Control system of artificial heart, artificial heart system, and method of controlling rotary blood pump
Patent Information
- Authority / Receiving Office
- EP · EP
- Patent Type
- Applications
- Current Assignee / Owner
- NAT CEREBRAL & CARDIOVASCULAR CENT
- Filing Date
- 2024-07-24
- Publication Date
- 2026-06-24
AI Technical Summary
The challenge of simultaneously and stably controlling the flow rates of two rotary blood pumps used in a total artificial heart or bi-ventricular assist device, mimicking the native heart's function, is extremely difficult due to variations in vascular resistance and atrial pressures.
A control system that includes a measurement device to monitor flow rates and atrial pressures, and a control device that calculates coefficients representing the slopes of logarithmic functions relating cardiac output to atrial pressures. This system adjusts the revolution speeds of the pumps under feedback control to maintain these coefficients close to target values, effectively stabilizing the balance between the flow rates.
The system achieves stable control of the rotary blood pumps, maintaining balance between flow rates and preventing suction and congestion, thereby ensuring effective blood circulation and physiological stability.
Smart Images

Figure JP2024026440_27022025_PF_FP_ABST
Abstract
Description
CONTROL SYSTEM OF ARTIFICIAL HEART, ARTIFICIAL HEART SYSTEM, AND METHOD OF CONTROLLING ROTARY BLOOD PUMP
[0001] The present disclosure relates to a control system of an artificial heart, an artificial heart system, and a method of controlling a blood pump.
[0002] A ventricular assist device (VAD) has been developed as a therapy for a terminally ill patient with a severe heart failure. In recent years, a rotary blood pump (RBP) including a centrifugal pump and an axial pump has been commonly used as a VAD, because RBPs are superior to a pulsatile blood pump in durability, compactness, and antithrombotic property. Development of the rotary blood pump enables long-term implantation as a left ventricular assist device (LVAD), and also serves as a bridge to transplantation or a bridge to a destination therapy.
[0003] In particular for a patient with a severe bi-ventricular failure or a patient with an exacerbated right heart failure caused by long-term use of LVAD, there is a case where a bi-ventricular assist device (BiVAD) therapy is applied or a case where a total artificial heart (TAH) is applied. Research and development for use of two RBPs, i.e. left and right RBPs as a TAH has also been promoted worldwide.
[0004] How the two RBPs should be controlled simultaneously and stably in the BiVAD or the TAH, as if the RBPs work like a native heart, however, is an extremely difficult issue.
[0005] The present disclosure was made to solve such a problem, and an object thereof is to automatically control in a stable manner, flow rates of two RBPs used in a TAH or a BiVAD.
[0006] A control system according to the present disclosure is a control system of an artificial heart including a first rotary pump that replaces or assists a function of a left heart of a subject and a second rotary pump that replaces or assists a function of a right heart of the subject, and includes a measurement device that measures a flow rate of the first pump, a flow rate of the second pump, a left atrial pressure, and a right atrial pressure and a control device that controls the first pump and the second pump. A cardiac output of the left ventricle of a subject is represented by a first curve of a logarithmic function of the left atrial pressure and a cardiac output of a right ventricle of the subject is represented by a second curve of a logarithmic function of the right atrial pressure. The control device calculates a first coefficient corresponding to a slope of the first curve based on a measurement value of the flow rate of the first pump and a measurement value of the left atrial pressure and calculates a second coefficient corresponding to a slope of the second curve based on a measurement value of the flow rate of the second pump and a measurement value of the right atrial pressure, and controls a revolution speed of the first pump under feedback control such that the first coefficient is closer to a first target coefficient and controls a revolution speed of the second pump under feedback control such that the second coefficient is closer to a second target coefficient.
[0007] An artificial heart system according to the present disclosure includes a first rotary pump that replaces or assists a function of the left heart of a subject, a second rotary pump that replaces or assists a function of the right heart of the subject, a measurement device that measures a flow rate of the first pump, a flow rate of the second pump, a left atrial pressure, and a right atrial pressure, and a control device that controls the first pump and the second pump. A cardiac output of the left ventricle of the subject is represented by a first curve of a logarithmic function of the left atrial pressure and a cardiac output of the right ventricle of the subject is represented by a second curve of a logarithmic function of the right atrial pressure. The control device calculates a first coefficient corresponding to a slope of the first curve based on a measurement value of the flow rate of the first pump and a measurement value of the left atrial pressure and calculates a second coefficient corresponding to a slope of the second curve based on a measurement value of the flow rate of the second pump and a measurement value of the right atrial pressure, and controls a revolution speed of the first pump under feedback control such that the first coefficient is closer to a first target coefficient and controls a revolution speed of the second pump under feedback control such that the second coefficient is closer to a second target coefficient. A control method according to the present disclosure is a method of controlling a blood pump including a first rotary pump that replaces or assists a function of the left heart of a subject and a second rotary pump that replaces or assists a function of the right heart of the subject, and includes measuring a flow rate of the first pump, a flow rate of the second pump, a left atrial pressure, and a right atrial pressure and controlling the first pump and the second pump. A cardiac output of the left ventricle of the subject is represented by a first curve of a logarithmic function of the left atrial pressure and a cardiac output of the right ventricle of the subject is represented by a second curve of a logarithmic function of the right atrial pressure. The controlling the first pump and the second pump includes calculating a first coefficient corresponding to a slope of the first curve based on a measurement value of the flow rate of the first pump and a measurement value of the left atrial pressure and calculating a second coefficient corresponding to a slope of the second curve based on a measurement value of the flow rate of the second pump and a measurement value of the right atrial pressure and controlling a revolution speed of the first pump under feedback control such that the first coefficient is closer to a first target coefficient and controlling a revolution speed of the second pump under feedback control such that the second coefficient is closer to a second target coefficient.
[0008] According to the present disclosure, flow rates of two rotary blood pumps used in a total artificial heart or a bi-ventricular assist device can automatically be controlled in a stable manner.
