File(s) under permanent embargo
Lipopolysaccharide binding protein and bone mineral density: Evidence from a male cohort of the Geelong osteoporosis study
conference contribution
posted on 2019-01-01, 00:00 authored by Jasmine Cleminson, Julie PascoJulie Pasco, Chiara BortolasciChiara Bortolasci, J M Hodge, K Anderson, Amanda StuartAmanda Stuart, Mark KotowiczMark Kotowicz, Lana WilliamsLana WilliamsObjectives: Increased risk of osteoporosis and related fractures are associated with gastrointestinal diseases, including Crohn’s disease and irritable bowel syndrome, yet a possible link between bacterial lipopolysaccharides (endotoxins), intestinal activity and downstream effects on bone remain unknown. We aimed to investigate lipopolysaccharide binding protein (LBP) levels and BMD in a randomly selected population-based cohort of men.
Methods: Serum LBP (ng/mL) was measured using enzyme-linked immunosorbent assay (ELISA; R&D Systems) for 1149 men (ages 20-96 y, median 61 y) enrolled in the Geelong Osteoporosis Study (GOS). BMD (g/cm2) was measured at the PA-spine, hip, total body and forearm using DXA (Lunar). Weight and height were measured; medication use, physical activity and smoking status were self-reported. LBP values were natural log transformed (ln-LBP) and associations between ln-LBP and bone measures were tested using Pearson’s correlation. Multivariable linear regression models were developed to test associations after adjusting for age, weight, height, physical activity and medications affecting bone (thyroid medication, bisphosphonates, oral glucocorticoids).
Results: Relationships between ln-LBP (median 16.5 ng/mL, IQR:11.5-23.1) and age (r2=0.06, p=0.04), weight (r2=-0.08, p=0.005), total body (r2=-0.09, p=0.003), spine (r2=-0.05, p=0.08), distal- (r2=-0.02, p=0.003) and mid-forearm BMD (r2=-0.15, p=<0.001) were significant, but not hip BMD (r2=-0.04, p=0.14). After adjustments, LBP was associated with decreased BMD at the mid-forearm (β -0.013, SE±0.004, p<0.001) and showed a trend with decreased spine (β -0.012±0.009, p=0.19) and total body BMD (β -0.006±0.004, p=0.13). No associations between LBP and hip (β -0.002±0.006, p=0.77) and distal-forearm BMD (β 0.001±0.003, p=0.66) were evident.
Conclusions: While acknowledging potential unrecognised confounding, these data suggest LBP is independently associated with mid-forearm, and potentially spine and total body BMD, in men. These results support a positive correlation with age, though a negative correlation with weight. Given the paucity of data, replication and future studies to confirm this relationship are warranted.
Methods: Serum LBP (ng/mL) was measured using enzyme-linked immunosorbent assay (ELISA; R&D Systems) for 1149 men (ages 20-96 y, median 61 y) enrolled in the Geelong Osteoporosis Study (GOS). BMD (g/cm2) was measured at the PA-spine, hip, total body and forearm using DXA (Lunar). Weight and height were measured; medication use, physical activity and smoking status were self-reported. LBP values were natural log transformed (ln-LBP) and associations between ln-LBP and bone measures were tested using Pearson’s correlation. Multivariable linear regression models were developed to test associations after adjusting for age, weight, height, physical activity and medications affecting bone (thyroid medication, bisphosphonates, oral glucocorticoids).
Results: Relationships between ln-LBP (median 16.5 ng/mL, IQR:11.5-23.1) and age (r2=0.06, p=0.04), weight (r2=-0.08, p=0.005), total body (r2=-0.09, p=0.003), spine (r2=-0.05, p=0.08), distal- (r2=-0.02, p=0.003) and mid-forearm BMD (r2=-0.15, p=<0.001) were significant, but not hip BMD (r2=-0.04, p=0.14). After adjustments, LBP was associated with decreased BMD at the mid-forearm (β -0.013, SE±0.004, p<0.001) and showed a trend with decreased spine (β -0.012±0.009, p=0.19) and total body BMD (β -0.006±0.004, p=0.13). No associations between LBP and hip (β -0.002±0.006, p=0.77) and distal-forearm BMD (β 0.001±0.003, p=0.66) were evident.
Conclusions: While acknowledging potential unrecognised confounding, these data suggest LBP is independently associated with mid-forearm, and potentially spine and total body BMD, in men. These results support a positive correlation with age, though a negative correlation with weight. Given the paucity of data, replication and future studies to confirm this relationship are warranted.
