Expiratory gas metabolism marker for evaluating attack severity of bronchial asthma patient and screening method

Abstract

The invention relates to an exhaled gas metabolism marker for evaluating attack severity of a bronchial asthma patient and a screening method of the exhaled gas metabolism marker. The marker is acetic acid. The screening method of the exhaled gas metabolism marker for evaluating the severity of the attack of the bronchial asthma patients comprises the following steps: respectively collecting exhaled gas of the bronchial asthma patients with mild continuous attack and moderate continuous attack as analysis samples; analyzing each exhaled gas sample by using HPPI-TOFMS to obtain an original mass spectrum of each exhaled gas sample; preprocessing the original mass spectrum of each expired gas sample to obtain data containing relative intensity information of metabolites; carrying out univariate and multivariate analysis on the data containing the relative intensity information of the metabolites, and screening out differential metabolites; a metabolite with the largest difference is screened out through binary logistic regression analysis, and the metabolite is the marker. The method is simple to operate, can efficiently and objectively evaluate the severity of the attack of the bronchial asthma patient, and has important clinical application value.

Description

Technical Field

[0001] The present invention belongs to the field of biomedical technology, and particularly relates to an exhaled breath metabolic biomarker for evaluating the severity of asthma attacks in patients and a screening method therefor. Background Art

[0002] Bronchial asthma is a chronic airway inflammatory disease characterized by reversible airflow limitation. Typical clinical manifestations include wheezing, shortness of breath, chest tightness, and coughing. Data from the Global Burden of Disease Study show that as of 2015, the number of asthma patients globally had reached 358 million, with more than 400,000 annual deaths. In China, the population base of asthma patients is huge, with approximately 45.7 million cases among people aged 20 and above. Among them, most patients have problems such as non-standard medical treatment or missed diagnosis. The diagnosis rate is only 28.8%, and the rate of standardized treatment is less than 5.6%. The causes of this phenomenon not only lie in the lack of homogenization in early screening and diagnosis, but are also closely related to the lack of accurate grading of the severity of asthma attacks. Scientific grading can dynamically evaluate the progression of the disease, guide the formulation of individualized treatment plans, effectively identify high-risk populations and implement early interventions, thereby reducing the frequency of acute attacks and the risk of death. It is a key link in improving the management level of asthma.

[0003] In current clinical practice, the grading of the severity of asthma attacks is mainly based on the standards of the Global Initiative for Asthma (GINA) guidelines, and is comprehensively evaluated through parameters such as symptom frequency (daytime / nighttime attack frequency), lung function indicators (percentage of FEV1 predicted value and diurnal variation rate), and history of acute exacerbation. However, this traditional grading system has significant limitations: First, symptom scoring depends on the patient's subjective perception and is easily affected by psychological factors and differences in expression. Second, lung function tests (such as FEV1) only reflect the end-state of airway obstruction and cannot capture the dynamic evolution process of the inflammatory cascade reaction. Third, the existing standards are difficult to achieve real-time monitoring, resulting in clinical interventions often lagging behind pathophysiological changes. These deficiencies together lead to a lagging deviation between clinical grading and the actual disease progression, and there is an urgent need for more accurate objective assessment methods.

[0004] Volatile Organic Compounds (VOCs) are volatile organic compounds produced during human metabolism, and the concentration of exhaled breath can reflect the pathophysiological state in the body. In the field of respiratory diseases, VOCs have shown important application value as novel biomarkers: the concentration of pentane in the exhaled breath of asthma patients is significantly correlated with the degree of airway remodeling, and the isoprene level can reflect the degree of eosinophil infiltration; the concentration of aromatic VOCs such as toluene and styrene in the exhaled breath of patients with acute exacerbation of chronic obstructive pulmonary disease (COPD) increases, which can be used to distinguish bacterial infections from non-infectious exacerbations. By detecting the organic compounds in the exhaled breath of bronchial asthma patients and constructing an assessment model for the severity of bronchial asthma attacks, it may be possible to efficiently and objectively evaluate the severity of bronchial asthma attacks. Summary of the Invention

[0005] Aiming at the deficiencies in evaluating the severity of asthma attacks in the prior art, the purpose of the present invention is to provide an exhaled breath metabolite marker for evaluating the severity of bronchial asthma attacks and a screening method thereof, in order to solve problems such as subjectivity and lag in the existing grading methods.

