EP0183829A4

EP0183829A4 - Techniques for obtaining waveform information associated with an individual's blood pressure.

Info

Publication number: EP0183829A4
Application number: EP19850903157
Authority: EP
Inventors: William Trevor Link
Original assignee: Norse Instruments Inc
Current assignee: American Hospital Supply Corp
Priority date: 1984-06-19
Filing date: 1985-06-17
Publication date: 1988-01-26
Also published as: WO1986000210A1; JPS62500077A; EP0183829A1

Abstract

Techniques for determining different parameters associated with an individual's blood pressure in a non-invasive manner. These techniques include generating a blood pressure waveform corresponding to the individual's actual waveform (20) whereby the mean blood pressure of the individual can be readily calculated.

Description

TECHNIQUES FOR OBTAINING WAVEFORM INFORMATION ASSOCIATED WITH AN INDIVIDUAL'S BLOOD PRESSURE

The present invention relates generally to blood pres- sure evaluation procedures and more particularly to noninvasive techniques for determining certain waveform information associated with blood pressure.

The most reliable ways presently known for obtaining information relating to an individual's blood pressure require invasive procedures. Such procedures are not carried out routinely but only under extreme circum¬ stances, for example during heart surgery. Under less critical conditions, blood pressure information including specifically an individual's systolic (maximum) and diastolic (minimum) blood pressures is obtained non-invasively. There are two well known non-invasive techniques presently being used today, one is commonly referred to as auscultation and the other is based on oscillometry. Both of these non-invasive techniques use the standard arm cuff which most people are familiar with. However, in the auscultatory method, the systolic and diastolic pressures are determined by listening to certain sounds (Korotkoff sounds) which occur as a result of the cuff first being pressurized and then depressurized whereas oscillometry actually measures changes in pressure in the cuff as a result of changes in blood pressure as the cuff is first pressurized and then depressurized. As will be seen hereinafter, the various embodiments of the present invention are based on oscillometry. In order to more fully appreciate these embodiments, reference is made to applicant's own United States Patent 3,903,872 (the Link patent) for obtaining blood pressure information non-invasively. This patent which is incorporated herein by reference describes, among other things, a way of obtaining the diastolic pressure of an individual in accordance with a technique which will be discussed in more detail hereinafter. In United States Patents 4,009,709 and 4,074,711 (Link et al) which are also incorporated herein by reference, non-invasive techniques using oscillometry are disclosed for obtaining the systolic pressure of an individual. These techniques will also be discussed hereinafter.

While the various procedures described in the Link and Link et al patents just recited and other patents held by applicant are satisfactory for their intended purposes, it is an object of the present invention to provide additional uncomplicated and yet reliable techniques for obtaining different types of informa¬ tion relating to an individual's blood pressure.

A more specific object of the present invention is to provide a different uncomplicated and yet reliable technique for generating non-invasively a waveform closely approximating an individual's true blood pres¬ sure waveform which, heretofore, has been obtainable by invasive means only.

Another particular object of the present invention is to provide a new way for measuring and calculating the mean arterial pressure of an individual. As will be described in more detail hereinafter, the objects just recited are achieved by means of oscillometry. In accordance with this technique, a suitably sized cuff, for example one which is 20 inches, long and 5 inches wide, is positioned around the upper arm of an individual, a human being specif¬ ically or a mammal in general (hereinafter referred to as the patient) and initially pressurized to a level which is believed to be clearly greater than the pa- tient's systolic pressure, for example 180 Torr. It is assumed that this pressure will also cause the patient's artery within the sleeve to completely col¬ lapse. Thereafter, cuff pressure is gradually reduced toward zero during which time the cuff continuously changes in pressure in an oscillating fashion due to the combination of (1) the internal blood pressure changes in the patient's artery and (2) changes in cuff pressure. The latter at any given time in the procedure is known and oscillaroty changes in cuff pressure can be readily measured, for example with an oscilloscope. By using these two parameters in conjunction with information which may be made available from methods disclosed in the above-recited United States patents it is possible to achieve the foregoing objectives in an uncomplicated and reliable way utilizing the techniques of the present invention to be described hereinafter.

In this regard, it should be noted at the outset that the typically 5" wide pressure cuff entirely surrounds a corresponding 5" length of artery. The tissue of the arm is for the most part incompressible, and therefore any change in the volume of the artery, caused for example by pulsations of blood, results in a corresponding change in the volume of air in the air bladder which is within the cuff and therefore adjacent to the arm. This change in air volume produces a small but accurately measurable pressure change in the air. This equivalence of pressure pulsations in the cuff bladder to volume pulsations of the artery is the essence of oscillometry.

