WO2019122357A1

WO2019122357A1 - Device for the automated administering of a vasopressor and a vascular filling liquid

Info

Publication number: WO2019122357A1
Application number: PCT/EP2018/086648
Authority: WO
Inventors: Guillaume CHENEGROS; Nicolas LIBERT; Gilles CORDURIE; Ryad Benosman; Jacques DURANTEAU; Anatole HARROIS; Eric Vicaut
Original assignee: Universite Pierre Et Marie Curie (Paris 6); Centre National De La Recherche Scientifique (Cnrs); Institut National De La Sante Et De La Recherche Medicale (Inserm); Assistance Publique - Hopitaux De Paris; Universite Paris-Sud
Priority date: 2017-12-22
Filing date: 2018-12-21
Publication date: 2019-06-27
Also published as: FR3075652B1; FR3075652A1

Abstract

The invention relates to a device (5) for determining the quantities of filling liquid and vasopressor to be administered to a patient by means of respectively a first dispenser (3) and a second dispenser (4), characterised in that the device (5) is designed to determine a quantity of filling liquid and a quantity of vasopressor, on the basis of the previously measured blood pressure of a patient, such that said quantities depend on the ratio between quantities of filling liquid and vasopressor that have already been administered over a given period.

Description

AUTOMATED DELIVERY DEVICE FOR A VASOPRESSOR AND

A VASCULAR FILLING LIQUID

Field of the invention

The present invention relates to a device for administering a vasopressor and a vascular filling fluid, especially during a state of shock (septic or hemorrhagic) and severe head injury.

The invention also relates to a method for determining the amounts to be administered of a vasopressor and a vascular filling liquid.

State of the art

Serious injuries are a public health issue on a global scale. Resulting from aggression or accidents, they are the leading cause of death in young people.

These injuries associate tissue attrition lesions with significant bleeding. If the blood loss no longer meets the oxygen and nutrient requirements of the various organs, the body then reaches a state of hemorrhagic shock (it is a mismatch between intake and needs).

Hemorrhagic shock is the second leading cause of death due to severe trauma (40-50%), the first cause being catastrophic neurological damage to the head or spinal cord.

Also, currently the possibility of increasing survival by improving the management of hemorrhagic shock is in the foreground.

During a hemorrhagic shock, the initial care consists in controlling as soon as possible the origin of the bleeding. Initial resuscitation aims to maintain vital tissue oxygenation. It must be performed without delaying transport on a structure that can control lesions (surgery and interventional radiology). Resuscitation is guided by simple haemodynamic goals based on blood pressure. She rests on administration of vascular filling fluid more or less associated with vasopressor administration.

However, filling causes dilution of the blood, hypocalcemia (by dilution) and increases hypothermia. These deleterious effects will aggravate coagulopathy and increase bleeding.

If the filling is too abundant and normalizes hemodynamics, the increase in perfusion pressure may increase blood loss through vascular breccias.

The current strategy is therefore to limit vascular filling, dilution and the level of blood pressure ("permissive hypotension"). The European recommendations, recommend an objective of 80 to 90mmHg of systolic blood pressure (80mmHg of average arterial pressure in case of severe head trauma).

To maintain blood pressure without resorting to heavy filling, it is recommended to use vasopressors, such as norepinephrine.

However, if the vasopressor dosage is too high, there is a risk of increasing vasoconstriction and aggravating tissue hypoperfusion.

Although vasopressors are used systematically, there is no recommendation on the dosages to be used nor the timing of their implementation. In practice, physicians attempt to achieve a compromise between the volume of fluid filled and the dose of vasopressor administered. This compromise is currently based solely on the feelings of the doctor and his personal experience.

Several devices are known to control the administration of products during a hemorrhagic shock.

WO 2012/097129 discloses a hemodynamic control system of a patient for automatic and standardized administration of fluids, blood products, or drugs to the patient to assist physicians.

The system described in this document comprises physiological vital signal sensors that are analyzed and compared to standard predictive data and / or analyzed according to the reaction following the administration of the system. fluid by the pumps, to send instructions to the pumps to deliver fluids and / or drugs to patients at a fixed rate and time. The control of the administration of the products to the patient is thus ensured by closed-loop servocontrol.

In addition, a terminal makes it possible to measure the volume of vascular filling administered and the cardiac output can be used to determine the commands to be sent to the pumps.

However, the solution described in WO 2012/097129 does not provide a sufficiently adapted response to the patient's condition.

General presentation of the invention

An object of the invention is to allow to better adapt the amount of filling liquid and vasopressor depending on the patient's condition. A filling liquid is, for example, 0.9% NaCl crystalloid solute.

Another object of the invention is to propose a solution to assist a physician during the resuscitation of a patient by a joint determination of the amounts of liquid filling and vasopressor to administer to reduce the number of actions that must realize the doctor during said resuscitation, the amounts to be administered to be regulated stably.

For this purpose, according to a first aspect of the invention, there is provided a device for determining the amounts of filling liquid and vasopressor intended to be administered to a patient by means of a first dispenser and a second dispenser respectively. , characterized in that the device is configured for, from a blood pressure previously measured on a patient, determining a quantity of filling liquid and a quantity of vasopressor such that said amounts are functions of the ratio between the amounts of liquid of filling and vasopressor already administered over a given period.

The device according to the invention is advantageously completed by the following features, taken alone or in any of their technically possible combinations: the amounts to be administered of filling liquid and of vasopressor are determined according to several phases according to which the rate of administration of the filling liquid or of the vasopressor is fixed and the ratio is modified.

the amounts of filling fluid and vasopressor to be administered are determined according to several phases:

- first phase: the amount of vasopressor is constant and is determined according to the blood pressure and the amount of filling liquid is determined so that the measured blood pressure reaches a target value, the first phase stops either when the arterial pressure reaches the target arterial pressure value, the administration then goes into a phase of regulation of the vasopressor, ie when the quantity of administered filling liquid reaches a first threshold value, the administration then proceeds to a second phase;

second phase: the determination of the amounts of vasopressor and filling liquid is such that the value of arterial pressure is the target value, the second phase stops either when the arterial pressure is equal to the target arterial pressure value and the the amount of vasopressor determined is less than a first threshold, the administration then passes into the regulation phase of the vasopressor, ie when the quantity of administered filling liquid reaches a second threshold value greater than the first threshold value, the administration then passes in a third phase;

- third phase: the amount of filling fluid is zero and the amount of vasopressor is determined so that the blood pressure reaches the target blood pressure value, the third phase stops either when the blood pressure reaches the blood pressure value target and the amount of vasopressor is less than the first threshold, the administration then goes into the vasopressor regulation phase, ie when the vasopressor rate is greater than the first threshold, the administration then goes into a fourth phase;

fourth stage: the amounts of vasopressor and filling liquid are determined so that the value of the arterial pressure reaches the target value, the fourth phase stops either when the arterial pressure reaches the target blood pressure value and the flow rate of the blood pressure. vasopressor is below the first threshold, the administration then goes into the regulation phase of the vasopressor, ie when the amount of filling liquid determined reaches a third threshold value greater than the second threshold value, the administration then goes into a fifth phase. ;

