ChiCTR2600126515 版本V1.0 版本创建时间2026/06/10 11:50:39 中国临床试验注册中心

审核状态:

Project audit state:

通过审核

Successful

注册号:

Registration number:

ChiCTR2600126515 

最近更新日期:

Date of Last Refreshed on:

2026-06-10 11:50:27 

注册时间:

Date of Registration:

2026-06-10 00:00:00 

注册号状态:

预注册

Registration Status:

Prospective registration

注册题目:

脓毒症患者肾替代治疗启动时机模型构建与验证

Public title:

Model Development and Validation for the Timing of Renal Replacement Therapy Initiation in Patients with Sepsis

注册题目简写:

English Acronym:

研究课题的正式科学名称:

脓毒症患者肾替代治疗启动时机模型构建与验证

Scientific title:

Model Development and Validation for the Timing of Renal Replacement Therapy Initiation in Patients with Sepsis

研究课题代号(代码):

Study subject ID:

在二级注册机构或其它机构的注册号:

The registration number of the Partner Registry or other register:

申请注册联系人:

麦振华 

研究负责人:

麦振华 

Applicant:

Zhenhua Mai 

Study leader:

Zhenhua Mai 

申请注册联系人电话:

Applicant telephone:

+86 18813611088

研究负责人电话:

Study leader's
telephone:

+86 759 2387093

申请注册联系人传真 :

Applicant Fax:

研究负责人传真:

Study leader's fax:

申请注册联系人电子邮件:

Applicant E-mail:

zhenhuamai@gdmu.edu.cn

研究负责人电子邮件:

Study leader's E-mail:

maikson@126.com

申请单位网址(自愿提供):

Applicant website(voluntary supply):

研究负责人网址(自愿提供):

Study leader's website(voluntary supply):

申请注册联系人通讯地址:

广东省湛江市霞山区人民大道南57号

研究负责人通讯地址:

广东省湛江市霞山区人民大道南57号

Applicant address:

57 Renmin Avenue South, Xiashan District, Zhanjiang, Guangdong

Study leader's address:

57 Renmin Avenue South, Xiashan District, Zhanjiang, Guangdong

申请注册联系人邮政编码:

Applicant postcode:

研究负责人邮政编码:

Study leader's postcode:

申请人所在单位:

广东医科大学附属医院

Applicant's institution:

The Affiliated Hospital of Guangdong Medical University

研究负责人所在单位:

广东医科大学附属医院

Affiliation of the Leader:

Affiliated Hospital of Guangdong Medical University

是否获伦理委员会批准:

Approved by ethic committee:

Yes

伦理委员会批件文号:

Approved No. of ethic committee:

PJKT2026-043

伦理委员会批件附件:

Approved file of Ethical Committee:

查看附件View

批准本研究的伦理委员会名称:

广东医科大学附属医院临床科研伦理委员会

Name of the ethic committee:

Ethics Committee for Institutional Review Affiliated Hospitals of Guangdong Medical University

伦理委员会批准日期:

Date of approved by ethic committee:

2026-04-28 00:00:00

伦理委员会联系人:

王健丽

Contact Name of the ethic committee:

Wang Jianli

伦理委员会联系地址:

广东省湛江市霞山区人民大道南57号

Contact Address of the ethic committee:

57 Renmin Avenue South, Xiashan District, Zhanjiang, Guangdong

伦理委员会联系人电话:

Contact phone of the ethic committee:

+86 759 2386971

伦理委员会联系人邮箱:

Contact email of the ethic committee:

3221452@qq.com

研究实施负责(组长)单位:

广东医科大学附属医院

Primary sponsor:

Affiliated Hospital of Guangdong Medical University

研究实施负责(组长)单位地址:

广东省湛江市霞山区人民大道南57号

Primary sponsor's address:

57 Renmin Avenue South, Xiashan District, Zhanjiang, Guangdong

试验主办单位(项目批准或申办者):

Secondary sponsor:

国家:

中国

省(直辖市):

广东省

市(区县):

Country:

China

Province:

Guangdong

City:

单位(医院):

广东医科大学附属医院

具体地址:

广东省湛江市霞山区人民大道南57号

Institution
hospital:

Affiliated Hospital of Guangdong Medical University

Address:

57 Renmin Avenue South, Xiashan District, Zhanjiang, Guangdong

经费或物资来源:

自选课题(自筹)

Source(s) of funding:

Self-raised

研究疾病:

脓毒症  

Target disease:

Sepsis

研究疾病代码:

Target disease code:

研究类型:

观察性研究

Study type:

Observational study

研究所处阶段:

其它 

Study phase:

N/A

研究设计:

析因分组(即根据危险因素或暴露因素分组) 

Study design:

Factorial 

研究目的:

本研究计划基于MIMIC-Ⅳ数据库,采用多种机器学习方法,构建脓毒症患者RRT最佳启动时机的预测模型,通过医院电子病历系统(Electronic Medical Record,EMR)数据对预测模型进行验证,选择最优的RRT启动时机模型,并运用于临床。  

Objectives of Study:

This study intends to construct a prediction model for the optimal initiation timing of renal replacement therapy (RRT) in patients with sepsis by employing multiple machine learning algorithms based on the MIMIC-IV database. The prediction model will be validated using electronic medical record (EMR) data from hospital information systems. The optimal model for RRT initiation timing will be selected and further applied in clinical practice.