In order to more fully appreciate the various tech¬ niques of the present invention, the following more detailed background information is provided in con¬ junction with Figures 1-5 of the drawings where:

FIGURE 1 (corresponding to Figure 6 in United States Patent 3,903,872) diagram atically illustrates the shapes of successive cuff pressure versus time pulses (cuff pulses) as the measured cuff pressure changes from 90 Torr to 80 Torr to 70 Torr, assuming the patient has a diastolic pressure of 80 Torr;

FIGURE 1A diagrammatically illustrates a full series of cuff pulses corresponding to those in Figure 1 from a cuff pressure of 160 Torr to a cuff pressure of zero;

FIGURE 2 diagrammatically illustrates a curve corre- sponding to arterial or cuff volume (V) , that is, the volume of the patient's artery within the cuff (as measured by cuff volume) versus wall pressure (P ) across the artery wall within the cuff and, super¬ imposed on this curve, a curve which is intended to correspond to the actual blood pressure waveform of a patient, the two curves being provided together in order to illustrate the principles of oscillometry, as relied upon in the above-recited patents;

FIGURES 3 and 4 diagrammatically illustrate the cuff curve of Figure 1 in ways which display techniques for obtaining a given patient's systolic and diastolic blood pressures in accordance with the Link and Link et al patents recited above; and

FIGURE 5 diagrammatically illustrates a compliance curve for the patient's artery, that is, a curve which displays the ratio ΔV/ΔP against the arterial wall pressure P , where ΔV is the incremental change in the arterial volume correspongind to a preselected constant change in bloodpressure ΔP. This curve is initially determined in order to 'provide the cuff or arterial volume curve (V/P curve) of Figure 2 by means of integration, as will be seen.

Turning first to Figure 1, this figure diagrammatical¬ ly illustrates three successive waveforms lOh, lOi and 10j which correspond to the change in volume in a pressurized cuff, as described above, at three differ¬ ent cuff pressures, specifically cuff pressures of 90 Torr, 80 Torr and 70 Torr. In actual practice, a greater number of waveforms (hereinafter referred to as cuff pulses) are generated starting at a cuff pressure of 160 Torr and ending at a cuff pressure of zero, as will be seen in Figure 1A. By generating these waveforms at known cuff pressures, both the diastolic and systolic pressures of a patient can be determined in accordance with the above-recited patents. While this will be explained in more detail below, it is important to note initially that each waveform has what may be referred to as a systolic rise S at one end of the waveform, a diastolic decline D, at the opposite end and a maximum amplitude A.

While the systolic rise S is fairly consistent and distinctive from one cuff pulse 10 to another, both the diastolic decline D, and amplitude A vary from pulse to pulse for reasons to be explained hereinafter. It is because of these variations that the techniques disclosed in the Link and Link et al patents recited above are able to determine the dia¬ stolic and systolic pressures. Specifically, as will be seen, when the diastolic pressure of a patient is equal to the cuff pressure, the cuff pulse generated has a diastolic decline which is greater in slope than the diastolic decline of any of the other cuff pulses. Thus, assuming that the diastolic decline has a maximum slope at the cuff pulse lOi illustrated in Figure 1, the patient providing these waveforms would have a diastolic pressure of 80 Torr. At the same time, this patient's systolic pressure can be de¬ termined by first finding which of the cuff pulses displays a maximum amplitude A and then, moving up in cuff pressure, finding the cuff pulse having half that amplitude. The cuff pressure responsible for produc¬ ing this half-amplitude pulse will equal the patient's systolic blood pressure. In order to more fully understand these capabilities, reference is made to Figures 2-5 in conjunction with the above-recited Link and Link et al patents.

Turning now to Figure 2, attention is directed to the curves illustrated there in order to explain why the cuff pulses of Figure 1 result from changes in cuff pressure. The generally S-shaped curve 12 illustrated is shown within a horizontal/vertical coordinate system where the horizontal axis represents the wall pressure P across the artery wall of a given patient, within the confines of the applied cuff, and the vertical axis represents arterial volume V of the artery within the cuff, as measured by the internal volume of the cuff itself. In order to fully under¬ stand this V/P curve (hereinafter merely referred to as an arterial or a cuff curve) , it is important to keep in mind the definition of Pw. The wall pressure P of the artery of the patient at any given time is equal to the blood pressure P, of the patient within the artery at that time less the applied pressure of the cuff Pc. Thus:

For purposes of the present discussion, it will be assumed that pressure is measured in Torr (m Hg) and that the section of the horizontal axis to the right of the vertical axis represents positive wall pres- sures while the section of the axis to the left of the vertical axis represents negative wall pressures. As a result, when no pressure is applied to the cuff

(e.g. , Pw at any given point in time is equal to the blood pressure of the patient at that time. As the cuff is pressurized, Pw decreases (moves to the left along the horizontal axis) . When the cuff pressure P is equal to the blood pressure P. at any given point in time, P at that time is equal to zero

(e.g. at the vertical axis) . As the cuff pressure is increased beyond the blood pressure at any point in time, Pw at that time becomes more negative (moves further to the left on the horizontal axis) .

With the definitions of the vertical axis V and the horizontal axis Pw in mind,' attention is now directed to an interpretation of the generally S-shaped cuff curve 12 within this coordinate system. For the moment, it is being assumed that this curve is charac¬ teristic of the particular patient being evaluated. That is, it is being assumed that the patient's artery within the cuff and therefore the cuff itself will change in volume along the S-shaped curve and only along the curve with changes in P . Hereinafter, with regard to Figure 3, it will be shown that the arterial curve 12 of a given patient can be generated from his cuff pulses 10 and corresponding cuff pressures P . Thus, for the time being, it will be assumed that the arterial curve illustrated in Figure 2 corresponds to that of the given patient.

