JP3231846B2 - Blood pressure daily fluctuation approximation measuring device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a blood pressure measuring device for measuring blood pressure, pulse rate, etc. for, for example, 24 hours or more, and it is possible to examine the circadian rhythm of blood pressure using continuously measured blood pressure. The present invention relates to a blood pressure circadian variation approximation measuring device .

[0002]

2. Description of the Related Art Normally, a human has about 100,000 heart beats per day, and there are systolic and diastolic blood pressures corresponding to this number. In the case of the direct method of measuring the arterial pressure by inserting a catheter into the artery, all of these can be recorded, and both short-period fluctuation and long-period fluctuation can be examined. On the other hand, in the case of indirect measurement by the Korotkoff sound method or the oscillometric method, one measurement takes about several tens of seconds,
The shortest measurement interval requires more than 1 minute, and it is difficult to measure at an interval shorter than 15 minutes in an ordinary clinical setting in consideration of congestion and invasiveness. Moreover, clinical problems such as obstruction of sleep during the night are great, so that it is often necessary to measure at hourly intervals during sleep. Therefore, in the case of non-invasive blood pressure measurement, 24 hour blood pressure is generally 24 hours.
In many cases, the blood pressure fluctuation is determined based on the measured values of .about.48 points, but the burden on the measurement during daily activities is still large.

[0003] Further, when observing a large change in blood pressure, which is the basis of the human blood pressure, rather than a minute change in blood pressure due to an external factor, all of these measurement points are not necessarily required. It is undeniable that the measurement by the method is wasteful in terms of patient burden and cost performance.

[0004]

When examining the circadian variation in the blood pressure measurement by the conventional indirect method, the blood pressure measurement device is always worn on the body in consideration of the measurement frequency, even though the frequency is 24 to 48 points per day. I have to carry it. Thus very high restricted to the body, the weight is also heavy, and the like relatively larger in size, often affects the physical activity in everyday life. In addition, frequent cuff pressurization leaves subcutaneous bleeding and impairs sleep, and leaves problems in invasiveness.

[0005] The present invention has been made in view of such a problem that the conventional measuring device is restricted and invasive.
For example, information on these circadian rhythms is obtained using blood pressure values measured at six time points per day.

It is another object of the present invention to provide a blood pressure circadian variation approximation measuring device in which the weight and size of the device are reduced and the restraint and invasiveness of blood pressure measurement are greatly reduced.

[0007]

SUMMARY OF THE INVENTION An object of the present invention is to start from a time zone in which the blood pressure is most lowered, and from this time zone, approximately four times
Blood pressure measuring means for measuring the blood pressure for each time , and a blood pressure circadian variability approximation measuring device characterized by comprising a regression curve creating means for creating a regression curve synthesized from curves of different cycles based on the measured values Achieved.

[0008] Specifically, it is characterized in that several measurements are made at each of the above measurement points, and the average value is used.

When analyzing the data obtained by measuring the blood pressure and the pulse rate and displaying a fluctuation pattern of 24 hours, it is required
Periodic analysis is performed using data at intervals of 4 hours ± 15 minutes starting at hours ± 15 minutes.

The blood pressure and pulse rate measured over several days at intervals of 48 hours or more or 1 day or more
When the comparison is made in units of time and used for the analysis, the blood pressure and the pulse rate are compared at intervals of 4 hours ± 15 minutes starting at 3: 00 ± 15 minutes in the morning.

The reason why the measurement can be performed six times is that a regression curve created by synthesizing a curve having a cycle of 24 hours and a curve having a cycle of 12 hours using the data of the six measurements is used to measure the measured values. Due to the increase in the contribution to the measurement and the fact that the starting point of the measurement is at a specific time zone.

[0012]

EXAMPLE The following describes the basis obtained from the data at 6 time points of 4 hours ± 15 minutes starting at 3: 00 ± 15 minutes, giving the information necessary for the diagnosis of hypertension and the judgment of the therapeutic effect. Will be described.

