JP7725064B2 - Apparatus, method, and program for calculating blood pressure index or blood flow index - Google Patents

Description

Article 30, paragraph 2 of the Patent Act applies. (Part 1) Date of website publication: September 7, 2021 Website address: https://journals.lww.com/jhypertension/Abstract/9000/Central_to_peripheral_stiffness_gradients.96558.aspx (Part 2) Date of website publication: October 21, 2021 Website address: https://www.arterysociety.org/wp-content/uploads/2021/10/Abstract-Book-artery. PDF (Part 3) Date: October 21st, 2021 - October 23rd, 2021 (Published on October 21st, 2021) Name of the meeting, location: 21st Annual Meeting of the Society for Research in Arterial Structure and Physiology (ARTERY21), Hôpital européen Georges-Pompidou 20, rue Leblanc, 75015 Paris, France.

The present invention relates to a device, method, and program for calculating a blood pressure index or a blood flow index.

In healthy people, the arteries are softer in the central aorta and stiffer toward the periphery, resulting in a phenomenon known as pulse pressure amplification, in which peripheral pulse pressure is greater than central aortic pulse pressure. Measurement of pulse pressure amplification has traditionally been widely used to evaluate a subject's cardiovascular system, particularly the degree of arterial stiffness (Non-Patent Document 1).

Measuring pulse pressure amplification requires measuring two points: aortic blood pressure and peripheral blood pressure, making the measurement process complicated. Furthermore, because aortic blood pressure is located deep in the trunk, direct measurement is difficult, and in most cases it is estimated using a transfer function obtained by measuring blood pressure waveforms at the arm and wrist (Non-Patent Document 2). It would be desirable to have a blood pressure index and/or blood flow index that can more easily and accurately evaluate a subject's cardiovascular system. Furthermore, it would be even more desirable if pulse pressure amplification and aortic blood pressure could be measured more easily.

Avolio AP et al., Hypertension 2009; 54: 375-383 Williams B et al. Circulation 2006; 113: 1213-1225

The problem that this invention aims to solve is to provide a device, method, and program for calculating blood pressure and/or blood flow indices that can more easily and accurately evaluate the cardiovascular system.

As a result of extensive research to solve the above problems, the inventors unexpectedly focused on the diastolic waveform of peripheral arteries and calculated the diastolic blood pressure variability index (PFI d ) using the first negative peak and the next positive peak of a curve obtained by fitting a blood pressure waveform to an exponential decay curve. They found that this blood pressure index can be used to easily and accurately evaluate the cardiovascular system, leading to the completion of the present invention. Furthermore, surprisingly, they also found that when the diastolic blood flow variability index (FFI _d ) is calculated using the negative peak and the next positive peak in the blood flow _waveform , this blood flow index can also be used to easily and accurately evaluate the cardiovascular system.

The present invention includes the following embodiments.

Item 1. A device for calculating a blood pressure index of a test subject,
a residual pressure waveform generator that fits the diastolic portion of the blood pressure waveform of the peripheral artery under test with an exponential decay curve to generate a residual pressure waveform having an initial negative peak and a subsequent positive peak;
a diastolic blood pressure variability index calculation unit that calculates a diastolic blood pressure variability index (PFI _d ) expressed by the following formula (1), where _PR1 is the pressure at the negative peak of the residual pressure waveform, _PR2 is the pressure at the positive peak, and PP _Peri is the pulse pressure of the peripheral artery;

A blood pressure index calculation device equipped with

Item 2. The blood pressure index calculation device described in Item 1, further comprising a pulse pressure amplification estimation unit that estimates pulse pressure amplification based on the diastolic blood pressure variability index.

Item 3. The blood pressure index calculation device described in Item 1, further comprising a central aortic blood pressure estimation unit that estimates central aortic blood pressure based on the diastolic blood pressure variability index.

Item 4. The blood pressure index calculation device according to Item 1, further comprising a vascular stiffness evaluation unit that evaluates the stiffness of the blood vessels of the test subject based on the diastolic blood pressure variability index.

Item 5. A device for calculating a blood flow index of a test subject, comprising:
a diastolic blood flow variability index calculation unit that calculates a diastolic blood flow variability index (FFI d ) expressed by the following formula (3), where V _FW is the blood flow velocity at the maximum positive peak in the systolic period of the blood flow velocity waveform of the peripheral artery to be examined with respect to the time change, V _BW is the blood flow velocity at the negative peak, and _{V FW2} is the blood flow velocity at the next positive peak in the diastolic period;

A blood flow index calculation device equipped with

Item 6. The blood flow index calculation device according to Item 5, further comprising a pulse pressure amplification estimation unit that estimates pulse pressure amplification based on the diastolic blood flow variability index.

Item 7. The blood flow index calculation device described in Item 5, further comprising a central aortic blood pressure estimation unit that estimates central aortic blood pressure based on the diastolic blood flow variability index.

Item 8. The blood flow index calculation device according to Item 5, further comprising a vascular stiffness evaluation unit that evaluates the stiffness of the blood vessels of the test subject based on the diastolic blood flow variability index.

Item 9. A method for calculating a blood pressure index of a test subject, comprising:
A method for calculating a blood pressure index of a test subject, comprising:
a residual pressure waveform generating step of fitting a diastolic portion of the blood pressure waveform of the peripheral artery under test with an exponential decay curve to generate a residual pressure waveform having an initial negative peak and a subsequent positive peak;
a diastolic blood pressure variability index calculation step of calculating a diastolic blood pressure variability index (PFI d ) expressed by the following formula (1), where P _R1 is the pressure at the negative peak of the residual pressure waveform, P _R2 is the pressure at the positive peak, and PP _Peri is the pulse pressure _of the peripheral artery;

A method for calculating blood pressure indicators including:

Item 10. The blood pressure index calculation method described in Item 9, further comprising a pulse pressure amplification estimation step of estimating pulse pressure amplification based on the diastolic blood pressure variability index.

Item 11. The blood pressure index calculation method described in Item 9, further comprising a central aortic blood pressure estimation step of estimating central aortic blood pressure based on the diastolic blood pressure variability index.

Item 12. A method for calculating a blood flow index in a diastolic phase of a test subject, comprising:
a blood flow velocity variability index calculation step of calculating a blood flow velocity variability index (FFI d ) expressed by the following formula (3), where V _FW is the blood flow velocity at the maximum positive peak in the systolic period of the blood flow velocity waveform of the peripheral artery to be examined with respect to time change, V _BW is the blood flow velocity at the negative peak, and _{V FW2} is the blood flow velocity at the next positive peak in the diastolic period;

A blood flow calculation method.

Item 13. The blood flow index calculation method according to Item 12, further comprising a pulse pressure amplification estimation step of estimating pulse pressure amplification based on the diastolic blood flow variability index.

Item 14. The blood flow index calculation method according to Item 12, further comprising a central aortic blood pressure estimation unit that estimates central aortic blood pressure based on the diastolic blood flow variability index.

Item 15. A program for causing a computer to execute a process for calculating a blood pressure index of a test subject,
Computer,
a residual pressure waveform generating unit that generates a residual pressure waveform having a negative peak that appears first and a positive peak that appears next by fitting the diastolic portion of the blood pressure waveform with respect to the time change of the peripheral artery of the test subject to an exponential decay curve, and a diastolic blood pressure variability index calculating unit that calculates a diastolic blood pressure variability index (PFI _d ) expressed by the following formula ( ₁ ) when the pressure at the negative peak of the residual pressure waveform is P _{R1 , the pressure at the positive peak is P R2} , and the pulse pressure of the peripheral artery is PP _Peri

A blood pressure index calculation program that functions as a

Item 16. A program for causing a computer to execute a process for calculating a blood flow index of a test subject, comprising:
Computer,
a blood flow velocity variability index calculation unit that calculates a blood flow velocity variability index (FFI d ) expressed by the following formula (3), where V _FW is the blood flow velocity at the maximum positive peak in the systolic period of the blood flow velocity waveform of the peripheral artery to be examined with respect to the time change, V _BW is the blood flow velocity at the negative peak, and _{V FW2} is the blood flow velocity at the next positive peak in the diastolic period.

A blood flow index calculation program that functions as a

Item 17. A computer-readable recording medium on which the program described in Item 15 or 16 is recorded.

The present invention makes it possible to easily and accurately evaluate a subject's cardiovascular system.

