FR3075590A1 - METHOD OF MEASURING AVERAGE ARTERIAL PRESSURE - Google Patents

Abstract

The present invention relates to a method for evaluating the mean arterial pressure of a patient from plethysmography measurements by calculating an estimated value of the arterial pressure PAMest from the value of a parameter obtained at the time t and a Calib calibration value evaluated at a time t0.

Description

The present invention relates to the field of anesthesia, and more particularly to a new procedure and a new algorithm which can be used in particular for monitoring the general condition of a patient during anesthesia and in particular during the induction period. 'anesthesia.

PRIOR STATE OF THE ART

The role of the anesthesiologist-resuscitator is to watch over and ensure the control of the patient's major vital functions, during general anesthesia, throughout the duration of this anesthesia, and especially the period of induction, period critical where the use of anesthetic in high doses to induce loss of consciousness most often leads to low blood pressure.

This hypotension arises from the interaction between the direct vasodilating peripheral effect of hypnotic and morphinomimetic drugs, and the central anesthetic effect of blood pressure regulatory centers. Indeed, blood pressure (BP) is a physiological parameter finely regulated and compensated by complex reflex systems. During general anesthesia all these systems are globally inhibited and this in proportion to the depth of the anesthesia. The risk linked to low blood pressure has been demonstrated, but the threshold from which consequences for the perfusion of one or more organs appear varies according to the mechanism, associated anomalies (HR, cardiac output and oxygen transport) and patient's particular terrain.

The major role of average blood pressure in organ perfusion, particularly that of the organs most sensitive to hypoperfusion (brain, heart, digestive tract and kidneys) exposes you to the risk of failure if it drops.

In clinical practice, the prevention of intraoperative arterial hypotension which makes it possible to maintain an average arterial pressure close to the initial mean arterial pressure, that is to say before anesthesia, is widely practiced in order to reduce the vital risk and the 'improvement of the operating suites. In addition, vasoactive agents are used during general anesthesia to quickly correct any variation in potentially harmful average blood pressure, particularly phenylephrine, ephedrine and norepinephrine.

We can commonly measure three arterial pressures: The systolic (PAS) and diastolic (PAD) arterial pressures are measured using an arm cuff placed on the arm and a stethoscope, the doctor listening for the appearance and then the disappearance pulse. These pressures represent the maximum systolic and minimum diastolic pressures respectively during a cardiac cycle. Average blood pressure (MAP) is measured using a sphygmomanometer and is an indicator of the organ perfusion pressure.

The sphygmomanometer (or electronic blood pressure monitor) does not use the principle of the auscultatory or palpatory method, but an oscillometric measurement. When the sleeve placed at the patient's forearm deflates automatically, oscillations are recorded by the device. These oscillations start before the actual systolic value and end after the actual diastolic value. The maximum value of the oscillation represents the average blood pressure (Figure 1). From this PAM value and algorithms developed by the manufacturers, the PAS and PAD are calculated. It is this device that is used by anesthesiologists in operating theaters.

The average arterial pressure can be estimated from the systolic and diastolic arterial pressures by the formula PAM = 2/3 PAD + 1/3 PAS. However, this formula is only an estimate, discussed in the art, of other formulas having notably been developed (see in particular Razminia et al Catheter Cardiovasc Interv. 2004 Dec; 63 (4): 419-25). In a response given on Research Gâte, Gianni Losano (University of Turin) recalls the definition of average blood pressure (average of all the values that blood pressure presents during a whole cardiac cycle), and that it can sometimes be the average between systolic pressure and diastolic pressure.

https://www.researchgate.net/post/What_formulas_and_methods_exist_for_the_ca I had I ati o n_of_m ea n_arte ri a l_b I ood_p ress u re it therefore appears that the knowledge of systolic and diastolic pressures is insufficient to know the average blood pressure.

The objective of the anesthesiologist is therefore to protect the patient from physiological disturbances induced by the surgical act and the anesthetic agents. It requires continuous monitoring of the cardiovascular system (heart rate, blood pressure), the respiratory system (respiratory rate, pulse oximeter, expired CO2) and the temperature in the event of long-term intervention. This control must be adapted to the risks and contexts to limit these episodes of hemodynamic instability.

As a routine, minimum monitoring notably includes measuring at least every 5 minutes the average blood pressure, invasively or not, as well as continuous measurement of pulsed oxygen saturation (SpO2). These controls are imposed by regulations.

The measurement of non-invasive average blood pressure is now carried out by an oscillometric method using a cuff, as indicated above. However, one of the major limitations of this blood pressure control is its intermittent nature and therefore systematically lagging behind the hemodynamic variations caused by the anesthesia. Thus, it is not possible to obtain the blood pressure values continuously. In fact, the measurement of the blood pressure takes about a minute and cannot be repeated too often, so as not to damage the patient's arm by too repeated swelling.

The measurement of pulsed oxygen saturation is carried out by plethysmography with a sensor generally positioned at the fingertip. This measure is inexpensive and provides continuous information.

This sensor is called a pulse oximeter (plethysmograph), and contains two diodes emitting red light which must be located in front of a receiving area. The diodes emit two lights (red and infrared), whose absorption by the pulsating flow is measured. This absorption of red and infrared light is variable depending on whether it meets non-oxygenated reduced hemoglobin (Hb) or oxygenated oxyhemoglobin (HbO2). The output data is therefore absorption data. Figure 2 shows that the absorption measured by the pulse oximeter contains several components, the only variable observed under stable conditions being the arterial systolic flow.

