Risk assessment model for heart failure rehospitalization or death and its application

Abstract

This invention provides a risk assessment model for readmission or death due to heart failure and its application. Specifically, this invention provides a system for assessing the risk of readmission or death within one year for heart failure patients. The system includes: (a) an input module configured to input heart failure risk index data of the subject; (b) an assessment module configured to compare the input risk index data with pre-stored conditions or thresholds in the device to obtain a comparison result; obtain a risk score based on the comparison result; and obtain an assessment result based on the risk score; and (c) an output module configured to output the assessment result. The system of this invention can more effectively and accurately assist in assessing the risk of readmission or death within one year for heart failure patients.

Description

Technical Field

[0001] This invention relates to the field of medical diagnostics, and more specifically to a risk assessment model for readmission or death due to heart failure and its application. Background Technology

[0002] Heart failure (HF) is a complex pathophysiology affected by multiple systems, including hemodynamics, neurohormones, metabolism, and the kidneys.

[0003] Studies have shown that the gut plays a crucial role in the pathophysiology of heart failure, including the contribution of multiple co-occurring diseases, including the gut-heart axis, to heart failure. This adds a new dimension to our understanding of the pathophysiology of heart failure, but currently available heart failure risk scores do not incorporate the contribution of the gut-heart axis.

[0004] Therefore, constructing a multi-etiological prognostic risk assessment model that incorporates the gut-heart axis for assessing the prognosis (risk of readmission or death) of heart failure is of great significance in this field. Summary of the Invention

[0005] This invention provides a novel risk assessment model for heart failure readmission or death and its application.

[0006] In a first aspect of the invention, a system for assessing the risk of readmission or death within one year in patients with heart failure is provided, the system comprising:

[0007] (a) An input module configured to input heart failure risk index data of the subject to be tested;

[0008] The heart failure risk indicators include: chronic obstructive pulmonary disease (COPD), history of heart failure, use of loop diuretics, NYHA class III / IV, diabetes, gut microbiota metabolites, age, creatinine, N-terminal pro-B-type natriuretic peptide (NT-proBNP), diastolic blood pressure, and no use of beta-blockers; and the gut microbiota metabolites include: γ-butylbetaine, acetyl-L-carnitine, trimethylamine oxide, and L-carnitine;

[0009] (b) An evaluation module, which is configured to perform the following functions:

[0010] i) Compare the input risk indicator data (including data of each gut microbiota metabolite) with the conditions or thresholds pre-stored in the system to obtain the comparison result; if a certain input risk indicator data value meets the condition or is greater than its corresponding threshold, it is assigned a value of 1; if the risk indicator data value does not meet the condition or is less than its corresponding threshold, it is assigned a value of 0.

[0011] ii) Substitute the comparison results of gut microbiota metabolite data into the first-level risk scoring formula to obtain the first risk score (or the risk score value of gut microbiota metabolites); and compare the first risk score with the pre-set conditions or thresholds in the system to obtain the comparison result G of gut microbiota metabolites with their corresponding conditions or thresholds.

[0012] iii) The comparison result G is compared with the heart failure risk index data of other non-gut microbiota metabolites, and then substituted into the second risk scoring formula to calculate the second risk score, thereby obtaining the assessment result.

[0013] The scoring formulas for the first and second risks are expressed as follows:

[0014] Wherein, when the scoring formula is the first risk scoring formula (S1), Wi is the weight value of each gut microbiota metabolite, and Pi is the comparison result of each gut microbiota metabolite data with its corresponding conditions or thresholds;

[0015] When the scoring formula is the second risk scoring formula (S2), Wi is the weight value of each risk indicator; Pi is the comparison result of each risk indicator data with its corresponding conditions or thresholds;

[0016] The assessment based on the second score includes: when the second risk score is greater than or equal to the risk cutoff value, it indicates a high risk of readmission or death within one year for the subject; conversely, it indicates a low risk.

[0017] (c) Output module, which is configured to output the evaluation result.

[0018] In another preferred embodiment, the heart failure risk indicators are selected from the following group: chronic obstructive pulmonary disease (COPD), history of heart failure, use of loop diuretics, NYHA class III / IV, diabetes, gut microbiota metabolites, age, creatinine, N-terminal pro-B-type natriuretic peptide (NT-proBNP), diastolic blood pressure, no use of beta-blockers, or combinations thereof.

[0019] In another preferred embodiment, the intestinal flora metabolites are selected from the group consisting of γ-butylbetaine, acetyl-L-carnitine, trimethylamine oxide, L-carnitine, or combinations thereof.

[0020] In another preferred embodiment, the value of G is 1 or 0.

[0021] In another preferred example, the risk cutoff value for heart failure patients to be readmitted or die within one year is 11.

[0022] In another preferred embodiment, when the second risk score is ≤11, the risk of readmission or death within one year for the subject is low; when the second risk score is 12-23, the risk of readmission or death within one year for the subject is moderate; and when the second risk score is >23, the risk of readmission or death within one year for the subject is high.

[0023] In another preferred example, the risk cutoff value for heart failure patients to be readmitted or die within one year is 16.

[0024] In another preferred embodiment, when the second risk score is <16, the risk of readmission or death within one year for the subject is low; when the second risk score is ≥16, the risk of readmission or death within one year for the subject is high.

[0025] In another preferred embodiment, the first risk scoring formula is: S1 = 4*γ-butyl betaine + 3*acetyl L-carnitine + 2*trimethylamine oxide + 1*L-carnitine.

[0026] In another preferred embodiment, in the first risk scoring formula, the values ​​of γ-butyl betaine, acetyl L-carnitine, trimethylamine oxide, or L-carnitine are 1 or 0, respectively.

[0027] In another preferred embodiment, when the first risk score is ≥6, the comparison result G between the gut microbiota metabolites and their corresponding conditions or thresholds is assigned a value of 1, otherwise it is assigned a value of 0.

[0028] In another preferred embodiment, the second risk scoring formula is: S2 = 5 * Chronic obstructive pulmonary disease (COPD) + 5 * History of heart failure + 5 * Use of loop diuretics + 4 * NYHA class III / IV + 4 * Diabetes + 3 * Gut microbiota metabolites + 2 * Age + 2 * Creatinine + 2 * NT-proBNP + 2 * Diastolic blood pressure + 2 * No use of β-blockers.

[0029] In another preferred embodiment, in the second risk scoring formula, the values ​​for chronic obstructive pulmonary disease (COPD), history of heart failure, use of loop diuretics, NYHA class III / IV, diabetes, gut microbiota metabolites, age, creatinine, NT-proBNP, diastolic blood pressure, or no use of β-blockers are 1 or 0, respectively.

[0030] In another preferred embodiment, the first risk scoring formula and the second risk scoring formula can be automatically calculated by a computer-aided program.