[0009] Fig. 1 is a diagram schematically showing an overall configuration of an artificial heart system, according to example embodiments.Fig. 2 is a diagram schematically showing the circulation of blood in a subject implanted with the artificial heart system, according to example embodiments.Fig. 3 is a diagram showing relation between an atrial pressure and a cardiac output from a ventricle in the heart of a subject, according to example embodiments.Fig. 4 is a functional block diagram (No. 1) showing a control structure of a controller, according to example embodiments.Fig. 5 is a diagram (No. 1) showing an exemplary result of an experiment of control by the controller, according to example embodiments.Fig. 6 is a diagram showing relation between an atrial pressure and a pump flow rate in the result of the experiment shown in Fig. 5, according to example embodiments.Fig. 7 is a diagram conceptually showing a difference between control in a comparative example and control in the present disclosure, according to example embodiments.Fig. 8 is a diagram (No. 2) showing an exemplary result of an experiment of control by the controller, according to example embodiments.Fig. 9 is a diagram (No. 3) showing an exemplary result of an experiment of control by the controller, according to example embodiments.Fig. 10 is a functional block diagram (No. 2) showing a control structure of a controller, according to example embodiments.Fig. 11 is a diagram showing on a three-dimensional coordinate, an FSLcurve representing a left ventricle output, an FSRcurve representing a right ventricle output, and an integrated FS cardiac output curve FSi, according to example embodiments.Fig. 12 is a diagram showing a venous return surface VRS and integrated FS cardiac output curve FSi on a three-dimensional coordinate, according to example embodiments.Fig. 13 is a diagram (No. 4) showing an exemplary result of an experiment of control by the controller, according to example embodiments.Fig. 14 is a diagram (No. 5) showing an exemplary result of an experiment of control by the controller, according to example embodiments.Fig. 15 is a diagram showing relation between an atrial pressure and a pump flow rate in the result of the experiment shown in Fig. 14, according to example embodiments.Fig. 16 is a diagram (No. 6) showing an exemplary result of an experiment of control by the controller, according to example embodiments.
[0010] An embodiment of the present disclosure will be described in detail with reference to the drawings. The same or corresponding elements in the drawings have the same reference characters allotted and description thereof will not be repeated.
[0011] Fig. 1 is a diagram schematically showing an overall configuration of an artificial heart system 1 including a control system according to the present disclosure. Artificial heart system 1 includes a left pump (first pump) 10 and a right pump (second pump) 20 and a controller (control device) 100.
[0012] Left pump 10 is a rotary blood pump (RBP) that replaces a function of the left heart. Left pump 10 has an inlet connected to the left atrium or the left ventricle of a subject and an outlet connected to an aorta which is the exit of the left ventricle of the subject.
[0013] Right pump 20 is a rotary blood pump (RBP) that replaces a function of the right heart. Right pump 20 has an inlet connected to the right atrium or the right ventricle of the subject and an outlet connected to a pulmonary artery which is the exit of the right ventricle of the subject.
[0014] In general, the rotary blood pump is superior in durability, compactness, and antithrombotic property than a pulsatile blood pump. Therefore, by adoption of the rotary blood pump as left pump 10 and right pump 20 as in the present disclosure, left pump 10 and right pump 20 can be implanted into a living body of a patient for a long period. A structure of left pump 10 and right pump 20 is not particularly limited, so long as they are each a rotary pump such as a centrifugal pump or an axial pump.
[0015] A measurement device 11 including a flow rate sensor that measures a flow rate COSof left pump 10 and a pressure sensor that measures a left atrial pressure PLis attached to left pump 10. Measurement device 11 outputs a signal indicating a result (COS, PL) of measurement to controller 100.
[0016] A measurement device 21 including a flow rate sensor that measures a flow rate COPof right pump 20 and a pressure sensor that measures a right atrial pressure PRis attached to right pump 20. Measurement device 21 outputs a signal indicating a result (COP, PR) of measurement to controller 100.
[0017] Each sensor in measurement devices 11 and 21 may independently be arranged separately from pumps 10 and 20, instead of attachment to pumps 10 and 20.
[0018] Controller 100 includes a central processing unit (CPU) 110 and a memory 120 such as a read only memory (ROM) and a random access memory (RAM). Controller 100 generates command revolution speeds NLcomand NRcombased on measurement values (COS, PL, COP, and PR) from measurement devices 11 and 21 and outputs generated command revolution speeds NLcomand NRcomto left pump 10 and right pump 20, respectively. A revolution speed NLof left pump 10 is controlled to command revolution speed NLcomand a revolution speed NRof right pump 20 is controlled to command revolution speed NRcom. The control system according to the present disclosure includes measurement devices 11 and 21 and controller 100.
[0019] Though Fig. 1 shows an example in which left pump 10 and right pump 20 are used in the setting of the total artificial heart (TAH), left pump 10 and right pump 20 can be used in the setting of a bi-ventricular assist device (BiVAD). When the bi-ventricular assist device (BiVAD) is employed, the inlet of left pump 10 should only be connected to the left ventricle and the inlet of right pump 20 should only be connected to the right ventricle.
[0020] Fig. 2 is a diagram schematically showing the circulation of blood achieved by artificial heart system 1. The circulation of blood is classified into a systemic circulation, a pulmonary circulation, left pump 10 that performs the function of the left heart, and right pump 20 that performs the function of the right heart. Left pump 10 receives blood that flows back from the pulmonary circulation and ejects the blood to the systemic circulation. Right pump 20 receives the blood that flows back from the systemic circulation and ejects the blood to the pulmonary circulation.
[0021] <Frank Starling Law of Heart>
[0022] The heart of the subject has such a characteristic that a cardiac output increases as the atrial pressure increases, that is, the Frank Starling law of the heart (which is also referred to as the "FS law" below).
[0023] Fig. 3 is a diagram showing relation between an atrial pressure (unit; mmHg) and a cardiac output (unit; ml / min / kg) from a ventricle in the heart of the subject. The graph on the right in Fig. 3 shows relation between a left atrial pressure PLand a left cardiac output and the graph on the left in Fig. 3 shows relation between a right atrial pressure PRand a right cardiac output.
[0024] The left cardiac output has such a characteristic as increasing with increase in left atrial pressure PL. More specifically, the left cardiac output is represented by a curve in accordance with the FS law, that is, a curve (which is also referred to as an "FSLcurve" below) of a logarithmic function of left atrial pressure PL. The FSLcurve can be formulated in an expression (L) below where k1 and k2 are constants found in in-vivo experiments and SLrepresents a coefficient (first coefficient) corresponding to a slope of the FSLcurve.
