阅读说明：本技术 一种用于早期体腔内感染性并发症诊断的标记物及方法 (Marker and method for early diagnosis of infectious complications in body cavity ) 是由 季加孚 李子禹 吴舟桥 石晋瑶 陕飞 于 2019-11-28 设计创作，主要内容包括：本发明涉及一种体腔内感染性并发症早期诊断的标记物,以及相关的试剂盒和装置,同时本发明构建出一种对腔内感染性并发症的评分方法、对其危险等级的检测方法,以及提供一种筛选体腔内感染性并发症的标记物的方法。本发明实现了对体腔内感染的早期诊断,能够有效地指导临床干预,降低患者围手术期死亡率,具有极高的临床应用价值。(The invention relates to a marker for early diagnosis of infectious complications in body cavities, and a related kit and device, and also provides a method for scoring the infectious complications in the cavities, a method for detecting the risk level of the infectious complications in the cavities, and a method for screening the marker for the infectious complications in the body cavities. The invention realizes early diagnosis of infection in the body cavity, can effectively guide clinical intervention, reduces the death rate of patients in the perioperative period, and has extremely high clinical application value.)

1. A marker for early diagnosis of infectious complications in body cavities, characterized in that: the diagnostic marker comprises a series of inflammatory factors, the inflammatory factors are selected from one or a combination of more than two of cytokines, matrix metalloproteases, reactive oxygen species, vascular endothelial growth factors, tissue metalloprotease inhibitors, C-reactive proteins, leukocyte counts and the like, preferably the inflammatory factors are selected from one or a combination of more than two of cytokines and matrix metalloproteases, preferably the cytokines are selected from the group consisting of: one or more of interleukins, colony stimulating factors, interferons, tumor necrosis factors, chemokines or growth factors, and further preferably the inflammatory factor is selected from the group consisting of: one or more of interleukins, interferons, tumor necrosis factors or matrix metalloproteinases, and more preferably the inflammatory factor is selected from the group consisting of: one or more of IL-1 beta, IL-6, IL-10, TNF-alpha, MMP-2 and MMP-9, preferably one or more of IL-1 beta, IL-6, IL-10 and MMP-9, and more preferably one or more of IL-1 beta, IL-10 and MMP-9.

2. The marker composition for early diagnosis of infectious complications in body cavities according to claim 1, characterized in that: the inflammatory factor is derived from liquid in a body cavity, such as liquid in an abdominal cavity, a pelvic cavity, a thoracic cavity environment or a brain cavity, preferably the inflammatory factor is derived from liquid in the abdominal cavity, the pelvic cavity and the thoracic cavity environment or cerebrospinal fluid, further preferably the inflammatory factor is derived from lacunar effusion or liquid obtained by drainage, and preferably the abdominal effusion or the abdominal drainage fluid.

3. The marker composition for early diagnosis of infectious complications in body cavities according to any one of claims 1-2, characterized in that: the inflammatory factor is the inflammatory factor at 0-15 days after operation; preferably inflammatory factors on days 0-7 post-operatively; preferably inflammatory factors at day 0-5 post-surgery; preferably the inflammatory factor is present on days 0-3 after surgery, more preferably the inflammatory factor is present on day 0, 1, 2 or 3 after surgery.

4. The marker composition for early diagnosis of infectious complications in body cavities according to claims 1-3, characterized in that: the infection complication in the body cavity is selected from infection complications of abdominal cavity, pelvic cavity, thoracic cavity or brain cavity, more preferably the infection complications in the body cavity is selected from abdominal cavity infection, abdominal dropsy, peritonitis, celiac abscess, sepsis, anastomotic leakage, pancreatic fistula, duodenal stump fistula, other digestive tract fistula, Pedi superior to lymphatic fistula or chylomic fistula, and more preferably the infection complications in the body cavity is anastomotic leakage.

5. A kit for early diagnosis of infectious complications in body cavities, characterized in that: comprising a detection reagent for detecting the marker composition of any one of claims 1 to 4.

6. Use of a marker according to any of claims 1 to 4 for the preparation of a detection reagent or a detection kit for the early diagnosis of infectious complications in body cavities, preferably comprising a detection reagent or a detection device for the detection of said marker.

7. A scoring method for detecting an infectious complication in a body cavity of a subject, comprising the steps of:

1) detecting the markers of claims 1-4 and their amounts from a sample of a subject, and

2) calculating the content of the marker measured in the step 1 through a model 1 to obtain a value Y, namely a score of the infectious complications in the body cavity;

wherein the statistical model 1 is: y ═ X1 β 1+ X2 β 2+ … … + Xn β n + epsilon,

wherein Y is a score of the infectious complication score in the body cavity, X is the content of a certain marker at a certain time point, beta is a coefficient and is the weight of a corresponding variable X in the infectious complication score in the body cavity, epsilon is a constant, preferably the statistical method is LASSO regression, further preferably the object is a patient with a disease related to a body cavity organ or tissue, preferably the object is a postoperative patient with an abdominal cavity, a pelvic cavity, a thoracic cavity environment or a cranial cavity related disease, further preferably the object is a gastrointestinal postoperative patient, such as a patient subjected to an abdominal cavity operation, and further preferably the object is a patient clinically collecting abdominal cavity drainage fluid.

8. A method for screening a marker for diagnosing an infectious complication in a body cavity of a subject, comprising the steps of:

1) detecting the markers of claims 1-4 and their amounts from a sample of a subject, and

2) carrying out statistical analysis on the content of the marker determined in the step 1 through a model 1;

wherein, the statistical model 1 is: y ═ X1 β 1+ X2 β 2+ … … + Xn β n + epsilon, Y is the score of the infectious complication score in the body cavity, X refers to the content of a certain marker at a certain point in time, beta is a coefficient, which is the weight of the corresponding variable X in the score of the infectious complication in the body cavity, epsilon is a constant;

wherein, the statistical analysis method is preferably LASSO regression, X1 … Xn retained in the model 1 obtained by regression is the diagnostic marker obtained by screening, preferably, the object is a patient with coelomic organ or tissue related diseases, further preferably, the object is a postoperative patient with abdominal cavity, pelvic cavity, thoracic cavity environment or cranial cavity related diseases, further preferably, the object is a gastrointestinal postoperative patient, such as a patient after abdominal cavity operation, and most preferably, the object is a patient for clinically collecting abdominal cavity drainage fluid.

9. A method for detecting the risk level of an infectious complication in a body cavity in a subject, comprising the steps of:

1) detecting the markers of claims 1-4 and their amounts from a sample of a subject, and

2) calculating the content of the marker measured in the step 1 through a model 1 to obtain a value Y;

wherein the statistical model 1 is: y ═ X1 β 1+ X2 β 2+ … … … + Xn β n + epsilon,

wherein Y is the score of the infection complication in the body cavity, X is the content of a certain marker at a certain time point, beta is a coefficient which is the weight of a corresponding variable X in the score of the infection complication in the body cavity, and epsilon is a constant;

wherein the values of β and ε are determined by a statistical method, preferably LASSO regression;

3) comparing the calculated value Y with a corresponding reference value, wherein when Y is greater than the reference value, the risk level of the infectious complications in the body cavity is relatively high; when Y is less than or equal to the reference value, the risk level of the infectious complications in the body cavity is relatively low, preferably, the object is a patient with a disease related to organs or tissues of the body cavity, further preferably, the object is a postoperative patient with a disease related to the abdominal cavity, the pelvic cavity, the thoracic cavity environment or the cranial cavity, further preferably, the object is a gastrointestinal postoperative patient, such as a patient subjected to an abdominal cavity operation, and most preferably, the object is a patient for clinically collecting abdominal cavity drainage fluid.

