METHOD AND CARDIAC SUPPORT SYSTEM FOR DETERMINING OUTFLOW PRESSURE

Description

[0001] The subject matter of this application is a method for determining an outlet pressure present at an outlet of a cardiac assist system according to claim 1 and a cardiac assist system with a pump and a cannula connectable to the pump according to claim 9.

[0002] The heart support systems listed in this application are so-called ventricular assist devices (VADs). These can be designed for the left ventricle (LVAD), the right ventricle (RVAD), or both ventricles (BVAD). The subject matter of this application is described below primarily in the context of LVADs, although the respective pumps can also be used as RVADs or BVADs.

[0003] When a heart assist system is operated as an LVAD, the heart assist system is usually connected to the left ventricle and the aorta, bypassing the aortic valve. This creates a bypass that can pump blood through the heart assist system from the left ventricle into the aorta. In many cases, the heart assist system comprises a pump unit and an inlet and outlet cannula. Other variants involve coupling the pump unit directly to the heart and pumping the blood pumped by the pump unit from the pump outlet to the aorta via a cannula. Suitable pumps for this purpose are rotating blood pumps, which can be designed either as axial pumps or radial pumps. Axial pumps are pumps that pump the blood axially, i.e. they pump the blood using a predominantly axial propulsion force. Pumps for axial blood pumping can be connected to a radial outlet.In contrast to axial pumps, radial pumps have rotors that impose both an axial and a radial component on the blood. In many cases, the rotor expands radially, so that the rotor diameter at the pump inlet is usually smaller than the rotor diameter near the pump outlet. Radial pumps are sometimes also referred to as turbopumps.

[0004] To ensure safe operation of the ventricular assist system for the patient, data is continuously collected to measure the amount of blood delivered and the prevailing pressures.

[0005] US 2003 / 0045772 A1 describes a heart pump in which the speed of the pump is adjusted based on the inlet pressure measured in an inlet cannula.

[0006] One option for measuring the pressures relevant for ventricular assist systems is to incorporate absolute pressure sensors into the ventricular assist system. Relevant pressures in an LVAD system include ventricular pressure, pv, and aortic pressure, p ao. Ventricular pressure can be measured directly in the ventricle. In ventricular assist systems where the pump unit is connected directly to the ventricle, the pressure measured at the pump inlet is essentially equivalent to the ventricular pressure. Aortic pressure is more difficult to measure. One option is to place a pressure sensor in the aorta, for example. Due to the pressures, flow velocities, and difficulty accessing the aorta, anchoring a sensor is difficult and poses a significant health risk.

[0007] Therefore, there is a need to determine the aortic pressure or the pressure at the outlet of a cardiac assist system in another way. In addition to the method according to claim 1s and a cardiac assist system according to claim 9, the problem is also solved by arranging an absolute pressure sensor at the outlet of the cannula.

[0008] According to a method for determining an outlet pressure present at an outlet of a cardiac assist system, an inlet pressure present at an inlet of the cardiac assist system is first determined. The inlet pressure can be determined, for example, using a pressure sensor located at the inlet of the cardiac assist system. Alternatively, a pressure sensor separate from the cardiac assist system can be used, such as those used in pacemaker systems or implantable defibrillator systems. The recorded pressure data can then be transmitted wirelessly, for example, to a control unit of the cardiac assist system.

[0009] To determine the outlet pressure, a pressure difference between the inlet and outlet of the ventricular assist system is estimated. This means, in particular, that subsections of the ventricular assist system can be measured or estimated. Estimating a pressure difference between the inlet and outlet of the ventricular assist system can therefore be understood as estimating at least a pressure drop across a subsection of the ventricular assist system. The term "estimate" is to be understood here as meaning that the pressure drop or pressure difference is determined using variables that are not directly measured in this section.

[0010] In one embodiment, the heart assist system comprises a pump with a pump inlet and a pump outlet, and a cannula with a cannula inlet and a cannula outlet, the pump outlet being connected to the cannula inlet. The pressure difference between the inlet and the outlet of the heart assist system is composed of a pump pressure difference between the pump inlet and the pump outlet and a cannula pressure difference between the cannula inlet and the cannula outlet. The cannula connects the pump to the aorta and can, for example, comprise or be formed from a biocompatible material such as silicone or a graft material. Since the cannulas often have to be individually adapted to the patient during implantation, i.e. the cannulas may be shortened during implantation, the final length of the cannula is often only known after implantation.Using the methods and the ventricular assist system described in this application, the cannula pressure difference can be estimated using parameters or variables measured by the pump. Once the pressure difference between the inlet and outlet of the ventricular assist system has been estimated, the outlet pressure can then be determined from the inlet pressure and the estimated pressure difference.

