Does the Ventrica magnetic vascular positioner (MVP) for coronary artery bypass grafting significantly alter local fluid dynamics? A numeric study.

OBJECTIVE: Automatic devices have been recently introduced to make the anastomosis procedure quick and efficient when creating a coronary bypass on the beating heart. However, the implantation of these devices could modify the graft configuration, consistently affecting the hemodynamics usually found in the traditional anastomosis. As local fluid dynamics could play a significant role in the onset of vessel wall pathologies, in this article a computational approach was designed to investigate flow patterns in the presence of the Ventrica magnetic vascular positioner (Ventrica MVP) device. METHODS: A model of standard hand-sewn anastomosis and of automated magnetic anastomosis were constructed, and the finite volume method was used to simulate in silico realistic graft hemodynamics. Synthetic analytical descriptors -- i.e., time-averaged wall shear stress (TAWSS), oscillating shear index (OSI) and helical flow index (HFI) -- were calculated and compared for quantitative assessment of the anastomosis geometry hemodynamic performance. RESULTS: In this case study, the same most critical region was identified for the 2 models as the one with the lowest TAWSS and the highest OSI (TAWSS=0.229, OSI=0.255 for the hand-sewn anastomosis; TAWSS=0.297, OSI=0.171 for the Ventrica MVP(R)). However, the shape of the Ventrica MVP does not induce more critical wall shear stresses, oscillating flow and damped helicity in the graft fluid dynamics, as compared with conventional anastomosis. CONCLUSIONS: We found that the use of the Ventrica MVP for the case study under investigation was not associated with more critical fluid dynamics than with conventional hand-sewn anastomosis. Thereby, the device could facilitate beating heart and minimally invasive coronary artery bypass grafting without increasing local hemodynamic-related risks of failure.

A new electro-mechanical bioreactor for soft tissue engineering.

By enabling the maintenance of controlled chemical and physical environmental conditions, bioreactors proved that electro-mechanical stimulation improves tissue development in vitro, especially in the case of tissues which are subjected to stimuli during embryogenesis and growth (i.e. skeletal and cardiac muscle tissue). However, most of the bioreactors developed in the last 20 yrs, designed to suit specific applications, lack versatility. With the aim to provide researchers with a yielding, versatile tool, we designed and realized in this study an electro-mechanical stimulator capable of dynamically culturing four biological constructs, delivering assignable stretching and electrical stimulation patterns. The device has been conceived to be easy to handle and customizable for different applications, while ensuring sterility along with stimuli delivery. The gripping equipment, modular and adaptable to scaffolds of different consistencies, is provided with dedicated tools for supporting sample insertion into the culture chamber performed under a laminar flow hood. As to performance, a wide range of electro-mechanical stimulation patterns and their relative occurrence can be accomplished, permitting the adjustment of the dynamic culture parameters both to the specific cell species and to the developmental phase of the cultured cells.

Endothelial cell adhesion force estimation at the nanoscale.

Endothelial cell adhesion to a synthetic surface includes a definite set of molecular interactions. Cell adhesion is managed by fibronectin and vitronectin in extracellular matrix (ECM) that binds the receptor site of the trans-membrane protein dimers, the integrins. These proteins contain one of the binding sites (I-like domain receptor) for the Arg-Gly-Asp (RGD) peptides that are the established adhesion receptor sites in the ECM. A molecular approach can quantify the adhesion strength by ligand-receptor force computation. The molecular interaction energy between a polyethylene (PE) surface covalently grafted with the adhesion sites (RGD-containing) and trans-membrane integrin receptor (I-domain) was evaluated through a molecular model of a single ligand-receptor complex. The aims of this work were: (i) the generation of the receptor molecular model: the I-like domain; (ii) the evaluation of the greatest binding chemical affinity between the I-like domain and three RGD-containing peptides; (iii) the development of the molecular model of crystalline lamellae of PE; and (iv) the evaluation of the interac-tion energies and the interaction force between the I-like domain and the grafted biomaterial. The calculation of the interaction energies can provide an estimation of the adhesion force of the ligand-receptor complex and, finally, of the endothelial cell adhesion force. The calculated cell adhesion force is in agreement with experimental data.(Journal of Applied Biomaterials and Biomechanics 2005; 3: 42-9).

Flow dynamics of the St Jude Medical Symmetry aortic connector vein graft anastomosis do not contribute to the risk of acute thrombosis.