[0025] Left cardiac output = SLx{log(PL-k1)+k2} ... (L)
[0026] The right cardiac output has such a characteristic as increasing with increase in right atrial pressure PR. More specifically, the right cardiac output is represented by a curve in accordance with the FS law, that is, a curve (which is also referred to as an "FSRcurve" below) of a logarithmic function of right atrial pressure PR. The FSRcurve can be formulated in an expression (R) below where k3 and k4 are constants found in in-vivo experiments and SRrepresents a coefficient (second coefficient) corresponding to a slope of the FSRcurve.
[0027] Right cardiac output = SRx{log(PR-k3)+k4} ... (R)
[0028] As the heart of the subject has the FS law as shown in Fig. 3, the left cardiac output and the right cardiac output are stabilized at certain values substantially equal to each other and left atrial pressure PLand right atrial pressure PRare also stabilized at respective certain values and enters an equilibrium state.
[0029] The slope (coefficient SL) of the FSLcurve and the slope (coefficient SR) of the FSRcurve vary in accordance with ventricular contractility, heart rate, and the like. For example, when a blood flow required in the whole body increases as in intensified physical activities (for example, exercises), sympathetic activities are intensified and coefficients SLand SRincrease as compared with those in a normal condition. The blood flow required in the whole body is thus secured. When a heart failure occurs, on the other hand, coefficients SLand SRdecrease as compared with those in the normal condition.
[0030] <Control of RBPs (Left Pump 10 and Right Pump 20) of Left and Right Hearts>
[0031] How the two simultaneously driven RBPs of the left and right hearts should be controlled in the bi-ventricular assist device or the total artificial heart similarly to a native heart is an extremely difficult issue.
[0032] In the heart of the subject, under the FS law as described above, the left and right cardiac outputs are stabilized at certain values substantially equal to each other and the left and right atrial pressures are also stabilized in a certain equilibrium state. The flow rate of the RBP which is used in the artificial heart, however, is greatly dependent on a pressure at the outlet of the RBP (an aortic pressure or a pulmonary arterial pressure) and less dependent on a pressure at the inlet (the left atrial pressure or the right atrial pressure). Therefore, even when balance between flow rates of the RBPs of the left and right hearts can be optimized once in simultaneous drive of the RBPs in the left and right hearts, variation in vascular resistance due to occlusion or stenosis of the aorta or the pulmonary artery may easily lead to loss of the balance between the flow rates of the RBPs of the left and right ventricles. When the balance between the flow rates of the RBPs of the left and right ventricles is lost, blood distribution between the systemic circulation and the pulmonary circulation becomes abnormal. Then, there is a concern about suction which means suction of the inlet of the RBP to the atrial wall caused by lowering in atrial pressure or congestion or edema caused by abnormal increase in atrial pressure.
[0033] Therefore, for stable drive of the RBPs of the left and right hearts without such a problem as suction and congestion described above, the RBPs of the left and right hearts are desirably controlled as if they had the FS law of the native heart.
[0034] In order to solve the problem above, the inventors of the subject application developed a technique to control the RBPs (left pump 10 and right pump 20) of the left and right hearts as if they had the FS law of the native heart. Specifically, controller 100 according to the present disclosure controls left pump 10 and right pump 20 with a technique below as if they had the FS law of the native heart.
[0035] Controller 100 controls left pump 10 such that the flow rate of left pump 10 (which is also referred to as a "left pump flow rate" below) COSvaries along the FSLcurve described above and relation in an expression (L') below is always satisfied. Similarly, controller 100 controls right pump 20 such that the flow rate of right pump 20 (which is also referred to as a "right pump flow rate" below) COPvaries along the FSRcurve described above and relation in an expression (R') below is always satisfied.
[0036] COS= SLx{log(PL-k1)+k2} ... (L’)
[0037] COP= SRx{log(PR-k3)+k4} ... (R’)
[0038] The expression (L') results from substitution of the "left cardiac output" in the above-mentioned expression (L) that expresses the FSLcurve of the native heart with "left pump flow rate COS." The expression (R') results from substitution of the "right cardiac output" in the above-mentioned expression (R) that expresses the FSRcurve of the native heart with "right pump flow rate COP."
[0039] Transformation of the expression (L') provides an expression (1) below. Therefore, control of left pump 10 such that relation in the expression (L') is always satisfied is equivalent to control of left pump 10 such that coefficient SLcalculated in the expression (1) below is always constant. Similarly, transformation of the expression (R') provides an expression (2) below. Therefore, control of right pump 20 such that relation in the expression (R') is always satisfied is equivalent to control of right pump 20 such that coefficient SRcalculated in the expression (2) below is always constant.
[0040] SL= COS / {log(PL-k1)+k2} ... (1)
[0041] SR= COP / {log(PR-k3)+k4} ... (2)
[0042] Controller 100 according to the present disclosure calculates coefficient SLcorresponding to a slope of the FSLcurve by substituting the measurement value of left pump flow rate COSand the measurement value of left atrial pressure PLinto the expression (1) and calculates coefficient SRcorresponding to the slope of the FSRcurve by substituting the measurement value of right pump flow rate COPand the measurement value of right atrial pressure PRinto the expression (2).
[0043] Controller 100 then controls revolution speed NLof left pump 10 under feedback control such that coefficient SLis closer to a target coefficient SLt(first target coefficient) and controls revolution speed NRof right pump 20 under feedback control such that coefficient SRis closer to a target coefficient SRt(second target coefficient). Left pump 10 and right pump 20 can thus be controlled as if they had the FS law of the native heart.
[0044] Fig. 4 is a functional block diagram showing a control structure of controller 100. Controller 100 includes a target value generator 101, subtractors 102L and 102R, control units 103L and 103R, and a coefficient calculator 104. Control by controller 100 is in a form of negative feedback control as a whole.
[0045] Coefficient calculator 104 obtains the measurement value of each of left pump flow rate COS, left atrial pressure PL, right pump flow rate COP, and right atrial pressure PRfrom measurement devices 11 and 21 and calculates coefficients SLand SRrepresenting the slope of the FS curve by substituting the obtained measurement values into the expressions (1) and (2) described above.
[0046] Target value generator 101 obtains a target value COtof a pump flow rate, a target value PLtof the left atrial pressure, and a target value PRtof the right atrial pressure from memory 120 and calculates target coefficients SLtand SRtby substituting the obtained target values into expressions (3) and (4) below.
