Implantable pump

Abstract

A blood flow assist device comprising an axial flow rotary pump, said pump comprising a rotatable impeller assembly 3 comprising a hub and one or more impeller blades 15, and an electromagnetic stator

Description

Field of the Invention The present invention relates to implantable pumps, and particularly, although not exclusively, to intraarterial blood pumps. Background Heart failure is a common cause of death in the western world, resulting in approximately 10 % of all deaths in the UK per year. The leading cause for this is coronary artery disease. Coronary artery disease is caused by blockages of the arteries supplying the heart which can cause heart attacks and ischaemia, and death of parts of the muscle of the heart. Upon death of part of the heart muscle, the main pumping chamber of the heart (the left ventricle) dilates and becomes weaker, known as left ventricular systolic dysfunction (LVSD). Much like any pump, when the pump weakens, the pressure behind the pump increases. In this context, this causes an increase in pulmonary venous pressure. An elevation in pulmonary venous pressure will causes breathlessness and pulmonary oedema. Patients may also suffer from low forward-pressure, leading to hypoperfusion of the brain and kidneys, thereby resulting in postural hypotension and renal impairment. Currently, medical management of heart failure centres around management of fluid balance, and up-titration of prognostically significant medications. These medications (ACE-I, B blocker, MRA, ARNI) are well proven to improve the prognosis of LVSD associated heart failure, but also simultaneously reduce the blood pressure, and worsen renal function. This means that in severe heart failure, these medications can frequently not be used, as this worsens the symptoms of the syndrome of heart failure. This is termed medication refractory heart failure. As heart failure progresses, patients become more breathless, and this is classified on the NYHA (New York Heart Association) scoring system. NYHA class IV defines patients who are breathless at rest, and these patients have a median life expectancy of approximately 6 months. For some patients, heart transplantation is considered as a possible treatment option. However, due to the poor prognosis of end-stage heart failure, it is often necessary to provide a 'bridging’ treatment to support existing heart function until a suitable donor heart becomes available. For this reason, mechanical devices were developed to bridge people between end stage heart failure and transplantation. The device for supporting existing heart function in patients with end-stage heart failure is a Left Ventricular Assist Device (LVAD). This type of device is an externally electrically powered pump which is implanted in the left ventricular apex and accelerates blood into the systemic circulation, normally directly to the aorta. It is normally a continuous flow pump. Implantation of an LVAD device requires open-heart surgery in which one end of the LVAD pump is attached to the left ventricle and the other end to the aorta. Blood flows from the ventricle into the LVAD, and is then pumped by the device out into the aorta, where it then flows to the rest of the body. However, whilst use of an LVAD is an effective treatment in some patients, there are significant disadvantages: the surgery required for implantation of the device is extremely invasive and strenuous, and so unsuitable for frailer patients. Additionally, once an LVAD has been implanted, it is not possible to readily restore the original heart function, even if there is some subsequent recovery of the heart: in other words, implantation of an LVAD necessitates either life-long treatment via LVAD, or a future heart transplant. LVADs are currently complex &accordingly expensive to manufacture, and are unsuitable for a broad range of cardiac morphologies including smaller left ventricles, which limits their use-case to a relatively small cohort of patients. Some efforts have been made to develop alternative methods for treatment or support of cardiac pump function. For example, US20100268017A1 proposes an intracardiac pumping device comprising a pump connected at the proximal end with a catheter. In use, the pumping device is placed such that the pump is located in the aorta while the suction head lies in the left ventricle. A canula extends through the aortic valve, connecting the suction head and pump. Thus, the pump draws blood from the left ventricle and feeds into the aorta. This device is suitable for implantation in the heart using percutaneous insertion via the femoral artery, thereby obviating the need for open-heart surgery for implantation of the device. However, due to the configuration of this device (its size meaning that high RPM of the pump are required to provide suitable pumping pressure, thereby resulting in undesirable haemolysis and / or thrombosis), it is not suitable for long-term or permanent implantation, and is only used in a peri-procedural setting to improve the haemodynamics of a patient at the cost of significant haemolysis. EP4069347A1 proposes a blood flow assist system including an impeller assembly comprising a rotor assembly and an impeller coupled with the rotor assembly. A drive unit is provided proximal the impeller assembly, the drive unit comprising a drive magnet and a drive bearing between the drive magnet and the impeller assembly. The provision of a bearing surface between the drive magnet and impeller assembly may lead to undesirable haemolysis and / or thrombosis of the blood during use and so such devices may not be suitable for long-term or permanent implantation. Furthermore, this “umbrella” design to the impeller blades aids deployment, but requires a high RPM to generate sufficient forward flow to aid haemodynamics, resulting in significant haemolysis. US2021077687A1 proposes a wirelessly powered circulatory assist pump. In this proposal, the pump comprises an impeller which is powered either by a drive shaft, or by a wireless drive unit provided in the form of an external power belt an external power belt that fits around the patient's abdomen, which belt provides a magnetic field that drives and / or controls rotation of the impeller. Such arrangements are not suitable or practical for long-term implantation: in particular use of an external power belt provides a risk of misalignment of the belt with the pump, leading to loss of power to the device which could be harmful to patient safety. The present invention has been devised in light of the above considerations. Summary of the Invention The present inventors have realised that it would be possible to overcome some of the known problems with LVAD devices by providing a blood flow assist system having a different configuration to known devices. Specifically, the inventors have realised that by providing a device which is configured to engage an interior wall of, or replace a segment of, a vessel of a human body during use - and, in particular, engage or replace part of the aorta (e.g. part of the ascending aorta) in a human heart, the blood flow assist device can occupy a larger cross-sectional area within the vessel as compared with existing known devices which do not engage an interior wall of the vessel, thereby allowing the use of larger impeller blades which may reduce the risk of undesirable haemolysis and / or thrombosis during use of the device, and may additionally increase the efficiency of the device. Accordingly, in a first aspect, the present invention provides a blood flow assist device comprising an axial flow rotary pump, said pump comprising a rotatable impeller assembly comprising a hub and one or more impeller blades, and an electromagnetic stator operable to drive rotation of the impeller assembly; wherein the electromagnetic stator assembly comprises electromagnetic windings supported on a supporting structure, said structure being configured for engagement with, or replacement of, an interior wall of a vessel of a human body during use. Whilst the device may be suitable for implantation within any given vessel of a human body, the device is preferably adapted to be delivered to, and implanted within, a portion of the aorta, more preferably the ascending aorta. Accordingly, the device is preferably configured for engagement with, or replacement of, the interior wall of the aorta of a human body during use, preferably the ascending aorta. Where the device is configured to engage the interior wall of vessel, the engagement is preferably circumferential engagement: that is, preferably the supporting structure is configured to engage an interior wall of a vessel of a human body during use at multiple locations about the internal circumference of the vessel. Alternatively, where the device is configured to replace a segment of vessel (e.g. of the aorta), the device may interface with non-replaced sections of the aorta via a sleeve which can be sutured to the free edges of these remaining sections of the vessel (e.g. aorta) - other arrangements are contemplated, but it is suggested that this arrangement offers a particularly convenient mode of attachment of the device to the vessel. Techniques for attaching a sleeve in this manner are generally known from e.g. techniques used in aortic root replacement surgery. In either type of arrangement, the supporting structure may be configured such that at least one external dimension of the supporting structure substantially corresponds to an internal dimension of the vessel of a human body in which the device is intended to be located I which the device is intended to replace a part of during use. For example, an external diameter of the supporting structure may be selected to be substantially equal to an internal diameter of the vessel e.g. selected to be no more than ±20%, ±10%, or ±5% of the internal diameter of the vessel. In this way, the external dimension of the device is similar to the dimension of the vessel, thereby enabling a close fit between the device and the vessel, which may provide for appropriate device function. Further optional features relating to sizing of the device are discussed in more detail below. The present inventors have further realised that particular advantages may be achieved by providing a blood flow assist system at least part or all of which is transluminally deliverable, as well as being configured for engagement with an interior wall of a vessel of a human body during use. However, existing systems which are suitable for transluminal delivery also suffer from a range of problems, as identified above. Firstly, existing pumps that are small enough for transluminal delivery typically have impeller blades that are small, meaning that work is distributed over only a small area, thereby resulting in large point forces on the blade, and increased risk of haemolysis during use. Additionally, common axial flow pumps typically comprise components arranged along a central axis, which both obstructs flow and causes stasis regions. Some of these problems also apply in non-transluminally deliverable systems. Accordingly, it is a further aim of the present invention to provide a blood flow assist system which provides a relatively high surface blood contact area, and which occupies a significant proportion of a blood vessel but without obstructing flow within the vessel. However, it is contemplated that even in systems where not all components are transluminally deliverable, the impeller assembly may be deliverable percutaneously and transluminally to allow for ease of replacement of the impeller via a less invasive procedure. In some embodiments, the supporting structure may accordingly be an expandable supporting structure. The expandable supporting structure may be expandable between an insertion configuration and a deployed configuration. In the insertion configuration, the structure may be transluminally deliverable. In the deployed configuration, the structure may be configured to engage with an internal wall surface of a vessel of a human body. Alternatively, it is contemplated that many of the above advantages can also be achieved in devices which do not employ an expandable supporting structure. In some embodiments, the supporting structure may therefore have a fixed external circumference (e.g. fixed external profile of the device). The fixed external circumference could be provided e.g. by a rigid housing portion of the device. Embodiments having a fixed external profile may by particularly suitable for use in replacing a segment of a vessel. Typically, where the external profile of the device is fixed (i.e. the device is non-expandable) it will be required to be implanted