1 is a schematic diagram of the configuration of a blood pressure index calculation system. 4 is a flowchart showing the operation of a processing unit of the blood pressure index calculation device. 1 is a schematic diagram of the configuration of a blood flow index calculation system. 10 is a flowchart showing the operation of a processing unit of the blood flow index calculation device. (a) Graph of femoral artery blood pressure pulse waveform, (b) graph of diastolic residual pressure, (c) graph of blood flow velocity. Symbols in parentheses represent coordinates (x, y). T is time, P is pressure, V is velocity, FW is peak systolic forward flow, BW is peak retrograde (backward) flow, and FW2 is peak diastolic forward flow. Relationship between diastolic blood flow variability and pressure variability in the femoral artery (n=592). (a) Bland-Altman histograms for the time from systolic onset to peak backward (reverse) velocity (T _BW ) and end-systole (T _ES ) estimated from the pressure inflection point. The x-axis shows the mean ([T _ES +T _BW ]/2), the y-axis shows the difference (T _ES -T _BW ), and the z-axis shows the number of observations. The mean and 2SD difference are -0.002 and 0.046 s, respectively. (b) Graph showing the correlation between diastolic blood flow variability index and diastolic blood pressure variability index. (c) Bland-Altman histograms for the time from systolic onset to peak forward diastolic velocity (T _FW2 ) and second (positive) residual pressure peak (T _R2 ). The x-axis shows the mean ([T _R2 +T _FW2 ]/2), the y-axis shows the difference (T _R2 -T _FW2 ), and the z-axis shows the number of observations. The mean and 2SD difference are 0.071 and 0.091s, respectively. The effect of the arterial-leg arterial stiffness gradient [(a) PWV _FD /PWV _CF ratio] or pulse pressure amplification [(b) P _PD /P _PA ratio] on the diastolic blood pressure variability index (PFI _d ) mediated by _the diastolic flow variability index (FFI d ). For each mediation model, the effect is given as the standardized regression coefficient. The model number (i = 1, 2, ...) is indicated by a subscript (e.g., a _i ), where a _i , _bi , and _ci indicate direct effects, _ci is the total effect, and a _i * _bi is the indirect effect. Solid and dotted lines indicate significant and non-significant associations, respectively. **P less than 0.01, ***P less than 0.001.

(First embodiment)
First, the configuration of a blood pressure index calculation system 1 including a device for calculating a blood pressure index of a test subject according to a first embodiment of the present invention will be described. The blood pressure index calculation system 1 includes a blood pressure pulse wave measuring device 10 that measures the blood pressure waveform of the subject's peripheral artery, a blood pressure measuring device 20 that measures the subject's arterial blood pressure, and a blood pressure index calculation device 30 that includes a computer or the like that stores and processes data received from the blood pressure pulse wave measuring device 10 and the blood pressure measuring device 20.

The blood pressure index calculation device 30 includes a processing device 40 such as a CPU (Central Processing Unit), GPU (Graphics Processing Unit), ASIC (Application Specific Integrated Circuit), or FPGA (Field Programmable Gate Array) that performs calculations or judgments based on various data received from the blood pressure pulse wave measuring device 10 and the blood pressure measurement device 20 and/or data stored or generated internally; a storage device 50 such as a ROM, RAM, or hard disk that stores various data; and a display device 52 that displays the results of calculations performed by the processing device 40.

The blood pressure pulse wave measuring device 10 may be any blood pressure pulse wave measuring device capable of measuring blood pressure waveforms from the subject's peripheral arteries, including, for example, pulse pressure sensors such as tonometers and pulse oximeters. Peripheral arteries that can be measured with the blood pressure pulse wave measuring device 10 include, but are not limited to, the brachial artery, radial artery, carotid artery, femoral artery, and/or dorsalis pedis artery. The arterial blood pressure measured with the blood pressure pulse wave measuring device 10 includes the blood pressure waveform and pulse pressure of the peripheral arteries.

The blood pressure waveform stored in the blood pressure pulse wave measuring device 10 for a certain period of time (e.g., 5 to 30 seconds) is transmitted to the blood pressure index calculation device 30 and saved in the storage device 50 of the blood pressure index calculation device 30. In a preferred embodiment, a user of the blood pressure index calculation system 1 (hereinafter referred to as the user), who is a medical professional such as a doctor or nurse, can select data from a time period (e.g., 10 seconds) that the user determines to be stable from the stored blood pressure waveform data displayed on the time series axis on the display device 52, and select this data for subsequent calculation processing. In another preferred embodiment, the blood pressure waveform data to be sent from the blood pressure pulse wave measuring device 10 to the blood pressure index calculation device 30 is automatically selected by the blood pressure index calculation device 30.

The blood pressure measurement device 20 may be any device capable of measuring the subject's arterial blood pressure, such as a cuff-type oscillometer or a tonometer. Peripheral arteries that can be measured by the blood pressure measurement device 20 include, but are not limited to, the brachial artery, radial artery, and/or lower limb arteries. The arterial blood pressure measured by the blood pressure measurement device 20 includes the systolic blood pressure, diastolic blood pressure, mean blood pressure, and/or heart rate of the peripheral artery. Such blood pressure measurement methods are well known and can be performed by those skilled in the art. The blood pressure data measured by the blood pressure measurement device 20 is transmitted to the blood pressure index calculation device 30 and stored in the storage device 50 of the blood pressure index calculation device 30. In a preferred embodiment, the user of the blood pressure index calculation system 1 selects and/or averages blood pressure data that the user determines to be stable from the stored blood pressure data displayed on the display device 52, and uses the data for subsequent calculation processing. In another preferred embodiment, the selection of blood pressure data transmitted from the blood pressure measurement device 20 to the blood pressure index calculation device 30 is performed automatically by the blood pressure index calculation device 30.

The processing device 40 of the blood pressure index calculation device 30 comprises an averaged blood pressure waveform generation unit 41 that takes the ensemble average of the blood pressure waveform data received from the blood pressure pulse wave measuring device 10 over a predetermined period to generate an averaged blood pressure waveform, a residual pressure waveform generation unit 42 that fits the diastolic portion of the blood pressure waveform over time of the blood pressure waveform data of the peripheral artery of the test subject generated by the averaged blood pressure waveform generation unit 41 to an exponential decay curve to generate a residual pressure waveform having a first negative peak and a second positive peak, and a diastolic blood pressure variability index calculation unit ₄₃ that calculates the diastolic blood pressure variability index (PFI _d ) expressed by the following equation (1), where PR1 is the pressure at the first negative peak of the residual pressure waveform, _PR2 is the pressure at the second positive peak, and PP _Peri is the pulse pressure of the peripheral artery.

In the examples described below, an example is described in which a diastolic blood pressure variability index is calculated using blood pressure waveform data from the femoral artery. Figure 5(a) is a graph of an averaged arterial waveform obtained by taking the ensemble average of blood pressure waveform data from the subject's femoral artery measured by a blood pressure pulse wave measuring device 10, and Figure 5(b) is a graph of the residual pressure waveform that remains when the diastolic portion of the blood pressure waveform data in Figure 5(a) is fitted to an exponential decay curve.

The symbols in parentheses represent coordinates (x, y), T is time, and P is pressure. ES is end-systole, DN is dicrotic notch, and DW is dicrotic wave peak, all of which are well known to those skilled in the art of hemodynamics.

While previous research has often examined the systolic phase, this disclosure focuses on the diastolic phase of the blood pressure waveform.

When the diastolic blood pressure waveform is approximated to an exponential decay curve, the residual pressure has a first negative peak at (T _R1 , P _R1 ), a second positive peak at (T _R2 , P _R2 ), and a third negative peak at (T _R3 , P _R3 ), as shown in Figure 5(b). (T _DN , P _DN ) and ( _TDW , P _DW ) in Figure 5(a) correspond to (T _R1 , P _R1 ) and (T _R2 , P _R2 ), respectively. The peaks at (T _R1 , P _R1 ) and (T _R2 , P _R2 ) in Figure 5(b) were observed in all subjects examined in the examples. These facts were discovered for the first time by the inventors of the present application; conventionally, (T _ES , P _ES ) in Figure 5(a) was considered the boundary between systole and diastole, but equation (1) is characterized by measuring pressures _PR1 and _PR2 at two points different from (T _ES , P _ES ). The peripheral arterial pulse pressure PP _Peri is the difference between the systolic maximum and diastolic minimum values of the averaged arterial waveform (in Figure 5(a) 130 mmHg - 70 mmHg, which is approximately 60 mmHg). The peripheral arterial pulse pressure PP _Peri can be determined manually or automatically from the signal from the blood pressure pulse wave measuring device 10.

The exponential decay curve can be fitted using the well-known equation (2) (Liu Z et al., Am J Physiol 1986; 251 (3 pt 2):H588-600. and Kottenberg-Assenmacher E et al., Anesthesiology 2009; 110:370-379.), which can be optimized using the Levenberg-Marquardt algorithm.

where t is the time from end-systole, P(t) is the exponential pressure waveform at t, P ₀ is the estimated (optimized) pressure at end-systole, P _∞ is the optimized asymptotic pressure, and τ (tau) is the optimized time constant. τ can be a variable or a finite value, but is preferably a finite value for better fitting. Alternatively, P _∞ = 0 may be used.

