Scialla Julia J, Kao W H Linda, Crainiceanu Ciprian, Sozio Stephen M, Oberai Pooja C, Shafi Tariq, Coresh Josef, Powe Neil R, Plantinga Laura C, Jaar Bernard G, Parekh Rulan S
Department of Medicine, University of Miami, Miami, Florida;, †Department of Epidemiology,, §Department of Biostatistics, and, ‡Department of Medicine, Johns Hopkins University, Baltimore, Maryland;, ‖Department of Medicine, San Francisco General Hospital and University of California San Francisco, San Francisco, California;, ¶Department of Epidemiology, Emory University, Atlanta, Georgia;, *Nephrology Center of Maryland, Baltimore, Maryland, ††Hospital for Sick Children, University Health Network and University of Toronto, Toronto, Ontario, Canada.
Clin J Am Soc Nephrol. 2014 Apr;9(4):745-55. doi: 10.2215/CJN.05450513. Epub 2014 Jan 23.
Vascular calcification is common among patients undergoing dialysis and is associated with mortality. Factors such as osteoprotegerin (OPG), osteopontin (OPN), bone morphogenic protein-7 (BMP-7), and fetuin-A are involved in vascular calcification.
DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: OPG, OPN, BMP-7, and fetuin-A were measured in blood samples from 602 incident dialysis patients recruited from United States dialysis centers between 1995 and 1998 as part of the Choices for Healthy Outcomes In Caring for ESRD Study. Their association with all-cause and cardiovascular mortality were assessed using Cox proportional hazards models adjusted for demographic characteristics, comorbidity, serum phosphate, and calcium. An interaction with diabetes was tested because of its known association with vascular calcification. Predictive accuracy of selected biomarkers was explored by C-statistics in nested models with training and validation subcohorts.
Higher OPG and lower fetuin-A levels were associated with higher mortality over up to 13 years of follow-up (median, 3.4 years). The adjusted hazard ratios (HR) for highest versus lowest tertile were 1.49 (95% confidence interval [95% CI], 1.08 to 2.06) for OPG and 0.69 (95% CI, 0.52 to 0.92) for fetuin-A. In stratified models, the highest tertile of OPG was associated with higher mortality among patients without diabetes (HR, 2.42; 95% CI, 1.35 to 4.34), but not patients with diabetes (HR, 1.26; 95% CI, 0.82 to 1.93; P for interaction=0.001). In terms of cardiovascular mortality, higher fetuin-A was associated with lower risk (HR, 0.85 per 0.1 g/L: 95% CI, 0.75 to 0.96). In patients without diabetes, higher OPG was associated with greater risk (HR for highest versus lowest tertile, 2.91; 95% CI, 1.06 to 7.99), but not in patients with diabetes or overall. OPN and BMP-7 were not independently associated with outcomes overall. The addition of OPG and fetuin-A did not significantly improve predictive accuracy of mortality.
OPG and fetuin-A may be risk factors for all-cause and cardiovascular mortality in patients undergoing dialysis, but do not improve risk prediction.
血管钙化在接受透析的患者中很常见，且与死亡率相关。骨保护素（OPG）、骨桥蛋白（OPN）、骨形态发生蛋白-7（BMP-7）和胎球蛋白-A等因素参与血管钙化过程。
设计、地点、参与者及测量方法：作为“关爱终末期肾病患者健康结局选择研究”的一部分，对1995年至1998年期间从美国透析中心招募的602例新透析患者的血样进行OPG、OPN、BMP-7和胎球蛋白-A检测。使用经人口统计学特征、合并症、血清磷酸盐和钙调整的Cox比例风险模型评估它们与全因死亡率和心血管死亡率的关联。由于糖尿病与血管钙化已知的关联，对其相互作用进行了测试。通过在具有训练和验证亚队列的嵌套模型中使用C统计量来探索选定生物标志物的预测准确性。
在长达13年的随访（中位随访时间为3.4年）中，较高的OPG水平和较低的胎球蛋白-A水平与较高的死亡率相关。OPG最高三分位数与最低三分位数相比，调整后的风险比（HR）为1.49（95%置信区间[95%CI]，1.08至2.06），胎球蛋白-A为0.69（95%CI，0.52至0.92）。在分层模型中，OPG最高三分位数与无糖尿病患者的较高死亡率相关（HR，2.42；95%CI，1.35至4.34），但与糖尿病患者无关（HR，1.26；95%CI，0.82至1.93；交互作用P=0.001）。就心血管死亡率而言，较高的胎球蛋白-A与较低风险相关（每0.1 g/L的HR为0.85：95%CI，0.75至0.96）。在无糖尿病患者中，较高的OPG与较高风险相关（最高三分位数与最低三分位数的HR为2.91；95%CI，1.06至7.99），但在糖尿病患者或总体人群中并非如此。总体而言，OPN和BMP-7与结局无独立关联。加入OPG和胎球蛋白-A并未显著提高死亡率的预测准确性。
OPG和胎球蛋白-A可能是透析患者全因死亡率和心血管死亡率的危险因素，但不能改善风险预测。
