Introduction

With extremely prevalent rates of obesity currently affecting today’s children, the incidence of obesity-related obstructive sleep apnea (OSA) is also increasing¹. OSA is characterized by intermittent episodes of nocturnal hypoxia, hypercapnia and sleep disruption and is, therefore, a significant contributor to co-morbid disease in obese children. These features of OSA lead to oxidative damage of endothelial cells, increase cerebral blood flow (CBF) and therefore disrupts the vasodilatory capacity and increases the risk of cerebrovascular dysfunction². The obese OSA population also experience impaired neurocognitive ability and poor school performance which can be associated with cerebrovascular damage³. Non-invasive magnetic resonance imaging strategies can assess vessel distensibility, and therefore vasodilatory capacity, by quantifying the relative changes in CBF in response to a vasoactive stimulus⁴. The resulting measure, cerebrovascular reactivity (CVR), can help us understand the mechanism of cerebrovascular injury and can serve a useful tool in the clinical assessment of cerebrovascular health. We hypothesize that CVR measures will decrease with higher CBF and with increasing severity of OSA, determined by the Obstructive Apnea-Hypopnea Index (OAHI), in obese children.

Methods

33 obese patients between 9 and 18 years old were included and received overnight polysomnography tests to assess severity of OSA. They were classified as having no OSA (n=10, OAHI<1.5), mild-OSA (n=9, 1.5≤OAHI≤5.0) or moderate-to-severe OSA (n=14, OAHI>5.0). Each patient was then imaged on a 3T MRI scanner using a 32-channel head coil. CVR data were acquired using a blood-oxygen level dependent (BOLD) sequence during a computer-controlled administration of a vasoactive stimulus through a rebreathing mask. The BOLD CVR images were acquired with TR/TE = 2000/40ms, FOV = 220mm, matrix size = 64×64, slices = 25, slice thickness = 4.5mm, volumes = 240, time = 8 minutes. The stimulus paradigm alternated between 60 seconds of normocapnia (PETCO₂ = 40mmHg, PETO₂ = 100mmHg) and 45 seconds of hypercapnia (PETCO₂ = 45mmHg, PETO₂ = 100mmHg). CBF was measured using a PICORE arterial spin labeling sequence and quantified with the vendor pipeline. Additionally, a high resolution MPRAGE anatomical scan was performed for tissue classification. The BOLD CVR and CO₂ data were transferred to an independent workstation for post-processing and analysis. High resolution CVR maps were computed using FSL v1.4 as previously described⁵. CVR maps for each subject were co-registered to the corresponding anatomical scan and then average across grey and white matter regions. SPSS v23 was used to perform comparisons between OSA groups and Pearson correlation analysis on the resulting data.

Results

Patients in the moderate-to-severe OSA group had higher GM CBF measures compared to the no OSA and mild OSA groups (56.1, 50.4 and 50.9 mL/100g/min respectively, p<0.05). CVR measures were similar across the no, mild and moderate-to-severe groups (0.301, 0.273 and 0.285 %ΔMR/mmHg CO₂, respectively). Higher CBF was inversely associated with CVR in obese patients with moderate-to-severe OSA for both GM [Figure 1] and WM [Figure 2] (r=-0.331 and r=-0.590, respectively). Furthermore, in patients with moderate-to-severe OSA, a decline in CVR was observed for both GM and WM [Figure 3] when plotted against the Obstructive Sleep Apnea Index (OAHI) (r=-0.297 and r=-0.508, respectively).

Discussion

In this study, we have demonstrated higher CBF values in obese patients with moderate-to-severe OSA compared to those with no or mild OSA. Reduced CVR in children with obesity and moderate-to-severe OSA may be exposed to a higher risk of cerebrovascular damage. If left untreated, this may present behaviourally as neurocognitive deficits and serious vasculopathies such as stroke. Further studies are needed to confirm our results in a larger cohort and to assess the effectiveness of CPAP therapy.

Acknowledgements

No acknowledgement found.