Surprisingly, the diastolic blood pressure variability index PFI _d calculated by formula (1) has a strong correlation with pulse pressure amplification. Measurement of pulse pressure amplification requires measurements at two points, aortic blood pressure and peripheral blood pressure, making the measurement cumbersome. However, the diastolic blood pressure variability index PFI _d of the present disclosure can be calculated simply by measuring a single point in a peripheral artery, making it simple. By using the diastolic blood pressure variability index PFI _d of the present disclosure, the cardiovascular system of a subject can be evaluated more easily and accurately.

Furthermore, by using the diastolic blood pressure variability index PFI _d of the present disclosure, pulse pressure amplification and aortic blood pressure can be easily estimated without using a transfer function.

Furthermore, while pulse wave velocity (PWV) is a well-known indicator of vascular arteriosclerosis (Townsend RR et al. Hypertension 2015;66:698-722), the diastolic blood pressure variability index (PFI _d) disclosed herein also has a strong correlation with the ratio of PWVs of two peripheral arteries. Therefore, the diastolic blood pressure variability index (PFI _d ) disclosed herein can be used to easily evaluate arteriosclerosis without measuring pulse wave velocity.

The processing device 40 of the blood pressure index calculation device 30 may further include a pulse pressure amplification estimation unit 44 that estimates the pulse pressure amplification based on the diastolic blood pressure variability index PFI _d .

Specifically, the diastolic blood pressure variability index (PFI _d) is strongly correlated with pulse pressure amplification, which is the ratio of central aortic pulse pressure (PP _{A )} to peripheral arterial pulse pressure (PP peri) (PP _peri /PP _A ). Therefore, pulse pressure amplification can be estimated from the diastolic blood pressure variability index (PFI _d) .

In one example, the memory device 50 stores a function of the diastolic blood pressure variability index PFI _d and the pulse pressure amplification PP _peri /PP _A , and the pulse pressure amplification estimation unit 44 can estimate the pulse pressure amplification PP _peri /PP _A of a subject using this function from the diastolic blood pressure variability index of the subject.

In another example, the storage device 50 stores a reference value for the diastolic blood pressure variability index PFI _d and a reference value for the pulse pressure amplification PP _peri /PP _A , and when the diastolic blood pressure variability index of a certain subject is higher than the reference value for the diastolic blood pressure variability index PFI _d , the pulse pressure amplification estimation unit 44 can estimate that the pulse pressure amplification of the subject is higher than the reference value for the pulse pressure amplification PP _peri /PP _A. Note that the reference value for the diastolic blood pressure variability index PFI _d and the reference value for the pulse pressure amplification PP _peri /PP _A may be the average values of the diastolic blood pressure variability index PFI _d and the pulse pressure amplification PP _peri /PP _A of healthy subjects, respectively, or the average values of the diastolic blood pressure variability index PFI _d and the pulse pressure amplification PP _peri /PP _A of patients with heart disease, respectively.

In another example, the storage device 50 stores a reference value for the diastolic blood pressure variability index PFI _d and a reference value for the pulse pressure amplification PP _peri /PP _A , and when the diastolic blood pressure variability index of a certain subject is equal to or lower than the reference value for the diastolic blood pressure variability index PFI _d , the pulse pressure amplification estimation unit 44 can estimate that the pulse pressure amplification of the subject is equal to or lower than the reference value for the pulse pressure amplification PP _peri /PP _A. Note that the reference value for the diastolic blood pressure variability index PFI _d and the reference value for the pulse pressure amplification PP _peri /PP _A may be the average values of the diastolic blood pressure variability index PFI _d and the pulse pressure amplification PP _peri /PP _A of healthy subjects, respectively, or may be the average values of the diastolic blood pressure variability index PFI _d and the pulse pressure amplification PP _peri /PP _A of patients with heart disease, respectively.

In yet another example, the storage device 50 stores a reference value for the diastolic blood pressure variability index PFI _d and a reference value for the pulse pressure amplification PP _peri /PP _A , and the pulse pressure amplification estimation unit 44 can calculate an estimated value for the pulse pressure amplification of a certain subject from the relationship between the diastolic blood pressure variability index PFI _d and the reference value for the diastolic blood pressure variability index PFI d of the subject and the reference value for the pulse pressure amplification PP _peri /PP _A. Note that the reference values for the diastolic blood pressure variability index PFI _d and the pulse pressure amplification PP _peri /PP _A may be the average values of the diastolic blood pressure variability index PFI _d and the pulse pressure amplification PP _peri /PP _A of healthy subjects, respectively, or may be the average values of the diastolic blood pressure variability index PFI _d and the pulse pressure amplification PP _peri /PP _A of patients with heart disease, respectively.

The processing device 40 of the blood pressure index calculation device 30 may further include a central aortic blood pressure estimation unit 45 that estimates central aortic blood pressure based on the diastolic blood pressure variability index PFI _d . The central aortic blood pressure includes central aortic pulse pressure _PPA , central aortic systolic blood pressure, etc.

Specifically, since the diastolic blood pressure variability index PFI _d has a strong correlation with the pulse pressure amplification PP _peri /PP _A , the central aortic pulse pressure PP _A can be calculated from the values of the diastolic blood pressure variability index PFI _d and the peripheral arterial pulse pressure PP _peri . Furthermore, since the diastolic blood pressure of the peripheral artery and the central aorta are considered to be the same, the central aortic systolic blood pressure can be estimated from the central aortic pulse pressure PP _A. The peripheral arterial pulse pressure PP _peri can be measured by the blood pressure pulse wave measuring device 10 and transmitted to the blood pressure index calculation device 30.
When using brachial pulse pressure as the peripheral arterial pulse pressure PP _peri , it can be obtained directly from brachial cuff blood pressure measurement, etc. When calculating the peripheral arterial pulse pressure PP _peri on an artery other than the brachial artery, such as the femoral artery, the mean blood pressure and diastolic blood pressure obtained from brachial blood pressure measurement, etc., can be used as calibration blood pressures to calculate the peripheral arterial pulse pressure PP _peri from the peripheral arterial blood pressure waveform, assuming that the mean blood pressure and diastolic blood pressure are equivalent in all peripheral arteries.

In one example, the memory device 50 stores functions of the diastolic blood pressure variability index PFI _d , peripheral arterial pulse pressure PP _peri , and central aortic pulse pressure PP _A , and the central aortic blood pressure estimation unit 45 can estimate the central aortic pulse pressure PP _A of a subject from the diastolic blood pressure variability index and peripheral arterial pulse pressure of the subject, for example, using such functions.

Pulse pressure amplification PP _peri /PP _A = a × log(PFI _d ) + b × age + c × heart rate + d × BMI
(a, b, c, and d are constants that take unique values depending on the location of the peripheral artery.)
Central aortic pulse pressure PP _A = PP _peri ÷ pulse pressure amplification Central aortic systolic pressure = (PP _peri ÷ pulse pressure amplification) + peripheral arterial diastolic pressure

In another example, the storage device 50 stores a reference value for the diastolic blood pressure variability index PFI _d , a reference value for the peripheral arterial pulse pressure PP _peri , and a reference value for the central aortic pulse pressure PP _A , and the central aortic blood pressure estimation unit 45 can estimate whether a subject's central aortic pulse pressure PP _A is higher than the reference value for the central aortic pulse pressure PP _A based on the subject's diastolic blood pressure variability index and peripheral arterial pulse pressure. Note that the reference value for the diastolic blood pressure variability index PFI _d , the reference value for the peripheral arterial pulse pressure PP _peri , and the reference value for the central aortic pulse pressure PP _A may be the average values of the diastolic blood pressure variability index PFI d, the average values of the peripheral arterial pulse pressure PP _peri , and the average values of the central aortic pulse pressure PP _A of healthy subjects, respectively, or may be the average values of the diastolic blood pressure variability index PFI _d , the average values of the peripheral arterial pulse pressure PP _peri , and the average values of the central aortic pulse pressure PP _A of patients with cardiovascular disease _, respectively.

In yet another example, the storage device 50 stores a reference value for the diastolic blood pressure variability index PFI _d , a reference value for the peripheral arterial pulse pressure PP _peri , and a reference value for the central aortic pulse pressure PP _A , and the central aortic blood pressure estimation unit 45 can estimate the central aortic pulse pressure PP _A of a certain subject from the relationship between the diastolic blood pressure variability index and the reference value for the diastolic blood pressure variability index PFI _d of the subject, the relationship between the peripheral arterial pulse pressure of the subject and the reference value for the peripheral arterial pulse pressure PP _peri , and the reference value for the central aortic pulse pressure PP _A. The reference values for the diastolic blood pressure variability index PFI _d , the peripheral arterial pulse pressure PP _peri , and the central aortic pulse pressure PP _A may be the average values for the diastolic blood pressure variability index PFI _d , the peripheral arterial pulse pressure PP _peri , and the central aortic pulse pressure PP _A of healthy subjects, respectively, or may be the average values for the diastolic blood pressure variability index PFI _d , the peripheral arterial pulse pressure PP _peri , and the central aortic pulse pressure PP _A of patients with cardiovascular disease, respectively.

