Normal variation in brain volumetrics, CSF dynamics, and ocular structures from magnetic resonance images of healthy participants over two years.

Changes in ocular and brain structure and cerebral spinal fluid (CSF) dynamics from magnetic resonance imaging (MRI) data have been reported in astronauts following long-duration spaceflight. The purpose of this study was to quantify normal variation in these outcomes over a 2-year period in a healthy cohort similar in age to astronauts, but without spaceflight experience. Variation in MRI acquisition, observer measurement, or normal aging over a 2-year period was used to determine thresholds for interpreting measures following long-duration spaceflight. Ten healthy participants (5 male), aged 38 ± 8 yr (mean ± SD), underwent five MRI sessions at baseline, 2, 6, 12, and 24 mo to quantify intracranial volumetry, CSF dynamics, pituitary morphology, and ocular structures. Total within-person variation ([Formula: see text]), modeled as the sum of the variance components for each outcome measure, was used to define a sensitivity threshold of 2.00 × [Formula: see text]. Changes greater than these thresholds can be interpreted as a result of an intervention and have <5% probability of occurring due to expected variability. Thresholds for change due to all sources of variability were: 27 mL for gray matter volume, 16 mL for white matter volume, 0.9 mL for lateral ventricular volume, 26.0 µL for CSF aqueductal stroke volume, 3.2 cm/s for peak-to-peak aqueductal CSF velocity, 0.5 mm for pituitary height, 0.50 mm for optic nerve sheath diameter, and 10.2 mm for posterior optic globe volume displacement. Changes on brain MR images after an intervention need to exceed these thresholds to be attributable to that intervention. We report estimated thresholds of normal variability from methodological and physiological sources in brain and ocular outcomes in healthy participants. Outcomes that exceed these thresholds have less than a 5% probability of occurring due to these sources of variation. One of the present findings suggests that about 50% of the spaceflight-induced increase in lateral ventricular volume exceeds contributions of expected sources of measurement variability.