Watts Amber, Honea Robyn A, Billinger Sandra A, Rhyner Kathleen T, Hutfles Lewis, Vidoni Eric D, Burns Jeffrey M
Department of Clinical Psychology, University of Kansas, Lawrence, KS, USA.
Alzheimer's Disease Center, University of Kansas, Fairway, KS, USA.
J Alzheimers Dis. 2015;45(1):187-93. doi: 10.3233/JAD-142085.
Though hypertension is a commonly studied risk factor for white matter lesions (WMLs), measures of blood pressure may fluctuate depending on external conditions resulting in measurement error. Indicators of arterial stiffening and reduced elasticity may be more sensitive indicators of risk for WMLs in aging; however the interdependent nature of vascular indicators creates statistical complications.
The purpose of the study was to determine whether a factor score comprised of multiple vascular indicators would be a stronger predictor of WMLs than traditional measures of blood pressure.
In a sample of well-characterized nondemented older adults, we used a factor analytic approach to account for variance common across multiple vascular measures while reducing measurement error. The result was a single factor score reflecting arterial stiffness and reduced elasticity. We used this factor score to predict white matter lesion volumes acquired via fluid attenuated inversion recovery (FLAIR) magnetic resonance imaging.
The combined vascular factor score was a stronger predictor of deep WML (β = 0.42, p < 0.001) and periventricular WML volumes (β = 0.49, p < 0.001). After accounting for the vascular factor, systolic and diastolic blood pressure measurements were not significant predictors.
This suggests that a combined measure of arterial elasticity and stiffening may be a stronger predictor of WMLs than systolic and diastolic blood pressure accounting for the multicollinearity associated with a variety of interrelated vascular measures.
尽管高血压是白质病变(WMLs)常见的研究风险因素,但血压测量值可能会因外部条件而波动,从而导致测量误差。动脉僵硬度增加和弹性降低的指标可能是衰老过程中WMLs风险更敏感的指标;然而,血管指标之间的相互依存性质会产生统计上的复杂性。
本研究的目的是确定由多个血管指标组成的因子得分是否比传统的血压测量方法更能预测WMLs。
在一组特征明确的非痴呆老年人样本中,我们采用因子分析方法来解释多个血管测量指标共有的方差,同时减少测量误差。结果得到一个反映动脉僵硬度增加和弹性降低的单一因子得分。我们使用这个因子得分来预测通过液体衰减反转恢复(FLAIR)磁共振成像获得的白质病变体积。
综合血管因子得分是深部WML(β = 0.42,p < 0.001)和脑室周围WML体积(β = 0.49,p < 0.001)更强的预测指标。在考虑血管因子后,收缩压和舒张压测量值不是显著的预测指标。
这表明,综合动脉弹性和僵硬度的测量指标可能比收缩压和舒张压更能预测WMLs,同时考虑到与各种相互关联的血管测量指标相关的多重共线性。