[0006] To achieve the above purpose, the present invention specifically adopts the following technical solutions:

[0007] One of the purposes of the present invention is an exhaled breath metabolite marker for evaluating the severity of bronchial asthma attacks. The marker is acetic acid (m / z 79.038), and the severity of the attack refers to mild persistent and moderate persistent attacks of bronchial asthma.

[0008] Another purpose of the present invention is to provide a screening method for an exhaled breath metabolite marker for evaluating the severity of bronchial asthma attacks. The screening method includes:

[0009] (1) Collect the exhaled breaths of patients with mild persistent bronchial asthma and patients with moderate persistent bronchial asthma respectively as analysis samples;

[0010] (2) Analyze each exhaled breath sample using high-pressure photoionization time-of-flight mass spectrometry (HPPI-TOFMS) to obtain the original mass spectrum of each exhaled breath sample;

[0011] (3) Pretreat the original mass spectrum of each exhaled breath sample to obtain data containing metabolite relative intensity information;

[0012] (4) Perform univariate and multivariate analyses on the data containing metabolite relative intensity information, and screen out the differential metabolites between bronchial asthma patients with different severity levels of attacks according to the set thresholds (VIP>1 and FDR<0.05);

[0013] (5) The metabolites with the largest differences were screened out through binary logistic regression analysis, and this metabolite is the exhaled breath metabolite biomarker for evaluating the severity of asthma attacks in patients.

[0014] The screening method can also establish a receiver operating characteristic curve discrimination model based on the exhaled breath metabolite biomarker, and obtain the discrimination formula and discrimination threshold for patients with mild persistent asthma attacks and patients with moderate persistent asthma attacks.

[0015] Furthermore, when collecting the exhaled breath samples of patients, the collection population for exhaled breath sample collection is patients with clearly diagnosed mild persistent asthma attacks and moderate persistent asthma attacks, and there are acute symptom attacks; before collecting the exhaled breath of patients, they should fast, refrain from smoking, and remain calm for at least 3 hours. It is required that when collecting the exhaled breath, the patient first takes a deep breath through the nose, and then completely exhales through the mouth into a Tedlar collection bag with a volume of 3L by an exhaled breath sampling device. The exhaled air flow rate is set at 2800 - 3200 ml / min, and the Tedlar collection bag is sent to the laboratory within 24 hours for high-pressure photoionization time-of-flight mass spectrometry (HPPI-TOFMS) analysis; the exhaled breath samples of patients with mild persistent asthma attacks and patients with moderate persistent asthma attacks are all taken from the population aged 20 - 72 years; there are more than 10 exhaled breath samples for each type of patient.

[0016] Furthermore, when using high-pressure photoionization time-of-flight mass spectrometry (HPPI-TOFMS) to analyze each exhaled breath sample, first place the gas bag in an oven and heat it to 50 °C, then enter the TOFMS ionization analysis. The exhaled air flow rate is set at 2700 - 3100 ml / min. The ionization source uses a vacuum ultraviolet Kr lamp as the light source, the ionization source pressure is 200 - 300 Pa, the temperature is set at 100 °C, the ionization source repulsion voltage is 18 V, and the MCP voltage is 4100 V; the spectrum acquisition time is 120 s, and the mass-to-charge ratio range for spectrum data acquisition is 0 - 400 Da.

[0017] Preprocessing the original mass spectrometry diagram of the exhaled breath sample means: using Matlab software to perform mass calibration and signal peak intensity extraction on the obtained original mass spectrometry diagram, and then performing total normalization, data filtering, and logarithmic transformation on the data of each exhaled breath sample on the MetaboAnalyst website. Finally, a data table that can be used for statistical analysis is obtained: among them, the first row of the table is the metabolite mass number, the first column of the table is the exhaled breath sample number information, the second column is the exhaled breath sample grouping information, and each of the remaining columns is the relative signal intensity of each metabolite.

[0018] Univariate analysis refers to the Mann-Whitney U rank sum test, which was performed on the Multiple Experiment Viewer software. Differentially expressed metabolites with a false discovery rate (FDR) less than 0.05 were considered to be different between the two groups. Multivariate analysis refers to orthogonal partial least squares discriminant analysis, which was performed on the Simca-P software. The data was transformed by unit variance scaling, and metabolites with a variable importance in the projection (VIP) greater than 1 were selected. Finally, the metabolites common to the results of univariate analysis and multivariate analysis were taken as the differentially expressed metabolites for different severity grades of bronchial asthma attacks.