10. A method for detecting the probability of risk of developing an infectious complication in a body cavity in a subject, comprising the steps of:

1) detecting the markers of claims 1-4 and their amounts from a sample of a subject, and

2) calculating the content of the marker measured in the step 1 through a model 1 to obtain a value Y;

wherein the statistical model 1 is: y ═ X1 β 1+ X2 β 2+ … … … + Xn β n + epsilon,

wherein Y is the score of the infection complication in the body cavity, X is the content of a certain marker at a certain time point, beta is a coefficient which is the weight of a corresponding variable X in the score of the infection complication in the body cavity, and epsilon is a constant;

wherein the values of β and ε are determined by a statistical method, preferably LASSO regression;

3) recording and analyzing relevant clinical factors;

4) obtaining the occurrence risk probability p of the infectious complications in the body cavity through the analysis and calculation of the model 2;

wherein the model 2 is:

is established by Logistic regression, wherein, the independent variable x₁，x₂…x_nThe index refers to indexes such as various clinical factors and score values of the infectious complications in the body cavity, wherein w is a coefficient or weight, x is an independent variable, namely an index included in a formula, and the value of g (x) is calculated, so that the occurrence risk probability p of the infectious complications in the body cavity is obtained, preferably, the object is a patient with diseases related to body cavity organs or tissues, further preferably, the object is a postoperative patient with diseases related to abdominal cavity, pelvic cavity, thoracic cavity environment or cranial cavity, and further more preferablyIt is further preferred that the subject is a post-gastrointestinal patient, such as a patient undergoing abdominal surgery, and most preferably that the subject is a patient from whom an abdominal drainage fluid is clinically collected.

11. An apparatus for early diagnosis of infectious complications in a body cavity, characterized by: the device comprises an analysis unit 1, an analysis unit 2 and an analysis unit 3, wherein

The analytical unit 1 is used for detecting the markers and their corresponding amounts according to claims 1-4;

the analysis unit 2 is used for obtaining an analysis calculation result Y from one or more measured quantities obtained in the analysis unit 1 through the model 1;

the analysis unit 3 is used for comparing the calculation result Y in the analysis unit 2 with corresponding reference values to obtain the risk level of the infectious complications in the body cavity;

preferably, further comprising an analysis unit 4 and/or an analysis unit 5, wherein:

the analysis unit 4 is used for recording and analyzing relevant clinical factors;

the analysis unit 5 is used for combining the risk level obtained by the analysis unit 3 with the clinical factors in the analysis unit 4, and obtaining the risk probability p of occurrence of the infectious complications in the body cavity through analysis and calculation of the model 2;

each analysis unit comprises a corresponding computer-implemented algorithm, preferably, the object is a patient with coelomic organ or tissue related diseases, further preferably, the object is a postoperative patient with abdominal cavity, pelvic cavity, thoracic cavity environment or cranial cavity related diseases, further preferably, the object is a gastrointestinal postoperative patient, such as a patient after an abdominal cavity operation, and most preferably, the object is a patient for clinically collecting abdominal cavity drainage fluid.

Technical Field

The application relates to the field of biological medical treatment, in particular to a marker, a method and a device for early diagnosis of infectious complications in body cavities.

Background

Local infectious complications after the operation of the body cavity visceral organs are important factors for restricting postoperative rehabilitation of patients. Such as stoma leaks, which are one of the most serious postoperative complications of intra-abdominal surgery, such as gastrointestinal surgery, and which occur in connection with impaired healing of the stoma after reconstruction of the digestive tract. Once anastomotic leakage occurs, pelvic infection, abscess, panperitonitis and even sepsis can cause perioperative death of the patient. According to the report of the literature, the incidence rate of anastomotic leakage after gastric surgery is between 1 and 6 percent, and the incidence rate of colorectal anastomotic leakage is between 4 and 33 percent. Anastomotic leakage seriously affects the postoperative safety of patients and imposes a serious burden on the medical health system. From foreign documents and domestic summarized data, about one third of perioperative deaths are directly related to anastomotic leakage and are the leading cause of perioperative deaths of gastrointestinal tract surgery patients at present. Anastomotic leakage prolongs the hospitalization time and the hospitalization period economic expenses of the patient. Although there is still a lack of health economics analysis for stoma leakage in China, stoma leakage adds significantly to the cost of a patient hospitalization from U.S. relevant data analysis, with a stoma leakage of about $ 4.6, a colorectal leak of about $ 3.4, about 3 times as high as other patients, and a hospitalization time of 2 times as long as other patients. Even if the condition of the patient is relieved after treatment, the postoperative life quality of the patient with anastomotic leakage is obviously worse than that of other patients. Anastomotic leakage is also a high-risk factor of local tumor recurrence, and seriously affects the treatment effect and the survival time limit of tumor patients.

For local infectious complications of body cavities, clinical routine laboratory examinations have limited diagnostic efficacy. Leukocyte counts and CRP are widely used as routine laboratory test items in clinical diagnosis of infectious complications after surgery. In recent years, a great deal of research is carried out to find whether a certain indication relationship exists between laboratory examination items and the occurrence of stomal leakage so as to be used for early diagnosis of body cavity infectious complications such as stomal leakage or celiac abscess. However, most of the current relevant research results do not show satisfactory results, and the clinical values of the laboratory indexes such as white blood cell count, C-reactive protein, troponin and the like which are commonly used in clinic for predicting anastomotic leakage are limited, so that the diagnosis of a disease gradually changed from a local lesion to a systemic infection is difficult only through the systemic infection expression.

In recent years, a great deal of research is attempted to explore the risk factors of anastomotic leakage after gastrointestinal surgery. Two multicenter retrospective studies from the netherlands analyzed clinical data for 36900 and 600 patients with postoperative treatment for inflammatory bowel disease, respectively, and found that obesity and a high American Society of Anesthesia (ASA) score are independent risk factors for anastomotic leakage. A study from multiple domestic centers has collected data from over 300 patients with low resection in the rectum to show that men, diabetes, preoperative chemoradiotherapy and tumor location are independent risk factors for anastomotic leakage. One study from japan analyzed the independent risk factors for anastomotic leakage after gastrectomy with laparoscopic assistance of gastric cancer, and it was thought that patients with lower Prognostic Nutritional Index (PNI) were more likely to develop anastomotic leakage. A domestic research report states that high age (age is more than or equal to 65 years), anemia (hemoglobin is less than or equal to 8.0g/dL) and malnutrition are independent high-risk factors of stomal leakage after gastric cancer operation. Other risk factors for anastomotic leakage after gastrointestinal surgery include excessive smoking, low-grade anastomosis, advanced cancer surgery, emergency surgery, massive blood loss, long surgery time, and preoperative hormone use. However, a slight analysis reveals that in practice most patients will have one or more of the above-mentioned risk factors, and thus merely listing the risk factors may not be of practical utility for clinical work. For this type of research, there is a troublesome problem: these risk factors and related studies do not really provide a corresponding solution. When a patient with a plurality of risk factors is clinically met, whether to continue the operation or change the operation scheme still has insufficient clinical evidence to directly support, and a solution is provided.