via a surgical method. Accordingly, such embodiments may be referred to herein as ‘surgically-implantable’ designs. The present invention may reduce or mitigate a number of problems with known blood flow assist devices. Where the device is transluminally deliverable, allowing for transluminal delivery of the electromagnetic stator can allow for implantation of the device in a less invasive manner compared to various known blood flow assist devices, thereby allowing for use of the device in frailer patients. Additionally, as will be discussed in greater detail below, the device may be implanted by techniques similar to those used for transcatheter aortic valve implantation (TAVI), which is a technique familiar to many interventional cardiologists, reducing the ‘learn-in’ time required for operators to learn how to perform suitable implantation of the device. In alternative embodiments where the device is not transluminally deliverable, but rather is a device which replaces a segment of a vessel, advantages are still neverthesless provided over existing LVAD systems in view of the less invasive surgery required for implantation of the device as compared with that required for implantation of traditional devices. This surgery would be similar to ascending aortic root replacement surgery. In some embodiments, the rotatable impeller assembly of the device may be configured to be moveable between an insertion configuration and a deployed configuration. This may not be essential in some configurations of the device, as the relatively smaller size of the rotatable impeller assembly may in any case allow for transluminal delivery without the need for providing an adjustable configuration (and in particular may not be necessary where the device is intended for surgical implantation e.g. where the device has a fixed external profile). However, provision of an adjustable configuration for the rotatable impeller assembly can offer a number of technical advantages, including facilitation of implantation, as well as allowing for use of impellers having larger blade sizes than would be possible in devices now having an adjustable configuration. The rotatable impeller assembly may be transluminally deliverable in one or both of the insertion and the deployed configurations, however it is particularly preferred that it is transluminally deliverable in at least the insertion configuration. The insertion configuration may be a configuration in which the one or more impeller blades are at least partially folded relative to the deployed configuration. Alternatively or additionally, the insertion configuration may be a configuration in which the one or more impeller blades radially extend by a smaller amount relative to the radial extent of the one or more blades in the deployed configuration (the radial direction being a direction radially transverse to the longitudinal axis of the device). Where the support structure is an expandable supporting structure, both the expandable supporting structure and / or the rotatable impeller assembly may additionally be configured to move to positions between the insertion and the deployed configurations (e.g., they may be configured to move to a partially deployed position). As will be readily understood by the skilled person, these features may not be necessary where the device is intended for surgical implantation. As discussed above, an advantage of some embodiments of the present invention that multiple primary components of the blood flow assist device are transluminally deliverable or deliverable without operating on the left ventricle itself, thereby obviating the need for complex surgery on a contracting myocardium. This advantage is present in both a surgical and percutaneous approach to device implantation. The term “transluminally deliverable” is used herein to define that the specified component or structure is collapsible to a maximum diameter suitable for insertion through a blood vessel of a patient - for example, suitable for insertion through the femoral artery of a patient, although alternative implantation techniques are possible, as will be discussed in further detail below. The maximum diameter of a transluminally deliverable structure (diameter here being defined as a dimension perpendicular to the longitudinal axis of the structure) when in the insertion configuration may accordingly be 8 mm or less, 7 mm or less, or 6 mm or less. Providing a smaller diameter may allow for use of the device across a wider range of patients. For example, providing devices where the components thereof have a smaller diameter when in an insertion configuration may allow use of the device in patients having smaller-diameter arteries, or mild arterial restriction e.g. resulting from atherosclerosis. As noted above, the electromagnetic stator comprises electromagnetic windings supported on a supporting structure. In some embodiments, the supporting structure may be an expandable / collapsible supporting structure. In other embodiments, the supporting structure may be a substantially rigid structure (i.e. may not be expandable / collapsible structure). In some arrangements, the device may comprise a single set of windings. In other arrangements multiple set of windings may be provided: for example the device may comprise two or more separate sets of windings. Where the supporting structure is an expandable / collapsible supporting structure, the supporting structure may be provided by a stent. In such arrangements, the supporting structure may comprise a self expandable stent. Alternatively, or additionally the supporting structure may comprise a balloonexpandable stent. Where the support structure comprises an expandable stent, preferably the stent has a cross-strut locking design such that when in the deployed configuration, it provides a stable, substantially non-compressible cylinder. Where the supporting structure is substantially rigid structure, this may be provided by a rigid housing portion of the device - for example by a metal housing, e.g. a stainless steel case, or rigid stent or wireframe structure. The size of the supporting structure (when deployed for use, in relevant embodiments) may be limited by the intended application of the device - specifically, by the size and / or geometry of the vessel of the human body in which the device is intended for implantation. For devices having an expandable / collapsible supporting structure, as the deployed configuration is a configuration in which the expandable supporting structure is configured to engage with an internal wall surface of a vessel of a human body, the device may be configured by appropriate selection of the dimensions of the device in the deployed configuration such that suitable engagement occurs. Whilst the device may be suitable for implantation within any given vessel of a human body, the device is preferably adapted to be delivered to, and implanted within, a portion of the aorta, more preferably the ascending aorta. A typical human aortic diameter (without dilatation) may be in a range of from about 3 cm to about 4 cm. Accordingly, the maximum diameter of the device when in the deployed configuration may be 2 cm or more, or 3 cm or more. Providing a larger diameter for the supporting structure when in the deployed configuration may provide for a more secure engagement of the device within the vessel of the human body. However, the diameter should preferably not be so large that full deployment results in overstretching of the vessel in which the device is implanted. For devices having a rigid supporting structure, preferably the largest external diameter of the supporting structure is selected to be similar in diameter to the aorta to avoid compression of external structures (i.e. may be in a range of from about 2 cm to about 5 cm), but there may be more tolerance in the size. In some preferred arrangements, and in particular for arrangements where the device is designed for implantation in the aorta of a human patient, the diameter of the supporting structure when in the deployed configuration may be in a range of from 2 to 5 cm, more preferably 3 to 5 cm, suitably around 4 cm. If the device is designed to replace a section of the descending aorta, its largest external diameter could exceed this, to approximately 5cm. As the skilled person readily understands, different diameters may be appropriate where the device has a different intended implantation location. Different diameters may also be appropriate for use in different patients. Accordingly, the diameter of the supporting structure in the deployed configuration may be selected based on at least one parameter of the patient. For example, the diameter of at least one blood vessel of the patient may be measured (e.g. the diameter of the ascending aorta), and the diameter of the supporting structure in the deployed configuration may be selected to be substantially equal to, or within a given error margin of, this measured diameter, or selected to be within a given error margin the measured diameter, such as within 10%, or within 5% of the measured diameter. Where the supporting structure is an expandable supporting structure configured for transluminal delivery, moveable between an insertion configuration and a deployed configuration, the maximum diameter of the expandable supporting structure may be 8 mm or less, 7 mm or less, or 6 mm or less when in the insertion configuration, thereby allowing for transluminal delivery of the structure. In a fixed structure design, the diameter would remain substantially constant through implantation and use of the device, and would not be constrained to collapse to 8mm or less, 7mm or less, or 6mm or less. The length of the supporting structure (measured along its longitudinal axis) may be selected to be in a range of from 2 cm to 5 cm, more preferably in a range of from 3 to 4 cm, in some arrangements around 3.5 cm. It has been found that this length can provide suitable engagement of the device within the aorta of a typical patient. This corresponds to a relatively straight section of the aorta, without arterial branches, allowing engagement or replacement of this section with relative ease. Providing a device having a length greater than this may prevent the device from being appropriately located within a patient due to the natural curvature of the ascending aorta. The material of the supporting structure may be a material having suitable biocompatibility as well as magnetic flux capacity. Suitably, the supporting structure may include stainless steel (e.g. stainless steel 410), or a composite of iron and cobalt. The supporting structure may comprise one or more surface coatings. For example, it may be coated in a biocompatible polymer - to avoid direct contact of the underlying material of the supporting structure with the blood. The electromagnetic stator may be a single-phase, two-phase or three-phase stator. Preferably, the stator is a multi-phase stator, utilising higher order phases to provide magnetic stabilisation. A multiphase stator may include 6 phases or more. That is, the electromagnetic stator may comprise 6 or more pairs of electromagnetic windings, supported on the supporting structure. Where the stator is a singlephase, two-phase or three-phase stator in may comprise one, two or three electromagnetic windings, or pairs of windings, supported on the supporting structure. The electromagnetic windings may be supported on the structure by either being embedded in solid portions of the stent, between the expandible sections (for arrangement utilising an expandable supporting structure), or wound around the stator profiles (e.g. iron-cobalt stator profiles) of a rigid supporting structure. The electromagnetic windings may be arranged in a ‘star’ fashion - that is, each pair of windings may be oppositely arranged about a periphery of the stator, with each pair of windings being circumferentially offset by a predetermined angle. For example, each pair of windings may be circumferentially offset by angles of around 60° and 120° with respect to the other pairs of windings. The number of windings may be between 3 and 24. Alternative diamond-shaped windings may be used. The use of diamond-shaped winding may be preferred when the device employs a twin bearingless motor design, as discussed in further detail below. The windings may be formed from a wound wire element, as is conventionally known in the art. The electromagnetic windings may follow a traditional helical winding pattern. A helical winding may be preferred in a fixed circumference model due to ease of manufacture, and less strict constraints on the external diameter due to the replacement of the segment of vessel, and