[0019] Binary logistic regression analysis was performed on the obtained differentially expressed metabolites using SPSS software. The dependent variable was set as the grouping information for different severity grades of asthma attacks in patients, the covariates were the relative intensities of different metabolites, and the method was "Forward: Conditional". The top-ranked differentially expressed metabolite among all the obtained differentially expressed metabolites was the optimal differentially expressed metabolite, and this differentially expressed metabolite was the exhaled breath metabolite combination biomarker for diagnosing asthma. Then, the method was set to "Enter", the optimal metabolite was selected as the covariate, and binary logistic regression analysis was performed again to obtain the predicted probability of this metabolite. Finally, receiver operating characteristic (ROC) curve analysis was performed on the obtained differentially expressed metabolites in SPSS software. The predicted probability of the differentially expressed metabolite was set as the test variable, the grouping information was set as the status variable, and the value of the status variable was the disease grouping. Finally, the ROC curve was obtained, and the abscissa and ordinate were 1 - specificity and sensitivity, respectively.

[0020] By detecting the organic substances in the exhaled breath of patients with bronchial asthma, the present invention found that there were significant differences in the contents of 21 volatile organic compounds, namely methanol, dimethylamine, 2-furanone, acetone, acetic acid, ammonium carboxylate, isoprene, acrylic acid, 1,2-dioxolane, propionic acid, cyclohexadiene, 2-methylfuran, dihydroxyacetone, toluene, methyl cyanoacetate, diacetylamine, hexanal, 1-nitro-2-propanone, butyric acid, hexanamide, and amyl nitrate, among patients with different severity grades of attacks. Further, an evaluation model for the severity of bronchial asthma attacks was established based on the metabolite contents, which can efficiently and objectively evaluate the severity of bronchial asthma attacks in patients, is very important for patients to select drug treatment plans, and has important clinical application value. Compared with the prior art, the present invention has high accuracy. By detecting the concentration of characteristic VOCs in asthma, it can avoid the dependence on the patient's subjective perception in the traditional evaluation method, which is easily affected by psychological factors and expression differences. Description of the Drawings

[0021] Figure 1 It is a schematic structural diagram of the sampling device used for sampling exhaled breath.

[0022] Figure 2It is the score plot of orthogonal partial least squares discriminant analysis, where R2Y = 0.811 > 0.8 and Q2 = 0.541 > 0.5.

[0023] Figure 3 It is the OPLS-DA modeling validation plot based on 200 random permutations.

[0024] Figure 4 It is the Box plot of the relative concentration of acetic acid, where Mip-AS represents asthma patients with mild persistent severity (Mildpersistent, Mip-AS), and Mop-AS represents asthma patients with moderate persistent severity (Moderatepersistent, Mop-AS).

[0025] Figure 5 It is the ROC curve of the differential metabolite combination, with the area under the curve being 0.992, and the sensitivity and specificity being 90.9% and 100% respectively.

[0026] Figure 6 It is the ROC curve of the differential metabolite combination after 10-fold cross-validation of the subjects, with the area under the curve AUC = 0.913, and the sensitivity and specificity being 90.9% and 100% respectively. Detailed implementation manners

[0027] The following specific embodiments illustrate the implementation manners of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification.

[0028] Before further describing the specific implementation manners of the present invention, it should be understood that the protection scope of the present invention is not limited to the following specific implementation manners; it should also be understood that the terms used in the embodiments of the present invention are for describing specific implementation manners, rather than for limiting the protection scope of the present invention. For the test methods without specific conditions noted in the following embodiments, they are usually in accordance with conventional conditions or the conditions recommended by each manufacturer. When the embodiments give a numerical range, it should be understood that unless otherwise specified in the present invention, any value between the two endpoints of each numerical range and any value between the two endpoints can be selected. Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those skilled in the art of this technology. In addition to the specific methods, equipment, and materials used in the embodiments, according to the knowledge of those skilled in the art of this technology and the records of the present invention, any methods, equipment, and materials similar or equivalent to the methods, equipment, and materials described in the embodiments of the present invention can also be used to implement the present invention.

