The National Heart, Lung, and Blood Institute PediatricCirculatory Support Program

J. Timothy Baldwin, PhD; Harvey S. Borovetz, PhD; Brian W. Duncan, MD; Mark J. Gartner, ME;Robert K. Jarvik, MD; William J. Weiss, PhD; Tracey R. Hoke, MD, ScM

Abstract—Options for the circulatory support of pediatric patients under the age of 5 years are currently limited toshort-term extracorporeal devices, the use of which is often complicated by infection, bleeding, and thromboembolism.Recognizing this void, the National Heart, Lung, and Blood Institute solicited proposals for the development of novelcirculatory support systems for infants and children from 2 to 25 kg with congenital or acquired cardiovascular disease.Five contracts were awarded to develop a family of devices that includes (1) an implantable mixed-flow ventricularassist device designed specifically for patients up to 2 years of age, (2) another mixed-flow ventricular assist device thatcan be implanted intravascularly or extravascularly depending on patient size, (3) compact integrated pediatriccardiopulmonary assist systems, (4) apically implanted axial-flow ventricular assist devices, and (5) pulsatile-flowventricular assist devices. The common objective for these devices is to reliably provide circulatory support for infantsand children while minimizing risks related to infection, bleeding, and thromboembolism. The devices are expected tobe ready for clinical studies at the conclusion of the awards in 2009. (Circulation. 2006;113:147-155.)

Key Words: heart-assist device � pediatrics � congenital heart defects � heart diseases � heart failure

The need for mechanical circulatory support to treatchildren with congenital and acquired heart disease is

well established.1,2 Currently, 25% of the �36 000 babiesborn with congenital heart defects in the United Stateseach year require invasive treatment during the first year oflife.3 Although the annual mortality from heart defects inthe United States has declined by 39% since 1979, nearly1800 infants with congenital heart defects die each year.4

Acquired heart disease also affects this vulnerable pediat-ric population; based on National Heart, Lung, and BloodInstitute (NHLBI)-sponsored Pediatric CardiomyopathyRegistry data, each year in the United States nearly 350children under 1 year of age develop cardiomyopathy,*many of whom die or require cardiac transplantation.4,5

International registry data indicate that the majority ofthose receiving a heart transplant under 1 year of age havea diagnosis of congenital heart disease (67%), whereas theremainder have some form of cardiomyopathy (33%).6 Aparticular problem in this age group is the high mortalityrate suffered by infants awaiting transplantation; theUnited Network for Organ Sharing (UNOS) reports that ofthe more than 1600 infants added to the heart or heart/lung

transplant lists over the last decade, fewer than 50%received a donor organ.7 During this same period, 68% ofthe more than 3000 children between 1 and 18 years listedfor cardiac transplantation underwent successfultransplantations.

The use of mechanical circulatory support devices(MCSDs) as a bridge to transplantation has been shown todecrease waiting list mortality and improve the efficiency oforgan utilization in children.8 MCSDs have also been usedsuccessfully as a bridge to recovery in children, especially inthe management of acute fulminant myocarditis or postcar-diotomy heart failure.9–14 However, current MCSD optionsfor infants and children are quite limited, particularly inregard to duration of support. Extracorporeal membraneoxygenation (ECMO), which remains the only form ofmechanical circulatory support available at most pediatrictertiary care centers in the United States, is capable only ofproviding support for days to, at most, a few weeks. Thisbecomes a significant limitation when ECMO is used for theyoungest patients, those who often require significant time oncirculatory support until a suitable donor organ becomesavailable.

From the Division of Heart and Vascular Diseases, The National Heart, Lung, and Blood Institute (J.T.B., T.R.H.), Bethesda, Md; the Departmentsof Bioengineering, Surgery, and McGowan Institute for Regenerative Medicine, The University of Pittsburgh (H.S.B.), Pittsburgh, Pa; the Departmentsof Pediatric and Congenital Heart Surgery and Biomedical Engineering, The Cleveland Clinic Foundation (B.W.D.), Cleveland, Ohio; Ension, Inc(M.J.G.), Pittsburgh, Pa; Jarvik Heart, Inc (R.K.J.), New York, NY; and the Departments of Surgery and Bioengineering, The Pennsylvania StateUniversity, College of Medicine (W.J.W.), Hershey, Pa.

The online-only Data Supplement can be found at http://circ.ahajournals.org/cgi/content/full/113/1/147/DC1.Correspondence to J. Timothy Baldwin, PhD, Program Officer, Division of Heart and Vascular Diseases, NHLBI, Two Rockledge Centre, Room 9150

6701 Rockledge Dr, Bethesda, MD 20892-7940. E-mail baldwint@nhlbi.nih.gov*Based on the incidence rate of 8.34/100 000 for New England and the Central Southwest regions and extrapolated to the US national total of 4 113 000

live births for the 12 months ending November, 2004.© 2006 American Heart Association, Inc.

Circulation is available at http://www.circulationaha.org DOI: 10.1161/CIRCULATIONAHA.105.571422

147

Special Report

by guest on February 5, 2015http://circ.ahajournals.org/Downloaded from

Achieving effective mechanical circulatory support in chil-dren involves addressing important population-specific tech-nological challenges that do not exist in adults. One funda-mental design challenge arises from the need to providecirculatory support for the entire range of patient sizesencountered in pediatrics, from newborns to adolescents. Inaddition, because of the substantial growth that a child mayexperience while on circulatory support, pediatric MCSDsintended for chronic use should retain the capacity to increaseoutput to match circulatory requirements as the child grows.Also, owing to the small size of the vasculature in children,cannulation strategies must be capable of providing requiredcardiac support while preserving the integrity of small andoften fragile blood vessels.

