医学
股骨颈
骨矿物
骨质疏松症
弗雷克斯
骨量减少
核医学
置信区间
定量计算机断层扫描
骨髓
腰椎
骨密度
放射科
泌尿科
内科学
骨质疏松性骨折
作者
Fernando Uliana Kay,Vinh Ho,Edmund B. Dosunmu,Avneesh Chhabra,Keenan Brown,Xinhui Duan,Orhan K. Öz
标识
DOI:10.1097/rlu.0000000000003416
摘要
Purpose We assessed the prevalence of low bone mineral density (BMD) in oncologic patients undergoing 18 F-FDG PET/CT. Patients and Methods This is a retrospective analysis of 100 patients who underwent 18 F-FDG PET/CT at a single center from October 2015 till May 2016. Quantitative CT (QCT) was used to assess BMD at the lumbar spine (BMD QCT ) and femoral necks (BMD CTXA ). SUV max was used to evaluate metabolic activity of the bone marrow. Risk of osteoporosis-related fractures was calculated with femoral neck BMD CTXA and the FRAX algorithm, which was compared against measurements of CT attenuation of the trabecular bone at L1 (L1 HU ). Results Osteoporosis and osteopenia were respectively present in 16% and 46% of patients 50 years and older. Bone marrow SUV max was correlated with BMD at the lumbar spine (ρ = 0.36, P < 0.001). Increased age and low marrow SUV max were associated with low BMD QCT at the lumbar spine (both P < 0.001), whereas increased age, female sex, and low marrow SUV max were associated with low BMD CTXA at the femoral necks ( P < 0.001, P < 0.001, P = 0.01, respectively). L1 HU had an area under the curve of 0.95 (95% confidence interval [CI], 0.90–0.99) for detecting increased risk for osteoporosis-related fracture, with best threshold of 125.8 HU (95% CI, 115.7–144.9) yielding sensitivity of 100% (95% CI, 0.92–1.00), specificity of 0.90 (95% CI, 0.76–0.97), and accuracy of 0.91 (95% CI, 0.79–0.97). Conclusions Low BMD is frequent in oncologic patients undergoing 18 F-FDG PET/CT. Decreased 18 F-FDG avidity of the bone marrow correlates with decreased BMD, validating the link between osteoporosis and bone marrow fat. L1 HU could be a simple and accurate approach for detecting patients at risk for osteoporosis-related fractures using PET/CTdata.
科研通智能强力驱动
Strongly Powered by AbleSci AI