BACKGROUND: The efficacy of the St Jude Medical Symmetry aortic connector (St Jude Medical, Inc, St Paul, Minn) for coronary artery bypass is currently debated. Potential drawbacks are the biocompatibility of the endoluminal device, the need for graft manipulation during the procedure, and the 90 degrees offset of the vein graft from the ascending aorta, which may induce graft kinking and abnormal fluid dynamics. In this article, a computational approach was designed to investigate the fluid dynamics pattern at the proximal graft. METHODS: Four models of hand-sewn anastomoses and two models of automated anastomoses were constructed; a finite volume technique was used to simulate realistic graft fluid dynamics, including aortic compliance and proper aortic and graft flow rates. The anastomosis geometry performance was analyzed by calculating time-averaged wall shear stress and the oscillating shear index at the toe and heel regions of the proximal graft. RESULTS: Time-averaged wall shear stress was significantly lower in the hand-sewn anastomosis models than in the two models that simulated the use of the aortic connector (0.38 +/- 0.07 Pa vs 1.32 +/- 0.4 Pa). Higher oscillating shear index values were calculated in the hand-sewn anastomosis models (0.15 +/- 0.02 Pa vs 0.06 +/- 0.02 Pa). CONCLUSIONS: Automated anastomosis geometry is associated with less critical fluid dynamics than with conventional hand-sewn anastomosis: the shape of the proximal graft induces more physiological wall shear stresses and less oscillating flow, suggesting a lower risk of atherosclerotic plaque and intimal hyperplasia as compared with conventional anastomosis geometry. Therefore, the reported early thrombosis and late failure of the St Jude Medical aortic connector anastomoses are not related to unfavorable fluid dynamics.

Myosin head mechanical performance under different conformational change mechanisms.

The present paper puts forward a mathematical approach to model the conformational changes of the myosin head due to ATP hydrolysis, which determine the head swinging and consequent sliding of the actin filament. Our aim is to provide a simple but effective model simulating myosin head performance to be integrated into the overall model of sarcomere mechanics under development at our Laboratory (J. Biomech. 34 (2001) 1607). We began by exploring myosin head mechanics in recent findings about myosin ultrastructure, morphology and energetics in order to calculate the working stroke distance (WS) and the force transmitted to the actin filament during muscle contraction. Two different working stroke mechanisms were investigated, assuming that the swinging of the myosin head occurs either as a consequence of purely conformational changes (Science 261 (1993a) 58) or by thermally driven motion (ratchet mechanism) followed by conformational changes (Cell 99 (1999) 421). Our results show that force and WS values vary markedly between the two models. The maximum force generated is about 10 pN for the first model and 31 pN for the second model, and the WSs are about 13 and 4 nm, respectively. These results are then discussed and compared with published data. The experimental data used for comparison are scarce and non-homogeneous; hence, the final remarks do not lead to definite conclusions. In any event, relatively speaking, the first model is more coherent with experimental findings.

Possible role of decorin glycosaminoglycans in fibril to fibril force transfer in relative mature tendons--a computational study from molecular to microstructural level.

Experimental studies on immature tendons have shown that the collagen fibril net is discontinuous. Manifold evidences, despite not being conclusive, indicate that mature tissue is discontinuous as well. According to composite theory, there is no requirement that the fibrils should extend from one end of the tissue to the other; indeed, an interfibrillar matrix with a low elastic modulus would be sufficient to guarantee the mechanical properties of the tendon. Possible mechanisms for the stress-transfer involve the interfibrillar proteoglycans and can be related to the matrix shear stress and to electrostatic non-covalent forces. Recent studies have shown that the glycosaminoglycans (GAGs) bound to decorin act like bridges between contiguous fibrils connecting adjacent fibril every 64-68 nm; this architecture would suggest their possible role in providing the mechanical integrity of the tendon structure. The present paper investigates the ability of decorin GAGs to transfer forces between adjacent fibrils. In order to test this hypothesis the stiffness of chondroitin-6-sulphate, a typical GAG associated to decorin, has been evaluated through the molecular mechanics approach. The obtained GAG stiffness is piecewise linear with an initial plateau at low strains (<800%) and a high stiffness region (3.1 x 10(-11)N/nm) afterwards. By introducing the calculated GAG stiffness in a multi-fibril model, miming the relative mature tendon architecture, the stress-strain behaviour of the collagen fibre was determined. The fibre incremental elastic modulus obtained ranges between 100 and 475 MPa for strains between 2% and 6%. The elastic modulus value depends directly on the fibril length, diameter and inversely on the interfibrillar distance. In particular, according to the obtained results, the length of the fibril is likely to play the major role in determining stiffness in mature tendons.