[0029] Based on the current need for grading the severity of asthma attacks mentioned in the background art section, the present invention detected the organic substances in the exhaled breath of asthma patients with different severity grades of attacks, and found that there were significant differences in the contents of 15 volatile organic compounds, namely methanol, dimethylamine, 2-furanone, acetone, acetic acid, ammonium carboxylate, isoprene, acrylic acid, 1,2-dioxolane, propionic acid, cyclohexadiene, 2-methylfuran, dihydroxyacetone, toluene, methyl cyanoacetate, diacetylamine, hexanal, 1-nitro-2-propanone, butyric acid, hexanamide, and amyl nitrate, among patients with different severity grades of attacks. Therefore, these 15 volatile organic compounds have important clinical application value in evaluating the severity grade of asthma attacks. Further, an evaluation model for the severity of bronchial asthma attacks was established based on the content of acetic acid therein, which can efficiently and objectively evaluate the severity of bronchial asthma attacks in patients, is very important for patients to select drug treatment plans, and has important clinical application value. Based on the results of this research, the applicant proposed the technical solution of this application.

[0030] A screening method for exhaled breath metabolic markers for evaluating the severity of asthma attacks in patients, specifically including the following steps:

[0031] 1) Collect the exhaled breath of patients with mild persistent asthma attacks and patients with moderate persistent asthma attacks respectively as analysis samples;

[0032] 2) Analyze each exhaled breath sample using HPPI-TOFMS to obtain the original mass spectrometry of each exhaled breath sample;

[0033] 3) Pretreat the original mass spectrometry of each exhaled breath sample to obtain data containing metabolite relative intensity information;

[0034] 4) Perform univariate and multivariate analyses on the data containing metabolite relative intensity information, and screen out the differential metabolites among patients with different severity grades of asthma attacks according to the set thresholds (VIP>1 and FDR<0.05);

[0035] 5) Screen out the metabolite with the largest difference among patients with different severity grades of asthma attacks through binary logistic regression analysis, and this metabolite is the exhaled breath metabolic marker for evaluating the severity of asthma attacks in patients.

[0036] The screening method can also establish a receiver operating characteristic curve discrimination model based on a combination of differential metabolites, and obtain the discrimination formula and discrimination threshold for patients with mild persistent asthma attacks and patients with moderate persistent asthma attacks.

[0037] When collecting exhaled breath samples, the population to be collected from was healthy subjects and patients with a clear diagnosis of bronchial asthma. According to the GINA guideline standards, a comprehensive assessment of the attack severity grading was carried out based on parameters such as symptom frequency (number of daytime / nighttime attacks), lung function indices (percentage of FEV1 predicted value and diurnal variation rate), and history of acute exacerbation. It was required to maintain fasting, water abstinence, no smoking, and a calm state for at least 3 hours before exhaled breath collection to minimize the influence of factors such as diet and smoking on the test results. When collecting exhaled breath, it was required that the participants first take a deep nasal inhalation, and then completely exhale through the mouth and collect it into a 3L Tedlar collection bag by an exhaled breath sampling device until the bag was filled with exhaled air. The exhaled airflow rate of the participants was required to be 2700 - 3100 ml / min to eliminate the difference in metabolite concentration caused by inconsistent exhaled airflow rates. The Tedlar collection bag was sent to the laboratory within 24 hours for the next analysis.

[0038] When performing HPPI - TOFMS analysis on the exhaled breath samples to be tested, first place the gas bag in an oven and heat it to 50°C, and then enter the TOFMS ionization analysis. Connect the collection port of the Tedlar collection bag to the inlet of the mass spectrometer, and send the exhaled breath sample into the mass spectrometer through the pressure difference inside and outside the mass spectrometer. The injection flow rate is 30 - 50 ml / min. The ionization source uses a vacuum ultraviolet Kr lamp as the light source, the ionization source pressure is 200 - 300 Pa, the temperature is set at 100°C, the ionization source repulsion voltage is 18 V, and the MCP voltage is 4100 V. The spectrum acquisition time is 120 s, and the mass - to - charge ratio range for spectrum data acquisition is 0 - 400 Da.

[0039] Pre - processing the original mass spectrum of the exhaled breath samples refers to: using Matlab software to perform mass calibration and signal peak intensity extraction on the obtained original mass spectrum, and then performing total sum normalization, data filtering, and logarithmic transformation on the data of each exhaled breath sample on the MetaboAnalyst website. Finally, a data table for statistical analysis can be obtained. Among them, the first row of the table is the metabolite mass number, the first column of the table is the exhaled breath sample number information, the second column is the exhaled breath sample grouping information, and each of the remaining columns is the relative signal intensity of each metabolite.

