Am J Epidemiol. 2023 Jan 6;192(1):62-69. doi: 10.1093/aje/kwac161.
Vertebral fractures have been associated with increased mortality, but findings are inconclusive, and many vertebral fractures avoid clinical attention. We investigated this association in a general population of 2,476 older adults aged ≥55 years from Tromsø, Norway, who were followed over 2007-2020, using dual-energy x-ray absorptiometry (DXA) at baseline to evaluate vertebral fractures (mild, moderate, or severe). We used multiple Cox regression models to estimate hazard ratios (HRs) for all-cause mortality, adjusted for age, sex, body mass index, education, smoking, alcohol intake, cardiovascular disease, and respiratory disease. Mean follow-up in the cohort was 11.2 (standard deviation, 2.7) years; 341 participants (13.8%) had ≥1 vertebral fracture at baseline, and 636 participants (25.7%) died between baseline and follow-up. Full-adjustment models showed a nonsignificant association between vertebral fracture status (yes/no) and mortality. Participants with ≥3 vertebral fractures (HR = 2.43, 95% confidence interval: 1.57, 3.78) or ≥1 severe vertebral fracture (HR = 1.65, 95% confidence interval: 1.26, 2.15) had increased mortality compared with those with no vertebral fractures. Dual-energy x-ray absorptiometry-based screening could be a potent and feasible tool in detecting vertebral fractures that are often clinically silent yet independently associated with premature death. Our data indicated that detailed vertebral assessment could be warranted for a more accurate survival estimation.
椎体骨折与死亡率增加有关，但研究结果尚无定论，而且许多椎体骨折并未引起临床关注。我们在挪威特罗姆瑟的一个年龄≥55 岁的 2476 名老年人普通人群中进行了这项研究，这些人在 2007 年至 2020 年期间接受了双能 X 线吸收法（DXA）基线检查，以评估椎体骨折（轻度、中度或重度）。我们使用多 Cox 回归模型估计全因死亡率的危险比（HR），调整了年龄、性别、体重指数、教育程度、吸烟、饮酒量、心血管疾病和呼吸系统疾病。队列的平均随访时间为 11.2 年（标准差为 2.7 年）；341 名参与者（13.8%）在基线时有≥1 个椎体骨折，636 名参与者（25.7%）在基线和随访期间死亡。完全调整模型显示，椎体骨折状态（是/否）与死亡率之间无显著关联。与无椎体骨折的参与者相比，有≥3 个椎体骨折（HR=2.43，95%置信区间：1.57，3.78）或≥1 个严重椎体骨折（HR=1.65，95%置信区间：1.26，2.15）的参与者死亡率更高。基于双能 X 线吸收法的筛查可能是一种有效且可行的工具，可用于检测通常临床上无症状但与过早死亡独立相关的椎体骨折。我们的数据表明，详细的椎体评估可能需要更准确地估计生存率。
