背景: 腦血流是否會隨著生理老化而產生變化，老化所引起的腦血流變化與其他病理性腦血流變化如何分別一直是失智症的研究上令人感興趣的課題。本研究的目的在探討年齡與腦血流影像之間的相互關係。
方法: 研究對象為52位年齡75歲以上(28位男性，24位女性，年齡75-95歲，平均81±5歲)輕度阿茲海默症病患(依據NINCDS-ADRDA診斷為準則)且其薝床失智症指數(Clinical Dementia Rating, CDR)均為1。我們以每個人的年齡作為其變數(covariate)，在體素對體素的基準(voxel-by-voxel basis)上使用SPM99 (Statistical Parametric Mapping)速行鎝-99m-HMPAO腦血流灌注斷層掃描(cerebral perfusion SPECT)影像的簡單回歸分析(simple regression)分椷。
結果: 下列腦中的五個區域與年齡有明確的負相關性: (1)左前下腦回(left inferior frontal gyrsu) [未修正P值<0.0003，叢集體積(cluster size)=517體素]; (2)左腦島(left insula) (未修正P值<0.00003，叢集體積=350體素); (3)左丘腦 (left thalamus) (未修正P值<0.00002，叢集體積=337體素); (4)右前扣帶(right anterior cingulated) (未修正P值<0.00002，叢集體積=395體素); (5)右腦島(right insula)(未修正P值<0.00002，叢集體積=245體素)。如果將顯著水準(significance level)提高為修正P值<0.05 (corrected for multiple statistics)再進行分析時，則未發現任何叢集或體素和年齡有顯著的正或負的相關性(correlation)。
結論: 本研究結果雖然通過了未修正P值<0.001之顯著水準，但是仍無法通過修正P值<0.05之顯著水準因而提高了偽陽性的可能，所以未來有必要針對更大的抽樣群或是使用更靈敏的診斷儀器(如正子造影)做更進一步研究。

Background: It remains a matter of controversy as to whether cerebral perfusion declines with healthy aging. The present study was aimed at clarifying the correlation between age and CBF image.
Methods: Fifty-two patients (28 male, 24 female; age range, 75-95 years; mean ± SD, 81 ± 5 years) with mild Alzheimer's disease were included in this study. The clinical dementia rating (CDR) of these patients was scored 1. We performed simple regression analysis using SPM99 with the individual's age as a covariate on a voxel-by-voxel basis.
Results: Significant negative correlations with the individual's age were found in the following five areas: (1)left inferior frontal gyrus (uncorrected P<0.00003; cluster size=517 voxels); (2) left insula (uncorrected P<0.00003; cluster size=350 voxels); (3 left thalamus (uncorrected P<0.0002; cluster size= 337 voxels); (4)right anterior cingulated (uncorrected P<0.0002; cluster size=395 voxels); (5) right insula (uncorrected P<0.0002; cluster size=245 voxels). There was no any voxel showing significant correlation with age if P<0.05 (corrected for multiple statistics).
Conclusions: Since our results did not reach a correction to the significance of the t statistics which takes account of the multiple comparisons in the image, the possibility of false positive results could be high. Further study with larger sample size or with more sensitive diagnostic modalities, such as positron emission tomography, is necessary.