药物成份或治疗方案详述:

 

Description for medicine or protocol of treatment in detail:

 

纳入标准:

1. 年龄>18岁且符合脓毒症3.0诊断标准的,脓毒症定义为疑似感 染伴序贯器官衰竭评估(SOFA)评分急性增加≥2分,如果脓毒症在同一住院期间发生不止一次,则仅使用首次诊断的信息; 2. 患者入住时长>24小时且如果患者多次入住ICU,仅分析首次入院的数据。

Inclusion criteria

1. Included patients were aged over 18 years and met the Sepsis-3 diagnostic criteria. Sepsis was defined as suspected infection with an acute increase of ≥ 2 points in the Sequential Organ Failure Assessment (SOFA) score. For patients with multiple episodes of sepsis during the same hospitalization, only the data from the first diagnostic episode was retained for analysis. 2. Patients with an ICU stay longer than 24 hours were included. For patients with multiple ICU admissions, only the data from the first admission was analyzed.

排除标准:

1.孕妇、哺乳期女性;
2.临床资料严重缺失(核心预测变量缺失值>30%);
3.既往确诊慢性肾脏病(CKD)4-5 期或既往接受长期肾脏替代治疗;
4.入住 ICU 后 24 小时内放弃治疗、转入临终关怀或预期生存期<72 小时。

Exclusion criteria:

1.Pregnant or lactating women;
2.Severely missing clinical data (missing rate of core predictive variables > 30%);
3.Patients with a prior diagnosis of chronic kidney disease (CKD) stage 4–5 or a history of long-term renal replacement therapy;
4.Patients who withdrew treatment, were transferred to hospice care within 24 hours of ICU admission, or had an expected survival time of less than 72 hours.

研究实施时间:

Study execute time:

From 2026-05-01 00:00:00 To 2028-04-28 00:00:00  

征募观察对象时间:

Recruiting time:

From 2026-06-15 00:00:00 To 2028-04-28 00:00:00

干预措施:

Interventions:

组别:

肾脏替代治疗组

样本量:

250

Group:

Group of Renal Replacement Therapy

Sample size:

干预措施:

干预措施代码:

Intervention:

None

Intervention code:

组别:

不使用肾脏替代治疗组

样本量:

500

Group:

Group of Non-Renal Replacement Therapy

Sample size:

干预措施:

干预措施代码:

Intervention:

None

Intervention code:

研究实施地点:

Countries of recruitment and research settings:

国家:

中国

省(直辖市):

广东省 

市(区县):

 

Country:

China

Province:

Guangdong

City:

单位(医院):

广东医科大学附属医院 

单位级别:

三级甲等 

Institution
hospital:

Affiliated Hospital of Guangdong Medical University

Level of the institution:

Tertiary A

测量指标:

Outcomes:

指标中文名:

28 天死亡率

指标类型:

主要指标

Outcome:

28-day mortality

Type:

Primary indicator

测量时间点:

以患者入 ICU / 确诊脓毒症 首日为起点,随访至第 28 天,判定是否死亡,即为 28 天死亡率测量时间点。

测量方法:

电子病历回溯:从 EMR、MIMIC-IV 等数据库提取入院、诊断、死亡时间 院内住院系统:调取出院小结、死亡记录、病程记录核实生存状态 电话随访:出院患者通过电话回访确认 28 天生存情况

Measure time point of outcome:

patients are followed up until day 28 to determine survival status, which defines the time point

Measure method:

Retrospective review of electronic medical records: extract the time of admission, diagnosis and death from databases such as EMR and MIMIC-IV;Hospital inpatient system: verify the survival status by retrieving discharge summaries, death records and medical course records;Telephone follow-up: confirm the 28-day survival status of discharged patients via telephone interview.

指标中文名:

28 天无 RRT 治疗天数

指标类型:

次要指标

Outcome:

28-day RRT-free days

Type:

Secondary indicator

测量时间点:

以患者入住 ICU 首日或脓毒症确诊首日为观察起点,连续观察至第 28 天;统计 28 天内未接受任何肾脏替代治疗的总天数,即为 28 天无 RRT 治疗天数。

测量方法:

通过医院电子病历、ICU 病程记录、治疗操作记录及 MIMIC-IV 数据库,提取每日 RRT 实施与否的记录,汇总计算 28 天无 RRT 治疗天数。

Measure time point of outcome:

the total number of days without receiving any renal replacement therapy (RRT) within the 28-day

Measure method:

Relevant daily records on whether renal replacement therapy (RRT) was administered were extracted from hospital electronic medical records, ICU medical course records, treatment operation records and the MIMIC-IV database, and the 28-day RRT-free days were subsequently summarized and calculated.

指标中文名:

精确率

指标类型:

次要指标

Outcome:

Precision

Type:

Secondary indicator

测量时间点:

建模完成后,分别在训练集、内部验证测试集测算

测量方法:

统计所有模型预测需行 RRT 的样本中,真实临床实施 RRT 的样本占比:精确率 = 真阳性 /(真阳性 + 假阳性)

Measure time point of outcome:

Measured separately on the training set and internal validation test set post-modeling.

