RU2684191C2 - Method for predicting the sedation level during anaesthetic support - Google Patents

Abstract

FIELD: medicine.SUBSTANCE: invention relates to medicine, in particular to anesthesiology – resuscitation, and can be used in predicting the level of sedation during anesthesia. To do this, determine the indicators of the electrophysiological activity of the brain of the patient by the method of omegametry with the registration of ultralow brain biopotentials. Heart rate variability and hemodynamic parameters are analyzed to form a visual demonstration. Statistically significant factors are assigned numerical values, after which the prognostic factors F, Fand Fare determined by the original calculation formulas. If value Fis more than Fand Fa high probability of the first level of sedation is predicted, if Fis more then Fand F, a high probability of the second level of sedation is predicted, and if Fis more then Fand F, a high probability of the third level of sedation is predicted.EFFECT: method provides the possibility of timely registration of the development of a painful reaction to surgical aggression during anesthesia before the onset of its clinical manifestations and thus allows for adequate sedation and analgesia.1 cl, 4 dwg, 1 ex, 2 tbl

Description

The alleged invention relates to medicine, namely to anesthesiology - resuscitation and can be used when conducting anesthetic management of patients with various pathologies.

It is known that existing methods of monitoring the adequacy of sedation are aimed at recording changes in the physical or chemical parameters of the body that occur in response to pain irritation.

A known method of controlling the level of sedation during inhalation anesthesia, in which the level of sedation during anesthesia is assessed by the dynamics of pericortical oxygen saturation of brain tissue, recorded by cerebral oximetry - optical spectroscopy in the range of 700-1500 nm. With an increase in the regional saturation index (rSat) of more than 2% of the initial values, the adequacy of the sedation level is fixed (Application 2009136063 Russian Federation, IPC А61В 10/00. Method for controlling the level of sedation during inhalation anesthesia / Lazarev VV, Tsypin L.E. and dr. - claimed on September 30, 2009; published on April 10, 2011).

The disadvantages of this method include the low accuracy and specificity of assessing the adequacy of the level of sedation, which, according to the authors of the proposed method, is associated with a size of data variability from 2 to 100%. In addition, accuracy is also affected by a significant number of artifacts due to disposition of sensors and an admixture of extracerebral blood.

The closest in technical essence to the proposed one, is a method for predicting inadequate sedation with neuroaxial anesthesia, including recording the constant potential of the brain. 12-24 hours before the operation, the study of ultra-slow biopotentials is carried out. For this, the background constant potential (PP) is recorded by the Omega-4 hardware-computer complex using the omegametry method. The active electrode is located in the center point of the midline of the forehead 1 cm above the superciliary arches, the reference electrode - in the tenar of the hand of the dominant hand. Since the background value of PP reflects the level of operative rest and nonspecific resistance of the organism to stressful effects by preoperative omegagrams, the value of constant PP after reaching a plateau is estimated.

With PP values of more than 19 mV, the development of inadequate sedation is predicted. With constant potential values of less than 20 mV, sufficient sedation is predicted. This allows us to identify a group of patients prone to the development of psycho-emotional discomfort during neuroaxial anesthesia, and requiring preliminary individual correction of premedication (Pat. 2340280 Russian Federation, IPC A61B 5/0476. A method for predicting inadequate sedation with neuroaxial anesthesia / Zabolotskikh IB, Pesnyak E.V. et al .; No. 2007108184; claimed 05.03.2007, publ.: 10.12.2008).

The disadvantages of this method should also include its low specificity associated with the use of the subjective Ramsay scale, as well as the assessment of the adequacy of sedation - by the presence of nausea and vomiting in the post-narcotic period.

In addition, the known method allows to assess only the presence or absence of sedation and does not share its levels.

The objective of the invention is the development of a method for predicting the level of sedation during anesthesia, based on modeling the variability of the heart rhythm and hemodynamic parameters of the patient.

The technical result of the proposed method is to increase the accuracy and specificity of predicting the level of sedation during anesthesia, by analyzing the heart rate variability in real time (on line) and by registering a pain response in a patient even before its clinical manifestations.