Although size constraints are the most obvious challengefor the development of pediatric MCSDs, significant ana-tomic and physiological differences between pediatric andadult patients must also be considered during the design ofcirculatory support devices for children. For example, con-genital cardiac malformations demonstrate a vast number ofvarious combinations of anatomic abnormalities that maysignificantly affect cannulation. Transposition of the greatarteries, dextrocardia, and abnormalities of the systemic veinsare a few of the anatomic features that, when present alone orin combination, may require substantial creativity in planningfor appropriate cannulation strategies.15 Survival beyondinfancy is increasing in the population of patients withsingle-ventricle physiology who are successfully palliatedwith the Fontan procedure. The “Fontan pathway,” the mostcommon surgical sequence for single-ventricle patients, con-sists of a neonatal palliative procedure, followed by a bidi-rectional superior cavopulmonary anastomosis at �6 monthsof age, and the Fontan procedure at 2 to 4 years of age.Although the single ventricle may fail in children at any stageduring this sequence, it also fails in a significant numbers ofadolescents and young adults initially managed successfullywith the Fontan procedure. These patients provide substantialmanagement challenges owing to the multisystem nature ofcomplications that may develop secondary to systemic ve-nous hypertension (cirrhosis, protein-losing enteropathy) andbecause these patients often have some of the most significantanatomic abnormalities encountered. Finally, circulatory sup-port with left ventricular assist devices (LVADs), which isfrequently successful in the management of relatively isolatedleft ventricular failure in adults, may not be as likely tosucceed in children. Pediatric heart failure is more oftencharacterized by right ventricular or biventricular failure andpulmonary disease (especially pulmonary hypertension).Consequently, strategies that provide biventricular supportwithin the size limitations imposed by the small size of thepediatric thorax assume greater importance in MCSDs de-signed for pediatric applications.1,16

Devices Currently Available forPediatric Applications

The options for mechanical circulatory support for pediatricpatients in the United States are currently limited to a handfulof devices. Although no MCSD has received premarketapproval from the Food and Drug Administration (FDA) for

specific use in pediatric patients, various devices have beenused to provide acute and chronic circulatory support in thepediatric population. This is possible through off-label use ofFDA-approved adult devices, compassionate use of investi-gational adult ventricular assist devices (VADs), the use ofdevices that have received an FDA Humanitarian DeviceExemption, the use of devices fabricated from FDA-approvedcomponents (such as ECMO circuits), and devices that areapproved on a case-by-case basis by the FDA for emergencyuse.1,17 The current FDA-approved adult devices that havebeen used in pediatric patients include the Thoratec Ventric-ular Assist Device and the Heartmate LVAD (ThoratecLaboratories Corp), and the Abiomed BVS 5000 (Abiomed,Inc).18–21 Historically, these FDA-approved adult VADs havetypically been limited to pediatric patients with body surfaceareas of at least 0.7 m2 for extracorporeal devices and 1.4 m2

for implantable devices.1 However, because of their smallersizes, a number of the current group of investigationalimplantable adult devices, such as the MicroMed DeBakeyVAD (MicroMed Technologies, Inc), the Jarvik 2000 Flow-maker (Jarvik Heart, Inc), and the Thoratec Heartmate IIVAD, provide additional options for larger pediatric patients,for whom the use of such devices qualifies as a “compassion-ate treatment.”22–24 The Micromed DeBakey VAD Child wasgranted a “humanitarian device exemption” from the FDA inFebruary 2004 and was approved for use in pediatric patients5 to 16 years old who have a body surface area between 0.7and 1.5 m2 and who are in New York Heart Association classIV end-stage heart failure, are refractory to medical therapy,and are listed as candidates for cardiac transplantation.25 It isthe only intracorporeal circulatory support device available inthe United States exclusively designated for use in children.The device utilizes the same pump as the adult MicromedDeBakey VAD, although the inflow cannula and outflowgraft have been redesigned to allow implantation in smallerpediatric patients.

Although first reported as appropriate for prolonged circu-latory support in the early 1970s, ECMO remains the mostcommonly used method for mechanically supporting thecirculation of pediatric patients.1 ECMO circuits typicallyconsist of a roller or centrifugal pump that drives bloodthrough a membrane oxygenator, a blood warmer, and a filter.The number of cases of pediatric cardiac ECMO use in theUnited States reported to the Extracorporeal Life Support(ECLS) Registry has increased from approximately 400/yearin 1999 to 500/year in 2004, primarily owing to an increase inneonatal cardiac ECMO use during this time period.26–28

Although some centers report cardiac ECMO survival rates ashigh as 70%, the average has remained �40% since its earlyuse.1,26 However, because ECMO is only capable of provid-ing circulatory support in the acute setting, its usefulness islimited. ECMO circuits have been used for up to 30 days;however, the likelihood of adverse events increases signifi-cantly with longer duration of ECMO support. Most ECMOruns are limited to substantially shorter time periods (5 to 10days). Although ECMO has provided the only option formany pediatric patients with heart failure refractory to med-ical management, serious adverse events, such as hemorrhageand neurological complications, are common and may be

148 Circulation January 3/10, 2006

by guest on February 5, 2015http://circ.ahajournals.org/Downloaded from

catastrophic in any given case.9,14,16 In addition, patients onECMO are intubated, with restricted mobility, which pre-cludes significant physical rehabilitation during support, andthe complexity of the circuit usually requires continuallabor-intensive monitoring by specially trained experts.

