Calcium, phosphate, and parathyroid hormone levels in combination and as a function of dialysis duration predict mortality: Evidence for the complexity of the association between mineral metabolism and outcomes
Calcium, phosphate, and parathyroid hormone levels in combination and as a function of dialysis duration predict mortality: Evidence for the complexity of the association between mineral metabolism and outcomes
复制标题
DOI：
10.1097/01.asn.0000113243.24155.2f
复制
发表时间：
2004-03-01
影响因子：
13.6
通讯作者：
Levin, A
中科院分区：
文献类型：
--
作者：
Stevens, LA;Djurdjev, O;Levin, A
目前的文献表明透析人群中矿物质代谢异常（MM）与心血管疾病之间存在关联，但结果相互矛盾。 MM 生理机能很复杂；因此，假设反映这种复杂性的 MM 参数群可以预测死亡率，并且这种效应将通过透析持续时间 (DD) 进行修改。加拿大不列颠哥伦比亚省的流行透析患者在 2000 年 1 月至 3 月期间进行了钙 (Ca)、磷酸盐 (Pi) 和甲状旁腺激素 (iPTH) 的测量，并前瞻性随访至 2002 年 12 月。统计分析包括以 Ca、Pi 和 iPTH 单独和组合作为解释变量的 Cox 比例风险模型；分析按 DD 分层。本次分析中纳入的 515 名患者代表了不列颠哥伦比亚省和加拿大的透析人群：69% 接受血液透析，平均年龄为 60 +/- 17 岁，40% 为女性，34% 患有糖尿病。平均 Ca 和 Pi 值分别为 2.32 +/- 0.22 mmol/L 和 1.68 +/- 0.59 mmol/L，中位 iPTH 为 15.8 pmol/L（第 25 至 75 个百分位数：6.9 至 37.3 pmol/L）。调整人口统计学、透析类型和充分性、血红蛋白和白蛋白后，血清 Pi 可独立预测死亡率（风险比 [RR]，每 1 mmol/L 1.56；95% 置信区间 [CI]，1.15 至 2.12；P = 0.004）。当对参数组合进行建模时（总体 P = 0.003），高血清 Pi 和 Ca 与高 iPTH（RR，3.7 1；95% CI，1.53 至 9.03；P = 0.004）和低 iPTH（RR，4.30；95% CI，2.01 至 9.22；P < 0.001）的组合具有最高的死亡风险，如下所示：与高iPTH与正常血清Ca和Pi的组合相比，其死亡率最低，并被用作指标类别。这些影响因 DD 的不同阶层而异。该分析证明了检查 MM 参数组合（相对于单独的单个变量）的重要性以及 DD 的影响。这样做，时间和 MM 之间复杂的相互作用就可以开始被理解。还需要进一步探索。
Current literature suggests associations between abnormal mineral metabolism (MM) to cardiovascular disease in dialysis populations, with conflicting results. MM physiology is complex; therefore, it was hypothesized that constellations of MM parameters, reflecting this complexity, would be predictive of mortality and that this effect would be modified by dialysis duration (DD). Prevalent dialysis patients in British Columbia, Canada, who had measurements of calcium (Ca), phosphate (Pi), and parathyroid hormone (iPTH) between January and March 2000 were followed prospectively until December 2002. Statistical analysis included Cox proportional hazard models with Ca, Pi, and iPTH alone and in combination as explanatory variables; analyses were stratified by DD. The 515 patients included in this analysis represent British Columbia and Canadian dialysis populations: 69% were on hemodialysis, mean age was 60 +/- 17 yr, 40% were female, and 34% had diabetes. Mean Ca and Pi values were 2.32 +/- 0.22 mmol/L and 1.68 +/- 0.59 mmol/L, respectively, and median iPTH was 15.8 pmol/L (25th to 75th percentile: 6.9 to 37.3 pmol/L). Serum Pi, after adjusting for demographic, dialysis type and adequacy, hemoglobin, and albumin, independently predicted mortality (risk ratio [RR], 1.56 per 1 mmol/L; 95% confidence interval [CI], 1.15 to 2.12; P = 0.004). When combinations of parameters were modeled (overall P = 0.003), the combinations of high serum Pi and Ca with high iPTH (RR, 3.7 1; 95% CI, 1.53 to 9.03; P = 0.004) and low iPTH (RR, 4.30; 95% CI, 2.01 to 9.22; P < 0.001) had highest risks for mortality as compared with the combination of high iPTH with normal serum Ca and Pi that had the lowest mortality and was used as index category. These effects varied across different strata of DD. This analysis demonstrates the importance of examining combinations of MM parameters as opposed to single variables alone and the effect of DD. In so doing, the complex interaction of time and MM can begin to be understand. Further exploration is required.