- fifth phase: the amount of filling fluid is zero the amount of vasopressor is determined so that the blood pressure reaches the target blood pressure value, the fifth phase stops when the blood pressure reaches the target blood pressure value and the the amount of vasopressor is less than a second threshold greater than the first threshold, the administration then passes into the regulation phase of the vasopressor, ie when the arterial pressure has not reached the target arterial pressure value after a predetermined duration, the administration then goes into a sixth phase;

sixth phase: the amount of vasopressor is constant according to a predetermined value and the amount of filling liquid is determined so that the arterial pressure reaches the target arterial pressure value, the sixth phase stops only when the arterial pressure reaches the value target blood pressure, the administration then goes into the regulation phase of the vasopressor;

- Regulatory phase of the vasopressor: the amount of filling fluid is zero and the amount of vasopressor is determined so that the arterial pressure reaches the value of target blood pressure, the regulation phase of the vasopressor stops when the blood pressure is below a threshold blood pressure value, in this case determine a quantity of filling liquid if the amount of filling liquid administered is lower than the threshold value of the phase in which the administration before the vasopressor regulation phase, or proceed to the fifth phase if the amount of vasopressor is less than a predetermined maximum value or otherwise pass into the sixth phase;

the amount of vasopressor is determined so as not to increase when a lactate level of the patient reaches a threshold level. In particular, it is determined not to increase in case of negative or no clearance. Clearance is assessed every 30 minutes. It is noted that the clearance is the decrease in the lactate concentration;

the amount of filling liquid and / or the vasopressor is determined by closed-loop control of the PID type relative to the arterial pressure;

the amount of filling liquid and / or the vasopressor is determined by a closed-loop control with a fuzzy logic type algorithm with respect to the arterial pressure.

According to a second aspect, the invention relates to a system for delivering to a patient a filling liquid and a vasopressor comprising: an arterial pressure sensor configured to measure the patient's blood pressure; a first dispenser configured to administer the filling fluid to the patient; a second dispenser configured to deliver the vasopressor to the patient; a control device according to the first aspect of the invention.

According to a third aspect, the invention relates to a computer program product comprising program code instructions for performing the steps of a method of operation of the device of the first aspect of the invention, said method comprising a step determining a rate of administration of the filling fluid and the vasopressor from an arterial pressure of said patient while maintaining a determined ratio defined by a quantity of filling liquid already administered on a quantity of vasopressor already administered, when said program is run on a computer.

According to a fourth aspect, the invention relates to a system (5) for controlling the administration to a patient of a filling liquid and a vasopressor by means of a first dispenser (3) and a second dispenser respectively. distributor, characterized in that the device (5) control comprises a device according to the first aspect of the invention.

According to a fifth aspect, the invention relates to a device for use in the treatment of septic shock states or vasoplegic shock stages, said treatment being such that it comprises administering to a patient a filling liquid and a A vasopressor using a system according to the fourth aspect of the invention.

According to the invention, the stability of the regulation is a function of a ratio between a quantity of filling liquid already administered during a given period of time on a vasopressor mass flow already administered.

This ratio differs according to the states (hemorrhagic shock, septic shock, hypovolemic shock, severe sepsis, hypotension, perioperative period, ARDS, severe head trauma ...). At each cumulative volume per unit vascular filling time (RV) will correspond to a maximum norepinephrine (NA) plateau. Conversely when one reaches a threshold of vasopressor (ceiling) one increases the filling volume possible. And so on. We will have a new maximum filling volume and when it is reached we will have a new ceiling of vasopressor up to have a maximum ceiling of norepinephrine. This approach of the ratio of "vascular filling / vasopressor dose" or RV / NA will have different values according to the pathologies and the situations, in particular according to the phase of patient management (pre-hospital, early, late,. ..), depending on the duration of treatments already administered and the vascular filling strategies (liberal or restrictive).

As an example in hemorrhagic shock, in a restrictive pre-hospital approach, the following ratios can be used:

20 ml / kg RV for 0.85 μg / kg / min at 1 μg / kg / min maximum NA;

30ml_ / kg of RV for 2pg / kg / min at 4pg / kg / min of maximum NA. The goal is to have a strategy of saving vascular filling without increasing organ failures or sacrificing a perfusion territory.

This ratio corresponds to a caricatural approach of the ratio of a marker of the pre-load dependency type volemia to the dose of vasoconstrictor administered or vascular resistance. This takes up the notion of "dry or wet slope" literature. Different markers of blood volume or systemic vascular resistance can be used.

At least three situations can be distinguished:

- Absence of monitoring: The regulation is only performed on blood pressure (studies provided). We have an RV / NA "ratio" defined according to the resuscitation strategy chosen a priori with thresholds of maximum volumes per unit of time and mass flow rate of NA.

- Absence of monitoring and discontinuous or continuous measurement of lactate.

In this case, the lactate and its clearance will modify these thresholds: for example to increase the filling for a given level of NA max or to decrease the level of NA max so as to be able to administer vascular filling.

Presence of a measure of invasive blood pressure: integration of markers predicting the response to vascular filling. For this we analyze the heart rate, then use markers of pre-load dependence if ventilation is controlled). The measurement of arterial elastance is also included to guide the regulation of NA. The clearance of lactate is always taken into account and will further refine the decision: if the evolution is favorable, there is no therapeutic modification, conversely if the clearance is bad, the thresholds of RV (or pre- load dependency) or titration of NA are changed. Lactate and its clearance are also used to redefine the therapeutic target. Thus the target blood pressure level is changed. Presence of hemodynamic monitoring: each component of the monitoring is taken into account. Here again, lactate and lactate clearance are also used to redefine the therapeutic target (target pressure level). In case of insufficient lactate clearance and weighting of the maximum filling thresholds and NA, the blood pressure target will be increased.

Presentation of figures

Other features, objects and advantages of the invention will appear on reading the following description of various embodiments shown in the following drawings:

FIG. 1 represents a device for resuscitating a patient during a hemorrhagic shock for the control of the administration of a filling liquid and a vasopressor;

FIG. 2 represents a schematic representation of the different phases of the administration of the filling liquid and the vasopressor according to a possible variant;

FIG. 3 represents a schematic representation of a method of administering the filling liquid and the vasopressor;

FIG. 4 represents the method for evaluating the performances of the invention in rats;

FIG. 5 represents the performances of the different groups of rats; FIG. 6 represents the quantity of filling fluid administered, the hematocrit and lactate level at the end of the resuscitation for the different groups of rats;

FIG. 7 presents the method for evaluating the performance of the invention on pigs;

- Figure 8 shows the cardiac output, arterial oxygen transport, vascular resistance and volume for different groups of pigs.

Description of the invention

Control system As shown in FIG. 1, a system 1 for controlling the administration of a filling liquid and a vasopressor for the resuscitation of a patient during a hemorrhagic shock comprises:

a blood pressure sensor 2 which is configured to measure the patient's blood pressure;

a first dispenser 3 which is configured to administer a filling liquid to the patient;

a second dispenser 4 which is configured to administer a vasopressor to the patient;

a device 5 for controlling the administration of the filling liquid and the vasopressor which is connected to the arterial pressure sensor 2, to the first distributor 3 and to the second distributor 4, typically wired wirelessly (for example Wifi).

The device 5 is configured to control the first dispenser 3 and the second dispenser 4 by means of the data measured by the blood pressure sensor 2, and thus regulate the administration of the filling liquid and the vasopressor so as to reach the target blood pressure. desired.

The device 5 comprises a processor 51 and a memory 52 for implementing a method of operating the device which makes it possible to regulate the rate of administration of the filling liquid and the vasopressor.

The device 5 may for example be a computer, or a tablet, or a PCB SB Rio 9626 National Instrument®.