therefore the potential for the device to be larger than the vessel it replaces. Alternatively, the electromagnetic windings may follow an orthocyclic winding pattern. That is, the element(s) providing the winding may be wound in an orthocyclic manner. Use of an orthocyclic winding pattern may be preferred because it is particularly space efficient, thereby encroaching minimally on the internal lumen. The materials from which the windings are formed are not particularly limited, other than that the windings should comprise an electrically conductive material. Suitably, the windings should be formed from a material having a relatively high electrical conductivity. Preferably, the material should also be biocompatible, and have a low rate of thrombogenicity. The windings may comprise a material selected from: silver or copper. In some arrangements, the winding may comprise a surface coating. For example, the windings may be coated in enamel or any alternative coating material suitable for such use. Where a surface coating is provided, this may be a coating which improves blood compatibility of the winding material. For example, the winding may be heparin coated. This can reduce the thrombogenicity of the windings. In one particularly preferred arrangement, the windings comprise heparin-coated silver wire. It has been found that this arrangement provides a convenient combination of suitably high electromechanical performance as well as suitable biocompatibility. In preferred arrangements, the windings will be contained within the supporting structure of the stator. Where the windings are contained within the supporting structure this can allow them to be protected from contact with blood, e.g. by a metal barrier, such as steel or titanium, forming part of the body of the supporting structure. The rotatable impeller assembly and the electromagnetic stator assembly operable to drive rotation of the impeller may together be considered as a motor. The precise mode of operation of the motor is not particularly limited, however preferably the motor is selected from a brushless DC motor (BLDC), an AC induction motor, or a permanent magnet synchronous motor. Use of a BLDC motor may be preferred over the use of an AC induction motor because it is more efficient on a smaller scale, as well as achieving having torque independent rotational speed, allowing the device to synchronise more closely to the ventricular systole. A permanent magnet synchronous motor has similar advantages. In some arrangements, a bearingless motor design may be employed. Where the electromagnetic stator assembly comprises multiple distinct sets of windings, it will be appreciated that the rotatable impeller assembly and the electromagnetic stator assembly operable to drive rotation of the impeller may together be considered as a plurality of motors - e.g. it is contemplated that the device could employ a twin motor design, in which the electromagnetic stator assembly comprises two distinct sets of windings, arranged at respective ends of the stator assembly, and accordingly acts as two motors which function synchronously. Such a configuration would act to rotate and also provide active magnetic levitation for the impeller. One possible winding and power configuration for this would be a 6 phase, multiphase combined winding approach, although other conformations would also be suitable. Different winding conformations may be preferred, to optimise the motor efficiency. Diamond shaped windings may be preferred, to allow tessellation of the two motor windings together, and minimise space needed at the end of the windings and provide active axial support, although alternative winding configurations could also be used. Use of such an arrangement may be particularly advantageous, in that it may allow for magnetic levitation of the impeller assembly in a more stable manner than some other possible arrangement, thereby removing any need to provide end caps on the device to retain the impeller assembly within the electromagnetic stator assembly. The use of two bearingless motors would be provide stabilisation across 4 degrees of freedom for the impeller assembly, and up to 5 degrees of freedom if some winding configurations are used, such as diamond windings. Stability across the fifth degree of freedom (axial stabilisation) could also be achieved passively, by suitable configuration for the device structure, or actively, by suitable operation of the device. For example, the device may comprise one or more nonmagnetic laminations which separate the two motors: this could provide passive axial stabilisation. Alternatively or additionally, the current supplied to the two motors during use of the device may be varied relative to one another In a manner which provides for active axial stabilisation. It is also possible that the device could use a single motor, with windings running in different directions, to achieve 5 degrees of freedom of control. This may offer further advantages by improving the efficiency and power density of the motor. Another advantage provided by use of a single- or twin-bearingless motor design is that it may allow the device to have an open-ended structure (in view of the lack of requirements for end caps), thereby providing an additional flow path for blood - this, and the corresponding advantages are discussed in further detail below. A single- or twin- bearingless motor conformation confers significant advantages over both a passive magnetic and conventional bearing approach. Accordingly, in some configurations, the impeller may comprise one or more permanent magnets, e.g. three or more permanent magnets. This may allow for operation of the device as a brushless DC motor / permanent magnet synchronous motor. The permanent magnets may be provided on or in the hub of the rotatable impeller assembly. In alternative arrangements, the magnets may be provided elsewhere in the rotatable impeller assembly - for example, where the rotatable impeller assembly comprises an outer sheath (as discussed in further detail below), the magnets may be provided on or in the outer sheath of the assembly. The one or more permanent magnets may be arranged to provide between two and sixteen pole pairs circumferentially arranged about the rotatable impeller assembly. The pole pairs may alternate North (N) and South (S) poles. The polarity of these magnets may be organised as a Halbach array. The one or more permanent magnets may be formed from any suitable material. For example, they may be formed from a material comprising iron, nickel, cobalt, rare-earth metals, or alloys thereof. Preferably, the one or more permanent magnets comprise a rare earth metal or alloy thereof. Suitably the one or more permanent magnets may be formed from Neodymium (Nd), Samarium Cobalt (SmCo), or an alloy of Neodymium, Ferrite and Boron (NdFeB). These magnets will also interact with the stator to repel the stator and function as passive magnetic bearings. The blood flow assist device may comprise one or more bypass flow conduits. Preferably, at least the rotatable impeller assembly of the blood flow assist device may comprise one or more bypass flow conduits. Provision of one or more bypass flow conduits may allow for an increase in the mass flow rate possible during use of the blood flow assist device. It may also decrease the exist velocity of blood flowing through the blood flow assist device. Decreased exit velocity may lead to an increase in pump efficiency, and thereby allow the pump to operate at comparatively lower RPMs as compared with conventional pump designs. This design also allows for a failsafe aspect of the design, as endogenous blood flow can continue if the device fails to rotate. Where one or more such bypass flow conduits are provided, they may have an area (measured in a cross-section perpendicular to the flow direction through the device) of greater than 1cm2, for example in a range of from 1-2 cm2-where there are multiple bypass flow conduits, this may be the total area of all bypass flow conduits. Providing such an area for the bypass flow conduit can ensure that the device would not obstruct endogenous blood flow sufficiently to induce significant haemodynamic compromise in the case of device failure. Preferably, at least a first bypass flow conduit is provided which is substantially centrally located on the rotatable impeller assembly. That is, the bypass flow conduit may lie along a central longitudinal axis of the rotatable impeller assembly. In some arrangements, a second bypass flow conduit may be provided. The second bypass flow conduit may circumferentially surround the rotatable impeller assembly (i.e. may be located radially outwardly of the rotatable impeller assembly, between the rotatably impeller assembly and the stator assembly). Such an arrangement may be particularly relevant for devices which employ a twin bearingless motor assembly without end caps: the open-ended edges of a bearingless motor configuration allow for provision of this additional bypass flow path radially outwardly of the rotatable impeller assembly. Providing an additional bypass flow path in this manner may allow for a further increase in the mass flow rate possible during use of the blood flow assist device. This can in turn improve temperature control of the device, and additionally may reduce the thrombotic risk within the device. The hub of the rotatable impeller assembly may be substantially cylindrical or tubular. The hub may be provided by a stent or wireframe structure. In the expandible design the stent or wireframe structure may have a variable diameter. The stent or wireframe structure may be of a fixed size in the surgically implantable design. As noted above, the rotatable impeller assembly may comprise one or more impeller blades. These may alternatively be referred to as ‘vanes’. The impeller assembly may comprise two, three, four, five or more blades. Preferably, the impeller assembly comprises between three and five blades, e.g. four blades. This arrangement has been found to provide suitable mass flow through the device. The blade(s) extends radially outwardly along a radial axis that is radially transverse to the longitudinal axis of the device. The radial extent of the blade(s) may be 5mm or more, more preferably 5-15 mm. As noted above, in some embodiments (e.g. embodiments where the rotatable impeller assembly is moveable between an insertion configuration and a deployed configuration) the blade(s) may be foldable. The blade(s) may be arranged in a helical manner with respect of the impeller hub. The pitch of the blades may be selected to be approximately 1 full rotation in 200mm, although the range of acceptable pitches may be in a range of from 1 full rotation in 50 mm to 1 full rotation 200 mm. The blades may be formed from any suitable material. Preferably the material(s) from which the blades are formed have low thrombogenicity. Preferably the blades are flexible, although this is not essential in all embodiments - for example, in a surgically-implantable design the blades are not required to be as flexible as in an expandable, transluminally-deliverable design and the blades may consequently be more rigid in devices intended for implantation by surgical means. The blades may comprise (e.g. be formed entirely or partially of) a fabric material, an elastomeric material or a metal material (e.g. an alloy, such as a titanium alloy or a steel alloy). One suitable material for the blades is Dyneema®, however the blades may alternatively comprise a material selected from silicone rubber, Pebax® plastic, or any other suitable material. The materials selected for use should preferably comply with ISO 10993-1:2018 regulation for intra-arterial application. If the device is surgically implanted, metal blades such as stainless steel or titanium may offer superior durability compared to other materials. The blades may be attached to or supported on the hub in any suitable manner. In some embodiments, the blades may be attached to the hub by being sewn to the hub. In other embodiments, the blades may be attached to the hub by use of an adhesive. In yet other embodiments, the blades and the hub may be integrally formed. That is, the blades and the hub may constitute a single one unit. In such arrangements, the unit may be formed by e.g. machining or printing the blades &hub as a single unit, or alternatively by laser welding the blades and hub together to form a single integral unit (in particular this manufacturing method may be appropriate where the blades and hub are formed of metal materials). In some arrangements, the rotatable impeller assembly hub may comprise inner and outer cylindrical portions, with the blades