Central aortic pulse pressure PP _A = peripheral arterial pulse pressure PP _peri / pulse pressure amplification,
Since central aortic systolic blood pressure = central aortic pulse pressure PP _A + peripheral arterial diastolic blood pressure,
Central aortic systolic pressure = (peripheral arterial pulse pressure PP _peri / pulse pressure amplification) + peripheral arterial diastolic pressure,
Using this equation, the central aortic blood pressure estimation unit 45 can estimate the central aortic systolic blood pressure from the central aortic pulse pressure _PPA .

The processing device 40 of the blood pressure index calculation device 30 may further include a vascular stiffness evaluation unit 46 that evaluates the stiffness of the blood vessels of the test subject based on the diastolic blood pressure variability index.

In one example, the storage device 50 stores a reference value for the diastolic blood pressure variability index PFI _d , and the vascular stiffness assessment unit 46 can estimate that a subject's vascular sclerosis is likely to be advanced when the subject's diastolic blood pressure variability index PFI _d is equal to or lower than the reference value. The high probability can be displayed on the display device 52 in the form of a display, ranking, probability, or the like. The reference value for the diastolic blood pressure variability index PFI _d may be the average value of the diastolic blood pressure variability index PFI _d of patients with arteriosclerosis.

In another example, the storage device 50 stores a reference value for the diastolic blood pressure variability index PFI _d , and the vascular stiffness assessment unit 46 can estimate that the possibility of the subject having advanced vascular stiffness is low when the diastolic blood pressure variability index PFI _d of the subject is equal to or higher than the reference value for the diastolic blood pressure variability index PFI d. The low possibility can be displayed on the display device 52 in the form of a display, ranking, probability, or the like. The reference value for the diastolic blood pressure variability index PFI _d may be the average value of the diastolic blood pressure variability index PFI _d of healthy subjects.

2 is a flowchart showing the main operations of the processing device 40 of the blood pressure index calculation device 30 for calculating the blood pressure index of the subject. In step S1, the averaged blood pressure waveform generator 41 generates an averaged blood pressure waveform from the subject's peripheral arterial blood pressure waveform data. In step S2, the residual pressure waveform generator 42 fits the diastolic portion of the blood pressure waveform data received from the averaged blood pressure waveform generator 41 to an exponential decay curve to generate a residual pressure waveform having a first negative peak and a second positive peak. In step S3, the diastolic blood pressure variability index calculator 43 calculates the diastolic blood pressure variability index (PFI d ) expressed by the following equation (1), where _PR1 is the pressure at the first negative peak of the residual pressure waveform, _PR2 is the pressure at the second positive peak, and _PPPeri is the peripheral arterial _pulse pressure.

Optionally, in step S4, the pulse pressure amplification estimator 44 estimates the pulse pressure amplification based on the diastolic blood pressure variability index PFI _d .

Optionally, in step S5, the central aortic blood pressure estimation unit 45 estimates the central aortic blood pressure from the pulse pressure amplification estimated based on the diastolic blood pressure variability index PFI _d and the mean blood pressure (or systolic blood pressure) and diastolic blood pressure obtained by measuring the brachial blood pressure.

Optionally, in step S6, the vascular stiffness assessment unit 46 assesses the stiffness of the blood vessels of the test subject based on the diastolic blood pressure variability index PFI _d .

The method for calculating a blood pressure index for a test subject according to the first embodiment described above allows for the easy calculation of the diastolic blood pressure variability index, which is a blood pressure index. Furthermore, using this diastolic blood pressure variability index, it is possible to easily and accurately evaluate the subject's cardiovascular system, estimate pulse pressure amplification, estimate central aortic blood pressure, and evaluate the degree of vascular stiffness of the test subject.

In the first embodiment above, a blood pressure index calculation device 30 and a blood pressure index calculation method for a test subject were described, but the present invention is not limited to this and also includes a program for causing a computer to function as the processing device 40 of the blood pressure index calculation device 30, as well as a computer-readable recording medium on which the program is recorded.

The first embodiment can be modified as follows.
When measuring blood pressure waveform data, the measurement time may be determined manually by the user, or a certain time period (e.g., 5 to 30 seconds) may be preset, and all steps may be automatically measured by the processing device 22 so that measurement is automatically started and ended within that time period. Alternatively, the user may be able to switch between manual and automatic modes.
Instead of measuring blood pressure and blood flow separately, the blood pressure and blood flow of the subject may be measured simultaneously, and both sets of data may be synchronized and imported into a personal computer for automatic processing. In this case, more precise measurements are possible.
In the first embodiment, the blood pressure pulse wave measuring device 10 and the blood pressure measuring device 20 are described as separate devices, but the blood pressure pulse wave measuring device 10 and the blood pressure measuring device 20 may be the same device. In other words, the blood pressure waveform from the subject's peripheral artery and the subject's arterial blood pressure may be measured by a single device.
In the first embodiment, the mean blood pressure waveform generating unit 41, residual pressure waveform generating unit 42, diastolic blood pressure variability index calculating unit 43, pulse pressure amplification estimating unit 44, central aortic blood pressure estimating unit 45, and vascular stiffness evaluating unit 46 of the processing device 40 are all present within the processing device 40, but the units 41, 42, 43, 44, 45, and 46 may be present in different physically separated processing devices. In other words, the scope of the present invention includes not only the case where each unit within the processing device 40 is present within a processing device of the same hardware, but also the case where each unit is present in a separate processing device of different hardware.
In the first embodiment, the blood pressure index calculation device 30 includes the storage device 50 and the display device 52. However, the storage device 50 and/or the display device 52 may be configured to be separate from the processing device 40 and external to the blood pressure index calculation device 30, and may be communicatively connected to the processing device 40 via a communication line such as a LAN (Local Area Network) or the Internet. In this case, for example, the storage device 50 may store data processed in each step of the processing device 40, including the steps shown in FIG. 2. The storage device 50 may also store data generated by the blood pressure pulse wave measuring device 10, the blood pressure measurement device 20, or both, and send such data to the processing device 40 for subsequent processing in the processing device 40, including the steps shown in FIG. 2.

Second Embodiment
Next, a configuration of a blood flow index calculation system 101 including an apparatus for calculating a blood flow index of a test subject according to a second embodiment of the present invention will be described. As shown in Fig. 3, the blood flow index calculation system 101 includes a blood flow measurement device 110 that measures the blood flow waveform of a peripheral artery of a subject, a blood pressure measurement device 20 that measures the arterial blood pressure of the subject, and a blood pressure index calculation device 130 that includes a computer or the like that stores and processes data received from the blood flow measurement device 110 and the blood pressure measurement device 120.

The blood flow index calculation device 130 includes a processing device 40 such as a CPU (Central Processing Unit), GPU (Graphics Processing Unit), ASIC (Application Specific Integrated Circuit), or FPGA (Field Programmable Gate Array) that performs calculations or judgments based on various data received from the blood flow measurement device 110 and blood pressure measurement device 120 and/or data stored or generated internally; a storage device 150 such as a ROM, RAM, or hard disk that stores various data; and a display device 152 that displays the results of calculations performed by the processing device 140.

The blood flow measurement device 110 may be any device capable of measuring peripheral arterial blood flow, such as a known ultrasound device with a transducer. Using an ultrasound device with a transducer, it is possible to non-invasively collect aortic blood flow data, particularly waveform data of changes in blood flow velocity over time, from the subject's body surface.

Blood flow velocity or blood volume waveform data stored for a certain period of time (e.g., 5 to 30 seconds) in the blood flow measurement device 110 is transmitted to the blood flow index calculation device 130 and saved in the storage device 150. In a preferred embodiment, a user of the blood flow index calculation system 100 (hereinafter referred to as the "user"), who is a medical professional such as a doctor or nurse, can select data from a time period (e.g., 10 seconds) that the user determines to be stable from the saved blood flow velocity waveform data displayed on the time series axis on the display device 152, and select this data for subsequent calculation processing. In another preferred embodiment, the blood flow velocity waveform data to be sent from the blood flow measurement device 110 to the blood flow index calculation device 130 is automatically selected by the blood flow index calculation device 130.

The blood pressure measurement device 120 may be any device capable of measuring the subject's arterial blood pressure, such as a cuff-type oscillometer. Peripheral arteries that can be measured by the blood pressure measurement device 120 include, but are not limited to, the brachial artery, radial artery, and/or lower limb arteries. The arterial blood pressure measured by the blood pressure measurement device 120 includes the systolic blood pressure, diastolic blood pressure, mean blood pressure, and/or heart rate of the peripheral artery. Such blood pressure measurement methods are well known and can be performed by those skilled in the art with ordinary skill. The blood pressure data measured by the blood pressure measurement device 120 is transmitted to the blood flow index calculation device 130 and stored in the memory device 150 of the blood flow index calculation device 130. In a preferred embodiment, the user of the blood flow index calculation system 100 can select and/or average blood flow data that the user determines to be stable from the stored blood flow data displayed on the display device 152, and use the data for subsequent calculation processing. In another preferred embodiment, the selection of blood pressure data to be transmitted from the blood pressure measurement device 120 to the blood flow index calculation device 130 is performed automatically by the blood flow index calculation device 130.