Furthermore, the system 1 may comprise a screen to allow the display of data on the regulation of the administration of the filling liquid and the vasopressor, such as, for example, the current flow rate of the vasopressor and the filling liquid as well as the quantities administered. The vital constants of the patient can also be displayed on this screen.

The blood pressure sensor 2 may be a non-invasive blood pressure monitor, which has the advantage of not necessarily being implanted in the patient, but which does not allow the blood pressure to be measured continuously. So, the For example, blood pressure sensor 2 may be any type of inflatable pressure cuff of the PNI type (non-invasive pressure).

The blood pressure sensor 2 may also be an invasive tensiometer that is inserted into a patient's artery through a catheter. Such a sensor offers the advantage of measuring the blood pressure of the patient almost continuously, but nevertheless requires implantation via a catheter into the patient's artery. Thus, the blood pressure sensor 2 may for example be all types of invasive pressure sensors, for example those of Edwards® and Pulsion® brands.

The first distributor 4 for the administration of the filling liquid is typically a filling pump, such as for example a filling pump of the Flowpump model of Gamidatech®.

The filling liquid may for example be crystalloids, which offer the advantage of being inexpensive, or colloids, which offer the advantage of having a higher volume expansion capacity.

The second dispenser for administering the vasopressor is typically a syringe pump, such as Carefusion® GH models.

The vasopressor may for example be norepinephrine, phenylephrine, vasopressin, or selepressin.

The control device is configured to determine a rate of delivery of the filling fluid and vasopressor from an arterial pressure of said patient.

A target value to be reached for the blood pressure is defined by the user, and the device 5 regulates by a closed-loop control on the one hand the rate of administration of the filling liquid by the first distributor 3, and other the rate of administration of the vasopressor by the second dispenser 4, so that the blood pressure of the patient reaches said defined target value.

In addition, the device 5 is configured to regulate the delivery rate of the filling liquid and the vasopressor while maintaining a ratio determined by a quantity of filling liquid already administered on a quantity of vasopressor already administered.

The use of such a ratio makes it possible to obtain a stable control system, while ensuring a closed-loop servocontrol of the patient's arterial pressure with two substances that impact said patient's blood pressure.

Furthermore, the device 1 is configured so that the administration of the filling liquid and the vasopressor is divided into different phases in which the rate of administration of the filling liquid or vasopressor is adapted, the ratio being different at different phases. .

The fact of defining different phases in the administration of the filling liquid and the vasopressor makes it possible to ensure administration as close as possible to the clinical state of the patient. Indeed, a drop in blood pressure at the beginning of the resuscitation and a drop in blood pressure after one hour of resuscitation should not be treated in the same way. However, with a simple closed-loop control, these two drops in blood pressure are treated in the same way because the regulation does not take into account the pathophysiological evolution of the resuscitation.

The distribution in phase makes it possible to privilege therapeutic axes according to the physiological state and failures of the various components of the cardiovascular system as it is the case in clinical practice and thus to better choose the therapeutic response according to the state:

Pre-dependency

pre-load dependence and vasoplegia

pre-dependence, vasoplegia and heart failure

vasoplegia

vasoplegia and heart failure

overall heart failure

Diastolic left heart failure

left heart failure

right heart failure

loss of hemodynamic coherence (persistent microcirculatory failure despite correction of macro-hemodynamics. In practice, here it is a special case of preloading dependence (filling), correction of vascular resistances.

Can also be considered in these phases or to define the maximum doses of therapeutics (and therefore ratios): lactate and lactate clearance; the arteriovenous CO2 delta; hemoglobin concentration; arterial pH and venous pH; temperature, pulse pressure delta, preload addiction to treat hemorrhagic shock; septic shock; cardiogenic shock; anaphylactic shock; or other shock such as acute respiratory distress syndrome (ARDS) or acute renal failure (ARF).

According to a possible variant illustrated in FIG. 2, the phases may be the following:

- phase 1: administer the vasopressor with a constant flow determined according to the arterial pressure P _art and administer the filling liquid with a regulated flow to reach a target blood pressure value P _dbie , phase 1 stops either when the blood pressure P _art reaches the target blood pressure value P _dbie , the administration of the filling liquid and the vasopressor then passes into the vasopressor regulation phase, ie when the quantity of liquid filling liquid administered Qiiquide reaches a first threshold value S1 liquid, administration of the filling fluid and vasopressor then passes into phase 2;

- Phase 2: administering vasopressor and the filling liquid with each of a flow rate to achieve the target blood pressure value Pdbie, phase 2 stops either when the amount of filling liquid administered QiiQ _fluid reaches a second threshold value S2iiq _fluid, with the second threshold value S2 nq _fluid greater than the first threshold value S1 _liquid, the administration then passes in the Phase 3, when the blood pressure reaches the target blood pressure value P _DBIE and vasopressor D _will flow so is less than a first threshold S1 vaso, the administration then goes into the regulation phase of the vasopressor;

- Phase 3: Stop administer liquid fill and administer the vasopressor with a controlled rate so that the blood pressure P _s reaches the target blood pressure value Pdbie, Phase 3 stops either when the blood pressure P _has reached the target value P _dbie and vasopressor flow D _vaso is below the first threshold S1 _Vaso , the administration of the filling liquid and the vasopressor then goes into the regulation phase of the vasopressor, or when the flow rate vasopressor D _vaso is greater than the first threshold S1 _Vaso , the administration then goes into phase 4;

- Phase 4: administering vasopressor and the filling liquid with each of a flow rate for the blood pressure P reaches the _art blood pressure target value P _DBIE, phase 4 stops or when the amount of filling liquid administered QiiQ _fluid reaches a third value S 3 liquid threshold with the third value _quide S3n threshold greater than the second threshold value S2iiq _fluid, the administration then passes into phase 5, or when the blood pressure P _s reaches the target value P _DBIE and the vasopressor flow D _vaSo is below the first threshold S1 _Vaso , the administration then goes into the regulation phase of the vasopressor;

- Phase 5: Stop administer liquid fill and administer the vasopressor with a controlled rate so that the blood pressure P reaches the _art blood pressure target value P _DBIE, the fifth phase stops when the blood pressure P reaches _art the target blood pressure value P _dbie and the vasopressor flow D _vaSo is less than a second threshold S2 _vaSo , the second threshold S2 _vaSo being greater than the first threshold S1 _Vaso , the administration then goes into the regulation phase of the vasopressor, or when the blood pressure P _art has not reached the target blood pressure value P _dbie after a predetermined time T, the administration then goes into a sixth phase;

- sixth phase: administering vasopressor with a predetermined constant rate and administer the filling liquid with a controlled rate so that the blood pressure P reaches the _art blood pressure target value P _DBIE, the sixth phase stops only when the blood pressure The _art reaches the target blood pressure value P _dbie , the administration then goes into the regulation phase of the vasopressor;

- regulation phase of the vasopressor: Stop administer liquid fill and administer the vasopressor with a controlled rate so that the blood pressure P reaches the _art blood pressure target value P _DBIE, the vasopressor the regulation phase stops when blood pressure P _art is in below a blood pressure threshold value P _seuii, in this case administering fill liquid if the amount of filling liquid Qiiquide administered is less than the value S1 liquid threshold S2ii _quide or S3ii _quide which phase was administration before the regulation phase of the vasopressor, or go to phase 5 if the vasopressor flow D _vaso is less than a predetermined maximum value D _vasomax , or otherwise go to phase 6.