[0047] SLt= COt / {log(PLt-k1)+k2} ... (3)
[0048] SRt= COt / {log(PRt-k3)+k4} ... (4)
[0049] Target values COt, PLt, and PRtare freely determined in advance, for example, by medical experts (doctors, nurses, and the like) and stored in memory 120. Therefore, in the present disclosure, target values COt, PLt, and PRtare each a fixed value.
[0050] Subtractor 102L calculates a difference (= SLt-SL) between target coefficient SLtgenerated by target value generator 101 and coefficient SLcalculated by coefficient calculator 104. Subtractor 102R calculates a difference (= SRt-SR) between target coefficient SRtgenerated by target value generator 101 and coefficient SRcalculated by coefficient calculator 104.
[0051] Control unit 103L generates command revolution speed NLcomby performing calculation in connection with the difference (=SLt-SL) calculated by subtractor 102L by using a prescribed gain and outputs generated command revolution speed NLcomto left pump 10. Revolution speed NLof left pump 10 is thus controlled to command revolution speed NLcom, coefficient SLof left pump 10 is controlled to target coefficient SLt, and the blood at left pump flow rate COSis outputted to the systemic circulation of a subject 2.
[0052] Control unit 103R generates command revolution speed NRcomby performing calculation in connection with the difference (= SRt-SR) calculated by subtractor 102R by using a prescribed gain and outputs generated command revolution speed NRcomto right pump 20. Revolution speed NRof right pump 20 is thus controlled to command revolution speed NRcom, coefficient SRof right pump 20 is controlled to target coefficient SRt, and the blood at right pump flow rate COPis outputted to the pulmonary circulation.
[0053] A manner of control by control units 103L and 103R is not particularly limited. For example, the manner of control by control units 103L and 103R may be a manner of control by a proportional integral control unit (PI control unit), a manner of control by a proportional integral derivative control unit (PID control unit), or a manner of control such as model prediction control.
[0054] As a result of feedback control as above, the RBPs (left pump 10 and right pump 20) of the left and right hearts can be controlled as if they had the FS law of the native heart. Consequently, balance between the flow rates of the RBPs (left pump 10 and right pump 20) of the left and right hearts can accurately be stabilized and suction and congestion described above can appropriately be prevented.
[0055] The inventors of the subject application conducted experiments in which the left and right hearts of a test animal were replaced with left pump 10 and right pump 20 (which are also referred to as "pumps 10 and 20" below) and operations of pumps 10 and 20 were controlled by controller 100 according to the present first embodiment.
[0056] Fig. 5 is a diagram showing an exemplary result of a first experiment of control by controller 100. In the first experiment, blood transfusion and blood removal of the test animal were performed under the control by controller 100 shown in Fig. 4. In Fig. 5, the abscissa represents time and the ordinate represents arterial pressures {a blood pressure (systemic arterial pressure) and a pulmonary arterial pressure}, slopes (coefficients SLand SR) of the FS curves of the pumps, pump flow rates (COSand COP), and atrial pressures (PLand PR) sequentially from the top.
[0057] When the experiment is started, initially, coefficients SLand SRat the time point of start of control are calculated based on measurement values COS, PL, COP, and PRand the expressions (1) and (2), and target coefficients SLtand SRtat the time point of start of control are calculated based on target values COt, PLt, and PRtand the expressions (3) and (4). In the present embodiment, target values COt, PLt, and PRtare each a fixed value freely set by the inventors of the subject application.
[0058] Revolution speed NLof left pump 10 is controlled under feedback control such that coefficient SLis closer to target coefficient SLtand revolution speed NRof right pump 20 is controlled under feedback control such that coefficient SRis closer to target coefficient SRt.
[0059] It can be understood from the result of the first experiment shown in Fig. 5 that blood transfusion and blood removal causes variation in atrial pressure (PLand PR), and accordingly the pump flow rates (COSand COP) also vary, whereas the slopes (coefficients SLand SR) of the FS curves of the pumps are maintained substantially at constant values as a result of repetition of the feedback control described above.
[0060] Fig. 6 is a diagram showing relation between the atrial pressures (PLand PR) and the pump flow rates (COSand COP) in the result of the first experiment shown in Fig. 5. The graph on the left in Fig. 6 shows relation between left atrial pressure PLand left pump flow rate COSand the graph on the right in Fig. 6 shows relation between right atrial pressure PRand right pump flow rate COP.
[0061] It can be understood from the result of the first experiment shown in Fig. 6 that left pump flow rate COSvaries along the FSLcurve (the curve of a logarithmic function of left atrial pressure PL) having a constant value of slope SLand right pump flow rate COPvaries along the FSRcurve (the curve of a logarithmic function of right atrial pressure PR) having a constant value of slope SR. In other words, it can be understood from the result of the first experiment shown in Fig. 6 that left pump 10 and right pump 20 can be controlled as if they had the FS law of the native heart as a result of feedback control according to the present disclosure.
[0062] Furthermore, in feedback control according to the present disclosure (control according to the present disclosure), the slope of the FS curve of each of pumps 10 and 20 is expressed with a single indicator (coefficient SLor SR). The control system according to the present disclosure can thus achieve prompt and stable control in terms of control engineering.
[0063] Fig. 7 is a diagram conceptually showing a difference between control in a comparative example and control in the present disclosure. Fig. 7 shows transition of an operating point defined by the atrial pressure and the pump flow rate.
[0064] Under the control in the comparative example, when the atrial pressure varies from a value at a starting operating point, initially, in a first step, control for maintaining the pump flow rate constant is carried out, and in a subsequent second step, the operating point is controlled to make transition to a target point on the FS curve. Under such control, the operating point moves to the target point with an extra step (first step) being interposed once, and it takes time to stabilize the operating point at the target value.
[0065] In contrast, under the control in the present disclosure, coefficients SLand SRrepresenting the slopes of the FS curves of pumps 10 and 20 are continuously monitored (calculated) and pump revolution speeds NLand NRare controlled under feedback control such that coefficients SLand SRattain to respective target coefficients SLtand SRt. Thus, when the atrial pressure varies, the operating point can directly be moved to the target point. Therefore, in control in the present disclosure, the operating point can be stabilized at the target value more promptly than in control in the comparative example.
[0066] Fig. 8 is a diagram showing an exemplary result of a second experiment of control by controller 100. In the second experiment, a pressure at the exit of left pump 10 was abnormally increased by increase in vascular resistance by occlusion of the aorta under the control by controller 100 shown in Fig. 4.