[0031] In another preferred embodiment, the system further includes (d) a storage module configured to store data selected from the group consisting of: judgment or calculation result values, conditions or thresholds for each risk indicator, and risk cutoff values.

[0032] In another preferred embodiment, the system further includes (e) a control module configured to control the operation of the modules.

[0033] In another preferred embodiment, the subject of the test is a patient with heart failure.

[0034] In another preferred embodiment, the risk is the risk of a heart failure patient being readmitted and / or dying from heart failure within one year.

[0035] In a second aspect of the invention, a method for assessing the risk of readmission or death within one year of heart failure is provided, comprising the steps of:

[0036] (a) Provide data, including heart failure risk index data for the subjects being tested;

[0037] The heart failure risk indicators include: chronic obstructive pulmonary disease (COPD), history of heart failure, use of loop diuretics, NYHA class III / IV, diabetes, gut microbiota metabolites, age, creatinine, N-terminal pro-B-type natriuretic peptide (NT-proBNP), diastolic blood pressure, and no use of beta-blockers; and the gut microbiota metabolites include: γ-butylbetaine, acetyl-L-carnitine, trimethylamine oxide, and L-carnitine;

[0038] (b) Analysis and evaluation, which includes the following steps:

[0039] (s1) The input risk indicator data (including data of each gut microbiota metabolite) is compared with the conditions or thresholds pre-stored in the system to obtain the comparison result; if a certain input risk indicator data value meets the condition or is greater than its corresponding threshold, it is assigned a value of 1; if the risk indicator data value does not meet the condition or is less than its corresponding threshold, it is assigned a value of 0.

[0040] (s2) Substitute the comparison results of gut microbiota metabolite data into the first-level risk scoring formula to obtain its first risk score; and compare the first risk score with the pre-set conditions or thresholds in the system to obtain the comparison result G of gut microbiota metabolites and their corresponding conditions or thresholds; and

[0041] (s3) The comparison result G is compared with the heart failure risk index data of other non-gut microbiota metabolites, and then substituted into the second risk scoring formula for calculation to obtain the second risk score, thereby obtaining the assessment result.

[0042] The scoring formulas for the first and second risks are expressed as follows:

[0043] Wherein, when the scoring formula is the first risk scoring formula (S1), Wi is the weight value of each gut microbiota metabolite, and Pi is the comparison result of each gut microbiota metabolite data with its corresponding conditions or thresholds;

[0044] When the scoring formula is the second risk scoring formula (S2), Wi is the weight value of each risk indicator; Pi is the comparison result of each risk indicator data with its corresponding conditions or thresholds;

[0045] The assessment based on the second score includes: when the second risk score is greater than or equal to the risk cutoff value, it indicates a high risk of readmission or death within one year for the subject; conversely, it indicates a low risk.

[0046] (3) Output the results.

[0047] In another preferred embodiment, the heart failure risk indicators are selected from the following group: chronic obstructive pulmonary disease (COPD), history of heart failure, use of loop diuretics, NYHA class III / IV, diabetes, gut microbiota metabolites, age, creatinine, N-terminal pro-B-type natriuretic peptide (NT-proBNP), diastolic blood pressure, no use of beta-blockers, or combinations thereof.

[0048] In another preferred embodiment, the intestinal flora metabolites are selected from the group consisting of γ-butylbetaine, acetyl-L-carnitine, trimethylamine oxide, L-carnitine, or combinations thereof.

[0049] In another preferred embodiment, the value of G is 1 or 0.

[0050] In another preferred example, the risk cutoff value for heart failure patients to be readmitted or die within one year is 11.

[0051] In another preferred embodiment, when the second risk score is ≤11, the risk of readmission or death within one year for the subject is low; when the second risk score is 12-23, the risk of readmission or death within one year for the subject is moderate; and when the second risk score is >23, the risk of readmission or death within one year for the subject is high.

[0052] In another preferred example, the risk cutoff value for heart failure patients to be readmitted or die within one year is 16.

[0053] In another preferred embodiment, when the second risk score is <16, the risk of readmission or death within one year for the subject is low; when the second risk score is ≥16, the risk of readmission or death within one year for the subject is high.

[0054] In another preferred embodiment, the first risk scoring formula is: S1 = 4*γ-butyl betaine + 3*acetyl L-carnitine + 2*trimethylamine oxide + 1*L-carnitine.

[0055] In another preferred embodiment, in the first risk scoring formula, the value of γ-butylbetaine, acetyl-L-carnitine, trimethylamine oxide, or L-carnitine is 1 or 0.

[0056] In another preferred embodiment, when the first risk score is ≥6, the comparison result G between the gut microbiota metabolites and their corresponding conditions or thresholds is assigned a value of 1, otherwise it is assigned a value of 0.

[0057] In another preferred embodiment, the second risk scoring formula is: S2 = 5 * Chronic obstructive pulmonary disease (COPD) + 5 * History of heart failure + 5 * Use of loop diuretics + 4 * NYHA class III / IV + 4 * Diabetes + 3 * Gut microbiota metabolites + 2 * Age + 2 * Creatinine + 2 * NT-proBNP + 2 * Diastolic blood pressure + 2 * No use of β-blockers.

[0058] In another preferred embodiment, in the second risk scoring formula, the values ​​for chronic obstructive pulmonary disease (COPD), history of heart failure, use of loop diuretics, NYHA class III / IV, diabetes, gut microbiota metabolites, age, creatinine, NT-proBNP, diastolic blood pressure, or no use of beta-blockers are 1 or 0, respectively.

[0059] In another preferred embodiment, the subject of the test is a patient with heart failure.

[0060] In another preferred embodiment, the heart failure patient is either a treated or untreated heart failure patient.

[0061] In another preferred embodiment, the method is an in vitro method.

[0062] In another preferred embodiment, the method is non-diagnostic and non-therapeutic.

[0063] In a third aspect of the invention, a method for constructing a risk assessment model for readmission or death within one year of heart failure is provided, comprising the steps of:

[0064] (s1) Provide a first dataset containing heart failure risk indicator data, wherein the risk indicator data includes: chronic obstructive pulmonary disease (COPD), history of heart failure, use of loop diuretics, NYHA class III / IV, diabetes, gut microbiota metabolites, age, creatinine, N-terminal pro-B-type natriuretic peptide (NT-proBNP), diastolic blood pressure, and no use of β-blockers; and wherein the gut microbiota metabolites include: γ-butyl betaine, acetyl-L-carnitine, trimethylamine oxide, and L-carnitine;

[0065] (s2) Based on the data information of the first dataset, a first risk model is constructed using logistic regression, and then a second risk model is constructed based on the first risk model to assess the risk of readmission or death within one year of heart failure (second risk model).

[0066] In another preferred embodiment, the first risk model is constructed based on gut microbiota metabolite index data.

