2026 年第 8 期 第 21 卷

中性粒细胞百分比与血红蛋白比值联合血钾变异系数和N末端B型脑钠肽前体对慢性肾脏病患者发生急性心力衰竭的预测价值

Predictive value of neutrophil percentage to hemoglobin ratio combined with serum potassium coefficient of variation and N-terminal pro-brain natriuretic peptide for acute heart failure in patients with chronic kidney disease

作者：王春燕任威薛小磊王欢张泽

英文作者：Wang Chunyan Ren Wei Xue Xiaolei Wang Huan Zhang Ze

单位：中国人民解放军联勤保障部队第九八四医院心肾内科，北京100094

英文单位：Department of Cardiorenal Medicine the 984th Hospital of Joint Logistics Support Force of the Chinese People′s Liberation Army Beijing 100094 China

关键词：慢性肾脏病；急性心力衰竭；中性粒细胞百分比与血红蛋白比值；血钾变异系数；N末端B型脑钠肽前体

英文关键词：Chronickidneydisease;Acuteheartfailure;Neutrophilpercentagetohemoglobinratio; Serumpotassiumcoefficientofvariation;N-terminalpro-brainnatriureticpeptide

• 摘要：

• 目的 探讨中性粒细胞百分比与血红蛋白比值（NPHR）联合血钾变异系数和N末端B型脑钠肽前体（NT-proBNP）对慢性肾脏病（CKD）患者发生急性心力衰竭（AHF）的预测价值。方法 采用回顾性队列研究设计，收集2023年1月至2024年10月在中国人民解放军联勤保障部队第九八四医院心肾内科住院治疗的186例CKD患者的临床资料，根据随访期间是否发生AHF分为AHF组（42例）和非AHF组（144例）。收集患者基线临床资料、实验室检查指标，计算NPHR（经自然对数转换后的lnNPHR用于分析）和血钾变异系数（排除干扰因素后的稳定期变异）。采用单因素分析比较2组临床指标差异，多因素Logistic回归方法分析筛选CKD患者发生AHF的独立危险因素，绘制受试者工作特征（ROC）曲线评估各指标及联合模型的预测效能。同时分析3个指标间的相关性。结果 单因素分析显示，AHF组患者年龄、高血压病分级≥3级比例均高于非AHF组（均P＜0.05）。AHF组患者血肌酐、血尿素氮、血尿酸、中性粒细胞百分比、lnNPHR、血钾变异系数、NT-proBNP水平均高于非AHF组，估算肾小球滤过率、血红蛋白水平低于非AHF组（均P＜0.05）。相关性分析显示，NPHR与血钾变异系数呈正相关（r=0.412，P＜0.01），NPHR与NT-proBNP呈正相关（r=0.385，P＜0.01），血钾变异系数与NT-proBNP呈正相关（r=0.358，P＜0.01）。多因素Logistic回归分析显示，lnNPHR、血钾变异系数、NT-proBNP是CKD患者发生AHF的独立危险因素（均P＜0.001）。ROC曲线分析显示，lnNPHR、血钾变异系数、NT-proBNP单独预测CKD患者发生AHF的AUC分别为0.734（95%置信区间：0.592～0.876）、0.715（95%置信区间：0.592～0.837）、0.824（95%置信区间：0.704～0.944），三者联合预测的AUC为0.884（95%置信区间：0.794～0.973），显著高于各指标单独预测（均P＜0.05）。结论 lnNPHR、血钾变异系数、NT-proBNP均为CKD患者发生AHF的独立危险因素，三者联合构建的预测模型具有更高的预测效能，可为临床早期识别CKD患者AHF发生风险提供可靠参考。但本研究为单中心回顾性分析，样本量较小（尤其AHF组仅42例），结论需大样本前瞻性研究验证。

• Objective To explore the predictive value of neutrophil percentage to hemoglobin ratio (NPHR) combined with serum potassium coefficient of variation and N-terminal pro-brain natriuretic peptide (NT-proBNP) for acute heart failure (AHF) in patients with chronic kidney disease (CKD). Methods A retrospective cohort study was conducted. Clinical data of 186 CKD patients hospitalized in the Department of Cardiorenal Medicine, the 984th Hospital of Joint Logistics Support Force of the Chinese People′s Liberation Army from January 2023 to October 2024 were collected. Patients were divided into AHF group (42 cases) and non-AHF group (144 cases) according to whether AHF occurred during follow-up. Baseline clinical data and laboratory indicators were collected, and NPHR (naturally log-transformed lnNPHR was used for analysis) and serum potassium coefficient of variation (stable-phase variation after excluding confounding factors) were calculated. Univariate analysis was used to compare differences in clinical indicators between the two groups. Multivariate Logistic regression analysis was performed to screen independent risk factors for AHF in CKD patients. Receiver operating characteristic (ROC) curves were drawn to evaluate the predictive efficiency of single indicators and the combined model. Meanwhile, the correlations among the three indicators were analyzed. Results Univariate analysis showed that age and the proportion of hypertension grade ≥3 in the AHF group were higher than those in the non-AHF group (both P＜0.05). The levels of serum creatinine, blood urea nitrogen, serum uric acid, neutrophil percentage, lnNPHR, serum potassium coefficient of variation and NT-proBNP in the AHF group were higher than those in the non-AHF group, while estimated glomerular filtration rate and hemoglobin level were lower in the AHF group (all P＜0.05). Correlation analysis indicated that NPHR was positively correlated with serum potassium coefficient of variation (r=0.412, P＜0.01), NPHR was positively correlated with NT-proBNP (r=0.385, P＜0.01), and serum potassium coefficient of variation was positively correlated with NT-proBNP (r=0.358, P＜0.01). Multivariate Logistic regression analysis revealed that lnNPHR, serum potassium coefficient of variation and NT-proBNP were independent risk factors for AHF in CKD patients (all P＜0.001). ROC curve analysis demonstrated that the AUC values of lnNPHR, serum potassium coefficient of variation and NT-proBNP alone for predicting AHF in CKD patients were 0.734 ［95% confidence interval(CI): 0.592-0.876］, 0.715 (95%CI: 0.592-0.837) and 0.824 (95%CI: 0.704-0.944), respectively. The AUC of the combined prediction model of the three indicators was 0.884 (95%CI: 0.794-0.973), which was significantly higher than that of each single indicator (all P＜0.05). Conclusions lnNPHR, serum potassium coefficient of variation and NT-proBNP are all independent risk factors for AHF in CKD patients. The prediction model constructed by combining the three indicators presents superior predictive efficiency, which can provide reliable reference for early clinical identification of AHF risk in CKD patients. Nevertheless, this study is a single-center retrospective analysis with a small sample size (only 42 cases in the AHF group), and the conclusions need to be verified by large-sample prospective studies.