The technical result of the proposed method is achieved by the fact that to predict the level of sedation during anesthesia benefits determine the indicators of electrophysiological activity of the patient’s brain.

Distinctive techniques of the proposed method are that by the method of omegametry, ultra-slow brain potentials are recorded and an analysis of heart rate variability and hemodynamic parameters is carried out with the formation of a visual demonstration.

The difference of the method lies in the fact that numerically assigned assigned statistically significant factors.

The difference of the method also lies in the fact that the boundaries of sedation correspond to prognostic factors F ₁ , F ₂ and F ₃ : the first level is BIS 60-100% - the patient responds to the stimulus; the second level - BIS 40-60% - sufficient sedation and the third level - BIS less than 40% - deep sedation. Sedation levels are determined by linear discriminant analysis formulas:

F ₁ = -3.21 + 2.7 × x ₁ -0.78 × x ₂ + 1.52 × * x ₃ -3.41 × x ₄ + 3.6 × x ₅ -1.56 × x ₆ + 0.69 × x ₇ + 1.04 × x ₈ + 1.07 × x ₉ -0.23 × x ₁₀ -0.09 × x ₁₁ ;

F ₂ = -0.75-0.74 × x ₁ + 0.39 × x ₂ + 1.51 × x ₃ + 1.15 × x ₄ -1.22 × x ₅ + 0.67 × x ₆ + 0.11 × x ₇ -0.21 × x ₈ -0.29 × x ₉ + 0.14 × x ₁₀ -0.02 × x ₁₁ ;

F ₃ = -5.22-1.55 × x ₁ + 1.02 × x ₂ + 1.94 × x ₃ + 0.91 × x ₄ -1.13 × x ₅ + 0.65 × x ₆ + 1.02 × x ₇ -0.75 × x ₈ -0.26 × x ₉ + 0.87 × x ₁₀ + 0.36 × x ₁₁ , where:

F ₁ - 1st level of sedation - BIS 60-100% - mild sedation, the patient responds to an irritant;

F ₂ - 2nd level of sedation - BIS 40-60% - sufficient sedation;

F ₃ - 3rd level of sedation - BIS less than 40% - deep sedation;

x ₁ ... _11. statistically significant parameters of heart rate variability and hemodynamics:

x ₁ - heart rate (heart rate);

x ₂ - AMo - mode amplitude (fraction of CI corresponding to the mode value);

x ₃ - HELL syst. - systolic blood pressure

x ₄ - CV - coefficient of variation of RR-intervals;

x ₅ - Delta X (ΔX),

x ₆ - Mo - mode (the most common RR interval);

x ₇ - HELL diast. - diastolic blood pressure

x ₈ - SKOS As - asymmetry;

x ₉ - Min - the minimum cardio interval (CI) in the sample;

x ₁₀ - excess,

x ₁₁ - IN - voltage index of regulatory systems.

The authors of the proposed method found that if the absolute value of F _{1 is} greater than the absolute value of F ₂ and F ₃ then they predict a high probability of the patient's condition, which corresponds to mild sedation or a confident reaction to an irritant - the first level of sedation.

When the absolute value of F ₂ more than the absolute values of F ₁ and F ₃ predict a high probability of the patient's condition, corresponding to sufficient sedation or a stable lack of sensitivity to the stimulus - the second level of sedation.

If the absolute value of F _{3 is} greater than the absolute values of F ₁ and F ₂ predict a high probability of the patient's condition, which corresponds to deep sedation, bordering on severe inhibition of the cerebral cortex - the third level of sedation.

A comparative analysis with the prototype showed that the proposed method differs from the known above methods and, therefore, meets the criteria of the invention of "novelty."

From the analysis of patent and specialized literature, the authors found that the proposed method has features that distinguish it not only from the prototype, but also other technical solutions in this and related fields of medicine. In the available literature, the authors have not identified a method for predicting the level of sedation during anesthetic management of the above factors.

