目的  探讨维持性血液透析(maintenance hemodialysis，MHD)患者血清脂质运载蛋白2(lipocalin-2，LCN2)、趋化因子配体3(C-C chemokine ligand 3，CCL3)、C1q肿瘤坏死因子相关蛋白-3(C1q tumor necrosis factor-related protein-3，CTRP3)水平与血管钙化(vascular calcification，VC)的关系。 方法 选择衡水市第二人民医院2023年4月—2024年9月行MHD的患者155例作为MHD组，67例作为验证组，另纳入同期体检健康志愿者160例作为对照组。酶联免疫吸附试验(enzyme-linked immunosorbent assay，ELISA)法检测血清LCN2、CCL3、CTRP3水平；根据患者末次随诊VC情况分为VC组和非VC组；Pearson法分析相关性；Logistic回归分析影响VC发生的因素。采用相对危险度分析血磷、血钙、甲状旁腺激素(parathyroid hormone，PTH)、LCN2、CCL3、CTRP3水平对VC的影响；采用海斯过程分析程序分析血清指标在VC发生中的中介作用；受试者工作特征曲线(receiver operating characteristic curve，ROC)分析血清指标对VC的预测价值。 结果  本研究MHD组共纳入155例患者(其中VC组54例，非VC组101例)，对照组共纳入160例健康志愿者。MHD组较对照组血清LCN2、CCL3水平升高，CTRP3水平降低(t=26.402、50.873、27.085，均P＜0.001)；VC组较非VC组血磷、血钙、PTH、LCN2、CCL3水平升高，CTRP3水平降低(t=4.132、6.041、20.249、7.172、7.553、6.533，均P＜0.001)；MHD患者血磷、血钙、PTH水平与血清LCN2(r=0.399、0.410、0.404，均P＜0.001)、CCL3(r=0.376、0.382、0.413，均P＜0.001)呈正相关，与血清CTRP3(r=-0.366、-0.370、-0.358，均P＜0.001)呈负相关；血磷(HR=2.110，95% CI：1.158～3.844，P=0.015)、血钙(HR=1.677，95%CI：1.267～2.496，P=0.011)、PTH(HR=2.263，95%CI:1.197～4.279，P=0.012)、血清LCN2(HR=1.785，95%CI：1.106～2.880，P=0.018)、CCL3(HR=1.562，95% CI：1.019～2.395，P=0.041)、CTRP3(HR=0.630，95% CI：0.459～0.865，P=0.004)是VC发生的影响因素；血磷、血钙、PTH、LCN2、CCL3、CTRP3高表达患者发生VC的风险分别是低表达患者的2.227倍(95% CI：1.361～3.641，P＜0.001)、2.040倍(95% CI：1.262～3.298，P=0.002)、2.079倍(95% CI：1.299～3.327，        P=0.001)、1.772倍(95% CI：1.117～2.811，P=0.012)、4.800倍(95% CI：2.774～8.305，P=0.002)、0.581倍(95% CI：0.369～0.914，P=0.016)；LCN2、CCL3、CTRP3在MHD患者VC发生之间具有中介调控作用(均P＜0.001)；血清LCN2、CCL3、CTRP3水平单独及联合预测MHD患者发生VC的曲线下面积(area under the curve，AUC)分别为0.794、0.831、0.849，联合预测效能最优(Z=4.138、3.856、3.523，均P＜0.001)。 结论 MHD患者血清LCN2、CCL3水平升高，CTRP3水平降低，三者联合检测可以较好地预测MHD患者VC的发生。

Objective  To explore the relationship between serum lipocalin-2 (LCN2), C-C chemokine ligand 3 (CCL3), and C1q tumor necrosis factor-related protein-3 (CTRP3) and vascular calcification (VC) in maintenance hemodialysis (MHD) patients . Methods  Patients who underwent MHD at Hengshui Second People's Hospital from April 2023 to September 2024 were selected as the MHD group. 67 patients who underwent MHD were included as the validation group, and healthy volunteers who underwent physical examinations during the same period were included as the control group. The levels of serum LCN2, CCL3 and CTRP3 were measured by enzyme-linked immunosorbent assay (ELISA). According to the VC status at the last follow-up, patients were divided into the VC group (n=54) and the non-VC group (n=101). Pearson correlation analysis was used to assess correlations; logistic regression was employed to analyze factors influencing the occurrence of VC. The effects of serum phosphorus, calcium, PTH, LCN2, CCL3, and CTRP3 levels on VC were analyzed using relative risk analysis. In addition, the Hayes process analysis program was used to analyze the mediating role of serum indicators in the occurrence of VC. Receiver operating characteristic curve (ROC) curves were used to explore the predictive value of serum LCN2, CCL3 and CTRP3 levels for VC in MHD patients. Results  In this study, 155 patients were included in the MHD group (54 with VC and 101 without VC), and 160 healthy volunteers were included in the control group. Compared with the control group, serum LCN2 (t=26.402) and CCL3 (t=50.873) levels were significantly higher, while CTRP3 level (t=27.085) was lower in the MHD group (all P<0.001). Compared with the non-VC group, serum phosphorus (t=4.132), calcium (t=6.041), parathyroid hormone (PTH, t=20.249) and LCN2 (t=7.172), CCL3 (t=7.553) levels were significantly higher, while CTRP3 level (t=6.533) was significantly lower in the VC group (all P<0.001). Serum phosphorus, calcium, PTH levels were positively correlated with serum LCN2 and CCL3 levels, and negatively correlated with CTRP3 levels (all P<0.001). Multivariate COX regression analysis showed that phosphorus (HR=2.110, 95% CI: 1.158～3.844, P=0.015), calcium (HR=1.677, 95% CI: 1.267～2.496, P=0.011), PTH (HR=2.263, 95% CI 1.197～4.279, P=0.012), LCN2(HR=1.785, 95% CI: 1.106～2.880,P=0.018), and CCL3 (HR=1.562, 95% CI: 1.019～2.395, P=0.041) were independent risk factors for VC, while CTRP3 (HR=0.630, 95% CI: 0.459～0.865, P=0.004) were a protective factor for VC. The risks of developing VC in patients with high expression of phosphorus, calcium, PTH, LCN2, CCL3, and CTRP3 were 2.227 times (95% CI: 1.361～3.641, P<0.001), 2.040 times (95% CI: 1.262～3.298, P=0.002), 2.079 times (95% CI: 1.299～3.327, P=0.001), 1.772 times (95% CI: 1.117～2.811, P=0.012), 4.800 times (95% CI: 2.774～8.305, P=0.002), and 0.581 times (95% CI: 0.369～0.914, P=0.016), respectively, compared with patients with low expression of these biomarkers. The area under the curve (AUC) for predicting VC by serum LCN2, CCL3, and CTRP3 levels were 0.794, 0.831, and 0.849, respectively, and the AUC for the combination of the three was 0.942, with better predictive performance than each individual indicator (Z=4.138, 3.856, 3.523, all P<0.001).  Conclusion   Serum LCN2 and CCL3 levels are elevated, while CTRP3 level is decreased in MHD patients. Combined detection of the three biomarkers may effectively predict the occurrence of VC in MHD patients.