The pulse wave is generated by the heart with each beat. It causes variations in blood volume in the arteries, which contract and relax as it passes. This wave is dicrotic, reflecting (first peak) the blood expulsion from the ventricle in the cardiovascular system (systolic pressure), and (second peak, or change in the slope of decay of the pulse wave after the maximum) the residual propulsion observed at the closing of the sigmoid valves, and representing the relaxation (contraction) of the arteries close to the heart which had dilated to absorb the blood flow during blood propulsion.

It is also recalled that the blood pressure wave can be described according to its average component and its pulsatile component. The average component is MAP, which is considered constant from the aorta to large caliber peripheral arteries while the pulsatile component varies according to complex phenomena all along the arterial tree.

An application for smartphones already exists as an aid for monitoring certain physiological parameters (Captesia, described in particular by Desebbe et al, Anesth Analg 2016; 123: 105-13). This application requires taking a photograph of the pulse wave graph, as shown by plethysmography, and sending this photo to a server, with certain patient parameters. The application returns the variation of the pulse pressure, but not the average blood pressure in real time and continuous, because it requires operator intervention. This application is linked to patent application WO 2015/181622.

Application WO 2017/037369 describes a non-invasive device for measuring the aortic flow rate in a small mammal, using a device for plethysmography of the chest by inductance measurement, the acquisition and analysis of the signal of variation of section of each turn.

Application FR3024943 describes a method for determining the respiratory rate of an individual from his heart rate, by measuring the heart signal by plethysmography.

Application WO 2007/128518 describes a non-invasive device for continuous measurement of arterial pressure (PA) characterized in that it comprises means for indirect measurement from a signal (VS) representing directly or indirectly the variations the volume of blood in an organ or part of the body, said volume variation signal being calibrated using intermittent values of PA obtained by standard non-invasive methods, preferably without using one or more predefined constants, and in particular one or more physiological parameters assumed to be constant. The signal (VS) representing directly or indirectly the variations in blood volume in an organ or part of the body can be the plethysmographic wave of blood oxygen saturation (SpO2) measured using a saturometer at the level a finger of the hand or the foot, the ear, the forehead, the wings of the nose, or other organ or part of the body. This device can comprise means for detecting the amplitude of the signal VS and the time for which the signal VS reaches its maximum (VSmax) and / or its minimum (VSmin) and / or a predefined reference value (VSO), during each heartbeat (cycle), by calculating the following parameters in real time:

• the rise time (Tm) for each heartbeat (cycle), defined as the time interval between the VS signal passing from the VSmin value or the VSO value to the Vsmax value during its rise: Tm = t (VSmax ) -1 (VSmin) or Tm = t (VSmax) - t (VSO);

• and / or the descent time (Td) for each heartbeat (cycle), defined as the time interval between the VS signal passing from the VSmax value or the VSO value to the VSmin value during its descent: Tm = t (VSmin) - t (VSmax) or Tm = t (VSmin) - t (VSO).

However, this document only describes the assessment of systolic blood pressure using the signal amplitude and duration of the peak rise with cuff calibration. However, as mentioned, this calibration is not correct because the cuff only gives an estimate of the systolic pressure. Furthermore, this document is silent on how to obtain the function linking the parameters measured on the pulse wave and the systolic pressure. It therefore appears that the teaching of this document is not relevant to solving the problem addressed in the present application (measurement of mean arterial pressure, determining for organ perfusion) since the peripheral systolic pressure is not a good indicator of tissue perfusion. In addition, systolic pressure is very much dependent on where it is measured. It can thus be seen, in FIG. 1, that the systole on the plethysmography does not vary in the same way as the real systole. We also note that the equation proposed by this document uses a constancy, to take into account the diastolic pressure (the calculation is carried out on the pulsed pressure) and which more or less represents the volume of blood which was pumped and rejected by the heart. Using this constant adds to the method uncertainty of this document.

US Patent 5,269,310 describes a method of determining blood pressure by attaching a plethysmograph to a patient such that said plethysmograph interacts with an artery of said patient, said plethysmograph generating an output signal having a predetermined relationship with a characteristic of blood in said artery; calibrate said plethysmograph during a calibration period by determining the patient's real blood pressure by means other than said plethysmograph, then determine the value of a first arterial characteristic in a predetermined relationship between said first arterial characteristic, the arterial volume indicated by said plethysmograph output signal, a conversion value corresponding to arterial volume at infinite pressure, and said actual blood pressure during said calibration period; and analyzing said output signal from the plethysmograph during a measurement period to determine a blood pressure corresponding to said output signal in accordance with said predetermined relationship. This document teaches the measurement of electrical voltages corresponding to the systole and diastole (column 6) and to a calculation of an electrical voltage corresponding to the "average electrical voltage". The methods described do not mention obtaining or using average blood pressure.

US Patent 5309916 describes a method for evaluating blood pressure, based on two variables, and essentially describes the use of the pulse wave speed and the blood flow speed.

STATEMENT OF THE INVENTION

The invention relates to a new ex vivo method for continuously assessing in real time the average blood pressure of a patient, in particular of a patient undergoing anesthesia, which uses values obtained in real time. Preferably, these values were obtained by plethysmography. This method therefore allows the anesthesiologist to react immediately in the event of a pressure drop below a predetermined threshold. It should be noted that the method described below is not implemented on the human body, but is implemented ex vivo, and uses values previously measured. The invention does not include the measurement of values, but only their manipulation (described below) in order to obtain a reliable estimate of the patient's blood pressure.

