OBJECTIVE To determine whether migraineurs may have a systemic deficiency of magnesium. BACKGROUND Magnesium deficiency has been shown to play a potential role in the pathogenesis of migraine, but there are no data on total body magnesium status in migraineurs. METHODS An oral magnesium load test was performed by giving 3000 mg of magnesium lactate during a 24-hour interictal period to 20 patients with migraine (15 women and 5 men; mean age, 37.9 years) and 20 healthy volunteers (16 women and 4 men; mean age, 39.6 years). Baseline and postload magnesium concentrations were determined from serum and 24-hour urine specimens. RESULTS There was no significant difference between the groups in the baseline serum and urine magnesium concentrations, although the latter tended to be lower (P = .064) in the migraine group. The postload magnesium concentrations were significantly higher within both the migraine (P < .0001 and P < .0001) and the control (P = .0009 and P < .0001) groups compared to the baseline values. After loading, the 24-hour urinary magnesium excretions were significantly lower (P = .0007) in the patients with migraine than in the controls, but serum values did not differ. CONCLUSIONS Magnesium retention occurs in patients with migraine after oral loading, suggesting a systemic magnesium deficiency.