Screening of Bone Density at CT: An Overlooked Opportunity
2019; Radiological Society of North America; Volume: 291; Issue: 2 Linguagem: Inglês
10.1148/radiol.2019190434
ISSN1527-1315
Autores Tópico(s)Advanced X-ray and CT Imaging
ResumoHomeRadiologyVol. 291, No. 2 PreviousNext Reviews and CommentaryFree AccessEditorialScreening of Bone Density at CT: An Overlooked OpportunityAndrew D. Smith Andrew D. Smith Author AffiliationsFrom the Department of Radiology, University of Alabama at Birmingham, 619 19th St S, JTN 452, Birmingham AL 35249-6830.Address correspondence to the author (e-mail: [email protected]).Andrew D. Smith Published Online:Mar 26 2019https://doi.org/10.1148/radiol.2019190434MoreSectionsPDF ToolsImage ViewerAdd to favoritesCiteTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinked In See also the article by Jang et al in this issue.IntroductionOsteoporosis is a silent disease characterized by loss of bone and higher risk of fracture (1). Current screening efforts for detecting low bone density and osteoporosis are underused; more than half of insufficiency fractures occur in individuals who have never been screened. There are substantial economic costs, morbidity, and mortality associated with insufficiency fractures caused by low bone density and osteoporosis (1,2).More than 50 000 000 CT examinations of the chest and/or abdomen are performed annually in the United States, and many more are performed worldwide. In general, CT examinations that include images of the spine contain unused and hidden bone density information (3). Opportunistic bone density screening is the use of routine CT images to extract bone density information to help detect low bone density and osteoporosis. Evaluation is most common in the spine, and CT has the advantage of avoiding confounders like body size, atherosclerotic calcifications, and degenerative changes of the spine that are problematic for dual-energy x-ray absorptiometry. Opportunistic bone density screening requires no additional patient time or cost, scanner equipment, or radiation exposure.In my experience working with radiology residents, fellows, and faculty at different institutions and across different subspecialties of radiology, the most common method for evaluating bone density of the spine is to subjectively grade the severity of disease by a visual analysis of the gray-scale CT images, often performed with set image windowing. However, subjective visual assessment of spinal bone density at CT does not provide quantitative information, and has poor accuracy (accuracy, 58%) and poor interobserver agreement (intraclass correlation coefficient, 0.20) (4). This suggests that subjective visual assessment of spinal bone density on CT images should not be performed for the purpose of opportunistic bone density screening.In 2013, Pickhardt et al (5) introduced a simple method for performing opportunistic bone density screening in which an ovoid region of interest is placed on a single axial CT image in the anterior trabecular portion of the L1 vertebral body (avoiding the venous plexus) to quantify CT attenuation in Hounsfield units as a surrogate for bone density. Note that L1 is commonly included at both chest and abdominal CT examinations and is relatively spared from degenerative changes that affect the lower lumbar spine. This simple quantitative method is correlated with measured bone density at dual energy x-ray absorptiometry and quantitative CT, externally validated, and works well when applied to reconstructed sagittal CT images to more easily depict insufficiency fractures and avoid measurement of sclerosis related to degenerative changes (4–7).Whereas manual measurement of L1 trabecular bone attenuation takes seconds to perform and has high accuracy and interobserver agreement for assessing spinal bone density, it is uncommonly performed in clinical practice. Radiologists must remember to make the measurement and should not use subjective visual assessments as a pretest guide because subjective visual assessments have poor accuracy and interobserver agreement. Furthermore, it is unclear which attenuation threshold should be used in a screening setting and how L1 bone attenuation changes with age.In this issue of Radiology, Jang and colleagues (8) conducted a large retrospective single-center study to establish normal reference ranges for L1 trabecular attenuation values across all adult ages to measure bone density on routine chest and abdominal CT performed for other clinical indications. Their final cohort included more than 20 000 body CT examinations in women and men aged 18–100 years, evaluated between 2000 and 2018. All CT examinations