Yun Huifeng, Liu Ye, Curtis Jeffrey R, Saag Kenneth, D'Erasmo Giavanna, Haseltine Katherine, Stein Emily M
Department of Epidemiology, School of Public Health, The University of Alabama at Birmingham, 1665 University Boulevard, Birmingham, AL 35233.
Division of Clinical Immunology and Rheumatology, The University of Alabama at Birmingham, Faculty Office Tower, 510 20th St S #834, Birmingham, AL 35294.
J Bone Miner Res. 2025 Feb 2;40(2):176-183. doi: 10.1093/jbmr/zjae162.
Epidural steroid injections (ESIs) are a common and often effective treatment for radicular back pain. While oral glucocorticoids increase fracture incidence, little is known regarding fracture risk after ESI. This study investigated the incidence of fractures among individuals who received ESI and those who did not. We hypothesized that ESI exposure would be associated with an increased incidence of osteoporotic fractures and specifically vertebral fractures. Using 2005-2018 5% Medicare data, individuals with radicular pain who had ≥1 ESI and those who did not (non-ESI) were matched 1:10 by age, sex, and month of radicular pain diagnosis using exposure density sampling (EDS). Using a high-dimensional propensity score (HDPS) calculated based on the top 500 covariates across multiple data dimensions, ESI and non-ESI individuals were matched 1:1. Fractures were identified using validated ICD-9/10 diagnosis codes. Fracture incidence rate (IR) was calculated by group, and hazard ratios (HR) compared using Cox regression. 25 062 ESI patients and 221 735 non-ESI patients who met eligibility criteria were identified using EDS. Mean age was 76 yr (74% female). Among ESI-treated individuals, there were 2296 fractures, IR 49.1 (95% CI: 47.2-51.2) per 1000 person yr. For non-ESI individuals, there were 11 917 fractures, IR 35.2 (95% CI: 34.5-35.8). Individuals who received ESI had a greater hazard of fracture at typical osteoporotic sites, HR 1.39 (95% CI 1.33-1.46) by EDS and 1.32 (1.12-1.54) by HDPS, and a greater hazard of vertebral fracture, 1.54 (1.45-1.64) by EDS and 1.69 (1.38-2.07) by HDPS. Patients who received greater cumulative ESI doses (≥3 in 1 yr) had a higher risk of fractures within the first 6 mo of follow-up. ESI exposure in older individuals is associated with an increased risk of fracture, suggesting there may be lasting detrimental skeletal effects of ESI. Further research into strategies to reduce fracture risk in this population is warranted.
硬膜外类固醇注射(ESI)是治疗神经根型背痛的一种常见且通常有效的方法。虽然口服糖皮质激素会增加骨折发生率,但关于ESI后的骨折风险却知之甚少。本研究调查了接受ESI和未接受ESI的个体的骨折发生率。我们假设,接触ESI会增加骨质疏松性骨折尤其是椎体骨折的发生率。利用2005 - 2018年5%的医疗保险数据,采用暴露密度抽样(EDS)方法,将患有神经根型疼痛且接受过≥1次ESI的个体与未接受ESI的个体(非ESI个体)按年龄、性别和神经根型疼痛诊断月份进行1:10匹配。基于多个数据维度的前500个协变量计算高维倾向得分(HDPS),将ESI个体与非ESI个体按1:1匹配。使用经过验证的ICD - 9/10诊断编码识别骨折。按组计算骨折发生率(IR),并使用Cox回归比较风险比(HR)。通过EDS识别出25062例符合入选标准的ESI患者和221735例非ESI患者。平均年龄为76岁(74%为女性)。在接受ESI治疗的个体中,有2296例骨折,每1000人年的IR为49.1(95%置信区间:47.2 - 51.2)。对于非ESI个体,有11917例骨折,IR为35.2(95%置信区间:34.5 - 35.8)。接受ESI的个体在典型骨质疏松部位发生骨折的风险更高:通过EDS计算的HR为1.39(95%置信区间1.33 - 1.46),通过HDPS计算的HR为1.32(1.12 - 1.54);椎体骨折风险更高,通过EDS计算的HR为1.54(1.45 - 1.64),通过HDPS计算的HR为1.69(1.38 - 2.07)。接受累积ESI剂量较大(1年内≥3次)的患者在随访的前6个月内骨折风险更高。老年个体接触ESI与骨折风险增加相关,这表明ESI可能对骨骼有持久的有害影响。有必要对降低该人群骨折风险的策略进行进一步研究。