[0040] Univariate analysis refers to the Mann - Whitney U rank - sum test, which is carried out on the Multiple Experiment Viewer software. It is considered that the differential metabolites with a false discovery rate (FDR) less than 0.05 are different between the two groups. Multivariate analysis refers to orthogonal partial least squares discriminant analysis, which is carried out on the Simca - P software. The data is transformed by unit variance scaling, and metabolites with a variable importance in the projection (VIP) greater than 1 are screened out. Finally, the metabolites common to the results of univariate analysis and multivariate analysis are taken as the differential metabolites for different attack severity grades of bronchial asthma.

[0041] Perform binary logistic regression analysis on the obtained differential metabolites using SPSS software. Set the dependent variable as the classification information of different attack severity levels of asthma patients, the covariate as the relative intensity of different metabolites, and the method as "Forward: Conditional". The metabolite ranked first among all the obtained differential metabolites is the optimal differential metabolite. Then, set the method as "Enter", select the optimal metabolite as the covariate, and perform binary logistic regression analysis again to obtain the predicted probability of this metabolite. Finally, perform receiver operating characteristic curve (ROC) analysis on the obtained differential metabolites in SPSS software. Set the predicted probability of the differential metabolite as the test variable and the grouping information as the status variable, with the value of the status variable being the disease grouping. Finally, obtain the ROC curve, where the horizontal and vertical coordinates are 1 - specificity and sensitivity, respectively.

[0042] Example 1

[0043] Next, the present invention will be further explained through the following specific embodiments, and the advantages of the present invention will be further demonstrated.

[0044] A method for screening exhaled breath metabolic markers for evaluating the attack severity of bronchial asthma patients:

[0045] 1. Research subjects

[0046] 22 asthma patients diagnosed in Zhongshan Hospital Affiliated to Dalian University, aged between 20 - 72 years, and the information is shown in Table 1.

[0047] Table 1 Patient information data table

[0048]

[0049]

[0050] 2. Exhaled breath metabolomics detection by high - pressure photoionization time - of - flight mass spectrometry

[0051] 2.1 Exhaled breath sample collection and pretreatment

[0052] Collect the exhaled breath of all patients as analysis samples; it is required to keep fasting, abstaining from smoking, and being in a calm state for at least 3 hours before exhaled breath collection to minimize the influence of factors such as diet and smoking on the test results; it is required that when collecting exhaled breath, the participant first takes a deep breath through the nose and then completely exhales through the mouth and is sampled by an exhaled breath sampling device (such as Figure 1It is collected into a Tedlar collection bag with a volume of 3 L as shown in the figure, and the collection is repeated until the bag is filled with exhaled air. The exhalation flow rate of the subjects is required to be 2700 - 3100 ml / min to eliminate the difference in metabolite concentration caused by inconsistent exhalation flow rate. The Tedlar collection bag is sent to the laboratory within 24 hours. When performing HPPI - TOFMS analysis on the exhaled breath sample to be tested, the gas bag is first heated to 50 °C in an oven and then enters the TOFMS for ionization analysis.

[0053] As Figure 1 shown, the exhaled breath sampling device used includes: a pressure measurement system, a flow measurement system, a mode switching system, and a cleaning system.

[0054] The pressure measurement system includes: a sensor holder 3, a saliva chamber 5, and a pressure sensor 18, where:

[0055] The sensor holder 3 has a hollow structure and is connected to the mouthpiece 1 through the handle 2. A saliva chamber 5 is provided below the sensor holder 3, a pressure sensor 18 is provided above it, and a high - pressure detection port 17 is provided at the end of the sensor holder 3; a sealing screw 4 is provided below the saliva chamber 5. When the exhaled breath in - line sampling device is in the cleaning mode, the sealing screw 4 is removed to clean the saliva chamber 5.

[0056] The flow measurement system includes: a gas resistance pipeline 6, a flow sensor 16, a high - pressure detection port 17, and a low - pressure detection port 15, where:

[0057] The gas resistance pipeline 6 is connected to the sensor holder 3, blocking the exhaled breath path and driving the exhaled breath to enter the flow sensor 16 through the high - pressure detection port 17; the flow sensor 16 is a differential - pressure type flow sensor, with one end connected to the high - pressure detection port 17 and the other end connected to the low - pressure detection port 15.