The processing device 140 of the blood flow index calculation device 130 includes an average blood flow velocity waveform generation unit 141 that takes the ensemble average of blood flow velocity waveform data received from the blood flow measurement device 110 over a predetermined period to generate an averaged blood flow velocity waveform, and a diastolic blood flow variability index calculation unit 143 that calculates the diastolic blood flow variability index (FFI d ) expressed by the following equation (3), where V _FW is the blood flow velocity at the maximum positive peak during systole of the blood flow velocity waveform over time of the blood flow velocity waveform data of the peripheral artery under test generated by the averaged blood flow velocity waveform generation unit ₁₄₁ , _{V BW} is the blood flow velocity at the negative peak, and V FW2 is the blood flow velocity at the next positive peak during diastole.

In the examples described below, an example is described in which the diastolic blood flow variability index is calculated using blood flow velocity waveform data from the femoral artery. Figure 5(c) is a graph of an averaged arterial blood flow velocity waveform obtained by taking the ensemble average of blood flow velocity waveform data from the subject's femoral artery measured by the blood flow measurement device 110.

The symbols in parentheses represent coordinates (x, y), T is time, and V is velocity. The waveforms in Figure 5(c) and Figure 5(b) have similar shapes in the diastolic portion, but time T _BW in Figure 5(c) is slightly earlier than time T _R1 in Figure 5(b).

Surprisingly, the diastolic blood flow variability index FFI _d calculated by equation (3) has a strong correlation with pulse pressure amplification. Measurement of pulse pressure amplification requires measurements at two points, aortic blood pressure and peripheral blood pressure, making the measurement cumbersome. However, the diastolic blood pressure variability index PFI _d of the present disclosure can be calculated simply by measuring a single point in a peripheral artery, making it simple. By using the diastolic blood flow variability index FFI _d of the present disclosure, the cardiovascular system of a subject can be evaluated more easily and accurately.

Furthermore, by using the diastolic blood flow variability index FFI _d of the present disclosure, pulse pressure amplification and aortic blood pressure can be easily estimated without using a transfer function.

Furthermore, while pulse wave velocity (PWV) is a well-known index of vascular arteriosclerosis, the FFI _d of the present disclosure also has a strong correlation with the ratio of PWVs of two peripheral arteries. Therefore, the FFI _d of the present disclosure can be used to easily evaluate arteriosclerosis without measuring pulse wave velocity.

Furthermore, the diastolic blood flow variability index FFI _d was highly correlated with the diastolic blood pressure variability index PFI _d (R=0.63, P<0.001).

The processing device 140 of the blood flow index calculation device 130 may further include a pulse pressure amplification estimation unit 144 that estimates the pulse pressure amplification based on the diastolic blood flow variability index FFI _d .

Specifically, the diastolic blood flow variability index (FFI _d) is strongly correlated with pulse pressure amplification, which is the ratio of central aortic pulse pressure (PP _{A )} to peripheral arterial pulse pressure amplification (PP peri) (PP _peri /PP _A ). Therefore, pulse pressure amplification can be estimated from the diastolic blood flow variability index (FFI _d) .

In one example, the memory device 150 stores a function of the diastolic blood flow variability index FFI _d and the pulse pressure amplification PP _peri /PP _A , and the pulse pressure amplification estimation unit 144 can estimate the pulse pressure amplification PP _peri /PP _A of a certain subject from the diastolic blood flow variability index of the subject, for example, using this function.

Pulse pressure amplification PP _peri /PP _A = a × FFI _d + b × age + c × heart rate + d × BMI
(a, b, c, and d are constants that take unique values depending on the location of the peripheral artery.)

In another example, the storage device 150 stores a reference value for the diastolic blood flow variability index FFI _d and a reference value for the pulse pressure amplification PP _peri /PP _A , and when the diastolic blood flow variability index FFI d of a certain subject is higher than the reference value for the diastolic blood flow variability index FFI _d , the pulse pressure amplification estimation unit 144 can estimate that the pulse pressure amplification of the subject is higher than the reference value for the pulse pressure amplification PP _peri /PP _A. Note that the reference value for the diastolic blood flow variability index FFI _d and the reference value for the pulse pressure amplification PP _peri /PP _A may be the average values of the diastolic blood flow variability index FFI _d and the pulse pressure amplification PP _peri /PP _A of healthy subjects, respectively, or may be the average values of the diastolic blood flow variability index FFI _d and the pulse pressure amplification PP _peri /PP _A of patients with heart disease, respectively.

In another example, the storage device 150 stores a reference value for the diastolic blood flow variability index FFI _d and a reference value for the pulse pressure amplification PP _peri /PP _A , and when the diastolic blood flow variability index FFI d of a certain subject is equal to or less than the reference value for the diastolic blood flow variability index FFI _d , the pulse pressure amplification estimation unit 144 can estimate that the pulse pressure amplification of the subject is equal to or less than the reference value for the pulse pressure amplification PP _peri /PP _A. Note that the reference value for the diastolic blood flow variability index FFI _d and the reference value for the pulse pressure amplification PP _peri /PP _A may be the average values of the diastolic blood flow variability index FFI _d and the pulse pressure amplification PP _peri /PP _A of healthy subjects, respectively, or may be the average values of the diastolic blood flow variability index FFI _d and the pulse pressure amplification PP _peri /PP _A of patients with heart disease, respectively.

In yet another example, the storage device 150 stores a reference value for the diastolic blood flow variability index FFI _d and a reference value for the pulse pressure amplification PP _peri /PP _A , and the pulse pressure amplification estimation unit 144 can calculate an estimated value for the pulse pressure amplification of a certain subject from the relationship between the diastolic blood flow variability index FFI _d and the reference value for the diastolic blood flow variability index FFI d of the subject and the reference value for the pulse pressure amplification PP _peri /PP _A. Note that the reference values for the diastolic blood flow variability index FFI _d and the pulse pressure amplification PP _peri /PP _A may be the average values of the diastolic blood flow variability index FFI _d and the pulse pressure amplification PP _peri /PP _A of healthy subjects, respectively, or may be the average values of the diastolic blood flow variability index FFI _d and the pulse pressure amplification PP _peri /PP _A of patients with heart disease, respectively.

The processing device 140 of the blood flow index calculation device 130 may further include a central aortic blood pressure estimation unit 145 that estimates the central aortic blood pressure based on the diastolic blood flow variability index FFI _d .

Specifically, since the diastolic blood flow variability index FFI _d is strongly correlated with the ratio of the central aortic pulse pressure PP _A to the peripheral arterial pulse pressure PP _peri (PP _peri /PP _A ), the central aortic pulse pressure PP A can be calculated from the values of the diastolic blood flow variability index FFI _d and the peripheral arterial pulse pressure PP _{peri .} Furthermore, since the diastolic blood pressure of the peripheral artery and the central aorta is considered to be the same, the central aortic systolic blood pressure can be estimated from the central aortic pulse pressure PP _A. The peripheral arterial pulse pressure PP _peri can be measured by the blood pressure pulse wave measuring device 10 and transmitted to the blood flow index calculating device 130.

In one example, functions of the diastolic blood flow variability index FFI _d , peripheral arterial pulse pressure PP _peri , and central aortic pulse pressure PP _A are stored in the memory device 150, and the central aortic blood pressure estimation unit 145 can estimate the central aortic pulse pressure PP _A of a certain subject from the diastolic blood flow variability index and peripheral arterial pulse pressure of the subject, for example, using such functions.

Pulse pressure amplification PP _peri /PP _A = a × FFI _d + b × age + c × heart rate + d × BMI
(a, b, c, and d are constants that take unique values depending on the location of the peripheral artery.)
Central aortic pressure = (PP _peri ÷ pulse pressure amplification) + peripheral arterial diastolic pressure

In another example, the storage device 150 stores a reference value of the diastolic blood flow variability index FFI _d , a reference value of the peripheral arterial pulse pressure PP _peri , and a reference value of the central aortic pulse pressure PP _A , and the central aortic blood pressure estimation unit 145 can estimate, from the diastolic blood flow variability index and peripheral arterial pulse pressure of a certain subject, whether the central aortic pulse pressure PP _A is higher than the reference value of the central aortic pulse pressure PP _A. Note that the reference value of the diastolic blood flow variability index FFI _d , the reference value of the peripheral arterial pulse pressure PP _peri , and the reference value of the central aortic pulse pressure PP _A may be the average values of the diastolic blood flow variability index FFI _d , the average values of the peripheral arterial pulse pressure PP _peri , and the average values of the central aortic pulse pressure PP _A of healthy subjects, respectively, or may be the average values of the diastolic blood flow variability index FFI _d , the average values of the peripheral arterial pulse pressure PP _peri , and the average values of the central aortic pulse pressure PP _A of patients with cardiovascular disease, respectively.

