Younger age was associated with greater CBF in the hippocampus (r=−0.393; p=0.01) and brainstem (r=−0.289; p=0.013), and aerobic fitness was associated with sex such that females had lower fitness scores than males (r=0.311, p=0.007). Thus, we included age and sex as covariates in the regression model. Despite no significant correlations between hippocampal CBF and hippocampal volume (r=0.004, p>0.05), or brainstem CBF and brainstem volume (r=−0.015, p>0.05), we also included volume as a covariate to confirm that blood flow effects were independent of the size of the Nanaomycin A region.
Higher aerobic fitness predicted greater CBF in the hippocampus (β=0.235, t=2.085, p=0.041) when controlling for age, sex, and average hippocampal volume (Fig. 2). In particular, higher aerobic fitness predicted greater CBF in the posterior hippocampus (β=0.241, t=2.163, p=0.034), when controlling for age, sex, and posterior hippocampal volume (Fig. 2). Higher aerobic fitness was marginally associated with CBF in the anterior hippocampus (β=0.201, t=1.723, p=0.090), when controlling for age, sex, and anterior hippocampal volume (Fig. 2). As hypothesized, aerobic fitness did not predict CBF in the brainstem (β=0.017, t=0.136, p=0.892), when controlling for age, sex, and brainstem volume (Fig. 2).
Our results raise the possibility that aerobic fitness plays a role in vascularization of the hippocampus during childhood. Studies suggest a positive benefit of aerobic exercise on brain vasculature in animals (Black et al., 1990; Clark et al., 2009; Kleim et al., 2002; Rhyu et al., 2010) as well as cerebral blood flow measures in middle-aged and older humans (Bullitt et al., 2009; Burdette et al., 2010; Pereira et al., 2007). Here we suggest that these associations may extend to a child population during a critical period of maturation. In fact, angiogenesis has been directly coupled with cerebral blood volume (Dunn et al., 2004; Jiang et al., 2005; Lin et al., 2002; Maia et al., 2005; Sugahara et al., 1998). Although we measured cerebral blood flow rather than cerebral blood volume, we expect perfusion and blood volume to be closely related via the Central Volume Theorem (Newman et al., 2006; Stewart, 1893). A tissue with higher perfusion likely has higher blood volume to sustain the perfusion; however, flow also depends on how quickly the blood passes through the tissue, usually quantified in terms of mean transit time. We postulate that increased blood water delivery and availability in the hippocampus, as a function of higher aerobic fitness, may be due to more blood vessels in this region.
Yet as a variety of molecular and cellular cascades accompany hippocampal changes with aerobic exercise, we can only speculate about the biological mechanisms underlying increased perfusion. For example, in addition to changes in vasculature, aerobic exercise is known to increase cell proliferation and cell survival (Cotman and Berchtold, 2002; Ding et al., 2006), dendritic structure (Redila and Christie, 2006), growth factors (Neeper et al., 1996), and gliogenesis (Uda et al., 2006) in the hippocampus. In fact, angiogenesis and neurogenesis are tightly linked (Louissaint et al., 2002; Palmer et al., 2000). For instance, blocking the secretion of vascular endothelial growth factor (VEGF), a neurotrophic molecule involved in blood vessel growth (Lopez-Lopez et al., 2004), has been found to abolish exercise-induced neurogenesis (Fabel et al., 2003). Further, measures of cerebral blood volume have been said to provide an in vivo correlate of neurogenesis (Pereira et al., 2007). It is possible that some fitness-related differences in cerebral blood flow may be mediated, in part, by neurogenesis.
We are the first to explore the plasticity of perfusion in the anterior and posterior hippocampus in children. Our results do not suggest compelling specificity of fitness on anterior or posterior perfusion (independent of volume), with a significant positive association between fitness and posterior hippocampal CBF and a marginal positive relationship between fitness and anterior hippocampus. Given functional distinctions along the anterior/posterior axis of the hippocampus (Giovanello et al., 2009; Sperling et al., 2003), future studies should integrate a relational memory task to explore the links among aerobic fitness, cerebrovascular function in sections of the hippocampus, and cognitive function specific to the hippocampus.