At present, clinical prevention of anastomotic leakage has three aspects of preoperative operation, intraoperative operation and postoperative operation. Preoperative prevention refers to preoperative assessment, and a patient with high risk factors is selected for targeted treatment, so that the patient is guaranteed to receive full preoperative preparation, including preoperative smoking cessation, correction of water electrolyte disorder and hypoproteinemia, full treatment of a diabetic or anemia patient, preoperative 3-day oral administration of antibiotics and the like. The intraoperative prevention comprises fine operation in an operation, ensures good blood circulation of an anastomotic stoma, has no tension and the like, and is essentially the quality control and optimization of an operation layer. In addition, in clinic, a team uses the albumin glue to reinforce the anastomotic stoma to reduce the incidence rate of anastomotic leakage, but after comprehensive analysis, the albumin glue is found to have no exact effect on preventing anastomotic leakage, and meanwhile, due to high price, the application of the albumin glue in clinic is limited. The postoperative prevention mainly comprises postoperative preventive use of antibiotics, routine imaging examination such as gastrointestinal angiography and the like.

In contrast, early diagnosis of anastomotic leakage is also one of the prevention strategies for postoperative anastomotic leakage. Most of the existing diagnostic methods are routine laboratory tests, in which leukocyte counts and CRP are widely used as routine laboratory test items for clinical diagnosis of infectious complications after surgery. In recent years, there have been numerous studies attempting to find out whether there is some kind of indication relationship between laboratory examination items and the occurrence of stomal leakage in order to be used for early diagnosis of stomal leakage or celiac abscess. However, most of the current relevant research results do not show satisfactory results, and the clinical values of the laboratory indexes such as white blood cell count, C-reactive protein, troponin and the like which are commonly used in clinic for predicting anastomotic leakage are limited, so that the diagnosis of a disease gradually changed from a local lesion to a systemic infection is difficult only through the systemic infection expression. Meanwhile, the clinical diagnosis and treatment effect on anastomotic leakage is not satisfactory, and doctors can only wait until the anastomotic leakage is serious to a certain extent and obvious laboratory index abnormity appears and can confirm the diagnosis by combining with imaging or endoscopy. Thus, most anastomotic leaks are currently diagnosed 5-8 days or even later post-operatively, and about half of the patients diagnosed with a leak need to be treated again by surgery. It follows that the greatest obstacle to stomal leak diagnosis is the lack of "early" and "local" monitoring means.

Disclosure of Invention

Aiming at the necessity and the urgency of early diagnosis of the infectious complications in the body cavity, the invention provides a set of reliable, convenient and timely novel clinical early diagnosis method of the infectious complications in the body cavity, and a diagnostic marker, a diagnostic kit, a diagnostic device, a diagnostic method and the like related to the method.

The invention adopts the following technical scheme:

in a first aspect, there is provided a diagnostic marker comprising a panel of inflammatory factors selected from one or a combination of two or more of cytokines, matrix metalloproteinases, reactive oxygen species, vascular endothelial growth factor, tissue metalloproteinase inhibitor, C-reactive protein, white blood cell count, and the like, which marker is useful in diagnosing infectious complications in body cavities in a test sample or test site.

In a second aspect, a kit for early diagnosis of infectious complications in a body cavity is provided, for quickly, conveniently and timely diagnosing whether a patient is at risk of developing infectious complications in a body cavity, the kit comprising a detection reagent and/or a detection device for detecting an inflammatory factor.

In a third aspect, there is provided a use of a diagnostic marker for the preparation of a detection reagent or a detection kit for the early diagnosis of infectious complications in body cavities; the kit comprises a detection reagent or a detection device for detecting the inflammatory factor.

In a fourth aspect, there is provided a scoring method for detecting an infectious complication in a body cavity of a subject, for scoring a risk of the subject to develop the infectious complication in the body cavity, the method comprising the steps of:

1) detecting a marker and its amount from a sample of a subject, and

2) calculating the content of the marker measured in the step 1 through a model 1 to obtain a value Y, namely a score of the infectious complications in the body cavity;

wherein the statistical model 1 is: y ═ X1 β 1+ X2 β 2+ … … … + Xn β n + epsilon,

wherein Y is the score of the body cavity infectious complication score, X is the content (such as concentration) of a certain marker at a certain time point, beta is a coefficient which is the weight of a corresponding variable X in the body cavity infectious complication score, and epsilon is a constant.

Wherein the values of β and ε are determined by a statistical method, preferably LASSO regression.

A scoring system for intracavity infectious complications of constructs was developed by selecting the best index from inflammatory factors at different time points after surgery using a Least Absolute Shrinkage and Selection Operator (LASSO) regression. The correlation calculation is done by the R language "glmnet" package. LASSO regression is commonly used to construct high-latitude prediction models. The method uses L1 regularization penalty correction to shrink some regression coefficients to exactly zero. LASSO regression selects the best parameters from the high dimensional data, and over-fitting can be avoided while the accuracy of the prediction model is considered.

In a fifth aspect, there is provided a method for screening a marker for diagnosing an infectious complication in a body cavity of a subject, the method comprising the steps of:

1) detecting a marker and its amount from a sample of a subject, and

2) carrying out statistical analysis on the content of the marker determined in the step 1 through a model 1;

wherein, the statistical model 1 is: y ═ X1 β 1+ X2 β 2+ … … + Xn β n + e, Y is the score of the infectious complication score in the body cavity, X refers to the content (e.g. concentration) of a certain marker at a certain time point, β is a coefficient, which is the weight of the corresponding variable X in the score of the infectious complication in the body cavity, e is a constant;

wherein, the statistical analysis method is preferably LASSO regression, and X1 … Xn remained in model 1 obtained by regression is the diagnostic marker obtained by screening.

In a sixth aspect, there is provided a method for detecting a risk level of a subject developing an intra-body cavity infectious complication, for assessing the risk level of the subject developing the intra-body cavity infectious complication, the method comprising the steps of:

1) detecting a marker and its amount from a sample of a subject, and

2) calculating the content of the marker measured in the step 1 through a model 1 to obtain a value Y;

wherein the statistical model 1 is: y ═ X1 β 1+ X2 β 2+ … … … + Xn β n + epsilon,

wherein Y is the score of the body cavity infectious complication score, X is the content (such as concentration) of a certain marker at a certain time point, beta is a coefficient which is the weight of a corresponding variable X in the body cavity infectious complication score, and epsilon is a constant;

wherein the values of β and ε are determined by a statistical method, preferably LASSO regression;

3) comparing the calculated value Y with a corresponding reference value, wherein when Y is greater than the reference value, the risk level of the infectious complications in the body cavity is relatively high; when Y is equal to or less than the reference value, the risk level of infectious complications in the body cavity is relatively low.

Wherein the corresponding reference value is calculated by a statistical method, preferably a Cut-Off value, i.e. by ROC analysis. The receiver operating characteristic curve (ROC curve) is used for measuring the diagnostic efficacy of the diagnostic Index, and the maximum value of the Youden Index (Youden Index) is set as the optimal Cut-Off value of the ROC curve, and the Youden Index is calculated by the formula: jotan index-sensitivity + specificity-1.