[0011] Alternatively, in an example not part of the claimed invention, a pressure sensor for determining the aortic pressure may be provided at the outlet of the heart assist system connected to the aorta. If a pressure sensor is now arranged at the cannula outlet, it would be removed if necessary. Alternatively, in an example not part of the claimed invention, a pressure sensor may be provided within the cannula. However, the pressure measured in the cannula may differ from the aortic pressure, so that in this case, a pressure drop in the cannula section between the pressure sensor and the cannula outlet should again be estimated.

[0012] A cardiac assist system according to this application comprises a pump and a cannula connectable to the pump. Furthermore, the cardiac assist system comprises a pressure sensing system that is configurable to detect a pressure at the pump inlet and a pump pressure difference between the pump inlet and pump outlet. A control unit is present that is configured to estimate a cannula pressure difference between the cannula inlet and the cannula outlet. As already mentioned, the cannula pressure difference can be estimated based on variables that can be detected by the pump, its pressure sensing system, or other sensors of the pump.

[0013] The control unit, which can be embodied, for example, as a microprocessor, microcontroller, or programmable field array, is equipped with a command set or software with instructions that outputs an estimated value for the cannula pressure difference using the measured values ​​acquired by the pump and a model for estimating the cannula pressure difference. The commands or the software or firmware can be stored in a memory unit and / or can be stored and transferred to a control unit via a data carrier such as a CD, DVD, or portable flash memory. The control unit is configured such that the cannula pressure difference is calculated from variables determined by the pump using the model for the cannula pressure difference and the calibrated model parameters associated with the model.Furthermore, in some embodiments, the control unit comprises commands for carrying out a method for calibrating the model parameters.

[0014] In one embodiment, the model for estimating the cannula pressure difference can include one or more of the following pump-detectable variables as independent variables: volume flow, pump speed, or the pump pressure difference between the pump inlet and pump outlet. The cannula pressure difference or the cannula outlet pressure is estimated as the dependent variable.

[0015] In order to adapt a model for estimating the cannula pressure difference to a patient, it may be necessary to calibrate the model parameters using measurement data. The model parameters can be modeled using statistical methods, such as regression analysis. The parameters can be modeled during continuous pump operation, calibrated at regular maintenance intervals, or calibrated once, for example, shortly after implantation. The model parameters can be calibrated either by the control unit or by another, external control unit. The model parameters can be modeled time-independently or time-dependently.In the event that the calibration is repeated once or at regular intervals, the model parameters determined using statistical methods are stored in a memory connected to the control unit, such as a RAM or ROM, and used to determine the cannula differential pressure.

[0016] As already mentioned at the beginning, the cannula pressure difference can be estimated based on various variables detected by the pump. For example, a first variant can be used to estimate the cannula pressure difference for times at which the aortic valve is open, in particular at times shortly after the aortic valve opens and before the aortic valve closes. At these times, the cannula pressure difference is approximately equal to the pump pressure difference. For this first variant, no further parameter needs to be determined other than the time of aortic valve opening. The disadvantage of this method, however, is that the estimate is only satisfactorily accurate at a few times (shortly after the aortic valve opens).

[0017] In a second variant, the cannula pressure difference can be estimated using the volume flow delivered by the pump. For example, the cannula pressure difference can be estimated over a physiologically meaningful range using a model of a non-linear function, such as a polynomial or an exponential function, or a linear function. The volume flow, the pump speed, the pump pressure difference, or combinations of these variables, or variables proportional to these variables, can serve as independent variables. For example, a polynomial or an exponential function of the pump pressure difference and the rotor speed can be used to estimate the cannula pressure difference. Furthermore, with some prior art pumps, it is possible, for example, to determine the pump pressure difference based on a current applied to a control coil, whereby the control coil determines the position of a pump rotor.Since the position is changed due to the applied pressure or the prevailing volume flow through the pumped fluid, the current flowing through the control coil is proportional to a pump pressure difference.

[0018] In a third variant, the model of the cannula pressure difference can have a differential equation or a system of differential equations.

