Fresenius Medical Care, Global Medical Office, 920 Winter Street, Waltham, MA, 02451, USA.
Fresenius Medical Care, Global Medical Office, Bad Homburg, Germany.
BMC Nephrol. 2024 Oct 19;25(1):366. doi: 10.1186/s12882-024-03809-2.
Gastrointestinal bleeding (GIB) is a clinical challenge in kidney failure. INSPIRE group assessed if machine learning could determine a hemodialysis (HD) patient's 180-day GIB hospitalization risk.
An eXtreme Gradient Boosting (XGBoost) and logistic regression model were developed using an HD dataset in United States (2017-2020). Patient data was randomly split (50% training, 30% validation, and 20% testing). HD treatments ≤ 180 days before GIB hospitalization were classified as positive observations; others were negative. Models considered 1,303 exposures/covariates. Performance was measured using unseen testing data.
Incidence of 180-day GIB hospitalization was 1.18% in HD population (n = 451,579), and 1.12% in testing dataset (n = 38,853). XGBoost showed area under the receiver operating curve (AUROC) = 0.74 (95% confidence interval (CI) 0.72, 0.76) versus logistic regression showed AUROC = 0.68 (95% CI 0.66, 0.71). Sensitivity and specificity were 65.3% (60.9, 69.7) and 68.0% (67.6, 68.5) for XGBoost versus 68.9% (64.7, 73.0) and 57.0% (56.5, 57.5) for logistic regression, respectively. Associations in exposures were consistent for many factors. Both models showed GIB hospitalization risk was associated with older age, disturbances in anemia/iron indices, recent all-cause hospitalizations, and bone mineral metabolism markers. XGBoost showed high importance on outcome prediction for serum 25 hydroxy (25OH) vitamin D levels, while logistic regression showed high importance for parathyroid hormone (PTH) levels.
Machine learning can be considered for early detection of GIB event risk in HD. XGBoost outperforms logistic regression, yet both appear suitable. External and prospective validation of these models is needed. Association between bone mineral metabolism markers and GIB events was unexpected and warrants investigation.
This retrospective analysis of real-world data was not a prospective clinical trial and registration is not applicable.
胃肠道出血（GIB）是肾衰竭患者面临的一项临床挑战。INSPIRE 研究组评估了机器学习是否可以确定血液透析（HD）患者 180 天内 GIB 住院风险。
使用美国 2017-2020 年 HD 数据集，建立极端梯度提升（XGBoost）和逻辑回归模型。患者数据随机分为（50%训练、30%验证和 20%测试）。在 GIB 住院前 HD 治疗<180 天的病例被归类为阳性观察结果，其余为阴性。模型共考虑了 1303 个暴露/协变量。使用未见测试数据来衡量模型性能。
HD 人群中 180 天 GIB 住院率为 1.18%（n=451579），测试数据集为 1.12%（n=38853）。XGBoost 的曲线下接收者操作特征（AUROC）为 0.74（95%置信区间（CI）0.72，0.76），而逻辑回归的 AUROC 为 0.68（95%CI 0.66，0.71）。XGBoost 的灵敏度和特异度分别为 65.3%（60.9，69.7）和 68.0%（67.6，68.5），逻辑回归的灵敏度和特异度分别为 68.9%（64.7，73.0）和 57.0%（56.5，57.5）。对于许多因素，暴露的关联在两个模型中都是一致的。两个模型均显示 GIB 住院风险与年龄较大、贫血/铁指标紊乱、近期全因住院和骨代谢标志物有关。XGBoost 显示血清 25 羟维生素 D 水平对结果预测的重要性较高，而逻辑回归显示甲状旁腺激素（PTH）水平的重要性较高。
机器学习可用于早期检测 HD 患者 GIB 事件风险。XGBoost 优于逻辑回归，但两者似乎都适用。需要对这些模型进行外部和前瞻性验证。骨代谢标志物与 GIB 事件之间的关联出乎意料，值得进一步研究。