According to one possible example, the initialization rate of the vasopressor in step 1 determined as a function of the patient's blood pressure may be as follows: 0.25 μg / kg / min if the arterial pressure is less than 50 mmHg; 0.2 μg / kg / min if the blood pressure is between 50 mmHg and 70 mmHg; 0.1 μg / kg / min if the blood pressure is above 70 mmHg.

In one possible example, the target blood pressure P _dbie is 83 mmHg, and the threshold blood pressure value is 70 mmHg.

In one possible example, the amount of filling liquid administered QiiQ _fluid, the first value _liquid S1 threshold is 10 ml / kg, the second threshold value S2ii _quide is 20 ml / kg, and the third threshold value S3ii _quide is of 30 ml / kg.

According to a possible example, for vasopressor flow D _vaso , the first threshold S1 _vaso is 1 μg / kg / min, and the second threshold S2 _vaso and the predetermined maximum value D _vasomax are 4 μg / kg / min.

In addition, as illustrated in FIG. 1, the device 1 may also comprise a lactate sensor 6 which is configured to measure the arterial or venous concentration of lactate circulating in the patient during resuscitation, in order to be able to estimate the metabolic suffering caused. by the administration of the vasopressor.

The lactate sensor 6 emits a lactate measurement signal which is acquired by the device 5 to be taken into account for regulating the administration of the filling liquid and the vasopressor.

The lactate sensor 6 can for example measure continuously or discontinuously and can be measured by a device and the value entered manually in the system (a type of sensor is a Getinge® Eirus monitor). Thus, the device 1 is configured to regulate the administration of the vasopressor by preventing an increase in the vasopressor delivery rate when the clearance of lactate measured exceeds a predetermined threshold level. Conversely, if the pulsating pressure is low (less than 25% of the PAS), the maximum filling level will be increased by 20%.

This threshold level may be for example for hemorrhagic shock:

- Prior to surgical control: the maximum vasopressor flow rate will be decreased if lactate clearance is less than 20% per hour or 10% per 30 minutes;

- after the surgical control: the maximum vasopressor rate (D _va somax) will also be decreased if the lactate clearance is less than 20% after one hour (in the absence of vascular monitoring). The clearance of lactate will then guide the treatment according to the available monitoring.

In the presence of an invasive measure of blood pressure, the ratio RV / NA can therefore be corrected by the clearance of lactate as defined above (out of phase of uncontrolled acute hemorrhage).

The ratio RV / NA can also be corrected by taking into account the parameters of pre-charge dependence as the variation of pulsed pressure (delta PP) induced by the mechanical ventilation. For a ventilated patient, with a measure of invasive blood pressure and the conditions of use of delta PP:

if the delta PP is greater than 15%: the administered volume of the filling liquid is increased per unit of time or the maximum vasopressor flow rate is decreased according to the phase in which the algorithm is located.

If the PP delta is less than 9%: the target blood pressure is increased and the measurement of hemoglobin and evaluation of cardiac function are recommended.

If the delta PP is between 9 and 15%: a filling test of 3mL / kg is made. If the objective is not reached after two filling tests and the delta PP remains between 9 and 15%: the target of blood pressure is increased. In case of hemodynamic monitoring, the RV / NA ratio can also be corrected according to the hemodynamic parameters (cardiac output, vascular resistance, preload dependence or any other derived parameter). In all cases an alert is sent for medical re-evaluation.

Moreover, the control of the rate of administration of the filling liquid and / or the vasopressor can be achieved by a closed-loop servocontrol of the PID (Proportional Integral Derivative) type with respect to the arterial pressure.

One of the coefficients for the PID-type servo-control may be zero. Thus, for example, the closed-loop servocontrol can have a zero coefficient for the Derivative control, the servocontrol being thus servocontrolled PI (for Proportional Integral).

The use of a PID servo-control offers the advantage of being able to calculate the flow rate of filling liquid and / or vasopressor in response to the change in blood pressure with a discontinuous measurement signal taken at regular intervals. Such a discontinuous blood pressure measurement is typically performed with the non-invasive blood pressure monitors that are currently used in the pre-hospital setting. Thus, it is not necessary to use an invasive tensiometer requiring to be installed in a catheter.

According to one possible variant, the rate of administration of the filling liquid and / or the vasopressor is regulated by closed-loop control with a fuzzy logic type algorithm with respect to the arterial pressure.

The use of fuzzy logic algorithm servocontrol offers the advantage of being closer to human logic. Such a measurement of continuous or near-continuous blood pressure can be made with invasive blood pressure monitors installed in a catheter which are currently used in the hospital environment.

Closed-loop PID or fuzzy logic feedback uses the following data: patient's blood pressure, time, and difference between the patient's blood pressure and the target blood pressure (error from the set point).

According to one possible variant, the rate of administration of the filling liquid is regulated by a PID-type closed-loop control, and the flow rate of the vasopressor is regulated by a closed-loop control with a fuzzy logic algorithm.

Control process

As illustrated in FIG. 3, the administration of the filling liquid and the vasopressor for the resuscitation of a patient during a hemorrhagic shock is carried out with the following steps:

- E1: measure the patient's blood pressure;

E2: determine the rate of administration of the filling liquid and the vasopressor from a blood pressure of said patient while maintaining a determined ratio defined by a quantity of liquid already administered on a quantity of vasopressor already administered, when said program is executed on a computer, processor or microcontroller.

Advantageously, the correction coefficients used in the algorithm are valid for a time delta of 1 min between two therapeutic modifications.

The method is implemented on the one hand by the arterial pressure sensor 2 which implements step E1 by measuring the arterial pressure of the patient, and on the other hand by the control device 5 which implements the step E2 by controlling the first distributor 3 and the second distributor 4.

Examples of use

Rodent model:

The performance of the developed algorithms was evaluated on a rodent model.

To do this, the model used is a shock model controlled by spoliation with maintaining a fixed pressure (Wiggers model). The protocol has been validated by the Ethics Committee. The animals used were Wistar rats aged 12-16 weeks.

The performance was evaluated by providing resuscitation of the rats by administration of filling fluid and / or combination treatment filling and vasopressor by different methods, each method of administration thus forming a group of rats.

The evaluation method is illustrated in Figure 4.

The vasopressor used in this evaluation was norepinephrine

(N / A).

Three groups of rats received vascular filling:

- a first group resuscitated by a manual vascular filling by a resuscitator (RV Manu: administration only of filling liquid manually);

a second group resuscitated by automatic PID-controlled vascular filling (RV Auto PID: administration only of closed-loop closed-loop filling fluid of the PID type);

a third group resuscitated by an automated vascular filling by slaving with a fuzzy logic algorithm (RV Auto FL: only administration of automated filling fluid by closed-loop servocontrol by a fuzzy logic algorithm);

Two groups of rats were resuscitated with a combination of vascular filling and norepinephrine treatment:

the fourth group was resuscitated by vascular filling and noradrenaline administered manually (RVNA Manu: manual administration of filling liquid and norepinephrine);

the fifth group was resuscitated by vascular filling and norepinephrine administered in an automated manner (RVNA Auto: automated administration of filling and vasopressor fluid, PID-type closed-loop servocontrol for the filling liquid and fuzzy logic for the vasopressor).

Induction of anesthesia was performed by inhalation (isoflurane tank). Maintenance of the anesthesia was performed by a mixture of Ketamine and Xylazine (respectively 90mg / kg and 5mg / kg). Additional analgesia was provided by the subcutaneous injection of buprenorphine at 50 μg / kg.

After anesthesia, a tracheotomy cannula was put in place to free the airways.