[0019] What the second and third variants have in common is that the model includes model parameters, which are determined from measurement data using statistical methods, for example. Examples of methods for calibrating the models can be found in the working examples.

[0020] In another embodiment of the cardiac assist system, it includes a volume flow sensor. The volume flow sensor can be used to directly determine the volume flow. For example, an ultrasonic flow sensor, a Prandtl probe, an orifice plate, or a magnetic inductive flow sensor can be used as a volume flow sensor. A pump system for indirectly measuring the volume flow, for example, by moving a rotor, does not fall under the term "volume flow sensor."

[0021] In a further embodiment, the pressure detection system comprises a pressure sensor at the pump inlet and a pressure sensor at the pump outlet. The pump differential pressure can be determined using these absolute pressure sensors. A pressure sensor arranged at the pump inlet is suitable, for example, for determining the ventricular pressure of a cardiac assist system. Alternatively or in addition to a sensor at the pump inlet and / or outlet, a pump system for determining a differential pressure can be present. Such a system can, for example, be formed by a system for detecting and compensating a pump rotor offset. This requires, for example, a characteristic map that links the corresponding current flowing through at least one control coil of the system for displacing the pump rotor with the corresponding differential pressure or volume flow.

[0022] Further examples can be found in the examples listed below.

[0023] They show: Fig. 1 shows a schematic structure of a heart assist system; Fig. 2a shows an exemplary structure of a pump with a cannula connected to it; Fig. 2b shows a schematic structure of a control unit of the heart assist system; Fig. 3 shows a schematic flow diagram for a method for determining an outlet pressure present at an outlet of a heart assist system; Fig. 4 shows a schematic representation of the estimation of the pump pressure difference of a heart assist system with a pump and a cannula; and Fig. 5 shows a schematic overview of a method for determining model parameters for a model for determining the cannula pressure difference.

[0024] Figure 1shows the schematic structure of an implanted cardiac assist system 1. The exemplary cardiac assist system 1 comprises a pump 2, which is connected to a cannula 3 at its outlet. The pump unit or pump 2 can be either an axial pump or a radial pump. The pumps can be adjusted, for example, using the control unit 4, to deliver a predetermined amount of blood. However, other delivery programs are also possible, such as a delivery mode that more accurately replicates the beat of a natural heart.

[0025] The cannula 3 can be formed, for example, from silicone or a softer graft material, such as polyester, ePTFE, or mixtures of these materials. The control unit 4 can be either fully attached to the pump 2, partially attached to the pump 2, or completely remote from the pump unit 2. The control unit 4 can include electronic or electrical control elements, such as a processor, memory, signal processing circuits, motor control circuits, and monitoring systems.

[0026] The pump 2 shown is an axial pump whose inlet is directly coupled to a heart 5. In the present example, the cardiac assist system is an LVAD system. Physiologically, the left ventricle 6 is connected to the aorta 8 via the aortic valve 7. The cardiac assist system 1 represents a bypass from the left ventricle 6 to the aorta 8. In an RVAD system, the pump would be connected to the right ventricle and the pulmonary artery; in a BVAD system, a dual-chamber pump would be provided, forming an LVAD system with the first chamber and an RVAD system with the second chamber. Alternatively, two pumps can be present.

[0027] In the present example of an LVAD system, blood is pumped from the left ventricle 6 through the pump unit 2 and the cannula 3, with the outlet of the cannula 3 being connected to the aorta 8. In Figure 1A sewing ring 9 is also shown, with which the pump 2 is attached to the apex of the heart.

[0028] In the left ventricle 6, there is a ventricular pressure pv , and in the aorta, there is an aortic pressure p ao . Physiologically, the aortic valve 7 opens when the pressure pv in the ventricle is greater than the aortic pressure p ao . The aortic pressure p ao can be described as the sum of the ventricular pressure pv , the pressure drop Δp across the pump 2, and the pressure drop Δp c across the cannula 3, ie p ao = p v + Δ p + Δp c .

[0029] One possibility for determining the aortic pressure would be, for example, to place an additional pressure sensor in the aorta 8. However, this implantation is associated with a certain health risk. Alternatively, a pressure sensor can be arranged at or in the outlet of the cannula 3, which is connected to the aorta 8, to record the pressure at the cannula outlet. Although this variant allows direct measurement of the aortic pressure, the sensor arranged in the cannula requires a data connection to the control unit 4. Since the cannula 3 often has to be shortened during implantation and is only subsequently connected to the aorta 8, it cannot be guaranteed that a pressure sensor arranged at the outlet of the cannula 3 will still be present after the cannula has been cut to size. One possible solution is to relocate the pressure sensor into the cannula 3.In this case, however, at least the pressure drop between the location of the sensor and the outlet of the cannula 3 connected to the aorta 8 must be estimated.