The present inventors analyzed census curves of approximately 500 Japanese healthy subjects and approximately 1,000 hypertensive patients by analyzing the measured values of human 24-hour blood pressure (at approximately 30-minute intervals). In the case of a regression curve obtained by combining two periods of 24 hours and 12 hours, while the contribution rate of the regression curve obtained by fitting a cosine curve having a period of 24 hours to the actually measured value was 80% or less. Has a contribution rate of about 95-9
This is a marked improvement of 8%. In addition, when this fitting was performed, the relationship between the contribution rate to the number of measurement points and the relationship between the measurement time and the contribution rate were examined.

Here, the "contribution ratio" means the square of the correlation coefficient r (r ² ), where 0 ≦ r ² ≦ 1.

The contribution ratio is interpreted as indicating the ratio of the amount of information shared by the two data groups.

By summarizing the measured blood pressure data group for 24 hours into 24 points at regular intervals in time, the systolic and diastolic blood pressures are each 0.9% smaller than the original data group.
The contribution ratios of 74 and 0.982 were obtained, and the original data could be explained with high fidelity.

The number of collections of these data per 24 hours, that is, the number of blood pressure measurements, was set at 12, 8, 6, and 5 at equal intervals.
When the coincidence coefficient of the periodic function of the periodic circuit curve when the number of points was gradually decreased was examined for the phase and the amplitude, the results shown in Table 1 were obtained. When the number of measurement points is reduced to 5, the coefficient of coincidence of the amplitude of the periodic function is 0.961 to 0.657 when the coefficient is 6 points.
And deteriorated significantly. On the other hand, the phase coincidence coefficient kept a high value of 0.986.

Therefore, with respect to the number of measurement points, if the time is equal to or more than 6 at 24 hours, almost the same information as the data measured at 30 minutes can be obtained. As for the measurement time, 3:00 AM, 7:00 AM and 11: 0 as shown in FIG.
0AM, 3:00 PM, 7:00 PM, 11:00 PM
There are nadir (base value), peak, or steep changes of the regression curve in the vicinity, and almost all of the characteristics of the variation curve can be prompted by using the measurement points in the vicinity.

For the above reasons, it was found that the measurement was necessary and sufficient at six time points. In addition, it was found that it is effective to measure the blood pressure at the time when the blood pressure is most reduced. This time zone is usually around 3:00 am, the measurement is started with this time zone as a starting point, and measurement points are set at intervals of about 4 hours from there.

[0020]

[Table 1] Specifically, the blood pressure circadian variation, which is information necessary for the diagnosis of hypertension and the determination of the therapeutic effect, is changed at a time zone in which the blood pressure is most decreased, for example, at 3: 00 ± 15 minutes, and at intervals of, for example, 4 hours ± 15 minutes. For example, it is obtained by measuring blood pressure and pulse rate at six time points a day. That is, two cosine curves of 24 hours and 12 hours are fitted by cyclic regression analysis of the obtained blood pressure value and pulse rate at the 6 time points, and the level (median value) and pattern (by phase and amplitude) of the blood pressure change over 24 hours Stipulated).

Hereinafter, a specific description will be given with reference to the embodiments of the present invention.

FIG. 1 is a functional block diagram of a daily fluctuation approximation measuring apparatus according to an embodiment of the present invention. Reference numeral 1 denotes a blood pressure measurement unit having a blood pressure measurement function in a general non-invasive blood pressure monitor. 2
The operation unit accepts a patient's measurement start operation and the like from the operation switch 9 when measuring blood pressure during awakening. This switch is of a so-called momentary type, in which the contacts are closed only while a force is being applied. Reference numeral 3 denotes a display unit for displaying a measurement result and the like. A speaker 4 informs the patient that the measurement time has come by sound. Reference numeral 5 denotes a control unit which controls the operation of the entire apparatus, and performs an operation for obtaining a regression curve from the measurement results as needed. Reference numeral 6 denotes an input / output interface used for data exchange with the outside. This interface is capable of bidirectional communication such as RS-232. Reference numeral 7 denotes a clock which informs the controller of the time intermittently. This watch is an electronic watch. Reference numeral 8 denotes a data storage unit that stores measurement conditions, measurement results, patient information, and the like, and is a nonvolatile memory that retains its contents even when the power is turned off.