An arterial catheter (PE ED 1, 16mm) was inserted into the carotid artery to allow spoliation and to measure pressure during resuscitation.

A venous catheter (PE ED 0.92mm) was inserted into the jugular to allow vascular filling. An arterial catheter (PE ED 0.69mm) was inserted into the femoral artery to measure blood pressure continuously during spoliation.

During resuscitation, the pressure sensor was connected to the carotid catheter. In groups with manual resuscitation, the pressure sensor was connected to the Biopac® MP30 monitor. In the automated resuscitation groups, the sensor was connected to the automated resuscitation device and a second sensor was connected from the Biopac® MP monitor to the femoral catheter to record blood pressure.

Haemorrhagic shock was induced by blood spoliation in two phases. The first phase rapidly decreased the level of the average arterial pressure to 30mmHg in 5 minutes. The second phase maintained mean arterial pressure between 30 and 35 mmHg for 60 minutes by spoliating or re-transfusing the necessary blood. The total spoliation time was therefore 65min.

Vascular filling was performed using a GH Carefusion® syringe pump used as a filling pump.

In groups with manual filling, the infusion rate was 2mL.kg-1.min-1.

Norepinephrine was infused with a Harvard pump 33 (Harvard Apparatus®) syringe pump.

In the 4 ^th group, the group wherein the manual administration of the filling liquid is associated with norepinephrine, the filling liquid and norepinephrine were administered according to a protocol standardized method that takes into account an indexation of the noradrenaline flow rate on the volume of vascular filling fluid. The administration of norepinephrine and the administration of liquid fillers were initiated synchronously. The increases in the administration rates have been alternated and regulated. The administration of the products was administered manually depending on the pressure level, on / off mode for the filling liquid, and through flow modification for norepinephrine.

In the automated groups, the only interventions required from the user were the notification of the weight of the animal and click on the icon of launch of the resuscitation.

For all the groups, the resuscitation lasted one hour.

A biological evaluation was performed at the end of the spoliation and at the end of resuscitation. Arterial lactate was measured by the device THE EDGE of ApexBio®. The blood gases were measured by a RAPIDLab348EX SIEMENS Healthcare® monitor. The hematocrit was calculated after centrifugation of a blood sample by the ratio of red blood cells to whole blood.

At the end of the experiment, the animals were killed by pentobarbital sodium overdose.

The normality of the variables was assessed by the Agostino-Pearson test. Some variables had a non-normal distribution and non-parametric tests were used. The data are presented in the median-interquartile form.

¹ group, the group RV with vascular Manu manual filling was used as a control group to evaluate the performance of the 2 ^nd and 3 ^rd groups with automated blood volume (RV RV Auto Auto PID and FL).

4 ^th group, the group with vascular RVNA Manu filling and noradrenaline manual, was used as a control group to evaluate the performance of the ^5th group, the group RVNA Auto with automated filling and norepinephrine.

Blood pressure measurement was performed every 200 ms. During the time of resuscitation, the performance of the treatment was evaluated firstly, the time spent in the target zone (the time during which the blood pressure was equal to the target arterial pressure) and secondly the performance error PE with respect to the target systolic blood pressure .

The PE performance error was calculated as follows (with PAS for systolic blood pressure):

with: PEÿ which corresponds to the performance error between the measurement points i and j;

PASmesuréei _j which corresponds to the blood pressure measured at the measurement points.

PAScible, which corresponds to the target blood pressure at the point of measurement.

Other performance evaluation parameters were used. These other parameters include the median error (MDPE), the absolute value of the median error (MDAPE), the variability (WOBBLE), the divergence (DIV), and the overall score (GS).

These parameters were calculated as follows:

MDPE (median error):

MDAPE (absolute value of the median error):

WOBBLE (variability):

Divergence:

GS (overall score): (M DAPE _f + WOOBLE)

CS -

Target time

The time spent in the target zone, the performance variables, the volume of vascular filling and the number of interventions required for resuscitation were firstly compared with the Kruskal Wallis test between the 1 ^st , 2 ^nd and 3 ^rd groups (groups with only administration of filling liquid), and secondly compared with the Mann-Whitney test for the 4 ^th and 5 ^th groups which associate administration of filling liquid and norepinephrine.

In comparisons between the 2 ^nd and 3 ^rd groups, if significance test, multiple comparisons relative to group ¹ were carried out and adjusted by the Dunn test.

The effects of resuscitation on blood gases, lactate, and hematocrit were assessed by nonparametric 2-way analysis of variance (or non-parametric ANOVA) for repeated measures.

Given the performance of the groups, the effects of vasopressor were sought by comparing the 2 ^nd and 5 ^th Groups (PID Auto RV and Auto RVNA) by Mann-Whitney.

For the test, 52 rats were included. The median weight was 348 g (318-373). Baseline MAP was 89 mmHg (85-99). During hemorrhagic period, the rats were deprived of 24.8 mL.kg ¹ (23.4 to 25.8). Abstraction resulted in a moderate metabolic acidosis with pH 7.34 (7.31 -7.37), bicarbonate concentration 22.3 mmol.L ¹ (21, 2 to 23.7) and an excess base - 2.6 mmol.L ¹ (-4--0.8). Lactate was raised to 5.6 mmol.L ¹ (4, 6 to 7.1).

The results of the evaluation test on rats are illustrated in FIG. 5. FIG. 6 illustrates the amount of filling liquid administered during resuscitation, hematocrit, and lactate at the end of resuscitation for the different groups. of rats.

Three rats died during resuscitation in the ^1st group (Administration Manual liquid filling) and were excluded from the study. As illustrated in Figure 5, the rats of ¹ group spent 67.3% (59- 74.1) resuscitation time in the target area. The rats spent 26.5% (21.7-32) of the time below the target and 3.5% (0.5-13.2) of the time below.

The Global Score (GS) is 10.1 (8.6-14.5).

The performance of resuscitation is reported in Table 1.

On average for each rat, 28 manual interventions (23-40) were required to perform resuscitation for 1 hour.

No rats died during the resuscitation of the 2 ^nd and 3 ^rd groups.

As shown in Figure 5, rats in the 2 ^nd group spent 70.9% (55-76.4) of time in the target area, 25% (22.8-38) below and 0% (0- 2.1) above. The rats in the 3 ^rd group spent 58.8% (35.7 to 61, 3) time in the target area, 38.7% (25.7 to 49.7) and below 2.7% (0 -20,2) above.

Global Score (GS) 2 ^nd group was 9.9 (7.3 to 19.2) and the Overall Score (GS) of the ^3rd group was 14.7 (13.5 to 52.6) . The Global Score (GS) is significantly different between groups (p = 0.037).

The absolute value of the median error (MDAPE) is higher in the 3 ^rd group.

Only 2 interventions were required in the 2nd and 3rd groups (p <0.0001 for each group vs control).

Table 1 - Performance of resuscitation by filling

* p <0.05 vs manual control

In the control group (group ^1), during the resuscitation period of one hour, 33.6 mL.kg ¹ (from 22.2 to 41, 1) for filling liquid were required to maintain hemodynamic objectives .

Resuscitation corrected acidosis with a pH of 7.38 (7.32-7.44), improved lactate to 3.9 mmol.L ¹ (3, 3-5, 6) and decreased hematocrit at 27.5% (26-30.3).

The rats in the 2 ^nd group (PID control) received 25.5 mL.kg ¹ (23,9- 45,2) filling liquid while those of the 3 ^rd group (fuzzy servo) received 25.7 mL .kg ¹ (19.5-40.8) of filling liquid.