[0030] The method presented in this application is based on a possibility to determine the aortic pressure p ao, for example, by estimating the pressure drop Δp c. If the other variables can be measured, ie pv and Δp, the aortic pressure can be determined from the measured variables of the ventricular pressure and the pressure drop across the pump. Although in Figure 1 While the ventricular assist system 1 comprises only an outlet cannula, the ventricular assist system may also comprise an inlet cannula. The methods presented here can also be applied to such a ventricular assist system.

[0031] In Figure 2aan exemplary cardiac assist system 20 is shown in more detail. The cardiac assist system comprises a pump 21 with a pump inlet 22 and a pump outlet 24. The pump unit pumps blood from the pump inlet 22 past a pre-guide wheel 26, which is connected to a housing wall of the pump 21 via vanes 27, to a conveying element 28, which is designed, for example, as a rotor. In the present example, the conveying element 28 has a helix 30, which conveys the blood towards the post-guide element 32. The conveying element 28 is rotated by means of a motor 33. For this purpose, permanent magnets are located in the conveying element 28, which can be rotated by the motor current of the motor 33. The post-guide wheel or the post-guide element 32 is arranged in a spiral chamber, into which the axially conveyed blood is conveyed and then pushed out through the pump outlet 24.

[0032] Examples of such pumps can be found, among others, in the publications WO 2013 / 021014, WO 2012 / 150045, WO 2012 / 149946, WO 2011 / 054545, and WO 02 / 066837. Further exemplary pumps are shown, for example, in WO 2012 / 051454 or WO 2012 / 024493.

[0033] In the present embodiment, the pump 21 comprises a pressure detection system. This comprises, for example, a sensor coil 34, which can detect a distance between the conveying element 28 and the guide body 32. If, due to the conveyed blood flow, a force directed along the directions 35 acts on the conveying element 28, it is deflected from a rest position, thereby inducing a current in the sensor coil 34. This induced current is detected and processed in the control unit or another circuit such that, by applying a current to a control coil 36, the conveying element 28 is returned to a position between the pre-guide wheel 26 and the guide body 32 corresponding to the rest position. Based on the current applied to the control coil 36 (measured directly or indirectly), a pressure difference between the pump inlet 22 and the pump outlet 24 can be determined, for example.

[0034] Furthermore, the pressure detection system may comprise a pressure sensor 38 arranged at the inlet 22 of the pump 31. This pressure sensor detects the blood pressure at the inlet of the pump, so that, as in Figure 1 shown, for example, a ventricular pressure pv can be determined. The pressure detection system can further comprise a further pressure sensor 40, which detects the pressure at the outlet 24 of the pump. If the pump comprises two sensors at the inlet and outlet, the pump pressure difference can be determined from the two measured values. However, at least one of the two pressure sensors can also be replaced or supplemented by a differential pressure sensor system, such as the system comprising elements 34 and 36 for indirectly detecting the differential pressure of the pump. The various data detected using the pressure detection system are detected, processed, and used to estimate the aortic pressure by a control unit.

[0035] The heart support system 20 further comprises a cannula 50, which is made of silicone, for example. A cannula inlet 52 is connected to the pump outlet 24, for example, via a snap-fit ​​connector. At the cannula outlet 54, the cannula 50 can be connected, for example, by suturing, to an aorta. Before connecting the aorta to the cannula, it may be necessary to shorten the cannula, for example, to a length 56. Therefore, in some embodiments, no active sensor technology is provided in the cannula. However, the system or method presented here can also be used to complement active sensor technology in the cannula.

[0036] If no sensor components are present in the cannula, the pressure drop between the cannula inlet 52 and the cannula outlet 54 or the shortened cannula outlet 56 is estimated using a method explained in this application. Estimation here means that when determining the pressure drop Δp c along the cannula 50, only variables determined by the pump 21 or measured externally of the pump or cannula can be used. Various models for estimating the cannula pressure drop or cannula pressure difference are possible for this purpose.