Thus, the methods described below are preferably based on the measurement of the pulse wave by plethysmography, which is a consequence of systolic expulsion, and allow the deduction and monitoring of the variation in mean arterial pressure, including we saw above that it is not clearly and directly linked to systolic blood pressure. The measurement by plethysmography is carried out using a saturometer (pulse oximeter) at the level of a finger of the hand or the foot, the ear, the forehead, the wings of the nose, or other organ or part of the body. It is preferred when this measurement is made with a finger saturometer, or in the earlobe.

Since the relationship between mean pressure and systolic and diastolic pressure is complex, and the pulse wave represents systolic pressure and allows oxygen saturation to be calculated, it is surprising that the pulse wave to measure general average pressure. Such a result could not be envisaged in view of the prior art, and it can be noted that the documents mentioned above do not speak of it, even though the average pressure is the parameter which makes sense in anesthesia. to check for good organ perfusion.

The invention thus relates to a method (or process) carried out ex vivo for continuously evaluating the average arterial pressure of a patient, on the basis of values of a parameter measured continuously, comprising the steps:

I. Calculate a Calib calibration value from

at. From the value of blood pressure at a time tO

b. From the VpO value linked to the measurement of the parameter obtained in the patient at time tO,

II. Calculate the estimated value PAMest of the patient's blood pressure at a time t after tO by the formula PAMest = k x Calib x Vpt, in which Vpt is the value of the parameter measurement obtained at time t.

This method therefore provides an average blood pressure value for each heartbeat. This method is very favorably implemented by computer, therefore in silico.

Thus, it is possible to evaluate the mean arterial pressure continuously by using the correlation coefficient Calib calculated on the basis of an actual measurement of the mean arterial pressure at time tO and of a measurement of the physiological parameter which can be obtained in continuous at the same time tO.

It is preferred when the parameter measured continuously is by plethysmography. There are various methods of plethysmography, and the preferred method within the framework of the invention is photoplethysmography or plethysmography by photoelectric effect.

This method makes it possible to determine the excess volume of blood at each heartbeat ("Pl" for "perfusion index" or index of perfusion) linked to the blood expulsion after systole. Thus, Pl is expressed as a percentage relative to the “non-pulsating” blood volume of the finger.

This method also measures the relationship between deoxy- and oxyhemoglobin (known as SpO ₂ ).

The rendering of a photoplethysmography is a graph representing the variation in the volume measured, as well as the two values Pl and SpO ₂ mentioned above. This graph represents the pulse wave and reflects the blood expulsion at the time of systole. We can thus identify the two pressure peaks which send blood into the body (main peak at the exit of the heart and secondary peak (dicrote wave) after relaxation of the vessels close to the ventricle). This dicrotic wave can be represented, on the graph, by a second peak, appearing during the decrease observed for the main peak, by a plateau, or by a variation of slope during the decrease of the main peak (corresponding to the influx of blood, which is too weak to create a new peak or plateau, however). The form taken of this second pressure wave (dicrotic wave), at the level of the plethysmograph graph, depends on several factors, such as the nature and precision of the plethysmograph, the cardiac state of the patient, and / or the state of the patient's vessels. However, as mentioned above, it is still possible to detect this second pressure wave.

In a preferred embodiment, the value of the dicrotic wave is used as the value Vpt of the continuously measured parameter. We thus use the height of this dicrotic wave, if we can observe a peak or a plateau (figure 4.A and 4. B. If we observe only a break in the decay curve, we use, as the beginning of l dicrotic wave, the place of this break (figure 4.C). The height of the dicrotic wave used is very preferably the total height or absolute height of the plethysmography signal (Hd, figure 3) preferably to the measured height relative to the baseline of the pulsatile part (which corresponds to the diastolic height). In fact, the absorption value of this baseline is likely to vary over time (Figure 7.B), which can lead to bad evaluations if one does not take into account the absolute value of the absorption measured at the appearance of the dicrote wave.

In another embodiment, the value Vpt which can be calculated continuously is linked to the logarithm of the inverse of the perfusion index (PI) at time t.

In another embodiment, the value Vpt used is the time between the start of the pulse wave and the dicrotic wave (the break in slope observed during the decrease in the pulse wave) designated as T2 in Figure 3.

In another embodiment, the value Vpt used is the time between the start of the pulse wave and the maximum of the pulse wave, designated as T1 in FIG. 3.

In another embodiment, the value Vpt used is the period of the pulse wave (duration measured between the feet of two successive pulse odes), designated as T3 in FIG. 3. This parameter is essentially used as as a secondary parameter, when several variables are used, in particular to weight the MAP is calculated with another parameter.

In another embodiment, the value Vpt is the ratio of the value of the dicrotic wave relative to the value of the maximum systolic peak or the ratio of the value of the dicrotic wave relative to the value of the diastole (foot of the pulse wave).

As mentioned above, the pulse wave is the pressure wave that can be detected by an influx of blood to an organ, and which is therefore linked to heartbeat and systole.

In a particular embodiment, the value Vpt used at time t is an average value of several values measured during a predetermined period. Using such an averaged value makes it possible to get rid of any singular variation at an instant t. One can in particular use the value averaged over three systoles.

In a particular embodiment, the method described above is implemented on a patient during general anesthesia. This method is therefore implemented for a period of several tens of minutes, or even several hours. It may be appropriate to perform a new calibration from time to time, especially when taking blood pressure on the cuff.

