Detection and Elimination of Microemboli Related to ...

Detection and Elimination of Microemboli Related toCardiopulmonary Bypass

Robert C. Groom, MS, CCP; Reed D. Quinn, MD; Paul Lennon, MD; Desmond J. Donegan, MD;John H. Braxton, MBA, MD; Robert S. Kramer, MD; Paul W. Weldner, MD; Louis Russo, MD;Seth D. Blank, MD; Angus A. Christie, MD; Andreas H. Taenzer, MD; Richard J. Forest, CCP;

Cantwell Clark, MS, MD; Janine Welch, MS, CCP; Cathy S. Ross, MS;Gerald T. O’Connor, PhD, ScD; Donald S. Likosky, PhD; for the Northern New England

Cardiovascular Disease Study Group

Background—Neurobehavioral impairment is a common complication of coronary bypass surgery. Cerebral microemboliduring cardiopulmonary bypass (CPB) are a principal mechanism of cognitive injury. The aim of this work was to studythe occurrence of cerebral embolism during CPB and to evaluate the effectiveness of evidence-based CPB circuitcomponent and process changes on the exposure of the patient to emboli.

Methods and Results—M-Mode Doppler was used to detect emboli in the inflow and outflow of cardiopulmonary circuitand in the right and left middle cerebral arteries. Doppler signals were merged into a single display to allow real-timeassociations between discrete clinical techniques and emboli detection. One hundred sixty-nine isolated coronary arterybypass grafting (CABG) patients were studied between 2002 and 2008. There was no statistical difference in medianmicroemboli detected in the inflow of the CPB circuit, (Phase I, 931; Phase II, 1214; Phase III, 1253; Phase IV, 1125;F [3,158]�0.8, P�0.96). Significant changes occurred in median microemboli detected in the outflow of the CPB circuitacross phases, (Phase I, 702; Phase II, 572; Phase III, 596; Phase IV, 85; F [3,157]�13.1, P�0.001). Significant changesalso occurred in median microemboli detected in the brain across phases, (Phase I, 604; Phase II, 429; Phase III, 407;Phase IV, 138; F [3,153]�14.4, P�0.001). Changes in the cardiopulmonary bypass circuit were associated with an87.9% (702 versus 85) reduction in median microemboli in the outflow of the CPB circuit (P�0.001), and a 77.2% (604versus 146) reduction in microemboli in the brain (P�0.001).

Conclusions—Changes in CPB techniques and circuit components, including filter size and type of pump, resulted in areduction in more than 75% of cerebral microemboli. (Circ Cardiovasc Qual Outcomes. 2009;2:191-198.)

Key Words: surgery � cardiopulmonary bypass � cerebrovascular circulation � embolism � coronary disease

Approximately 427 000 patients undergo open heart sur-gery every year in the United States.1 Cardiopulmonary

bypass (CPB) is commonly used to provide circulatorysupport during these procedures. Neurological injury subse-quent to heart surgery was first reported in 1954 and wasattributed to stress and psychoses related to the procedure.2

Gilman in 1965, in a series of patients undergoing open heartprocedures using cardiopulmonary bypass, postulated anorganic origin stemming from reduced cerebral blood flowowing to microemboli.3 Subsequent research, including thatfrom our group, continues to implicate microemboli as theprincipal mechanism of neurological injury.4–11 These inju-ries may include strokes (incidence 1.3% to 4.3%)12 or more

prevalent neurobehavioral injuries (incidence 50% at dis-charge, 24% at 6 months, and 42% at 5 years).13

Whereas some emboli are generated as a result of manip-ulating the heart and great vessels14 or the reinfusion ofunprocessed shed blood from the surgical field,15,16 theleading source of emboli during surgery is from the cardio-pulmonary bypass circuit.17 Gaseous microemboli may enterthe cardiopulmonary bypass circuit from cannulation sites atthe surgical field and subsequently be infused into thepatient’s arterial circulation through the outflow of thecardiopulmonary bypass (CPB) circuit.18,19 Additionalsources of gaseous microemboli include the administration ofmedications into the circuit17,20 or drawing blood samples

Received July 2, 2008; accepted February 9, 2009.From Cardiac Surgery (R.C.G., R.D.Q., D.J.D., J.H.B., R.S.K., P.W.W., L.R., S.D.B., R.J.F., J.W.) and the Department of Anesthesia (P.L., A.A.C.),

Maine Medical Center, Portland; and the Departments of Surgery (D.S.L.) and Community & Family Medicine (D.S.L., C.S.R., G.T.O.), DartmouthMedical School, the Department of Anesthesiology (A.H.T., C.C.), Dartmouth-Hitchcock Medical Center, and the Dartmouth Institute for Health Policy& Clinical Practice (D.S.L., G.T.O.), Dartmouth College, Lebanon, NH.