Measure method:

Proportion of true RRT recipients among model-predicted RRT cases; formula: Precision = TP/(TP+FP)

指标中文名:

准确率

指标类型:

次要指标

Outcome:

Accuracy

Type:

Secondary indicator

测量时间点:

建模完成后,分别在训练集、内部验证测试集测算

测量方法:

全部研究样本中模型预测结果与实际 RRT 结局一致的样本占比:准确率 =(真阳性 + 真阴性)/ 总样本量

Measure time point of outcome:

Measured separately on the training set and internal validation test set post-modeling.

Measure method:

Proportion of samples with consistent predicted and actual RRT status across all subjects; formula: Accuracy=(TP+TN)/Total cases

指标中文名:

召回率

指标类型:

次要指标

Outcome:

Recall

Type:

Secondary indicator

测量时间点:

建模完成后,分别在训练集、内部验证测试集测算

测量方法:

统计全部真实接受 RRT 治疗的样本里,被模型预判为需要 RRT 的样本占比:召回率 = 真阳性 /(真阳性 + 假阴性)

Measure time point of outcome:

Measured separately on the training set and internal validation test set post-modeling.

Measure method:

Percentage of actual RRT recipients predicted as RRT-positive; formula: Recall = TP/(TP+FN)

指标中文名:

F1 值

指标类型:

次要指标

Outcome:

F1

Type:

Secondary indicator

测量时间点:

建模完成后,分别在训练集、内部验证测试集测算

测量方法:

F1 = 2 * (精准率 * 召回率) / (精准率 +召回率)

Measure time point of outcome:

Measured separately on the training set and internal validation test set post-modeling.

Measure method:

F1 = 2 * (Precision * Recall) / (Precision + Recall)

指标中文名:

受试者工作特征曲线下面积

指标类型:

次要指标

Outcome:

area under the curve

Type:

Secondary indicator

测量时间点:

建模完成后,优选训练集最优模型,同步在训练集、内部验证测试集验证

测量方法:

通过R 语言的pROC 包进行绘制

Measure time point of outcome:

Optimal model screened on training set and validated in training and internal test sets.

Measure method:

Plotted using the pROC package in R.

采集人体标本:

Collecting sample(s)
from participants:

标本中文名:

血液

组织:

Sample Name:

Blood

Tissue:

人体标本去向

使用后销毁  

说明

Fate of sample:

Destruction after use  

Note:

标本中文名:

尿液

组织:

Sample Name:

Urine

Tissue:

人体标本去向

使用后销毁  

说明

Fate of sample:

Destruction after use  

Note:

征募研究对象情况:

Recruiting status:

尚未开始

Not yet recruiting

年龄范围:

Participant age:

最小 Min age 18 years
最大 Max age years

性别:

男女均可

Gender:

Both

随机方法(请说明由何人用什么方法产生随机序列):

Randomization Procedure (please state who generates the random number sequence and by what method):

None

是否公开试验完成后的统计结果:

Calculated Results after the Study Completed public access:

不公开/Private

盲法:

Blinding:

None

是否共享原始数据:

IPD sharing

否No

共享原始数据的方式(说明:请填入公开原始数据日期和方式,如采用网络平台,需填该网络平台名称和网址):

涉及到院内研究数据,未经广东医科大学附属医院医院伦理委员会批准不得共享。

The way of sharing IPD”(include metadata and protocol, If use web-based public database, please provide the url):

All institutional research data involved shall not be shared without the approval of the Ethics Committee of Affiliated Hospital of Guangdong Medical University.

数据采集和管理(说明:数据采集和管理由两部分组成,一为病例记录表(Case Record Form, CRF),二为电子采集和管理系统(Electronic Data Capture, EDC),如ResMan即为一种基于互联网的EDC:

CRF 填写:由经统一培训的研究人员严格按照纳排标准筛选研究对象,逐例完整、准确填写 CRF,不得随意涂改,确需修改时保留原始记录并标注修改原因与时间。 EDC 录入:以纸质 / 电子 CRF 为依据,采用双人独立录入方式将数据同步录入 EDC 系统,系统自动比对双录差异并提示核查,确保录入数据与原始病历、CRF 完全一致。

Data collection and Management (A standard data collection and management system include a CRF and an electronic data capture:

CRF completion:Uniformly trained researchers screen eligible participants strictly according to the inclusion and exclusion criteria, and complete the Case Report Form (CRF) item by item in a full and accurate manner. Random alteration is not allowed. Any necessary revisions shall keep the original records intact, with clear annotation of the revision reason and date.EDC entry:Based on paper or electronic CRFs, data are independently entered into the Electronic Data Capture (EDC) system by two researchers separately. The system automatically compares discrepancies between the two entries and prompts verification, ensuring full consistency among the entered data, original medical records and CRFs.

数据与安全监察委员会:

Data and Safety Monitoring Committee:

无/No

注册人:

Name of Registration:

 2026-06-10 11:50:27