extending between these portions. The hub may further comprise a housing portion configured to house one or more sub-elements of the hub - for example, the housing portion may be configured to support a magnet array, e.g. a plurality of magnets organised as a Halbach array arranged on its outer surface. The rotatable impeller assembly may comprise an outer sheath surrounding the hub. Where present, the outer sheath may conveniently be formed from heat-shrink polymer, or metal. The rotatable impeller assembly may be magnetically levitated within the electromagnetic stator. That is, the rotatable impeller assembly may be supported by permanent radial magnetic forces. In this way, there may be no mechanical contact between the rotatable impeller assembler and the electromagnetic stator. This may reduce the rate of haemolysis and / or thrombosis resulting from operation of the device, in particular in comparison to known devices, have one or more bearing surfaces at which the rotor and stator of the pump contact one another. Permanent radial magnetic forces may be provided by provision of one or more permanent magnets provided on the supporting structure. In some arrangements, a plurality of permanent magnets may be circumferentially arranged about a diameter of the supporting structure as a magnet array. Such an arrangement can provide suitable magnetic levitation of the rotatable impeller assembly. The electromagnetic stator may be operable to drive rotation of the impeller in a conventional manner. That is, the pump may be configured to receive power to energize the windings of the electromagnetic stator in a predetermined sequence, thereby resulting in rotation of the impeller by interaction of the generated electromagnetic field of the stator with the magnetic field of a magnet array provided on the rotatable impeller. The device may comprise one or more hall sensors, eddy current sensors, or inductive sensors, said sensors being configured to sense the position of the rotatable impeller assembly, and energize the windings of the electromagnetic stator in a predetermined sequence based on the detected position of the rotatable impeller assembly. In one suitable arrangement, the device may comprise three or more Hall sensors provided on the stator. In another suitable arrangement, the device may comprise four or more eddy current sensors placed at each end of the stator of the device. The use of at least four sensors may be preferable to allow more accurate determination of the radial position of both ends of the motor most accurately. In some arrangements, more than four sensors may be used to achieve redundancy in the event of sensor failure. Eddy current sensors may be preferred due to their well-established high accuracy. Alternatively, the device may be sensorless, and rely on the induced electromotive force in the windings to measure the radial displacement of the rotor relative to the stator (in other words, physical motor angle sensors are not present, and their functionality is instead provided by signal processing of the motor’s winding currents). In some arrangements, it is contemplated that the device may rely on endogenous circulation to start the motor. The pump may be a continuous flow plump (i.e. it may be operable to provide a continuous flow). Alternatively, the pump may be a pulsative pump (i.e. operable to provide a pulsate flow). In some arrangements, the pump may be operable to operate as a continuous flow pump in a first mode, and to operate as a pulsatile pump in a second mode. The pump may be operable to switch between the first and second modes in response to an input signal (e.g. from a control module of the device). The device may comprise a sensor configured to sense ventricular depolarisation. The device may be driven in response to the output of the sensor, to assist systolic ejection of blood from the left ventricle. The device may be configured to rotate the rotatable impeller assembly at a speed of from 600 RPM to 6000 RPM with a preferred speed of 3000 RPM. It has been found that operating the device at such RPM can provide for suitable assistance of blood flow, but with reduced risk of undesirable haemolysis and / or thrombosis during use of the device. The device may comprise one or more magnetic bearings for stabilisation of the impeller. The bearings may be passive bearings or active bearings. Preferably, a passive magnetic bearing arrangement is provided. A passive magnetic bearing arrangement may include one or more magnets provided on the impeller assembly and one or more magnets provided on the stator assembly, wherein the magnets on the impeller assembly and the magnets on the stator assembly are arranged such that their magnetic axes oppose one another (i.e. a pair of cooperating magnets / magnetic arrays). For example, in one convenient arrangement a circular magnet (or a circular magnetic array comprising a plurality of magnets arranged in a circular manner) may be provided one each of the impeller assembly and the stator assembly to provide a pair of cooperating magnets / magnetic arrays. The magnets / magnet array provided on the stator assembly may be positioned such that the magnetic axis of the magnet(s) are positioned facing inwards towards the impeller at a 45 degree angle. The magnets / magnetic array provided on the impeller assembly may be positioned such that the magnetic axes of the magnet(s) are positioned facing outwards towards the stator assembly at a 45 degree angle. These cooperating magnets I magnetic arrays may be provided at each end of the device, in a longitudinal direction. This conformation can act to stabilise the impeller in all planes of movement, and magnetically levitate the rotatable impeller within the electromagnetic stator assembly. The magnetic bearings may comprise a single piece magnet attached at both ends of the stator, or may comprise multiple smaller magnets equally spaced around the stator circumference. Additional support may be offered by active bearings, and / or by hydrodynamic support from the blood. In some arrangements, the electromagnetic stator assembly may comprise first and second end caps configured to be fixedly connected to one another to secure one or more components of the stator assembly together. For example, the end caps may be configured to be fixedly connected to one another via one or mor rod members (e.g. bars) extending between the end caps. The rod members may be threadably attachable to each of the first and second end caps. This may provide a particularly convenient attachment arrangement. Where such end caps are provided, each of the end caps may have a smallest internal diameter that is smaller than the largest external diameter of the impeller assembly. This may provide a further safeguard preventing the impeller assembly from being displaced from within the stator assembly. The means of providing power to the pump is not particularly limited. Preferably, the pump is powered by an implanted power supply module which is configured to be implanted in the patient’s body. However, whilst not preferred, it is nevertheless contemplated that use of an external power supply module may be possible. Where the power supply module is intended for implantation, it may be configured for implantation in the pectoral area of a patient, e.g. in a pre-pectoral pocket. This location is the same as typically used for the implantation of a cardiac pacemaker or defibrillator, thereby reducing the ‘learn-in’ time required for operators to learn how to perform suitable implantation of the device. The power supply module may be provided separately to the blood flow assist device but used together as part of a blood flow assist system. Accordingly, a further aspect of the invention is a blood flow assist system comprising a blood flow assist device and a power supply module configured for connection to the blood flow assist device to supply power to the device. The power supply module may be implantable (that is, configured for implantation in the human body). The power supply module may comprise a battery. The battery may suitably be e.g. a lithium-ion battery, or nickel-cadmium or alternative battery conformation. The battery may be rechargeable. This can reduce the need for battery replacement procedures to be performed after implantation of the device. The power supply module may be configured for transcutaneous charging. For example, the power supply module may employ a system for transcutaneous charging as disclosed in Karim et al “Transcutaneous Pulsed RF Energy Transfer Mitigates Tissue Heating in High Power Demand Implanted Device Applications: In Vivo and In Silico Models Results”, Sensors (Basel), 2022 Oct 13;22(20):7775. Alternatively or additionally a percutaneous charging port may be provided. Providing a device configured for transcutaneous charging may be preferred, due to the lower infection risk as compared with providing a percutaneous charging port. However providing a percutaneous charging port may provide alternative advantages such as ease of use, and reduced manufacturing complexity. The blood flow assist device may comprise at least one power supply wire arranged to connect the axial flow rotary pump to a power supply module. The power supply wire may be configured to extend between the electromagnetic stator and the power supply module. The power supply wire may have first and second ends, the first end being configured for electrical connection to the electromagnetic stator of the axial flow rotary pump, and the second end being configured for electrical connection to the power supply module. During implantation of the blood flow assist device, the second end may be disconnected from the power supply, thereby allowing for transluminal delivery of the blood flow assist device to the desired implantation location. The second end of the power supply wire may subsequently be connected to the power supply module in a subsequent procedure. The device may comprise a control module which is configured to control operation of the device. The control module may be provided in the same unit as the power supply module, i.e. these components may share a common housing. The feedback of the device may be mediated by biosensors e.g. sensor which can detect the ventricular depolarisation and / or pressure and frequency of blood pulsations from the heart, and use this to determine how fast the device should rotate. These feedback sensors may be implanted at the proximal and distal ends of the device. There may be internal sensors to the motor which are configured to determine one or more factors relating to device stability, orientation and / or damage. In addition, the device may be configured to respond dynamically to signs of haemodynamic insufficiency. For example, the device may be configured to detect pulmonary impedance e.g. by measuring electrical resistance between the device and the battery site, to detect deterioration in heart function. The device may be further configured to modulate the output power of the device based on the detected level of pulmonary impedance - e.g. if the pulmonary impedance increases above a predetermined threshold, the device may be configured to increase the output of the device, and therefore improve cardiac output. Another aspect of the invention provides a method of treatment of an individual comprising implanting a blood flow assist device as described herein within the body of the individual, such that: (I) the supporting structure of the blood flow assist device is engaged with an interior wall of a vessel within the body of the individual, or (ii) the supporting structure of the blood flow assist device replaces part of a vessel within the body of the individual. Related aspects provide blood flow assist device as described herein for use in a method of treatment of heart failure in an individual, or for use in a method for supporting existing heart function in an individual. The device may be implanted in a patient by any suitable means. Preferably, the device should be insertable via minimally invasive surgery, to minimise surgical risk to a patient. At a broad level, the method may include steps of: establishing arterial access, positioning the device in the implantation location (e.g. in the ascending aorta), and connecting the device to a power supply unit. The step of positioning the device in the implantation location may include separate steps of positioning the electromagnetic stator, and positioning the rotatable impeller assembly within the stator. In some embodiments, the device may be implanted using thoracoscopic techniques, i.e. in a minithoracotomy or median sternotomy procedure. A step-wise description of such