Lan Qingzhan, He Shanshan, Lu Chao, Huan Cheng
Clin Lab. 2025 Jul 1;71(7). doi: 10.7754/Clin.Lab.2025.240936.
Sepsis is the leading cause of death for critically ill patients worldwide, and blood purification technology is an effective method for rapidly improving and treating sepsis. At present, there is a lack of sufficient clinical data on the timing of blood purification intervention for sepsis patients in China. This study aimed to develop an effective and straightforward tool for predicting the need for blood purification in the sepsis population using an evaluation model.
A total of 346 patients were enrolled in the study. The patients were divided into two groups: the blood purification group (n = 80) and the non-blood purification group (n = 266). Demographic information, medical history, clinical performance, laboratory results, and treatment characteristics were extracted from the medical records of all participants. The optimal predictive risk factors were selected using the least absolute shrinkage and selection operator (LASSO) method to reduce the high-dimensional data. Multivariate logistic regression analysis and the creation of a nomogram were performed using R software (3.1.1). The model's discrimination, calibration, and clinical utility were evaluated using the C-index, calibration plot, and decision curve analysis, respectively. The 95% confidence interval (CI) for the calculated odds ratio (OR) was also estimated.
The novel predictive nomogram, developed using β2-microglobulin (BMG), urea nitrogen (BUN), acute kidney injury (AKI), neutrophil gelatinase-associated lipocalin (NGAL), uric acid (URIC), and estimated glomerular filtration rate (eGFR), could be easily applied to predict the appropriate timing for blood purification. Using the nomogram to predict the risk of requiring blood purification provided greater benefits than the standard method.
Our findings provide an effective prediction model that will assist clinicians in identifying the optimal time for blood purification.
脓毒症是全球危重症患者的主要死因，血液净化技术是快速改善和治疗脓毒症的有效方法。目前，中国缺乏关于脓毒症患者血液净化干预时机的充分临床数据。本研究旨在使用评估模型开发一种有效且简便的工具，用于预测脓毒症人群对血液净化的需求。
本研究共纳入346例患者。患者分为两组：血液净化组（n = 80）和非血液净化组（n = 266）。从所有参与者的病历中提取人口统计学信息、病史、临床表现、实验室检查结果和治疗特征。使用最小绝对收缩和选择算子（LASSO）方法选择最佳预测风险因素，以减少高维数据。使用R软件（3.1.1）进行多变量逻辑回归分析并创建列线图。分别使用C指数、校准图和决策曲线分析评估模型的辨别力、校准度和临床实用性。还估计了计算的比值比（OR）的95%置信区间（CI）。
使用β2-微球蛋白（BMG）、尿素氮（BUN）、急性肾损伤（AKI）、中性粒细胞明胶酶相关脂质运载蛋白（NGAL）、尿酸（URIC）和估计肾小球滤过率（eGFR）开发的新型预测列线图可轻松用于预测血液净化的合适时机。使用列线图预测需要血液净化的风险比标准方法更具优势。
我们的研究结果提供了一种有效的预测模型，将有助于临床医生确定血液净化的最佳时机。