The online-only Data Supplement can be found at http://circep.ahajournals.org/cgi/content/full/10.1161/CIRCOUTCOMES.108.803163/DC1.Correspondence to Robert C. Groom, MS, CCP, Associate Vice President of Cardiac Services, Director of Cardiovascular Perfusion, Maine Medical

Center, 22 Bramhall Street, Portland, ME 04102. E-mail [email protected]© 2009 American Heart Association, Inc.

Circ Cardiovasc Qual Outcomes is available at http://circoutcomes.ahajournals.org DOI: 10.1161/CIRCOUTCOMES.108.803163

191

by guest on February 14, 2018http://circoutcom

es.ahajournals.org/D

ownloaded from

http://circoutcomes.ahajournals.org/

from the circuit.17,20,21 Numerous studies have shown thatCPB systems are not capable of completely removing thesemicroemboli.8,22,23 The detection of microemboli requiressophisticated monitoring modalities, such as Doppler ultran-sonography, and as such may not be readily available to thesurgical team during the major parts of the operation.4,8

We used a continuous neurological and systemic monitor-ing system to prospectively study microemboli in the CPBcircuit and in the brain during coronary artery bypass grafting(CABG) surgery. We evaluated the effectiveness of the CPBcircuit, subsequent to evidence-based changes to the circuit orchanges in techniques, in mitigating microemboli delivery tothe patient.

MethodsStudy ModelA monitoring system was designed capable of providing real timeassociations between discrete clinical events/techniques and thedetection of microemboli in the CPB circuit and the middle cerebralarteries.19 Blood flow velocity and microemboli were recorded every8 milliseconds in the cerebral arteries and the inflow and outflow ofthe CPB circuit, using Doppler ultrasound (TCD 100 mol/L DigitalTranscranial Power M-Mode Doppler [Spencer Technologies], Fig-ure 2).24 A tripod mounted digital video camera was aimed at thesurgical field to record audio and video signals of surgical events andtechnique.

A video was created that synchronized outputs from the Dopplersignal of the cerebral arteries and cardiopulmonary bypass circuit(CPB), as well as the surgical video (Figure 2). Images werecombined onto one screen in real time using a video splitting device(Keywest Technology).

Data CollectionOur protocol received approval from the Institutional Review Board,and written informed consent was received before enrolling patientsinto the study.

We enrolled 169 patients between the age of 40 and 89 yearsundergoing isolated CABG procedures between October 2002 andJanuary 2008. A data form was used to collect intraoperativevariables. This data were supplemented with the Northern NewEngland Cardiovascular Disease Study Group’s (NNECDGS) car-diac surgery registry. The NNECDGS (http://www.nnecdsg.org/) is aregional collaborative composed of 8 medical centers in northernNew England. These centers collaborate with the goal of improvingcare of patients with cardiovascular disease. Our definitions forvariables have previously been reported.25 All authors had full accessto the data and take responsibility for its integrity. All authors haveread and agree to the manuscript as written.

Method of Conducting Cardiopulmonary Bypass

ManagementVenous drainage consisted of either gravity or augmented siphonfrom a 35-cm height differential from the right atrium to the inlet ofa rigid polycarbonate venous reservoir. A multiple stage singlevenous cannula was used. Arterial blood return was returned to theascending aorta with either a 7- or an 8-millimeter diameter soft-flowarterial perfusion cannula (Terumo Cardiovascular). Myocardialprotection was accomplished with either intermittent cold or contin-uous warm blood cardioplegia delivered antegrade (via the aorticroot) or retrograde (via the coronary sinus) or both. Anticoagulationwas accomplished by administering 4 milligrams of heparin perkilogram of body weight as a loading dose. Activated clotting times(ACT) were measured, and additional heparin was delivered asneeded to maintain the ACT above 480 seconds. Mean arterial bloodpressure during CPB was maintained between 55 mm Hg and75 mm Hg. The perfusion flow rate was maintained between 2.2 to

2.6 L/min/M2 to maintain a mixed venous oxygen saturation ofgreater than 60%. Phenylephrine was used to increase and nitroglyc-erin or isoflurane to decrease the arterial blood pressure during CPB.A continuous online blood gas monitor (CDI 500, Terumo Cardio-vascular Inc) was used to maintain Alpha-Stat blood gas strategy andfor surveillance of oxygenation. The CPB system consisted of ahollow fiber membrane oxygenator, an open venous reservoir, andSMARxT coated tubing (The Sorin Group). A Cell Saver device wasused to collect and process blood shed from the surgical field.(Hemonetics Corporation).