Thus, the method described above (actual measurement of blood pressure, measurement of the parameter, calculation of the Calib value, subsequent evaluation of the blood pressure using this Calib value) can be repeated several times at predetermined intervals. In particular, we can repeat this method each time we take the blood pressure at the cuff (i.e. essentially recalculate a Calib value), or every two times, or every three times. This avoids a possible drift of the estimated value of the average blood pressure between two real measurements. It is also possible to carry out this calculation of the Calib value if the estimated value of the average arterial pressure is too far away (variation of 10% or 5%) from the actual value measured from time to time.

In another embodiment, it is also possible to refine the value PAMest evaluated at time t, by calculating several of these values (using different continuously measurable parameters) and by performing an interpolation (evaluation) of the real PAMest value, as a function of the several results obtained. Thus, we make n individual measurements with n parameters (for example dicrotic wave, Pl, time difference between the peaks ...) then we measure the estimated PAMest for each of the parameters, and we calculate a final PAMest, probabilistically.

We thus describe an ex vivo method for continuously assessing the mean arterial pressure in a patient, characterized in that

at. we repeat the method described above for different parameters, in order to obtain several PAMest values at time t (each one being linked to a particular parameter)

b. we calculate a final PAM value by statistical estimation taking into account

i. different PAM values are calculated at time t ii. one or more final PAM values calculated before time t.

In step a), it is thus possible to calculate a PAMest at time t on the basis of the measured values of the following parameters chosen from the height of the dicrotic wave, the logarithm of the inverse of the perfusion index (Pl) (to which 1 is added to avoid obtaining a negative value), the duration between the start of the pulse wave and the dicrotic wave, and the duration between the start of the pulse wave and the maximum of the pulse. You can also use the ratio (height of the dicrotic wave / maximum height of the pulse wave) and / or the ratio (height of the dicrotic wave / height of the diastolic part (foot of the pulse wave) .

In particular, it is thus possible to calculate a PAMest at time t on the basis of the following combinations of the measured values of the dicrotic height and logarithm parameters of (the inverse of the perfusion index (PI) + 1) height of the dicrote wave and duration between the start of the pulse wave and the dicrote wave height of the dicrote wave and duration between the start of the pulse wave and the maximum of the pulse wave height of the dicrote wave , logarithm of (the inverse of the perfusion index (PI) + 1), and duration between the start of the pulse wave and the dicrotic wave height of the dicrotic wave, logarithm of (the inverse of the perfusion index (PI) + 1), and duration between the start of the pulse wave and the maximum of the pulse wave height of the dicrote wave, duration between the start of the pulse wave and the maximum of pulse wave, and duration between the start of the pulse wave and the dicrotic wave height of the dicrotic wave, logarithm of (the inverse of the indi ce of infusion (PI) + 1), duration between the start of the pulse wave and the dicrote wave, and duration between the start of the pulse wave and the maximum of the pulse wave

Other combinations not including the height of the dicrotic wave can also be envisaged and / or integrating the ratios (height of the dicrotic wave / maximum height of the pulse wave) and / or (height of the dicrotic wave / height of the diastolic part).

It is preferred when the height of the dicrotic wave (or the ratio height of the dicrotic wave / height of the systole, or the ratio height of the dicrotic wave / height of the systole) is one of the variables measured and used, the other variables being chosen from those mentioned above. In fact, and as shown in the examples, the parameters linked to the dicrotic wave (and in particular to its height) make it possible to obtain a very good value of the average arterial pressure, and will essentially have the most important weight in determination of the final MAP, the other variables being used essentially to weight the variables linked to the dicrotic wave which may prove to be interesting for certain patients.

At the end of the implementation of step a), we thus obtain n estimated values of the mean arterial pressure (n being the number of parameters selected).

Step b) consists of evaluating an average blood pressure on the basis of these n estimated values, and of the value estimated at the previous time.

Such an evaluation can be performed by any statistical method known in the art, and in particular using a Kalman filter in a discrete context. The Kalman filter in discrete context is a recursive estimator. This means that to estimate the state at a time t, we only use the estimate of the previous state and the measurements at time t. The history of observations and estimates is therefore not required.

The use of the values of several parameters and of the statistical probability as well as proposed by filters such as the Kalman filter makes it possible to increase the reliability of the displayed measurement of the average arterial pressure, compared to a measurement evaluated only on the base of a single parameter. This allows in particular not to give too much weight to outliers that could be obtained for a parameter at a given time (for whatever reason).

In another embodiment, an ex vivo method is thus described for continuously assessing the mean arterial pressure in a patient, characterized in that

at. we repeat the method described above for different parameters (Vpnt), in order to obtain several PAMest values (PAMestn) at time t (each one being linked to a particular parameter)

b. a final PAMest value is calculated by combining the different PAMest values calculated at time t

The combination described in step b) is preferably a linear regression, and the final PAM is written alPAMestl + a2 PAMest2 + ... (PAMestn corresponding to the value PAMest obtained for the parameter Vpnt measured at time t.

This linear regression is performed by any method known in the art, taking into account the relative weight of the PAMestn values for each parameter. The factors a1, a2 ... are preferably recalculated with each actual measurement of average cuff pressure.

Preferably, this method is used when there is already a certain amount of data, which makes it possible to refine the quality of the coefficients. It is thus possible to use, as initial coefficients, coefficients previously calculated on a cohort of other patients (at least 50, preferably at least 100 patients). Even if there is inter-patient variability and the coefficients obtained on this cohort are not necessarily the best for the patient in question, these previously calculated coefficients can be used before being refined based on the data obtained. for the patient. Thus, during each calibration, the PAMest is calculated with the coefficients previously used (the initial coefficients (from the cohort) during the first calibration), this value is compared to the measured PAM value, the coefficients are readjusted by giving more and more weight to the values measured for the patient, in the regression as we have the data of MAP measured in this patient.

