The effect of obesity, diabetes, and epidural steroid injection on regional volumetric bone mineral density measured by quantitative computed tomography in the lumbosacral spine

• Published in: European spine journal 2021, Vol.30 (1), p.13-21
• Main Authors: Okano, Ichiro, Salzmann, Stephan N., Jones, Conor, Reisener, Marie-Jacqueline, Ortiz Miller, Courtney, Shirahata, Toshiyuki, Shue, Jennifer, Carrino, John A., Sama, Andrew A., Cammisa, Frank P., Girardi, Federico P., Hughes, Alexander P.
• Format: Article
• Language: English
• Subjects: Axial skeleton; Body mass index; Bone density; Bone mass; Bone mineral density; Computed tomography; Diabetes; Diabetes mellitus; Epidural; Glucose tolerance; Intolerance; Medicine; Medicine & Public Health; Neurosurgery; Obesity; Original Article; Patients; Sacrum; Skeleton; Spine (lumbar); Steroid hormones; Surgical Orthopedics
• ISSN: 0940-6719; 1432-0932
• DOI: 10.1007/s00586-020-06610-5

Purpose High body mass index (BMI) is positively correlated with bone mineral density (BMD) in healthy adults; however, the effect of BMI on regional segmental BMDs in the axial skeleton is unclear. In addition, obese patients often have glucose intolerance and patients with lumbar spine pathology commonly have a history of epidural steroid injections (ESIs). The purpose of this study is to evaluate the effect of these patient factors on regional differences in BMD measured by quantitative computed tomography (QCT) in a lumbar fusion patient cohort. Methods The data were obtained from a database comprised of clinical and preoperative CT data from 296 patients who underwent primary posterior lumbar spinal fusion from 2014 to 2017. QCT-vBMDs of L1 to L5, S1 body, and sacral alae were measured. Multivariate linear regression analyses were performed with setting vBMDs as the response variables. As explanatory variables, age, sex, race, current smoking, categorized BMI, diabetes, and ESI were chosen a priori. Results A total of 260 patients were included in the final analysis. Multivariate analyses demonstrated that obese and morbidly obese patients had significantly higher vBMD in the sacral alae (SA). Diabetes showed independent positive associations with vBMDs in L1, L2, and the SA. Additionally, patients with an ESI history demonstrated significantly lower vBMD in the SA. Conclusions Our results demonstrate that obesity, diabetes, and epidural steroids affected vBMD differently by lumbosacral spine region. The vBMD of the SA appeared to be more sensitive to various patient factors than other lumbar regions.