Endocrine Abstracts

Introduction

Recent evidence suggests a possible link between prediabetes and lower density in postmenopausal populations. Moreover, earlier data suggest a link between diagnosis of diabetes and lower levels of bone density in women after the menopausal transition. We aimed to evaluate the possible association between insulin resistance and a diagnosis of lumbar or hip osteoporosis and/or prevalent fractures of the lumbar spine.

Methods

This cross-sectional study included 322 postmenopausal women, retrieved from the Menopause Clinic of Aretaieio Hospital. Women with insulin resistance (homeostasis model assessment, HOMA-IR > 5) clinically overt cardiovascular disease, diabetes mellitus, untreated thyroid dysfunction, gynecological malignancy or endometrial thickness of more than 5 mm were excluded. Blood samples were obtained to perform biochemical and hormonal assessment including markers of bone turnover. Bone density at the lumbar spine (LS) and femoral neck (FN) was estimated using DXA. Lumbar spine plain radiographs were performed to assess for the prevalence of vertebral fractures (VFs).

Results

The prevalence of VFs was 7.5% (24/322), while LS-osteoporosis was identified in 8.8% (19/216) and FN-osteoporosis was identified in 14.8% (31/209) women. Bone density measures correlated with HOMA-IR (LS, T-score r-coefficient = 0.149, P-value = 0.028; FN, BMD r-coefficient = 0.143, P-value = 0.040; FN, T-score r-coefficient = 0.147, P-value = 0.033), age, menopausal age. Markers of bone turnover did not associate with HOMA-IR, glucose or insulin levels. Osteoporosis vs normal bone density/osteopenia was related with lower levels of HOMA-IR, especially at the lumbar spine (LS: 1.2 ± 0.4 vs 1.7 ± 0.9, P-value = 0.018; FN: 1.3 ± 0.6 vs 1.6 ± 0.9, P-value = 0.091), as well as with higher age and follicular stimulating hormone levels. Logistic regression analysis showed that the prevalence of VFs associated inversely with HOMA-IR levels (1^st vs 4^th quartile, 2.5% vs 7.5%, P-value 0.018 after bootstrapping for 1000 samples), adjusted for age, BMI, FSH, E2. Presence of LS-osteoporosis associated inversely with HOMA-IR levels (1^st vs 4^th quartile, 13.7% vs 1.8%, P-value = 0.030 after bootstrapping for 1000 samples) as well as menopausal age, adjusted for the same variables. Presence of FN-osteoporosis associated inversely with HOMA-IR levels (1st vs 4th quartile, 16.9% vs 4.5%, P-value = 0.040 after bootstrapping for 1000 samples), adjusted for the same variables.

Conclusion

The development of VFs, LS- as well as FN-osteoporosis is inversely associated with insulin resistance, as estimated by levels of HOMA-IR, in a non-diabetic postmenopausal population.