AGE-RELATED DIFFERENCES IN IMMUNO-HEMATOLOGIC PROFILES AND THEIR ASSOCIATION WITH ALL-CAUSE MORTALITY
AGE-RELATED DIFFERENCES IN IMMUNO-HEMATOLOGIC PROFILES AND THEIR ASSOCIATION WITH ALL-CAUSE MORTALITY
复制标题
DOI：
10.1093/geroni/igz038.388
复制
发表时间：
2019-11-08
影响因子：
7
通讯作者：
Gunzler DD
中科院分区：
文献类型：
--
作者：
Dalton JE;Zidar DA;Krieger NI;Perzynski AT;Gunzler DD
免疫血液功能（IHF）越来越被认为是老年人健康状况的核心组成部分。在这项研究中，我们试图确定 IHF 与全因死亡率关系的同质特征。然后，我们研究了这些资料在 65 岁以上个体中的分布。我们使用了 30,828 名 NHANES 参与者的数据，包括 10 个具有差异成分的基线全血细胞计数 [例如淋巴细胞、白细胞、红细胞分布宽度 (RDW)] 和全因死亡率。我们使用潜在特征分析 (LPA) 同时优化 CBC 成分的簇内同质性和簇间生存差异。 LPA（使用 MPlus 8.2）允许根据拟合优度标准对不同解决方案进行实证比较。 LPA 模型收敛后，9 类解决方案平衡了拟合优度标准和结果类的可解释性。最大的3类占样本的83.7%，其中1、2和3类分别占32.1%、28.6%和23.6%。与 1 类和 3 类相比，2 类的淋巴细胞、单核细胞、中性粒细胞和血小板较低。1 类和 2 类（Cox 模型风险比，HR=0.85；P=0.012）和 2 类与 3 类（HR=1.18；P=0.001）之间的生存率不同。其余 6 个类别通常具有较高 RDW 和较低血红蛋白的共同特征，但也存在显着的生存差异。多项逻辑回归显示，在 7,173 名 65 岁以上参与者的子集中，相对于第 2 类 (P<0.001) 和第 3 类 (P<0.001)，年龄较大与第 1 类成员资格显着相关。这些结果表明有可能开发出指示加速衰老的免疫标记谱。
Immuno-hematologic function (IHF) is increasingly being recognized as a central component of health status in older age. In this study, we sought to identify homogeneous IHF profiles regarding their relationship to all-cause mortality. We then studied the distribution of these profiles among individuals over age 65. We used data on 30,828 NHANES participants, including 10 baseline complete blood count with differential components [e.g., lymphocytes, leukocytes, red cell distribution width (RDW)] and all-cause mortality. We used latent profile analysis (LPA) to simultaneously optimize intra-cluster homogeneity on CBC components and inter-cluster survival differences. LPA (using MPlus 8.2) allowed for the empirical comparison of different solutions based on goodness-of-fit criteria. After LPA model convergence, a 9-class solution balanced goodness-of-fit criteria and interpretability of the resulting classes. The largest 3 classes accounted for 83.7% of the sample, with classes 1, 2 and 3 comprising 32.1%, 28.6% and 23.6%. Class 2 had lower lymphocytes, monocytes, neutrophils and platelets relative to classes 1 and 3. Survival rates were different between classes 1 and 2 (Cox model hazard ratio, HR=0.85; P=0.012) and 2 vs 3 (HR=1.18; P=0.001). The remaining 6 classes, which generally shared in common characteristics of higher RDW and lower hemoglobin, also were involved with significant survival differences. Multinomial logistic regression revealed that, among the subset of 7,173 participants over 65, older age was significantly associated with membership in class 1 relative to classes 2 (P<0.001) and 3 (P<0.001). These results point toward the possibility of developing immune marker profile indicative of accelerated aging.