were performed by using a variety of multidetector CT scanners from a single vendor (GE Healthcare, Waukesha, Wis) at a constant peak voltage of 120 kV with variable tube current values (in milliamperes). They included CT with (21% of CT examinations in the study) or without intravenous contrast agent and with a broad range of image noise (noise index range, 14–60), which extended the applicability of their study. The authors measured L1 trabecular bone attenuation on a single axial CT image by using a manual method (45%) and a fully automated tool. The fully automated method was adapted from previous work and had a success rate greater than 99%.In their study, Jang et al (8) found that age was the dominant determinate of Hounsfield unit decrease and that mean L1 Hounsfield unit values decreased linearly, with age at a rate of 2.5 HU per year (R2 = 0.99). The decline in mean L1 Hounsfield unit values in women was more rapid after menopause. The mean L1 Hounsfield unit values in women were higher than in men until age 54 years, after which both groups had similar L1 Hounsfield unit values. This is particularly striking because most current efforts to detect osteoporosis by traditional active screening methods are directed at women who are asymptomatic or men with known risk factors. Whereas women tend to have a higher rate of insufficiency fractures across all bone locations, the rate of spinal insufficiency fractures between women and men is surprisingly similar in patients older than 50 years, which further substantiates the study's (9) results (10). Also, surprisingly, intravenous contrast agent had a negligible effect on L1 attenuation values in patients older than 40 years. This indicates that opportunistic bone density by measuring L1 trabecular attenuation is broadly applicable to most CT examinations of the chest and/or abdomen.The study by Jang et al (8) also showed that a fully automated algorithm for CT-based opportunistic bone density screening is feasible in a large study cohort. Implementation of a fully automated method for opportunistic CT screening for osteoporosis has the potential to identify a substantial number of patients with osteoporosis who otherwise would not be screened by using current guidelines. However, in a small subanalysis, the automated method consistently provided higher L1 Hounsfield unit values than did the manual method, specifically leading to attenuation values that were on average 21 HU higher for the automated method. The authors showed that this difference was because of the location of the region of interest in the L1 vertebral body. For the automated method, the region of interest was placed in the center of the spine with respect to craniocaudal location. However, for the manual method, the region of interest was placed off midline in a more superior location to avoid the natural horizontal sclerotic central aspect of the vertebral body. The authors indicate that a future version of the fully automated method will use a similar approach to their manual method.With this study, Jang et al (8) advanced the science of opportunistic bone density screening. The strengths of their study include the large sample size, broad age range, wide range of CT image acquisition parameters, and development of a robust fully automated and quantitative method to assess L1 vertebral body attenuation. The fully automated algorithm is of substantial interest and has potential for future research. It allows for replication of results in additional patient populations, facilitates important follow-up studies associating L1 attenuation with outcomes, and provides optimization of the technique and threshold for measuring L1 trabecular attenuation while extending the application beyond research and into clinical practice.The study had several limitations. The single-center study was retrospective in design and had uneven distributions by patient age, sex, use of intravenous contrast agent, and method of measurement (ie, manual vs automated). The population characteristics are typical of a radiology practice in the midwestern United States. All imaging was performed by using CT scanners from a single vendor (though results are likely applicable to other scanner types because of limited variability in Hounsfield unit measurements across different scanners) (10). The authors did not correlate with a reference standard method for assessment of bone density or outcomes such as current or future insufficiency fractures, though these are logical next steps. There was a consistent difference between the manual and automated methods for measuring L1 attenuation, which will likely be eliminated in a future