Effect of recent alcohol intake on parathyroid hormone and mineral metabolism in men.

The mechanisms by which alcohol intake, particularly moderate alcohol intake, effects bone metabolism are poorly defined. We have examined the relationship between mineral metabolism and recent self-reported alcohol intake (SRAI) across a wide range of such intakes in a series of 104 men aged 32 to 78 years of age in an outpatient setting. A morning nonfasting urine, serum specimen and recent SRAI were obtained from each subject. SRAI was reported as between 0 and 45 oz/week. SRAI correlated positively with liver function tests, including serum bilirubin (r = 0.30, p = 0.002), alkaline phosphatase (r = 0.30, p = 0.004), and aspartate aminotransferase (SGOT) (r = 0.29, p = 0.006). SRAI correlated with serum calcium corrected for albumin (r = -0.39, p < 0.001), estradiol (r = 0.43, p < 0.001), and immunoreactive parathyroid hormone (iPTH) (r = -0.51, p < 0.001), as well as urinary calcium (per 100 mg of creatinine) (r = 0.55, p < 0.001). We have arbitrarily divided the participants into two groups on the basis of their reported alcohol intake. Individuals in the first group had intakes ranging from none to moderate intake (drank 8.4 oz or less of ethanol per week, equivalent to an average of two drinks daily or less). Those in the second group had moderate or heavier intake, with >8.4 oz of ethanol intake/week. Mean serum iPTH was significantly greater in those in the first group (none to moderate), compared with the second group (moderate or heavier) (56.0 +/- 3.4 and 39.9 +/- 2.0 pM/liter, respectively). Calcium corrected for serum albumin was significantly greater in individuals in the first, compared with the second, group (9.23 +/- 0.05 vs. 8.88 +/- 0.07 mg/dl, respectively). In addition, urinary calcium (corrected per 100 mg of creatinine) was significantly lower in the former, compared with the latter (3.1 +/- 0.4 vs. 8.4 +/- 1.1 mg/100 mg of creatinine, respectively). Similarly, urinary excretion of collagen crosslinks (corrected per 100 mg of creatinine) was significantly less in men in the second group, compared with the first group (316 +/- 38 vs. 530 +/- 78 nM/100 mg of creatinine, respectively). Not surprisingly, a series of correlations between iPTH and age, 250-hydroxyvitamin D, and testosterone were significant in individuals with none to moderate SRAI, but not moderate or heavier SRAI. Significant independent predictors of serum iPTH in the entire group of men were age (beta = 0.215, p = 0.025), SRAI (beta = -0.281, p = 0.003), 250-hydroxyvitamin D (beta = -0.309, p = 0.002), and testosterone (beta = -184, p = 0.048). We have concluded that, in free-living men, alcohol intake >8.4 oz/week was associated with decreased serum iPTH concentrations.