Improvement PlanOur underlying philosophy for improvement was based on the modelpreviously described by Batalden and colleagues (Figure 1).26 Thismodel involves providing both generalizable evidence from thepublished literature and also context knowledge to the multidisci-plinary team as a catalyst for improvement. We disseminatedgeneralizable knowledge gleaned from the peer-reviewed literatureconcerning the association between CPB circuit design and micro-emboli. Context knowledge related to emboli was obtained from ourintensive monitoring model and was also made available to the teamfor analysis and discussion. To that end we studied the generalizableevidence related to neurological injury associated with cardiacsurgery and context knowledge about the performance of our systemrelative to microembolic activity (supplemental Table I). The pub-lished evidence and data concerning effectiveness of changes madeto the circuit in reducing microemboli (context knowledge) weredisseminated to members of the multidisciplinary team at monthlystaff meetings, at our regional NNECDSG meetings held 3 times peryear, and at regional and national conferences. We used statisticalprocess control charts to display the effectiveness and to monitor thesustainability of the change in our system. Our expectation was thatusing such a strategy would provide us with ample opportunities forgaining grassroots interest and momentum for reducing microembol-ic activity during CPB.

A timeline of changes made to the CPB circuit components isdescribed in Table 1.

During Phase I (10/2002 to 4/2004), we conducted the observa-tional portion of the study. Phase II (4/2004 to 12/2005) consisted ofthe development of our quality improvement group. Additionally,our team stopped initiating CPB with an empty venous line. Ourteam began discussing findings from each monitored case beforeleaving the operating room during Phase III (12/2005 to 12/2006).Our team developed and implemented a series of changes duringPhase IV (12/2006 to 1/2008), including a smaller pore arterial linefilter, a venous reservoir with a smaller pore size integrated screenfilter, and a centrifugal pump (incorporated at the end of 2007;previous phases used a nonocclusive roller pump).

During Phases I-III, the CPB circuit consisted of a nonocclusiveroller pump, 40-micron arterial line filter and an Optima XPoxygenator (The Sorin Group). This oxygenator used a 105-micronvenous reservoir sock. During Phase IV, the CPB circuit consisted ofa 27-micron arterial line filter and a PrimO2x oxygenator (The SorinGroup). The PrimO2x oxygenator used a 30-micron venous reservoir

Figure 1. Foundation for Developing and Sustaining Improvementformula for sustaining improvement. Adapted from Batalden PB,Davidoff F. What is “quality improvement” and how can it trans-form health care? Qual Saf Health Care. 2007;16:2–3.

192 Circ Cardiovasc Qual Outcomes May 2009



ownloaded from


screen filter. The pump used during the latter half of Phase IV wasa Revolution centrifugal blood pump (The Sorin Group).

Statistical AnalysisAll analyses were performed using the STATA 10.0 program (StataCorporation).27 Comparisons between time points and conduct ofCPB were assessed using �2 tests, and continuous variables withStudent t test and Wilcoxon Rank Sum. Microemboli were log-transformed to achieve a near normal distribution. We tested theassociation between microemboli detected in the circuit and Phasethrough a linear regression analysis, adjusting for both patient ageand sex. We repeated this analysis with the dependent variable beingmicroemboli detected in the brain.

ResultsTable 2 displays pre-, intra-, and postoperative characteristicsfor the 169 individuals enrolled in the study. We stratifiedthese characteristics by Phase of the study, and found nostatistically significant differences. Half of the patients hadmild aortic disease. The rate of neurological injury (stroke ortransient ischemic attack) or mortality was low, as would beexpected given enrollment of 169 patients.

There was no statistical difference in microemboli detectedin the inflow of the CPB circuit, (Phase I, 931; Phase II, 1214;Phase III, 1253; Phase IV, 1125; F [3,158]�0.8, P�0.96).

Significant changes occurred in median microemboli detectedin the outflow of the CPB circuit across phases, (Phase I, 702;Phase II, 572; Phase III, 596; Phase IV, 85; F [3,157]�13.1,P�0.001). Significant changes also occurred in medianmicroemboli detected in the brain across phases, (Phase I,604; Phase II, 429; Phase III, 407; Phase IV, 138;F[3,153]�14.4, P�0.001). Significant reductions in micro-emboli in the outflow of the CPB circuit and brain wereattributed primarily to Phase IV (Figures 3 and 4).

At the beginning of Phase II, modifications were made tothe conduct of CPB among cases using controlled vacuumassisted venous drainage. During our observational phase, wenoted that there was a trend toward higher levels of micro-emboli in the inflow of CPB among cases using an empty(mean, 1702; median, 1329) versus primed (mean, 666;median, 598) venous line, P�0.08. Subsequently, the venousline was primed before the onset of CPB among cases usingVAVD. The target median VAVD pressure during CPB wasmodified during this first phase of the study. During Phase I,68% of cases using VAVD had higher levels of vacuum(more than �20 mm Hg), 53% during Phase II, 25% duringPhase III, and 55% during Phase IV (P�0.05). Our observa-tion of the effect of high vacuum exacerbating inflow emboli

Figure 2. Monitoring system schematic,depicting sources of Doppler signals inthe CPB circuit and in the brain with dis-play of Doppler signals and synchronousvideo of the surgical procedure.