[0067] A result in the case of control in the present disclosure is shown on the left in Fig. 8 and a result in the case of control in the comparative example is shown on the right in Fig. 8. During control in the comparative example shown on the right in Fig. 8, unlike control in the present disclosure, control for maintaining pump revolution speeds NLand NRat constant values is carried out.
[0068] During control in the comparative example, pump revolution speeds NLand NRare maintained at the constant values. When the aorta was occluded under such control, occurrence of increase in left atrial pressure which may be a cause for congestion of the lungs and lowering in right atrial pressure which may cause suction was observed.
[0069] In contrast, during control in the present disclosure, even when the pressure at the exit of left pump 10 is abnormally increased by occlusion of the aorta, left pump revolution speed NLis automatically increased owing to feedback control to maintain coefficient SLat the target value, and hence pump flow rates COSand COPof two pumps and atrial pressures PLand PRare maintained in a stable manner. Consequently, occurrence of increase in left atrial pressure and lowering in right atrial pressure can appropriately be suppressed.
[0070] Fig. 9 is a diagram showing an exemplary result of a third experiment of control by controller 100. In the third experiment, the pressure at the exit of right pump 20 was abnormally increased by increase in vascular resistance by pulmonary artery stenosis under the control by controller 100 shown in Fig. 4.
[0071] A result in the case of control in the present disclosure is shown on the left in Fig. 9 and a result in the case of control in the comparative example is shown on the right in Fig. 9. During control in the comparative example shown on the right in Fig. 9, unlike control in the present disclosure, control for maintaining pump revolution speeds NLand NRat constant values is carried out.
[0072] During control in the comparative example, pump revolution speeds NLand NRare maintained at the constant values. When the pulmonary artery is narrowed under such control, occurrence of increase in right atrial pressure which may be a cause for edema and lowering in left atrial pressure which may cause suction was observed. Furthermore, great lowering in blood pressure (systemic arterial pressure) and collapse of blood circulation were observed.
[0073] In contrast, in the control system in the present disclosure, even when the pressure at the exit of right pump 20 is abnormally increased by pulmonary artery stenosis, right pump revolution speed NRis automatically increased owing to feedback control to maintain coefficient SRat the target value, and hence pump flow rates COSand COPof two pumps and atrial pressures PLand PRare maintained in a stable manner. Consequently, occurrence of increase in right atrial pressure and lowering in left atrial pressure can appropriately be suppressed.
[0074] Since target values COt, PLt, and PRtare fixed values in the first embodiment described above, target coefficients SLtand SRtwhich are the target values of the slopes of the FS curves of pumps 10 and 20 are maintained at the constant values.
[0075] In the native heart, however, as shown in Fig. 3 described above, the slopes (coefficients SLand SR) of the FS curves of the left and right hearts are increased and / or decreased depending on a physical activity level. The blood flow necessary and sufficient required in the whole body is secured by such a characteristic of the native heart, without excessive increase and / or lowering in left and right atrial pressures PLand PR.
[0076] Therefore, the slopes (coefficients SLand SR) of the FS curves of the RBPs of the left and right hearts are desirably varied depending on the physical activity level also in the artificial heart including the RBPs, as in the native heart.
[0077] In view of such aspects, a controller 100A according to the present second embodiment varies the slopes (coefficients SLand SR) of the FS curves in accordance with a degree of variation in physical activities, with a technique below. Thus, even when the physical activities or sympathetic activities are varied, left and right pump flow rates COSand COPcan be increased and / or decreased to a physiologically appropriate level without excessive increase and / or lowering in left and right atrial pressures PLand PR.
[0078] Fig. 10 is a functional block diagram showing a control structure of controller 100A according to the present second embodiment. Controller 100A is obtained by addition of a control structure that sets target values COt, PLt, and PRtto controller 100 described above in accordance with the degree of variation in physical activities. Specifically, controller 100A is obtained by addition of a blood volume calculator 105, a visualizer 106, and setting units 107 and 108 to controller 100 described above. Since controller 100A is identical in other control structures and objects to be controlled thereby to above-described controller 100, detailed description will not be repeated.
[0079] <Circulatory Equilibrium Theory>
[0080] Prior to explanation of a technique for setting of target values COt, PLt, and PRtby controller 100A, initially, a circulatory equilibrium theory applied to controller 100A will be described.
[0081] The inventors of the subject application established the "circulatory equilibrium theory", which is a physiological theory to describe a physiological mechanism that determines a cardiac output and left and right atrial pressures with a minimum necessary parameter. According to this circulatory equilibrium theory, the whole circulatory system is divided into a heart portion consisting of the left and right hearts and a vascular portion consisting of the whole body (systemic circulation) and the pulmonary circulation, and an equilibrium state of the heart portion and the vascular portion is analyzed on a three-dimensional coordinate where a cardiac output CO, left atrial pressure PL, and right atrial pressure PRare defined as coordinate axes.
[0082] Cardiac output CO provided from the heart portion to the vascular portion increases with increase in left and right atrial pressures PLand PRin accordance with the FS law, and it can be expressed as one curve on the three-dimensional coordinate. A curve which represents cardiac output CO on the three-dimensional coordinate is also referred to as an "integrated FS cardiac output curve (FSi)" below. Integrated FS cardiac output curve FSi is a curve, on which cardiac output CO increases in proportion to logarithmic functions of atrial pressures PLand PR.
[0083] Fig. 11 is a diagram showing on a three-dimensional coordinate, the FSLcurve representing the left cardiac output, the FSRcurve representing the right cardiac output, and integrated FS cardiac output curve FSi. Though the FSLcurve and the FSRcurve are shown as curves on a two-dimensional coordinate in Fig. 3, they are shown as curved surfaces on the three-dimensional coordinate as shown on the left in Fig. 11. Integrated FS cardiac output curve FSi corresponds to a line of intersection between the curved surface representing the FSLcurve and the curved surface representing the FSRcurve on the three-dimensional coordinate, and shown as one curve as shown on the right in Fig. 11.
[0084] A blood volume that can be sent back from the vascular portion to the heart portion (which is also referred to as a "venous return COv" below) is considered. In the flow of blood, the left and right atriums are located upstream of the vascular portion. Therefore, venous return COvdecreases with increase in left and right atrial pressures PLand PR, and it can be expressed as a single plane on the three-dimensional coordinate. The plane that expresses venous return COvon the three-dimensional coordinate is also referred to as a "venous return surface VRS" below. Venous return surface VRS is a plane formed such that venous return COvdecreases with increase in atrial pressures PLand PR.