[0067] In another preferred embodiment, the risk indicators input to the first risk model are: γ-butyl betaine, acetyl L-carnitine, trimethylamine oxide, and L-carnitine.

[0068] In another preferred embodiment, the scoring formula for the first risk model is:

[0069] Wherein, Wi is the weight value of each gut microbiota metabolite, Pi is the comparison result of each gut microbiota metabolite data with its corresponding condition or threshold, and if the input gut microbiota metabolite data meets the condition or is greater than its corresponding threshold, it is assigned a value of 1, and if the gut microbiota metabolite data does not meet the condition or is less than its corresponding threshold, it is assigned a value of 0.

[0070] In another preferred embodiment, the first risk scoring formula is: S1 = 4*γ-butyl betaine + 3*acetyl L-carnitine + 2*trimethylamine oxide + 1*L-carnitine.

[0071] In another preferred embodiment, the scoring formula for the heart failure readmission or death risk assessment model (second risk model) within one year is as follows:

[0072] Wherein, Wi is the weight value of each risk indicator; Pi is the comparison result of each risk indicator data with its corresponding condition or threshold, and if the input gut microbiota metabolite data meets the condition or is greater than its corresponding threshold, it is assigned a value of 1, and if the gut microbiota metabolite data does not meet the condition or is less than its corresponding threshold, it is assigned a value of 0.

[0073] In another preferred embodiment, when the first risk score is ≥6, the comparison result G between the gut microbiota metabolites and their corresponding conditions or thresholds is assigned a value of 1, otherwise it is assigned a value of 0.

[0074] In another preferred embodiment, the second risk scoring formula is: S2 = 5 * Chronic obstructive pulmonary disease (COPD) + 5 * History of heart failure + 5 * Use of loop diuretics + 4 * NYHA class III / IV + 4 * Diabetes + 3 * Gut microbiota metabolites + 2 * Age + 2 * Creatinine + 2 * NT-proBNP + 2 * Diastolic blood pressure + 2 * No use of β-blockers.

[0075] In another preferred example, the risk cutoff value for heart failure patients to be readmitted or die within one year is 11.

[0076] In another preferred embodiment, when the second risk score is ≤11, the risk of readmission or death within one year for the subject is low; when the second risk score is 12-23, the risk of readmission or death within one year for the subject is moderate; and when the second risk score is >23, the risk of readmission or death within one year for the subject is high.

[0077] In another preferred example, the risk cutoff value for heart failure patients to be readmitted or die within one year is 16.

[0078] In another preferred embodiment, when the second risk score is <16, the risk of readmission or death within one year for the subject is low; when the second risk score is ≥16, the risk of readmission or death within one year for the subject is high.

[0079] It should be understood that, within the scope of this invention, the above-described technical features of this invention and the technical features specifically described below (such as in the embodiments) can be combined with each other to form new or preferred technical solutions. Due to space limitations, they will not be described in detail here. Attached Figure Description

[0080] Figure 1 The flowchart shows the development strategy for a heart failure readmission or death risk assessment model within one year.

[0081] Figure 2 Kaplan-Meier survival curves (top) and ROC curves (bottom) are shown, using clinical risk models for one-year outcomes of death and / or readmission due to heart failure using (A) the training cohort and (B) the validation cohort.

[0082] Figure 3Kaplan-Meier survival curves for the training cohort (A) and validation cohort (B) are shown, based on clinical risk scores according to the Youden index, for low and high risk groups of death and / or rehospitalization due to heart failure within one year. Detailed Implementation

[0083] This invention, based on extensive and in-depth research, is the first to develop a more effective and accurate assessment model for the risk of readmission or death within one year of heart failure, and its application. Specifically, the inventors constructed a model using specific heart failure risk indicator data for heart failure risk assessment. These indicators include: chronic obstructive pulmonary disease (COPD), history of heart failure, use of loop diuretics, NYHA class III / IV, diabetes, gut microbiota metabolites, age, creatinine, NT-proBNP, diastolic blood pressure, and lack of β-blocker use. The model of this invention can effectively classify low-risk and high-risk individuals; and compared with existing models, the model of this invention represents a certain degree of optimization and improvement (i.e., improved classification or assessment capabilities). Based on this, this invention was completed.

[0084] the term

[0085] The terminology used in this invention has the meanings commonly understood by those skilled in the art. However, for a better understanding of this invention, some definitions and related terms are explained below:

[0086] As used herein, the terms “comprising,” “including,” and “containing” are used interchangeably and include not only closed definitions but also semi-closed and open definitions. In other words, the terms include “consisting of” and “substantially consisting of”.

[0087] As used in this article, the term "LC-MS" is short for "high performance liquid chromatography-mass spectrometry," and the terms "LC-MS" and "high performance liquid chromatography-mass spectrometry" are used interchangeably.

[0088] The following descriptions are given using acetyl-L-carnitine, L-carnitine, and trimethylamine oxide as examples.

[0089] As used in this article, the term "acetyl-L-carnitine" is abbreviated as ALC, and "acetyl-L-carnitine" and "ALC" are used interchangeably.

[0090] As used in this article, the term "L-carnitine" is abbreviated as carnitine, and "L-carnitine" and carnitine are used interchangeably.

[0091] As used in this article, the term "trimethylamine oxide" is abbreviated as TMAO, and "trimethylamine oxide" and "TMAO" are used interchangeably.

[0092] As used in this article, the term "ultra-high performance liquid chromatography" is abbreviated as UPLC, and the terms "ultra-high performance liquid chromatography" and "UPLC" are used interchangeably.

[0093] As used herein, “mass spectrometry” (MS) refers to analytical techniques for the mass identification of compounds. MS techniques generally involve (1) ionizing compounds to form charged compounds; and (2) detecting the molecular weight of the charged compounds and calculating the mass-to-charge ratio (m / z). Compounds can be ionized and detected by any suitable means. A “mass spectrometer” generally includes an ionizer and an ion detector.

[0094] The term “approximately” as used herein refers to the value shown plus or minus 10% when referring to a quantitative measurement.

[0095] According to the present invention, the term "biomarker set" refers to a single biomarker or a combination of two or more biomarkers.

[0096] According to the present invention, the term "intestinal flora metabolites" refers to metabolites produced by the intestinal flora in an organism, such as choline, betaine aldehyde, betaine, L-carnitine, croton betaine, γ-butyl betaine, acetyl L-carnitine, trimethyllysine, etc. According to the present invention, the term "risk" refers to the risk of a heart failure patient being readmitted and / or dying from heart failure within one year of being diagnosed with heart failure.

[0097] According to the present invention, the first-level risk score is the risk score of gut microbiota metabolites.