[0058] The mode switching system includes: a solenoid valve holder 7, a sampling port 9, an exhaust port 10, and a two - way three - way solenoid valve 14, where:

[0059] The solenoid valve holder 7 has a hollow structure. The front end of the solenoid valve holder 7 is connected to the low - pressure detection port 15. The two - way three - way solenoid valve 14 is arranged inside the solenoid valve holder 7. The two - way three - way solenoid valve 14 includes two paths, NO and NC. In the detection mode, NO is closed and NC is open, connecting the sampling port 9. In the cleaning mode, NC is closed and NO is open, connecting the exhaust port 10.

[0060] The cleaning system includes: a three - way pipeline 11, a cleaning pump 13, and a normally open two - way solenoid valve 12, where:

[0061] The three ports of the three-way pipe 11 are respectively connected to the exhaust port 10, the cleaning pump 13, and the normally open two-way solenoid valve 12; before sampling, the exhaled breath online sampling device evacuates the gas in the exhaled breath online sampling device through the normally open two-way solenoid valve 12;

[0062] The cleaning pump 13 is used in the cleaning mode to blow high-pressure gas from the cleaning system through the mode switching system, the flow measurement system, the pressure measurement system, the handle 2, and out of the nozzle 1 to complete the cleaning.

[0063] Before sampling, the NC of the two-position three-way solenoid valve 14 is closed and the NO is opened to evacuate the gas in the exhaled breath online sampling device through the normally open two-way solenoid valve 12; then the NO of the two-position three-way solenoid valve 14 is closed and the NC is opened;

[0064] Blow air into the exhaled breath online sampling device through the nozzle 1, and the exhaled air flows through the pressure sensor 18 to measure the exhaled breath pressure, and the saliva is concentrated and collected into the saliva bin 5;

[0065] The exhaled breath flows from the high-pressure detection port 17 to the low-pressure detection port 15, and the exhaled breath flow rate is measured by the flow sensor 16;

[0066] The exhaled breath flows from the low-pressure detection port 15 to the solenoid valve holder 7, and flows through the passage NO to the sampling port 9, and is collected by the sample container 8a. The sample container 8a is a 3L Tedlar collection bag.

[0067] The exhaled breath flow rate is controlled to be 2700 - 3100 ml / min by the flow measurement system.

[0068] 2.2 High-pressure photoionization time-of-flight mass spectrometry detection of exhaled breath

[0069] For mass spectrometry detection, a self-developed high-pressure photoionization time-of-flight mass spectrometer is used. The sampling port of the Tedlar collection bag is connected to the inlet of the mass spectrometer, and the exhaled breath sample is sent into the mass spectrometer through the air pressure difference inside and outside the mass spectrometer. Its injection flow rate is 45 ml / min. The ionization source uses a vacuum ultraviolet Kr lamp as the light source, the temperature is set at 100 °C, the ionization source air pressure is 200 - 300 Pa, the ionization source repulsion voltage is 18 V, and the MCP voltage is 4100 V; the spectrum acquisition time is 120 s, and the mass-to-charge ratio range for spectrum data acquisition is 0 - 400 Da.

[0070] 3. Data preprocessing

[0071] The Matlab software is used to perform mass calibration and signal peak intensity extraction on the obtained original mass spectrum, and then on the MetaboAnalyst website, the data of each exhaled breath sample is subjected to total normalization, data filtering, and logarithmic transformation. Finally, a discriminant analysis score map based on orthogonal partial least squares method is obtained, as Figure 2As shown, the sample groups that can be well separated are the mild persistent patient group (11 people) and the moderate persistent patient group (11 people). Q2Y = 0.811 > 0.8, Q2 = 0.541 > 0.5, indicating that the discriminant model established above has relatively good reliability.

[0072] 4. Differential metabolite screening

[0073] Based on the obtained data table containing metabolite relative intensity information, it was imported into Simca software for orthogonal partial least squares-discriminant analysis (OPLS-DA). The OPLS-DA score plot is as Figure 2 shown. It can be seen the differences and classification trends in the metabolic profiles between asthma patients and healthy people. In the OPLS-DA score plot, R2Y = 0.811 > 0.8, Q2 = 0.541 > 0.5, indicating that the model is reliable and has good predictive ability.

[0074] In addition, 200 permutation validations were performed, and the results are as Figure 3 shown (R2 = 0.227 < 0.4, Q2 = -0.207 < 0.15), indicating that overfitting did not occur during OPLS-DA analysis.