a procedure is described below: a) Establish cardiopulmonary bypass: cardioplegia may be established, or bypass and clamping could be performed. b) Perform minimally invasive ascending aorta access e.g. from 2 cm below the right clavicle. c) Resect approximatively 4cm of the ascending aorta and sew the device in. Alternatively implant the device into the inner aortic wall and secure with sutures. d) Connect the device to the powering cables and bring out of the aortic lumen into the mediastinum. e) Remove free air from the lumen of the device, and submerge the device entirely in blood. f) Test run the device for stability and performance. g) Close the aortic lumen with sutures. h) Re-establish endogenous circulation and retest the device for in-blood performance and stability. I) Close the chest cavity and bring the power cables out through the anterior chest wall. j) Create a pocket for the battery in the pre-pectoral fat. k) Secure the device in the pre-pectoral pocket and close the subcutaneous tissue and skin. Alternatively, the device may be implanted via median sternotomy, using techniques similar to ascending aorta replacement / aortic root replacement. The wires may be tunnelled out of the thorax and into the pre-pectoral space, whereby the device can be implanted, using similar techniques as pacemaker battery insertion. In other embodiments, the device may be implanted percutaneously &transluminally delivered to the implantation site. A step-wise description of such a procedure is described below: • Establishing arterial access o Arterial access may be established as per transcatheter aortic valve implantation (TAVI) procedures. An ultrasound guided arterial needle may be introduced into the common femoral artery below the inguinal ligament, aiming for an artery at least ~7mm in diameter. Following this, a wire can be advanced to secure the position, and sheaths introduced to dilate the puncture site. o Alternatively, this could be achieved through arterial cutdown technique with closure suture. • Positioning electromagnetic stator in ascending aorta o High rate pacing: A common technique used with TAVI implantation. A temporary wire is implanted into the right ventricular to pace the heart at high rates, this reduces the stroke volume and blood pressure for a few seconds, so the device is not moved around excessively by the systolic pulsation o Implantation of electromagnetic stator comprising electromagnetic windings supported on a supporting structure &testing of EM field: the supporting structure (which may be provided by e.g. a stent) is transluminally delivered to the implantation location in an insertion configuration. It is then deployed (e.g. via balloon-assisted deployment) to actively stretch the aortic wall. As the wall relaxes onto the supporting structure, it will ‘lock’ in place at a fixed cylindrical size. This is different to classical stents which are typically designed to bend and move with the artery. Instead, the supporting structure used in the present invention is configured to maintain a cylindrical shape upon implantation to provide a stable EM field. o Stop pacing - Once the stator is implanted, the pacing can be stopped o Test stability of stator (this may be achieved e.g. using cough manoeuvres) o Removal of expanding equipment (e.g. balloon). • Connecting the stator to a power supply unit o As the wires will trail behind from the stator, they will need to be tunnelled out e.g. to a pre-pectoral pocket, where pacemakers normally sit. To do this through arterial circulation the following procedure can be followed: ■ Establish pre-pectoral pocket, as is done with traditional pacemaker implantation ■ Either cut-down approach to subcostal artery, or alternate redundant artery (axillary may be appropriate in some cases) ■ Sheath artery, and advance wire with magnetic tip (neodymium magnet) to approach the power supply wire of the EM stator, which has a magnetic portion ■ When the two magnets connect, create slack on the femoral string ■ Gently retract the subcostal wire, pulling the cabling out of the subcostal artery, &once it is free from the subcostal artery, tie off the subcostal artery ■ Secure the power supply wire in place, attach to the power supply module and close pre-pectoral pocket • Positioning rotatable impeller assembly within stator &test performance o Progress rotatable impeller assembly in insertion configuration &expand inside EM stator. This can be a temperature dependent process which simply relies on unsheathing of the structure, with an initial phase of passive expansion, followed by an active phase of balloon expansion. These will form rigid locking stent structures which creates a stable rotor. o Removal of expanding equipment (e.g. balloon). o Performance testing: ■ Test stability, and ECG sensing, and hall monitor sensing -Initial electrical testing to ensure stability of the field and rotor and magnetic bearings ■ Test peak output, and before / after pressure > Haemodynamic testing running at the rate of velocities, active measurement of coronary diastolic pressure and ECG features of ischaemia to ensure coronary perfusion is preserved. • Close any access points (e.g. close femoral wound with femstop, close access point for cutdown technique using stitches, or if Seidinger technique used then angioseal + pressure and subsequent imaging for pseudoaneurysm if necessary) The invention includes the combination of the aspects and preferred features described except where such a combination is clearly impermissible or expressly avoided. Summary of the Figures Embodiments and experiments illustrating the principles of the invention will now be discussed with reference to the accompanying figures in which: Figure 1 shows (a) a schematic longitudinal cross section through a blood flow assist device according to the present invention and (b) an exploded perspective view of the same device. Figure 2 shows a first schematic exploded perspective view of the impeller assembly of the device shown in Fig. 1. Figure 3 shows a second schematic exploded perspective view of the impeller assembly, showing the arrangement of magnets on the impeller hub. Figure 4 is a schematic cross-section through the magnet array of the impeller assembly, showing the direction of magnetisation of each of the magnets in the array. Figure 5 shows (a) a schematic end view; (b) a schematic cross-sectional view; and (c) a schematic exploded perspective view of an end cap of the supporting structure of the device of Fig. 1. Figures 6(a) and (b) show schematic representations of two alternative methods of implantation of the device of Fig. 1. Figure 7 is a graph showing modelled extrapolation of optimal hub ratio for a design of an axial rotary flow pump having a bypass channel, against specific speed of the pump. Figure 8 shows the calculated hub ratio for one embodiment of a suitable rotary impeller. Figure 9 shows various initial design concepts for rotary impellers which may be suitable for use in the present invention. Figure 10 is a graph showing modelled result of how velocity varies along the Z axis in various designs an axial rotary flow pump that vary in bypass radius, modelled at 1000RPM. Figure 11 is a graph showing modelled result of how velocity varies along the Z axis in various designs an axial rotary flow pump that vary in bypass radius, modelled at 2100RPM. Figure 12 is a graph showing the effect of RPM on Mass flow rate at various impeller speeds (rad / s): 62, 225, 104.72 and 162 Figures 13(a), (b) and (c) are graphs showing the velocity profile across the length (z-axis) of the pump for designs using 3, 4 and 5 blades respectively. Figure 14 shows a schematic longitudinal cross section through a blood flow assist device which employs a twin bearingless motor design according to the present invention. Detailed Description of the Invention Aspects and embodiments of the present invention will now be discussed with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned in this text are incorporated herein by reference. Fig. 1-6 show various views of a blood flow assist device 1 according to the present invention, or of various sub-components of the blood flow assist device 1. Fig. 1 shows (a) a schematic longitudinal cross section through a blood flow assist device according to the present invention and (b) an exploded perspective view of the same device. The blood flow assist device 1 comprises an axial flow rotary pump comprising a rotatable impeller assembly 3 and an electromagnetic stator assembly 5 operable to drive rotation of the impeller assembly. The embodiment shown is a non-expandable / collapsible embodiment. That is, the stator assembly provide a substantially rigid structure on which the electromagnetic windings are supported. In view of this, the arrangement shown is generally best adapted for replacement of a segment of a vessel in a human body - and, more particularly, is most preferably adapted for replacement of a segment of the ascending aorta in a human heart - further specific discussion on how the device may be implanted in a vessel for use is provided below in relation to Fig. 6(a) and Fig. 6(b). The immediately following discussion will focus on the form and structure of the device. The electromagnetic stator 5 comprises a supporting structure including: core stack 7 which comprises a plurality of laminations of a material having high magnetic permeability; a plurality of threaded bars 9 extending through said core stack; and first and second end caps 11, configured for engagement with the threaded bars 9 to fix the stator assembly together. A plurality of electromagnetic windings (not shown) are supported on the supporting structure. They are secured to the core stack via an adhesive / potting material which encloses the windings and provides stability to their conformation. The rotatable impeller assembly 3 comprises a hub 13 and a plurality of blades 15. The hub 13 comprises inner 17 and outer 19 cylindrical portions, with the blades 15 extending between these portions. In the embodiment shown in these figures, four helical blades are conveniently provided. In the arrangement shown, the pitch of the blades is selected to be approximately 1 full rotation in 200mm, although other arrangements are contemplated. The inner cylindrical portion 17 defines a bypass flow conduit 21. The presence of a bypass flow conduit 21 may allow for an increase in the mass flow rate possible during use of the blood flow assist device, and may additionally decrease the exit velocity of blood flowing through the blood flow assist device, thereby leading to increased pump efficiency, allowing the pump to operate at comparatively lower RPMs as compared with conventional pump designs. This design also allows for a failsafe aspect of the design, as endogenous blood flow can continue via this bypass conduit if the device fails to rotate. The outer cylindrical portion 19 engages with a housing portion 23 of the hub. The outer cylindrical portion 19 is fixed to an inner cylindrical surface of housing portion 23. The housing portion 23 is further arranged to support a magnet array 25, best seen in relation to Fig. 2, Fig. 3 and Fig. 4. The magnet array comprises a plurality of magnets organised as a Halbach array, as shown in Fig. 4, where the direction of magnetization of each magnet in the magnet array is indicated with an arrow. The magnet array sits within a recess provided on housing portion 23 of the hub and are held in place by sheath 27 (which may be referred to as an ‘impeller sheath’. The impeller sheath 27 provides an external surface of the impeller assembly. The impeller sleeve may be formed from a heat-shrink polymer, for ease of manufacture, or a thin metal sheath. Whilst in this embodiment the housing portion 23 of the hub and the outer cylindrical portion 19 are separate components, it is contemplated that other embodiments may utilise a single outer cylindrical portion which may directly support the magnet array 25. The impeller assembly further comprises impeller end caps 29 which comprise magnetic bearings for stabilisation of the impeller. Each end cap comprises a circular shim 31 having a substantially L-shaped cross-sectional profile. A plurality of magnets 33 having a substantially triangular cross-section (in practice, a truncated triangle cross-section, as can be seen most clearly in Fig. 1) are arranged to sit within the recess provided by the L-shaped profile of the shim - in the figures this plurality of magnets is visually represented as a single magnetised ring, but in practice may conveniently be provided as a plurality of individual magnets that collectively act as a magnetised ring, with the axis of