Thus, the methods described above are based on the fact that the mean arterial pressure can be measured as a function of parameters which can be measured continuously, preferably from a plethysmography measurement. In particular, the height of the dicrotic wave (or the ratio compared to the total value of the systolic wave and / or the diastolic wave) or the logarithm of the inverse of the perfusion index (increased by 1 ) are proportional to the average blood pressure. The method for obtaining average blood pressure is therefore much simpler than those described in the prior art, while being reliable.

As mentioned above, the methods described are particularly advantageous for monitoring the status of a patient's blood pressure continuously when the latter is in a state of general anesthesia. This allows the doctor to be able to act quickly in the event of too low pressure, without having to wait for the actual measurement obtained from the cuff.

Thus, it is preferred when these methods are implemented continuously throughout the duration of a patient's general anesthesia.

Furthermore, and in order to ensure patient safety, it is possible that a signal will be emitted when the PAMest value is less than a predetermined threshold (this can be considered if the average blood pressure is less than 65 mm Hg) . Such a signal emission step can be integrated into a method as described above. The signal can be a graphic signal (such as a representation of the average blood pressure by a color code different from the classic color code (red in the event of an alert instead of green or yellow). One can also consider displaying the value with different color codes on the follow-up monitor, to alert the doctor to a risk of hypotension.

The alert signal emitted can also or alternatively be an audible signal (long beep or other) when the average arterial pressure falls below a predetermined value. This also alerts the surgeon to a problem, and the need, for the anesthesiologist to perform the medical procedure to correct this hypotension.

The invention also relates to a computer product / program comprising program code instructions recorded on a medium readable by a computer, for implementing the steps of the methods described above, when said program is executed on a computer.

This program can also contain code instructions allowing the display of the mean blood pressure value on a monitor. It can also include code instructions allowing the emission of a visual and / or audible signal if the estimated average blood pressure value is below a predetermined threshold (pre-programmed or entered by the practitioner). It can also include code instructions for an actual blood pressure measurement to be performed on the patient if the estimated value is less than a predetermined value. Thus, the program can request and order a blood pressure measurement before the regulatory time, without the intervention of a human being.

The invention also relates to a computer-readable recording medium on which a computer program is recorded comprising program code instructions for executing the steps of the methods as described above or of programs such as described above.

The invention may also include a device for implementing a method as described above, comprising:

means for receiving data for measuring the average arterial pressure, in particular as taken by non-invasive method (cuff) means for receiving data for one or more parameters measured continuously (in particular the height of the wave dicrote, or the perfusion index) means of calculation making it possible to calculate a coefficient Calib at each real measurement of average arterial pressure means of calculating making it possible to calculate an average arterial pressure at each instant t as a function of the value of the parameters at this instant t, according to the methods described above, means for presenting the average arterial pressure calculated at each instant t (optionally including alert means in the event that the calculated arterial pressure is less than a predetermined value) possibly means for automatic actuation of the device for measuring arterial pressure the mean, if the calculated mean blood pressure is less than a predetermined value.

The computing means are essentially processors allowing the execution of the computer products / programs as mentioned above. The means for receiving data, for presenting the average arterial pressure, or for actuating the sphygmomanometer are conventional means in the art.

The methods described above are of primary interest in the field of anesthesia and post-interventional monitoring room, but can also be used in other fields such as resuscitation (in particular hyperventilated patients), cardiology, city medicine, or emergency medicine (pre-hospital and inter-hospital). The methods and devices can also be used in sports medicine. Methods and devices can also be used to assess average blood pressure in patient stress tests.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1: example of a graph of a sphygmomanometer signal (obtained from the site According to https://www.infirmiers.com/etudiant-en-ifsi/cours/cours-cardiologie-lapression-arterielle-et-sa-mesure .html)

Figure 2: Principle of pulse oximetry. (1): variable light absorption linked to the variation in arterial blood volume. (2) constant light absorption linked to the non-pulsatile part of arterial blood. (3): constant light absorption linked to venous blood. (4): constant light absorption linked to tissues, bones ... According to Feissel, Réanimation 16 (2007) 124-131.

Figure 3: Representation of the variables usable in the context of the invention, from a pulse oximeter plot. T1: duration between the start and the maximum of the pulse wave; T2: duration between the start of the pulse wave and the dicrotic wave; T3: total duration of the pulse wave; Hd: height of the dicrotic wave.

Figure 4: representations of different types of plethysmography signals with location of the dicrotic wave.

Figure 5: flowchart representing the implementation of a method according to the invention.

Figure 6: flowchart representing the implementation of another embodiment of a method according to the invention.

Figure 7: A. Representation, in a representative patient, of the MAP measured by invasive method (ARTm, large dotted lines), the MAP estimated by a method according to the invention on the basis of the height of the dicrotic wave (PlethoMAP, small dotted lines), the MAP measured by the cuff sphygmomanometer (NBPm, black dots) and the calibration factor (Calib, solid line). B. Representation, for the same patient and the same period, the evolution of the systolic peak (solid line), the dicrote (small dotted lines) and the diastole (large dotted lines) over time.

A graphical representation of the superimposed pulse wave is also shown in this figure, only to improve the understanding of the three points (what each value corresponds to). Note: the time scales of this representation of the pulse wave and of the evolution of the peaks over time are different and this representation of the pulse wave (generally lasting 1 sec or less) is only present d 'from an informative point of view.