With the foregoing in mind, the arterial curve of

Figure 2 will now be examined. Let it first be assumed that no pressure is applied to the patient's cuff so that Pc equals zero. As a result, Pw equals the blood pressure P. of the patient. In this regard, it is important to note that P. varies with time between the patient's diastolic blood pressure _h(D) and his systolic blood pressure P, (S). For purposes of this discussion, let it be assumed that these values are known and that specifically the patient's diastolic blood pressure is 80 Torr and his systolic blood pressure is 120 Torr. Thus, with no pressure in the cuff,' Pw oscillates back and forth with time between P_b(D) and P, (S), that is, between 80 Torr and 120 Torr. This 40 Torr measuring band is illustrated by dotted lines in Figure 2 at 14 and actually repre¬ sents the patient's pulse pressure ΔP which is equal to 40 Torr in this case.

The patient's actual blood pressure waveform 15 is superimposed on the V/P coordinate system in Figure 2 within the pulse pressure band 14. As seen there, this waveform is made up of a series of actual blood pressure pulses 16, each of which corresponds to a single beat of the patient's heart. Note that each pulse starts at a minimum pressure (the diastolic pressure of the patient) and sharply increases along its leading edge which is the systolic rise S until it reaches a maximum (the patient's systolic blood pressure) , at which time it drops back down along a trailing edge which includes a dichrotic notch and a diastolic decline D, to the minimum pressure again. At those points in time when the patient's blood pressure is at a minimum (that is, at the diastolic ends of pulses 16), the volume of the patient's artery and therefore the volume of the cuff is fixed by the arterial curve at the value indicated at V 1, .

On the other hand, whenever the patient's blood pressure is maximum (at the systolic end of each blood pressure pulse 16) , the arterial curve fixes arterial and therefore cuff volume at the slightly higher value indicated at V (P =120). Therefore, it should be apparent that for each heart beat, assuming a cuff pressure P of zero, the volume V (the cuff volume) moves between the values V, and V_, thereby generating a series of cuff pulses lOq corresponding to those illustrated in Figure 1 but at a cuff pressure P =0, as shown in Figure 1A. Thus, as the patient's blood pressure rises from a minimum to a maximum, the volume of the artery rises from V.. to V₂ in a generally corresponding manner and as the patient's blood pressure drops back down to a minimum, the arterial volume falls from V- to V₁ in a generally corre¬ sponding manner. Thus, each of the arterial pulses 10 in Figure 2 has a systolic rise S and a diastolic decline D, corresponding to the systolic rise and diastolic decline of each blood pressure pulse 16.

Having shown how the cuff pulses lOq are dependent upon the volume curve at a cuff pressure of zero, we will now describe how the arterial curve causes these arterial pulses to change with applied cuff pressure. Let us assume now a cuff pressure of 50 Torr. Under these conditions, Pw oscillates back and forth between

30 Torr and 70 Torr. The 30 Torr value is determined by subtracting the cuff pressure P of 50 Torr from the diastolic blood pressure P_b(D) of 80 Torr and the 70 Torr value is determined by subtracting the same P of 50 Torr from the systolic blood pressure^" P_b(°) °f 120 Torr. Thus, the entire 40 Torr band has merely been shifted to the left an amount equal to 50 Torr as indicated by the band 14'. Under these circumstances,

P oscillates back and forth along a steeper segment of the arterial curve so as to cause the volume of the patient's artery and therefore the volume of the cuff to oscillate between the values V, and V.. This results in the production of arterial pulses 101 at a

Pc of 50 Torr. Note that the amplitude of each cuff pulse LOl is greater than the amplitude of each cuff pulse lOq. This is because the 40 Torr band 14' at a cuff pressure- of 50 Torr is on a steeper part of the volume slope than the band 14 at a cuff pressure of zero. Indeed, as we increase the cuff pressure P (which decreases Pw) and therefore move the pressure band to the left on the horizontal axis, we first continue to move along steeper sections of the arte¬ rial curve and thereafter less steep sections. There¬ fore, the amplitude A (see Figures 1 and 1A) of the corresponding cuff pulses 10Q, 10L and so on will first increase to a maximum and then decrease again. At a cuff pressure P of 100, the entire 40 Torr pressure band is shifted to the left so as to uniform¬ ly straddle opposite sides of the vertical axis, as indicated at 14". This results in a corresponding cuff pulse lOg having approximately a maximum ampli¬ tude (ΔVmax in Figure 2) .

