目的 探讨维持性血液透析(maintenance hemodialysis，MHD)患者血清胎球蛋白-A(fetuin-A)及血清钙化倾向(以钙蛋白颗粒半数转变时间T50表示)与新发心血管疾病(cardiovascular disease，CVD)事件及事件后死亡风险的关联，明确二者在CVD不同病程阶段的预测价值。 方法  选取2020年1月—2024年12月在沧州市中心医院接受治疗的MHD患者为研究对象，检测基线血清Fetuin-A及T50水平，随访自基线检测日起至2024年12月31日。Fetuin-A按三分位数分为低Fetuin-A组、中Fetuin-A组和高Fetuin-A组，T50按三分位数分为短T50组、中T50组和长T50组。观察终点为新发CVD事件及CVD事件后全因死亡。采用Kruskal-Wallis H检验或χ²检验比较基线特征；采用Gray检验及Fine-Gray竞争风险回归模型分析Fetuin-A和T50与新发CVD事件的关联；采用Kaplan-Meier法、Log-rank检验及多因素COX比例风险回归模型分析二者与CVD事件后死亡风险的关系。 结果  本研究共纳入513例患者。中位随访30.0(18.0，44.0)个月。按Fetuin-A三分位分组，3组年龄(H=42.136，P＜0.001)、透析龄(H=22.584，P＜0.001)、糖尿病肾病(χ²=11.256，P=0.004)、既往CVD病史(χ²=10.674，P=0.005)、血脂异常(χ²=6.124，P=0.047)、血钙(H=9.210，P=0.010)、红细胞生成刺激剂(erythropoiesis-stimulating agent，ESA)剂量(H=8.270，P=0.016)、静脉铁使用(χ²=7.501，P=0.024)、体质量指数(body mass index，BMI)(H=18.907，P＜0.001)、白蛋白(H=32.514，P＜0.001)及C反应蛋白(C-reactive protein，CRP)(H=48.762，P＜0.001)比较差异有统计学意义。按T50三分位分组，3组年龄(H=8.149，P=0.017)、既往CVD病史(χ²=7.755，P=0.021)、高血压(χ²=6.873，P=0.032)、血钙(H=8.846，P=0.012)、血磷(H=20.436，P＜0.001)、全段甲状旁腺激素(intact parathyroid hormone，iPTH)(H=6.648，P=0.036)、ESA剂量(H=7.726，P=0.021)、白蛋白(H=13.816，P=0.001)及CRP(H=32.887，P＜0.001)比较差异有统计学意义。Gray检验显示Fetuin-A三分位组间新发CVD事件累积发生率比较差异有统计学意义(χ²=24.613，P＜0.001)，随Fetuin-A水平降低而升高；T50三分位组间新发CVD事件累积发生率比较差异有统计学意义(χ²=8.743，P=0.013)，随T50缩短而升高。与高Fetuin-A组相比，低Fetuin-A组和中Fetuin-A组新发CVD事件风险升高(低Fetuin-A组：sdHR=1.650，95% CI：1.090～2.510，P=0.019；中Fetuin-A组：sdHR=1.820，95% CI：1.230～2.690，P=0.003)。以长T50组为参照，在调整混杂因素及Fetuin-A后，短T50组和中T50组与新发CVD事件风险均无显著关联(短T50组：sdHR=1.210，95% CI：0.770～1.890，P=0.405；中T50组：sdHR=1.290， 95% CI：0.880～1.890，P=0.192)。Log-rank检验显示低/中Fetuin-A组与高Fetuin-A组CVD事件后死亡累积发生率比较差异无统计学意义(χ²=1.892，P=0.169)；短/中T50组高于长T50组(χ²=7.862，P=0.005)。多因素COX回归分析显示短/中T50组CVD事件后死亡风险高于长T50组(HR=3.240，95% CI：1.370～7.650，P=0.009)；低/中Fetuin-A组与高Fetuin-A组比较差异无统计学意义(HR=0.900，95% CI：0.380～2.160，P=0.810)。 结论  低Fetuin-A主要与随访期间新发CVD事件风险升高相关，短T50主要与CVD事件后死亡风险升高相关，联合评估二者有助于MHD患者CVD分阶段风险分层。

Objective To investigate the association of serum fetuin-A (Fetuin-A) and serum calcification propensity (quantified by calciprotein particle semi-maturation time, T50) with incident cardiovascular disease (CVD) events and post-event mortality in maintenance hemodialysis (MHD) patients, and to clarify the predictive value of Fetuin-A and T50 at different stages of the CVD course.  Method A total of 513 MHD patients treated at Cangzhou Central Hospital from January 2020 to December 2024 were included and followed up from baseline measurement to December 31, 2024. Baseline serum Fetuin-A and T50 levels were measured. According to tertiles, Fetuin-A was categorized into low, intermediate, and high Fetuin-A groups, while T50 was categorized into short, intermediate, and long T50 groups. The endpoints were incident CVD events and all-cause mortality after CVD events. Baseline characteristics were compared using the Kruskal-Wallis H test or χ² test. Gray’s test and Fine-Gray competing risk regression were used to analyze the association of Fetuin-A and T50 with incident CVD events. Kaplan-Meier analysis, Log-rank test, and multivariate Cox proportional hazards regression were used to evaluate the association of Fetuin-A and T50 with mortality risk of post-CVD event.  