In a seventh aspect, there is provided a method for detecting a risk probability of developing an infectious complication in a body cavity in a subject, for evaluating the risk probability of developing an infectious complication in a body cavity in the subject, the method comprising the steps of:

1) detecting a marker and its amount from a sample of a subject, and

2) calculating the content of the marker measured in the step 1 through a model 1 to obtain a value Y;

wherein the statistical model 1 is: y ═ X1 β 1+ X2 β 2+ … … … + Xn β n + epsilon,

wherein Y is the score of the body cavity infectious complication score, X is the content (such as concentration) of a certain marker at a certain time point, beta is a coefficient which is the weight of a corresponding variable X in the body cavity infectious complication score, and epsilon is a constant;

wherein the values of β and ε are determined by a statistical method, preferably LASSO regression;

3) recording and analyzing relevant clinical factors;

4) obtaining the occurrence risk probability p of the infectious complications in the body cavity through the analysis and calculation of the model 2;

wherein the model 2 is:

is established by Logistic regression, wherein, the independent variable x₁，x₂…x_nThe index refers to indexes such as various clinical factors and score values of the infectious complications in the body cavity, wherein w is a coefficient or weight, x is an independent variable, namely an index included in a formula, and the value of g (x) is calculated, so that the occurrence risk probability p of the infectious complications in the body cavity is obtained.

Wherein, whether the anastomotic leakage occurs or not is a classification variable, 0 represents 'none', 1 represents 'present', and the risk probability of occurrence of the infectious complications in the body cavity obtained by the model is between 0 and 1. In the formula, g (x) is a continuous variable, and the value range of 0-1 is not applicable, so that the g (x) is converted into the risk probability between 0 and 1 by using Logistic transformation. The analysis and calculation are completed by SPSS statistical software.

In an eighth aspect, there is provided an apparatus for early diagnosis of infectious complications in a body cavity, the apparatus comprising an analysis unit 1, an analysis unit 2, and an analysis unit 3, wherein:

the analysis unit 1 is used for detecting the diagnostic markers and their corresponding contents (e.g. concentrations);

the analysis unit 2 is used for obtaining an analysis calculation result Y from one or more measured quantities obtained in the analysis unit 1 through the model 1;

the analysis unit 3 is used for comparing the calculation result Y in the analysis unit 2 with corresponding reference values to obtain the risk level of the infectious complications in the body cavity;

preferably, further comprising an analysis unit 4 and/or an analysis unit 5, wherein:

the analysis unit 4 is used for recording and analyzing relevant clinical factors;

the analysis unit 5 is used for combining the risk level obtained by the analysis unit 3 with the clinical factors in the analysis unit 4, and obtaining the risk probability p of occurrence of the infectious complications in the body cavity through analysis and calculation of the model 2.

Each of the analysis units contains a corresponding computer-implemented algorithm.

Further, in the analyzing unit 1, the content (e.g., concentration) of the diagnostic marker may be inputted; or one or more detection reagents or kits for detecting the amount (e.g.concentration) of said marker, for determining the amount of the corresponding marker in the sample;

further, in the analysis unit 2, the statistical model 1 is: y ═ X1 β 1+ X2 β 2+ … … … + Xn β n + epsilon, where Y is the score of the infectious complication score in the body cavity, X refers to the content (e.g. concentration) of a certain marker at a certain point in time, beta is a coefficient, which is the weight of the corresponding variable X in the score of the infectious complication in the body cavity, epsilon is a constant;

furthermore, in the analysis unit 3, the corresponding reference value is calculated by a statistical method, preferably a Cut-Off value, that is, the maximum value of the john index obtained by ROC analysis; comparing the calculation result Y obtained by the analysis unit 2 with a corresponding reference value, wherein when Y is larger than the reference value, the risk level of the infectious complications in the body cavity is relatively high; when Y is less than or equal to the reference value, the risk level of the infectious complications in the body cavity is relatively low;

further, in the analysis unit 5, the model 2 is:

wherein the independent variable x₁，x₂…x_nThe indexes such as various clinical factors, score values of infectious complications in body cavities and the like are indicated, wherein w is a coefficient or weight and is obtained through Logistic regression; x is an independent variable, namely an index included in the formula, and the value of g (x) is calculated, so that the risk probability p of occurrence of the infectious complications in the body cavity is obtained.

Wherein, whether the anastomotic leakage occurs or not is a classification variable, 0 represents 'none', 1 represents 'present', and the risk probability of occurrence of the infectious complications in the body cavity obtained by the model is between 0 and 1. In the formula, g (x) is a continuous variable, and the value range of 0-1 is not applicable, so that the g (x) is converted into the risk probability between 0 and 1 by using Logistic transformation. The analysis and calculation are completed by SPSS statistical software.

The product and the method can predict whether the infection complication in the body cavity occurs in advance, guide the clinical adoption of more effective intervention measures for the infection complication in the body cavity, and can be used for reducing the perioperative mortality of the operation patient. Doctors can effectively distinguish low-risk patients and high-risk patients with the infection complications in the body cavity according to the scores of the infection complications in the body cavity, and determine whether the patients can resume eating and discharge from hospital or need to give further imaging examination and antibiotic treatment in time. On the basis, the risk probability of the infection complications in the body cavity can be predicted by further combining clinical characteristics or clinical risk factors, and the risk of the infection complications in the body cavity of the postoperative patient is specified. The product and the method have extremely high clinical application value for early diagnosis of the infectious complications in the body cavity.

Drawings

FIG. 1: content of inflammatory factor in drainage liquid of abdominal cavity

The concentration is converted to a logarithm with a base of 10. (a) IL-1 beta level in the abdominal drainage fluid after gastric tumor surgery; (b) IL-6 level in abdominal drainage after gastric tumor surgery; (c) IL-10 level in abdominal drainage after gastric tumor operation; (d) TNF-alpha level in abdominal drainage after gastric tumor operation; (e) MMP-2 level in abdominal drainage after gastric tumor operation; (f) MMP-9 level in abdominal drainage liquid after gastric tumor operation.

FIG. 2: dynamic change of inflammatory factor content in peritoneal drainage fluid

(a) Dynamic change of IL-1 beta in abdominal drainage fluid after gastric tumor operation; (b) dynamic change of IL-6 in abdominal drainage after gastric tumor operation; (c) dynamic change of IL-10 in abdominal drainage fluid after gastric tumor operation; (d) dynamic change of TNF-alpha in abdominal drainage fluid after gastric tumor operation; (e) dynamic change of MMP-2 in abdominal drainage after gastric tumor operation; (f) dynamic change of MMP-9 in abdominal drainage after gastric tumor operation. P < 0.05; p < 0.01; p < 0.001.

FIG. 3.1: ROC curve for diagnosing anastomotic stoma leakage by inflammatory factors in abdominal drainage fluid on day of operation

FIG. 3.2: ROC curve for diagnosing anastomotic leakage of inflammatory factors in abdominal drainage fluid on postoperative day 1

FIG. 3.3: ROC curve for diagnosing anastomotic leakage of inflammatory factors in abdominal drainage fluid on postoperative day 2

FIG. 3.4: ROC curve for diagnosing anastomotic leakage of inflammatory factors in abdominal drainage fluid on postoperative day 3

FIG. 4: screening inflammatory factor indicators using LASSO regression

(a) Screening inflammatory factor indexes of all time nodes within 3 days after gastric tumor operation; (b) screening inflammatory factor indexes on the 1 st day after gastric tumor operation; (c) screening inflammatory factor indexes on the 2 nd day after gastric tumor operation; (d) screening inflammatory factor indexes on the 3 rd day after gastric tumor operation; (e) and (4) screening inflammatory factor indexes on the 3 rd day after gastric tumor operation.

The black vertical lines in the cross validation graph respectively represent the lambda value (Minimum criterion) corresponding to the Minimum target parameter mean value and the optimal lambda value (1-SE criterion) corresponding to the Minimum independent variable equation obtained in one variance range. The red vertical bar in the LASSO regression plot indicates the optimal lambda value for obtaining the optimal equation.