Moving still further to the left, at for example, a cuff pressure P of 160 Torr, the entire 40 Torr band is moved a substantial distance to the left of the vertical axis, as indicated at 14*'' such that the resultant change in volume (amplitude of the corre¬ sponding cuff pulse 10a) is quite small. By increas¬ ing the cuff pressure to even a greater amount, the band is moved still further to the left, eventually producing very small changes in volume V. From a II

physical standpoint, this represents a collapsed artery. In other words, sufficient cuff pressure P is being applied over and above the internal blood pressure P. to cause the wall of the artery to col¬ lapse. At the other extreme, that is, when the cuff pressure P is zero, there are no external constraints placed on the artery and the latter is free to fluctu¬ ate back and forth based on its internal pressure P. only. Between these extremes, the amplitude A of cuff pulse 10 (e.g. ΔV) will increase to a maximum and then decrease again, as stated. It is this characteristic of the volume curve which is used to determine the patient's systolic pressure in accordance with the previously recited Link et al patents, as will be described with regard to Figures 3 and 4.

As previously mentioned, it should be noted that a blood pressure increase causes an arterial volume increase. This arterial volume increase causes a cuff bladder air volume decrease which in turn causes a cuff bladder air-pressure increase. Therefore a blood pressure increase results in a cuff air pressure increase. This is emphasized as follows:

blood arterial ^■*^■ cuff air cuff air pressure volume volume pressure increase increase decrease increase

Thus blood cuff air pressure pressure increase increase

Referring to Figure 3, the same arterial curve 12 illustrated in Figure 2 is again shown but with a single superimposed pressure band 14'' ' ' at a cuff pressure P of 120 Torr. Assume again that the diastolic pressure of the patient is 80 Torr and his systolic pressure is 120 which means that P is equal to the patient's systolic pressure. Under these circumstances,' Pw oscillates back and forth within band 14^{•' •' •'}' between wall pressures of -40 Torr and zero, as shown. This results in a change in arterial volume ΔV (e.g., the amplitude A of a corresponding cuff pulse) which is approximately equal to one-half of the maximum change in arterial volume (e.g., max cuff pulse amplitude) . It may be recalled that a maximum change in volume ΔV max (and therefore a maximum cuff pulse amplitude Amax) results from a cuff pressure P of about 100 Torr (e.g. the pressure band 14" in Figure 2) . Thus, when the cuff pressure P is equal to the patient's systolic blood pressure P, (S), the amplitude A of the resultant cuff pulse 10 is about one-half of the amplitude of the cuff pulse having a maximum amplitude. Therefore, a patient's systolic blood pressure can be determined by first generating a series of cuff pulses across the cuff pressure spectrum, as in Figure 1A. From these pulses, the one having maximum amplitude Amax is determined and then the cuff pulse having half that amplitude (at a greater cuff pressure) is found. The cuff pressure P used to generate that pulse corre¬ sponds to the patient's systolic pressure. In other words, by evaluating the amplitudes of the various cuff pulses, the one corresponding to the band 14'''' illustrated in Figure 3 can be found. Once that pulse is found, its associated cuff pressure is assumed to be equal to the patient's systolic pressure. This is discussed in more detail in Link et al United States Patents 4,009,709 and 4,074,711 and means are provided in these latter patents for electronically making these evaluations. Returning to Figure 2, it should be noted that the actual blood pressure waveform 15 is shown having a uniform repetition rate, for example 60 pulses/minute, and that each blood pressure pulse 16 making up this waveform is identical to the next one. Both of these aspects of the waveform are assumed for purposes herein. Moreover, each pulse has its own systolic rise S and diastolic decline D,, as mentioned hereto¬ fore. It should also be noted that the arterial curve 12 dictates the relationship between V and P at each and every point on the waveform 15 of individual blood pressure pulse 16, not merely at the extreme diastolic and systolic end points of each pulse. Thus, one could measure the change in volume ΔV at two different cuff pressures along the diastolic decline only. In this case, the measuring band (e.g. the pressure difference between the two measuring points) is substantially narrower than band 14. As best illus¬ trated in Figure 4, ΔV., ' is determined for a cuff pressure P of zero using the pressure band 18 which encompasses a small part of the diastolic decline of each blood pressure pulse 16. ΔV-' is determined for a cuff pressure of P of 50 Torr by shifting the band to 18' and, ΔV,' is determined for a cuff pressure P of 80 Torr (e.g. the patient's diastolic blood pres¬ sure) by shifting the band to 18". Note that ΔV is maximum when the cuff pressure P is equal to the patient's diastolic blood pressure. Therefore, by determining the change in volume ΔV at the end of the diastolic slope of the patient's actual blood pressure waveform for each and every cuff pressure, the one cuff pressure producing a maximum change will corre¬ spond to the patient's diastolic blood pressure. The lowest pressure part of the diastolic decline D, forming part of each pulse 16 is particularly suitable for this purpose since it can be readily located during each cycle of the waveform. This is because it immediately precedes the systolic rise S which is readily distinguishable each time it appears. This procedure is described in more detail in the previous¬ ly recited Link Patent 3,903,872 along with means for carrying out this procedure electronically.