Table 1. Structural and Process-Level Changes by Time

Phase

QualityImprovementWorkgroup

Discuss Findingsof Case Prior to

LeavingOperating Room

Method of Venous Drainage

Arterial LineFiltration Oxygenator Pump

Primed vs DryVenous Line

Degree ofVacuum

I October 2002 to April 2004 No No Primed or dry Target�40 mm Hg

40-�m 105-�m venousreservoir screen sock

Roller

II April 2004 to December 2005 Yes No Primed Target�20 mm Hg


Roller

III December 2005 to December 2006 Yes Yes Primed Target�20 mm Hg


Roller

IV December 2006 to January 2008 Yes Yes Primed Target�20 mm Hg

27-�m 30-�m venous reservoirscreen filter

Roller and centrifugal

Table depicts time intervals where process were changed or components to the CPB circuit were changed.

Groom et al Detection and Elimination of CPB Emboli 193



ownloaded from


prompted our minimizing the amount of vacuum used to alevel no higher than necessary to achieve adequate decom-pression of the right atrium. Overall, the use of either a27-micron arterial line filter or 30-micron venous reservoirsock (versus their larger pore counterparts) was associatedwith a 84% reduction in median outflow microemboli overthe course of all phases (579 versus 95, P�0.001). WithinPhase IV, the use of a centrifugal pump (n�7) was associatedwith an 86% (95 versus 13, P�0.02) reduction in outflowmicroemboli, relative to changes in arterial line filter andoxygenator.

Over the course of the entire study, changes in thecardiopulmonary bypass circuit were associated with an87.9% (702 versus 85) reduction in median microemboli inthe outflow of the CPB circuit (P�0.001), and a 77.2% (604versus 146) reduction in microemboli in the brain (P�0.001).

DiscussionThere is a growing body of evidence implicating the perni-cious nature of microemboli occurring during heart surgery.We sought to reduce the occurrence of microemboli using a2-fold strategy that involved clinical application of knowl-edge gained from the published work of others with a robustmodel for monitoring and contextualizing microembolic ac-tivity in the CPB circuit and brain. Although numerousclinical studies have documented microemboli during sur-gery, our study further contexualized microembolic activityenhancing our understanding of cerebral emboli emanatingfrom the CPB circuit and those related to surgical technique.Through this approach, we were able to assess the effective-ness of changes in technique and technology aimed atreducing microembolic activity. Our initiatives to redesignthe cardiopulmonary bypass circuit through both clinical

Table 2. Pre-, Intra-, and Postoperative Characteristics

Variable Phase I (n�35) Phase II (n�60) Phase III (n�37) Phase IV (n�37) P Value

Preoperative

Age in years, mean (SD) 64 (11) 66 (9) 67 (12) 64 (9) 0.25

Female, % 20 20 14 14 0.75

Diabetes, % 29 37 41 32 0.73

Vascular disease, % 14 22 27 24 0.60

BMI,* mean (SD) 29 (6) 31 (5) 29 (5) 29 (4) 0.50

�3 Distal anastomoses 89 90 84 84 0.75

Aortic disease†

None, % 44 75 53 53

Mild, % 50 75 53 56

Moderate, % 3 9 3 0

Severe, % 3 0 0 0

Not studied, % 6 11 3 5 0.89

Renal failure or creatinine 0 2 1 0

� 2 mg/dl, %

Priority, %

Elective 4 6 8 6

Urgent 96 94 92 94 0.18

Intraoperative

Pump time in minutes, mean (SD) 108 (26) 108 (27) 104 (26) 108 (37)

Clamp time in minutes, mean (SD) 65 (25) 65 (22) 64 (18) 67 (24)

Postoperative

Length of stay

Operation date to discharge, mean (SD) 6 (2) 7 (8) 8 (7) 6 (2) 0.24

Admission date to discharge, mean (SD) 9 (5) 10 (8) 11 (8) 11 (6) 0.59

Intensive care unit, mean (SD) 31 (30) 34 (30) 50 (53) 33 (31) 0.26

Extubation time in hours, mean (SD) 6 (5) 7 (9) 10 (14) 8 (10) 0.60

�2 inotropes at 48 hours (%) 3 2 8 3 0.31

Neurologic injury

TIA or stroke, % 0 1 1 0 1.00

Atrial fibrillation, % 14 22 24 22 0.75

In-hospital mortality, % 0 2 2 0 0.33

Table depicts intraoperative and postoperative patient characteristics.*kg/m2.†By epiaortic echocardiography updated October 6, 2008.




ownloaded from


process and circuit redesign resulted in an 87.9% reduction inmicroemboli detected in the outflow of the CPB circuit, anda 77.2% reduction in microemboli detected in the cerebralarteries. Ultimately changes in the CPB circuit lowered theCPB outflow count presently to below the cerebral microem-boli count. This latter observation suggests the greatestremaining opportunity for reducing cerebral microemboli isthrough the redesign of processes of surgical care.

