Cerebral venous oxygenation (Yv) is an important physiological marker of the brain, as it has a direct association with brain oxygen metabolism. Recent studies have suggested that Yv may be a sensitive marker in differentiating normal aging from Alzheimer’s disease, namely that Yv shows a pronounced decrease with age but is elevated in patients with Mild Cognitive Impairment (MCI) [1]. A plausible hypothesis is that, the brain tries to compensate the negative impact of aging by increasing its oxygen consumption, leaving less oxygen in the vein; but with neurodegeneration, neural activity and oxygen consumption are diminished, resulting in more oxygen remaining in the vein. In this study, we further test and substantiate this mechanism by demonstrating that Yv is higher in ApoE4 carriers who have increased risks of AD and that higher Yv is associated with poorer cognitive performance in healthy older individuals.

Participants

65 cognitively normal elderly subjects (Male/Female: 28/37; Age: 70.44 ± 8.25 yrs) from a longitudinal cohort of BIOCARD (Biomarkers for Older Controls at Risk for Dementia) study were recruited. All subjects were carefully screened and reported no cognitive impairment or dementia.

MRI Experiment

All subjects were studied on a 3T Philips System. Global brain Yv was measured from the superior sagittal sinus (SSS) using the novel T2-relaxation-under-spin-tagging (TRUST) MRI approach (Figure 1) [2]. TRUST MRI is based on the principle that T2 of the blood has a well-known and calibratable relationship with Yv. The scan duration of TRUST MRI was 1.2 min and its imaging protocol followed that of a recent multi-site trial [3].

Genotype, Cognitive Assessment, and CSF biomarkers

Other tests were performed as part of the longitudinal BIOCARD protocol. ApoE genotype was determined by standard procedures. Four cognitive tests were conducted on the participants, including Wechsler Adult Intelligence Scale – Revised (WAIS-R) Digit Symbol Test, Logical Memory test, Paired Associates Subtests of Wechsler Memory Scale – Revised (WMS-R) and Boston Naming Test (BNT). A-beta 42 concentration of each participant was measured using CSF samples obtained from lumbar punctures.

Statistical Analysis

A multivariate linear regression model was used to examine the relationship among Yv, age and sex. Furthermore, the association between Yv and ApoE was evaluated by dividing the participants into one-copy-of-ApoE4 carriers (N=20), two-copies-of-APOE4 carriers (N=5) and non-ApoE4 carriers (N=39), with age and sex as covariates. Since we are testing a priori hypotheses, a one-tail p value of 0.05 or less is considered significant.

Figure 2 shows a scatter plot between Yv and age across subjects. Consistent with previous reports, cerebral venous oxygenation manifested a significant age-related decrease (R2=0.14, p=0.002). However, the age change itself does not tell whether the decrease in Yv is an indication of neural degradation (i.e. a bad sign) or a neural compensation (i.e. a good sign). Therefore, we further studied how a well-known genetic risk factor, ApoE4, influences Yv. Figure 3 shows Yv values of the participants by ApoE categories. It was found that, after equating for age, ApoE4 carriers have a significant higher Yv (p=0.003) compared to non-ApoE4 carriers, especially two-copies-of-APOE4 carriers. This finding suggests that, in older adults, decreased Yv represents a neural compensatory process and higher Yv may be an indication of early neurodegeneration.

We next examined how Yv may be correlated with cognitive performance. Linear regression analysis revealed that, after accounting for age, participants with higher Yv had lower scores in WAIS-R Digit Symbol Test (R2=0.10, p=0.047). This again suggests that higher Yv may be an indication of neurodegeneration in elderly individuals.

Finally, we studied how Yv may be related to A-beta 42 in CSF. Lower A-beta 42 in CSF is known to be a hallmark of early Alzheimer’s disease [4]. We observed that individuals with lower A-beta 42 tend to have higher Yv, although the relationship did not reach a statistical significance (p=0.111).

Most imaging biomarkers (e.g. brain atrophy) cannot effectively differentiate between normal aging and neurodegeneration, as their changes are often of the same direction. Here, by examining the relationship between Yv and several Alzheimer-specific hallmarks, we added more supporting evidence that, in older adults, a lower Yv value represents an active neural compensation process whereas a high Yv is an indication of neurodegeneration. Therefore, assessment of Yv with non-invasive MRI methods may present a potential simple opportunity to identify the transition point from normal to pathological aging.