Although ECMO has proved to be an effective form oftherapy for acute circulatory support in small children, theoptions for chronic circulatory support in children in theUnited States with body surface areas �0.7 m2 are quitelimited.1,17 The Berlin Heart Excor VAD (Berlin Heart AG),an extracorporeal device made in sizes ranging from 10 to 80mL, is available only for emergency use in the United States,and as such, it requires a petition to the FDA before eachimplantation. However, because of the absence of othersuitable devices, a limited experience that involves �20pediatric implants in the last 3 years in the United States hasbeen accumulated on the device. Because it is available in awide range of sizes, the Excor VAD provides circulatorysupport options for pediatric patients ranging from 2.5-kginfants to adolescents. The device is intended to addresssevere ventricular failure resulting from acute fulminantmyocarditis, cardiomyopathy, postcardiotomy failure, post-transplantation graft failure, and end-stage congenital heartdisease.29–31 This extracorporeal device utilizes cannulascovered with a polyester velour coating at the site of skinentry to minimize the risk of infections. Patients supported bythe Excor VAD have been managed with warfarin sodiumanticoagulation in the majority of cases.29–31

The NHLBI Pediatric Circulatory Support Program

Overview of the ProgramRecognizing the limitations of circulatory support devices forsmall children and the common complications of infection,thromboembolism, and cannulation problems, the NHLBIissued a Broad Agency Announcement (BAA) entitled “Pe-diatric Circulatory Support” in November 2002. This BAAsolicited contract proposals to develop novel circulatorysupport systems for infants and children with congenital andacquired cardiovascular disease who experience cardiopul-monary failure and circulatory collapse. The types of devicesintended to be developed under the BAA included left andright VADs, ECMO systems, and other novel bioengineeredsystems for children ranging in weight from 2 to 25 kg.Among the technical requirements detailed in the solicitationwere that each system should: (1) be able to be routinelydeployed and functioning in less than 1 hour after thedecision to initiate support; (2) minimize the priming vol-umes; (3) include cannulation strategies to accommodatepotential variations in patient anatomy; (4) minimize expo-sure to blood products; (5) minimize risks of infection,bleeding, hemolysis, and thrombosis; and (6) be capable ofproviding support for up to 6 months, depending on theintended application.

Evaluation Criteria, Review, and AwardsThirteen proposals were submitted to NHLBI and subse-quently reviewed by a panel of scientists with expertise inpediatric cardiology, pediatric cardiac surgery, biomedicalengineering, biocompatibility, and biostatistics. To objec-

tively rank the proposals with regard to their potential tosuccessfully develop the proposed devices while meeting theclinical and performance goals, each proposal was evaluatedon the following criteria: (1) concept and theory; (2) innova-tion; (3) research, design, development, and manufacturing;(4) plan for testing and evaluation; and (5) capabilities ofofferor. Per the BAA, secondary consideration of the costs ofindividual projects was made to fund an overall program thatprovided the best value to the government.

On the basis of the evaluation of the proposals, 5 contractswere awarded by the NHLBI to provide a comprehensiveprogram for the development of a family of novel pediatriccirculatory support devices. The 5 projects awarded includean implantable mixed-flow VAD designed specifically forpatients up to 2 years of age (the PediaFlow VAD), anothermixed-flow VAD that can be implanted intravascularly orextravascularly depending on patient size (the PediPump),compact integrated pediatric cardiopulmonary assist systems(the pCAS), apically implanted axial-flow VADs (the Pedi-atric Jarvik 2000 Flowmaker), and pulsatile-flow VADs (thePenn State PVAD). The 5 contracts totaling $22 399 727 forthe 5-year program duration were awarded on March 30,2004. The pediatric circulatory support devices funded by theNHLBI contracts are described below.

Descriptions of the Programs Supported by theNHLBI’s Pediatric Circulatory Support Program

The PediaFlow VAD (University of Pittsburgh)The PediaFlow VAD (Figure 1) is an implantable, magneti-cally suspended mixed-flow turbodynamic blood pump beingdeveloped by a consortium consisting of the University ofPittsburgh, Carnegie Mellon University, Children’s Hospitalof Pittsburgh, LaunchPoint, LLC, and World Heart Corpora-

Figure 1. Left ventricular placement of PediaFlow system. Thedevice is a mixed-flow turbodynamic blood pump for patientsup to 2 years old.

Baldwin et al The NHLBI Pediatric Circulatory Support Program 149

by guest on February 5, 2015http://circ.ahajournals.org/Downloaded from

tion. It is being designed and developed to provide chronic (6months) circulatory support to patients from birth to 2 yearsof age (3 to 15 kg body weight) with congenital or acquiredheart disease. Anticoagulation therapy while patients are onthe device is intended to be limited to antiplatelet medications(with the option for warfarin sodium if clinically indicated).The device is being designed to provide a flow rate rangefrom 0.3 to 1.5 L/min and will be limited to a maximumweight of 30 g, a maximum volume of 5 mL, and a maximumpriming volume 0.5 mL to help achieve the goal of beingfully implantable. Only a single percutaneous lead crossingthe skin for energy transmission will be required. ThePediaFlow VAD’s “smart” sensor-based controller will con-tinuously monitor cardiac status for potential “bridge-to-recovery” applications. The controller will also continuouslymonitor the performance of the PediaFlow device and supporteither continuous or pulsatile flow modes. The PediaFlowcannula sets will be designed to be suitable for both left andright ventricle cannulation.

The PediaFlow pump is based on rotary blood pumptechnology. The device blood flow path will be designed tooptimize hemocompatibility; minimize shear-induced bloodtrauma, flow separation, and stagnation regions; and avoidcavitation. The PediaFlow pump is intended to be implantedin the left upper abdomen, in the anterior abdominal wallbehind the left rectus abdominus muscle. As shown in Figure1, inflow to the pump is through the left ventricular apex, withoutflow to the ascending aorta. The same pump may be usedfor either left or right heart assistance, or 2 pumps may beused together to provide biventricular support. The pumplocations will allow easy access for component change in theevent of pump malfunction or for removal in case ofventricular recovery, thus obviating the need for a repeatsternotomy.