Although both gravity venous drainage and VAVD mayresult in gaseous entrainment through cannulation sites andother incisions in the heart, the risk may be somewhat greaterwith VAVD, because it uses a higher negative pressure andexacerbates air entrainment. The difference in degree of airentrainment was the focus of previous in vitro investigations.18,23

In response to this data, our clinical team made theevidence-based changes shown in supplemental Table I overthe course of the study. We consistently observed microem-boli entrained into the venous line that was subsequentlyintroduced into the patient’s arterial circulation. This obser-vation was particularly evident during the onset of CPB andduring the introduction of the coronary sinus catheter forretrograde cardioplegia solution, even when attempts were

made to fill the heart to prevent microemboli entrainmentduring this maneuver. To prevent microemboli during theonset of CPB, the venous line of the circuit was fluid filledduring cases using VAVD sporadically during Phase I, andconsistently from Phase II through Phase IV. Without prim-ing the venous line, air within the venous cannula will bedelivered to the arterial circulation in a bolus manner.28

Additional changes included reducing the lower limit ofVAVD pressure required to drain the right atrium andmaintain suitable conditions for revascularization.28

To prevent any remaining microemboli from being deliv-ered to the patient, we coupled our efforts at redesigning theconduct of CPB with the redesign of our circuit components.The newer circuit (Phase IV) incorporates a 30-micronvenous reservoir screen filter and 27-micron arterial linefilter. This reduced pore size where the blood enters theoxygenator resulted in substantial reduction in the outflowemboli. Standards for blood oxygenators promulgated by theAssociation for the Advancement of Medical Instrumentationdo not set forth performance characteristics for blood oxy-genators or reservoirs with regard to microemboli. Conse-quently manufacturers are not required to comply with a

Figure 3. Median embolic signals overtime, showing median embolic signalsover time, collapsed into groups of 20patients each.




ownloaded from


benchmark nor do they provide performance characteristicsof their products to their customers with regard to microem-boli handling. Dickinson compared the effectiveness of man-ufacturers circuits in an in vitro study model and reportedvariation in circuit performance related to microemboli han-dling.22 The authors did not disclose the model or manufac-turer of the circuits tested. A reduction in the pore size of thearterial line filter, the last opportunity for microemboliremoval before blood returning to the patient, appeared inPhase IV to prevent most of the microemboli from the venouscirculation from traveling to the patient (as evidenced by astatistically significant drop in outflow microemboli withouta concurrent change in inflow microemboli during Phase IV).More recently, Riley assessed the pressure drop and filtrationefficiency of arterial line filters.29 This work showed that,generally, smaller pore size resulted in improved filtration ofmicroemboli. Exceptions included the Pall LeukoGuard-6Leukocyte Reduction Arterial Blood Filter, a 40-micronarterial line filter, which performed similarly to a 20-micronfilter with regard to microemboli elimination.

Although we are pleased with our present findings, wecannot rule out whether other circuit component changeswould be as effective as those used in the current report.Additionally, our data does not allow us to identify whichchanges in circuit design and components were most effectiveat reducing microemboli load to the patient. Nonetheless, ourfindings suggest that the greatest reduction in microembolileaving the outflow of the CPB circuit was attributed to PhaseIV of our study.

In the current study, additional reductions in microembolifrom the CPB circuit were realized through the use of acentrifugal pump. A number of studies have comparedcentrifugal and roller pumps in relation to microemboligeneration, inflammatory markers, and clinical out-comes.30–33 Reductions in microemboli likely occur throughminimization of gaseous microemboli through its cyclic flowof blood relative to the potential generation and entrainmentof gaseous microemboli through negative pressure and cavi-

tation in a roller pump. In a small trial by Wheeldon andcolleagues, significantly less microemboli generation, lesscomplement activation, and better preservation of plateletcount was observed in patients randomized to a centrifugalpump.30 Scott identified more neurobehavioral test deficitsamong the roller pump group, however this difference did notreach statistical significance.34 A study of 3438 consecutivepatients revealed that the use of the centrifugal pump showeda trend toward risk reduction for adverse neurologicalevents.31 Randomized trials with neurological measures as aprimary outcome variable, however, have not demonstratedsignificant differences in neurobehavioral outcomes or tissue-level markers of brain injury (S100 beta) between types ofCPB pumps.34,35 In the largest randomized trial, Klein andcolleagues assigned 1000 adult cardiac patients to manage-ment with a roller pump or a centrifugal pump, with reduc-tions in blood loss, renal function, and neurological outcomedemonstrated in the centrifugal group.32