In the inventive method, the authors use a model for predicting the level of sedation during anesthesia, based on heart rate variability (HRV) and the "gold standard" of sedation adequacy - the bispectral index (BIS) obtained from the electroencephalogram, followed by a graphical plot in real time. At the same time, the model included both HRV parameters and hemodynamic parameters. The statistically significant parameters established, which were calculated by analyzing the rhythmogram in the ORTO Science and MATLAB 6.1 software, are presented in Table 1 below.

where x ₁ - heart rate (heart rate);

x ₂ - AMo - mode amplitude (fraction of CI corresponding to the mode value);

x ₃ - blood pressure syst.- systolic blood pressure;

x ₄ - CV - coefficient of variation of RR-intervals;

x ₅ - Delta X (AX);

x ₆ - Mo - mode (the most common RR interval);

x ₇ - HELL diast. - diastolic blood pressure;

x ₈ - SKOS As - asymmetry;

x ₉ - Min - the minimum KI in the sample;

x ₁₀ - excess;

x ₁₁ - IN - voltage index of regulatory systems (IN = AMo / 2ΔX × Mo).

To visually confirm the results, a demonstration of the change in the sedation level in the coordinate system is carried out using the canonical linear discriminant function (KLDF), where the coefficients K _X and K _Y correspond to the abscissa and ordinates (Fig. 1-4, appendices to the description):

K _X = 0.21-1.26 × x ₁ + 0.43 × x ₂ -0.71 × x ₃ + 1.65 × x ₄ -1.75 × x ₅ + 0.81 × x ₆ -0 , 2 × x ₇ -0.46 × x ₈ -0.5 × x ₉ + 0.14 × x ₁₀ + 0.05 × x ₁₁

K _Y = -0.04 + 0.6 × x ₁ -0.4 × x ₂ + 1.25 × x ₃ -0.07 × x ₄ + 0.16 × x ₅ -0.09 × x ₆ - 0.47 × x ₇ + 0.35 × x ₈ + 0.04 × x ₉ -0.42 × x ₁₀ -0.19 × x ₁₁

Clinical observations of the authors of the proposed method indicate that the use of the proposed technical solution allows us to predict the level of sedation during anesthetic management. At the same time, the forecast accuracy of the first level is 85.7%, for the second level - 93.3%) and for the third level - 46.3%.

The total percentage of reliability of the classification matrix was 90.1%). The low reliability result for deep sedation is the third level, easily explained by the small number of observations associated with the inappropriateness of achieving such a depth of anesthesia with endoscopic interventions.

This allows us to conclude that the technical solution meets the criterion of "inventive step".

A method for predicting the level of sedation during anesthetic benefits that make up the claimed invention is intended for use in healthcare. The implementation of its capabilities is confirmed by the methods and means described in the application. From the above material it follows that the claimed invention meets the condition of patentability "industrial applicability".

The proposed method is as follows.

The heart rate values are entered into a personal computer using HeartSense equipment (produced by NPP Living Systems) complying with GOST 11.040 РМЭК 601-1-1-96 (wireless equipment for recording heart rhythm), which consists of cardioelectrodes and a radio signal receiver. information, in the form of the distance between two contractions of the ventricles (RR interval) in milliseconds for each contraction of the heart, is digitally transmitted via com (usb) - port of a personal computer, decoded software and is recorded.

By moving data through macros, the resulting parameters are calculated - the linear discriminant function (LDF) for level 1 is mild sedation, the patient responds to an irritant - BIS 60-100%, for level 2 - sufficient sedation - BIS 40-60% and 3 - deep sedation - BIS less than 40%. From the obtained LDF values, choose the maximum, which is the corresponding level of sedation.

Also in parallel mode, the calculation of the canonical linear discriminant function (CLDF) and the construction of a point graph are performed. After a preliminary calculation of the coordinates of the "centroids" (the average value of the canonical estimates for each sedation level), the coordinates of the points X and Y with each R-R interval are calculated. As a result, by the position of the found point on the axes in the X and Y coordinates, the distance from the “centroids” of the CLDF values of a patient with mild sedation — response to stimulus — BIS 60–100%, sufficient sedation — BIS 40–60%, and deep sedation — BIS is estimated less than 40%. Depth of sedation is established by the "centroid", from which the smallest removal is obtained. Each new value obtained as a result of continuous cardiointervalometry and recording of hemodynamics allows you to see a picture of simulated changes in BIS by HRV.

