摘要
背景：人工智能（AI）通过机器学习（ML）模型似乎可以在某些临床环境中提供准确和精确的急性肾损伤（AKI）风险分类，但其在真实世界环境中的表现和实施尚未得到证实：方法：检索了截至 2023 年 8 月的 PubMed、EMBASE、Web of Science 和 Scopus。使用与 AKI、AI 和 ML 相关的文本词检索了报道经外部验证的预测住院成人和儿童患者 AKI 发病、AKI 严重程度和 AKI 后并发症的模型的文章。两名独立审稿人筛选了文章标题、摘要和全文。使用接收者操作特征曲线下面积（AUC）比较模型的区分度，并使用随机效应模型进行汇总：在初步确定和筛选的 4816 篇文章中，95 篇被纳入，代表了 380 万例入院患者。KDIGO-AKI标准是最常用的AKI定义标准（72%）。我们确定了 302 个模型，其中最常见的是逻辑回归（37%）、神经网络（10%）、随机森林（9%）和 XGBoost（9%）。报告最多的住院AKI事件预测因素是年龄、性别、糖尿病、血清肌酐和血红蛋白。内部和外部验证的 AKI 发病集合 AUC 分别为 0.82（95% CI，0.80-0.84）和 0.78（95% CI，0.76-0.80）。在多种临床环境、AKI 严重程度和 AKI 后并发症中，内部验证的汇总 AUC 为 0.78 至 0.87，外部验证为 0.73 至 0.84。虽然数据有限，但儿科人群的结果与在成人中观察到的结果一致。所有结果的研究间异质性都很高（I2>90%），根据预测模型偏倚风险评估工具（Prediction model Risk Of Bias ASsessment Tool），大多数研究的偏倚风险都很高（86%）：大多数外部验证模型在预测成人和儿童住院患者的 AKI 发病、AKI 严重程度和 AKI 后并发症方面表现良好。然而，临床环境、研究人群和预测因素的异质性限制了这些模型的通用性和在床旁的应用。Background: Artificial Intelligence (AI) through machine learning (ML) models appears to provide accurate and precise acute kidney injury (AKI) risk classification in some clinical settings, but their performance and implementation in real-world settings has not been established.
Methods: PubMed, EMBASE, Web of Science, and Scopus were searched until August 2023. Articles reporting on externally validated models for prediction of AKI onset, AKI severity, and post-AKI complications in hospitalized adult and pediatric patients were searched using text words related to AKI, AI, and ML. Two independent reviewers screened article titles, abstracts, and full texts. Areas under the receiver operating characteristic curves (AUCs) were used to compare model discrimination and pooled using a random-effects model.
Results: Of the 4816 articles initially identified and screened, 95 were included representing 3.8 million admissions. The KDIGO-AKI criteria were the most frequently used to define AKI (72%). We identified 302 models, with the most common being logistic regression (37%), neural networks (10%), random forest (9%), and XGBoost (9%). The most frequently reported predictors of hospitalized incident AKI were age, sex, diabetes, serum creatinine, and hemoglobin. The pooled AUCs for AKI onset were 0.82 (95% CI, 0.80-0.84) and 0.78 (95% CI, 0.76-0.80) for internal and external validation, respectively. Pooled AUCs across multiple clinical settings, AKI severities, and post-AKI complications ranged from 0.78 to 0.87 for internal validation and 0.73 to 0.84 for external validation. Although data were limited, results in the pediatric population aligned with those observed in adults. Between-study heterogeneity was high for all outcomes (I2 >90%), and most studies presented high-risk of bias (86%) according to the Prediction model Risk Of Bias ASsessment Tool.
Conclusions: Most externally validated models performed well in predicting AKI onset, AKI severity, and post-AKI complications in hospitalized adult and pediatric populations. However, heterogeneity in clinical settings, study populations, and predictors limits their generalizability and implementation at the bedside.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