The ultimate safety and efficacy of the PediaFlow VADwill depend on the hemocompatibility of the flow path. Thisis particularly challenging considering the program’s goal forminimal anticoagulation and a wide range of operation. ThePediaFlow design will be optimized with advanced 3D designmethods and experimental and computational fluid dynamics(CFD) analysis that will incorporate new blood-damagemodels based on neonatal blood rheology and coagulation. Tothis end, a blood shearing instrument has been developed bythe investigators to simulate human hemodynamic conditionsexpected with the PediaFlow pump and to refine blood-damage models.32

The blood-contacting surface plays a significant role in thebiocompatibility of the pump. Thus, in addition to theTi6Al4V ELI and commercially pure titanium alloys currentlyused in most rotary VADs, other materials are being consid-ered. These include injection-moldable ceramics and poly-mers, as well as alternative titanium alloys, such as Ti6Al7Nb.Injection-moldable materials may offer manufacturability andcost advantages over machined titanium alloys. Over 30potential coatings and surface treatments have been identifiedfor application to the base material selected from the optionsmentioned above to enhance biocompatibility.

In vivo evaluation of the PediaFlow VAD will be con-ducted in the pediatric ovine model. Initially, 15-kg lambs

will be used, with progressively smaller animals to be usedfor subsequent studies to simulate the intended clinicalpopulation. A number of assays to measure hemolysis,erythrocyte fragility, blood cell activation, thrombosis, andinflammation will be applied in these studies to assessPediaFlow hemocompatibility in addition to standard mea-sures of end-organ function and cellular counts.

The PediPump (The Cleveland Clinic Foundation)The PediPump VAD is a magnetic bearing–supported, rotarydynamic circulatory support pump designed specifically forchildren.33 The enabling technology for the PediPump isderived from an adult “catheter” pump in development byThe Cleveland Clinic’s Department of Biomedical Engineer-ing and Foster-Miller Technologies, Inc. Because its capabil-ities are packaged into a small size, the resulting basic pumpdesign provides support for the entire range of patient sizesencountered in pediatrics.

The pump rotating assembly consists of an impeller in thefront, front and rear radial magnetic bearings, and a motorrotor magnet in its center (Figure 2). Blood enters axially atthe inlet and is turned to exit the pump at an intermediateangle through the pump’s outside diameter. The use ofpassive, radial magnetic bearings to support the rotor isexpected to result in exceptional device durability. Titaniumshells seal all potentially corrodible components from bloodand tissue. The initial PediPump prototype measures approx-imately 7�75 mm with a priming volume of 0.6 mL,imparting �10% of the physical displacement of currentlyavailable axial flow pumps.

Two configurations are currently envisioned for deploy-ment of the PediPump based on patient size. For largerchildren (�15 kg), the small size of the PediPump may allowcompletely intravascular implantation (Figure 3A). Forsmaller children (�15 kg), extravascular, intracorporeal im-plantation may be performed by standard cannulation strate-gies used for existing axial flow pumps, with inflow andoutlet cannulas configured as needed (Figure 3B). The samebasic pump is anticipated for use as a right VAD, LVAD, ora biventricular assist device.

One aim of the PediPump Program is to determine thebasic engineering requirements for hardware and controllogic, including design analysis for system sizing, establish-ment of the analytical model to evaluate control concepts, andbench testing of prototypes. Initial hydraulic designs for the

Figure 2. Diagram of the PediPump demonstrating centralmotor magnet, impeller, and radial magnetic bearings. Bloodenters axially at the inlet and is then turned in the impeller toexit the pump at an intermediate angle through the pump outershell (blood flow path is demonstrated by the arrows). Reprintedwith permission from The Cleveland Clinic Foundation. Copy-right 2005.

150 Circulation January 3/10, 2006

by guest on February 5, 2015http://circ.ahajournals.org/Downloaded from

first intravascular and extravascular prototypes have alreadybeen generated and are being used to optimize impellerdesign with solid 3D computer-aided design modeling andCFD studies. The best-performing impeller and stator havebeen manufactured in stereolithography, and their combinedperformance has been evaluated in a hydraulic test stage withan externalized motor. Design of test loops for pump proto-types is also proceeding. The magnetic bearing design hasbeen reviewed and detailed for the prototype. Device assem-bly will begin shortly after the finishing details of the thrustbearings are finalized on the basis of the durability testingresults.

The second aim of the program is to perform preclinicalanatomic fitting studies using computed tomography–based3D modeling. Archived computed tomography scans fromchildren with congenital heart disease have been convertedinto 3D digital models with Mimics software (Materialise).These digital renderings have been used subsequently togenerate to-scale physical models by a variety of rapidprototyping techniques. Standard stereolithography methodshave been used to create rigid physical models, whereasflexible physical models have been generated with a 3Dprinter, ZPrinter 310 System (Z Corporation), that utilizesstarch-based powder to create rigid, thin-walled physicalmodels derived from the on-screen renderings. The accuratemodeling provided by these tools is expected to enablepreoperative planning on a case-by-case basis, which will beof particular importance for patients with anatomic abnormal-ities of the great arteries and veins.

The last aim of the program is to perform animal implan-tations for characterization and reliability testing of thedevice. Careful assessment of the physiological impact of the

PediPump and the ensuing host response will be determinedfrom the studies. Animal implantations will begin after theongoing basic engineering and anatomic studies.

Pediatric Cardiopulmonary Assist System (Ension, Inc)Ension’s pCAS is being developed in collaboration betweenEnsion, Inc, the University of Louisville, Seare BioMatrixSystems, and Fluent, Inc. The goal of this program is toimprove on the current state of ECMO by decreasing clinicalcomplications of this therapy and improving the possibilityfor parent-child contact and bonding during support.

The blood-contacting component of the pCAS, shown inFigure 4, is based around a compact, paracorporeal rotaryflow device capable of simultaneously pumping blood andproviding oxygenation. The device rotor is fabricated fromlayers of microporous hollow fibers that include a customcoating to both increase fiber life and minimize requirementsfor systemic anticoagulation.