magnetisation being arranged as approximately 45° to the longitudinal axis of the device. The axis of magnetisation is selected to be parallel to a direction that is substantially perpendicular to a major face of the magnets, about the circumference of the end cap. The magnets provided on the impeller end caps 29 interact with corresponding magnets provided on each of stator end caps 11. Ina corresponding manner to the impeller end caps, the stator end caps 11 each comprise a circular shim 35 having a substantially L-shaped cross-sectional profile. A plurality of magnets 37 having a substantially triangular cross-section are arranged to sit within the recess provided by the L-shaped profile of the shim - in the figures this plurality of magnets is visually represented as a single magnetised ring, but in practice may conveniently be provided as a plurality of individual magnets that collectively act as a magnetised ring, with the axis of magnetisation being arranged as approximately 45° to the longitudinal axis of the device. The axis of magnetisation is selected to be parallel to a direction that is substantially perpendicular to a major face of the magnets, about the circumference of the end cap. In use, when the impeller assembly and stator assembly are assembled together to form the device, the magnets 33 on each impeller end cap 29 and the magnets 37 on each stator end cap are arranged such that their axes of magnetisation oppose one another. In this manner, the magnets on the impeller end caps and the stator end caps collectively act as passive magnetic bearings which stabilise the impeller assembly within the stator assembly. It will be understood that the illustrated arrangement is just one suitable method of stabilising the impeller - in other embodiments, it is contemplated that active bearings or other suitable stabilising arrangements could be employed. Fig. 5(a), (b) and (c) are figures showing the stator end cap 11 in greater detail. Fig. 5 (a) is a schematic end view, Fig. 5 (b) is a schematic cross-sectional view (taken along line A-A of Fig. 5(a)), and Fig. 5 (c) is a schematic exploded perspective view of a stator end cap 11. The end cap has an end cap body 39, on which shim 35 is supported - Fig. 5(b) and (c) show the shim in an exploded configuration relative to the end cap body 39. The end cap body has a plurality of threaded holes 41 formed therethrough (eighteen) for receipt of end portion of threaded bars 9. In this manner, end stator end caps on each end of the device can act to hold the device together via the threaded bars. The end cap bodies have an internal diameter that is smaller at the smallest point than the largest external diameter of the impeller assembly. In this manner, as the impeller assembly is wider than the gap that each end cap provides, the impeller assembly is trapped between the end caps, and it is not possible for the impeller assembly to move outside of the stator assembly without disassembly of the device. This acts as another failsafe preventing unwanted movement of the impeller assembly after implantation. Fig. 6(a) and (b) show schematic representations of two alternative methods of implantation of the device of Fig. 1 within a vessel 100 of a human body (e.g. within the ascending aorta of a human heart). Fig. 6(a) shows an arrangement in which the device 1 is configured to be disposed within the vessel 100, and engage an interior wall of the vessel. Here it can be seen that the largest outer diameter of the device is selected to substantially correspond with the inner diameter of the vessel. In this way, the device can fit within the vessel whilst providing a close fit with the vessel walls. This can minimise the risk of the device becoming misplaced within the vessel after implantation. Fig. 6(b) shows an alternative arrangement, in which the device is configured to replace a segment of the vessel on implantation. In this arrangement, the device comprises a sleeve 43 which can be sutured to the free edges of these remaining sections of the vessel (e.g. aorta) - other arrangements are contemplated, but it is suggested that this arrangement offers a particularly convenient mode of attachment of the device to the vessel. Techniques for attaching a sleeve in this manner are generally known from e.g. techniques used in aortic root replacement surgery. An arrangement for supplying power to the device is not shown in the above figures, however in one convenient arrangement, wires for connecting to the electromagnetic windings may extend through one or more end caps of the stator assembly. And may extend to a power supply module (e.g. to a transcutaneously charged battery that sits under the skin). Fig. 14 shows a schematic longitudinal cross section through a blood flow assist device according to the present invention which employs a twin bearingless motor design. The blood flow assist device 100 comprises an axial flow rotary pump comprising a rotatable impeller assembly 103 and an electromagnetic stator assembly 105 operable to drive rotation of the impeller assembly. The embodiment shown is a non-expandable / collapsible embodiment. That is, the stator assembly provide a substantially rigid structure on which the electromagnetic windings are supported. In view of this, the arrangement shown is generally best adapted for replacement of a segment of a vessel in a human body - and, more particularly, is most preferably adapted for replacement of a segment of the ascending aorta in a human heart. The electromagnetic stator 105 comprises a supporting structure including a core stack 107 which comprises a plurality of laminations of a material having high magnetic permeability such as iron. Two separate sets of electromagnetic windings 108a,b are supported on the supporting structure, shown schematically as separate blocks in this figure. A thin sleeve 110 is provided to protect the windings (located radially inwardly of the windings). The rotatable impeller assembly 103 is broadly similar to that described in relation to the arrangement shown in Fig. 1. Similarly to that arrangement, the rotatable impeller assembly 103 comprises a central bypass flow conduit 121, which has the same advantages as described above. One difference is that the rotatable impeller assembly 103 comprises two set of magnet arrays 125a, b, each comprising a plurality of magnets arranged circumferentially amount an outer circumference of the impellor (vs the one set of magnets provided in the magnet array of the arrangement of Fig. 1). The magnet arrays 125 a, b are held in place by a sheath 127 which conveniently comprises a thin steel sleeve. The sheath holds the magnets in place, and conveniently also presents a target for non-contact position sensors 112 - whilst this figure shows two position sensors being provided at each end of the device, it will be appreciated that in fact a plurality of such sensors can be provide at each end (e.g. four or more at each end). One difference between this arrangement and the arrangement shown in Fig. 1 is that an additional bypass conduit 122 is provided which circumferentially surrounds the rotatable impeller assembly 103. (i.e. which is located radially outwardly of the rotatable impeller assembly, but radially inwardly of the stator assembly). This is possible due to the lack of end caps in this deice design: the open-ended edges of a bearingless motor configuration allow for provision of this additional bypass flow path radially outwardly of the rotatable impeller assembly. Providing an additional bypass flow path in this manner may allow for a further increase in the mass flow rate possible during use of the blood flow assist device. This can in turn improve temperature control of the device, and additionally may reduce the thrombotic risk within the device. It has been found that the use of a twin bearingless configuration for the device, as shown in this schematic figure may offer significant advantages over both a passive magnetic and conventional bearing approach, for a number of reasons including improved stability for the impeller assembly, and the possibility of providing the additional circumferential bypass flow between the impeller assembly and the stator assembly. Feasibility Modelling To assess the feasibility of a blood flow assist device as described in the present disclosure, and determine various parametric features of the device, a mathematical study was done, evaluating a proposed design specification against a mathematical approach as outlined in “Centrifugal and Axial Flow Pumps 2nd Edition by A.J. Stepanoff ISBN-13 : 978-0894647239”. The book states that there are four key steps in the procedure for developing an appropriate impeller. These are as follows: (1) Calculation of head capacity and specific speeds From C&A, the initial stage of the pump is to equate a desired head capacity. Then to select a desired speed for the impeller (revolution per minute [RPM]). This can then be used to equate specific speed of the impeller according to the following equation: Where Ns is the specific speed, RPM is the desired revolutions per minute, Q is the discharge rate and Ht is the total head. The specific speed of the impeller is a ratio between the desired RPM and pump head. Through experimental testing, it has been found that by using the pump specific speed, hub ratio, total efficiency, vane spacing and vane angle of attack [AOA] can all be roughly determined. (2) Selection of hub ratio and vane spacing after a specific speed has been obtained, a decision can be made about a hub ratio and vane spacing. Through practical experiments, this is a standardised decision made of empirical data. This gives a rough estimate to start and can be altered as evaluation of the design develops. (3) Selection of speed and capacity constants In much the same way that hub ratio and vane spacing is chosen, speed and capacity constants can be selected from a standardised set of empirical data related to the specific speed calculated. (4) Selection of blade / vane profiles After speeds and dimensional constraints have been picked (RPM and bypass ratio), Euler’s velocity profiles can be developed. This will give AOA of both incoming and exiting flow. From this the AOA of the impeller chord can be decided, allowing for the foil design to be undertaken. Vane Specifics As explained in (4) above, a vanes AOA is picked once Euler’s velocity triangles are solved. There should then be further design work on specifics such as, vane setting angle at the outside diameter, chord spacing ratio [Z / t] and number of vanes. This is done with experience and previous experimental data collected by previous tests. Due to equation [2] identified below it can be seen that the AOA of the vane changes with distance from the hub. Therefore, the setting degree at the outer edge has been studies and related to specific speed. Equation [2] show that for greater values of velocity (v and vave) the value for C; reduces. As apparent velocity will increase with the radius (tip speed is >central bypass speed). This implies a reduction in AOA along the radius to decrease the lift coefficient. 1 V / £ \v£ve) hJ \sin sin ^ave + X) ) [2] Hub ratio 0.702 0.573 0.426 Total efficiency, per cent 84 85 82 Vane setting degrees at outside diameter 20 21 23 Specific speed 5550 8500 14750 Dimensionless specific speed 1.133 1.7 2.46 Table 1: Performance data for a range of Ns As seen in Table 1, as specific speed reduces, so does the setting angle for the vanes at the outside diameter. Specific speeds tested for such parameter stop at Ns = 5550 as much lower than this, centrifugal pumps are used. Despite this, the correlation implies that the angle would continue to reduce with speed. Chord spacing ratio The chord-spacing ratio is a ratio of the distance between vanes [t] and the chord length [ / ]. As axial pumps increase specific speed, the ration | falls closer to 1. Once specific speeds exceed that of 10,000, the ratio will often fall below unity. The choice of l / t has no numerical method to be equated but must be picked by the experience of the engineer. There have been numerous experiments correlating relations between the number of vanes and bypass ratio. These experiments have shown a relationship between lower specific speeds and increase in bypass ratio, number of vanes and vane spacing. This also implies a lower AOA for the vanes as to allow for the increase in number. Vane spacing can be numerically calculated once a choice of the number of vanes has been made. tiD [3] Where Z here is the number of vanes desired. Number of vanes Similarly, to the chord spacing ratio, there is no numerical way to calculate the most efficient number of vanes. Various pieces of literature have shown through experimental tests that a projected frontal area of 63 % shows the highest efficiency. It has then further been shown that the number of vanes should increase with a reduction of specific speeds. Pump mathematics - picking design parameters Fig. 7 is a graph showing the extrapolation of performance data for an axial rotary flow pump based on Ns (Ns being specific speed of the pump). Specifically, Fig. 7 is a graph showing modelled extrapolation of optimal hub ratio for a design of an axial rotary flow pump having a bypass channel, against specific speed of the pump. As can be seen the hub ratio seems to extrapolate in an almost linear mode to represent a hub ratio of approximately 0.836. Noting that the hub ratio is defined herein as a ratio of the impeller bypass diameter to the outer diameter of the impeller assembly. Dh — = HR 4 Do Dt -!