The operation of the apparatus is roughly classified into three processes S1, S2 and S3 as shown in FIG.
This is performed in the order of 2, S3. S1 is an initial setting, S2 is a long-time blood pressure measurement, and S3 is an output of data such as a measurement result to the outside. In S1 and S3, the operation is performed by a doctor, a nurse, or the like, and the patient needs to operate only when instructing the apparatus to recognize the measurement time at the measurement time and start the measurement at the time of awakening in S2. 9 is 1
A single push button switch is sufficient. When the patient is awake, knowing the measurement time point by the promotion sound, he once closes the operation switch 9 to prepare for blood pressure measurement, and when ready, closes the switch 9 to start measurement. During sleep, blood pressure measurement is performed completely automatically if measurement preparations are made in advance. In this case, no promotion sound is generated so as not to hinder sleep.

Each of S1, S2, and S3 will be described in detail below.

First, the initial setting process S1 will be described below with reference to FIGS. FIG. 3 is a diagram showing paths of data or signals at the time of initial setting. FIG. 8A shows S1.
FIG. 8B is a diagram showing the types and order of data or signals received from the outside at the time of initial setting. First, an initialization command shown in FIG. 8B is externally transmitted via the path A1 in FIG.
Is transferred to the control unit 5 through In step S101, the completion of the reception of this command is waited. Then, in step S102, the contents of the data storage unit 8 are erased via the path A3. Subsequently, the data of the patient information / measurement condition is sent from the outside in the same manner, the completion of reception is waited for in step S103, and the patient information / measurement condition is stored in the data storage unit 8 via the route A3 in step S104. Subsequently, the initial data for the time adjustment of the clock is received in step S105, and the time adjustment of the clock is executed via the path A4 in step S106 using this initial data. During this time, the operation state may be displayed on the display unit 3 via the path A5 as needed.

Next, the details of the long time measurement processing S2 will be described below with reference to FIGS. FIG. 4 shows a data or signal path when measuring blood pressure during awakening, and FIG. 5 shows a data or signal path when measuring blood pressure during sleep. FIG. 9 is a flowchart of S2.

In step S201, the time information obtained from the clock 7 intermittently via the route B1 in FIG. 4 to the route C1 in FIG. The measurement condition is read from the data storage unit 8 via the paths B2 to C2 in step S202 recognized as the measurement time point, and it is determined whether the measurement time point is the setting at the time of awakening or the setting at the time of sleep. In S206, in the case of sleep, the process proceeds to S203.

In the case of awakening, the route B is determined in step S206.
The sound of the speaker 4 is made to sound as the sound for promoting the blood pressure measurement. When the patient recognizes the prompting sound, the patient closes the operation switch 9 to stop the prompting sound and sends a recognition signal to the control unit 5 via the operation unit 2 via the path B4. When this is detected in step S207, the sound of the speaker 4 is stopped in step S208. Next, in step S209, the process waits until a signal for starting the measurement is given via the path B4 by the operation of the switch 9 of the patient, and if the start operation is not performed even after 10 minutes elapse in the step S201, the processing S2 for generating the promotion sound again.
Return to 06. When the start of measurement is recognized, the process proceeds to step S203. In the following, substantially the same processing is performed during awakening and during sleep.

In step S203, the blood pressure measurement unit 1 is caused to perform automatic blood pressure measurement via the routes B to C3, and a blood pressure measurement result is obtained via the routes B6 to C4. Processing S204, S
According to 205, automatic measurement is performed three times at three minute intervals. When awake, the measurement result may be displayed on the display unit 3 via the route B7. When the three measurements are completed, the result is stored in the data storage unit 8 via the paths B8 to C5 in step S211. Next, in step S212, it is determined whether or not the blood pressure measurement at the determined measurement point has been completed. If the measurement has been completed, the entire operation of S2 ends, otherwise, the process in step S20
Return to step 1 and wait until the next measurement.

Next, details of the measurement data transfer processing S3 will be described below with reference to FIGS.

FIG. 6 shows data or signal paths when blood pressure measurement data and the like are transferred to the outside. FIG. 10 is a flowchart of S3. First, the apparatus waits for a command for data transfer from the outside in step S301. In this state, a data transfer command is sent from the outside via the path D1 and received by the control unit 5 via the path D2. When the command is recognized, the measurement data, the patient information, and the like stored in the data storage unit 8 via the path D4 are read out in step S302.