version of the automated method. Finally, the L1 Hounsfield unit threshold for triggering a change in patient care remains unclear.In conclusion, Jang and colleagues (8) substantiated a fully automated technique for measuring L1 trabecular attenuation on routine CT images and provided normative values of L1 trabecular attenuation across a full age range of adults. They found a slow and steady decline of L1 Hounsfield unit values with age. They also showed that L1 Hounsfield unit values were similar between men and women older than 54 years. Surprisingly, the effect of intravenous contrast on L1 attenuation measurements was negligible in patients older than 40 years, broadening the applicability of opportunistic bone density screening. In current clinical practice, radiologists manually measure L1 trabecular attenuation on axial or sagittal CT images with or without intravenous contrast enhancement. This manual approach to opportunistic bone density screening may be replaced by an automated method in the future.Disclosures of Conflicts of Interest: A.D.S. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: disclosed money paid to institution for grant from GE; disclosed patent for Color Enhanced Detection, which is related to opportunistic bone density screening by using color CT images. Other relationships: disclosed no relevant relationships.References1. U.S. Preventive Services Task Force. Screening for osteoporosis: U.S. preventive services task force recommendation statement. Ann Intern Med 2011;154(5):356–364. Crossref, Medline, Google Scholar2. Tosteson AN, Melton LJ 3rd, Dawson-Hughes B, et al. Cost-effective osteoporosis treatment thresholds: the United States perspective. Osteoporos Int 2008;19(4):437–447. Crossref, Medline, Google Scholar3. IMV. IMV 2017 CT Market Outlook Report. IMV Medical Information Division; Des Plaines, IL: 2017. https://www.imvinfo.com/index.aspx?sec=def&sub=dis&pag=dis&ItemID=200081. Accessed March 21, 2018 Google Scholar4. Smith A, Khan M, Varney E, et al. Opportunistic bone density screening for the abdominal radiologist using colored CT images: a pilot retrospective study. Abdom Radiol (NY) 2019;44(2):775–782. Crossref, Medline, Google Scholar5. Pickhardt PJ, Pooler BD, Lauder T, del Rio AM, Bruce RJ, Binkley N. Opportunistic screening for osteoporosis using abdominal computed tomography scans obtained for other indications. Ann Intern Med 2013;158(8):588–595. Crossref, Medline, Google Scholar6. Lee SJ, Binkley N, Lubner MG, Bruce RJ, Ziemlewicz TJ, Pickhardt PJ. Opportunistic screening for osteoporosis using the sagittal reconstruction from routine abdominal CT for combined assessment of vertebral fractures and density. Osteoporos Int 2016;27(3):1131–1136. Crossref, Medline, Google Scholar7. Buckens CF, Dijkhuis G, de Keizer B, Verhaar HJ, de Jong PA. Opportunistic screening for osteoporosis on routine computed tomography? An external validation study. Eur Radiol 2015;25(7):2074–2079. Crossref, Medline, Google Scholar8. Jang S, Graffy PM, Ziemlewicz TJ, Lee SJ, Summers RM, Pickhardt PJ. Opportunistic osteoporosis screening at routine abdominal and thoracic CT: normative L1 trabecular attenuation values in more than 20 000 adults. Radiology 2019;291:360–367. Link, Google Scholar9. Watts NB, Adler RA, Bilezikian JP, et al. Osteoporosis in men: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2012;97(6):1802–1822. Crossref, Medline, Google Scholar10. Engelke K, Lang T, Khosla S, et al. Clinical use of quantitative computed tomography-based advanced techniques in the management of osteoporosis in adults: the 2015 ISCD official positions-Part III. J Clin Densitom 2015;18(3):393–407. Crossref, Medline, Google ScholarArticle HistoryReceived: Feb 25 2019Revision requested: Mar 4 2019Revision received: Mar 4 2019Accepted: Mar 4 2019Published online: Mar 26 2019Published in print: May 2019 FiguresReferencesRelatedDetailsCited ByLumbar Bone Mineral Density Estimation From Chest X-Ray Images: Anatomy-Aware Attentive Multi-ROI ModelingFakaiWang, KangZheng, LeLu, JingXiao, MinWu, Chang-FuKuo, ShunMiao2023 | IEEE Transactions on Medical Imaging, Vol. 42, No. 1Opportunistic measures of bone mineral density at multiple skeletal sites during whole-body CT in polytrauma patientsZvonimirKutleša, IvanOrdulj, IvaPerić, KristianJerković, DinoPoljak, VladoGavrilović, VesnaČapkun, ŠimeDevčić, DanijelaBudimir Mršić2023 | Osteoporosis InternationalMulti-View Computed Tomography Network for Osteoporosis ClassificationDong HwanHwang, So HyeonBak, Tae-JunHa, YoonKim, Woo JinKim, Hyun-SooChoi2023 | IEEE Access, Vol. 11Opportunistic osteoporosis screening using chest CT with artificial