[0037] In Figure 2bAn exemplary control unit 60 is schematically illustrated. The control unit 60 can, for example, be arranged in an external housing outside the body. Data acquired by the pump unit is transmitted to the control unit 60, for example, via a data bus. However, the control unit can also be attached to the pump unit, for example, integrated into a pump housing. The control unit illustrated here comprises a CPU 62, which is configured to operate the ventricular assist system. This means that it is particularly capable of adjusting the speed of the ventricular assist system.

[0038] The CPU 62 is connected to a first memory unit 64, in which various commands and control programs are stored. In the present example, the control commands also include instructions for determining an estimated aortic pressure and instructions for determining model parameters of a model for aortic pressure estimation or cannula pressure difference estimation. The CPU 62 can also be connected to another memory unit 66, which is configured, for example, to record data and allow the user to later read the data.

[0039] A signal processing stage 68 is connected upstream of the CPU, which filters, processes, and conditions incoming signals arriving at the data input 70 so that the data can be processed by the CPU 62. In addition to filters, the signal processing stage 68 can include, for example, analog / digital converters or the like.

[0040] Signals 72 are received at data input 70. For example, these signals include measured motor currents 74 for determining the speed, signals from the absolute pressure sensor 38 in the form of signals 76, and further signals 78, which represent signals from the absolute pressure sensor 40. Signals 80 represent the current present in the control coil 36, which can be converted into a pump differential pressure by the CPU 62. Optionally, the Figure 2a The pump system shown may also include a volume flow sensor, whose signals 82 are also input to the data input 70. The CPU 62 may also return signals to the pump, such as signals 86, which are shown as control currents 88 for the motor or currents for the control coil 36.

[0041] Based on the Figure 3A method for determining the aortic pressure (also called afterload of the ventricular assist system) will be explained. The method 100 for determining the aortic or outlet pressure of the ventricular assist system includes step 110, in which a pressure in the ventricle or at the ventricular assist system inlet is first measured. This measurement can be performed, for example, using an absolute pressure sensor at the ventricular assist system inlet. A differential pressure between the ventricular assist system inlet and outlet is then estimated using a calibrated model. This occurs in step 120. The aortic pressure or outlet pressure can then be determined in step 130 from the estimated pressure difference and the measured inlet pressure.

[0042] According to the invention, step 120 is carried out in a heart support system with a pump and a cannula connected thereto as in Figure 4illustrated in such a way that a pump pressure difference is first measured in step 122 and then a cannula pressure difference is estimated based on the quantities measured using the pump and a model of the cannula pressure difference (step 124).

[0043] Based on the Figure 5 Different models of the cannula pressure difference are to be explained, possibilities for calibrating or determining the possible model parameters are to be given and it is to be explained how the model calibrated with the model parameters is used in the cardiac support system.

[0044] The exemplary method 300 first includes determining a model of the cannula pressure difference 310. Various models can be considered here. Some models are listed here as examples.

[0045] In a first model, the cannula pressure difference Δp c can be modeled at selected times by the pump pressure difference Δp. For example, it is known that at the time of aortic valve opening, the ventricular pressure pv is briefly essentially equal to the aortic pressure p ao. For these times, the model can therefore be set up that the cannula pressure difference is equal to the pump pressure difference. However, this model is only applicable at a few times during a single cardiac cycle, since the deviations between the aortic and ventricular pressure at times other than the time of aortic valve opening differ significantly from one another. This means that the times at which the model is valid must be determined continuously and in this sense form the model parameters. For this purpose, a method is necessary that can reliably determine the opening and, in particular, the time of opening of the aortic valve.Subsequently, a time interval from the opening must be determined during which the model is applied. If the pump or its control unit is capable of determining the aortic valve opening, a control unit of a ventricular assist system can determine the aortic pressure at least at the times of the valve opening. The disadvantage of the model, however, is that continuous measurement of the aortic pressure does not provide satisfactory results.

[0046] Another model for determining the cannula pressure difference is a relationship between a static volume flow, which can be determined, for example, using variables determined by the pump. For example, the pressure difference Δp c can be approximated by a nonlinear function. One example is a third-order polynomial of the volume flow. Alternatively, other nonlinear functions, such as an exponential function, can also be used. The polynomial can be represented, for example, as follows: Δ p c = k 3 Q ˙ p 3 + k 2 Q ˙ p 2 + k 1 Q ˙ p 1 , where Q represents the volume flow through the pump and the model parameters k 1 , k 2 and k 3 must be determined.