The paracorporeal design of the pCAS affords modularity,which allows a separate device for each target patientpopulation and the possibility of device exchange during thesupport period. Such modularity also permits the patient to be“upgraded” from the neonatal-sized device to the child-sizeddevice as growth or support needs change. In addition tosteady flow, the pCAS will include an option to deliverpulsatile flow at the higher heart rates of the intended patientpopulation.

The hardware and control schemes of the pCAS willregulate both pump flow and oxygenation on the basis ofinputs from a suite of sensors. These sensors include anoff-the-shelf oxygen analyzer integrated with a custom-designed, photo-acoustic carbon dioxide analyzer developedat Ension. An ultrasonic-based blood flow sensor will be usedin conjunction with a high-fidelity pressure transducermounted distal to the device to provide input needed for theblood flow control scheme. Because oxygenation is directly

Figure 3. A, Intravascular PediPump deployed as a biventricularassist device for patients weighing �15 kg. Pumps are demon-strated placed across both semilunar valves to position the inletbelow valve level while blood is ejected above the valve to pro-vide biventricular support. B, Extravascular, intracorporealPediPump deployed as a biventricular assist device for patientsweighing �15 kg. Biventricular support is provided by inlet can-nulation of the right and left ventricles with outlet cannulation ofthe corresponding great artery. Corresponding LVAD and rightVAD deployment of either the intravascular or extravascular ver-sions would be achieved by removing the pump from theunsupported ventricle. Reprinted with permission from TheCleveland Clinic Foundation. Copyright 2005.

Figure 4. The Ension pCAS system. The blue and red arrowsshow the blood inflow and outflow paths, respectively. Gasexchange occurs in the rotor’s microporous hollow fibers.

Baldwin et al The NHLBI Pediatric Circulatory Support Program 151

by guest on February 5, 2015http://circ.ahajournals.org/Downloaded from

dependent on blood flow, a complex control algorithm toregulate this interdependency will be developed.

Ension and its partners are also conducting research en-compassing (1) rheological characterization of human andanimal neonatal and pediatric blood; (2) computational mod-eling of blood flow, mass exchange, and blood damage withinthe pCAS device; and (3) in vitro and animal testing includ-ing inflammatory response assays. Researchers at the Univer-sity of Louisville are focusing on rheological characterizationas the basis of an improved fluid model for computationaldesign work and the functional evaluation of pCAS proto-types. This work includes in vitro evaluations in a mockcirculatory loop approximating the anatomic and hemody-namic features of a 1-year old infant, while also allowing theoption of different cannulation sites and sizes, and thecreation of ventricular failure conditions for the evaluation. Inaddition, the University of Louisville will also be responsiblefor all facets of the comprehensive acute and chronic animalevaluations of the pCAS device.

Fluent, Inc, a flow-modeling company with relevant expe-rience in biological flows, is performing the CFD modeling ofthe pCAS device for Ension. The goal of this work is to usea CFD model as a design tool that may be leveraged to guiderefinement of pCAS design iterations. The team’s develop-ment program focuses first on calculation and validation ofthe blood flow field, forming a foundation for subsequentoxygenation, decarbonation, and blood-damage calculations.

Seare Biomatrix Systems, Inc, a company that specializesin surface modifications to promote healthy device-tissueinterface, has partnered with Ension to texture the exteriorcannula surface used in the pCAS application. The spatiallyspecific, porous material with an integrated biologic compo-nent (SeareMatrix) is intended to promote the development ofa soft tissue in-growth with neovascularity. This tissue-engineered interface should provide anchoring of the cannu-las and an infection-resistant barrier while being easy tosurgically dissect at the time of cannula removal. Earlyresearch efforts will focus on optimizing the interface mate-rial porosity for the size of the cannulas to be used forpercutaneous placement in children.

The Pediatric Jarvik 2000 (Jarvik Heart)Two models of a Pediatric Jarvik 2000 axial-flow bloodpump are being developed by Jarvik Heart, Inc in collabora-tion with the University of Maryland, Mississippi StateUniversity, and Whalen Biomedical, Inc. The design of thePediatric Jarvik 2000 is based on the Jarvik 2000 VAD,which has been implanted in more than 100 patients in theUnited States and Europe over the past 5 years. During thefirst 3 years of the program, the work will be focused ondeveloping and testing a child-size pump for long-termsupport in children of �15 to 25 kg. The child-size model isa 35-g, 10-mL miniaturized version of the Jarvik 2000 bloodpump. In years 4 and 5 of the program, the work will shifttoward developing a still smaller 12-g, 4-mL infant-sizeblood pump. The infant-size model is intended for 3- to 15-kgchildren. These small child and infant Jarvik 2000 models areintended to be versatile. These devices, through variouscreative surgical techniques, are being designed to be im-

plantable in any of the normal heart’s 4 chambers to providethe chronic mechanical left, right, or biventricular supportneeded by children with compromised cardiac function.

Prototypes of the child and infant Jarvik 2000 models areshown in Figure 5 with the adult Jarvik 2000 VAD. Despitethe similarity in appearance, the pediatric devices will notsimply be scaled-down models of the adult device. Thepediatric models require new blade designs for the lower flowand pressure requirements of children and infants. Hydrody-namic test results (Figure 6) demonstrate that prototype infantand child-size Jarvik 2000 models provide desired cardiacoutputs while functioning under expected operating condi-tions. These results have been confirmed in initial animal testresults that showed that the child model pumps approximately1 to 2.5 L/min. The animal study also revealed that the pumpfits well in the apex of small sheep, the animal model selectedto simulate the size and anatomy of a 15- to 25-kg child.

Figure 5. The adult, child, and infant Jarvik 2000 devices. Theinfant model is still at the conceptual prototype stage ofdevelopment.

Figure 6. In vitro flow/pressure curves of the child-size Jarvik2000.