An approach to computer automation of the extracorporeal circulation.

In order to move towards extracorporeal circulation (ECC) automation, a virtual simulation of the process was designed. The ECC model is composed of a virtual patient linked to a virtual ECC circuit. A user interface panel allows to set control parameters for the simulation and to visualize results. It is possible to switch between manual and automatic control. Meaningful hemodynamic and hematochemical variables are continuously shown along with a score (from 0 to 10). The virtual model can play a crucial role in educating and training the personnel devoted to the managing of the heart-lung machine.

Myosin cross-bridge mechanics: geometrical determinants for continuous sliding.

Advances in experimental techniques have provided new details on the molecular mechanisms governing the cross-bridge kinetics. Nevertheless, the issue of micromechanics of sliding is still debated. In particular, uncertainty exists regarding the myosin filament arrangement and structure and the mechanics of the myosin head with respect to the working stroke distance (WS) and the duty ratio (r), i.e. the fraction of the ATPase cycle time the myosin head is attached to the actin filament. The object of the present work is to provide a theoretical framework to correlate different features of cross-bridge mechanics; the main hypothesis is that the attachment between the actin filament and the surrounding myosin filaments has to be continuous through the sliding (continuous sliding hypothesis) in order to maximise the effect of the myosin head performance. A 3-D model of the sliding mechanism based on a geometrical approach is presented, which is able to identify the architectures that accomplish the continuous sliding under unloaded conditions. About 200 different configurations have been simulated by changing the myosin head binding range, i.e. its ability to reach an actin binding site from its rest position, WS, the myosin head orientation and the actin filament orientation. Only few configurations were consistent with the continuous sliding hypothesis. Depending on the parameter set adopted, the percentage of attached heads (%AH) calculated ranges between 4% and 28%, r between 0.08 and 0.02s(-1), and the sliding velocity between 0.7 and 10.6 microm/s. In all the cases, results were not affected by the WS value.

The pumping oxygenator: design criteria and first in vitro results.

A new project is presented, the pumping oxygenator, functionally integrating pulsatile pumping and blood oxygenation in a single device. Solid, semipermeable silicone membranes allow gas exchange and simultaneously transfer energy from pressurized gas to blood thanks to their distensibility and to inlet and outlet 1-way valves. Two small-sized (1 m2 exchange surface area) prototypes were designed, constructed, hydraulically characterized, and subjected to gas transfer evaluation tests. Blood flow rates (Q(b)) up to 1,250 ml/min were obtained with 30 mm Hg static preload and 130 mm Hg afterload with 0.7 m upstream and 2.1 m downstream 3/8 inch pipes. Physiological oxygen transfer (VO2 = 5 ml/dl, ml of transferred O2/dl of treated blood) was delivered at Q(b) < 900 ml/min, about 4 ml/dl at Q(b) = 1,250 ml/min. VO2 also was significantly increased by increasing percent systolic time. CO2 transfer decreased regularly with increasing Q(b) from VCO2 = 4.8 ml/dl at Q(b) = 400 ml/min to VCO 2 = 2.1 ml/dl at Q(b) = 1,250 ml/min. The results confirm the possibility of integrating oxygenation and pulsatile pumping. The pumping oxygenator represents a promising project deserving further improvements.

Ventricular motion during the ejection phase: a computational analysis.

In the present paper, the study of the ventricular motion during systole was addressed by means of a computational model of ventricular ejection. In particular, the implications of ventricular motion on blood acceleration and velocity measurements at the valvular plane (VP) were evaluated. An algorithm was developed to assess the force exchange between the ventricle and the surrounding tissue, i.e., the inflow and outflow vessels of the heart. The algorithm, based on the momentum equation for a transitory flowing system, was used in a fluid-structure model of the ventricle that includes the contractile behavior of the fibers and the viscous and inertial forces of the intraventricular fluid. The model calculates the ventricular center of mass motion, the VP motion, and intraventricular pressure gradients. Results indicate that the motion of the ventricle affects the noninvasive estimation of the transvalvular pressure gradient using Doppler ultrasound. The VP motion can lead to an underestimation equal to 12.4 +/- 6.6%.

Poroelastic finite element analysis of a bone specimen under cyclic loading.