FIG. 5: ROC curve for scoring and diagnosing anastomotic leakage of each time node

FIG. 6: specific distribution of anastomotic leakage scores in gastric tumor surgery patients on day 3 post-surgery

FIG. 7: Cut-Off value of postoperative 3 days anastomotic leakage score

FIG. 8: calibration curve of probability model of anastomotic leakage risk

FIG. 9: ROC curve of postoperative anastomotic leakage risk probability model and clinical risk factor assessment model

FIG. 10: clinical benefit and availability analysis of anastomotic leakage risk probability model and clinical risk factor assessment model

(a) Decision curves of the anastomotic leakage risk probability model and the clinical risk factor evaluation model after gastric tumor operation; (b) a clinical prediction curve of the anastomotic leakage risk probability model; (c) clinical risk factor evaluation model clinical impact curve.

FIG. 11: dynamic change of routine laboratory examination index of postoperative infectious complication patients

(a) The leucocyte count of the patients with the gastric tumor postoperative infectious complications dynamically changes; (b) patients with postoperative infectious complications of gastric neoplasia have dynamic changes in CRP levels. P <0.05

FIG. 12: dynamic change of routine laboratory examination index of postoperative anastomotic leakage patient

(a) The leucocyte count of the anastomotic leakage patient after gastric tumor surgery dynamically changes; (b) the CRP level of a gastric tumor postoperative anastomotic leakage patient dynamically changes. P <0.05

FIG. 13: gastric tumor postoperative anastomotic leakage risk prediction nomogram

The specific implementation mode is as follows:

the inflammatory factor in the invention is selected from one or the combination of more than two factors of cell factors, matrix metalloproteases, reactive oxygen species, vascular endothelial growth factors, tissue metalloprotease inhibiting factors, C-reactive proteins, leukocyte counts and the like.

Further, the cytokine is selected from the group consisting of: interleukin (IL), Colony-Stimulating Factor (CSF), Interferon (IFN), Tumor Necrosis Factor (TNF), Chemokine (CK), Growth Factor (GF), or the like.

Further, the interleukin is selected from the group consisting of: IL-1 α, IL-1 β, IL10, IL11, IL12A, IL12B, IL13, IL15, IL16, IL17A, IL17B, IL17C, IL18C, IL1F C, IL 1C, IL22F 1C, IL22F 2C, IL22F 3C, IL22F 4C, IL22F 5C, IL 23C, IL36L 1C, IL36L 2C, IL36, IL C, etc.; the colony stimulating factor is selected from: CNTF, CSF1, CSF2, CSF3, CTF1, or the like; the interferon is selected from: IFNA1, IFNA2, IFNA3, IFNA4, IFNB, IFNG, IFNL1, IFNL2, and the like; the tumor necrosis factor is selected from: TNF, TNF α, TNF β, LTA, LTB, TNFSF4, CD40LG, FASLG, CD70, TNFSF8, TNFSF9, TNFSF10, TNFSF10L, TNFSF11, TNFSF13La, TNFSF13B, TNFSF14, TNFSF15, TNFSF18, or EDA, etc.; the chemokine is selected from: CCFAMIY, CCLD 26, CCLD1a, CCLD2a, CCLD3a, CCLD4a, CCLD5a, CCLD6a, CCLD7a, CCLD8a, CCLD9a, CCLD10a, CCLD11a, CCLD12a, CCLD13a, CCLD14a, CCLD15a, CCLD16a, CCL17, CCL19, CCL20, CCL21, CCL22, CCL25, CCL27, CCL family, CXC family, CXCLD1 27, CXCLD2 27, CXCL10LAa, CXCL10LBa, CXCL 27, CXCL13 27, CXCL 72, CXCL 27, CXXCXCXCXC 27, LACX 3, LBFa 3, etc.; the growth factor is selected from TGFB1, TGFB2, TGFB3 or VEGFA, etc.

Further, the matrix metalloprotease is selected from the group consisting of: one or more of MMP-1, MMP-2, MMP-3, MMP-4, MMP-5, MMP-6, MMP-7, MMP-8, MMP-9, and MMP-10; preferably one or more of MMP-2, MMP-3, MMP-6 and MMP-9.

Further, the inflammatory factor is selected from one or more of IL-1 beta, IL-6, IL-10, TNF-alpha, MMP-2 and MMP-9;

further, the inflammatory factor is selected from one or more of IL-1 beta, IL-6, IL-10 and MMP-9;

further preferably one or more of the inflammatory factors IL-1. beta., IL-10 and MMP-9.

Further, the detection sample of the inflammatory factor is liquid in a body cavity, and the body cavity is selected from an abdominal cavity, a pelvic cavity, a thoracic cavity environment and/or a brain cavity and the like; the fluid in the body cavity is selected from fluid in the abdominal cavity, pelvic cavity, thoracic cavity environment and/or cerebrospinal fluid; further, the liquid is lacuna effusion or drainage-obtained liquid; further, the liquid is preferably an effusion or drainage of the abdominal cavity.

Further, the detection site of the inflammatory factor is an infected area in the body cavity or the periphery thereof; further, for a patient undergoing a surgery, for example, a body cavity surgery, the inflammatory factor is selected from the group consisting of inflammatory factors at 0-15 days after the surgery; preferably inflammatory factors from day 0 to 7 post-surgery, preferably inflammatory factors from day 0 to 5 post-surgery, preferably inflammatory molecules from day 0 to 3 post-surgery; further preferred is inflammatory factor at day 0, day 1, day 2 or day 3 after surgery; further preferred is inflammatory factor on day 3 after surgery.

The subject in the present invention is a mammal including human, domestic animal, pet, laboratory animal, etc., wherein human includes patients of various age and sex characteristics, further patients who have undergone a first operation or multiple operations. Furthermore, the object is a patient with coelomic organ or tissue related diseases, further a patient with abdominal cavity, pelvic cavity, thoracic cavity environment or cranial cavity related diseases, further a postoperative patient with abdominopelvic cavity or thoracic cavity viscera such as a digestive system, a urinary system, a reproductive system, a respiratory system or a cardiovascular system; furthermore, the patients are patients after operation of related diseases of organs such as digestive tract, liver, gallbladder, pancreas, spleen, kidney, ureter or bladder, or patients after operation of respiratory tract and heart; the patient is a patient after gastrointestinal surgery, is a patient after abdominal cavity surgery, and is a patient for clinically collecting abdominal cavity drainage fluid.

Wherein the digestive system diseases include: digestive tract diseases such as digestive tract tumor, digestive tract inflammation, digestive tract ulcer, etc.; such as chronic active gastritis, chronic atrophic gastritis, gastric ulcer, duodenal ulcer, ulcerative colitis, inflammatory bowel disease, ulcerative colitis, crohn's disease (, collagenous colitis, lymphocytic colitis, ischemic colitis, diversion colitis, beset's syndrome, infectious colitis, indeterminate colitis, ulcerative colitis, familial adenomatous polyposis, congenital megacolon disease, intestinal stenosis, proctitis, rectal mucositis, colon cancer, rectal cancer, fistula, intestinal obstruction, mechanical intestinal obstruction, paralytic intestinal obstruction, digestive fistula, pancreatic fistula, other unnatural fistulous tracts (including rectal vaginal fistula, rectal cystitis, intestinal fistula, etc.), ischemic intestinal necrosis, cecum cancer, rectosigmoid carcinoma, or gastric cancer.

