Skin Autofluorescence, a Noninvasive Biomarker for Advanced Glycation End-Products, Is Associated With Prevalent Vertebral and Major Osteoporotic Fractures: The Rotterdam Study

Advanced glycation end-products (AGEs), which bind to type 1 collagen in bone and skin, have been implicated in reduced bone quality. The AGE reader (TM) measures skin autofluorescence (SAF), which might be regarded as a marker of long-term accumulation of AGEs in tissues. We investigated the association of SAF with bone mineral density (BMD) and fractures in the general population. We studied 2853 individuals from the Rotterdam Study with available SAF measurements (median age, 74.1 years) and with data on prevalent major osteoporotic (MOFs: hip, humerus, wrist, clinical vertebral) and vertebral fractures (VFs: clinical + radiographic Genant's grade 2 and 3). Radiographs were assessed 4 to 5 years before SAF. Multivariate regression models were performed adjusted for age, sex, BMI, creatinine, smoking status, and presence of diabetes and additionally for BMD with interaction terms to test for effect modification. Prevalence of MOFs was 8.5% and of VFs 7%. SAF had a curvilinear association with prevalent MOFs and VFs and therefore, age-adjusted, sex stratified SAF quartiles were used. The odds ratio (OR) (95% confidence interval [CI]) of the second, third and fourth quartiles of SAF for MOFs were as follows: OR 1.60 (95% CI, 1.08-2.35;p= .02); OR 1.30 (95% CI, 0.89-1.97;p= .20), and OR 1.40 (95% CI, 0.95-2.10;p= .09), respectively, with first (lowest) quartile as reference. For VFs the ORs were as follows: OR 1.69 (95% CI, 1.08-2.64;p= .02), OR 1.74(95% CI, 1.11-2.71;p= .01), and OR 1.73 (95% CI, 1.12-2.73;p= .02) for second, third, and fourth quartiles, respectively. When comparing the top three quartiles combined with the first quartile, the OR (95% CI) for MOFs was 1.43 (95% CI, 1.04-2.00;p= .03) and for VFs was 1.72 (95% CI, 1.18-2.53;p= .005). Additional adjustment for BMD did not change the associations. In conclusion, there is evidence of presence of a threshold of skin AGEs below which there is distinctly lower prevalence of fractures. Longitudinal analyses are needed to confirm our cross-sectional findings. (c) 2020 The Authors.Journal of Bone and Mineral Researchpublished by American Society for Bone and Mineral Research.