152 Circulation January 3/10, 2006

by guest on February 5, 2015http://circ.ahajournals.org/Downloaded from

Further pump design refinements will be made usingadvanced CFD methods together with flow visualization andex vivo hemolysis testing to achieve acceptably low blooddamage in these very small pumps. Also as part of the project,a small microprocessor-based programmable control systemwill be developed with consideration of human factors so thatit can be worn by children with adult supervision.

The encouraging clinical performance of the adult Jarvik2000 VAD has resulted in high expectations for the PediatricJarvik 2000 models. The first destination-therapy recipient ona Jarvik 2000 VAD remains in excellent health after 5 yearsof circulatory assistance, the longest of any patient with anysingle heart-assist device. To date, no bearing failures haveoccurred in any patient, the few patients who experienceddevice-related infections were all successfully treated, and theincidences of thrombus, thromboembolism, and hemolysishave been acceptably low.34 Given the design and clinicalresults to date for the adult device, the reliability of thepediatric Jarvik 2000 VADs is targeted to exceed 10 years.

Pediatric Ventricular Assist Device (Penn State)The Penn State Pediatric Ventricular Assist Device (PVAD),is a pulsatile, pneumatically actuated blood pump based onthe design principles of the adult-sized Pierce-Donachy VAD,which was developed at Penn State and is now known as theThoratec VAD, a product of the Thoratec Corporation.35 ThePVAD, which underwent initial development in 1986 underthe direction of William S. Pierce, MD, in collaboration with3M Corporation, is intended to be used for left, right, orbiventricular support for up to 6 months. The device isintended primarily for paracorporeal placement (Figure 7) butwill also be implantable for bridge-to-transplantation appli-cations.36 The PVAD (Figure 8) is being developed in 2 sizes:a 12-mL dynamic stroke volume size for infants ranging inweight from 5 to 15 kg and a 25-mL stroke volume size forchildren 15 to 35 kg in weight. The flow rate ranges of theinfant- and child-size devices are 0.5 to 1.3 L/min and 1.3 to

3.3 L/min, respectively. Each of the pumps will consist of aseamless segmented polyether polyurethaneurea (SPEUU)blood sac positioned within a rigid titanium case, mechanicalheart valves, and cannula connectors. Both bileaflet andmonoleaflet mechanical heart valves in the 17- to 21-mm sizerange for the infant-size device and the 19- to 23-mm sizerange for the child-size device will be evaluated for use in thedevices. The pump shape will encourage the development ofa vortex flow pattern, which will efficiently maintain highwall shear rates to help prevent thrombus formation. ThePVAD will be controlled by a new portable biventricularpneumatic driver, which will be developed in collaborationwith Minnetronix, Inc based on actuator technology devel-oped by Penn State for electrically powered implantableblood pumps.

Device biocompatibility of the PVAD systems is anotherarea of focus. In addition, using advanced fluid dynamictechniques to identify areas that are prone to thrombusdeposition, hemolysis, or platelet activation, surface-modifiedSPEUU materials (Polymer Technology Group) are beinginvestigated in an effort to increase thromboresistance with-out compromising fatigue life. Polymer surface morphologywill be measured by scanning electron micrography, and bulkmechanical properties will be measured by dynamic mechan-ical analysis.37 In vitro methods of assessing blood compat-ibility are being used to supplement animal experiments.Platelet adhesion and reactivity under a range of shear rates ismeasured with a rotating disk system with immunofluores-cent labeling of the cell membrane protein CD41, the �IIb

integrin chain of the glycoprotein IIb/IIIa complex. Furthertests that assess the activity of biomaterial-adherent plateletsin supporting formation of the coagulation complexes are alsobeing developed. Material-induced activation of the intrinsicpathway of blood coagulation is being assessed by an in vitroassay that looks at the response of plasma to procoagulantmaterial stimuli. Small tubes for this testing will be fabricatedfrom the test polymers using the same procedures by whichblood sacs are fabricated. Coagulation, measured by time toformation of a fibrin clot, will be assessed by testing thesamples in recalcified, citrated human plasma.

During the development of the PVAD devices, animalstudies will be performed to evaluate hemodynamic perfor-mance and biological effects, primarily thrombogenesis and

Figure 7. The Penn State PVAD, shown in the paracorporealplacement with left atrial-to-aorta cannulation.

Figure 8. Penn State PVADs are designed to produce rotationalblood flow patterns within the pump chambers. The 12-mLinfant-size VAD is shown.

Baldwin et al The NHLBI Pediatric Circulatory Support Program 153

by guest on February 5, 2015http://circ.ahajournals.org/Downloaded from

hemolysis. The 12-mL PVAD will be evaluated in 5- to 15-kggoats and the 25-mL PVAD in 15- to 35-kg goats. Pumps willbe implanted in the preperitoneal space via a left thoracot-omy, with cannulation of the left ventricle and descendingaorta. Hematologic studies will be performed to detectalterations to blood-formed elements, evidence of activationof the hemostatic system, and hemolysis. Thrombus forma-tion on the blood sac will be assessed by (1) postexplantationgross examination, (2) histological analysis, and (3) multi-scale surface analysis (including scanning electron micros-copy, immunofluorescent labeling of platelets and fibrinogen,and atomic force microscopy). Embolization will be assessedby clinical laboratory values throughout the study and exam-ination of the major organs at autopsy for evidence ofinfarctions.