[0075] Then the Multiple Experiment Viewer software was used to perform the Mann-Whitney U test to screen for differential metabolites. In the OPLS-DA analysis, metabolites with variable importance in projection (VIP) greater than 1 were selected, and in the Mann-Whitney U test, metabolites with a false discovery rate (FDR) less than 0.05 were selected.

[0076] Finally, 21 differential metabolites were screened out, as shown in Table 2.

[0077] Table 2 Information data table of 21 differential metabolites

[0078]

[0079]

[0080] 5. Establishment of differential metabolite combination model

[0081] Based on the 21 selected differential metabolites, binary logistic regression analysis was performed using the logistic regression analysis function of SPSS software to select the optimal differential metabolite. The dependent variable was set as the grouping information of healthy people and disease patients, the covariate was the relative intensity of different metabolites, and the method was "Forward: Conditional". The metabolite ranked first among all the metabolites obtained was the optimal differential metabolite. Then, the method was set to "Enter", the optimal metabolite was selected as the covariate, and binary logistic regression analysis was performed again to obtain the predicted probability of this combination. Finally, acetic acid (m / z 79.038) was the optimal differential metabolite, and its relative concentration between asthma patients with mild persistent and moderate persistent attack severity was as Figure 4 shown.

[0082] In SPSS software, receiver operating characteristic curve (ROC) analysis was performed on the obtained optimal differential metabolite. The predicted probability of acetic acid (m / z 79.038) was set as the test variable, the grouping information was the status variable, and the value of the status variable was the disease grouping. The results were as Figure 5 shown. The area under the ROC curve (AUC) value was 0.983, and the sensitivity and specificity were 90.9% and 100% respectively. The threshold corresponding to the maximum sum of sensitivity and specificity was the discrimination threshold, and the discrimination threshold was 0.702. The discrimination formula was y = exp(x / (x + 1)), where x = 0.012*I 乙酸 - 7.519, where I was the relative intensity of the metabolite acetic acid and y was the obtained threshold. When the obtained threshold was higher than 0.702, the sample was an asthma patient with mild persistent attack, and when the threshold was lower than 0.702, the sample was an asthma patient with moderate persistent attack.

[0083] The model was internally validated by the method of 10-fold cross-validation. The results were as Figure 6 shown. The area under its ROC curve (AUC) = 0.913, and the sensitivity and specificity were 90.9% and 100% respectively. This result further proved that the model was well established and could better distinguish between the moderate persistent patient group and the mild persistent patient group.

[0084] 6. Conclusion

[0085] As can be seen from the above verification, the model constructed by the method of the present invention can efficiently and objectively evaluate the severity of asthma attacks in patients, which is very important for patients to select drug treatment plans and has important clinical application value.

Claims

1. An exhaled gas metabolic marker for evaluating the severity of asthma attacks in patients, characterized in that, The marker is acetic acid (m / z 79.038).

2. The exhaled gas metabolic marker for evaluating the severity of asthma attacks in patients according to claim 1, wherein: The severity of the attack refers to mild persistent attack and moderate persistent attack of bronchial asthma.

3. A screening method for exhaled metabolite markers for evaluating the severity of asthma attacks in patients with bronchial asthma, characterized in that, The screening method includes: (1) Collect the exhaled breath of patients with mild persistent bronchial asthma and patients with moderate persistent bronchial asthma respectively as analysis samples; (2) Analyze each exhaled breath sample using high-pressure photoionization time-of-flight mass spectrometry (HPPI-TOFMS) to obtain the original mass spectrometry diagram of each exhaled breath sample; (3) Pretreat the original mass spectrometry diagram of each exhaled breath sample to obtain data containing metabolite relative intensity information; (4) Perform univariate and multivariate analyses on the data containing metabolite relative intensity information, and screen out the differential metabolites between patients with different severity levels of bronchial asthma according to the set thresholds (VIP>1 and FDR<0.05); (5) Screen out the metabolite with the largest difference through binary logistic regression analysis, and this metabolite is the exhaled breath metabolite marker for evaluating the severity of the attack in patients with bronchial asthma.

4. The method according to claim 3, wherein: The screening method can also establish a receiver operating characteristic curve discrimination model based on the exhaled breath metabolite marker, and obtain the discrimination formula and discrimination threshold for patients with mild persistent bronchial asthma and patients with moderate persistent bronchial asthma.