This data display allows you to observe in 2 charts: linear (LDF) and point (KLDF) the patient’s response: mild sedation - BIS 60-100%, sufficient sedation - BIS 40-60% and deep sedation - BIS less than 40%, which precedes the clinical manifestations of the pain reaction, depending on the level of sedation.

The proposed dynamic assessment of stress stress in the regulatory systems of the body, namely: the shift of lines (LDF) to the 1st, 2nd or 3rd level of sedation or the emergence of new points (KLDF) in relation to the "centroids", allows you to timely notice the development of pain response to surgical aggression during anesthesia even before the appearance of its clinical manifestations and, therefore, timely conduct adequate sedation and analgesia.

The proposed method for predicting the level of sedation during anesthetic management is illustrated by examples of specific performance.

Clinical example 1. Patient A., was admitted with a diagnosis of Diverticular disease of the colon. During the preparatory period, the values of hemodynamic parameters and their standardized values, which are presented in table 2, were determined.

To determine the state of anesthesia, the established discriminant analysis formulas were used:

F ₁ = -3.21 + (2.7 * l, 66) - (0.78 * 0.96) + (l, 52 * 4.22) - (3.41 * -0.15) + ( 3.6 * -0.45) - (l, 56 * -1.66) + (0.69 * -0.48) + (1.04 * 0.73) + (1.07 * -0, 16) + (0.23 * -0.85) - (0.09 * -0.23) = 8.5

F ₂ = -0.75- (0.74 * 1.66) + (0.39 * 0.96) + (1.51 * 4.22) + (1.15 * -0.15) - ( 1.22 * -0.45) + (0.67 * -1.66) + (0.11 * -0.48) - (0.21 * 0.73) - (0.29 * -0, 16) + (0.14 * -0.85) - (0.02 * -0.23) = 3.8

F ₃ = -5.22- (1.55 * 1.66) + (1.02 * 0.96) + (1.94 * 4.22) + (0.91 * -0.15) - ( 1.13 * -0.45) + (0.65 * -1.66) + (1.02 * -0.48) - (0.75 * 0.73) - (0.26 * -0, 16) + (0.87 * -0.85) + (0.36 * -0.23) = - 1.2

The absolute value of F _{1 is} greater than the absolute values of F ₂ and F ₃ , therefore, they predict a high probability of the patient's condition, which corresponds to mild sedation - a confident reaction to the stimulus.

Demonstration of changes in the sedation level in the coordinate system occurs when using KLDF:

K _X = 0.21- (1.26 * 1.66) + (0.43 * 0.96) - (0.71 * 4.22) + (1.65 * -0.15) - (1 , 75 * -0.45) + (0.81 * -1.66) - (0.2 * -0.48) - (0.46 * 0.73) - (0.5 * -0.16 ) + (0.14 * -0.85) + (0.05 * -0.23) = - 5.56

K _Y = -0.04 + (0.6 * 1.66) - (0.4 * 0.96) + (1.25 * 4.22) - (0.07 * -0.15) + ( 0.16 * -0.45) - (0.09 * -1.66) - (0.47 * -0.48) + (0.35 * 0.73) + (0.04 * -0, 16) - (0.42 * -0.85) - (0.19 * -0.23) = 6.8

In FIG. 1 of the appendix to the description, the patient’s sedation status is displayed in the coordinate system when using KLDF, relative to centroids that determine sedation levels according to BIS. The maximum approximation to the centroid corresponding to the reaction to the stimulus was revealed, which "says" about BIS 60-100%.

The closest approach to the centroid, corresponding to sufficient sedation, indicates a BIS of 40-60% - the second level of sedation (Fig. 2 appendix to the description).