The foregoing discussions for obtaining a given pa¬ tient's systolic and diastolic blood pressures have assumed that the patient's arterial curve corresponded to the one illustrated in Figures 2, 3 and 4. While this assumption is reasonably valid, it is possible to determine the patient's own volume ^" curve using the principles associated with Figure 4. Specifically, using the narrower bands 18, 18' and so on as measuring bands, the change in volume ΔV (e.g., the change in cuff volume) resulting from different cuff pressures P is plotted, as shown in Figure 5. Thus at a cuff pressure P of zero, there is a relatively small change in volume ΔV, as evidenced by the small ΔV. ' in Figure 4. As the cuff pressure P increases, the change in volume ΔV continues to increase to a maximum (ΔV ' in Figure 4) and then decreases. In mathematical terms, this curve represents incremental changes in volume with incremental changes in pressure or dV/dP (Figure 5) . By integrating this curve we obtain the cuff curve or the V/P curve of Figures 2-4.

Having discussed Figures 1-5 in regards to the prior art techniques for obtaining diastolic and systolic blood pressures for a given patient in accordance with the techniques described in the above-recited Link and Link et al patents, attention is now directed to the various aspects of the present invention, as discussed briefly above, in conjunction with remaining Figures 6-9 where: FIGURE 6 diagrammatically illustrates an actual blood pressure pulse for a given patient;

FIGURE 7 diagrammatically illustrates a plotted waveform which approximates the actual blood pressure pulse of Figure 6 and which is generated non-invasively in accordance with the present invention;

FIGURE 8 diagrammatically illustrates an S-shaped cuff or arterial (PV) curve similar to the one illustrated in Figures 2-4 but exaggerated along the vertical slope with enlarged portions of the diastolic decline forming part of an actual blood pressure waveform superimposed thereon:

FIGURES 9 (a)-(d) diagrammatically illustrate four blood pressure waveforms haing different blood pressure constants; and

FIGURE 10 schematically illustrates an arrangement for providing a curve which closely approximates a patients actual blood pressure waveform and also provides the patients mean pressure and blood pressure constant.

Turning first to Figures 6-8, attention is now direct¬ ed to a technique provided in accordance with the present invention for generating a waveform which closely approximates an individual patient's actual blood pressure waveform (for example, the blood pressure pulse 16 in Figure 2) without using an invasive device. For purposes of illustration, the assumed actual blood pressure pulse is shown graph- ically in Figure 6 at 20 where the patient's blood pressure (the vertical axis) is measured against time

(the horizontal axis) . From previous discussions and a knowledge of the subject matter of Link Patent 3,903,872 and Link et al Patents 4,009,709 and 4,074,711, the individual patient's diastolic and systolic pressures _b(D) and P_fa(S) can be first deter- mined, thereby providing the minimum and maximum vertical points on the pulse as seen in Figure 6. At the same time, since the systolic rise (leading edge) of the actual pulse is readily detectable, its begin¬ ning and end points timewise (t and t ') are also readily detectable. Therefore, using the same coordi¬ nate system in Figure 7 we will now describe a method for non-invasively determining the important features of the blood pressure pulse shown in Figure 6 and thus providing a useful approximation of the whole pulse. As appears in Figure 6, the horizontal (time) axis can be made to represent the patient's diastolic blood pressure with t being provided at the vertical axis, indicating the beginning of the waveform and t ' a second point indicating the end of that waveform and the beginning of the next one. A dotted horizontal line L- can then be drawn above the time axis at the patient's systolic blood pressure. A single point on this horizontal line can then be established as to"

This corresponds to the time, which is readily appar- ent from the measured cuff pressure pulses, at which the peak systolic pressure P_fa(s) is reached. See for example the cuff pulses of Figure 1. The peak ampli¬ tude of each of these pulses corresponds to P_h(s) at time to" in the frame of reference j^Just established,

Having graphically provided the time and pressure axes t and P, the points to and to' and the dotted horizon- tal line L. at the patient's systolic blood pressure, and the part to" on the horizontal line, attention is now directed to the way in which the waveform is completed in a non-invasive manner. To this end, reference is directed to Figure 8 which is similar to Figure 4, with one exception. It may be recalled from the previous discussion of Figure 4 in conjunction with Link Patent 3,903,872, a patient's diastolic blood pressure can be determined electronically by varying cuff pressure P and detecting the resultant change in cuff volume ΔV within the same pressure or measuring band (e.g., the band 18, 18' and so on in Figure 4) in the patient's actual blood pressure waveform, that is, along the diastolic decline just before the systolic rise. The cuff pressure P resulting in a maximum change in arterial volume ΔVmax corresponds to the patient's diastolic blood pressure. The physical components for carrying out this process are described in the '872 patent and reference is made thereto. These components include means serving as a generator for detecting the beginning of the patient's systolic rise and measuring back therefrom, for example 50 milliseconds, in order to provide repeated¬ ly a 50 millisecond measuring band. During that time the change in cuff volume ΔV is measured by cooperat¬ ing means for each cuff pressure P . Means are also provided for determining when ΔV is at a maximum and for reading out the cuff pressure P at ΔVmax, this latter cuff pressure corresponding to the patient's diastolic pressure.