EXAMPLES

Example 1. Determination of the parameters that can be used to measure blood pressure continuously

In cardiology, it is therefore possible to measure the variation of the pulse wave non-invasively by plethysmography. We thus obtain a curve (graph) representing the excess volume due to the systolic expulsion.

The perfusion index (PI) reflects the amount of blood flow measured locally and in part of the pulsed arterial flow, the volume of systolic ejection. The perfusion index represents the area under the curve mentioned above. In the case of photoelectric plethysmography, an SpO ₂ value is also obtained, representing the pulsed oxygen saturation.

The curve represents the profile of the pulse wave and makes it possible to see the dicrotic wave leaving the heart at the systole, (a second peak (possibly two peaks), a plateau or a break in decay) in the organ. .

This plethysmography signal can be used for the measurement of mean arterial pressure continuously.

We thus use the variations of peak size (of the dicrotic wave), of area (value of the perfusion index) or of timing between two events of the profile of the pulse wave. In particular, we can actually see the decrease of the dicrotic wave, that is to say that we can detect this dicrotic wave with certainty in the plethysmography signal.

In order to determine these parameters, it is possible to analyze the pulse wave.

Pulse wave foot

We can, at first, detect the foot of the pulse wave. The bottom of the pulse wave is characterized by a rapid rise in the signal, which results in a peak in the second derivative of the signal. The analysis of this second derivative of the signal obtained by the plethysmograph makes it possible to obtain a signal (peak of the second derivative) at each rising edge of the pulse wave, and therefore to detect the foot of the pulse wave, and therefore the moment corresponding to the start of the signal.

Systole (peak pulse wave)

From the bottom of the wave, for each cycle, we look for the maximum value of the cycle. To do this, you can split the signal into several time windows and look at the maximum value in each time window. It is thus possible to determine the maximum local value corresponding to the maximum of the pulse wave. This gives the value of the maximum (absorbance value given by the pulse oximeter corresponding to the peak of the systolic wave) when this maximum is reached

We can therefore calculate the time between the start of the pulse wave and the maximum of the pulse wave.

Dicrote wave

Once the peak of the pulse wave has been identified, the second derivative of the signal is analyzed over predetermined time windows (between 50 ms and 300 ms). We are looking for the local maximum of the second derivative which is located after the systolic peak, in order to determine an area of interest in which the minimum in absolute value of the first derivative is sought. The dicrotic point corresponding to the dicrotic wave corresponds to this point for which the absolute minimum of the first derivative is reached. We can then measure its value (absorbance value given by the pulse oximeter), as well as the duration between the foot of the pulse wave and this point.

Example of practical application of this method

Signal collection

The signal collection was carried out in real time from a standard patient monitor capable of providing a photoplethysmography wave as well as non-invasive blood pressure by blood pressure cuff. The connection to the monitor is usually made via an RS232 serial port or a network connection via

Ethernet or WIFI. In addition to these two essential parameters, the value of the perfusion index (PI) was also used. Communication with the monitor can be bidirectional and allow, for example, to request a new non-invasive pressure test on demand.

Signal analysis

The signal processing is based on an online algorithm which measures the values and produces a beat by beat result in real time.

Pulse wave foot

The software uses a heartbeat detection algorithm which is based on the detection of the foot of the pulse wave. The bottom of the pulse wave is characterized by a rapid rise in the signal, which results in a peak in the second derivative of the signal.

Pulse wave foot detection is based on the second derivative of the signal. This second derivative is weighted by the first derivative to focus only on the increasing part of the signal (that is to say that the second derivative is given a zero value when the first derivative is not positive) . The values obtained are squared then integrated in a floating manner on a centered 240ms window (average on values 120 ms before and 120 ms after the desired point). This integrated signal provides a strong signal at each rising edge of the pulse wave.

This signal is compared to a threshold value. This threshold value depends on the patient, the equipment used, the shape of the plethysmography signal and the measurement noise. Since the measurement noise is not constant, the threshold is necessarily adaptive in real time. To calculate the threshold, we use the integral (calculated above) to which a floating average is applied with a window of 3s (centered or not), the result being multiplied by 1.5. The threshold thus obtained makes it possible to define an area of interest where the integral exceeds the threshold. Within this area of interest, the peak of the second derivative defines the bottom of the wave.

Systole (peak pulse wave)

From the bottom of the wave, for each cycle, the detection of the dicrotic wave is carried out in two stages. The first step is to find the systole and therefore the maximum value of the cycle. For this, the signal is split into 50 ms windows and the signal is advanced window by window as long as higher values are found. Once the highest value has been exceeded, the local maximum value (corresponding to the maximum value of the pulse wave, or systole value).

Dicrote wave

Once the peak of the pulse wave has been identified, progress is made from this point by analyzing the second derivative of the signal, progressing through 150ms windows. We thus seek the local maximum of the second derivative which is located after the systolic peak. Once this peak is identified, it informs of an area of interest in which the dicrotic wave is located. From this point, the first derivative is analyzed and the minimum in absolute value of the first derivative is sought in a window of 8 ms from the peak of the second derivative.

The minimum of the first derivative close to the peak of the second derivative is thus obtained. This point is defined as the dicrotic point corresponding to the dicrotic wave and its value can be measured (total absorbance value given by the pulse oximeter).