[0047] In order to determine the model parameters k 1 , k 2 , k 3 , one embodiment provides that the pressure drop Δp c across the cannula at specific measuring points is known, the volume flow is approximately constant and at least three different measuring points are known. However, if the model parameters have been determined or calibrated, the model can be used to continuously estimate the cannula pressure difference, i.e. it also delivers satisfactory results in cardiac cycle segments with a closed aortic valve. The advantage here is that the model only needs to be calibrated once, whereby calibration can be carried out at specific, for example regular time intervals such as days, weeks or months.

[0048] As already described in the first model, at least at times shortly after the opening of the aortic valve, the pressure drop in the cannula Δp c is essentially equal to the pressure drop in the pump. Therefore, the cannula pressure difference can be determined at least for these times. At these times, the pressure difference between the ventricle and the aorta is approximately constant. Therefore, the volume flows through the aortic valve and through the pump are also approximately constant. Since the volume flow during the ejection phase of the ventricle is approximately constant but not absolutely constant, limit values ​​can be set to determine whether the volume flow drop during the ejection phase of the heart is satisfactorily constant.Here, for example, the standard deviation of the volume flow during the expulsion phase can be determined and a limit value can be set below which the expulsion phase interval can be used to determine the model parameters.

[0049] In order to further determine the parameters for a wide pump speed range, measurements are performed at several, for example, three, different speeds n, and the associated data or time series are recorded. The model parameters k 1 , k 2 , k 3 are then determined for the time series of the pump pressure difference and the volume flow through the pump recorded during the respective speed, in conjunction with the criteria that the volume flow is satisfactorily constant and the cannula pressure difference can be essentially equated with the pump pressure difference. The following relationship can apply, for example: k → = Δ p t → ⋅ Q ˙ ^ + t

[0050] Here, Q ˙ ^ + the pseudoinverse of the matrix Q ˙ ^ , which recorded values ​​at different measurement times Q ˙ p 3 , Q ˙ p 2 and Q ˙ p 1 as entries. While the k-vector in this embodiment is three-dimensional, the pump pressure difference vector is m-dimensional, provided that m measurement points are present. The matrix Q ˙ ^ is an [m×3] matrix. Using statistical methods such as regression, the parameters k 1 , k 2 , k 3 can be determined and validated. Once the model parameters k 1 , k 2 , k 3 have been determined, the cannula pressure difference can be estimated by the model using the volume flow through the pump at any time during the cardiac cycle.

[0051] Thus, in step 330, the times of aortic valve opening at different speeds are first determined. The corresponding measurement data of the volume flow and the pump pressure difference are acquired in steps 340 and 350. Subsequently, a calibration of the model parameters is performed using the acquired data under the described boundary conditions, e.g., a determination of the model parameters with the desired accuracy, so that the model parameters are calibrated in step 320. The model can then be applied by the control unit (step 360), so that by measuring the volume flow, the pump pressure difference, and the pump inlet pressure, a statement can be made about the pump or cannula outlet pressure, which almost corresponds to the aortic pressure.

[0052] Another model is the use of a differential equation system to determine the cannula pressure difference. While the model described in the previous section is essentially modeled according to a static cannula pressure difference, a differential equation can also capture a dynamic cannula pressure difference term. One way to model the cannula pressure difference Δp c is, for example, Δ p c = L c d Q ˙ p dt + R c Q ˙ p ⋅ Q ˙ p , where R c ( Q̇ p ) · Q̇ p can correspond to the static cannula pressure difference term of the second variant (R c corresponds to a resistance term in an equivalent circuit) and L c models a mass inertia of the fluid volume of the cannula and the pump. The mass inertia of the fluid volume in the pump can be determined based on the geometry of the pump. The mass inertia L c * of the fluid volume of the cannula can be determined, for example, for a cannula with a round inner cross-section and length l, by the term L c = 4 ρl πD i 2 where ρ is the blood density, l is the cannula length, and D i is the cannula diameter. Other cannula geometries can be captured, for example, by adjusting the inertia term (e.g., by integrating all sections of the cannula) and are obvious to the person skilled in the art based on the above explanations. The inertia can also be determined experimentally as a numerical value for known cannulas.

[0053] When using this model, a similar procedure can be followed as in the previous case, first determining the static parameters k 1 to k 3 . Then, for example, the length of the cannula and its known inner diameter are determined and transmitted to the control unit. Based on the collected information, the cannula pressure difference can then be estimated.