In accordance with the present invention, the genera¬ tor just recited is used to provide a series of identical measuring bands of width Δt moving back in time from the systolic rise at t ' , specifically from

t_χ - - to t_χ + ψ- ; from t₂ - -^ to t₂ + —■ ; and generally from tn - -2=— to tn + -2=—

as shown in Figure 8, rather than a single band as in Figure 4. These bands are indicated at 22, 22', 22" and so on. The generator means forming part of the components described in the '872 patent can be readily used or easily modified to provide these different points in time and therefore the different bands 22,

22', 22" and so on. During each measuring band 22, the cuff pressure Pc is varied in order to find the cuff pressure which results in a maximum change in cuff volume ΔVmax. That cuff pressure corresponds to the patient's actual blood pressure, for example P.. at a time t., in the patient's waveform where t. is temporally the center point of the band 22. This can be repeated (preferably all in one run) for the measuring bands 22' and 22" and so on for providing pressure points P, _ and P. _ and so on corresponding to time t_ and time , and so on in the graph of Figure 7 where t_ and t.. and so on are temporally the center points of their respective bands, while Figure 8 shows only three such bands along the diastolic decline, they can extend entirely back to the begin¬ ning of the pulse (time t ) , except along any segments of the pulse where the slope is zero, as for example at the systolic pressure point or at the dichrotic notch, as will be pointed out below. All of the intermediate points (except at zero slopes) between t and to' can be found this wa-y-.

Only if the measuring bands 22, 22' and so on are selected to fall on parts of the waveform of Figure 6 which have finite positive or negative slope, will this procedure be accurate. If by example the measur¬ ing band were located at time t, at a position of zero slope on the waveform of Figure 6, the resulting slope of the cuff pressure pulse at the time t. would also be zero for all values of the cuff pressure in this latter case the sensitivity of the method becomes zero when the slope of the true waveform becomes zero. With the exception of all these zero slope points on i y

the waveform; any other position on the waveform at general time t can be located by Observing the general cuff pressure P (n) at which the slope of the cuff pressure pulse has a maximum at time tn. The observed cuff pressure P (n) may then be plotted against t as a valid point on the waveform. This is made clear in Figure 7 in which the areas of the waveform which are unmeasurable or poorly measurable by this method are indicated by dotted lines.

The foregoing has been a discussion of how a particu¬ lar patient's actual blood pressure waveform can be closely approximated without an invasive device. This may be an important diagnostic tool to a doctor, especially if it turns out that his patient has an irregular waveform. This is best exemplified in Figures 9a-d which diagrammatically illustrate a number of waveforms having different mean values. The mean pressure P (πt) of a blood pressure waveform is equal to the diastolic blood pressure P_fa(D) plus a particular fraction K of the pulse pressure which is the difference between the patient's systolic blood pressure ^p (S) ^an<3 his diastolic blood pressure. Equation 2A shows this and equation 2B show the same thing in a convenient short hand notation.

P_b(m) = P_b(D) K(P- (s) -P_fa(D)) (2A)

M = D + K(S-D) . (2B)

Noting that the mean pressure M can be calculated by integrating the waveform (its amplitude pressure P) over time T (the duration of the waveform) so that:

-T

/ Pdt M = o ... (3) and :

-T

/ ^' Pdt

K = M - D = o - D . . . (4)

S - D

With the above equations in mind, the Figure 9a waveform can be shown to have a value (which is commonly referred to as the blood pressure constant) of about 0.50. The Figure 9b waveform approximates a K value of 0.6 while the Figure 9c waveform approxi¬ mates a K value of 0.2. Finally, the Figure 9d waveform approximates a K value of 0.33. This latter waveform most closely corresponds to a healthy blood pressure waveform and therefore some diagnostic devices of the prior art purport to calculate mean blood pressures by assuming a value of 0.33. With this assumption of =0.33 along with the patient's diastolic and systolic blood pressures, a Figure 9d waveform can be generated. Of course, this can be quite dangerous if the particular patient actually has a blood pressure constant of, for example, 0.60 or 0.20. However, in accordance with another aspect of the present invention, by generating the approximated waveform illustrated in Figure 7, all guess work regarding the patient's mean blood pressure and blood pressure constant K is eliminated. In fact, once the approximated waveform is determined, it can be inte- grated electronically so as to calculate the mean pressure P. (M) which might be helpful to the doctor and from this the blood pressure constant K can be readily calculated. Suitable means can readily be provided to make these various calculations.

As a result of the various aspects of the present invention described, a diagnostic tool can be provided which not only provides for a patient's diastolic and systolic blood pressures non-invasively but also a close approximation of the patient's actual blood pressure waveform as well as his mean pressure and blood pressure constant, again non-invasively. This tool or arrangement is shown in Figure 10 including suitable cuff means generally indicated at 30 in position around the arm of a patient in the normal operating manner and maintained at different pressure levels from zero pressure to, for example, 160' Torr. The resultant cuff pulses are monitored by transducer 34. Suitable and readily providable electronic means 35 serve to receive these pulses and from this information can provide the patient's diastolic and systolic blood pressures along with his arterial pressure-volume curve in accordance with the Link and Link et al patents. Means 35 also includes readily providable circuitry for providing the intermediate pressure points at times t., t_, t-, and so on from Figures 6-8 and with this information further readily providable circuitry for graphing the waveform in Figure 7 on an oscilloscope or monitor 38. In addition means 35 can also include circuitry for calculating the mean pressure P, (M) and blood pressure constant K from this waveform and equations 2-4 above.