Diseases of the liver and gallbladder include: cholelithiasis, cholecystitis, cholangitis, chronic hepatitis or liver cancer;

pancreatic diseases include: pancreatitis, pancreatic cancer, pancreatic fistula; kidney: kidney cancer, nephritis, renal calculus, pyelonephritis, or renal pelvis cancer;

spleen diseases include: spleen infarction, etc.;

bladder diseases include: bladder cancer, etc.;

diseases of the reproductive system include: uterus and ovary diseases, such as endometrial cancer, ovarian cancer, uterine myoma, endometriosis, adenomyosis or chocolate cyst, etc.; vaginal diseases such as rectovaginal fistula and the like;

the abdominal cavity diseases include: infectious peritonitis, idiopathic peritonitis, tuberculous peritonitis, or ascites without clear cause;

chest diseases include: bronchial fistula, pneumonia, atelectasis, pleural effusion, etc.;

neurological disorders include: nervous system infections, such as intracranial infections, and the like.

The infectious complications in the body cavity are selected from infectious complications of abdominal cavity, pelvic cavity, thoracic cavity or brain cavity, and further relate to abdominal cavity infection, abdominal dropsy, peritonitis, abdominal abscess, sepsis, anastomotic leakage, pancreatic fistula, duodenal stump fistula, other digestive tract fistulas, lymphatic fistula, chylemia fistula and the like; further, anastomotic leakage is preferred.

The clinical factors described in the present invention are: patient birthday, operation date, age, sex, height, weight, BMI, ASA score, diabetes, smoking, alcoholism, medication (hypotensive, hypolipidemic, corticosteroid, anticoagulant, non-steroidal anti-inflammatory drug), cardiac history, cardiovascular symptoms, peripheral vascular disease, respiratory disease history, respiratory symptoms, pre-operative ileus, neoadjuvant radiotherapy, neoadjuvant chemotherapy, prophylactic antibiotic use. Patient mode of surgery (laparoscopic/laparoscopic resection), change of mode of surgery, reason for change, elective/emergency surgery, scope of resection, anastomotic stoma (type, initial, type of placement, site, manual suture), surgical indication, time of surgery, anesthesia, intraoperative complications, amount of blood loss, time of admission, time of discharge, hospital period, drainage, stoma (location, type), air leakage test (including test results), surgeon (number and expertise), etc.

In the present invention, the statistical methods include methods generally used in the art, such as: continuous variable data conforming to normal distribution can be recorded by adopting an average value +/-standard deviation, classified variable data can be recorded in the form of quantity and percentage, comparison among normal distribution continuous variable groups can adopt a t test, comparison among non-normal distribution continuous variable groups can adopt a non-parametric test, comparison between two groups adopts a Mann-Whitney test, comparison among classified variable groups or single-factor analysis adopts a line x row x2 test and a Fisher accurate test, and factors (P > 0.1) with prediction value of single factors are included in multi-factor analysis. The inflammatory factors in the postoperative abdominal drainage fluid can be screened and analyzed by adopting Least Absolute Shrinkage and Selection Operator (LASSO) regression. Multivariate analysis of categorical variables can be performed using Logistic regression analysis, and the results are presented in a Nomogram. The reliability of the nomogram can be evaluated by using a calibration curve, and the clinical application value of the nomogram can be evaluated by using a decision curve and a clinical influence curve. A receiver operating characteristic curve (ROC curve) can be used to measure the diagnostic efficacy of a diagnostic index. The maximum value of the Youden Index (Youden Index) can be set as the optimal Cut-Off value of the ROC curve, and the statistical significance is considered to be achieved when P <0.05 on both sides of all statistical results. The above analytical calculations can be done using SPSS 20.0 statistical software, R language "rms" and "rmda" packages, and the like.

To further illustrate the method for early diagnosis of early intrinsic infectious complications and the effect thereof, the following examples are given, which are only illustrative of the method of the present invention and do not limit the subject and scope of the present invention, and other equivalent techniques within the scope of the inventive concept are also within the scope of the present invention.

Example 1: quantitative analysis of inflammatory factors in peritoneal drainage fluid

1. Material

1.1 patient Condition and clinical data Collection

Patient conditions: after gastric tumor surgery, there is abdominal drainage fluid.

Clinical data: patient birthday, operation date, age, sex, height, weight, BMI, ASA score, diabetes, smoking, alcoholism, medication (hypotensive, hypolipidemic, corticosteroid, anticoagulant, non-steroidal anti-inflammatory drug), cardiac history, cardiovascular symptoms, peripheral vascular disease, respiratory disease history, respiratory symptoms, pre-operative ileus, neoadjuvant radiotherapy, neoadjuvant chemotherapy, prophylactic antibiotic use. Patient mode of surgery (laparoscopic/laparoscopic resection), change of mode of surgery, reason for change, elective/emergency surgery, scope of resection, anastomotic stoma (type, initial, type of placement, site, manual suture), surgical indication, time of surgery, anesthesia, intraoperative complications, amount of blood loss, time of admission, time of discharge, hospital period, drainage, stoma (location, type), air leakage test (including test results), surgeon (number and expertise), etc.

Information acquisition in the art: surgical procedure (laparoscopic/laparoscopic resection), surgical procedure change, reason for change, phase selection/emergency surgery, scope of resection, anastomotic stoma (type, initial, type of placement, site, manual suture), surgical indication, time of surgery, anesthesia, intraoperative complications, amount of blood loss, time of admission, time of discharge, period of hospitalization, drainage, stoma (location, type), air leakage test (including test results), surgeon (number and expertise).

And (3) postoperative information acquisition: routine laboratory test results (leucocytes, CRP and the like), drainage liquid shape, color, drainage quantity, survival condition within 30 days after operation, condition of hospital re-admission and condition of secondary operation.

Postoperative complication enrollment of patients in cohort: prospective registration is carried out on postoperative complications of patients to be enrolled, researchers and clinicians independently register and input various databases, and the databases are regularly checked and summarized.

Registration content and criteria: 21 complications were included in the category of the registration and the severity of the complications was graded using a Clavien-Dindo grading (CD grading) as a standard, as shown in Table 1.

TABLE 1 points of diagnosis of postoperative complications of gastric tumors

1.2 Collection and pretreatment of Abdominal drainage liquid specimen

Abdominal drainage liquid of a patient in three days after operation (including the day after operation to the 3 rd day after operation) is collected at a fixed time point every day, and 20ml is collected every time. After sampling the abdominal cavity drainage liquid specimen, centrifuging for 10 minutes at 4 ℃ and 2800g, separately subpackaging the supernatant and the precipitate and storing in a refrigerator at minus 80 ℃.

2. Method and steps

Selecting representative inflammatory factors IL-1 beta, IL-6, IL-10, TNF-alpha, MMP2 and MMP9 in the drainage liquid of abdominal cavity for quantitative analysis

The content of each factor is quantitatively analyzed by using a multi-factor enzyme-linked immunoassay. Wherein the amounts of IL-1 β, IL-6, IL-10 and TNF- α are measured using HSTCMAG-28SK (EMD Millipore, USA) kit; the amounts of MMP2 and MMP9 were determined using the HMMP2MAG-55K (EMD Millipore, USA) kit.

3. Results

3.1 clinical results: 263 patients after gastric tumor surgery, wherein 209 patients are male and 54 patients are female; median age 62(54-67) years; among them, a total of 81 patients were diagnosed with postoperative complications within 30 days after the operation, and the complication rate was 30.8%. 24 of the patients were diagnosed with infectious complications, including 17 patients with anastomotic leakage, with a incidence of anastomotic leakage of 6.46%; in addition, there were more than two infectious complications in 5 patients; one patient died within 30 days of surgery.