Resuscitation normalized to pH 7.4 (7.35-7.42) in the 2 ^nd group and 7.38 (7.36-7.41) in the 3 ^rd group.

Filling also decreased lactate with no difference between groups.

As shown in Figure 5, the rats of the ^4th group (manual administration filling liquid and vasopressor) spent 49.6% (37,6- 64,9) time in target zone, 38.4% ( 29.3-59) below and 3.8% (2.2-16.2) above (see also Table 2).

Global Score was 19.1 (10.8 to 42) for the rats of 4 ^th group.

The rats in the 5 ^th group (administration automated filling liquid and vasopressor) spent 74.1% (55.8 to 81, 9) of time in target zone (higher than the 4 ^th group, p = 0.0217) , 14.8% (11 to 25.8) of the time below (lower than the 4 ^th group, p = 0.0037) and 5.6% (3.2 to 17) of the time above.

Global Score was 9.8 (6.6 to 19.3) for the rats of the 5 ^th group.

The number of interventions is significantly lower in the automated group (p <0.0001).

Table 2 - Performance of resuscitation by filling and vasopressor

No rat of the 4 ^th and 5 ^th Groups has died during resuscitation.

During resuscitation hour, the rats of the ^4th group received 19.9 mL.kg ¹ (from 13.9 to 21, 6) for filling fluid. Resuscitation increased pH 5 to 7.38 (7.33-7.4) and decreased lactate to 4.1 mmol.L ¹ (2, 8-6, 6). At the end of resuscitation the hematocrit was 37.3% (31-40).

In the 5 ^th group, the rats received 17.7 mL.kg ¹ (10-29.9) of filling liquid. Resuscitation normalized the pH to 7.41 (7.38-7.46) and decreased the lactate 3.3 mmol.L ¹ (2, 8-4, 3). At the end of resuscitation, the hematocrit decreased

36.1% (35-37).

In terms of performance, the only difference observed between the 2 ^nd and 5 ^th groups is the time spent above the target zone which was higher in the 5 ^th group (p <0.05).

The volume required for resuscitation is much higher in the

2 ^nd group (p <0.001). This is associated with lower dilution in the 5 ^th group with higher hematocrit (p = 0.011). This saving of vascular fluid did not result in more metabolic dysfunction (see Figure 6 and Table 3).

20

Table 3 - Biological effects of spoliation and resuscitation: blood gas, lactate and hematocrit.

This small animal study showed the feasibility

Automated guided resuscitation on Blood Blood Pressure (PAS) in a model of severe haemorrhagic shock. The performance of automated resuscitation is excellent. It makes it possible to maintain the PAS in the target area over time in a manner comparable to the optimized manual filling with a number of interventions that would probably not be feasible in current practice (this required a staff entirely dedicated to this task which provided surveillance permanent level of PAS and intervened in real time on the filling pump and / or the administration of norepinephrine).

The performance of the automated system is achieved with a minimal number of interventions, boiling down to two clicks (start and stop resuscitation). The performance of the automated combined treatment is superior to the manual treatment with optimal support in the target pressure zone.

The administration of a vasopressor (here norepinephrine) allowed a reduction in the amount of filling liquid administered and resulted in less dilution. This vascular filling savings did not alter the correction of metabolic dysfunctions.

In addition, this reduction in the amount of fill fluid administered through the vasopressor was not associated with a higher arterial lactate level and corrected the pH in a manner similar to vascular filling.

Pig model:

The performance of the complete system retained for later use in humans was also evaluated on a porcine model.

A shock model controlled by spoliation with a fixed pressure level maintained (Wiggers model) was used.

The protocol has been validated by the ethics committee. The animals used were pigs from a cross of the Large-white and Landrace breeds aged 11 to 12 weeks.

The animals were sedated by IM ketamine. A peripheral venous catheter was positioned in an atrial peripheral vein. Induction of anesthesia was performed intravenously with propofol 2mg.Kg ^1. Maintenance of anesthesia was performed by inhalation (Sevoflurane 2.5% Vol). Additional analgesia was provided by Buprenorphine injection at 0.5 mg / kg ¹ . The animals were curarized with atracurium at 0.5 mg.Kg ¹ .

After anesthesia, the animals were intubated and ventilated. An arterial catheter (Pulsiocath 7F) was inserted into the right femoral artery to measure blood pressure and cardiac output.

A central venous catheter (Arrow®) was inserted into the jugular to allow norepinephrine infusion. A filler pad was inserted into the brachiocephalic venous trunk to allow for spoliation and vascular filling. The vascular approaches were performed according to the Seldinger technique under ultrasound guidance.

A cutaneous temperature sensor was positioned on the proximal portion of the right lower leg. Electrocardiogram (ECG), blood pressure and distal temperature were recorded continuously via the MP150 Biopac® acquisition system.

The haemorrhagic shock was induced by a spoliation in two phases. The first quickly lowered the average blood pressure level to 30mmHg in 5 minutes. The second maintained mean arterial pressure between 30 and 35 mmHg for 85 minutes by spoliating or re-transfusing the necessary blood.

As illustrated in Figure 7, three groups were used in this study.

In the first group, the administration of the vascular filling fluid is performed manually (RV Manu, by opening and closing the infusion as in routine clinical practice (gravimetric infusion with a pocket positioned 1.5 m above In case of haemodynamic deterioration, an acceleration by compression of the bag of solute filling is carried out.The infusion is administered manually according to the level of pressure by a mode "on" (below 80mmHg of arterial pressure blood) and "stop" (above 83mmHg blood pressure). The 2 ^nd and 3 ^rd groups of pigs were resuscitated with the automated system: the 2nd group (Auto RV) by automated filling alone and the 3rd group (RVNA Auto) by automated vascular filling and norepinephrine injection.

In the 2 ^nd and 3 ^rd groups, an additional sensor was connected to the autonomous resuscitation system to measure the pressure on the same arterial line.

In the 2 ^nd and 3 ^rd groups, the vascular filling was performed by a pump (Flowpump, Gamida®).

For the ^3rd group, the administration of vasopressors (noradrenaline) was performed by an electric syringe pump GH (Carefusion®). A saline infusion vector was used to stabilize norepinephrine infusion. The flow rate of the vector was automatically indexed to the noradrenaline flow rate to ensure a stable minimum flow rate.

In the 2 ^nd and 3 ^th groups, the only necessary interventions were the notification of the weight of the animal and the click on the icon of launch of the resuscitation.

The resuscitation lasted one hour for all groups.

The cardiac output was measured by transpulmonary thermodilution with the Picco 2 monitor (Pulsion®) after stabilization at basal level, 1 hour of spoliation, end of spoliation and end of resuscitation. Arterial blood temperature and distal skin temperature were recorded continuously.

Arterial lactate, ionized calcium, a blood gas and a hemoglobin assay were performed at the end of spoliation and at the end of resuscitation (ABL 800, Radiometer®).

At the end of the experiment, the animals were killed by overdose with pentobarbital sodium and potassium chloride.

To standardize the sampling for measuring the blood pressure between the group ¹ and the 2 ^nd and 3 ^rd groups, the blood pressure values restraints were measured with the same material and extracted from the Picco 2 monitor. Blood pressure was measured every 20 seconds.

The primary endpoint was the Global Score (GS).