Example 2. Calibration and continuous estimation of mean arterial pressure (MAP / MAP)

The calibration requires the value (Vp) of at least one of the following parameters height of the dicrotic wave value of the logarithm (natural or decimal) of the inverse of the perfusion index Pl (In (1 / PI + 1) ). Add 1 to the inverse of the perfusion index to avoid taking the logarithm of a value less than 1, and to obtain a negative value of the duration between the foot of the pulse wave and the dicrotic wave and value of the duration between the foot of the pulse wave and the maximum of the pulse wave ratio "height of the dicrotic wave / height of the pulse wave" and / or "height of the dicrote wave / height of the diastole wave ". total duration of the pulse wave

These values can be obtained beat by beat (i.e. for each pulse wave). Figure 3 shows how these variables are measured.

The calibration also requires the mean arterial pressure (MAP) value which can be obtained in particular by a non-invasive blood pressure cuff.

You can use an average over several cycles (2, 3, 4 5, 6, 8 or 10 cycles) of the value of the chosen parameter. This eliminates disturbances like respiratory variability in pressure and irregular heart cycle. This mean is preferably the statistical median rather than the arithmetic mean.

A Calib calibration factor is estimated when taking the non-invasive blood pressure and is obtained by Calib = PAM / Vp. We understand that the Calib value depends on the chosen parameter and that the Calib value obtained if we choose the value of the dicrotic wave will be different from the Calib value if we choose the logarithm of (the inverse of the perfusion index (PI) + 1).

Once the Calib value has been obtained, the mean arterial pressure is estimated beat by beat using the photoplethysmography signal as the only source of data.

The estimated PAM (PAMest) at time t is calculated by the formula

PAMest = Calib x VPt, in which Vpt is the value (possibly averaged) of the chosen parameter.

Application example (the parameter being the dicrotic wave)

The calibration requires the values of the dicrotic wave beat by beat as well as intermittent values of average arterial pressure for example by a non-invasive blood pressure cuff. The dicrotic pressure value obtained by the dicrotic wave is averaged over several cycles.

The number of cycles over which the value is averaged is adjustable (for example 5 cycles). This eliminates disturbances like respiratory variability in pressure and irregular heart cycle. The averaging is done using the statistical median rather than the arithmetic mean in order to be more robust in the presence of noise. A calibration factor is estimated when taking the non-invasive blood pressure and is based on the current average value of the dicrote (Pdic) and the average measured blood pressure (MAP). The calibration factor is obtained by Calib = PAM / Pdic.

Continuous mean arterial pressure assessment (MAP / MAP)

Once the calibration is done, the average blood pressure is estimated beat by beat using the absorption measured by photoplethysmography as the only signal source. The estimated PAM (PAMest) is calculated by PAMest = Calib-Pdic, where Pdic is value averaged over the value of the dicrotic wave as described above.

Another example (Use of the perfusion index), in particular as a duality of the signal

The Pl value is used as an indicator of the signal quality. Indeed, a Pl value of less than 0.1% indicates a bad photoplethysmography signal and makes it possible to indicate to the user a poor quality of estimation and the more frequent need for a calibration. Signal quality can generally be improved in these cases by correctly repositioning the sensor on the patient.

Integration in the estimate

The Pl gives information on the hemodynamic state in mean arterial pressure in the same way as the dicrotic wave and in a complementary manner. Indeed, Pl generally evolves in the opposite direction from mean arterial pressure.

We use a measure that we called mPI (for modified Pl) and which is calculated as follows: mPI = 10 x ln (1 / PI + 1). The mPI thus obtained varies in the same direction as the average arterial pressure is the behavior is found linearized compared to the exponential behavior of Pl.

A Calib calibration is carried out with the measured mean arterial pressure and the mPI at time 0 and the mean arterial pressure is calculated at time t using PAMest = Calib x mPI (t).

Using multiple parameters

Several parameters can be used (height of the dicrotic wave, mPI, durations indicated above).

We can regularly calculate a Calib for each parameter calculate the PAMest for each of the parameters define the PAMest by statistical mean (Kalman filter) by weighting these PAMest values and using the value calculated at the previous time

Example 3. Exemplification in real conditions

These results were obtained on the basis of the height of the dicrotic wave, similar results can be obtained with the other parameters.

A study was performed in the operating room for neurosurgery or during an interventional neuroradiology procedure. The patients benefited from the usual basic care for this type of intervention including:

monitoring

Non-invasive hemodynamic monitoring of blood pressure by oscillometry, as well as continuous monitoring of the ECG Continuous monitoring of pulsed oxygen saturation (SpO2) by photoplethysmography, as well as monitoring of expired CO2.

Monitoring the depth of anesthesia by the bispectral index (BIS)

Induction and maintenance of general anesthesia according to an intravenous concentration objective method (AIVOC) including propofol and remifentanil, and curarization before orotracheal intubation by atracurium besilate.

All monitors were connected to a Philips monitor, hypotension was defined as a drop in average blood pressure (MAP) of at least 20% compared to basal MAP.

When hypotension was noted, the anesthesiologist in charge of the patient was free to alleviate the anesthesia, administer a vascular filling or a vasoconstrictor (Ephedrine 9mg, Phenyphine 50mcg or Noradrenaline 10 mcg).

Experimental protocol

Phase 1: Preoxygenation (baseline) - Anesthetic induction.

• Pre-oxygenation for 2 min: It is during this phase and before any injection that the basic values of all the parameters are recorded (average of 2 values, baseline) • Remifentanil at 5 ng / ml of target concentration for 1 min.

• Propofol 5 pg / ml of target concentration.