Future DirectionsThe objective of the NHLBI Pediatric Circulatory SupportProgram is to support the development of the family ofpediatric MCSDs described above so that these novel deviceswill eventually provide effective options for clinical therapy.As evidenced in their descriptions, the PediPump, PediatricJarvik 2000 Flowmaker, PediaFlow VAD, Penn State PVAD,and pCAS each have unique characteristics to address thebiocompatibility, cannulation, control, capacity, infection,reliability, and deployment issues specific to pediatric pa-tients with congenital or acquired heart disease. As a result,each of the devices is expected to have unique applicationsbased on patient size, expected duration of support, intentionof therapy as a bridge to transplantation or to aid in recovery,and type of heart disease. Together, these 5 MCSDs areexpected to provide the missing effective options for thevulnerable, small pediatric patients who are currently limitedto short-term extracorporeal VADs and cumbersome conven-tional ECMO. Clinical evaluations are expected to begin at orbefore the conclusion of the development program in 2009.

AcknowledgmentsPlease see the online-only Data Supplement at http://circ.ahajournals.org/cgi/content/full/113/1/147/DC1.

DisclosuresDrs Baldwin and Hoke are employees of NHLBI and are the ProjectOfficers for NHLBI’s Pediatric Circulatory Support Program. DrBorovetz is the Principal Investigator (PI) for the contract awarded tothe University of Pittsburgh. The University of Pittsburgh licensestechnology to WorldHeart, Inc. Dr Borovetz does not receive anyproceeds from the licenses. Dr Duncan is the PI for the contractawarded to The Cleveland Clinic and is a member of the AdvisoryBoard for the DeBakey VAD Child, a product of MicroMedTechnologies, Inc. Mr Gartner is the PI for the contract awarded toEnsion, Inc, the company that intends to commercialize the pCAS.He is also President and retains ownership interest in Ension, Inc. DrJarvik is the PI for the contract awarded to Jarvik Heart, Inc and isPresident and a shareholder of Jarvik Heart, Inc. Dr Weiss is the PIfor the contract awarded to The Pennsylvania State University.

References1. Duncan BW. Mechanical circulatory support for infants and children with

cardiac disease. Ann Thorac Surg. 2002;73:1670–1677.2. Fuchs A, Netz H. Ventricular assist devices in pediatrics. Images Paediatr

Cardiol. 2002;9:24–54.

3. AHA 2005 Statistical Reference Book. Dallas, Tex: American HeartAssociation; 2005.

4. Boneva RS, Botto LD, Moore CA, Yang Q, Correa A, Erickson JD.Mortality associated with congenital heart defects in the United States:trends and racial disparities, 1979 –1997. Circulation. 2001;103:2376–2381.

5. Lipshultz SE, Sleeper LA, Towbin JA, Lowe AM, Orav EJ, Cox GF,Lurie PR, McCoy KL, McDonald MA, Messere JE, Colan SD. Theincidence of pediatric cardiomyopathy in two regions of the UnitedStates. N Engl J Med. 2003;348:1647–1655.

6. Boucek MM, Edwards LB, Keck BM, Trulock EP, Taylor DO, Hertz MI.Registry for the International Society for Heart and Lung Transplantation:seventh official pediatric report: 2004. J Heart Lung Transplant. 2004;23:933–947.

7. The Organ and Transplantation Procurement Network Web site 2005.Available at: http://www.optn.org. Accessed May 16, 2005.

8. Goldman AP, Cassidy J, de Leval M, Haynes S, Brown K, Whitmore P,Cohen G, Tsang V, Elliott M, Davison A, Hamilton L, Bolton D, WrayJ, Hasan A, Radley-Smith R, Macrae D, Smith J. The waiting game:bridging to paediatric heart transplantation. Lancet. 2003;362:1967–1970.

9. Black MD, Coles JG, Williams WG, Rebeyka IM, Trusler GA, Bohn D,Gruenwald C, Freedom RM. Determinants of success in pediatric cardiacpatients undergoing extracorporeal membrane oxygenation. Ann ThoracSurg. 1995;60:133–138.

10. Chen YS, Yu HY, Huang SC, Chiu KM, Lin TY, Lai LP, Lin FY, WangSS, Chu SH, Ko WJ. Experience and result of extracorporeal membraneoxygenation in treating fulminant myocarditis with shock: whatmechanical support should be considered first? J Heart Lung Transplant.2005;24:81–87.

11. Duncan BW. Mechanical circulatory support in infants and children withcardiac disease. In: Zwischenberger JB, Bartlett RH, eds. ECMO Extra-corporeal Cardiopulmonary Support in Critical Care. 2nd ed. Ann Arbor,Mich: Extracorporeal Life Support Organization; 2000.

12. Duncan BW, Bohn DJ, Atz AM, French JW, Laussen PC, Wessel DL.Mechanical circulatory support for the treatment of children with acutefulminant myocarditis. J Thorac Cardiovasc Surg. 2001;122:440–448.

13. Grinda JM, Chevalier P, D’Attellis N, Bricourt MO, Berrebi A, GuibourtP, Fabiani JN, Deloche A. Fulminant myocarditis in adults and children:bi-ventricular assist device for recovery. Eur J Cardiothorac Surg. 2004;26:1169–1173.

14. Walters HL III, Hakimi M, Rice MD, Lyons JM, Whittlesey GC, KleinMD. Pediatric cardiac surgical ECMO: multivariate analysis of riskfactors for hospital death. Ann Thorac Surg. 1995;60:329–337.

15. Sidiropoulos A, Hotz H, Konertz W. Pediatric circulatory support.J Heart Lung Transplant. 1998;17:1172–1176.

16. Duncan BW, Hraska V, Jonas RA, Wessel DL, Del Nido PJ, Laussen PC,Mayer JE, Lapierre RA, Wilson JM. Mechanical circulatory support inchildren with cardiac disease. J Thorac Cardiovasc Surg. 1999;117:529–542.

17. Throckmorton AL, Allaire PE, Gutgesell HP, Matherne GP, Olsen DB,Wood HG, Allaire JH, Patel SM. Pediatric circulatory support systems.ASAIO J. 2002;48:216–221.

