Park Nicollet Clinic and HealthPartners Institute, 3800 Park Nicollet Blvd., Minneapolis, MN, 55416, USA.
Division of Health Policy and Management, University of Minnesota, Minneapolis, MN, USA.
Osteoporos Int. 2018 Jan;29(1):49-59. doi: 10.1007/s00198-017-4273-z. Epub 2017 Oct 31.
Among older men, characteristics that predict longitudinal changes in trabecular bone score (TBS) are different from characteristics that predict changes in bone mineral density (BMD). Most notably, weight loss is strongly associated with concomitant loss in BMD but with concomitant increases in TBS, when measured on Hologic densitometers.
Our objective was to compare and contrast predictors of changes in TBS, total hip BMD, and lumbar spine BMD.
Our study population was 3969 Osteoporotic Fractures in Men (MrOS) cohort participants (mean age 72.8 years) with repeat measures of TBS, lumbar spine and total hip BMD, body mass index (BMI) less than 37 kg/m, and no use of bisphosphonate or glucocorticoid medications. TBS was scored (Med-Imaps Software version 2.1) and BMD measured on Hologic densitometers.
One thousand four hundred forty-four men had a TBS decrease > 0.04 units (estimated least significant change for TBS), 795 men had a TBS increase > 0.04 units, and 1730 men had TBS change ≤ 0.04 units over mean follow-up of 4.6 years. Older age was not associated with TBS change, but was associated with greater decline in lumbar spine and total hip BMD. Compared to stable weight, > 10% weight loss was strongly associated with an increase in TBS [effect size = 1.24 (95% CI 1.12, 1.36)] and strongly associated with a decrease in total hip BMD [- 1.16 (95% CI - 1.19, - 1.03)]. Other predictors discordant for longitudinal changes of TBS and BMD included baseline BMI, walk speed, and ACE inhibitor use.
Predictors of changes in TBS are different from predictors of changes in lumbar spine and total hip BMD. At least when assessed on Hologic densitometers, weight loss is associated with subsequent declines in spine and total hip BMD but subsequent increase in TBS. Faster walk speed may protect against loss of hip BMD, but is not associated with longitudinal changes of TBS.
目的：比较和对比骨小梁评分（TBS）、全髋骨密度（BMD）和腰椎 BMD 变化的预测因素。
方法：我们的研究人群是 3969 名男性骨质疏松性骨折（MrOS）队列参与者（平均年龄 72.8 岁），他们的 TBS、腰椎和全髋 BMD 进行了重复测量，体重指数（BMI）<37kg/m2，且未使用双膦酸盐或糖皮质激素药物。使用 Med-Imaps 软件版本 2.1 对 TBS 进行评分，使用 Hologic 密度仪测量 BMD。
结果：1444 名男性 TBS 下降>0.04 个单位（TBS 估计最小有意义变化），795 名男性 TBS 增加>0.04 个单位，1730 名男性 TBS 变化≤0.04 个单位，平均随访 4.6 年。年龄较大与 TBS 变化无关，但与腰椎和全髋 BMD 下降更相关。与体重稳定相比，>10%的体重减轻与 TBS 增加呈强相关（效应大小=1.24（95%置信区间 1.12，1.36）），与全髋 BMD 下降呈强相关[-1.16（95%置信区间-1.19，-1.03）]。TBS 和 BMD 纵向变化的其他预测因素包括基线 BMI、行走速度和 ACE 抑制剂使用。
结论：TBS 变化的预测因素与腰椎和全髋 BMD 变化的预测因素不同。至少在 Hologic 密度仪上评估时，体重减轻与脊柱和全髋 BMD 的随后下降相关，但与 TBS 的随后增加相关。较快的步行速度可能有助于防止髋部 BMD 丢失，但与 TBS 的纵向变化无关。