These pieces of information are stored in the route D
5 and transferred to the outside via D6. Next, a regression curve is obtained in step S304. An example of specific means for the contents of the processing S304 will be described below.

A regression curve is determined by the least squares method using Fourier analysis. Regression curve y to be found
When (t) is expressed by Fourier series expansion, generally,

[0034]

(Equation 1) It is expressed as Here, t represents time, and T / k represents a cycle. Here, assuming that up to k = 2 and an error represented by k ≧ 3 is ε, the equation (1) is as follows.

[0035]

(Equation 2) It can be expressed as.

Here, β ₀ = a _0/2 , β ₁ = a ₁ , β
₂ = b ₁ , β ₃ = a ₂ , β ₄ = b ₂ , x ₀ = 1, x ₁ (t) = cos (2πt / T), x ₂ (t) = si
n (2πt / T) x ₃ (t) = cos (4πt / T), x ₄ (t) = si
n (4πt / T), Equation (2) becomes

[0037]

(Equation 3) Where y (t) is the target variable and ｛x ₁ (t), x ₂ (t), x ₃ (t), x ₄
(T) corresponds to a multiple regression model with｝ as an explanatory variable, and β corresponds to a partial regression coefficient. When equation (3) is applied to p (≧ 4) measurement points, using vectors and matrices (hereinafter, vectors and matrices corresponding to scalar values in lower case are expressed in upper case),

[0038]

(Equation 4) However,

[0039]

(Equation 5) Can be expressed as Here, y represents the measured values at p measurement points, and ε also represents the error. From equation (4), generally, the least-squares solution that minimizes ^t E · E →

[0040]

(Equation 6) It is known to be given as In the present invention, the number of measurement points p is 6, and each element of the matrix X of 6 rows and 5 columns is T = 24 hours, t ₁ = 3, t ₂ = 7, t ₃ = 11,
Assuming that t ₄ = 15, t ₅ = 19, and t ₆ = 23 (hour),

[0041]

(Equation 7) It can be determined more easily. In this way, X is obtained as follows.

[0042]

(Equation 8) Therefore, ^t XX

[0043]

(Equation 9) And this inverse matrix is

[0044]

(Equation 10) Can be calculated.

Therefore, when these are applied to the equation (6),

[0046]

[Equation 11] Β can be easily obtained by giving a measured value to y, that is, a ₀ , a ₁ , a ₂ , b ₁ , b ₂ in the equation (2).

Calculation is actually executed for the following two examples as the measured value y.

[0048]

(Equation 12) For y _1,

[0049]

(Equation 13) That is, a / 2 = 120, a ₁ = 0, a ₂ = 0, b ₁ =
0, b ₂ = 0, resulting in only a DC component. For y _2,

[0050]

[Equation 14]That is, a₀ / 2 = 119.2, a₁ = -11.47, b
₁ = −7.387, a _Two = -2.887, b_{twenty two}= -6.
667.

The expression (1) can be expressed only by a cosine function. That is,

[0052]

(Equation 15) _{Here, A 0 = a 0/2} ( Mesa: mesor), A _{k
^{= √a k 2 + b k 2}} ( ^{_{amplitude: amplitude) θ k = tan -1}} (b k / a k) ( phase angle: Acroph
case) ω _k = 2πk / T (angular frequency).

A ₀ , a for y ₂ obtained in the above example
_1, application of a _2, ab _1, b ₂ a _{(1) ', A 0 =} a 0/2, A 1 = √a 1 2 + b 1 2 ≒ 13.6 A 2
= √a ₂ ² + b ₂ ² ２７7.27 θ ₁ = tan ^-1 (b ₁ / a ₁ ) = tan ^-1 (-7.387 / -11.47) ≒ 0.572 + n
π (n = 0,1,2, ...) θ ₂ = tan ^-1 (b ₂ / a ₂ ) = tan ^-1 (-6.667 / -2.887) ≒ 1.162 + n
π (n = 0, 1, 2,...) w ₁ = π / 12 (rad / hour), w ₂ = π / 6 (rad
/ H) θ ₁ and θ ₂ are n = 1 in consideration of the signs of a ₁ , b ₁ , a ₂ and b ₂ , and θ ₁ ≒ 3.714 (rad), θ ₂ ≒
4.304 (rad) are obtained. Therefore, the regression by y ₂ curve y ₂ (t) is

[0054]

(Equation 16) Becomes (The unit of t is hour and the unit of angle is rad.) A ₀ (meser) represents the average of the blood pressure fluctuation over 24 hours, and corresponds to “level” described herein. θ ₁ and θ ₂ similarly correspond to “phase”, and A ₁ and A ₂ correspond to “amplitude”.