In yet another example, the storage device 150 stores a reference value of the diastolic blood flow variability index FFI _d , a reference value of the peripheral arterial pulse pressure PP _peri , and a reference value of the central aortic pulse pressure PP _A , and the central aortic blood pressure estimation unit 145 can estimate the central aortic pulse pressure PP _A of a certain subject from the relationship between the diastolic blood flow variability index and the reference value of the diastolic blood flow variability index FFI _d of the subject, the relationship between the peripheral arterial pulse pressure of the subject and the reference value of the peripheral arterial pulse pressure PP _peri , and the reference value of the central aortic pulse pressure PP _A. The reference values of the diastolic blood flow variability index FFI _d , the peripheral arterial pulse pressure PP _peri , and the central aortic pulse pressure PP _A may be the average values of the diastolic blood flow variability index FFI _d , the peripheral arterial pulse pressure PP _peri , and the central aortic pulse pressure PP _A of healthy subjects, respectively, or may be the average values of the diastolic blood flow variability index FFI _d , the peripheral arterial pulse pressure PP _peri , and the central aortic pulse pressure PP _A of patients with cardiovascular disease, respectively.

The processing device 140 of the blood flow index calculation device 130 may further include a vascular stiffness evaluation unit 146 that evaluates the stiffness of the blood vessels of the test subject based on the diastolic blood flow variability index FFI _d .

In one example, the storage device 150 stores a reference value for the FFI _d , and the vascular stiffness assessment unit 146 can estimate that there is a high possibility that the subject's vascular sclerosis is progressing when the subject's vascular stiffness index is equal to or lower than the reference value for the FFI _d . The high possibility can be displayed on the display device 52 in the form of a display, ranking, probability, or the like. The reference value for the FFI _d may be the average value of the FFI _d of patients with arteriosclerosis.

In another example, the storage device 150 stores a reference value for the FFI _d , and the vascular stiffness assessment unit 146 can estimate that the possibility of the subject having advanced vascular stiffness is low when the FFI _d of the subject is equal to or higher than the reference value. The low possibility can be displayed on the display device 52 in the form of a message, a ranking, a probability, or the like. The reference value for the FFI _d may be the average value of the FFI _d of healthy subjects.

4 is a flowchart showing the main operations of the processing device 140 of the blood flow index calculation device 130 for calculating the blood flow index of the test subject. In step S101, the average blood velocity waveform generation unit 141 generates an average blood flow velocity waveform from blood flow velocity waveform data of the subject's peripheral artery. In step S102, the diastolic blood flow variability index calculation unit 143 calculates the diastolic blood flow variability index (FFI d ) expressed by the following equation (3), where V _FW is the blood flow velocity at the maximum positive systolic peak of the average blood flow velocity waveform, V _BW is the blood flow velocity at the negative peak, and V _FW2 is the blood flow _velocity at the next positive diastolic peak.

Optionally, in step S103, the pulse pressure amplification estimator 144 estimates the pulse pressure amplification based on the diastolic flow variability index FFI _d .

Optionally, in step S104, the central aortic blood pressure estimator 145 estimates the central aortic blood pressure based on the diastolic flow variability index FFI _d .

Optionally, in step S105, the vascular stiffness assessment unit 146 assesses the stiffness of the blood vessel of the test subject based on the diastolic flow variability index FFI _d .

The method for calculating a blood flow index of a test subject according to the second embodiment described above allows for the easy calculation of the diastolic blood flow variability index, which is a blood flow index. Furthermore, using this diastolic blood flow variability index, it is possible to easily and accurately evaluate the subject's cardiovascular system, estimate pulse pressure amplification, estimate central aortic blood pressure, and evaluate the degree of vascular stiffness of the test subject.

In the second embodiment above, a blood flow index calculation device 130 and a blood flow index calculation method for a test subject were described, but the present invention is not limited to this and also includes a program for causing a computer to function as the processing device 140 of the blood flow index calculation device 130 described above, and a computer-readable recording medium on which the program is recorded.

The second embodiment can be modified as follows.
When measuring blood flow waveform data, the measurement time may be determined manually by the user, but a certain time period (for example, 5 to 30 seconds) may be preset, and all steps may be automatically measured by the processing device 22 so that measurement is automatically started and ended within that time period. Also, a configuration may be adopted in which the user can switch between manual and automatic operation.
Instead of measuring blood pressure and blood flow separately, the blood pressure and blood flow of the subject may be measured simultaneously, and both sets of data may be synchronized and imported into a personal computer for automatic processing. In this case, more precise measurements are possible.
In the second embodiment, the blood pressure pulse wave measuring device 10 and the blood pressure measuring device 20 are described as separate devices, but the blood pressure pulse wave measuring device 10 and the blood pressure measuring device 20 may be the same device. In other words, a single device may measure the blood pressure waveform from the subject's peripheral artery and the subject's arterial blood pressure.
In the second embodiment, the mean blood flow waveform generating unit 141, the diastolic blood flow variability index calculating unit 143, the pulse pressure amplification estimating unit 144, the central aortic blood pressure estimating unit 145, and the vascular stiffness evaluating unit 146 of the processing device 140 are all present within the processing device 140, but the units 141, 143, 144, 145, and 146 may be present in different physically separated processing devices. In other words, the scope of the present invention includes not only the case where each unit within the processing device 140 is present within a processing device of the same hardware, but also the case where each unit is present in a separate processing device of different hardware.
In the second embodiment, the blood flow index calculation device 130 includes the storage device 150 and the display device 152. However, the storage device 150 and/or the display device 152 may be configured to be separate from the processing device 140 and external to the blood flow index calculation device 130, and to be communicatively connected to the processing device 140 via a communication line such as a local area network (LAN) or the Internet. In this case, for example, the storage device 150 may store data processed in each step of the processing device 40, including the steps shown in FIG. 4. The storage device 150 may also store data generated by the blood flow measurement device 110, the blood pressure measurement device 120, or both, and send such data to the processing device 140 for subsequent processing in the processing device 140, including the steps shown in FIG. 4.

(Third embodiment)
A blood pressure and blood flow index calculation system according to a third embodiment of the present invention includes both a blood pressure index calculation system 1 including an apparatus for calculating a blood flow index of a test subject according to the first embodiment, and a blood flow index calculation system 101 including an apparatus for calculating a blood flow index of a test subject according to the second embodiment. The blood pressure and blood flow index calculation system according to the third embodiment of the present invention includes a processing device that performs calculations or determinations based on various data received from the blood pressure pulse wave measuring device 10, the blood flow measuring device 110, and the blood pressure measuring devices 20 and 120 and/or data stored or generated internally, a storage device such as a ROM, RAM, or hard disk that stores various data, and a display device that displays the results of calculations performed by the processing device.

The processing device of the third embodiment of the present invention includes the processing device 40 of the first embodiment and the processing device 140 of the second embodiment, the storage device of the third embodiment is one or more storage devices that perform the functions of the storage device 50 of the first embodiment and the storage device 150 of the second embodiment, and the display device of the third embodiment is one or more display devices that perform the functions of the display device 52 of the first embodiment and the display device 152 of the second embodiment.

The present invention will be explained in more detail below using examples, but the present invention is not limited to these.

1. Patients: The patients in this study were 592 adult patients referred to the inventor's department at Tohoku University Hospital for clinical evaluation of hypertension and related cardiovascular risk. Among these, patients with overt heart failure, valvular heart disease, stroke, end-stage renal disease, peripheral arterial disease (ankle-brachial index <0.9 or >1.4, or documented), aortic aneurysm, persistent atrial fibrillation, and acute cardiovascular events within the past 6 months were excluded. None of the patients had overt clinical symptoms in the extremities. The study protocol was approved by the Tohoku University Ethics Committee, and all participants provided written informed consent.

2. Measurement of Arterial Pressure and Stiffness. A series of noninvasive hemodynamic measurements were performed in a quiet, temperature-controlled room as detailed in Hashimoto J. et al., Hypertension 2010; 56:926-933 and Hashimoto J. et al., Hypertension 2011; 58:839-846. Briefly, after 20 minutes of supine rest, patients' arm blood pressure was measured twice using a cuff-oscillometric device (HEM-907; Omron Healthcare, Kyoto, Japan). Pressure waves were recorded in the radial, common femoral, and dorsalis pedis arteries using applanation tonometer sensors (SPT-301; Milllar Instruments, Houston, TX, USA), and the respective beat-to-beat pulse waveforms were ensemble-averaged. Aortic waveforms were estimated from the corresponding radial waveforms using a generalized transfer function (SphygmoCor, AtCor Medical, West Ryde, Australia). Radial waveforms were calibrated using brachial systolic and diastolic pressures to calculate mean arterial pressure (MAP). Aortic, femoral, and dorsalis pedis waveforms were also calibrated by averaging the MAP and diastolic pressure levels of the aortic and peripheral arterial signals (Avolio AP et al., Hypertension
2009; 54:375-383. and Hashimoto J. et., Hypertension 2010; 56:926-933).