With a further set of KI, there is a calculation of all the studied parameters and the construction of KLDF (Fig. 3 of the appendix to the description).

The data obtained allow us to conclude that the level of sedation according to BIS coincides with the response to the stimulus (BIS 60-100%) both in the pre-anesthetic period according to LDF and KDLF, and with a further transition, after administration of drugs for anesthesia, to the level of sufficient anesthesia - BIS 40 -60%, and then a return to the state of reaction to the stimulus -BIS 60-100% is noted.

Based on the calculation formulas and additional software, an interval visual BIS image is presented in Fig. 4 (appendices to the description), which shows the level of anesthesia of the patient in the coordinate system using LDF, in dynamics with BIS monitoring.

A clinical study of the proposed method was conducted on the basis of the Irkutsk Clinical and Consultative Diagnostic Center since 2008. A total of 114 people were examined, of which 50 were women and 64 were men. An additional group that did not participate in the creation of the model consisted of 52 patients (men - 31 people, women - 21 people) who were examined for gastrointestinal tract pathology under intravenous anesthesia. Correct recognition was 81%. The analysis revealed a high correlation of most indicators of HRV and blood pressure with levels of mild sedation and sufficient sedation during short-term anesthesia with endoscopic interventions.

The proposed method can be used in clinical practice.

Claims (20)

A method for predicting the level of sedation during anesthetic management, including determining the electrophysiological activity of the patient’s brain, characterized in that the ultra-slow biopotentials of the brain are recorded by omegametry, an analysis of heart rate variability and hemodynamic parameters with the formation of a visual demonstration, statistically significant factors are assigned numerical values then prognostic factors F ₁ , F ₂ and F ₃ determine according to the formulas: F ₁ = -3.21 + 2.7 × x ₁ -0.78 × x ₂ + 1.52 × x ₃ -3.41 × x ₄ + 3.6 × x ₅ -1.56 × x ₆ + 0.69 × x ₇ + 1.04 × x ₈ + 1.07 × x ₉ -0.23 × x ₁₀ -0.09 × x ₁₁ ; F ₂ = -0.75-0.74 × x ₁ + 0.39 × x ₂ + 1.51 × x ₃ + 1.15 × x ₄ -1.22 × x ₅ + 0.67 × x ₆ + 0.11 × x ₇ -0.21 × x ₈ -0.29 × x ₉ + 0.14 × x ₁₀ -0.02 × x ₁₁ ; F ₃ = -5.22-1.55 × x _l + 1.02 × x ₂ + 1.94 × x ₃ + 0.91 × x ₄ -1.13 × x ₅ + 0.65 × x ₆ + 1.02 × x ₇ -0.75 × x ₈ -0.26 × x ₉ + 0.87 × x ₁₀ + 0.36 × x ₁₁ , where F ₁ - 1st level of sedation - BIS 60-100% - mild sedation, the patient responds to an irritant; F ₂ - 2nd level of sedation - BIS 40-60% - sufficient sedation; F ₃ - 3rd level of sedation - BIS less than 40% - deep sedation; x _{1 ... 11} - statistically significant parameters of heart rate variability and hemodynamics: x ₁ - heart rate (heart rate); x ₂ - AMo - the amplitude of the mode (fraction of CI, corresponding to the value of the mode); x ₃ - HELL syst. - systolic blood pressure; x ₄ - CV - coefficient of variation of RR-intervals; x ₅ - Delta X (ΔX); x ₆ - Mo - mode (the most common RR interval); x ₇ - HELL diast. - diastolic blood pressure; x ₈ - SKOS As - asymmetry; x ₉ - Min - the minimum cardio interval (CI) in the sample; x ₁₀ - excess; x ₁₁ - IN - voltage index of regulatory systems, and with F ₁ greater than F ₂ and F _{3, a} high probability of the first sedation level is predicted, with F ₂ greater than F ₁ and F ₃ , a high probability of the second sedation level, and with F ₃ greater than F ₁ and F ₂ , a high probability of the third sedation.

Images