From equation (10), t = 3, 7, 11, 1
Table ₂ shows y ₂ (t) obtained for 5, 19, and 23 and a comparison with the original measured value.

[0056]

[Table 2] It can be seen that each of them has an error of less than 1%. Table 1
Considering the contents of the above, it can be seen that there is no inferiority to data having more measurement points.

The data of the regression curve thus obtained is transferred to the outside in step S305.

[0058]

Since the present invention is constructed as described above, the time period when blood pressure is most reduced, for example, only 6 time points of 4 hours ± 15 minutes at 3 ± 15 minutes in a day, starting at 3 ± 15 minutes. By using the measured value, there is no need to always wear the blood pressure measurement device on the body except when the patient is sleeping,
In addition, even during normal sleep, only measurement at one or two time points is performed, so that restraint and invasiveness can be significantly reduced as described below.

(1) Since it is not necessary to keep the device worn on the body, there is no influence on the operation of daily life.

(2) Since the frequency of pressurizing the cuff is low, sleep disorders especially at night are extremely small, and subcutaneous hemorrhage hardly occurs.

(3) Since the time during which the cuff is worn is limited to during sleep, it is unlikely to cause scuffing and rash.

(4) Since the energy consumption of the apparatus can be reduced, the size of the battery of the power supply can be reduced, so that the apparatus can be reduced in size and weight, and the burden on the portable device can be significantly reduced. This is particularly beneficial for elderly, female and pediatric patients with poor physical fitness.

(5) Since the burden on the patient is small, the measurement can be continued for many consecutive days, or the daily fluctuation can be measured several times in order to confirm the effects of various therapeutic agents for hypertension.

(6) Since the measurement can be easily performed,
A large population such as schools and local residents can be screened for hypertension, which greatly contributes to early detection of hypertension.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of a measurement device that performs measurement according to a daily fluctuation measurement method according to an embodiment of the present invention.

FIG. 2 is a diagram showing an average of actually measured values at each time of systolic and diastolic blood pressures and a periodic regression curve.

FIG. 3 is a diagram showing paths of data or signals at the time of initialization in the functional block diagram of FIG. 1;

FIG. 4 is a diagram showing a data flow path at the time of pressure measurement at the time of awakening in the functional block diagram of FIG. 1;

FIG. 5 is a diagram showing a data flow path at the time of measuring blood pressure during sleep in the functional block diagram of FIG. 1;

6 is a diagram showing a data flow path at the time of transferring blood pressure data in the functional block diagram of FIG. 1;

FIG. 7 is a diagram showing basic processing executed by the device according to the embodiment.

FIG. 8 is a flowchart showing details of an initial setting process of FIG. 7;

FIG. 9 is a flowchart showing details of a long-time measurement process of FIG. 7;

FIG. 10 is a flowchart illustrating details of a measurement data transfer process of FIG. 7;

────────────────────────────────────────────────── ─── Continuation of the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) A61B 5/00-5/0295 JICST file (JOIS)

Claims (3)

(57) [Claims] 1. Starting from a time zone in which blood pressure is most reduced,
Blood pressure measuring means for measuring blood pressure approximately every four hours from this time zone; and regression curve creating means for creating a regression curve synthesized from curves of different periods based on the measured values. Daily fluctuation approximation measuring device. 2. The time zone in which the blood pressure is most decreased,
2. The blood pressure circadian variability approximate measurement apparatus according to claim 1, wherein the apparatus is used around 3:00 am. Wherein the different periods, blood pressure circadian variation approximation measuring device according to claim 1 or 2, characterized in that it is 24 hours and 12 hours.