[0085] Fig. 12 is a diagram showing venous return surface VRS and integrated FS cardiac output curve FSi on the three-dimensional coordinate. As described above, venous return surface VRS is the plane, on which venous return decreases with increase in atrial pressures PLand PR, whereas integrated FS cardiac output curve FSi is a curve, on which cardiac output increases with increase in atrial pressures PLand PR. Therefore, as shown in Fig. 12, venous return surface VRS and integrated FS cardiac output curve FSi intersect with each other at a certain point on the three-dimensional coordinate. It can be analyzed or predicted that cardiac output CO and atrial pressures PLand PRof a subject can be determined from a coordinate of the intersection (which is also referred to as a "circulatory equilibrium point" below) between venous return surface VRS and integrated FS cardiac output curve FSi.
[0086] A blood volume that contributes to blood circulation in the whole circulatory system of the subject is referred to as a "stressed blood volume V." Venous return COvis then associated in a linear combination expression of stressed blood volume V and atrial pressures PLand PRas in an expression (C) below where k5 to k7 are constants found in living body tests or the like.
[0087] COv= V / k5-k6xPL-k7xPR... (C)
[0088] Venous return COvexpressed in the expression (C) is shown as above-described venous return surface VRS on the three-dimensional coordinate. Therefore, venous return surface VRS is shifted upward and downward on the three-dimensional coordinate shown in Fig. 12, with increase and / or decrease in stressed blood volume V. Specifically, since stressed blood volume V increases when the physical activities are intensified (when sympathetic activities are intensified as in exercises), venous return surface VRS is shifted upward. When the physical activities are reduced (when sympathetic activities are reduced as in sleep), stressed blood volume V decreases and hence venous return surface VRS is shifted downward.
[0089] <Setting of Target Values COt, PLt, and PRtbased on Circulatory Equilibrium Theory>
[0090] The technique for setting of target values COt, PLt, and PRtby controller 100A shown in Fig. 10 will now be described. As described above, controller 100A includes blood volume calculator 105, visualizer 106, and setting units 107 and 108 as the functional blocks for setting of target values COt, PLt, and PRt. Blood volume calculator 105, visualizer 106, and setting units 107 and 108 set target values COt, PLt, and PRtto values in accordance with the physical activities, based on the circulatory equilibrium theory described above.
[0091] Blood volume calculator 105 obtains a measurement value of each of left pump flow rate COS, left atrial pressure PL, right pump flow rate COP, and right atrial pressure PRfrom measurement devices 11 and 21 and calculates stressed blood volume V by substituting the obtained measurement values into an expression (5) below. The expression (5) is obtained by replacement of "COv" in the expression (C) above with "(COS+COP) / 2" and transformation of the resultant expression to an expression that derives stressed blood volume V.
[0092] V = {(COS+COP) / 2+k6xPL+k7xPR}xk5 ... (5)
[0093] Stressed blood volume V of the subject, into which pumps 10 and 20 were implanted, can thus be calculated in real time.
[0094] Visualizer 106 visualizes venous return surface VRS corresponding to stressed blood volume V calculated by blood volume calculator 105 by drawing the same on the three-dimensional coordinate. Venous return surface VRS is a plane that expresses the combination of cardiac output CO and atrial pressures PLand PRthat satisfies an expression (6) below.
[0095] CO = V / k5-k6xPL-k7xPR... (6)
[0096] Setting unit 107 sets an optimal point OP on venous return surface VRS visualized by visualizer 106. A target coordinate point (CO, PL, PR) can again optionally be set on visualized venous return surface VRS. Therefore, setting unit 107 sets a coordinate point among coordinate points on venous return surface VRS, at which cardiac output CO attains to a physiologically appropriate value, as optimal point OP without excessive increase and / or decrease in left and right atrial pressures PLand PR.
[0097] Setting unit 108 sets the coordinate (CO, PL, PR) at optimal point OP set by setting unit 107 to target values COt, PLt, and PRt.
[0098] Explanation of visualizer 106 and setting units 107 and 108 above is a conceptual description of the technique for setting of target values COt, PLt, and PRtbased on the circulatory equilibrium theory. In actual, visualizer 106 and setting units 107 and 108 can extract, for example, an optimal combination (optimal point OP) from a plurality of combinations (venous return surface VRS) of CO, PL, and PRthat satisfy the expression (6) above and set the extracted combination to target values COt, PLt, and PRt.
[0099] For example, by way of example of the technique for setting of target values COt, PLt, and PRt, correspondence between stressed blood volume V and optimal point OP (CO, PL, PR) that satisfies the expression (6) above is mapped in advance and stored in memory 120, optimal point OP (CO, PL, PR) corresponding to stressed blood volume V calculated by blood volume calculator 105 is calculated by referring to the map, and calculated optimal point OP (CO, PL, PR) can be set to target values COt, PLt, and PRt.
[0100] In another exemplary technique for setting of target values COt, PLt, and PRt, while target values PLtand PRtare fixed to constant values within a normal range, remaining target value COtmay be varied in accordance with stressed blood volume V. Specifically, in humans and animals, when physical activities are intensified by exercises or the like, stressed blood volume V increases as compared with that in a resting state. At this time, the left and right atrial pressures are maintained substantially constant, whereas the cardiac output varies within a range from two times to three times larger. In accordance with such a function of the subject, among three target values COt, PLt, and PRt, while target values PLtand PRtof the left and right atrial pressures are fixed to constant values, remaining target value COtof the pump flow rate may be updated continuously based on the expression (6) above.
[0101] Target values COt, PLt, and PRtset by setting unit 108 are used for calculation of target coefficients SLtand SRtby target value generator 101. Subsequent processing is the same as that by controller 100 described above.
[0102] As processing as above is repeated, target values COt, PLt, and PRtcan be set in accordance with the physical activities (stressed blood volume V). Thus, in accordance with the physical activities (stressed blood volume V) of the subject, target coefficients SLtand SRtwhich are target values of the slopes of the FS curves of pumps 10 and 20 can appropriately be varied. Consequently, even when stressed blood volume V of the subject increases or decreases when the physical activities vary, left and right pump flow rates COSand COPcan be increased or decreased to the physiologically appropriate level while left and right atrial pressures PLand PRare maintained within the normal range without excessive increase or decrease thereof.