It had been suggested that the fluid embodied in bone lacunar-canalicular porosity may play an important role in bone remodelling [Weinbaum et al., 1994. Journal of Biomechanics 27, 339-360]. In this paper a finite element model of a poroelastic prismatic solid of rectangular cross-section is considered to simulate bone behaviour, precisely as in the previous work by Zhang and Cowin [Zhang and Cowin, 1994. Journal of Mechanical Physics of Solids 42, 1575-1599]. This solid is subject to combined cyclic axial and bending loads at its end. The objectives of the study are: (1) to verify the accuracy of the simplifying hypotheses underlying the analytical solutions established by the above authors; (2) to provide further insight into the behaviour of that solid; (3) to test the advantages in generality and versatility and the computing costs of general-purpose finite element codes in poroelastic analysis. The study is parametric with respect to the fluid leakage coefficient, to the ratio of the bending moment and axial load, and to the ratio of the characteristic relaxation time of the pore pressure over the excitation period. Results show that, for all the cases considered, the pore pressure distribution along the section height of the poroelastic beam exhibits a very good matching with previous analytical results. Stresses transversal with respect to the beam axis (assumed as constant or zero in previous analytical solutions) are evaluated. The analysis pointed out that: (1) the effects due to end-loads with zero resultants practically extinguish within a distance from the beam end almost equal to a typical length of the cross-section; (2) cross-sections remain plane above that distance; (3) the transversal total stresses are three orders of magnitude lower than axial stress.

Intraventricular pressure drop and aortic blood acceleration as indices of cardiac inotropy: a comparison with the first derivative of aortic pressure based on computer fluid dynamics.

This paper presents a computational approach to ventricular fluid mechanics to evaluate three inotropic indices of early ejection: the intraventricular pressure drop (deltap). the first derivative of aortic flow rate (df/dt) and the first derivative of aortic pressure dp/dt. dp/dt is one of the most frequently used indices for assessing myocardial inotropy. Deltap and df/dt are characteristic of inertia driven flows and reflect the impulsive nature of the flow inside the ventricle during the ejection phase. The study is based on an axisymmetric fluid dynamics model of the left ventricle, developed according to the finite element approach. The fluid cavity is bounded by a shell containing two sets of counter-rotating contractile fibres. Two simulation sets were performed: the former to investigate the sensitivity of deltap and df/dt peaks (deltap(max) and df/dt(max)) with respect to changes in the inotropic state of the fibre. The latter allows the evaluation of the dependency of deltap(max) and df/dt(max) on afterload by means of two supravalvular stenoses of 50% and 70%. The model simulates the inertial features of ventricle behaviour. The calculated values of the indices investigated are in close agreement with those reported in the literature. The sensitivities of deltap(max) df/dt(max) and dp/dt(max) are calculated for the two simulation sets. Data are normalised with respect to the maximum values reached in the simulation set. The comparison indicates that deltap(max) has a greater sensitivity (3.4 vs. 3.1 ) and a more linear pattern than dp/dt(max) for changes in the inotropic state of the fibre. df/dt(max), shows a sensitivity close to dp/dt(max). Results confirm that the afterload does not affect dp/dt(max), in accordance with experimental observations, while deltap(max) and, to a major degree, df/dt(max) decrease when the afterload is increased.

Factors affecting the respiratory ratio during cardiopulmonary by-pass.

Despite the wide use of hypothermic cardiopulmonary bypass (CPB) during open heart surgery there is little information about the patient metabolism. In particular no complete studies addressed the assessment of the respiratory ratio (RR) during CPB at different core temperatures. Therefore a clinical study was performed in order to evaluate the oxygen consumption (VO2) and carbon dioxide production (VCO2) in adult patients with valvular or coronary heart disease undergoing CPB. Twenty-five patients (16 male, 9 female) aged between 26 and 76 (54.2+/-12.4 mean +/- SD) were the subjects of this study. Measurements (102) were taken at various perfusion flow rates (from 1.6 to 2.9 L/min(-1) x m(-2)) and temperatures (from 24 to 37 degrees C). Arterial and mixed venous gas analyses were performed and O2 and CO2 concentrations were calculated, including the carbamate contribute. We calculated VO2, VCO2 and then RR from artero-venous differences in O2 and CO2 contents. Both VO2 and VCO2 showed a positive linear correlation with temperature (r = 0.82 and r = 0.59 respectively) and with blood flow rate (r = 0.61 and r = 0.29 respectively). The mean RR was 0.78+/-0.28 and more than 84% of RR values fell in the range 0.5-1.2. No significant correlation between RR and temperature and blood flow rate was observed. VCO2 and RR showed a positive linear correlation with the gas to blood flow rate ratio (r = 0.37 and r = 0.49 respectively).