Aorto-femoral pulse wave amplitude, aorto-dorsalis pedis pulse wave amplitude, and aorto-radial pulse wave amplitude (PP _F /PP _A , PP _D /PP _A , and PP _R /PP _A ) were calculated as the ratio of each peripheral arterial pulse pressure to the aortic pulse pressure (PP _A ) (Hashimoto J. et., Hypertension 2010; 56:926-933). PP _F , PP _D , and PP _R represent the pulse pressures of the femoral artery, dorsalis pedis artery, and radial artery, respectively.

Pulse wave velocity (PWV) was calculated at the carotid-femoral, femoral-dorsalis pedis, and carotid-radial sites (i.e., _PWVCF , _PWVFD , and _PWVCR ) as previously described (Hashimoto J. et al., Hypertension 2010; 56:926-933 and Hashimoto J. et al., Hypertension 2011; 58:839-846). PWV _CF reflects aortic (elastic) stiffness, while PWV _FD and PWV _CR reflect peripheral (muscular) stiffness. The ratios of PWV _FD and PWV _CR to PWV _CF (i.e., PWV _FD /PWV _CF and PWV _CR /PWV _CF ) have previously been considered to represent the aortic-to-peripheral (i.e., elastic-to-muscular) arterial stiffness gradient (London GM et al., Hypertension 2019; 74:1366-1372. and Hashimoto J. et al., Hypertension 2013; 62:542-549).

3. Diastolic Blood Pressure Waveform Analysis: The diastolic portion of the femoral artery pressure waveform was fitted to an exponential decay curve using a variable asymptote (Figure 5(a)). First, the onset of diastole (i.e., end-systole, time T _ES ) was defined as the second inflection point of the pulse wave in the entire cardiac period, which corresponds to the second positive peak of the second derivative wave (Balmer J et al., J Clin Monit Comput 2021; 35:79-88). The end of diastole was determined as the foot of the next cardiac cycle. Then, the first 95% of the diastolic period was extracted as the diastolic optimization window to avoid artifacts related to the pre-ejection period.

Curve fitting was performed according to the following equation (2) (Liu Z et al., Am J Physiol 1986;251 (3 pt 2):H588-600. and Kottenberg-Assenmacher E et al., Anesthesiology 2009;110:370-379.) and optimized using the Levenberg-Marquardt algorithm (OriginPro 2018; Origin Lab, Northampton, MA, USA).

where t is the time from end-systole, P(t) is the exponential pressure waveform at t, P ₀ is the estimated (optimized) pressure at end-systole, P _∞ is the optimized asymptotic pressure, and τ (tau) is the optimized time constant. In this example, τ was set to finite. The goodness of fit was evaluated using the coefficient of determination (R ² ) and the root mean square error (RMSE; i.e., the average difference in instantaneous pressure between the measured and estimated curves).

The residual pressure waveform was then derived as the difference between the measured and estimated curves (Fig. 5(b)). Typically, the residual pressure waveform had three pressure peaks: the first negative peak ( _PR1 ), the second positive peak ( _PR2 ), and the third negative peak ( _PR3 ). These peaks were automatically detected as local minima or maxima using peak analysis software (OriginPro). The times from the systolic onset of the original waveform to these peaks were also determined ( _TR1 , _TR2 , and _TR3 ). Therefore, the amplitude of fluctuations in the residual waveform (i.e., diastolic pressure fluctuation amplitude, _PFAd ) was calculated according to the following equation (4):

The diastolic blood pressure fluctuation amplitude was then normalized to the total amplitude of the original pulse wave (i.e., femoral artery pulse pressure), and the diastolic blood pressure fluctuation index (PFI _d ) was calculated according to the following formula (1): _{PP Peri} = femoral artery pulse pressure PP _F .

It should be noted that the calculation of PFI _d is solely waveform-dependent and is not affected by any pressure calibration. Furthermore, following the algorithm previously reported (Oppenheim MI et al., Comput Biomed Res 1995; 28:154-170), the diploic notch was identified as a notch (trough) or downward bending (deflection) between the end-systolic and diploic waves (Figure 5(a)). Specifically, a dipping notch (trough) was considered to be present if there was a zero-crossing point from negative to non-negative values in the corresponding first-derivative wave. If no trough was present, a downward zero-crossing point in the second-derivative wave was instead detected as a deflection notch. In this example, because a clear trough could not be detected in a significant number of waveforms, the dipping notch and deflection notch are collectively referred to as the diploic notch. The (hidden) peak of the diploic wave was then defined as the positive shoulder after the diploic prominence (Figure 5(a)), and the onset time was determined using the residual pressure after the first-order lead method (OriginPro). Instantaneous pressure and time from the onset of systole were calculated for the dicrotic notch (P _DN and T _DN ) and dicrotic wave peak (P _DW and T _DW ).

4. Blood flow velocity waveform analysis. Blood flow velocity in the common femoral artery was recorded using duplex ultrasonography (Vivid i, GE Healthcare, Tokyo, Japan) as previously described (Hashimoto J et al., Hypertension 2010; 56:926-933). Beat-to-beat pulse velocity waveforms were mathematically ensemble-averaged over 10 consecutive pulses. The femoral artery pulse waveform was essentially triphasic, and the following parameters (Figure 5(c)) were determined: systolic forward flow peak velocity ( _VFW ), reverse (backward) flow peak velocity (VBW), diastolic forward flow peak velocity ( _VFW2 ), end-diastolic velocity, and time-averaged velocity. The time from systolic onset was also calculated. Calculations were made _for the systolic forward flow ( _TFW ) peak, reverse flow (T _BW ) peak, and diastolic forward flow ( _TFW2 ) peak. The diastolic flow variability index ( _FFId ) was then calculated as the relative ratio of diastolic velocity variability amplitude (i.e., _VFW2 - _VBW ) to the total velocity pulse amplitude according to the following equation (3):

In calculating FFI _d , if the waveform was biphasic and there was no diastolic forward flow, V _FW2 was considered to be 0 (Hashimoto et al. Hypertension 2010; 56:926-933.).

5. Statistical Analysis Data are presented as mean ± standard deviation (if normal distribution is assumed), median [interquartile range] for continuous variables (if asymmetric), or percentage for categorical variables. Asymmetric data were log-transformed to a normal distribution for subsequent statistical analysis.

Univariate linear correlations were assessed using Pearson's coefficient (r). Paired t-tests or repeated-measures analysis of variance were used to compare individual matched data for pressure, velocity, and time parameters. Correspondence for time parameters was assessed using Bland-Altman plot analysis. Unpaired data were compared using Student's t-test. Multivariate linear regression analysis was used to compare the independent correlations of various pressure and stiffness gradient parameters with _PFId and _FFId . _PFId and _FFId were adjusted for covariates including age, sex, height, BMI, MAP, heart rate, fasting glucose, LDL cholesterol, HDL cholesterol, eGFR, treatment with calcium channel blockers, renin-angiotensin system inhibitors, β-blockers, α-blockers, diuretics, aldosterone receptor blockers, antidiabetic agents, and antihyperlipidemic agents. The independent determinants of PFI _d and FFI _d were assessed in a stepwise procedure using PP _D /PP _A and PWV _FD /PWV _CF as well as the same covariates as above as candidates.

Mediation analyses were performed based on the results of linear regression to examine the hypothesized associations between amplitude or stiffness gradient, diastolic blood flow variability, and diastolic blood pressure variability when appropriate. More specifically, indirect effects, direct effects, and confidence intervals were estimated in a model minimally adjusted for age and sex by bootstrapping with 5,000 resamples using the PROCESS program. Mediation was considered established if the indirect effect was significantly different from zero. Mediation rates were assessed as the ratio of the indirect effect to the total effect.

All statistical analyses were performed using SPSS Statistics (version 25.0; IBM, Armonk, NY, USA). A P value of less than 0.05 (two-tailed test) was considered statistically significant.

