physiologic systems that interact with blood gas monitors and creating an interactive program that would display the values achieved with normal blood gas analysis. The physiologic models were represented by a set of equations that were entered into MATLAB® as a function within a programmed graphical user interface (GUI). When the code initiated the function, all relevant values were collected from the interactive control and used in the calculations to determine dependent blood gas values. All values were then displayed.
Testing of the system through trial runs specific to the programmed models found that the virtual simulation would accurately represent the normal physiologic responses to changes in blood gas values, as well as the effects of cardiopulmonary bypass on those mechanisms. The addition of the simulated continuous blood gas monitor has created a more immersive simulation that allows students to practice assessing blood gas values and improve situational awareness that is necessary when these values change.

Contact the Library
Phone:

(414) 277-7180

Address:

500 E. Kilbourn Ave.
Milwaukee, WI 53202

Library Facebook

Library Twitter

Ask a Librarian

Library Staff

Library Director:
Gary Shimek

Librarian I:
Sarah Rowell

Librarian II and Archivist:
Denise Gergetz

Senior Assistant Librarians:
Karen Bolton
Pam Gorzalski