[0098] According to the present invention, the second-level risk score is a heart failure risk score that incorporates gut microbiota metabolites into other internationally recognized scoring systems. International scoring systems include ADHERE (Acute Decompensated Heart Failure National Registry), OPTIMIZE-HF (Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients with Heart Failure), and GWTG-HF (GWTG-HF: Get With The Guidelines-Heart Failure, the American Heart Association (AHA) "Follow the Guidelines - Heart Failure" risk scoring system, etc.

[0099] As used in this article, B-type natriuretic peptide (BNP) and N-terminal pro-B-type natriuretic peptide (NT-proBNP) are both commonly used biomarkers for heart failure. They are interconvertible; for example, the NT-proBNP threshold used in this article is 4623 ng / L, which is equivalent to a BNP threshold of 1008 ng / L. (ISHIHARA S et al.,(2022). New Conversion Formula Between B-Type Natriuretic Peptide and N-Terminal-Pro-B-Type Natriuretic Peptide. Circulation Journal,2022,1-9(doi:10.1253 / circj.CJ-22-0032)

[0100] According to the present invention, the content of intestinal microbiota metabolites is indicated by the normalized value of the mass spectrometry signal area.

[0101] According to the present invention, as is known from the prior art, the training set and validation set have the same meaning. In one embodiment of the present invention, the training set refers to the modeling cohort, which is a set of heart failure risk indicators, including the content of gut microbiota metabolites, in biological samples from heart failure patients. In one embodiment of the present invention, the validation set refers to the dataset used to test the performance of the training set. In one embodiment of the present invention, the content of gut microbiota metabolites can be represented as an absolute or relative value depending on the measurement method. For example, when mass spectrometry is used to measure the level (e.g., content) of gut microbiota metabolites, the peak intensity or area can represent the level of gut microbiota metabolites; when PCR is used to measure the level of biomarkers, the copy number of a gene or the copy number of a gene fragment can represent the level of the biomarker.

[0102] In one embodiment of the invention, the reference value refers to the reference value or normal value of a healthy control. Those skilled in the art will understand that, given a sufficiently large sample size, the range of normal (absolute) values ​​for each gut microbiota metabolite can be obtained through testing and calculation methods. Therefore, when biomarker levels are detected using methods other than mass spectrometry, the absolute values ​​of these biomarker levels can be directly compared to normal values, thereby evaluating the diagnosis of heart failure or early diagnosis of heart failure. Statistical methods can also be used in this invention.

[0103] According to the present invention, the term "biomarker," also known as "biological marker," refers to a measurable indicator of an individual's biological state. Such biomarkers can be any substance in an individual, as long as they are related to a specific biological state (e.g., disease) of the individual being examined, such as nucleic acid markers (e.g., DNA), protein markers, cytokine markers, chemokine markers, carbohydrate markers, antigen markers, antibody markers, species markers (markers of species / genus), and functional markers (KO / OG markers), etc. Biomarkers, after measurement and evaluation, are frequently used to examine normal biological processes, pathogenic processes, or pharmacological responses to therapeutic interventions, and are useful in many scientific fields.

[0104] According to the present invention, the term "individual" refers to an animal, particularly a mammal such as a primate, and preferably a human.

[0105] According to the present invention, the term "plasma" refers to the liquid component of whole blood. Depending on the separation method used, plasma may be completely free of cellular components or may contain varying amounts of platelets and / or small amounts of other cellular components.

[0106] According to the present invention, terms such as “a,” “an,” and “this” refer not only to a singular number of individuals, but also to a general class that can be used to describe a particular implementation.

[0107] It should be noted that the explanations of the terminology provided herein are only to enable those skilled in the art to better understand the invention and are not intended to limit the invention.

[0108] The NYHA classification of heart failure, developed by the New York Heart Association (NYHA), is a method for assessing cardiac function impairment based on the degree of activity that triggers heart failure symptoms. The NYHA classification divides heart failure into four classes, each representing a different degree of cardiac function impairment and clinical manifestations. Class I: The patient has heart disease but is not limited in their daily activities; general physical activity does not cause excessive fatigue, palpitations, shortness of breath, or angina. Class II: The patient has heart disease that significantly limits their physical activity. They are asymptomatic at rest, but general physical activity causes excessive fatigue, palpitations, shortness of breath, or angina. Class IV: The patient has heart disease that significantly limits their physical activity. They are asymptomatic at rest, but less than normal physical activity causes excessive fatigue, palpitations, shortness of breath, or angina. Class IV: The patient has heart disease and cannot engage in any physical activity; heart failure symptoms occur even at rest and worsen with physical activity.

[0109] COPD, or chronic obstructive pulmonary disease, is a group of lung diseases characterized by airflow limitation that is not fully reversible, progresses, and primarily affects the lungs. COPD is more common in heart failure than in other cardiovascular diseases, with a prevalence ranging from 10% to 40%, and is associated with prolonged hospital stays, readmission risk, heart failure decompensation, and can independently predict heart failure mortality. It is also noteworthy that the interaction between COPD and heart failure is common, and patients with both COPD and heart failure experience more severe impairment in vascular function and cardiac autonomic regulation compared to either condition alone.

[0110] Heart failure

[0111] Heart failure (HF) is a complex disease involving multiple systems, including hemodynamics, neurohormones, metabolism, and renal effects, as well as the recently reported gut microbiome. Notably, studies have shown that non-cardiac comorbidities in HF are associated with higher mortality rates, and there is limited optimization of guideline-based treatment prescriptions and medications.

[0112] However, currently available clinical risk scoring and risk stratification algorithms for heart failure are mainly based on traditional cardiovascular and hemodynamic factors (i.e., clinical parameters and circulating biomarkers - primarily natriuretic peptides), and rarely consider the impact of comorbidities (such as kidney injury, respiratory diseases, and metabolic diseases).

[0113] Furthermore, the heart failure-promoting effects of gut microbiota metabolites have added a new dimension to our understanding of heart failure. In particular, the trimethylamine oxide-related metabolites derived from gut microbiota (including acetyl-L-carnitine, γ-butyrate, and L-carnitine) have been shown to promote the severity and prognosis of heart failure.

[0114] ROC-AUC

[0115] ROC-AUC is a method for evaluating model accuracy. The ROC curve is the Receiver Operating Characteristic Curve, a coordinate graph composed of the false positive rate on the horizontal axis and the true positive rate on the vertical axis. It is a comprehensive indicator reflecting the sensitivity and specificity of continuous variables. AUC is the area under the ROC curve. ROC-AUC values ​​between 1.0 and 0.5 are considered accurate. The closer to 1, the better the predictive performance. Values ​​between 0.5 and 0.7 indicate lower accuracy, between 0.7 and 0.9 indicate some accuracy, and values ​​above 0.9 indicate high accuracy. An AUC of 0.5 indicates that the predictive method is completely ineffective. An AUC < 0.5 does not reflect reality and is extremely rare in practice.