3.2 levels of inflammatory factors in the drainage fluid of the abdominal cavity after gastric tumor surgery

The concentrations of inflammatory factors in the postoperative abdominal drainage fluid of patients with gastric tumor are shown in fig. 1, the levels of IL-1 beta, IL-10 and MMP-9 in the postoperative 3 days of patients with anastomotic leakage are obviously higher than those of other patients, and the differences have statistical significance (Table 2).

TABLE 2 quantitative analysis results of inflammation factors of abdominal drainage after gastric tumor operation

Inflammatory factor concentrations are expressed as median (quartile); differences were statistically significant and indicated in bold (P < 0.05).

3.3 dynamic Change in inflammatory factor levels in Abdominal drainage following gastric tumor surgery

The levels of inflammatory factors in the postoperative peritoneal drainage fluid of patients with gastric tumors showed dynamic changes (fig. 2), wherein the level of IL-1 β in the peritoneal drainage fluid of patients with stomal leakage continued to increase from postoperative day 1 (fig. 2a), while the level of IL-10 decreased in the early postoperative phase and increased again from postoperative day 2 (fig. 2 c). The MMP-9 content in the drainage fluid of the abdominal cavity is gradually reduced every other day, but the concentration of the patients with anastomotic leakage is higher than that of other patients within 3 days after the operation (figure 2 f).

Example 2: diagnostic value of inflammatory factors in abdominal drainage liquid for anastomotic leakage

1. Diagnosis efficiency of single inflammatory factor in abdominal cavity drainage fluid after gastric tumor operation on anastomotic stoma leakage

The diagnosis effectiveness of a single inflammatory factor in the drainage fluid of the abdominal cavity on anastomotic leakage is evaluated by adopting an ROC curve, and the ROC analysis result of each inflammatory factor on the diagnosis of anastomotic leakage within 3 days after gastric tumor operation is shown in figures 3.1-3.4 and tables 3-6.

TABLE 3 ROC analysis of inflammatory factors in peritoneal drainage fluid for diagnosis of anastomotic leakage on the day of gastric tumor surgery

Area under the AUC curve; differences were statistically significant and indicated in bold (P < 0.05).

TABLE 4 ROC analysis of inflammatory factors in peritoneal drainage fluid for diagnosis of anastomotic leakage on day 1 after gastric tumor surgery

Area under the AUC curve; differences were statistically significant and indicated in bold (P < 0.05).

TABLE 5 ROC analysis of inflammatory factors in peritoneal drainage fluid diagnosis of anastomotic leakage on day 2 after gastric neoplasia

Area under the AUC curve; differences were statistically significant and indicated in bold (P < 0.05).

TABLE 6 ROC analysis of inflammatory factors in peritoneal drainage fluid for diagnosis of anastomotic leakage on day 3 after gastric tumor surgery

Area under the AUC curve; differences were statistically significant and indicated in bold (P < 0.05).

The greater the area under the curve AUC value, the greater the diagnostic efficacy, often AUC greater than 0.8 is clinically instructive. As can be seen by the diagnostic efficacy analysis of a single inflammatory factor in the abdominal cavity drainage fluid after gastric tumor operation, the diagnostic efficacy of IL-1 beta, IL-10 and MMP-9 on anastomotic leakage at the 3 rd day after gastric tumor operation is the highest, and the AUC is 0.76(p <0.01), 0.77 (p <0.01) and 0.75(p <0.01), respectively. The diagnostic efficacy of TNF-alpha on anastomotic leakage reached its highest on day 2 post-gastric neoplasia with an AUC of 0.73(p < 0.01). IL-6 and MMP-2 were less effective in diagnosing stomal leakage, with maximum AUC 0.62(p ═ 0.11) and 0.59(p ═ 0.22) for stomal leakage diagnosis within 3 days after gastric neoplasia surgery, respectively. The results also show that the diagnosis efficacy of a single inflammatory factor on anastomotic leakage of a patient after gastric tumor operation does not meet the clinical application requirement.

Example 3: parameter selection and construction of anastomotic leakage scoring system

1. Parameter selection of gastric tumor postoperative anastomotic leakage scoring system

In order to further optimize the diagnosis effect of inflammatory factors in the abdominal drainage fluid after gastric tumor operation on anastomotic leakage, Least Absolute kringle and Selection Operator (LASSO) regression is adopted to screen the inflammatory factors in the abdominal drainage fluid at different time points of 3 days after gastric tumor operation, and the optimal index is selected to construct an anastomotic leakage prediction model. The method comprises the steps of screening 6 indexes (figure 4a) from 24 inflammation factor indexes of nodes at different time within 3 days after operation, screening 2 indexes (figure 4b) from 6 inflammation factor indexes on the day of operation, screening 2 indexes (figure 4c) from 6 inflammation factor indexes on the 1 st day after operation, screening 2 indexes (figure 4d) from 6 inflammation factor indexes on the 2 nd day after operation, screening 3 indexes (figure 4e) from 6 inflammation factor indexes on the 3 rd day after operation, respectively using the indexes as parameters to establish an anastomotic leakage scoring system, and calculating models of anastomotic leakage scoring of the nodes at different time after gastric neoplasm operation are shown in a table 7.

TABLE 7 gastric tumor postoperative anastomosis leakage scoring model at each time node

D0 day of surgery; d1 day 1 after surgery; day 2 post-surgery D2; day 3 post-surgery D3.

2. Diagnosis efficiency of anastomotic leakage scoring system for gastric tumor postoperative anastomotic leakage

The ROC curve is used to analyze the diagnosis effectiveness of the anastomotic leakage score on the anastomotic leakage after gastric tumor surgery, and the ROC analysis results of the anastomotic leakage scores at different time nodes on the anastomotic leakage diagnosis are shown in fig. 5 and table 8. The stoma leak score was highest for the diagnostic efficacy of stoma leaks on day 3 post-surgery, with an AUC of 0.87(p < 0.01).

TABLE 8 ROC analysis of stomal leak score diagnosis for stomal leak at each time node after gastric neoplasia

Area under the AUC curve; differences were statistically significant and indicated in bold (P < 0.05).

3. Setting of score of anastomotic leakage score Cut-Off value and efficacy of anastomotic leakage diagnosis on postoperative day 3

And selecting a value-2.801 corresponding to the maximum value of the Yoden index in ROC analysis as a Cut-Off value of the anastomotic leakage score on the 3 rd day after gastric tumor operation, and dividing the gastric tumor postoperative patients into a high group and a low group. The distribution of anastomotic leakage scores on day 3 after gastric tumor surgery is shown in fig. 6 and 7, and the differentiating effect of Cut-Off values on patients after gastric tumor surgery is shown in table 9.

Therefore, all patients are divided into a stoma leakage high-risk group and a stoma leakage low-risk group by taking-2.801 as the Cut-Off value of the stoma leakage score on the 3 rd day after operation, wherein 181 patients in the low-risk group have stoma leakage in 2 people, 43 patients in the high-risk group have stoma leakage in 11 people, and 39 patients are not classified into the high-risk group or the low-risk group due to lack of parameters required for calculating the stoma leakage score on the 3 rd day after operation and comprise 4 patients with stoma leakage. Compared with the clinical practical result, the sensitivity and the specificity of the scoring method of the invention exceed 80 percent, and the negative predictive value is as high as 98.5 percent.

TABLE 9 diagnostic efficacy of anastomotic leakage score on postoperative day 3 anastomotic leakage

Example 4: clinical verification of postoperative anastomotic leakage scoring system

And the reliability and application value of the anastomotic leakage scoring system and the risk grade evaluation system are tested through clinical time.

