Cai Min, Deng Yue, Hu Tianyang
Department of Nephropathy and Rheumatism, Yongchuan Hospital of Chongqing Medical University (The Fifth Clinical College of Chongqing Medical University), Chongqing, People's Republic of China.
Department of Respiratory and Critical Care Medicine, The Fifth People's Hospital of Chongqing, Chongqing, People's Republic of China.
Int J Chron Obstruct Pulmon Dis. 2024 Mar 4;19:619-632. doi: 10.2147/COPD.S444888. eCollection 2024.
Acute kidney injury (AKI) is a common complication of acute exacerbation of chronic obstructive pulmonary disease (AECOPD), and inflammation is the potential link between AKI and AECOPD. However, little is known about the incidence and risk stratification of AKI in critically ill AECOPD patients. In this study, we aimed to establish risk model based on white blood cell (WBC)-related indicators to predict AKI in critically ill AECOPD patients.
For the training cohort, data were taken from the Medical Information Mart for eICU Collaborative Research Database (eICU-CRD) database, and for the validation cohort, data were taken from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database. The study employed logistic regression analysis to identify the major predictors of WBC-related biomarkers on AKI prediction. Subsequently, a risk model was developed by multivariate logistic regression, utilizing the identified significant indicators.
Finally, 3551 patients were enrolled in training cohort, 926 patients were enrolled in validation cohort. AKI occurred in 1206 (33.4%) patients in training cohort and 521 (56.3%) patients in validation cohort. According to the multivariate logistic regression analysis, four WBC-related indicators were finally included in the novel risk model, and the risk model had a relatively good accuracy for AKI in the training set (C-index, 0.764, 95% CI 0.749-0.780) as well as in the validation set (C-index, 0.738, 95% CI: 0.706-0.770). Even after accounting for other models, the critically ill AECOPD patients in the high-risk group (risk score > 3.44) still showed an increased risk of AKI (odds ratio: 4.74, 95% CI: 4.07-5.54) compared to those in low-risk group (risk score ≤ 3.44). Moreover, the risk model showed outstanding calibration capability as well as therapeutic usefulness in both groups for AKI and ICU mortality and in-hospital mortality of critical ill AECOPD patients.
The novel risk model showed good AKI prediction performance. This risk model has certain reference value for the risk stratification of AECOPD complicated with AKI in clinically.
急性肾损伤（AKI）是慢性阻塞性肺疾病急性加重（AECOPD）的常见并发症，炎症是AKI与AECOPD之间的潜在联系。然而，关于重症AECOPD患者中AKI的发病率和风险分层知之甚少。在本研究中，我们旨在建立基于白细胞（WBC）相关指标的风险模型，以预测重症AECOPD患者的AKI。
对于训练队列，数据取自电子重症监护病房协作研究数据库（eICU-CRD）数据库，对于验证队列，数据取自重症监护医学信息集市-IV（MIMIC-IV）数据库。该研究采用逻辑回归分析来确定WBC相关生物标志物对AKI预测的主要预测因素。随后，利用确定的显著指标通过多变量逻辑回归建立风险模型。
最终，3551例患者纳入训练队列，926例患者纳入验证队列。训练队列中有1206例（33.4%）患者发生AKI，验证队列中有521例（56.3%）患者发生AKI。根据多变量逻辑回归分析，新型风险模型最终纳入了4个与WBC相关的指标，该风险模型在训练集（C指数，0.764，95%CI 0.749-0.780）和验证集（C指数，0.738，95%CI：0.706-0.770）中对AKI具有相对较好的预测准确性。即使在考虑了其他模型之后，与低风险组（风险评分≤3.44）相比，高风险组（风险评分>3.44）的重症AECOPD患者发生AKI的风险仍然增加（比值比：4.74，95%CI：4.07-5.54）。此外，该风险模型在两组中对AKI以及重症AECOPD患者的ICU死亡率和住院死亡率均显示出出色的校准校准能力能力和治疗实用性。
新型风险模型显示出良好的AKI预测性能。该风险模型对临床上AECOPD合并AKI的风险分层具有一定的参考价值。
Zotero 插件