[0116] The main advantages of this invention include:

[0117] (a) This invention provides an effective and accurate method for assisting in the assessment of the risk of readmission or death in patients with heart failure.

[0118] (b) Metabolites in the carnitine-trimethylamine oxide metabolic pathway associated with gut microbiota, whether singly or in combination, are closely related to the prognosis of heart failure. There is a bidirectional communication network between the gut and the heart, namely the "gut-heart axis". This invention is the first to use the OR value in logistic regression to weight or score gut microbiota metabolites individually or as a whole to assess the overall contribution of gut microbiota metabolites to the prognosis of heart failure.

[0119] (c) This invention is the first to incorporate multiple coexisting diseases, including the gut-heart axis (such as renal insufficiency, COPD, diabetes, and heart failure drug treatment), into a new risk model and score for risk stratification of adverse outcomes (rehospitalization and death) in heart failure patients within one year.

[0120] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Experimental methods in the following embodiments, unless otherwise specified, are generally performed under conventional conditions, such as those described in Sambrook et al., Molecular Cloning: A Laboratory Manual (New York: Cold Spring Harbor Laboratory Press, 1989), or as recommended by the manufacturer. Unless otherwise stated, percentages and parts are weight percentages and parts by weight.

[0121] Example 1: Indicator Screening and Risk Model Construction

[0122] 1.1 Experimental Subjects

[0123] This invention incorporated data from two cohorts. The first cohort was a single-center cohort of 806 patients with acute heart failure admitted to a European hospital. The primary endpoint was a composite of one-year all-cause mortality and / or readmission due to heart failure (death / HF). Endpoints were obtained by reviewing local hospital databases, national statistical data, and by contacting patients, and were verified by reviewing medical records. Each patient provided written informed consent for blood samples and outcome assessment during at least one year of follow-up. Biomarkers, including those related to the gut microbiome, were detected as previously described and reported (e.g., NT-proBNP, trimethylamine oxide, acetyl-L-carnitine, γ-butyrate, and L-carnitine). Glomerular filtration rate (eGFR) for all patients was calculated using a simplified dietary correction formula for kidney disease.

[0124] The second cohort was a multicenter heart failure cohort from multiple countries, including 1265 patients with progressively worsening or new-onset heart failure. As previously described, biomarkers, including gut microbiota-associated metabolites (trimethylamine oxide, acetyl-L-carnitine, γ-butylbetaine, and L-carnitine), were detected. NT-proBNP was measured using Roche's NT-proBNP assay.

[0125] 1.2 Model Construction

[0126] 1.2.1 Data Processing

[0127] Based on the data from the aforementioned cohorts, a risk assessment model for acute heart failure prognosis (readmission or death within one year) was developed, incorporating the contribution of the gut microbiota metabolome. Due to the heterogeneity of the two cohorts and differences in sampling and enrollment times, the two independent cohorts (n1 = 806, n2 = 1265) were merged for data integration. The integrated cohort samples were then randomly split into a training cohort (70%) and a validation cohort (30%), with a 70:30 allocation ratio.

[0128] The clinical variables and biomarker data for the training cohort (n=1444) and validation cohort (n=627) were obtained from the two independent cohorts mentioned above.

[0129] Patient demographics in these cohorts showed no significant differences in the variables tested between the training and validation cohorts. Overall, the median age of patients in both cohorts was 74 years, 68% were male, 80% were NYHA class III / IV, and the one-year mortality and / or readmission rate was 36%. Table 1 summarizes the patient demographics of the cohorts.

[0130] Table 1. Patient demographics for the training and validation cohorts.

[0131]

[0132]

[0133] Note: For continuous variables, data are reported as median (interquartile range); for categorical variables, data are reported as percentages. NT-proBNP: N-terminal pro-B-type natriuretic peptide; NYHA: New York Heart Association.

[0134] 1.2.2 Model Construction Method

[0135] The flowchart of the development strategy for the heart failure readmission or death risk assessment model within one year of this invention is as follows: Figure 1 As shown.

[0136] Univariate logistic regression analysis was used to analyze the relationship between clinical and biochemical indicators and the composite endpoint (death and / or readmission within one year) in the training cohort. Multivariate analysis showed significance of the variables (p < 0.05). Subsequently, logistic regression was used to include gut metabolites as a single predictor probability. A final multivariate model was built using stepwise logistic regression (backward selection). Variables with p > 0.05 were removed from the final model.

[0137] The variables were divided into two groups based on the Youden index. Then, the variables were ranked and assigned scores based on the odds ratio (OR) obtained from multivariate logistic regression.

[0138] Using the formula (maximum value - minimum value) / 3, we obtained a 3-point scale (low, medium, and high risk).

[0139] A dichotomous scoring method was further developed, dividing participants based on the Youden index and simplifying each group into low-risk and high-risk groups. The χ² trend test was used to assess changes in event incidence across the entire scoring model. Kaplan-Meier survival analysis was employed to test the utility of the risk model, and the log-rank test was used to examine changes in outcomes at each level.

[0140] The final model uses continuous logistic regression as a single predicted probability combined with the Net Reclassification Index (NRI), and compares this score with other risk scores (BIOSTAT compact and extended, ADHERE and OPTIMIZE-HF). The c-statistic is calculated based on the area under the curve (AUC) of the receiver operating characteristic (ROC) curve analysis when the same variables are combined linearly from the Cox regression.

[0141] Specifically, univariate logistic regression analysis was performed on all 31 available common variables in the training and validation queues, as shown in Table 1. (There were no significant differences between the 31 variables tested in the training and validation queues; therefore, there were no significant differences in variable selection between the training and validation queues. All variables are key indicators of clinical heart failure. To maximize the accuracy of selection, large datasets from two arrays were used for selection, which does not affect subsequent modeling.) The gut microbiota metabolites were categorized into a variable for predicting probabilities (i.e., gut microbiota metabolites). Then, significant variables were obtained through backward logistic regression, as shown in Table 2.

[0142] Further screening was conducted based on the significant variables in Table 2, and finally 11 significant variables (age, diastolic blood pressure, NT-proBNP, creatinine, COPD, history of HF, use of β-blockers, use of loop diuretics, diabetes, NYHA score, and gut microbiota metabolites) were obtained for the construction of the final model, as shown in Table 3.

[0143] Table 2. Outcomes of death and / or readmission within 1 year based on univariate logistic regression.

[0144]

[0145]

[0146] *Significant variables are included in the backward logistic regression analysis used to build the risk model.

[0147] NT-proBNP: N-terminal pro-B-type natriuretic peptide (NT-proBNP); NYHA: New York Heart Association.

[0148] Table 3. Risk model after multivariate inverse logistic regression for adverse outcomes of heart failure readmission and / or death within one year.