Neurological injuries, predominantly embolic in origin, areunwanted sequellae of CABG surgery. The source of micro-emboli may be eliminated in most cases by the redesign of thecardiopulmonary bypass circuit. The changes appear to sub-stantially reduce microemboli delivered to the patient while atthe same not appreciably increasing the cost of deliveringcare. Changes in the CPB circuit have lowered the CPBoutflow count presently to below the cerebral microembolicount. Subsequent work will focus on identifying the sourcesof these microemboli, which likely may be associated withaortic management techniques.

Reductions in microemboli have likely resulted in partthrough the feedback of data from our monitoring modalities.Modifications in circuit design and techniques for conductingcardiopulmonary bypass are made every day in operatingrooms across this country. The assessment of their effective-ness in reducing adverse sequelae, such as emboli-handlingability, is at times inadequate. The collection of data concern-ing the principal mechanisms of brain injury, and subsequentfeedback of information to the clinical team, has enabled ourteam to assess the effectiveness of our redesign initiatives.

Figure 4. Median embolic signals overtime by phase, showing median embolicsignals by phase of the study.




ownloaded from


Figure 4 demonstrates that while improving our ability tominimize gaseous microemboli returning to the patient fromthe cardiopulmonary bypass circuit, there remains anothersource of microemboli: the aorta. These microemboli may begaseous or particulate. The sources of the gaseous microem-boli include entrained air and cavitation related to the arterialcannula. The release of particulate microemboli is likelyrelated to aortic manipulation, especially secondary to aorticcrossclamps, both partial and complete.36

Having minimized gaseous emboli from the CPB circuit,we can now focus on changes in processes of care to decreasethe aortic microemboli sources. Aortic manipulation using apartial occlusion clamp has been associated with cerebralmicroemboli.8 Aortic cannulae insertion technique and care-fully deairing the aorta when necessary will likely furtherreduce the cerebral microembolic count. An aortic arterialcannula with side holes and lower exit velocity, such as theone used in this study, may reduce embolization from thesand-blasting effect characterized by open end hole arterialcannulae.37

This study model has provided us with the requisiteknowledge necessary to realize significant reduction in mi-croembolic activity. Our strategic use of generalizable evi-dence from the literature, and modifications of the CPBcircuit based on information about the timing and source ofmicroemboli, resulted in a significant and sustainable reduc-tion in the exposure of CABG patients to cerebralmicroemboli.

AcknowledgmentsThe authors acknowledge Norma Albrecht for her assistance withmanuscript preparation and Catherine LaVopa, RN for assistancecoordinating this study. Equipment used for this study was loaned bySpencer Technologies (Seattle, Wash), and Terumo CardiovascularSystems Inc (Tustin, Calif). The authors thank Dr Gerald T.O’Connor for his steadfast support and guidance on this project.

Sources of FundingDr Likosky was supported by a grant from the Agency for HealthcareResearch and Quality (1K02HS015663-01A1). Equipment used for thisstudy was loaned by Spencer Technologies (Seattle, Wash), and TerumoCardiovascular Systems, Inc (Tustin, Calif). Sorin Group (Arvada,Colo) provided an unrestricted grant to support study-related ex-penses. Somanetics Corporation (Troy, Mich) provided funds tosupport a research coordinator. This work was partially funded bythe Northern New England Cardiovascular Disease Study Group.

DisclosuresNone.

References1. Heart Disease and Stroke Statistics - 2008 Update. Am Heart Association,

2008.2. Fox H, Rizzon N, Gifford S. Psychological observations of patients

undergoing mitral surgery: A study of stress. Am Heart J. 1954;48:645–670.

3. Gilman S. Cerebral disorders after open-heart operations. N Engl J Med.1965;272:489–498.

4. Likosky DS, Marrin CA, Caplan LR, Baribeau YR, Morton JR,Weintraub RM, Hartman GS, Hernandez F Jr, Braff SP, CharlesworthDC, Malenka DJ, Ross CS, O’Connor GT. Determination of etiologicmechanisms of strokes secondary to coronary artery bypass graft surgery.Stroke. 2003;34:2830–2834.

5. Hogue CW Jr, Murphy SF, Schechtman KB, Davila-Rom:an VG. Riskfactors for early or delayed stroke after cardiac surgery. Circulation.1999;100:642–647.

6. Blauth CI. Macroemboli and microemboli during cardiopulmonarybypass. The Ann Thorac Surg. 1995;59:1300–1303.

7. McKhann GM, Grega MA, Borowicz LM Jr, Baumgartner WA, SelnesOA. Stroke and encephalopathy after cardiac surgery: an update. Stroke.2006;37:562–571.

