RESUMEN
Leukocyte telomere length (LTL) and bone mineral density (BMD) are associated with health and mortality. Because osteoporosis is an age-related condition and LTL is considered to be a biomarker of aging, we hypothesized that shorter LTL could predict lower BMD. The aim of our study was to assess whether there is an association of LTL with BMD and to determine whether this possible association is independent of age. The BMDs of the lumbar spine (LS), femoral neck (FN) and total hip (TH) were evaluated in 460 women using DXA. LTL was analyzed using quantitative polymerase chain reaction. The women completed a health and lifestyle questionnaire. The associations were estimated by regression models that considered age, body mass index (BMI), menopause, physical activity, alcohol consumption and smoking habits. We found a statistically significant unadjusted association between LTL and age (estimate and 95% confidence interval (CI): -0.003 (-0.005; -0.002)); and between BMI adjusted age and logarithmic transformed BMD. Estimates and 95% CI were as follows: LS: -0.13 (-0.26; -0.01); right TH: -0.44 (-0.53; -0.34); left TH: -0.38 (-0.48; -0.28); right FN: -0.57 (-0.67; -0.46) and left FN: -0.51 (-0.62; -0.40). There were no statistically significant associations between BMD and LTL (both logarithmically transformed) with or without age adjustments. The age-adjusted estimates and CI were as follows: LS: -0.10 (-0.71; 0.52); right TH: -0.13 (-0.66; 0.41); left TH: -0.13 (-0.67; 0.42); right FN: -0.03 (-0.58; 0.52) and left FN: 0.09 (-0.47; 0.66). In conclusion, we found no statistically significant associations between BMD and LTL, although the estimates of the crude associations were all positive, indicating hypothesis consistency; that shorter LTL predict lower BMD values. This positive association was no longer apparent after adjusting for age. As expected, age was statistically significantly associated with both telomere length and BMI adjusted BMD.