• After the BIS has dropped below 50 and verification of the absence of ciliary reflex and the patient is ventilated manually: curarization with 0.5 mg / kg of tracrium.

• Wait 3 minutes with manual ventilation.

Phase 2: Laryngoscopy-Intubation-Manual ventilation.

• Direct laryngoscope.

• Orotracheal intubation.

• Manual ventilation and fixing of the probe.

Phase 3: Mechanical Ventilation-Anesthesia Maintenance.

• Connect the patient to the ventilator and start the mechanical ventilation.

• Lower targets for Remifentanil and Propofol to 3.5 ng / ml and 4 pg / ml respectively.

• Continuation of the collection for 3 min.

Phase 4: Correction Possible hypotension by vasoconstrictor.

• Hypotensive episode treated by administration of vasoconstrictor.

• Continuation of the collection one minute after effect of the vasoconstrictor.

• End of collection.

During phases 1, 2 and the start of 3, the cuff pressure was taken every minute for an estimated average duration of 15 min. The anesthesiologist in charge of the patient was free to deviate from the initial protocol at any time if he deemed that the clinical situation required it.

Data collection was carried out using Extrend (Ixellence) data capture software collecting signals at a frequency of 125Hz and all numerical values.

A point of collection of all the following parameters was carried out every minute:

- Height of the dicrotic wave: the calculation of the height of the dicrotic wave on the pletysmography signal.

- PI: the perfusion index was collected beat by beat.

Results:

patients were included in the study (55 years of median age, 32.7% male / 67.3% female).

of 61 patients have had at least one episode of hypotension. The incidence of hypotension, defined by a decrease in MAP> 20%, in our population was 88.5%). The time spent with a MAP <20% is on average 5.2 min during the induction, 44% of the time.

Evolution of the values during the entire induction.

The average duration of the entire induction phase was 12 ± 4 min.

Evolution P AM and height of the dicrotic wave

The variations in MAP and the variations in the height of the dicrotic wave were strongly correlated in a linear fashion over the entire duration of the induction (see FIG. 7, in particular the stability of the Calib value (FIG. 7.A)).

Analysis based on continuous average blood pressure measurement by invasive arterial catheterization.

The arterial catheter is a device allowing arterial access in order to measure blood pressure, invasively and continuously, and to take arterial blood samples.

Bloody or invasive blood pressure is an invasive technique for monitoring intravascular blood pressure through an arterial catheter.

A continuous measurement of the arterial pressure was carried out by arterial catheter and by the method according to the invention (calculation via the height of the dicrotic wave measured by plethysmography).

There was a perfect correlation of the variations in MAP measured by arterial catheter and by the method and this after the use of vasopressor drugs. The correlation was r = 0.88 with a 96% agreement between the variations obtained by the technique and the actual measurement of MAP (Figure 7.A).

In Figure 7.A, we can see in particular at the end a significant variation in pressure which is not detected by the cuff (due to the time between two pressure taps), but is by the PlethoMAP signal, and which therefore emphasizes the informative nature of the method using the dicrote wave.

In FIG. 7.B, it is clear that only the measurement of the height of the dicrotic wave makes it possible to obtain a value of the MAP, and that the evolution of the two other parameters (systolic value or diastolic value) does not is not informative enough. We can also see (T = 1.15h) that the drop in height of the dicrotic wave is indicative of the real drop in MAP (Figure 7.A), while the systolic and diastolic pressures do not vary. The combination of these two pressures would therefore not have been able to detect the drop in average blood pressure, and to have enabled the anesthetist to take any corrective action.

Claims (6)

1. Ex vivo method for evaluating the average arterial pressure of a patient, on the basis of values of a parameter calculated by plethysmography, comprising the steps: I. Calculate a Calib calibration value from at. From the value of the mean arterial pressure measured at a time tO b. Of the VpO value linked to said parameter, measured in the patient at time tO, II. Calculate the estimated value PAMest of the patient's blood pressure at a time t after tO by the formula PAMest = Calib x Vpf, in which Vpf is the value of the parameter measurement obtained at time t. 2. Ex vivo method according to claim 1, characterized in that the value Vpt of the parameter was obtained by plethysmography taken on the finger or the earlobe. 3. Ex vivo method according to claim 1 or 2, characterized in that the value Vpt of the parameter is chosen from the height of the dicrotic wave, the logarithm of the inverse of the perfusion index + 1, the duration between the start of the pulse wave and the dicrotic wave, the time between the start of the pulse wave and the maximum of the pulse wave, the ratio between the height of the systolic wave and the height of the pulse 'dicrote wave. 4. Ex vivo method according to one of claims 1 to 3, characterized in that the Vpt value used is a value averaged from several values measured over a predetermined period. 5. Ex vivo method according to one of claims 1 to 4, characterized in that the value Vpf of the parameter is the height of the dicrotic wave. 6. Ex vivo method for assessing the mean arterial pressure in a patient, characterized in that at. the method of one of claims 1 to 4 is repeated, for different parameters, in order to obtain several PAMest values at time t b. we calculate a final PAM value by statistical estimation taking into account i. different PAM values are calculated at time t ii. one or more final PAM values calculated before time t. Ex vivo method according to claim 6, characterized in that the final PAM value is calculated by Kalman filter in a discrete context. Computer product / program comprising program code instructions, recorded on a computer-readable medium, for implementing the steps of the method according to one of claims 1 to 7, when said program is executed on a computer. Computer-readable recording medium on which a computer program is recorded comprising program code instructions for executing the steps of the method according to one of claims 1 to 7. 1/5