8. Stump DA, Rogers AT, Hammon JW, Newman SP. Cerebral emboli andcognitive outcome after cardiac surgery. J Cardiothorac Vasc Anesth.1996;10:113–118; quiz 118–119.

9. Kincaid EH, Jones TJ, Stump DA, Brown WR, Moody DM, Deal DD,Hammon JW Jr. Processing scavenged blood with a cell saver reducescerebral lipid microembolization. Ann Thorac Surg. 2000;70:1296–1300.

10. Likosky DS, Roth RM, Saykin AJ, Eskey CJ, Ross CS, O’Connor GT.Neurologic injury associated with CABG surgery: outcomes, mech-anisms, and opportunities for improvement. Heart Surg Forum. 2004;7:E650–E662.

11. Grega MA, Borowicz LM, Baumgartner WA. Impact of single clampversus double clamp technique on neurologic outcome. Ann Thorac Surg.2003;75:1387–1391.

12. Charlesworth DC, Likosky DS, Marrin CA, Maloney CT, Quinton HB,Morton JR, Leavitt BJ, Clough RA, O’Connor GT. Development andvalidation of a prediction model for strokes after coronary artery bypassgrafting. Ann Thorac Surg. 2003;76:436–443.

13. Newman MF, Kirchner JL, Phillips-Bute B, Gaver V, Grocott H, JonesRH, Mark DB, Reves JG, Blumenthal JA. Longitudinal assessment ofneurocognitive function after coronary- artery bypass surgery. N EnglJ Med. 2001;344:395–402.

14. Wolf LG, Abu-Omar Y, Choudhary BP, Pigott D, Taggart DP. Gaseousand solid cerebral microembolization during proximal aortic anastomosesin off-pump coronary surgery: the effect of an aortic side-biting clampand two clampless devices. J Thorac Cardiovasc Surg. 2007;133:485–493.

15. Moody DM, Brown WR, Challa VR, Stump DA, Reboussin DM, LegaultC. Brain microemboli associated with cardiopulmonary bypass: a his-tologic and magnetic resonance imaging study. Ann Thorac Surg. 1995;59:1304–1307.

16. Brooker RF, Brown WR, Moody DM, Hammon JW Jr, Reboussin DM,Deal DD, Ghazi-Birry HS, Stump DA. Cardiotomy suction: a majorsource of brain lipid emboli during cardiopulmonary bypass. Ann ThoracSurg. 1998;65:1651–1655.

17. Taylor RL, Borger MA, Weisel RD, Fedorko L, Feindel CM. Cerebralmicroemboli during cardiopulmonary bypass: increased emboli duringperfusionist interventions. Ann Thorac Surg. 1999;6:89–93.

18. Willcox TW, Mitchell SJ, Gorman DF. Venous air in the bypass circuit:a source of arterial line emboli exacerbated by vacuum-assisted drainage.The Ann Thorac Surg. 1999;68:1285–1289.

19. Rider SP, Simon LV, Rice BJ, Poulton CC. Assisted venous drainage,venous air, and gaseous microemboli transmission into the arterial line: anin-vitro study. J Extra-Corp Techn. 1998;30:160–165.

20. Borger MA, Peniston CM, Weisel RD, Vasiliou M, Green RE, FeindelCM. Neuropsychologic impairment after coronary bypass surgery: effectof gaseous microemboli during perfusionist interventions. J ThoracCardiovasc Surg. 2001;121:743–749.

21. Rodriguez RA, Williams KA, Babaev A, Rubens F, Nathan HJ. Effect ofperfusionist technique on cerebral embolization during cardiopulmonarybypass. Perfusion. 2005;20:3–10.

22. Dickinson TA, Riley JB, Crowley JC, Zabetakis PM. In vitro evaluationof the air separation ability of four cardiovascular manufacturer extracor-poreal circuit designs. J Extra-Corp Techn. 2006;38:206–213.

23. Jones TJ, Deal DD, Vernon JC, Blackburn N, Stump DA. Does vacuum-assisted venous drainage increase gaseous microemboli during cardiopul-monary bypass? The Ann Thorac Surg. 2002;74:2132–2137.

24. Moehring MA, Spencer MP. Power M-mode Doppler (PMD) forobserving cerebral blood flow and tracking emboli. Ultrasound Med Biol.2002;28:49–57.