Claims

It

WHAT IS CLAIMED IS:

1. A non-invasive method of providing a waveform approximating the actual blood pressure pulse in a particular artery of a given mammal over a specific period of time during which a number of such pulses successively occur, each having diastolic and systolic pressure points, one edge defining a systolic rise and a second edge including a diastolic decline, said method comprising the steps of: (a) determining by non-invasive means

(i) the diastolic and systolic pressure point of said actual blood pressure pulse,

(ii) the beginning and end points of said actual blood pressure pulse, and (iii) a plurality of intermediate pressure points on said actual blood pressure pulse at differ¬ ent fixed points in time prior to the end point of said pulse, at least some of said intermediate pres¬ sure points being located on the diastolic decline of said actual blood pressure pulse and all of said intermediate points being located on a positively or negatively sloping part of said actual pulse but not a part of said actual pulse have zero slope; and

(b) using said diastolic and systolic points, said beginning and ending points and said intermediate points, graphically providing a waveform having

(i) minimum and maximum amplitude points corresponding to said diastolic and systolic points, respectively, (ii) beginning and end points corresponding to the beginning and end points of said pulse,

(iii) a plurality of intermediate points respectively corresponding to the intermediate points of said pulse. 2. A method according to Claim 1 wherein only such diastolic and systolic points, said beginning and end points and said intermediate points of said graphical¬ ly actual blood pressure pulse are used to provide said waveform.

3. A method according to Claim 1 including the step of determining the mean value of said waveform whereby to provide an approximated mean value for said actual blood pressure pulse.

4. A method according to Claim 1 including the step of determining the blood pressure constant of said waveform whereby to provide an approximated blood pressure constant for said actual blood pressure pulse.

5. A method according to Claim 1 wherein some of said intermediate points are located on the systolic rise of said actual blood pressure pulse and also along the segment of said actual pulse between its systolic rise and diastolic decline.

6. A non-invasive method of providing a waveform approximating the actual blood, pressure pulse in a particular artery of a given mammal over a specific period of time during which a number of such pulses successively occur, each having diastolic and systolic pressure points, a back edge defining a systolic rise and a front edge including a diastolic decline, said method comprising the steps of:

(a) determining by non-invasive means

(i) the diastolic and systolic pressure points of said actual blood pressure pulse,

(ii) the beginning and end points of said actual blood pressure pulse, and (iii) a plurality of intermediate pressure points on said actual blood pressure pulse at differ¬ ent fixed points in time on its diastolic decline prior to the end point of said pulse; (b) using said diastolic and systolic points, said beginning and ending points and said intermediate points, graphically providing a waveform having

(i) minimum and maximum amplitude points corresponding to said diastolic and systolic points, respectively,

(ii) beginning and end points corresponding to the beginning and end points of said pulse,

(iii) a plurality of intermediate points respectively corresponding to the intermediate points of said pulse and a front edge defined by the waves end point and its intermediate points,

(iv) a back edge defined by the waves beginning point and the maximum amplitude point; and (c) whereby said graphically provided waveform approximates the actual blood pressure waveform in configuration and mean value.

7. A non-invasive method of providing a waveform approximating the actual blood pressure pulse in a particular artery of a given mammal over a specific period of time during which a number of such pulses successively occur, each having diastolic and systolic pressure points, a front edge including a diastolic decline and a front edge serving as a systolic rise, said method comprising the steps of: (a) determining by non-invasive means

(ii) the beginning and end points of said actual blood pressure pulse, and (iii) an intermediate pressure point on said actual blood pressure pulse at a fixed point in time on its diastolic decline prior to the end point of said pulse; and

(b) providing a waveform having minimum and maximum amplitude points corresponding graphically to the diastolic and systolic points of said actual blood pressure pulse, beginning and ending points corre¬ sponding graphically to the beginning and ending points of said pulse, a front edge extending from the end point of the waveform to a second point corre- sponding graphically to said intermediate pressure point and a back edge extending from the beginning point of the waveform to the said maximum amplitude point, whereby said waveform approximates said blood pressure pulse.

8. A non-invasive method of graphically plotting within a two-axis Cartesian coordinate system, where one axis represents pressure and the other represents time, a pressure versus time waveform approximating the actual blood pressure pulse in a particular artery of a given mammal over a specific period of time during which a number of such pulses successively occur, each having diastolic and systolic pressure points, a front edge including a diastolic decline, a back edge serving as a systolic rise and a dichrotic notch there between, said method comprising the steps of:

(a) determining by non-invasive means the diastolic and systolic pressure points of said actual blood pressure pulse and, using these pressure points as the minimum and maximum amplitudes of the waveform, graphically plotting a first line and a first point within said coordinate system representing the minimum and maximum pressures of said waveform, respectively;

(b) determining by non-invasive means the begin- ning and end points of said actual blood pressure pulse and, using these end points as the end points of said waveform, graphically plotting the beginning and ending points of said waveform on said first line within the coordinate system;