Secondary evaluation criteria include:

the percentage of time spent in the target zone (the blood pressure is in the desired range);

- the percentage of time spent below the objective (the pressure is too low);

- the percentage of time spent above the objective (the blood pressure is too high);

- the number of resuscitation interventions (start and stop infusion);

- the indexed cardiac output;

the volume of filling administered (the amount of filling liquid administered);

- extra-vascular pulmonary water (EPEV);

- temperature ;

lactate;

- hemoglobin;

- the gases of the blood.

The normality of the distribution of the variables was evaluated by D'Agostino Pearson's test.

The majority of the variables were either too small or non-normal and non-parametric tests were used.

The data are presented in the median-interquartile form. The performance of the automated resuscitation system (2 ^nd and 3 ^rd groups) was compared to that of manual resuscitation (1 ^st group).

Performance of automated resuscitation algorithms with or without norepinephrine were compared (comparison of the 3 ^rd group compared to the 2 ^nd group).

Time spent in the target area, performance variables, volume of vascular filling and the number of interventions required for resuscitation were compared between a control group and ¹ group (manual filling) by Mann-Witney test.

The performance error and the other performance evaluation parameters (median error or MDPE, the absolute value of the median error or MDAPE, the variability or WOBBLE, the divergence and overall score or GS) were calculated according to the methodology described in the test on rats.

In Group ^1, one animal died during the measurement of cardiac output at the end of resuscitation. In order not to influence the statistical analysis with this animal, the hemodynamic variables and the biological dosages of this animal were replaced by the average values of the RVM group.

Given the groups' performance, the effects of norepinephrine were investigated by comparing the 3 ^rd group with the 1 ^st and 2 ^nd groups. For this, the 1 ^st and 2 ^nd groups were grouped into one group (fill group RV).

The effects of resuscitation on hemodynamics, extravascular pulmonary water (EPEV), temperature, blood gases, lactate, ionized calcium and hematocrit were assessed by analysis of variance (or ANOVA for 2-factor nonparametric "analysis of variance" for repeated measurements.

In this test, 22 pigs were included. The median weight was 35.5 kg (32.7-38). The basal blood pressure was 93/57 (73) mmHg. The basal indexed cardiac output was 5.205 Lmin ¹ nr ² (4.383-5.603), TaO 2 was 670.3 mL.min ¹ nr ² (564-649.2) and the EPEV 9.283 mL. kg- ¹ (8.667-1 1), blood temperature of 36.95 ° C (36.48-37.4), hemoglobin of 9.9 g.dL 1 (9.2-10.7), the pH at 7.55 (7.52-7.58) and the base basal excess at 7.9 mmol.L ¹ (7.1 - 9.2). Biological data are reported in Table 4. Table 4 - Biological Parameters of the Different Groups

During hemorrhagic period, pigs were deprived of mL.kg 36.95 ¹ (32.66 to 42.16). Spoliation decreased the pH to 7.41 (7.3-7.47) and the excess base to -2.1 mmol.L ¹ (-5, 1-1.1). Lactate increased to 6.6 mmol.L ¹ (4,8-12.3). Hemoglobin remained stable at 10.5 g.dL ¹ (9-11.2). Ionized calcium decreased to 1.22 mmol.L ¹ (1.16-1.24).

The indexed cardiac output decreased to 1.48 L.min ¹ nr ² (1.15-1.78), Ta0 ₂ to 257 mL.min- ¹ m ² (181-292.8), and EPEV at 7.73 mL.kg ¹ (6.32-8.79). The temperature increased to 37.4 ° C (36.6-37.9) .2 pigs died at the end of the spoliation before the start of resuscitation. Regarding the ^1st group (manual filling group), 1 pig died at the end of resuscitation during cardiac output measurement. During the resuscitation period of one hour, 95.4 mL.kg ¹ (75.5 to 135) for filling liquid were needed to maintain hemodynamic goals.

Resuscitation failed to correct acidosis with a pH of 7.28

(7.18-7.44), but decreased lactate to 5.1 mmol.L ¹ (2.9-9) and hemoglobin at 5.2 g.dL ¹ (4, 3-6, 3).

The indexed cardiac output increased to 6.05 L.min ¹ nr ² (4.91-7.74) and the EPEV to 9.06 mL.kg ¹ (7.55-13.16). TaO 2 decreased to 1 14.4 mL.min- ¹ m ² (66.6-232.1) as well as the temperature to 35.6 ° C (34.7-36.3).

Pigs spent 66.1% (34.4-78.2) of resuscitation time in the target area, 42% (24.9-69.7) of time below target, and 2.2% (0-4,5) above.

The Global GS Score is 12.7 (10.2-54.2). The performance of resuscitation is reported in Table 5.

35 manual interventions (13.8-45.5) were necessary to perform resuscitation for 1 hour.

Table 5 - Resuscitation Performance by Groups

^* p <0.05 vs RVM - ^*** p <0.001 vs RVM - $$$ p <0.001 vs RV Auto

Regarding the 2 ^nd group, no pig died during the resuscitation.

During the resuscitation time, pigs from the 2 ^nd group received 86

5 mL.kg ¹ (74.3-95.3) of filling. Resuscitation failed to normalize the pH to 7.37 (7.34-7.44). Filling decreased lactate 3.3 mmol.L ¹ (2.8-5) and hemoglobin to 4.4 g.dL ¹ (3.5-5-5.9).

The indexed cardiac output increased 5.98 L.min ¹ nr ² (5.27-6.65) and the EPEV 9.28 mL.kg- ¹ (8.56-10.47).

10 The TA02 decreased to 100.3 mL.min- ¹ .nr ² (81 4 to 122.1) and the temperature decreased to 35.4 ° C (34.7 to 35.5).

Animals in the 2 ^nd group (RVA) spent 79.2% (59.4-86.3) of the time in the target zone, 19.1% (14.6-52.5) of the time below and 0% (0-2,2) above. The Global GS Score is 7.6 (5.6-15.3).

In the 2 ^nd group, only 2 interventions were necessary to achieve resuscitation (p <0.0001).

Regarding the ^3rd group, no pig has died during resuscitation. During the resuscitation time, the pigs received 31.3 mL.kg- ¹ (27.7-42.1) of filling fluid, which is lower than the control group

(P <0.01).

Resuscitation did not improve the pH to 7.34 (7.26-7.39) but decreased the lactate to 7 mmol.L ¹ (4.2-9.1). Hemoglobin decreased to 9.1 g.dL ^-1 (7.8-10.2), which is lower than the control group (p <0.01). The indexed cardiac output increased to 3.17 L.min ^-1 .nr ² (3.09-3.54), which is less than in the

Control (p <0.05). TaC> 2 decreased similarly to the control group at 138.9 mL.min ¹ nr ² (90.3-212.1). The EPEV remained stable at 7.28 (6.94-8.08) mL.kg- ¹ . The temperature remained stable at 37.5 (35.9-38.5) ° C.

The pigs of the 3 ^rd group spent 71, 8% (55.4 to 76.8) of the time in target zone, 35.1% (23.3 to 69) of the time and below 0.2% (0- 8.6) above. The Global GS Score is 8.3 (7.7-15.9).

Only 2 interventions were necessary to achieve resuscitation

(p <0.0001).

The serum calcium at the end of resuscitation was superior to 1 ^st and 2 ^nd groups (this is significant taking calcemia end resuscitation without repeated measurements, p <0.05).

Regarding the comparison between the 1 ^st and 2 ^nd groups compared to the 3 ^th group, at the end of resuscitation the heart rate is higher in the 3 ^th group than in the 2 ^nd group (p <0.0001).

PPV (pulse pressure variation dependence precharge marker) is higher in the 3 ^rd end resuscitation group (p <0.001).

