Impact of Glucose Tolerance States on Serum 25-Hydroxyvitamin D Level and Bone Mineral Density

Presentation Number: SUN 589
Date of Presentation: April 2nd, 2017

Rudruidee Karnchanasorn*1, Horng-Yih Ou2, Wei Feng3, Raynald Samoa3, Lee-Ming Chuang4 and Ken C Chiu3
1The University of Kansas Medical Center, Kansas City, KS, 2National Cheng-Kung University Medical College and Hospital, Tainan, Taiwan, 3City of Hope National Medical Center, Duarte, CA, 4National Taiwan Univesity Hospital, Taipei, Taiwan


Diabetes mellitus (DM) is associated with the increased risk of fracture, although the underlying mechanism remains unknown. The effect of diabetes on bone mineral density (BMD) remains conflicting as well as serum 25-hydroxyvitamin D (25OHD) level. To address these issues, we compared to serum 25OHD level and BMD among different states of glucose tolerance in a representative US population.

This study included 11,651 subjects in 5 major racial/ethnic groups: Mexican American (MA), other Hispanic (OH), non-Hispanic white (NHW), non-Hispanic black (NHB), and others (OT). States of glucose tolerance were defined by fasting and 2-hour postchallenged plasma glucose levels, HbA1c, or history of established diabetes as normal glucose tolerance (NGT), abnormal glucose tolerance (AGT), and diabetes mellitus (DM). The femoral neck BMD, determined by dual-energy X-ray absorptiometry, was compared among three groups with consideration of covariates, including age, body mass index (BMI), serum 25-hydroxyvitamin D level, smoking status, alcohol consumption, education level, and poverty index. Subjects who received treatment of osteoporosis were excluded from this study. The analyses were performed by genders and race/ethnicity to reduce their impact of BMD.

A significant difference in serum 25OHD levels among 3 groups based glucose tolerance states was only noted in the DM groups in the MA men (P=0.0004), NHW men (P=0.0099), NHW women (P<0.000001), and NHB women (P=0.0002). The major determinants of serum 25OHD levels were age and BMI. Age had a significant impact on serum 25OHD levels in all groups (P=0.007 to <0.000001) except for NHW men (P=0.83). BMI had a significant impact on serum 25OHD levels in all groups (P=0.02 to <0.000001) except for OH men (P=0.40), and OT men (P=0.65) and women (P=0.91). After adjustment for covariates, the difference in serum 25OHD levels was only noted in MA men (P=0.00005), OH men (P=0.002), and NHW men (P=0.03). A significant lower BMD was noted in the DM subjects in all groups (P<0.02-0.000001), expect the other male and female subjects in the OT groups (P=0.06 and P=0.74, respectively). The major determinants for BMD were age and BMI in all groups (P<0.000001). After adjustment for covariates, the difference in BMD was only noted in NHW women (P=0.002) and OT women (P=0.03).

In this representative US population, there is no uniform observation in serum 25OHD levels and femoral neck BMD among NGT, AGT, and DM subjects. The observed difference can be explained mostly by the differences in age and BMI. Thus, the increased fracture risk in DM subjects cannot be explained by serum 25OHD levels and femoral neck BMD.


Nothing to Disclose: RK, HYO, WF, RS, LMC, KCC