Virtual extracorporeal circulation process.

Virtual instruments for an extracorporeal circulation (ECC) process were developed to simulate the reactions of a patient to different artificial perfusion conditions. The computer simulation of the patient takes into account the hydraulic, volume, thermal and biochemical phenomena and their interaction with the devices involved in ECC (cannulae dimensions, oxygenator and filter types, pulsatile or continuous pump and thermal exchangers). On the basis of the patient's initialisation data (height, weight, Ht) and perfusion variables (pump flow rate, water temperature, gas flow rate and composition) imposed by the operator, the virtual ECC monitors simulated arterial and venous pressure tracings in real time, along with arterial and venous flow rate tracings, urine production tracing and temperature levels. Oxyhemoglobin arterial and venous blood saturation together with other related variables (pO2, pCO2, pH, HCO3 are also monitored. A drug model which allows the simulation of the effect of vasodilator and diuretic drugs is also implemented. Alarms are provided in order to check which variables (pressure, saturation, pH, urine flow) are out of the expected ranges during the ECC simulation. Consequently the possibility of modifying the control parameters of the virtual devices of the ECC in run-time mode offers an interaction mode between the operator and the virtual environment.

Computational evaluation of intraventricular pressure gradients based on a fluid-structure approach.

The dynamics of intraventricular blood flow, i.e. its rapid evolution, implies the rise of intraventricular pressure gradients (IPGs) characteristic of the inertia-driven events as experimentally observed by Pasipoularides (1987, 1990) and by Falsetti et al. (1986). The IPG time course is determined by the wall contraction which, in turn, depends on the load applied, namely the intraventricular pressure which is the sum of the aortic pressure (i.e., the systemic net response) and the IPG. Hence the IPGs account, at least in part, for the wall movement. These considerations suggest the necessity of a comprehensive analysis of the ventricular mechanics involving both ventricular wall mechanics and intraventricular fluid dynamics as each domain determines the boundary conditions of the other. This paper presents a computational approach to ventricular ejection mechanics based on a fluid-structure interaction calculation for the evaluation of the IPG time course. An axisymmetric model of the left ventricle is utilized. The intraventricular fluid is assumed to be Newtonian. The ventricle wall is thin and is composed of two sets of counter-rotating fibres which behave according to the modified version of Wong's sarcomere model proposed by Montevecchi and Pietrabissa and Pietrabissa et al. (1987, 1991). The full Navier-Stokes equations describing the fluid domain are solved using Galerkin's weighted residual approach in conjunction with finite element approximation (FIDAP). The wall displacement is solved using the multiplane quasi-Newton method proposed by Buzzi Ferraris and Tronconi (1985). The interaction procedure is performed by means of an external macro which compares the flow fields and the wall displacement and appropriately modifies the boundary conditions to reach the simultaneous and congruous convergence of the two problems. The results refer to a simulation of the ventricular ejection with a heart rate of 72 bpm. In this phase the ventricle ejects 61 cm3 (ejection fraction equal to 54 percent) and the ventricular pressure varies from 78 mmHg to 140 mmHg. The IPG show an oscillating behaviour with two major peaks at the beginning (11.09 mmHg) and at the end (4.32 mmHg) of the ejection phase, when the flow rate hardly changes, according to the experimental data. Furthermore the wall displacement, the wall stress and strain, the pressure and velocity fields are calculated and reported.

A numerical fluid mechanical study of repaired congenital heart defects. Application to the total cavopulmonary connection.

A computational fluid dynamics study based on the application of the finite element method has been performed to investigate the local hemodynamics of the total cavopulmonary connection. This operation is used to treat congenital malformations of the right heart and consists of a by-pass of the right ventricle. In this paper the adopted methodology is presented, together with some of the preliminary results. A three-dimensional parametric model of the connection and a lumped-parameter mechanical model of the pulmonary circulation have been developed. The three-dimensional model has been used to simulate the local fluid dynamics for different designs of the connection, allowing a quantitative evaluation of the dissipated energy in each of the examined configurations. The pulmonary afterload of the three-dimensional model has been reproduced by coupling it with the pulmonary mechanical model. The results show that, from a comparative point of view, the energetic losses can be greatly reduced if a proper hydraulic design of the connection is adopted, which also allows control of the blood flow distribution into the lungs.