The indexed cardiac output (CI, cardiac index) is less in the ^3rd end of resuscitation group (p <0.001). On the other hand, the Ta0 ₂ was not different, as can be seen in Figure 8.

Vascular resistances are higher at the end of resuscitation in the 3 ^rd group NA (p <0.001) but remain below the basal values.

The indexed-diastolic heart volume (GEDI) and the intrathoracic blood volume indexed (ITBI the acronym "intrathoracic blood volume index") are lower in the ^3rd end of resuscitation group (p <0.001).

The extravascular lung water (EVLW) is lower in the 3 ^rd end resuscitation group (p <0.05).

The administration of a vasopressor, here norepinephrine, allowed a saving of filling (reduction of the amount of filling fluid administered) and resulted in less dilution: hemoglobin and calcium are higher (p <0.001 for Hb, significant only for the end-of-resuscitation value for Ca2 +). Similarly, the temperature is higher in the ^3rd group at the end of resuscitation and end of transfusion (p <0.0001). This vascular filling savings is not associated with a higher arterial lactate level and corrects the pH in a manner similar to vascular filling. The paC> 2, the paCC> 2 and the excess base are not different.

This large animal study confirms the feasibility of automated resuscitation in haemorrhagic shock with equipment and dosages close to what is found in humans.

The use of a closed filling loop guided on the systolic blood pressure makes it possible to maintain the pressure in the target zone over time in a manner comparable to the optimized manual filling.

The performance of the automated resuscitation is achieved with a minimal number of interventions, reduced to 2 clicks of start and stop resuscitation.

This study confirms the results obtained on small animals. The use of noradrenaline in combination allows a saving of vascular filling and a decrease of the dilution without increasing of metabolic suffering. The temperature was significantly higher in the 3 ^rd group (group with administration of pressor), of the order of 2 ° C. This effect is important because hypothermia is a component of the lethal triad and a factor promoting coagulopathy.

Claims

1. Device (5) for determining the amounts of filling liquid and vasopressor for administration to a patient by means of a first dispenser (3) and a second dispenser (4) respectively, characterized in that the device (5) is configured for, from a blood pressure previously measured on a patient, determining a quantity of filling liquid and a quantity of vasopressor such that said amounts are functions of the ratio between the amounts of filling liquid and vasopressor already administered over a period of time.

2. Device (5) according to claim 1, wherein the amounts to be administered filling liquid and vasopressor are determined according to several phases in which the rate of administration of the liquid filling or vasopressor is fixed and the ratio is changed .

3. Device (5) according to claim 2, wherein the amounts of filling liquid and vasopressor to be administered are determined according to several phases:

- first phase: the amount of vasopressor a is constant and is determined according to the blood pressure (Part) and the amount of filling liquid is determined so that the measured blood pressure reaches a target value (Pdbie), the first phase stops either when the blood pressure (Part) reaches the value of target blood pressure (Pdbie), the administration then goes into a phase of regulation of the vasopressor, either when the amount of filling liquid administered (Qiiquide) reaches a first threshold value ( _liquid S1), the administration then goes into a second phase;

second phase: the determination of the amounts of vasopressor and filling liquid is such that the value of arterial pressure is the target value (Pdbie), the second phase stops either when the arterial pressure (Part) is equal to the target arterial pressure value (Pdbie) and the amount of vasopressor (D _va so) determined is less than a first threshold (S 1 _V aso), the administration then goes into the regulation phase of the vasopressor, either when the quantity of administered filling liquid (Qnquide) reaches a second threshold value (S2ii _qU ide) greater than the first threshold value (S 1 liquid), then the administration proceeds to a third phase;

- third phase: the amount of filling liquid is zero and the amount of vasopressor is determined so that the arterial pressure (Part) reaches the value of target blood pressure (Pdbie), the third phase stops either when the blood pressure ( Part) reaches the value of the target arterial pressure value (Pdbie) and the amount of vasopressor (D _va so) is lower than the first threshold (S 1 _V aso), the administration then passes into the vasopressor regulation phase, either when the vasopressor flow (D _va so) is greater than the first threshold (S 1 _V aso), the administration then goes into a fourth phase;

fourth stage: the amounts of vasopressor and filling liquid are determined so that the value of the arterial pressure reaches the target value (Pdbie), the fourth phase stops either when the arterial pressure (Part) reaches the pressure value arterial target (Pdbie) and the vasopressor flow rate (D _va so) is less than the first threshold (S 1 _V aso), the administration then passes into the regulation phase of the vasopressor, ie when the quantity of filling liquid is determined ( Qnquide) reaches a third threshold value (S3iiq _U of) greater than the second threshold value (S2ii _qU ide), the administration then goes into a fifth phase;

- fifth phase: the amount of filling liquid is nule the amount of vasopressor is determined so that the arterial pressure (Part) reaches the value of target blood pressure (Pdbie), the fifth phase stops when the blood pressure (Part) reaches the target arterial pressure value (Pdbie) and the amount of vasopressor (D _va so) is less than a second threshold (S2 _va so) greater than the first threshold (S 1 _V aso), the administration then goes into the regulation of the vasopressor, ie when the blood pressure (Part) has not reached the target arterial pressure value (Pdbie) after a predetermined duration (T), the administration then proceeds to a sixth phase; - sixth phase: the amount of vasopressor is constant according to a predetermined value and the amount of filling liquid is determined so that the arterial pressure (Part) reaches the target arterial pressure value (Pdbie), the sixth phase stops only when the blood pressure (Part) reaches the target blood pressure value (Pdbie), the administration then goes into the regulation phase of the vasopressor;

- vasopressor regulation phase: the amount of filling fluid is zero and the amount of vasopressor is determined so that the arterial pressure (Part) reaches the target arterial pressure value (Pdbie), the vasopressor regulation phase stops when the blood pressure (Part) is below a threshold blood pressure value (Pseuii), in this case determine a quantity of filling fluid if the quantity of filling liquid administered (Qnquide) is below the threshold value ( S 1 liquid, S 2 liquid, S 3 _liquid ) of the phase in which was the administration before the regulation phase of the vasopressor, or go into the fifth phase if the amount of vasopressor (D _va so) is less than a value predetermined maximum (D _vasomax ), or otherwise go into the sixth phase.

4. Device (5) according to any one of claims 1 to 3, wherein the amount of vasopressor is determined not to increase when the lactate clearance of the patient reaches a threshold level.

5. Device (5) according to any one of claims 1 to 4, wherein the amount of filling liquid and / or vasopressor is determined by closed loop control of PID type relative to the arterial pressure.

6. Device (5) according to any one of claims 1 to 4, wherein the amount of filling liquid and / or vasopressor is determined by a closed loop control with a fuzzy logic type algorithm with respect to the pressure blood.

7. System (1) for administering to a patient a filling liquid and a vasopressor comprising:

an arterial pressure sensor (2) configured to measure the patient's blood pressure;

a first dispenser (3) configured to administer the filling liquid to the patient;

a second dispenser (4) configured to administer the vasopressor to the patient;

- A control device (5) according to one of claims 1 to 6.

8. Operating method of the device (5) control according to one of claims 1 to 6, said method comprising a step of determining a rate of administration of the filling liquid and the vasopressor from a blood pressure said patient while maintaining a determined ratio defined by a quantity of filling liquid already administered on a quantity of vasopressor already administered.

9. Computer program product comprising program code instructions for performing the steps of a method of operation of the control device (5) according to the preceding claim.