Studies suggest a link between magnesium status and osteoporosis. One barrier to more conclusive research on the potential relation is measuring intestinal magnesium absorption (MgA), which requires the use of stable isotopes and a ≥6-d stool or 3-d urine collection. We evaluated alternative methods of measuring MgA. We administered 2 stable magnesium isotopes to 15 postmenopausal women (cohort 1) aged 62 ± 8 y with a dietary magnesium intake of 345 ± 72 mg/d. Participants fasted from 1200 h to 0700 h and then consumed breakfast with ∼23 mg of oral ²⁶Mg and ∼11 mg of i.v. ²⁵Mg. We measured magnesium isotope concentrations in 72-h urine, spot urine (36, 48, 60, and 72 h), and spot serum (1, 3, and 5 h) samples collected after isotope dosing. We calculated MgA using the dose-corrected fraction of isotope concentrations from the 72-h urine collection. We validated new methods in 10 postmenopausal women (cohort 2) aged 59 ± 5 y with a dietary magnesium intake of 325 ± 122 mg/d. In cohort 1, MgA based on the 72-h urine collection was 0.28 ± 0.08. The 72-h MgA correlated most highly with 0-24 h urine MgA value alone (ρ = 0.95, P < 0.001) or the mean of the 0-24 h urine and the 3-h (ρ = 0.93, P < 0.001) or 5-h (ρ = 0.96, P < 0.001) serum MgA values. In cohort 2, Bland-Altman bias was lowest (-0.003, P = 0.82) using means of the 0-24 h urine and 3-h serum MgA values. We conclude that means of 0-24 h urine and 3-h serum MgA provide a reasonable estimate of 72-h MgA. However, if researchers seek to identify small changes in MgA, we recommend a 3-d urine or extended stool collection.
Trial registration: ClinicalTrials.gov NCT01593501.