25. O’Connor GT, Plume SK, Olmstead EM, Coffin LH, Morton JR,Maloney CT, Nowicki ER, Tryzelaar JF, Hernandez F, Adrian L, CaseyKJ, Soule DN, Marrin CAS, Nugent WC, Charlesworth DC, Clough R,Katz S, Leavitt BJ, Wennberg JE. A regional prospective study ofin-hospital mortality associated with coronary artery bypass grafting. TheNorthern New England Cardiovascular Disease Study Group [seecomments]. JAMA. 1991;266:803–809.




ownloaded from


26. Batalden PB, Davidoff F. What is “quality improvement” and how can ittransform healthcare? Qual Saf Health Care. 2007;16:2–3.

27. Stata. Stata Statistical Software: Release 8.0. College Station, TX: StataCorporation; 2003.

28. Willcox TW. Vacuum-assisted venous drainage: to air or not to air, thatis the question. Has the bubble burst? J Extra-Corp Techn. 2002;34:24–28.

29. Riley JB. Arterial line filters ranked for gaseous micro-emboli sepa-ration performance: an in vitro study. J Extra-Corp Techn. 2008;40:21–26.

30. Wheeldon DR, Bethune DW, Gill RD. Vortex pumping for routinecardiac surgery: a comparative study. Perfusion. 1990;5:135–143.

31. Alamanni F, Parolari A, Zanobini M, Porqueddu M, Dainese L, BerteraA, Costa C, Fusari M, Spirito R, Biglioli P. Centrifugal pump andreduction of neurological risk in adult cardiac surgery. J Extra-CorpTechn. 2001;33:4–9.

32. Klein M, Dauben HP, Schulte HD, Gams E. Centrifugal pumping duringroutine open heart surgery improves clinical outcome. Artificial Organs.1998;22:326–336.

33. Baufreton C, Intrator L, Jansen PG, te Velthuis H, Le Besnerais P, VonkA, Farcet JP, Wildevuur CR, Loisance DY. Inflammatory response tocardiopulmonary bypass using roller or centrifugal pumps. The AnnThorac Surg. 1999;67:972–977.

34. Scott DA, Silbert BS, Doyle TJ, Blyth C, Borton MC, O’Brien J L, de LHorne DJ. Centrifugal versus roller head pumps for cardiopulmonarybypass: effect on early neuropsychologic outcomes after coronary arterysurgery. J Cardiothorac Vasc Anesth. 2002;16:715–722.

35. Ashraf S, Bhattacharya K, Zacharias S, Kaul P, Kay PH, Watterson KG.Serum S100beta release after coronary artery bypass grafting: rollerversus centrifugal pump. The Ann Thorac Surg. 1998;66:1958–1962.

36. Hammon JW, Stump DA, Butterworth JF, Moody DM, Rorie K, DealDD, Kincaid EH, Oaks TE, Kon ND. Single crossclamp improves6-month cognitive outcome in high-risk coronary bypass patients: theeffect of reduced aortic manipulation. The J Thorac Cardiovasc Surg.2006;131:114–121.

37. Groom RC, Hill AG, Kuban B, Oneill W, Akl BF, Speir AM, Kon-ingsberg J, Sprissler GT, Shakoor M, Massimiano PS, et al. Aorticcannula velocimetry. Perfusion. 1995;10:183–188.




ownloaded from


Donald S. Likosky and for the Northern New England Cardiovascular Disease Study GroupTaenzer, Richard J. Forest, Cantwell Clark, Janine Welch, Cathy S. Ross, Gerald T. O'Connor,

S. Kramer, Paul W. Weldner, Louis Russo, Seth D. Blank, Angus A. Christie, Andreas H. Robert C. Groom, Reed D. Quinn, Paul Lennon, Desmond J. Donegan, John H. Braxton, Robert

Detection and Elimination of Microemboli Related to Cardiopulmonary Bypass

Print ISSN: 1941-7705. Online ISSN: 1941-7713 Copyright © 2009 American Heart Association, Inc. All rights reserved.

Greenville Avenue, Dallas, TX 75231is published by the American Heart Association, 7272Circulation: Cardiovascular Quality and Outcomes

published online May 5, 2009;Circ Cardiovasc Qual Outcomes.

http://circoutcomes.ahajournals.org/content/early/2009/05/05/CIRCOUTCOMES.108.803163World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circoutcomes.ahajournals.org//subscriptions/

at: is onlineCirculation: Cardiovascular Quality and Outcomes Information about subscribing to Subscriptions:

http://www.lww.com/reprints

Information about reprints can be found online at: Reprints:

document. Question and AnswerPermissions and Rightspage under Services. Further information about this process is available in the

which permission is being requested is located, click Request Permissions in the middle column of the WebCopyright Clearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of theCirculation: Cardiovascular Quality and Outcomesin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:



ownloaded from

http://circoutcomes.ahajournals.org/content/early/2009/05/05/CIRCOUTCOMES.108.803163

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circoutcomes.ahajournals.org//subscriptions/


Detection and Elimination of Microemboli Related to ...

Documents

Transcript of Detection and Elimination of Microemboli Related to ...