(c) determining by non-invasive means intermedi- ate pressure points on said actual blood pressure pulse at a number of fixed points in time along its diastolic taper prior to the end of said pulse and, using these intermediate pressure points as intermedi¬ ate pressure points on said waveform at the same fixed points in time prior to the end point of said waveform, plotting said last-mentioned intermediate pressure points within said coordinate system; and

(d) graphically providing within said coordinate system a second line defined by the end and intermedi- ate points of said waveform and a third line through said beginning point of said waveform and said first point, whereby to graphically complete said waveform except for the space between said second and third lines such that said third line corresponds to the systolic rise of said actual blood pressure pulse, said second line corresponds to the diastolic decline of the actual pulse and said space corresponds to the dichrotic notch of said actual pulse.

9. A non-invasive method of approximating the blood pressure constant K associated with the actual blood pulse in a particular artery of a given mammal, said method comprising the steps of:

(a) by non-invasive means, determining the dia¬ stolic and systolic pressure points D and S, respec- tively, of said actual blood pressure pulse;

(b) by non-invasive means, providing a waveform which approximates said actual pulse;

(c) finding the mean value M of said waveform; and (d) solving the equation

K = M - D S - D.

10. A non-invasive method of approximating the mean value M associated with the actual blood pressure pulse in a particular artery of a given mammal, said method comprising the steps of:

(a) providing a waveform which approximates said actual pulse; and

(b) integrating said waveform and solving for the equation

J Pdt M = O

where M is said mean value, T is the duration in time of said waveform and P is the waveforms' amplitude in pressure.

11. A non-invasive apparatus for providing a waveform approximating the actual blood pressure pulse in a particular artery of a given mammal over a specific period of time during which a number of such pulses successively occur, each having diastolic and systolic pressure points, a leading edge defining a systolic rise and a trailing edge including a diastolic taper, said apparatus comprising:

(a) means for determining by non-invasive means (i) the diastolic and systolic pressure points of said actual blood pressure pulse,

(ii) the beginning and end points of said actual blood pressure pulse, and (iii) a plurality of intermediate pressure points on said actual blood pressure pulse at differ¬ ent fixed points in time prior to the end point of said pulse; at least some of said intermediate pres- sure points being located on the diastolic decline of said actual blood pressure pulse and all of said intermediate points being located on a positively or negatively sloping part of said actual pulse but not a part of said actual pulse have zero slope; and (b) means responsive to said diastolic and sys¬ tolic points, said beginning and ending points and said intermediate points, for graphically providing a waveform having

(ii) beginning and end points corresponding to the beginning and end points of said pulse,and

(iii) a plurality of intermediate points respectively corresponding to the intermediate points of said pulse.

12. An apparatus according to Claim 10 wherein only such diastolic and systolic points, said beginning and end points and said intermediate points of said graph- ically actual blood pressure pulse are used to provide said waveform.

13. An apparatus according to Claim 11 including means for determining the mean value of said waveform whereby to provide an approximated mean value for said actual blood pressure pulse.

14. An apparatus according to Claim 11 including means for determining the blood pressure constant of said waveform whereby to provide an approximated blood pressure constant for said actual blood pressure pulse.

15. An apparatus according to claim 11 wherein some of said intermediate points are located on the systolic rise of said actual blood pressure pulse and also along the segment of said actual pulse between its systolic rise and diastolic decline.

16. A non-invasive apparatus for providing a waveform approximating the actual blood pressure pulse in a particular artery of a given mammal over a specific period of time during which a number of such pulses successively occur, each having diastolic and systolic pressure points, a front edge including diastolic taper and a back edge serving as a systolic rise, said apparatus comprising:

(ii) the beginning and end points of said actual blood pressure pulse, and

(iii) a plurality of intermediate pressure point on said actual blood pressure pulse at a fixed point in time on its diastolic taper prior to the end point of said pulse; and (b) means for providing a waveform having minimum and maximum amplitudes points corresponding graphically to the diastolic and systolic points of said actual blood pressure pulse, beginning and ending points corresponding graphically to the beginning and ending points of said pulse, a front edge extending from the end point of the waveform through a series of said second points corresponding graphically to said intermediate pressure points and a leading edge extending from the beginning point of the waveform to the waveforms* maximum amplitude point. u

17. A non-invasive apparatus for approximating the blood pressure constant K associated with the actual blood pulse in a particular artery of a given mammal, said apparatus comprising: (a) non-invasive means for determining means the diastolic and systolic pressure points D and S, respectively, of said actual blood pressure pulse;

(b) non-invasive means for providing a waveform which approximates said actual pulse; (c) means for finding the mean value M of said waveform; and

(d) means for solving the equation

K = M - D S - D.

18. A non-invasive apparatus for approximating the mean value M associated with the actual blood pressure pulse in a particular artery of a given mammal, said apparatus comprising:

(a) non-invasive means for providing a waveform which approximates said actual pulse; (b) means for integrating said waveform and solving for the equation τ

Pdt

M = o

where M is said mean value, T is the duration in time of the waveform and P is the waveforms' amplitude in pressure.