Department of Medicine and Bioregulatory Science, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan.
Medical Information Center, Kyushu University Hospital, Fukuoka, Japan.
PLoS One. 2022 Jul 12;17(7):e0271179. doi: 10.1371/journal.pone.0271179. eCollection 2022.
Previous reports have demonstrated the association of serum bilirubin levels with the progression of diabetic nephropathy. The objective of this study is to assess the association of basal bilirubin levels with progressive renal decline (PRD) and end-stage kidney disease (ESKD).
A total of 298 patients with diabetes who visited Kyushu University Hospital (Japan) were recruited and followed up for 10 years. PRD was defined as a negative change in estimated glomerular filtration ratio (eGFR) >3.7%/year, 2.5th percentile. Logistic regression analysis was performed to evaluate the association of total bilirubin levels with PRD and its cut-off point was determined by receiver operating characteristic (ROC) analysis. Kaplan-Meier method and Cox hazard regression analysis were used to evaluate the predictive ability of its cut-off point for ESKD.
Logistic regression model showed that total bilirubin levels were significantly associated with PRD, and ROC analysis showed that its cut-off point was 0.5 mg/dL. Kaplan-Meier method showed that the percent of patients who reached two endpoints, composite endpoint (ESKD or doubling of creatinine level) or 30% eGFR decline, was significantly higher in the low bilirubin group than in the high bilirubin group (18.5% vs 11.0%, P = 0.045; 49.1% vs 42.1%, P = 0.045, respectively, log-rank test). Cox hazard regression models confirmed the independence of the predictive ability of its cut-off point.
Serum total bilirubin levels were negatively associated with PRD in diabetic nephropathy and its cut-off point was 0.5 mg/dL. It may be clinically useful for identifying patients at high risk of ESKD.
先前的报告表明血清胆红素水平与糖尿病肾病的进展有关。本研究旨在评估基础胆红素水平与进行性肾下降（PRD）和终末期肾病（ESKD）的相关性。
共招募了 298 名在日本九州大学医院就诊的糖尿病患者，随访 10 年。PRD 定义为估算肾小球滤过率（eGFR）的负变化>3.7%/年，第 2.5 百分位。使用逻辑回归分析评估总胆红素水平与 PRD 的相关性，并通过接收者操作特征（ROC）分析确定其截断点。使用 Kaplan-Meier 方法和 Cox 风险回归分析评估其截断点对 ESKD 的预测能力。
逻辑回归模型表明，总胆红素水平与 PRD 显著相关，ROC 分析表明其截断点为 0.5mg/dL。Kaplan-Meier 方法表明，在低胆红素组中达到两个终点（ESKD 或肌酐水平翻倍）或 eGFR 下降 30%的患者比例明显高于高胆红素组（18.5%比 11.0%，P=0.045；49.1%比 42.1%，P=0.045，分别为对数秩检验）。Cox 风险回归模型证实了其截断点的预测能力的独立性。
血清总胆红素水平与糖尿病肾病的 PRD 呈负相关，其截断点为 0.5mg/dL。它可能对识别 ESKD 高危患者具有临床意义。