5. The method according to claim 3, characterized in that: When collecting the exhaled breath sample of the patient, 1) When collecting the exhaled breath sample, the collected population is patients with clearly diagnosed mild persistent bronchial asthma and patients with moderate persistent bronchial asthma, and there are acute symptoms; 2) The subject should fast, abstain from smoking and keep calm for at least 3 hours before collecting the exhaled breath; 3) It is required that when collecting the exhaled breath, the subject first takes a deep breath through the nose, and then completely exhales through the mouth into a Tedlar collection bag with a volume of 3L by the exhaled breath sampling device. Set the exhaled air flow rate to 2700-3100 ml / min, and the Tedlar collection bag is sent to the laboratory within 24 hours for high-pressure photoionization time-of-flight mass spectrometry (HPPI-TOFMS) analysis. 4) The exhaled breath samples of patients with mild persistent bronchial asthma and patients with moderate persistent bronchial asthma are all taken from people aged 20-72; there are more than 10 exhaled breath samples for each type of patient.

6. The method according to claim 3, characterized in that: When analyzing each exhaled breath sample using high-pressure photoionization time-of-flight mass spectrometry (HPPI-TOFMS), first place the air bag in an oven and heat it to 50°C, then enter the TOFMS ionization analysis. Set the exhaled air flow rate to 2800-3200 ml / min, the ionization source uses a vacuum ultraviolet Kr lamp as the light source, the ionization source pressure is 200-300 Pa, the temperature is set to 100°C, the ionization source repulsion voltage is 18V, and the MCP voltage is 4100V; the spectrum acquisition time is 120s, and the mass-to-charge ratio range for spectrum data acquisition is 0-400 Da.

7. The method according to claim 3, wherein: Preprocessing the original mass spectrometry of exhaled breath samples means: using Matlab software to perform mass correction and signal peak intensity extraction on the obtained original mass spectrometry, and then performing total normalization, data filtering, and logarithmic transformation on the data of each exhaled breath sample on the MetaboAnalyst website. Finally, a data table for statistical analysis is obtained. Among them, the first row of the table is the metabolite mass number, the first column of the table is the exhaled breath sample number information, the second column is the exhaled breath sample grouping information, and each of the remaining columns is the relative signal intensity of each metabolite.

8. The method according to claim 3, characterized in that: Performing univariate and multivariate analyses on the data containing metabolite relative intensity information means: Univariate analysis refers to the Mann-Whitney U rank sum test, which is performed on the MultipleExperimentViewer software. Differentially expressed metabolites with a false discovery rate (FDR) less than 0.05 are considered to be different between the two groups. Multivariate analysis refers to orthogonal partial least squares discriminant analysis, which is performed on the Simca-P software. The data is transformed by unit variance scaling, and metabolites with a variable importance in the projection (VIP) greater than 1 are selected. Finally, the metabolites common to the results of univariate analysis and multivariate analysis are taken as the differentially expressed metabolites for different severity grades of bronchial asthma attacks.

9. The method according to claim 3, wherein: Screen out the metabolite with the largest difference through binary logistic regression analysis. Perform binary logistic regression analysis on the obtained differentially expressed metabolites in the SPSS software. Set the dependent variable as the grouping information of patients with mild persistent bronchial asthma and moderate persistent bronchial asthma, and the covariates as the relative intensities of different metabolites. The method is "Forward: Conditional". The metabolite ranked first is the optimal differentially expressed metabolite, and this differentially expressed metabolite is the exhaled breath metabolite combination biomarker for diagnosing asthma.

10. According to the method described in claim 3, wherein: Establishing a receiver operating characteristic curve discrimination model based on differentially expressed metabolites means: Perform binary logistic regression analysis on the obtained differentially expressed metabolites in the SPSS software. Set the dependent variable as the grouping information of different severity grades of asthma attacks, and the covariates as the relative intensities of different metabolites. The method is "Forward: Conditional". The metabolite ranked first among all the obtained differentially expressed metabolites is the optimal differentially expressed metabolite. Then, set the method as "Enter", select the optimal metabolite as the covariate, and perform binary logistic regression analysis again to obtain the predicted probability of this metabolite. Finally, perform receiver operating characteristic curve (ROC) analysis on the obtained differentially expressed metabolites in the SPSS software. Set the predicted probability of the differentially expressed metabolite as the test variable, and the grouping information as the status variable. The value of the status variable is the disease grouping. Finally, an ROC curve is obtained, and the abscissa and ordinate are 1 - specificity and sensitivity, respectively.

Images