[0103] <Result of Experiments of Control by Controller 100A>
[0104] The inventors of the subject application conducted first to third experiments in which left and right ventricles of a test animal were replaced with pumps 10 and 20 and operations of pumps 10 and 20 were controlled by controller 100A according to the present second embodiment.
[0105] Fig. 13 is a diagram showing an exemplary result of a first experiment of control by controller 100A. In the first experiment, stressed blood volume V was maintained constant without transfusion and blood removal of the test animal under the control by controller 100A shown in Fig. 10. Target values COt, PLt, and PRtshown in Fig. 13 are not necessarily medically appropriate.
[0106] It could be verified that, under control by controller 100A while stressed blood volume V was maintained constant, as shown in Fig. 13, left and right pump flow rates COSand COP, left atrial pressure PL, and right atrial pressure PRcould accurately be controlled to respective target values COt, PLt, and PRtwith a minor error.
[0107] Furthermore, target values COt, PLt, and PRtare values at optimal point OP on venous return surface VRS set based on stressed blood volume V of the subject as described above. Therefore, it could be verified from the result of the first experiment that left and right pump flow rates COSand COP, left atrial pressure PL, and right atrial pressure PRcould accurately be controlled to optimal point OP on venous return surface VRS set based on stressed blood volume V of the subject.
[0108] Fig. 14 is a diagram showing an exemplary result of a second experiment of control by controller 100A. In the second experiment, increase and / or decrease in stressed blood volume V during exercises and in a resting state of a test animal was simulated by increasing and / or decreasing stressed blood volume V by transfusion and blood removal of the test animal under the control by controller 100A shown in Fig. 10.
[0109] Fig. 15 is a diagram showing relation between atrial pressures PLand PRand pump flow rates COSand COPin the result of the second experiment shown in Fig. 14. An upper graph in Fig. 15 shows relation between left atrial pressure PLand left pump flow rate COS, and a lower graph in Fig. 15 shows relation between right atrial pressure PRand right pump flow rate COP. "P1" and "P2" shown in Fig. 15 represent a lower control limit value and an upper control limit value of left atrial pressure PL, respectively. Therefore, a range from lower limit value P1 to upper limit value P2 is a normal range of left atrial pressure PL. "P3" and "P4" shown in Fig. 15 represent a lower control limit value and an upper control limit value of right atrial pressure PR, respectively. Therefore, a range from lower limit value P3 to upper limit value P4 is a normal range of right atrial pressure PR.
[0110] In Fig. 14, setting is again made (see an arrow A1) such that, in response to left atrial pressure PLbecoming higher than the normal range as a result of transfusion, target coefficients SLtand SRtincrease based on stressed blood volume V at that time point. Since the slopes (coefficients SLand SR) of the FS curves thus increase as shown in Fig. 15, the "circulatory equilibrium point" which is the intersection between venous return surface VRS and integrated FS cardiac output curve FSi moves over the three-dimensional coordinate in a direction in which atrial pressures PLand PRdecrease and cardiac output CO increases. Consequently, left atrial pressure PLcan be lowered to upper limit value P2 or lower and maintained in the normal range while left and right pump flow rates COSand COPare increased.
[0111] Setting is again made (see an arrow A2) such that, in response to left atrial pressure PLbecoming lower than the normal range as a result of blood removal, target coefficients SLtand SRtdecrease based on stressed blood volume V at that time point. Since the slopes (coefficients SLand SR) of the FS curves thus decrease as shown in Fig. 15, the "circulatory equilibrium point" which is the intersection between venous return surface VRS and integrated FS cardiac output curve FSi moves over the three-dimensional coordinate in a direction in which atrial pressures PLand PRincrease and cardiac output CO decreases. Consequently, left atrial pressure PLcan be increased to lower limit value P1 or higher and maintained in the normal range while left and right pump flow rates COSand COPare lowered.
[0112] As set forth above, under the control by controller 100A, even when stressed blood volume V of the subject is increased or decreased due to variation in physical activities, left and right pump flow rates COSand COPcan automatically be increased or decreased to the physiologically appropriate level while left and right atrial pressures PLand PRare maintained in the normal range without excessive increase or decrease thereof.
[0113] In the second experiment shown in Fig. 14, variation in three target values COt, PLt, and PRtin accordance with stressed blood volume V at the timing of deviation of left atrial pressure PLfrom the normal range (the timing shown with arrows A1 and A2 in Fig. 14) is permitted.
[0114] As described above, however, in humans and animals, at the time of variation in stressed blood volume V due to physical activities, cardiac output CO is varied while left and right atrial pressures PLand PRare maintained substantially constant. Therefore, in order to drive pumps 10 and 20 more physiologically, among three target values COt, PLt, and PRt, while target values PLtand PRtof the left and right atrial pressures are fixed to constant values, remaining target value COtof the pump flow rate is desirably always updated to a value in accordance with stressed blood volume V so as to satisfy the expression (6) described above. In order to verify this point, inventors of the subject application conducted a third experiment of control by controller 100A.
[0115] Fig. 16 is a diagram showing an exemplary result of the third experiment of control by controller 100A. In the third experiment, under the control by controller 100A shown in Fig. 10, stressed blood volume V was increased or decreased by transfusion and blood removal of a test animal, and while target values PLtand PRtof the left and right atrial pressures were fixed to values at the time of start of control, stressed blood volume V was calculated in short cycles (for example, every several seconds) and target value COtof the pump flow rate was always updated to a value in accordance with stressed blood volume V.
[0116] When stressed blood volume V is increased by transfusion, as shown in Fig. 16, coefficients SLand SRare increased with increase in stressed blood volume V, whereas atrial pressures PLand PRare substantially accurately maintained at the value at the time of start of control.
[0117] When stressed blood volume V is decreased by blood removal, as shown in Fig. 16, coefficients SLand SRare decreased with decrease in stressed blood volume V, whereas atrial pressures PLand PRare substantially accurately maintained at the value at the time of start of control.
[0118] By thus varying target value COtin accordance with stressed blood volume V while target values PLtand PRtof the left and right atrial pressures are fixed to constant values in spite of variation in stressed blood volume V, left and right pump flow rates COSand COPcan be controlled as in the subject.
