CYTOKINE PRODUCTION AND BONE-MINERAL DENSITY AT THE LUMBAR SPINE AND FEMORAL-NECK IN PREMENOPAUSAL WOMEN

Cytokines such as interleukin-1 (IL-1 beta), interleukin-6 (IL-6) and tumor necrosis factor (TNF-alpha) can influence both bone resorption a nd bone formation. The objective of this cross-sectional study was to examine the relationship between cytokine production by peripheral blo od mononuclear cells (PBMC) and bone mineral density (BMD); the annual rate of change in BMD was examined. Subjects participating in a rando mized clinical trial entitled the Women's Healthy Lifestyle Project in Allegheny County, Pennsylvania were used. They included 50 healthy pr emenopausal women, aged 45-52 years, who had regular menses within the past 3 months and were not on replacement estrogens. Dual-energy X-ra y absorptiometry measurements at the AP lumbar spine and femoral neck were made at baseline and at the first annual exam using a Hologic QDR 2000 densitometer. Cytokine production of IL-1 beta, IL-6, and TNF-al pha, by PBMC was measured at the annual exam. The median values for st imulated cytokine production by PBMC were 3.92 ng/ml, 31.3 ng/ml, and 1.05 ng/ml, for IL-1 beta, IL-6, and TNF-alpha, respectively. There we re modest correlations between cytokine production and cross-sectional BMD, ranging from r = -0.30 to r = -0.13. Trends of greater spinal bo ne loss were observed in women with ''high'' (greater than or equal to 75th percentile) cytokine production of stimulated IL-1 beta and IL-6 (IL-1 beta: ''high'' = -1.56% +/- 0.70 versus ''low'' (<75th percenti le) = -0.56% +/- 0.35, P = 0.21). In contrast, greater annual gains in femoral neck BMD were observed in those with high cytokine production of IL-1 beta and IL-6 (IL-1 beta: high = 3.39% +/- 1.16 versus low = -0.85 +/- 0.58, P = 0.002). There was no association between stimulate d TNF production and annual change in BMD. In this population of healt hy premenopausal women, the relationship between cytokine production b y PBMC and the rate of change in BMD was significantly different for t he lumbar spine and femoral neck, possibly reflecting differences in t he proportion of trabecular and cortical bone at these sites.