-= 0.836 ••• Dh = 0.025 5 0.03 h Fig. 8 is a figure showing the calculated hub ratio for one embodiment of a suitable rotary impeller. The outer diameter of the impeller assembly is 30 mm. The diameter of the bypass is 21 mm. The lowest reasonable data from axial pumps at a specific speed of 5550 was used as the starting design point. Pump Geometry decisions The decision was made to set the geometry based on the lowest specific speeds used in axial pumps. The following design parameters were chosen: Design Parameters Hub ratio 0.702 Total efficiency, per cent 84 Vane setting degrees at outside diameter 20 Specific speed 5500 The next stage is to develop the AOA of the vane where it meets the hub, which will be done using Euler’s velocity triangle. To calculate this velocity triangle, the angular velocity induced at the edge of the bypass will be taken as the x component of the velocity and the free stream velocity will be taken as q — m x r 6 m = RPM x 0.1 = 600 x 0.1 = 60 7 0.021 / m\ c, = 60 X----= 0.63(-) 8 2 's ' vrei ~ + ci “ V1 + 0.63 = 1.28 —) 9 / 0.63\ «rei = (r™) = 29.48° 10 xl.Zo / Where arel is the angle between vrel and Knowing the angle of incoming air gives a good approximation for an air foil with zero AOA from the relative velocity. This can later be iteratively developed with CFD modelling till an optimum value is found. The number of vanes can now be found knowing that from experimental data 5 vanes is most efficient for low-speed axial pumps. This allows for vane spacing to then be calculated from equation [3], nx 0.021 = 0.013m 11 From equation [2] it was shown that C; can be related to the apparent velocity and free stream velocity. This is then simplified in C&A to the following: Where: CL rel gHt Cu2 = = 0.1662 't= — = 0.0288m eh With an assumed eh = 85%. 2 x 0.1662 1.28 Then assuming a zero AOA with respect to the relative velocity of oncoming air, I can be equated allowing for C; to be found. 12 13 14 15 0.03 I =-------;------ = 0.034 15 cos cos (29.48) 0.26 -^= 0.091132 = 6, 16 V0.013 / As it is known that the C, value will change with position along the radius, the value needed along the wing will be found. With the conclusions from this modelling, the initial design was done. The philosophy led to several design ideas, as shown in Fig. 9. The various design ideas contemplated included: impellers having a diffuser, or having no diffuser; impellers having a bypass channel, or having no bypass channel; and impellers having a range of different blade pitches. Due to the mechanism of integration back into the device, no diffuser was preferred, and an axial flow rather than helical configuration was preferred for the impeller blade shape. Following generation of these initial design ideas, computational flow modelling was performed. A computational domain was developed as to envelop the entire model and allow for downstream flow properties to be analysed. The domain size was as follow:32 x 32 x 80 . The pump is placed with the centre of the pump at 15.5 x 15.5 x 30 as to allow for flow properties downstream of the pump to be observed. The fluid properties are entered into CFX, Fluent and solid works as follows: Value Unit Name Blood Density 1003 kg / mA3 Specific Heat 4182 j / (kg*k) Thermal conductivity 0.6 W / (m*k Viscosity Power law model Consistency Coefficient 0.012171 Pa*s Max Dynamic Viscosity 0.012171 Pa*s Min Dynamic Viscosity 0.003038 Pa*s power law Index 0.79981 A parametric study was then performed, within SolidWorks flow simulation, to compare specific design parameters efficiently. The study was run with varying geometry (specifically, varying bypass radius) at varying RPM. The range of RPM was from 600 RPM to 2100RPM. For each geometry change, mass flow rate, max velocity and a velocity profile along the pump’s z axis was recorded. To help visualise flow properties, a cut plot was produced, showing key areas of interest. Goal Design 1 Design 2 Designs Design 4 Designs Design 6 Design 7 Designs Design 9 Design 10 (Value) Bypass radius 5 6 7 8 9 10 11 12 13 14 [mm] The first parametric study was performed with varying bypass channel sizes, ranging from 10mm to 26mm in diameter (radius of from 5-14 mm). These 10 variants were repeated at 4 points of RPM in the range of 600 RPM to 2100RPM. The results are shown in Fig. 10 and Fig. 11, which are graphs showing velocity along Z at 1000 RPM and 2100 RPM respectively. These figures show how the velocity changes along the z axis of the pump for each hub bypass diameter tested. It can be seen that the inlet speed at both RPMs are roughly 1.2m / s. It should be noted that there is small negative changed in both inlet speed and mass flow rate for the smaller hub bypass diameters. This is due to the smaller bypass creating a back pressure if not of sufficiently large. Further to this, it is important to acknowledge that this is a measurement of the velocity through the bypass section of the pump. The fluid enters the computational domain at z = -0.03m, depending on the size of the hub, the fluid either starts to accelerate towards the pump, or reduces in speed as explained. There is then a significant increase in speed as the fluid enters the pump. As the pump starts to diverge (at around z = 0.007 m) the fluid starts to decelerate, this deceleration becomes steady at around z = 0.015, which is the final exit of the pump into the residual flow. With further analysis, it can be noted that the greatest exit velocity is generate by design point 4, 5 and 6. These points correlate to 16,18 and 20mm hub bypass diameter. It should also be taken note that an increase in RPM did not correlate to a notable increase in exit velocity but did lead to an increase in the internal velocity through the pump. Fig. 12 is a graph showing the effect of RPM on Mass flow rate at various impeller speeds (rad / s): 62, 225, 104.72 and 162. This shows how the mass flow rate increase with RPM. There is a minimal but not notable increase in efficiency 600RPM to 1000RPM, with a notable increase of mass flow rate at 1500 RPM. However, there is a significant increase when the pump is sped up to 2100RPM. Blade geometry An initial test is performed comparing a pitch of 30 x 1 vs 100 x 0.3. The original pitch had only one blade, while a reduced pitch uses 3. From a review of the flow dynamics it was found that the use of three blades of shorter pitch resulted in more even velocity distribution, with a significant increase of peak velocity from 2.07m / s to 3.358 m / s. It also showed a significant increase in mass flow rate with an increase from 0.4 kg / s to 1,09kg / s. It was therefore theorised that an increase in blade number with a decrease of pitch may provide for improved performance. This was tested. Fig. 13(a), (b) and (c) show respectively the effect of number of blades on the velocity profile across the length (z-axis) of the pump - Fig. 13(a) is a graph for a design with 3 blades, Fig. 13(b) is a graph for a design with 4 blades, and Fig. 13(c) is a graph for a design with 5 blades. As can be seen from these graphs, the number of blades does not have a significant effect on the velocity profile across the length of the pump, however it can be seen that an increase in blade number from 3 to 4 does lead to a mild increase in exit velocity but with a lower max velocity. It has also been noted that an increase in blade number also lead to a decrease in total maximum mass flow rate. Increasing the number of blades from 4 to 5 shows no significant increase in either velocity or mass flow rate. From this we can conclude that the optimal number of blades may be 3. A further study into the effect of the pitch was then performed, using 3 blades (due to it showing the highest mass flow rate) and varying the pitch from 50 to 250 mm. Pitch is defined as the length in mm for 1 full rotation of the vane about the hub. This analysis can then be used to pick the most efficient blade geometry. The results are set out in the table below: Pitch (mm) Max Vel (m / s) @225 rad / s 50 2.112 100 2.327 150 2.559 200 2.761 250 2.755 300 2.733 At 225rads / s (-2100 RPM), it was seen that the maximum velocity increased with increasing pitch until the pitch exceeded 200. Past this we see a decrease in maximum velocity, implying a decrease in efficiency. From this we can conclude that the optimal pitch of the blades may be around 200, where the intended RPM of the device is -2100 RPM. Pitch (mm) Max Vel (m / s) @62 rad / s 50 1.621 100 1.592 150 1.554 200 1.517 250 1.557 300 1.578 A similar set of results are seen at the proposed design speed of 600 RPM, but this time with a decrease in performance as soon as the pitch increases from 50. This implies that the optimum pitch is reliant on the specific speed of the pump. After this, it was questioned whether there is also an optimum number of blades depending on both specific speed and chosen pitch. The below table show the change in maximum velocity with no. of blades in a further design iteration. Max Vel No. of blades (m / s) 3 1.621 4 1.874 5 1.665 It was observed that the higher RPMs generated greater acceleration at lower pitches, with fewer blades. While, at lower speeds, a much higher pitch and more blades were optimum. Conclusions The feasibility modelling successfully shows that it is theoretically feasible to produce a blood flow assist device within the design parameters given to meet, and if not exceed, the output requirements. It is considered from the above analysis that an optimum RPM of such a device during operation may be around 2000 RPM. This may allow for greater efficiency of the device, and lower shear on the blood due to the significant increase in fluid velocity observed at such RPMs. Inclusion of a bypass flow section of the pump is found to be advantageous in allowing the pump to run efficiently at such comparatively low RPMS, as well as in allowing for the flow of blood to not be blocked by the pump and only assisted. Whilst the above modelling was performed assuming no diffuser portion on the impeller, a diffuser may be a possible future consideration if it is deemed necessary to increase the pressure of the blood instead of just the velocity and mass flow rate. *** The features disclosed in the foregoing description, or in the following claims, or in the accompanying drawings, expressed in their specific forms or in terms of a means for performing the disclosed function, or a method or process for obtaining the disclosed results, as appropriate, may, separately, or in any combination of such features, be utilised for realising the invention in diverse forms thereof. While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope of the invention. For the avoidance of any doubt, any theoretical explanations provided herein are provided for the purposes of improving the understanding of a reader. The inventors do not wish to be bound by any of these theoretical explanations. Any section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described. Throughout this specification, including the claims which follow, unless the context requires otherwise, the word “comprise” and “include”, and variations such as “comprises”, “comprising”, and “including" will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps. It must be noted that, as used in the specification and the appended claims, the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from “about” one particular value, and / or to “about” another particular value. When such a range is expressed, another embodiment includes from the one particular value and / or to the other particular value. Similarly, when values are expressed as approximations, by the use of the antecedent “about,” it will be understood that the particular value forms another embodiment. The term “about” in relation to a numerical value is optional and means for example + / - 10%.

