Oral magnesium supplementation improves insulin sensitivity in non-diabetic subjects with insulin resistance. A double-blind placebo-controlled randomized trial.

OBJECTIVE: Although hypomagnesemia reduces insulin sensitivity, benefits of magnesium supplementation to non-diabetic insulin resistant subjects has not been established. Our purpose was to determine whether oral magnesium supplementation with magnesium chloride (MgCl2) 2.5 g daily modify insulin sensitivity in non-diabetic subjects. MATERIAL AND METHODS: This study was a 3 months randomized double-blind placebo-controlled trial. Apparently healthy subjects were eligible to participate if they had insulin resistance (HOMA-IR index equal or greater than 3.0) and hypomagnesemia (Serum magnesium levels equal or lower than 0.74 mmol/l). Subjects were randomized to receive either, MgCl2 2.5 g daily or placebo by 3-months. RESULTS: At baseline there were not significant anthropometric or laboratory differences between both groups. At ending of the study, magnesium-supplemented subjects significantly increased their serum magnesium levels (0.61 +/- 0.08 to 0.81 +/- 0.08 mmol/l, p<0.0001) and reduced HOMA-IR index (4.6 +/- 2.8 to 2.6 +/- 1.1, p<0.0001), whereas control subjects did not (0.62 +/- 0.08 to 0.61 +/- 0.08 mmol/l, p=0.063 and 5.2 +/- 1.9 to 5.3 +/- 2.9, p=0.087). CONCLUSIONS: Oral magnesium supplementation improves insulin sensitivity in hypomagnesemic non-diabetic subjects. Clinical implications of this finding have to be established.

Prevalence of abnormal adrenocortical function in human immunodeficiency virus infection by low-dose cosyntropin test.

Recent evidence suggests that 10 microg cosyntropin test has higher sensitivity for detecting hypothalamus-hypophysis-adrenal axis (HHA-A) dysfunction. Our objective was to determine prevalence of glucocorticoid insufficiency with the 10 microg cosyntropin test and the level of the HHA-A defect. One hundred and four HIV-infected patients underwent the 10 microg cosyntropin test. In abnormal and borderline respondents, insulin-induced hypoglycaemia test and human corticotropin releasing hormone test were used to confirm and localize the level of the HHA-A defect. Thirty-two patients with HIV infection and 72 with AIDS were identified. Prevalence of glucocorticoid insufficiency by the 10 microg cosyntropin test was 21.2%. By clinical categories, the frequency in AIDS and HIV infection patients was 26.4% and 9.4%, respectively. Confirmed glucocorticoid insufficiency by insulin-induced hypoglycaemia test was found in 16 out of 19 cases. Twelve cases had primary glucocorticoid insufficiency, 7 had secondary glucocorticoid insufficiency and 3 were false positive. In conclusion, adrenocortical dysfunction occurs in approximately 20% of the cases with HIV disease. Clinical findings commonly occurring in HIV disease as well as adrenocortical insufficiency are not reliable indicators for performing adrenocortical laboratory assessment. Our results suggest screening all AIDS patients with the 10 microg cosyntropin test.

A high-sensitivity test in the assessment of adrenocortical insufficiency: 10 microg vs 250 microg cosyntropin dose assessment of adrenocortical insufficiency.

The short cosyntropin (synthetic ACTH) test is recognized as the best screening manoeuvre in the assessment of adrenocortical insufficiency. Recent data, however, suggest that i.v. administration of 250 microg cosyntropin could be a pharmacological rather than a physiological stimulus, losing sensitivity for detecting adrenocortical failure. Our objective was to compare 10 vs 250 microg cosyntropin in order to find differences in serum cortisol peaks in healthy individuals, the adrenocortical response in a variety of hypothalamic-pituitary-adrenal axis disorders and the highest sensitivity and specificity serum cortisol cut-off point values. The subjects were 83 healthy people and 37 patients, the latter having Addison's disease (11), pituitary adenomas (7), Sheehan's syndrome (9) and recent use of glucocorticoid therapy (10). Forty-six healthy subjects and all patients underwent low- and standard-dose cosyntropin testing. In addition, 37 controls underwent the low-dose test. On comparing low- and standard-dose cosyntropin testing in healthy subjects there were no statistical differences in baseline and peaks of serum cortisol. In the group of patients, 2 out of 11 cases of Addison's disease showed normal cortisol criterion values during the standard test but abnormal during the low-dose test. In our group of patients and controls, the statistical analysis displayed a better sensitivity of the low-dose vs standard-dose ACTH test at 30 and 60 min. In conclusion, these results suggest that the use of 10 microg rather than 250 microg cosyntropin i.v. in the assessment of suspicious adrenocortical dysfunction gives better results.