18. Ashton RC Jr, Oz MC, Michler RE, Champsaur G, Catanese KA, HsuDT, Addonizio LJ, Quaegebeur JM. Left ventricular assist device optionsin pediatric patients. ASAIO J. 1995;41:M277–M280.

19. Helman DN, Addonizio LJ, Morales DL, Catanese KA, Flannery MA,Quagebeur JM, Edwards NM, Galantowicz ME, Oz MC. Implantable leftventricular assist devices can successfully bridge adolescent patients totransplant. J Heart Lung Transplant. 2000;19:121–126.

20. Korfer R, El-Banayosy A, Arusoglu L, Minami K, Korner MM, Kizner L,Fey O, Schutt U, Morshuis M, Posival H. Single-center experience withthe Thoratec ventricular assist device. J Thorac Cardiovasc Surg. 2000;119:596–600.

21. McBride LR, Naunheim KS, Fiore AC, Moroney DA, Swartz MT.Clinical experience with 111 Thoratec ventricular assist devices. AnnThorac Surg. 1999;67:1233–1239.

22. Wieselthaler GM, Schima H, Hiesmayr M, Pacher R, Laufer G, Noon GP,DeBakey M, Wolner E. First clinical experience with the DeBakey VADcontinuous-axial-flow pump for bridge to transplantation. Circulation.2000;101:356–359.

23. Frazier OH, Myers TJ, Jarvik RK, Westaby S, Pigott DW, Gregoric ID,Khan T, Tamez DW, Conger JL, Macris MP. Research and developmentof an implantable axial-flow left ventricular assist device: the Jarvik 2000Heart. Ann Thorac Surg. 2001;71:S125–S132.

154 Circulation January 3/10, 2006

by guest on February 5, 2015http://circ.ahajournals.org/Downloaded from

24. Griffith BP, Kormos RL, Borovetz HS, Litwak K, Antaki JF, Poirier VL,Butler KC. HeartMate II left ventricular assist system: from concept tofirst clinical use. Ann Thorac Surg. 2001;71:S116–S120.

25. Morales DL, Dibardino DJ, McKenzie ED, Heinle JS, Chang AC, LoebeM, Noon GP, Debakey ME, Fraser CD Jr. Lessons learned from the firstapplication of the DeBakey VAD Child: an intracorporeal ventricularassist device for children. J Heart Lung Transplant. 2005;24:331–337.

26. ECLS Registry Report. Ann Arbor, Mich: Extracorporeal Life SupportOrganization; January 2005.

27. Conrad SA, Rycus PT, Dalton H. Extracorporeal Life Support RegistryReport 2004. ASAIO J. 2005;51:4–10.

28. Thiagarajan RR, Nelson DP. Should we be satisfied with currentoutcomes for cardiac extracorporeal life support? Pediatr Crit Care Med.2005;6:89–90.

29. Hetzer R, Loebe M, Potapov EV, Weng Y, Stiller B, Hennig E, Alexi-Meskishvili V, Lange PE. Circulatory support with pneumatic paracor-poreal ventricular assist device in infants and children. Ann Thorac Surg.1998;66:1498–1506.

30. Hetzer R, Loebe M, Weng Y, Alexi-Meskhishvili V, Stiller B. Pulsatilepediatric ventricular assist devices: current results for bridge to transplan-tation. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu. 1999;2:157–176.

31. Stiller B, Dahnert I, Weng YG, Hennig E, Hetzer R, Lange PE. Childrenmay survive severe myocarditis with prolonged use of biventricular assistdevices. Heart. 1999;82:237–240.

32. Wu J, Antaki JF, Snyder TA, Wagner WR, Borovetz HS, Paden BE.Design optimization of blood shearing instrument by computational fluiddynamics. Artif Organs. 2005;29:482–489.

33. Duncan BW, Lorenz M, Kopcak MW, Fukamachi K, Ootaki Y, ChenHM, Chapman PA, Davis JJ, Smith WA. The PediPump: a new ventric-ular assist device for children. Artif Organs. 2005;29:527–530.

34. Frazier OH, Shaw AS, Myers TJ, Robertson KD, Gregoric ID, DelgadoR. Use of the Flowmaker (Jarvik 2000) left ventricular assist device fordestination therapy and bridging to transplantation. Cardiology. 2004;101:111–116.

35. Pierce WS, Rosenberg G, Donachy JH, Landis DL, Weiss W, Hensley F,Larach D. Postoperative cardiac support with a pulsatile assist pump:techniques and results. Artif Organs. 1987;11:247–251.

36. Daily BB, Pettitt TW, Sutera SP, Pierce WS. Pierce-Donachy pediatricVAD: progress in development. Ann Thorac Surg. 1996;61:437–443.

37. Liu Q, Runt J, Felder G, Rosenberg G, Snyder AJ, Weiss WJ, Lewis J,Werley T. In vivo and in vitro stability of modified poly(urethaneurea)blood sacs. J Biomater Appl. 2000;14:349–366.

Baldwin et al The NHLBI Pediatric Circulatory Support Program 155

by guest on February 5, 2015http://circ.ahajournals.org/Downloaded from

William J. Weiss and Tracey R. HokeJ. Timothy Baldwin, Harvey S. Borovetz, Brian W. Duncan, Mark J. Gartner, Robert K. Jarvik,

The National Heart, Lung, and Blood Institute Pediatric Circulatory Support Program

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 2006 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/CIRCULATIONAHA.105.571422

2006;113:147-155Circulation. 

http://circ.ahajournals.org/content/113/1/147World Wide Web at:

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

http://circ.ahajournals.org/content/suppl/2006/02/02/113.1.147.DC1.htmlData Supplement (unedited) at:

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer this process is available in the

click Request Permissions in the middle column of the Web page under Services. Further information aboutOffice. Once the online version of the published article for which permission is being requested is located,

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the EditorialCirculationin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on February 5, 2015http://circ.ahajournals.org/Downloaded from