Effect of high-intensity interval training on hippocampal metabolism in older adolescents

Abstract

Although well-evidenced in older adults, the effects of exercise on the hippocampus in youth are relatively unknown. This study examined the impact of a 6-month school-based physical activity intervention on hippocampal metabolism in adolescents using magnetic resonance spectroscopy. A subset of lower fit older adolescents [N = 56, 61% female, 16.1 ± 0.4 years] was included from four secondary schools (10 classes) in New South Wales, Australia, who were participating in a larger cluster randomized controlled trial. Participants were randomized to the Burn 2 Learn (B2L) intervention (five classes, 30 participants) or a control group (five classes, 26 participants). Changes in hippocampal metabolism were assessed using linear mixed models adjusted for clustering at the class level. We observed group-by-time effects for the B2L intervention on N-acetylaspartate (NAA) (+2.66 mmol/L, 95% CI 0.20 to 5.11, d = 0.66) and glutamate+glutamine (Glx) (+3.38 mmol/L, 95% CI 0.34 to 6.42, d = 0.67) in the left hippocampus. Increases in left hippocampal NAA and Glx concentrations were associated with improvements in cardiorespiratory fitness (NAA: r_s = 0.52, p = .016; Glx: r_s = 0.57, p = .007), lower body muscular fitness (NAA: r_s = 0.49, p = .018; Glx: r_s = 0.59, p = .003), and working memory (NAA: r_s = 0.42, p = .032; Glx: r_s = 0.43, p = .028) in the intervention group. Our findings suggest physical activity may improve hippocampal metabolism in lower fit older adolescents with implications for working memory. Further studies involving larger samples are needed to replicate our findings.

Keywords: adolescents; cognition; exercise; working memory.

FIGURE 1

Position of the coronal 15 mm slice with voxel size of 15 × 10 × 10 mm³ (AP‐RL‐HF) within the VOI shown in axial, sagittal, and coronal T1 MPRAGE planes with the bilateral hippocampus highlighted in red

FIGURE 2

Spectra modeling in LCModel shows the fit‐curve calculated to estimate the chemical shift of signal peaks corresponding to molecular structure and the peak intensity corresponding to metabolite concentration

FIGURE 3

Flow of clusters, participants, and MRS data through the study