Results The median follow-up duration was 30.0 months (P25-P75: 18.0-44.0 months). Based on Fetuin-A tertiles, significant differences among the 3 groups were observed in age (H=42.136, P＜0.001), dialysis vintage (H=22.584, P＜0.001), diabetic kidney disease (χ²=11.256, P=0.004), previous CVD history (χ²=10.674, P=0.005), dyslipidemia (χ²=6.124, P=0.047), serum calcium (H=9.210, P=0.010), erythropoiesis-stimulating agent dose (H=8.270, P=0.016), intravenous iron use (χ²=7.501, P=0.024), body mass index (H=18.907, P＜0.001), albumin (H=32.514, P＜0.001), and C-reactive protein (H=48.762, P＜0.001). Based on T50 tertiles, significant differences among the 3 groups were observed in age (H=8.149, P=0.017), previous CVD history (χ²=7.755, P=0.021), hypertension (χ²=6.873, P=0.032), serum calcium (H=8.846, P=0.012), serum phosphorus (H=20.436, P＜0.001), intact parathyroid hormone (H=6.648, P=0.036), erythropoiesis-stimulating agent dose (H=7.726, P=0.021), albumin (H=13.816, P=0.001), and C-reactive protein (H=32.887, P＜0.001). Gray’s test showed that the cumulative incidence of incident CVD events differed significantly among the 3 groups of Fetuin-A tertiles (χ²=24.613, P＜0.001) and increased with decreasing Fetuin-A levels; it also differed significantly among the 3 groups of T50 tertiles (χ²=8.743, P=0.013) and increased with shorter T50. Fine-Gray competing risk regression showed that compared with the high Fetuin-A group, the low and intermediate Fetuin-A groups had higher risks of incident CVD events (low Fetuin-A group: sdHR=1.650, 95% CI: 1.090～2.510, P=0.019; intermediate Fetuin-A group: sdHR=1.820, 95% CI: 1.230～2.690, P=0.003),  whereas with the long T50 group as the reference, the short and intermediate T50 groups were not significantly associated with incident CVD events after adjustment for confounders and Fetuin-A (short T50 group: sdHR=1.210, 95% CI: 0.770～1.890, P=0.405; intermediate T50 group: sdHR=1.290, 95% CI: 0.880～1.890, P=0.192). Log-rank test showed no significant difference in cumulative mortality of post-CVD event between the combined low/intermediate Fetuin-A group and the high Fetuin-A group (χ²=1.892, P=0.169), whereas the combined short/intermediate T50 group had a higher cumulative mortality than the long T50 group (χ²=7.862, P=0.005). Multivariate Cox proportional hazards regression showed that the combined short/intermediate T50 group had a higher mortality risk of post-CVD event than the long T50 group (HR=3.240, 95% CI: 1.370～7.650, P=0.009), while no significant difference was observed between the combined low/intermediate Fetuin-A group and the high Fetuin-A group (HR=0.900, 95% CI: 0.380～2.160, P=0.810).  Conclusion Low Fetuin-A was mainly associated with an increased risk of incident CVD events during follow-up, whereas short T50 was mainly associated with an increased mortality risk of post-CVD event. Combined assessment of these two biomarkers may facilitate stage-specific stratification of CVD risk in MHD patients.