It is unclear if advanced glycation end products (AGEs) are involved in the bone fragility of type 1 diabetes (T1D). We evaluated whether skin AGEs by skin autofluorescence and serum AGEs (pentosidine, carboxymethyl-lysine [CML]) are independently associated with BMD by DXA (lumbar spine, hip, distal radius), trabecular bone score (TBS), serum bone turnover markers (BTMs: C-terminal crossed-linked telopeptide of type 1 collagen, CTX; procollagen type 1 N-terminal propeptide, P1NP; osteocalcin), and sclerostin in participants with and without T1D. Linear regression models were used, with interaction terms to test effect modification by T1D status. In participants with T1D, correlations between skin and serum AGEs as well as between AGEs and 3-year HbA1C were evaluated using Spearman’s correlations. Data are mean±SD or median(interquartile range). We included individuals who participated in a cross-sectional study and had BMD and TBS assessment (106 T1D/65 controls, 53.2% women, age 43±15 years, BMI 26.6±5.5 kg/m2). Participants with T1D had diabetes for 27.6±12.3 years, a mean 3-year HbA1C of 7.5±0.9% and skin AGEs of 2.15±0.54 arbitrary units. A subgroup of 65 T1D/57 controls had BTMs and sclerostin measurements, and those with T1D also had serum pentosidine (16.8[8.2-32.0] ng/mL) and CML [48.0±16.8] ng/mL) measured. Femoral neck BMD, TBS, and BTMs were lower while sclerostin levels were similar in participants with T1D vs controls. T1D status did not modify the associations between AGEs and bone outcomes. Skin AGEs were significantly associated with total hip and femoral neck BMD, TBS, BTMs, and sclerostin before, but not after adjustment for confounders. Serum AGEs were not associated with any bone outcome. There were no significant correlations between skin and serum AGEs, or between AGEs and 3-year HbA1C. In conclusion, skin and serum AGEs are not independently associated with BMD, TBS, BTMs, and sclerostin in participants with relatively well-controlled T1D and participants without diabetes.