Fluid-structure interaction problems in bio-fluid mechanics: a numerical study of the motion of an isolated particle freely suspended in channel flow.

In this paper a problem belonging to the moving boundary class is tackled with a 2-D application of computational fluid dynamics techniques. The motion of an isolated rigid particle freely suspended in an incompressible Newtonian fluid in a narrow channel is studied numerically at a low Reynolds number, yet different from zero. The actual problem consists of two coupled problems: the motion of the viscous fluid and that of the rigid particle suspended and convected with the fluid. The full Navier-Stokes equations (i.e. both transient and convective terms are included) are solved in the fluid domain by means of the finite element method, while the motion of the particle is determined on the basis of a rigid act of motion. Results from simulations corresponding to differential initial positions of the particle are shown in this paper: they allow one to study the rotational motions of the particle as well as its displacements. The goal of the paper is to analyse the lateral displacement behaviour of the particle, already observed in experimental studies in microcirculation. In particular, lateral migrations are supposed to be due to inertial forces acting in the fluid around the moving particle combined with the proximity of the resting wall (wall effect). Preliminary results are in fairly good agreement with those available in the literature.

Preliminary design and optimization of an ECC blood pump by means of a parametric approach.

This study concerns the development of an analytical parametric model of a centrifugal disk pump. The advantage of this kind of approach is to have an adaptable tool as a first step for the design of a pump device. The method allows the evaluation of the velocity profiles and the shear stresses within the impeller disks in the flow domain along with the performance of the device in terms of torque, mechanical power, power loss, head-flow performance, pump efficiency, and hemolytic index. Some simplifying hypotheses are assumed: steady state condition, laminar flow, Newtonian and incompressible fluid. The radial velocity profiles are assumed to be uniform and the flow cross-sectional area is assumed to be constant along the radius. The influence of the housing and secondary flows caused by recirculation are neglected. To test the approach reliability, the model was used to simulate a pump with the following characteristics: an external and internal radius of 50 mm and 5 mm, respectively, and a channel height of 2.5-0.25 mm (h) from inlet to outlet section. The angular velocity omega was varied in the range 500-3,000 rpm. The flow rate has been varied from 1 to 5 L/min. The results show that when the flow rate is increased, head performances obtained using this pump model vary from 411 to 100 mm Hg, and its efficiency varies from 48 to 15%. A parallel simulation has been carried out by means of a Finite Element Method model with an angular velocity equal to 2,000 rpm.(ABSTRACT TRUNCATED AT 250 WORDS)

Mechanical characterization of a model of a multicomponent cardiac fibre.

We have developed a model of a cardiac fibre composed of several contractile units in series and activated in succession; each unit behaves according to Wong's model. The main difference between the multicomponent model and the classic monocomponent model is that it is possible to take into account both the dynamic phenomena due to the propagation of the activation signal along the fibre and the contractility of each unit from which the fibre is constructed. Isometric and isotonic contractions have been simulated under different conditions in terms of preload, afterload, frequency and number of inhibited units. The analysis of the results allows us to assert that the multicomponent fibre behaviour is in good agreement with experimental results from the literature. We believe that the multicomponent cardiac fibre should be regarded as a powerful tool linking the sarcomere contraction with that of the whole ventricle.

Clinical experience with a new pulsatile pump for infant and pediatric cardiopulmonary bypass.

A pulsatile pump of new concept has been developed for infant and pediatric cardiopulmonary bypass (cpb) (Parenzan-Fumero pump). A segment of elastic tubing is compressed by a pneumatically driven pushing plate under control of a microprocessor. Flow parameters such as pulse rate and stroke volume can be set. The pump can be synchronized with the patient's ECG for counterpulsation heart assist. A total of 87 open-heart procedures were performed using randomly either a conventional roller pump or the Parenzan-Fumero pump (respectively 39 and 48 patients). A previously published cpb protocol and anesthetic regimen were adopted in all cases. The results show increased cooling and rewarming rate (p less than 0.05) and urinary output, decreased vascular resistance, intensive care unit time and need for blood transfusion in the pulsatile group compared to the continuous perfusion group. In the pulsatile group, mortality was significantly lower (10.4% vs 25.6%) and low cardiac output syndrome was less frequent in the post-operative course.