[0149]

[0150]

[0151] (1) Construction of the first model based on gut microbiota metabolites

[0152] First, based on the odds ratio (OR) (logistic regression) data in Table 2, the gut microbiota metabolites were weighted and scored to assess their contribution, resulting in the weight scores (weights) of each gut microbiota metabolite as shown in Table 4.

[0153] Table 4. Risk ranking and risk score of each gut microbiota metabolite

[0154] Gut microbiota metabolites OR / HR rankings Weighted scoring (weight) γ-Butylbetaine 1 4 Acetyl-L-carnitine 2 3 Trimethylamine oxide (TMAO) 3 2 L-carnitine 4 1 total 10

[0155] The highest OR was for γ-butylbetaine (OR, 1.30), rated 4 points; followed by acetyl-L-carnitine (OR, 1.03), trimethylamine oxide (TMAO) (OR, 1.01), and L-carnitine (OR, 1), rated 3, 2, and 1 points respectively. A value of 1 was assigned when the measured value of each metabolite was greater than or equal to a threshold, and 0 otherwise. In other words, a weighted value was obtained when the level of gut microbiota metabolites was greater than or equal to a threshold level, and 0 otherwise.

[0156] The final model score was S1 = 4 * γ-butyl betaine + 3 * acetyl L-carnitine + 2 * trimethylamine oxide + 1 * L-carnitine; the maximum value of S1 was 10 points.

[0157] The weight scores and thresholds of each gut microbiota metabolite are shown in Table 5.

[0158] Table 5. Weight scores and thresholds for metabolites of each gut microbiota.

[0159]

[0160] For example: If the concentration of γ-butyl betaine in a clinical sample is 1.6 μmol / L (greater than the threshold, assigned a value of 1), the concentration of acetyl-L-carnitine is 8.8 μmol / L (less than the threshold, assigned a value of 0), the concentration of trimethylamine oxide is 5.6 μmol / L (greater than the threshold, assigned a value of 1), and the concentration of L-carnitine is 84 μmol / L (equal to the threshold, assigned a value of 1), then S1 = 4*1 + 3*0 + 2*1 + 1*1 = 7 points.

[0161] (2) Construction of the second model based on 11 variables

[0162] Based on the odds ratio (OR) (logistic regression) data in Table 3, the 11 variable indicators are weighted and scored (the higher the OR value, the higher the score, and vice versa), and the contribution of each of the 11 variable indicators is evaluated to obtain the weight scores (weights) of the 11 variable indicators as shown in Table 6.

[0163] Table 6 shows the weighted scores of each variable indicator.

[0164]

[0165] When each of the 11 variables meets the conditions or its level is greater than or equal to the threshold level, a weighted value (assigned as 1, weighted value is 1 * weight) is obtained; otherwise, a value of 0 (assigned as 0, weighted value is 0 * weight) is obtained.

[0166] In this invention, the scores of various gut microbiota metabolites in the patient are integrated into an 11-variable model. In the 11-variable model, the gut microbiota metabolites have a weighted score or a weight value of 3 in the second model. Specifically, if the summed score of a gut microbiota metabolite in the first model reaches a passing score of 6 or higher, it is assigned a value of 1, and its risk score in the 11 variables = 1 × 3 = 3 points; conversely, it is assigned a value of 0, and its risk score in the 11 variables = 0 × 3 = 0 points.

[0167] The weight scores and thresholds of each variable indicator in the 11 variables are shown in Table 7.

[0168] Table 7 shows the weight scores and thresholds for each variable indicator.

[0169] Variables or indicators Weighted scoring Conditions or thresholds COPD 5 yes Previous history of heart failure 5 yes Loop diuretics use 5 yes NYHA Classification III / IV 4 yes diabetes 4 yes Gut microbiota metabolites 3 <![CDATA[S1 score ≥ 6]]> age 2 75 years old Creatinine 2 133.5 μmol / L NT-proBNP 2 4625pg / mL diastolic blood pressure 2 68.5 mmHg No beta-blockers used 2 yes

[0170] The final second model score S2 was obtained as follows: S2 = 5 * COPD + 5 * history of heart failure + 5 * use of loop diuretics + 4 * NYHA class III / IV + 4 * diabetes + 3 * gut microbiota metabolites + 2 * age + 2 * creatinine + 2 * NT-proBNP + 2 * diastolic blood pressure + 2 * no use of β-blockers; the maximum value of S2 was 36 points.

[0171] Based on the second model constructed in this invention, Kaplan-Meier analysis in the training cohort showed that there were significant differences in survival rates among the increased risk groups (chi-square 73.572-165.696, p<0.001), and the event incidence gradually increased (low risk = 17%, medium risk = 40%, high risk = 72%).

[0172] These results indicate that it is possible to differentiate at-risk groups based on comorbidity risk scores. Figure 2 A). The area under the curve / C statistic in the receiver operating characteristic (ROC) curve analysis in the training cohort was 0.71 (95% CI 0.69–0.74, p < 0.001).

[0173] (3) Model Validation

[0174] Survival analysis of the validation cohort of 627 patients showed a significant correlation between increasing risk group and worse survival (p<0.001). The high-risk group showed the lowest survival rate (58%) compared to the low-risk (chi-square 43.310, p<0.001) and intermediate-risk (chi-square 6.499, p=0.011) groups. Significant differences also existed between the intermediate-risk and low-risk groups (chi-square 29.977, p<0.001; low-risk event rate, 19%, intermediate-risk event rate, 44%). Figure 2 B).

[0175] The area under the curve / C statistic in the receiver operating characteristic (ROC) curve analysis in the validation cohort was 0.70 (95% CI 0.65–0.74, p < 0.001). Figure 2 As shown in B.

[0176] (4) Dichotomy scoring

[0177] This invention further establishes a dichotomous scoring system, simplifying each group into low-risk and high-risk groups to facilitate clinical use and dissemination (ease of use). The Youden Index (risk score <16 for low risk; ≥16 for high risk) was used to segment risk scores. Both the training and validation cohorts showed significant differences between the low-risk and high-risk groups (chi-square 40.107–155.481, p<0.001), with an approximately 2-fold increase in event rates between the low-risk and high-risk groups (22% low-risk and 52% high-risk in the training cohort; 26% low-risk and 50% high-risk in the validation cohort). Figure 3 AB).

[0178] Example 2: Comparison of the model of the present invention with other models

[0179] Net weight classification analysis was performed by comparing the model of this invention with other models (ADHERE, OPTIMIZE-HF, GWTG-HF, BIOSTAT-compact, and BIOSTAT-extended) applicable to the queue data used in this invention.