The inclusion conditions and diagnostic criteria were as before. 66 patients with gastric tumor surgery were used in all, 48 men and 18 women; median age 57 years; a total of 7 people were diagnosed with anastomotic leakage.

The third day anastomotic leakage scoring model in example 3 was used: -3.10+0.000241 × IL-1 β (D3) +0.00183 × IL-10(D3) +0.000000853 × MMP-9(D3), and 66 patients were scored and the diagnosed AUC was 0.83(p <0.01) as the Cut-Off value for the stoma leak score on day 3 after gastric neoplasia, based on the score-2.801 corresponding to the maximum value of the jotan index in the ROC analysis, and the Cut-Off values obtained using the training cohort were used to classify the patients after gastric neoplasia into high and low groups.

A total of 17 of 66 were rated as anastomotic leakage high groups, of which 6 had anastomotic leakage, and 49 were rated as anastomotic leakage low groups, of which 48 had not. The accuracy, specificity, sensitivity, etc. of the diagnosis are shown in Table 10.

TABLE 10 diagnostic efficacy analysis results of anastomotic leak scoring system

Practice ofExample 5: constructing a model of probability of anastomotic leakage risk

In order to enable the diagnosis effect of the inflammatory factors on the anastomotic leakage to be closer to clinical practice, the inventor further combines the clinical characteristics of the gastric tumor surgery patient with the score result of the anastomotic leakage, and then constructs an anastomotic leakage risk probability model containing the score result of the anastomotic leakage and clinical risk factors by adopting Logistic regression. By constructing an anastomotic leakage risk probability model, the probability value of anastomotic leakage of a patient is calculated by means of a nomogram, and whether corresponding clinical intervention is adopted or not is determined by comparing with a decision curve.

1. Collecting patient clinical basic data

Acquiring preoperative information: birthday, date of surgery, age, sex, height, weight, BMI, ASA score, diabetes, smoking, alcoholism, medication (hypotensive, hypolipidemic, corticosteroid, anticoagulant, non-steroidal anti-inflammatory drug), cardiac history, cardiovascular symptoms, peripheral vascular disease, respiratory disease history, respiratory symptoms, preoperative ileus, neoadjuvant radiotherapy, neoadjuvant chemotherapy, prophylactic antibiotic use.

Information acquisition in the art: surgical procedure (laparoscopic/laparoscopic resection), surgical procedure change, reason for change, phase selection/emergency surgery, scope of resection, anastomotic stoma (type, initial, type of placement, site, manual suture), surgical indication, time of surgery, anesthesia, intraoperative complications, amount of blood loss, time of admission, time of discharge, period of hospitalization, drainage, stoma (location, type), air leakage test (including test results), surgeon (number and expertise).

And (3) postoperative information acquisition: routine laboratory test results (leucocytes, CRP and the like), drainage liquid shape, color, drainage quantity, survival condition within 30 days after operation, condition of hospital re-admission and condition of secondary operation.

2. Clinical risk factor exploration

263 patients after the concurrent gastric tumor surgery, wherein 209 patients are male and 54 patients are female; median age 62(54-67) years; the median Body Mass Index (BMI) is 24(22-26) kg/m 2; patient specific baseline information, ASA score, surgical information, etc. are shown in table 11. Among them, a total of 81 patients were diagnosed with postoperative complications within 30 days after the operation, and the complication rate was 30.8%. 24 of the patients were diagnosed with infectious complications, including 17 patients with anastomotic leakage, with a incidence of anastomotic leakage of 6.46%; in addition, a total of 5 patients presented with more than two infectious complications (table 12); one patient died within 30 days of surgery.

TABLE 11 clinical data of patients after gastric tumor surgery

Age and BMI are indicated by median (quartile); the differences are statistically significant and indicated in bold (P)<0.05); Mann-Whitney U test.Comprises a Billroth-II formula and an Uncut Roux-en-Y formula

TABLE 12 gastric tumor postoperative infectious complications enrollment results

On the basis, the age (p is 0.02), the tumor position (p <0.01), the resection range (p <0.01) and the anastomosis mode (p <0.01) are related to the occurrence of the postoperative complications of the gastric tumor through single factor analysis; for infectious complications, it is associated with age (p 0.01), tumor location (p 0.01), range of resection (p 0.04), and mode of anastomosis (p 0.05); while anastomotic leakage is related to age (p ═ 0.03), tumor location (p <0.01), mode of operation (p ═ 0.01), range of resection (p <0.01), and mode of anastomosis (p < 0.01). The factors are included in the multi-factor analysis, and the result shows that the age (p is 0.03) and the anastomosis mode (p is 0.03) are independent risk factors of postoperative complications of gastric tumors; age (p 0.03) tumor location (p 0.01) and resection range (p 0.04) are independent risk factors for infection complications; tumor location (p ═ 0.02) and surgical mode (p <0.01) are independent risk factors for stomal leakage, as shown in table 13.

TABLE 13 analysis of independent risk factors for postoperative complications of gastric neoplasia

The differences are statistically significant and indicated in bold (P)<0.05); "-" does not apply;including Billroth-II formula and Uncut Roux-en-Y formula.

3. Constructing a model of probability of anastomotic leakage risk

And establishing an anastomotic leakage risk probability model after gastric tumor operation through Logistic regression, wherein included independent variables comprise an anastomotic mode, an operation mode, a tumor part and an anastomotic leakage score on the 3 rd day after the operation. And presenting the model prediction result through a nomogram. See fig. 13 alignment chart of prediction of stomal leakage risk after gastric tumor surgery.

4. Reliability assessment of anastomotic leakage risk probability model

The reliability of the anastomotic leakage risk probability model is evaluated through the correction curve, the Mean Absolute Error (Mean Absolute Error) is 0.016 (figure 8), and the predicted risk probability and the actual observation probability of the model have better consistency.

5. Clinical application value evaluation of anastomotic leakage risk probability model

And 3 clinical indexes of the anastomosis mode, the operation mode and the tumor part are brought into Logistic regression to establish a gastric tumor postoperative anastomotic leakage clinical risk factor evaluation model.

The diagnostic efficacy of the anastomotic leakage risk probability model was compared to the clinical risk factor assessment model by ROC analysis (fig. 9).

Wherein, the AUC of the anastomotic leakage risk probability model is 0.93(p <0.01), and the AUC of the clinical risk factor evaluation model is 0.86(p < 0.01). The decision curve is adopted to compare the clinical benefit conditions of the anastomotic leakage risk probability model and the clinical risk factor evaluation model, and the results show that the net benefit of the anastomotic leakage risk probability model is higher than that of the clinical risk factor evaluation model when the risk threshold is less than 0.7, as shown in fig. 10a, fig. 10b and fig. 10 c.

Therefore, compared with a clinical risk factor evaluation model constructed by clinical risk factors alone, the postoperative anastomosis leakage score of 3 days after the incorporation can improve the diagnostic efficacy of the model. From the decision curve analysis result, when the threshold value of the anastomotic leakage risk is in the range of 5% -70%, the net gain of whether clinical intervention is adopted is determined to be higher than that of the clinical risk factor evaluation model according to the probability model of the anastomotic leakage risk.

The foregoing shows and describes the general principles and features of the present invention, rather than limitations thereof. It will be understood by those skilled in the art that various equivalent modifications or changes may be made by those skilled in the art without departing from the spirit and scope of the present invention, and still be covered by the appended claims.