Claims

:

1. A blood flow assist device comprising an axial flow rotary pump, said pump comprising a rotatable impeller assembly comprising a hub and one or more impeller blades, and an electromagnetic stator operable to drive rotation of the impeller assembly;wherein the electromagnetic stator assembly comprises electromagnetic windings supported on a supporting structure, said supporting structure being configured for engagement with an interior wall of, or replacement of a segment of, a vessel of a human body during use,wherein the blood flow assist device has a single- or twin-bearingless motor configuration and is configured to provide stabilisation of the impeller assembly across 5 degrees of freedom.

2. The blood flow assist device according to claim 1 wherein the supporting structure has a substantially fixed external circumference, and wherein the device is configured to replace a section of the vessel during use.

3. The blood flow assist device according to claim 2 wherein the supporting structure has a maximum external diameter in a range of from 2 to 5 cm.

4. The blood flow assist device according to claim 2 or claim 3 wherein the device comprises one or more sleeve portions configured to be attached to free edges of the vessel either side of the segment of vessel to be replaced.

5. The blood flow assist device according to claim 1 wherein the supporting structure is an expandable supporting structure, said structure being expandable from an insertion configuration in which the structure is transluminally deliverable, to a deployed configuration in which the structure is configured to engage with an internal wall surface of a chamber or vessel of a human body.

6. The blood flow assist device according to claim 5 wherein when the supporting structure is in a state of being engaged with an interior wall of the vessel during use, the supporting structure has a diameter in a range of from 2 to 5 cm.

7. The blood flow assist device according to claim 5 or claim 6 wherein in the insertion configuration, the maximum diameter of the expandable supporting structure is 8 mm or less.

8. The blood flow assist device according to any one of the preceding claims wherein the rotatable impeller assembly is moveable between an insertion configuration and a deployed configuration, wherein in the insertion configuration, the one or more impeller blades radially extend by a smaller amount relative to the radial extent of the one or more blades in the deployed configuration.

9. The blood flow assist device according to any one of the preceding claims wherein the rotatable impeller assembly and the electromagnetic stator assembly act as a motor, and wherein the mode of operation of the motor is selected from a brushless DC motor (BLDC), an AC induction motor, or a permanent magnet synchronous motor.

10. The blood flow assist device according to any one of the preceding claims wherein the rotatable impeller assembly comprises one or more permanent magnets, optionally wherein the magnets are organised as a Halbach array.01 04 2511. The blood flow assist device according to any one of the preceding claims wherein the blood flow assist device comprises one or more bypass flow conduits, optionally wherein: (i) the bypass flow conduit lies along a central longitudinal axis of the rotatable impeller assembly and / or (ii) the bypass flow conduit circumferentially surrounds the rotatable impeller assembly.

12. The blood flow assist device according to claim 11 wherein the bypass flow conduit has an area measured in a cross-section perpendicular to the flow direction through the device in a range of from 1 -2 cm2.

13. The blood flow assist device according to any one of the preceding claims wherein the one or more blades comprise a flexible fabric material, an elastomeric material or a metal material, optionally wherein the blades are formed from Dyneema® silicone rubber, Pebax® plastic, stainless steel, or titanium.

14. The blood flow assist device according to any one of the preceding claims wherein the device is configured such that the rotatable impeller assembly is magnetically levitated within the electromagnetic stator assembly during use.

15. The blood flow assist device according to any one of the preceding claims wherein the windings comprise a material selected from: silver or copper, optionally wherein the windings comprise a coating configured to improves blood compatibility of the winding material.

16. The blood flow assist device according to any one of the preceding claims wherein the blood flow assist device comprises a power supply wire comprising first and second ends, the first end being connected to the electromagnetic stator of the axial flow rotary pump, and the second end being configured for electrical connection to the power supply module.

17. The blood flow assist device according to any one of claims 1 to 16 wherein the electromagnetic stator assembly comprises first and second end caps configured to be fixedly connected to one another to secure one or more components of the stator assembly together.

18. The blood flow assist device according to claim 17 wherein each of the end caps have a smallest internal diameter that is smaller than the largest external diameter of the impeller assembly.

19. A blood flow assist system comprising the blood flow assist device of any one of claims 1 to 18, and a power supply module configured for connection to the blood flow assist device to supply power to the device.

20. The blood flow assist system according to claim 19 wherein the power supply module comprises a rechargeable battery and is configured for transcutaneous charging.

Images