[0180] The results are shown in Table 8. Compared with other models, the risk model of this invention shows an overall improvement in reclassification (95% CI 0.4-33.3, p≤0.042). Compared with ADHERE, OPTIMIZE-HF, and GWTG-HF (NRI 37.6-45.1), the overall reclassification NRI scores of this risk model are similar, meaning that compared with the above three models, the model of this invention has a similar degree of improvement (i.e., improved prediction accuracy for event occurrence). Compared with the BIOSTAT-compact and -extended models, the NRI scores are also similar (NRI 16.2 and 12.4, respectively), meaning that compared with the above two models, the model of this invention has a similar degree of improvement (i.e., improved prediction accuracy for event occurrence).

[0181] The results above show that the increase in NRI reclassification values ​​compared to other models demonstrates that the model of this invention has better classification performance than existing models. It can accurately assess the risk of readmission and / or death of a heart failure patient within one year, thus more effectively assisting clinicians in making appropriate clinical decisions.

[0182] Table 8 compares the risk model with currently available clinical risk models, showing the net weight classification index of all-cause mortality and / or rehospitalization (death / HF) due to heart failure in one year.

[0183] Note: NRI, Net Weight Classification Index.

[0184] discuss

[0185] This invention develops a heart failure readmission and / or mortality risk assessment model / scoring system that combines known comorbidities (such as renal insufficiency, COPD, diabetes) and the contribution of the gut microbiome to stratify the prognosis of acute / worsening heart failure.

[0186] Comorbidities of heart failure lead to more complex management and poorer outcomes, with an increased risk of readmission and death. The multi-comorbidity heart failure risk score (heart failure readmission and / or death risk assessment model / scoring system) of this invention includes additional values ​​considering non-cardiac comorbidities (i.e., renal insufficiency, COPD, diabetes) and gut microbiota metabolites. Furthermore, its risk score shows an association with adverse outcomes of heart failure (i.e., readmission and heart failure death), and the reclassification of this risk model is generally an improvement over currently used risk scores (i.e., ADHERE, OPTIMIZE, and GWTG).

[0187] With the aging population leading to increased prevalence and hospitalization rates of heart failure, comorbidities are becoming more common. More than half of patients with congestive heart failure have at least one comorbidity (i.e., diabetes, kidney impairment, obesity, and / or respiratory disease caused by smoking) and age-related impairments (age, frailty and falls, and confusion), which represents a clinical challenge necessitates the development of a risk score that takes into account heart failure comorbidities for HF management.

[0188] The final risk model in this study consisted of 11 variables, 10 of which were routinely collected in clinical practice, and 1 new variable was derived from a combined gut microbiota metabolome. Of these 11 variables, 6 (age, diabetes, creatinine level, NYHA class, natriuretic peptide, and history of heart failure) are commonly used in validated acute heart failure risk models and are known to be independently associated with adverse outcomes.

[0189] In the model of this invention, COPD is a major factor. Chronic obstructive pulmonary disease is more prevalent in heart failure than in other cardiovascular diseases, with a prevalence ranging from 10% to 40%, and is associated with prolonged hospital stays, readmission risk, heart failure decompensation, and can independently predict heart failure mortality. It is also noteworthy that the interaction between COPD and heart failure is common, and patients with both COPD and heart failure experience more severe impairment in vascular function and cardiac autonomic regulation compared to either alone. Furthermore, recent Mendelian randomization analyses have shown a causal relationship between COPD and heart failure, but no evidence of the opposite (heart failure to COPD). This may reflect progressive loss of exercise capacity, muscle atrophy, increased sedentary status, and frailty as the disease progresses, leading to increasingly severe symptoms in patients with either COPD or heart failure. The 2021 European Heart Failure Guidelines state that COPD management can improve cardiac function (by initiating bronchodilators and beta-blockers according to the severity of disease in each case and adjusting the dosage accordingly), and improve heart failure hospitalization rates, and that COPD is well tolerated in the treatment of heart failure.

[0190] Type 2 diabetes (T2D) is another major contributor to the model of this invention, and has been shown to increase the risk of heart failure hospitalization and death by two times compared to heart failure patients without diabetes. Furthermore, in terms of the effects of T2D itself, renal insufficiency is common in T2D, which is a further factor contributing to adverse outcomes. Notably, heart failure patients with T2D have a worse prognosis compared to heart failure patients with normoglycemia, exhibiting progressive right ventricular dysfunction and dyskinesia.

[0191] Heart failure medication (i.e., using loop diuretics and not using beta-blockers) also contributes to this model. Loop diuretics are first-line drugs for treating congestion, and long-term use or higher doses are associated with poorer heart failure outcomes and worsening renal insufficiency. In patients with advanced heart failure, heart failure mortality is dose-dependently related to loop diuretic dosage. The contribution of not using beta-blockers to mortality / heart failure is also evident, consistent with evidence that beta-blockers can improve survival and are recommended for their use in guideline-based heart failure treatment.

[0192] An integral-based risk model / score was developed, categorized into three risk groups (low / intermediate / high risk) and simplified to two groups (low risk and high risk) associated with the composite endpoint (1-year mortality / heart failure). When categorized into two and three groups, the survival rate of patients in the high-risk group was approximately half that of patients in the low-risk group, demonstrating its ability to stratify high-risk adverse outcomes.

[0193] Compared with other previously reported clinical risk models, the model of this invention (heart failure comorbidity model) (C-statistic 0.71) performed better in some cases (C-statistics for heart failure comorbidity score, Seattle model, Framingham score, CHARM score and SENIORS trial were 0.70-0.72, 0.72, 0.69 / male and 0.72 / female, 0.74 and 0.69, respectively).

[0194] Furthermore, the established risk score has been applied to patients with acute / exacerbating heart failure. Most similar risk models are applicable to patients with acute heart failure presenting in the emergency department, with reported C-statistics ranging from 0.74 to 0.84; however, these models are primarily designed for risk stratification of short-term / in-hospital outcomes, from 7-day mortality to 30-day mortality, rather than long-term outcomes. Our multi-disease comorbid heart failure risk score includes the same variables as previously reported models and can stratify risk of mortality and heart failure readmission for up to 1 year.

[0195] The carnitine-TMAO pathway, a gut microbiota metabolite, acts as a carrier in the transmembrane transfer of activated acyl and acetyl groups, participating in fatty acid oxidation. More specifically, in heart failure, carnitine dysregulation is common in advanced heart failure, and its effects are associated with cardiac cachexia / sarcopenia. Previous studies have shown that single or combined carnitine-TMAO pathway gut microbiome markers are closely associated with heart failure mortality, heart failure severity, and prognosis, with increased levels of related metabolites observed in patients with poorer prognoses. The described model explains this novel pathophysiology's additional contribution to heart failure.

[0196] All documents mentioned in this invention are incorporated herein by reference as if each document were individually incorporated by reference. Furthermore, it should be understood that after reading the foregoing teachings of this invention, those skilled in the art can make various alterations or modifications to this invention, and these equivalent forms also fall within the scope defined by the appended claims.