[0119] It should be understood that the embodiments disclosed herein are illustrative and non-restrictive in every respect. The scope of the present disclosure is defined by the terms of the claims rather than the description above and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims. Features exemplified in the present embodiment and features exemplified in the modification can be combined as appropriate.
[0120] 1 artificial heart system; 2 subject; 10 left pump; 11, 21 measurement device; 20 right pump; 100, 100A controller; 101 target value generator; 102L, 102R subtractor; 103L, 103R control unit; 104 coefficient calculator; 105 blood volume calculator; 106 visualizer; 107, 108 setting unit; 110 CPU; 120 memory
Claims
1. A control system of an artificial heart including a first rotary pump that replaces or assists a function of a left heart of a subject and a second rotary pump that replaces or assists a function of a right heart of the subject, the control system comprising: a measurement device that measures a flow rate of the first pump, a flow rate of the second pump, a left atrial pressure, and a right atrial pressure; and a control device that controls the first pump and the second pump, wherein a cardiac output of a left ventricle of the subject is represented by a first curve of a logarithmic function of the left atrial pressure and a cardiac output of a right ventricle of the subject is represented by a second curve of a logarithmic function of the right atrial pressure, and the control device calculates a first coefficient corresponding to a slope of the first curve based on a measurement value of the flow rate of the first pump and a measurement value of the left atrial pressure and calculates a second coefficient corresponding to a slope of the second curve based on a measurement value of the flow rate of the second pump and a measurement value of the right atrial pressure, and controls a revolution speed of the first pump under feedback control such that the first coefficient is closer to a first target coefficient and controls a revolution speed of the second pump under feedback control such that the second coefficient is closer to a second target coefficient.
2. The control system of the artificial heart according to claim 1, wherein the control device calculates the first coefficient SLand the second coefficient SRbased on expressions (1) and (2) below SL= COS / {log(PL-k1)+k2} ... (1) SR= COP / {log(PR-k3)+k4} ... (2) where COSrepresents the measurement value of the flow rate of the first pump, COPrepresents the measurement value of the flow rate of the second pump, PLrepresents the measurement value of the left atrial pressure, PRrepresents the measurement value of the right atrial pressure, SLrepresents the first coefficient, SRrepresents the second coefficient, and k1 to k4 represent first to fourth constants, respectively, calculates the first target coefficient SLtand the second target coefficient SRtbased on expressions (3) and (4) below SLt= COt / {log(PLt-k1)+k2} ... (3) SRt= COt / {log(PRt-k3)+k4} ... (4) where COtrepresents a target value of a pump flow rate, PLtrepresents a target value of the left atrial pressure, PRtrepresents a target value of the right atrial pressure, SLtrepresents the first target coefficient, and SRtrepresents the second target coefficient, and controls the revolution speed of the first pump under feedback control such that the first coefficient SLis closer to the first target coefficient SLtand controls the revolution speed of the second pump under feedback control such that the second coefficient SRis closer to the second target coefficient SRt.
3. The control system of the artificial heart according to claim 2, wherein the control device calculates a stressed blood volume based on the measurement values COS, COP, PL, and PR, the stressed blood volume representing a blood volume that contributes to blood circulation in the subject, and sets at least one of the target values COt, PLt, and PRtbased on the stressed blood volume.
4. The control system of the artificial heart according to claim 3, wherein the control device calculates the stressed blood volume V based on an expression (5) below V = {(COS+COP) / 2+k6xPL+k7xPR}xk5 ... (5) where V represents the stressed blood volume and k5 to k7 represent fifth to seventh constants, respectively.
5. The control system of the artificial heart according to claim 4, wherein the control device extracts one combination of a plurality of combinations of CO, PL, and PRthat satisfy an expression (6) below and sets the extracted combination as the target values COt, PLt, and PRt CO = V / k5-k6xPL-k7xPR... (6) where CO represents a cardiac output or pump flow.
6. The control system of the artificial heart according to claim 5, wherein the control device fixes the target values PLtand PRtin the expression (6) while the control device varies the target value COtin accordance with the stressed blood volume V.
7. The control system of the artificial heart according to claim 3, further comprising a memory where correspondence between the stressed blood volume and a combination of the target values COt, PLt, and PRtis stored in advance, wherein the control device sets the target values COt, PLt, and PRtcorresponding to the stressed blood volume by referring to the correspondence stored in the memory.
8. An artificial heart system comprising: a first rotary pump that replaces or assists a function of a left heart of a subject; a second rotary pump that replaces or assists a function of a right heart of the subject; a measurement device that measures a flow rate of the first pump, a flow rate of the second pump, a left atrial pressure, and a right atrial pressure; and a control device that controls the first pump and the second pump, wherein a cardiac output of a left ventricle of the subject is represented by a first curve of a scaled logarithmic function of the left atrial pressure and a cardiac output of a right ventricle of the subject is represented by a second curve of a scaled logarithmic function of the right atrial pressure, and the control device calculates a first coefficient corresponding to a slope of the first curve based on a measurement value of the flow rate of the first pump and a measurement value of the left atrial pressure and calculates a second coefficient corresponding to a slope of the second curve based on a measurement value of the flow rate of the second pump and a measurement value of the right atrial pressure, and controls a revolution speed of the first pump under feedback control such that the first coefficient is closer to a first target coefficient and controls a revolution speed of the second pump under feedback control such that the second coefficient is closer to a second target coefficient.
9. A method of controlling a blood pump including a first rotary pump that replaces or assists a function of a left heart of a subject and a second rotary pump that replaces or assists a function of a right heart of the subject, the method comprising: measuring a flow rate of the first pump, a flow rate of the second pump, a left atrial pressure, and a right atrial pressure; and controlling the first pump and the second pump, wherein a cardiac output of a left ventricle of the subject is represented by a first curve of a scaled logarithmic function of the left atrial pressure and a cardiac output of a right ventricle of the subject is represented by a second curve of a scaled logarithmic function of the right atrial pressure, and the controlling the first pump and the second pump includes calculating a first coefficient corresponding to a slope of the first curve based on a measurement value of the flow rate of the first pump and a measurement value of the left atrial pressure and calculating a second coefficient corresponding to a slope of the second curve based on a measurement value of the flow rate of the second pump and a measurement value of the right atrial pressure, and controlling a revolution speed of the first pump under feedback control such that the first coefficient is closer to a first target coefficient and controlling a revolution speed of the second pump under feedback control such that the second coefficient is closer to a second target coefficient.