[0054] As an extension of these models, the pressure drop across the aortic valve could also be integrated. This could be estimated, for example, based on an assumed geometry of the aortic valve or one measured using imaging techniques.

[0055] Furthermore, the influence of the elasticity of the cannulas has been ignored in the models presented so far, as the cannulas are highly elastic and therefore no wave reflections are to be expected in the frequency range of interest. However, it is conceivable to include a term for the elasticity of the cannula in the region of the aortic anastomosis in the models; this could potentially improve the accuracy of the estimation.

[0056] Although the models presented here for determining the cannula pressure difference only include models that include the volume flow as an independent variable, similar models can also be set up for a pump pressure difference or a speed or a combination of these variables.

[0057] Once a model for the cannula pressure difference has been established, the outlet or aortic pressure can be determined from the pump inlet or ventricular pressure, the pump pressure drop and the model.

Claims

1. A method for determining an outlet pressure (pao), in particular an aortic pressure, prevailing at an outlet of a heart assist system (1; 20) comprising a pump and a cannula connected thereto, wherein the method comprises the following steps: - determining an inlet pressure (pv) which prevails at the inlet of the heart assist system; - determining a pump pressure difference (Δp) between the pump inlet and the pump outlet of the pump of the heart assist system; - estimating a pressure difference between the inlet and the outlet of the heart assist system based on the pump pressure difference (Δp) and an estimated cannula pressure difference (Δpc); - determining the outlet pressure from the inlet pressure and the pressure difference.

2. The method according to claim 1, wherein the inlet pressure is determined using a pressure sensor.

3. The method according to claim 1 or 2, wherein the determining of the pump pressure difference takes place using a current prevailing at a control coil, wherein the control coil determines a position of the pump rotor.

4. The method according to claim 1 or 2, wherein the pump pressure difference is detected by two pressure sensors arranged at the inlet and outlet of the pump or by a differential pressure sensor arranged in the pump.

5. The method according to claim 1, wherein the pressure difference is estimated by a non-linear function, or wherein the pressure difference is approximated by a non-linear function of a volume flow, wherein parameters of the non-linear function are determined from measurement points at which the pressure difference is known.

6. The method according to any one of the preceding claims, wherein a dynamic pressure difference term is determined by a differential equation.

7. A method according to one of the preceding claims, wherein a model of the pressure difference is determined and at least one parameter of the model is determined by way of statistical methods.

8. A method according to one of the preceding claims, wherein the pressure difference is estimated by way of a higher order polynomial and wherein the polynomial is preferably a function of a variable which is measurable by the heart assist system.

9. A heart assist system (1; 20) with a pump (2; 21), for the assistance of the heart and with a cannula (3; 50) which is connectable to the pump, wherein the pump comprises a pump inlet (22) and a pump outlet (24), and the cannula comprises a cannula inlet (52) and a cannula outlet (54), and the heart assist system comprises a pressure detection system which is configurable in a manner such that a pressure prevailing at the pump inlet is detectable and that a pump pressure difference between the pump inlet and the pump outlet is detectable, and wherein a control unit (60) is present, which is configured in a manner such that a cannula pressure difference (Δpc) between the cannula inlet and the cannula outlet can be estimatedusing the pump pressure difference.

10. A heart assist system according to claim 9, wherein the control unit comprises a memory unit (64, 66), and the memory unit stores commands and / or parameters for estimating the cannula pressure difference.

11. A heart assist system according to one of the claims 9 to 10, wherein the control unit is connected to the pressure detection system so that data detected by the pressure detection system can be transferred to the control unit.

12. A heart assist system according to one of the claims 9 to 11, wherein the control unit is designed in a manner such that at least one variable detected by the pump is used for estimating the cannula pressure difference.

13. A heart assist system according to claim 12, wherein the control unit is designed in a manner such that the cannula pressure difference is determined in dependence on a volume flow flowing through the pump, wherein the control unit is designed such that the volume flow is preferably determined on the basis of the pump pressure difference.

14. A heart assist system according to one of claims 9 to 13, wherein the pump pressure difference is determined by way of a displacement of a pump rotor.

15. A heart assist system according to one of the claims 9 to 14, wherein the pressure detection system comprises a pressure sensor (38) at the pump inlet and / or a pressure sensor (40) at the pump outlet and / or a differential pressure sensor in the pump.