6. Results (1) Patient characteristics
The clinical and hemodynamic characteristics of 592 patients (mean age, 55 ± 14 years) are shown in Table 1. The majority of patients (91%) were treated with antihypertensive drugs, so their blood pressure was well controlled (brachial blood pressure (131 ± 18/74 ± 12 mmHg)). Diabetes and hypercholesterolemia were observed in 24% and 43% of all patients. The median PWV _CF was 7.8 m/s, and the mean PWV _FD /PWV _CF (aorta-leg stiffness gradient) and mean PWV _CR /PWV _CF (aorta-brachial stiffness gradient) ratios were 1.2 ± 0.3 and 1.0 ± 0.2, respectively. The pulse pressure amplification ratio between the aorta and dorsalis pedis (PP _D /PP _A ) was 1.8 ± 0.3, and between the aorta and radial artery (PP _R /PP _A ) was 1.3 ± 0.2.

(2) Femoral artery diastolic pressure waveform The time from the onset of systole to the second inflection point (T _ES ) was 309 ± 3 ms (Fig. 5(a) and Table 1). A dipping dicrotic notch (i.e., trough) was present in only 92 patients, occurring 428 ± 50 ms from the onset of systole, but the time of occurrence was not different from the deflective dicrotic notch in the remaining 500 patients (436 ± 4 ms, P = NS (not statistically significant)). In 517 patients (87%), a dicrotic wave peak or shoulder (T _DW ) was identified 63 ± 4 ms after the dicrotic notch (T _DN ).

The diastolic waveforms of all patients were successfully fitted to a monoexponential decay curve (Fig. 5(a)), and the goodness of fit was satisfactory in terms of ^R2 (median, 0.99) and RMSE (0.9 mmHg). The median τ and asymptote were estimated to be 0.7 [interquartile range: 0.4-1.7] and 50 [14-64] mmHg, respectively.

Despite the excellent model fit, the residuals exhibited finite oscillatory fluctuations (Fig. 5(b)).
Residual waves were either triphasic (n = 492, 83%) or biphasic (n = 100, 17%) in pattern; i.e., an initial negative wave (R1) was always followed by a second positive wave (R2) and almost always by a third negative wave (R3) (Table 1). Absolute wave heights increased in ascending order of their appearance (|P _R1 | > |P _R2 | > |P _R3 |; P < 0.001). The diastolic pressure variability index (PFI _d ) was 4.9 [3.4-6.6]%. The R1 peak occurred slightly earlier than the dicrotic notch (difference: 49 ± 39 ms; P < 0.001), and the R2 peak occurred later than the dicrotic notch (difference: 86 ± 62 ms; P < 0.001), but they were closely correlated with each other (r = 0.58, r = 0.62; P < 0.001).

(3) Femoral Artery Diastolic Blood Flow Waveforms Velocity pulse waveforms were triphasic, with backward and forward diastolic flow in 549 patients (93% and Fig. 5(c)), whereas they were biphasic, lacking forward diastolic flow, in 43 patients (7%). The mean diastolic flow variability index ( _FFId ) was 34 ± 8% (Table 1). Bland-Altma histograms revealed that the time of the backward velocity peak (T _BW ) was well aligned with the time of the second inflection point of the original compression wave (T _ES, Fig. 6(a)). The mean and 2SD of the difference were -2 and 46 ms, respectively.

(4) Correlation between diastolic residual pressure and blood flow waveform
There was a close correlation between FFI _d and PFI _d (Fig. 6(b)). The time of peak diastolic velocity (T _FW2 ) corresponded well with the time of peak R2 (T _R2 , r = 0.68, P < 0.001), but the former preceded the latter by 71 ms on average (Fig. 6(c)).

(5) Determinants of Diastolic Blood Pressure and Blood Flow Variability Table 2 shows the multivariate-adjusted associations between individual stiffness and pressure measurements and diastolic blood flow and blood pressure variability. Even when adjusted for numerous covariates (including age, sex, MAP, heart rate, anthropometric parameters, biochemical parameters, and various medical therapies), higher PWV _CF and lower PWV _FD or PWV _CR were associated with smaller FFI _d . Accordingly, the PWV _FD /PWV _CF ratio and the PWV _CR /PWV _CF ratio (i.e., elastic-to-muscular arterial pressure gradient) had closer positive correlations with FFI _d . Similarly, larger pulse pressure amplification ratios (i.e., P _PD /P _PA , P _PF /P _PA , and P _PR /P _PA ) were associated with larger FFI _d . Of these, P _PD /P _PA was most closely correlated with FFI _d . Similar to FFI _d , PFI _d had significant independent associations with measures of pulse wave amplitude and stiffness gradient (Table 2). The results were essentially the same when PFI _d was replaced with the absolute amplitude of the residual pressure wave (PFA _d ).

The independent determinants of _FFId and _PPId were assessed using a stepwise linear regression model, which included the PWV _FD /PWV _CF ratio and the P _PD /P _PA ratio, as well as other potential covariates (Table 3). Even when considered simultaneously, both P _PD /P _PA and PWV _FD /PWV _CF were found to be independent determinants of _FFId . The combination of these two alone accounted for 67% of the total explanatory variance. Similar to _FFId , _PFId was independently determined by P _PD /P _PA and PWV _FD /PWV _CF , in addition to age and sex.

(6) Mediation Analysis Mediation analysis was performed to examine whether diastolic blood flow variability explained the correlation between arterial stiffness gradient and diastolic blood pressure variability (Figures 7(a) and (b)). Figure 7(a) shows the correlation between a higher PWV _FD /PWV _CF ratio and a larger PFI _d . This correlation was evaluated separately in terms of the direct and indirect effects mediated through a larger FFI _d . Mediation analysis revealed that the indirect effect was highly significant (P<0.001). FFI _d was estimated to mediate 54% (95% confidence interval: 30-100%) of the total observed effect of PWV _FD /PWV _CF on PFI _d . When _{the P PD} /P _PA ratio was evaluated as a predictor variable (Figure 7(b)), FFI _d was estimated to mediate 57% (42-80%) of the correlation between the P _PD /P _PA ratio and PFI _d .

30... Blood pressure index calculation device, 42... Residual pressure waveform generation unit, 43... Diastolic blood pressure variability index calculation unit, 44, 144... Pulse pressure amplification estimation unit, 45, 145... Central aortic blood pressure estimation unit, 46, 146... Vascular stiffness evaluation unit, 143... Diastolic blood flow variability index calculation unit,

Claims (9)

An apparatus for calculating a blood pressure index of a test subject,
a residual pressure waveform generator that fits the diastolic portion of the blood pressure waveform of the peripheral artery under test with an exponential decay curve to generate a residual pressure waveform having an initial negative peak and a subsequent positive peak;
a diastolic blood pressure variability index calculation unit that calculates a diastolic blood pressure variability index (PFI _d ) expressed by the following formula (1), where _PR1 is the pressure at the negative peak of the residual pressure waveform, _PR2 is the pressure at the positive peak, and PP _Peri is the pulse pressure of the peripheral artery;
A blood pressure index calculation device comprising: The blood pressure index calculation device of claim 1 further comprising a pulse pressure amplification estimation unit that estimates pulse pressure amplification based on the diastolic blood pressure variability index. The blood pressure index calculation device of claim 1 further comprises a central aortic blood pressure estimation unit that estimates central aortic blood pressure based on the diastolic blood pressure variability index. The blood pressure index calculation device according to claim 1 , further comprising a vascular stiffness evaluation unit that evaluates a stiffness of the blood vessels of the test subject based on the diastolic blood pressure variability index. A method for calculating a blood pressure index of a test subject, comprising:
a residual pressure waveform generating step of fitting a diastolic portion of the blood pressure waveform of the peripheral artery under test with an exponential decay curve to generate a residual pressure waveform having an initial negative peak and a subsequent positive peak;
a diastolic blood pressure variability index calculation step of calculating a diastolic blood pressure variability index (PFI d ) expressed by the following formula (1), where P _R1 is the pressure at the negative peak of the residual pressure waveform, P _R2 is the pressure at the positive peak, and PP _Peri is the pulse pressure _of the peripheral artery;
A blood pressure index calculation method including: The blood pressure index calculation method according to claim 5 , further comprising a pulse pressure amplification estimation step of estimating a pulse pressure amplification based on the diastolic blood pressure variability index. The blood pressure index calculation method according to claim 5 , further comprising a central aortic blood pressure estimation step of estimating a central aortic blood pressure based on the diastolic blood pressure variability index. A program for causing a computer to execute a process for calculating a blood pressure index of a test subject,
Computer,
a residual pressure waveform generating unit that generates a residual pressure waveform having a negative peak that appears first and a positive peak that appears next by fitting the diastolic portion of the blood pressure waveform with respect to the time change of the peripheral artery of the test subject to an exponential decay curve, and a diastolic blood pressure variability index calculating unit that calculates a diastolic blood pressure variability index (PFI _d ) expressed by the following formula ( ₁ ) when the pressure at the negative peak of the residual pressure waveform is P _{R1 , the pressure at the positive peak is P R2} , and the pulse pressure of the peripheral artery is PP _Peri
A blood pressure index calculation program that functions as a A computer-readable recording medium on which the program according to claim 8 is recorded.