Claims

1. A system for assessing the risk of readmission or death within one year in patients with heart failure, characterized in that, The system includes: (a) An input module configured to input heart failure risk index data of the subject to be tested; The heart failure risk indicators include: chronic obstructive pulmonary disease (COPD), history of heart failure, use of loop diuretics, NYHA class III / IV, diabetes, gut microbiota metabolites, age, creatinine, N-terminal pro-B-type natriuretic peptide (NT-proBNP), diastolic blood pressure, and no use of beta-blockers; and the gut microbiota metabolites include: γ-butylbetaine, acetyl-L-carnitine, trimethylamine oxide, and L-carnitine; (b) An evaluation module, which is configured to perform the following functions: i) Compare the input risk indicator data (including data of each gut microbiota metabolite) with the conditions or thresholds pre-stored in the system to obtain the comparison result; if a certain input risk indicator data value meets the condition or is greater than its corresponding threshold, it is assigned a value of 1; if the risk indicator data value does not meet the condition or is less than its corresponding threshold, it is assigned a value of 0. ii) Substitute the comparison results of gut microbiota metabolite data into the first-level risk scoring formula to obtain the first risk score; and compare the first risk score with the pre-set conditions or thresholds in the system to obtain the comparison result G of gut microbiota metabolites and their corresponding conditions or thresholds. iii) The comparison result G is compared with the heart failure risk index data of other non-gut microbiota metabolites, and then substituted into the second risk scoring formula to calculate the second risk score, thereby obtaining the assessment result. The scoring formulas for the first and second risks are expressed as follows: Wherein, when the scoring formula is the first risk scoring formula (S1), Wi is the weight value of each gut microbiota metabolite, and Pi is the comparison result of each gut microbiota metabolite data with its corresponding conditions or thresholds; When the scoring formula is the second risk scoring formula (S2), Wi is the weight value of each risk indicator; Pi is the comparison result of each risk indicator data with its corresponding conditions or thresholds; The assessment based on the second score includes: when the second risk score is greater than or equal to the risk cutoff value, it indicates a high risk of readmission or death within one year for the subject; conversely, it indicates a low risk. (c) Output module, which is configured to output the evaluation result.

2. The system as described in claim 1, characterized in that, The risk cutoff value is 16.

3. The system as described in claim 2, characterized in that, The assessment based on the second score includes: when the second risk score is ≥16, the subject has a high risk of readmission or death within one year; when the second risk score is <16, the subject has a low risk of readmission or death within one year.

4. The system as described in claim 1, characterized in that, The first risk scoring formula is: S1 = 4 * γ-butyl betaine + 3 * acetyl L-carnitine + 2 * trimethylamine oxide + 1 * L-carnitine.

5. The system as described in claim 4, characterized in that, In the first risk scoring formula, the values ​​of γ-butyl betaine, acetyl L-carnitine, trimethylamine oxide, or L-carnitine are 1 or 0, respectively.

6. The system as described in claim 1, characterized in that, When the first risk score is ≥6, the gut microbiota metabolites are assigned a value of 1 (G); otherwise, they are assigned a value of 0.

7. The system as described in claim 1, characterized in that, The second risk scoring formula is: S2 = 5 * Chronic obstructive pulmonary disease (COPD) + 5 * History of heart failure + 5 * Use of loop diuretics + 4 * NYHA class III / IV + 4 * Diabetes + 3 * Gut microbiota metabolites + 2 * Age + 2 * Creatinine + 2 * NT-proBNP + 2 * Diastolic blood pressure + 2 * No use of β-blockers.

8. The system as described in claim 7, characterized in that, In the second risk scoring formula, the values ​​for chronic obstructive pulmonary disease (COPD), history of heart failure, use of loop diuretics, NYHA class III / IV, diabetes, gut microbiota metabolites, age, creatinine, NT-proBNP, diastolic blood pressure, or no use of beta-blockers are 1 or 0, respectively.

9. A method for assessing the risk of readmission or death within one year of heart failure, comprising the following steps: (a) Provide data, including heart failure risk index data for the subjects being tested; in, The heart failure risk indicators include: chronic obstructive pulmonary disease (COPD), history of heart failure, use of loop diuretics, NYHA class III / IV, diabetes, gut microbiota metabolites, age, creatinine, N-terminal pro-B-type natriuretic peptide (NT-proBNP), diastolic blood pressure, and no use of beta-blockers; and the gut microbiota metabolites include: γ-butylbetaine, acetyl-L-carnitine, trimethylamine oxide, and L-carnitine; (b) Analysis and evaluation, which includes the following steps: (s1) The input risk indicator data (including data of each gut microbiota metabolite) is compared with the conditions or thresholds pre-stored in the system to obtain the comparison result; if a certain input risk indicator data value meets the condition or is greater than its corresponding threshold, it is assigned a value of 1; if the risk indicator data value does not meet the condition or is less than its corresponding threshold, it is assigned a value of 0. (s2) Substitute the comparison results of gut microbiota metabolite data into the first-level risk scoring formula to obtain a first risk score; and compare the first risk score with the pre-set conditions or thresholds in the system to obtain the comparison result G of gut microbiota metabolites and their corresponding conditions or thresholds; and (s3) The comparison result G is compared with the heart failure risk index data of other non-gut microbiota metabolites, and then substituted into the second risk scoring formula for calculation to obtain the second risk score, thereby obtaining the assessment result. The scoring formulas for the first and second risks are expressed as follows: Wherein, when the scoring formula is the first risk scoring formula (S1), Wi is the weight value of each gut microbiota metabolite, and Pi is the comparison result of each gut microbiota metabolite data with its corresponding conditions or thresholds; When the scoring formula is the second risk scoring formula (S2), Wi is the weight value of each risk indicator; Pi is the comparison result of each risk indicator data with its corresponding conditions or thresholds; The assessment based on the second score includes: when the second risk score is greater than or equal to the risk cutoff value, it indicates a high risk of readmission or death within one year for the subject; conversely, it indicates a low risk. (3) Output the results.

10. A method for constructing a risk assessment model for readmission or death within one year in patients with heart failure, characterized in that, Including the following steps: (s1) Provide a first dataset containing heart failure risk indicator data, wherein the risk indicator data includes: chronic obstructive pulmonary disease (COPD), history of heart failure, use of loop diuretics, NYHA class III / IV, diabetes, gut microbiota metabolites, age, creatinine, N-terminal pro-B-type natriuretic peptide (NT-proBNP), diastolic blood pressure, and no use of β-blockers; and wherein the gut microbiota metabolites include: γ-butyl betaine, acetyl-L-carnitine, trimethylamine oxide, and L-carnitine; (s2) Based on the data information of the first dataset, a first risk model is constructed using logistic regression, and then a second risk model is constructed based on the first risk model to assess the risk of readmission or death within one year of heart failure (second risk model).

Images