intelligenceJinrongYang, ManLiao, YaolingWang, LeqingChen, LinfengHe, YingyingJi, YaoXiao, YichenLu, WenliangFan, ZhuangNie, RuiyunWang, BenlingQi, FanYang2022 | Osteoporosis International, Vol. 33, No. 12Screening for osteoporosis using L1 vertebral density on abdominal CT in an Australian populationH.Abbouchie, N.Raju, A.Lamanna, C.Chiang, N.Kutaiba2022 | Clinical Radiology, Vol. 77, No. 7Imaging Evaluation of Inflammatory Bowel Disease ComplicationsDavid J.Bartlett, Guilherme PiovezaniRamos, Joel G.Fletcher, David H.Bruining2022 | Gastrointestinal Endoscopy Clinics of North America, Vol. 32, No. 4Identification of osteoporosis using ensemble deep learning model with panoramic radiographs and clinical covariatesShintaroSukegawa, AiFujimura, AkiraTaguchi, NorioYamamoto, AkiraKitamura, RyosukeGoto, KeisukeNakano, KiyofumiTakabatake, HotakaKawai, HitoshiNagatsuka, YoshihikoFuruki2022 | Scientific Reports, Vol. 12, No. 1Combination of Quantitative Computed Tomography and Blood Biochemistry for Evaluating the Relationship Between Nutrition and Bone Mineral Density in Patients on Maintenance HemodialysisShuoWang, WenjuanTang, JiajunZhou, FengYou, FengLei, YingyingYang, XiaojuanZha, YunfengZhou2022 | Journal of Renal Nutrition, Vol. 32, No. 6CORR Insights®: Poor Bone Quality, Multilevel Surgery, and Narrow and Tall Cages Are Associated with Intraoperative Endplate Injuries and Late-onset Cage Subsidence in Lateral Lumbar Interbody Fusion: A Systematic ReviewSergio A.Mendoza-Lattes2022 | Clinical Orthopaedics & Related Research, Vol. 480, No. 1Cetaceans Humerus Radiodensity by CT: A Useful Technique Differentiating between Species, Ecophysiology, and AgeFrancesco Maria AchilleConsoli, YaraBernaldo de Quirós, ManuelArbelo, StefaniaFulle, MarcoMarchisio, MarioEncinoso, AntonioFernandez, Miguel A.Rivero2022 | Animals, Vol. 12, No. 14Lecture Notes in Computer ScienceFakaiWang, KangZheng, YiruiWang, XiaoyunZhou, LeLu, JingXiao, MinWu, Chang-FuKuo, ShunMiao2021 | , Vol. 12928Prospective validation of a rapid CT-based bone mineral density screening method using colored spinal imagesElliotVarney, AsserAbou Elkassem, MajidKhan, EllenParker, ToddNichols, DavidJoyner, Seth T.Lirette, CandaceHoward-Claudio, Andrew D.Smith2021 | Abdominal Radiology, Vol. 46, No. 4Opportunistic diagnosis of osteoporosis, fragile bone strength and vertebral fractures from routine CT scans; a review of approved technology systems and pathways to implementationVeenaAggarwal, ChristinaMaslen, Richard L.Abel, PinakiBhattacharya, Paul ABromiley, Emma M.Clark, Juliet E.Compston, NicolaCrabtree, Jennifer S.Gregory, Eleni P.Kariki, Nicholas C.Harvey, Kate A.Ward, Kenneth E. S.Poole2021 | Therapeutic Advances in Musculoskeletal Disease, Vol. 13Risk of vertebral fractures: evaluation on vertebral trabecular attenuation value and hydroxyapatite concentration in patients by chest spectral CTQiangMa, XinmengHou, XiaoyueCheng, YuangangYou, ZhenghanYang, DaqingMa, ZhenchangWang2021 | The British Journal of Radiology, Vol. 94, No. 1119Automatic opportunistic osteoporosis screening using low-dose chest computed tomography scans obtained for lung cancer screeningYalingPan, DejunShi, HanqiWang, TongtongChen, DeqiCui, XiaoguangCheng, YongLu2020 | European Radiology, Vol. 30, No. 7Adjustment of DXA BMD measurements for anthropometric factors and its impact on the diagnosis of osteoporosisLingWang, LimeiRan, XiaojuanZha, KaipingZhao, YingyingYang, QingShuang, YandongLiu, KarenHind, XiaoguangCheng, Glen M.Blake2020 | Archives of Osteoporosis, Vol. 15, No. 1Utilizing Artificial Intelligence to Determine Bone Mineral Density Via Chest Computed TomographyRock H.Savage, Marlyvan Assen, Simon S.Martin, PooyanSahbaee, Lewis P.Griffith, DanteGiovagnoli, Jonathan I.Sperl, ChristianHopfgartner, RainerKärgel, U. JosephSchoepf2020 | Journal of Thoracic Imaging, Vol. 35, No. Supplement 1Automatic Measurement of Subregional Vertebral Bone Mineral Density via Deep Learning of Quantitative Computed Tomography ImagesChentianLi, ChiMa, XianglongZhuo, WeiWang, LiLi, Wing-YukIp, BingLi, TaoLi, SongjianLi, FengZhu, William W.Lu2020 | International Journal of Orthopedics and Rehabilitation, Vol. 7Dépistage du cancer broncho-pulmonaire: les « big 3 » et les autresP.-Y.Brillet2019 | Revue des Maladies Respiratoires Actualités, Vol. 11, No. 2Accompanying This ArticleOpportunistic Osteoporosis Screening at Routine Abdominal and Thoracic CT: Normative L1 Trabecular Attenuation Values in More than 20 000 AdultsMar 26 2019RadiologyRecommended Articles Opportunistic Osteoporosis Screening at Routine Abdominal and Thoracic CT: Normative L1 Trabecular Attenuation Values in More than 20 000 AdultsRadiology2019Volume: 291Issue: 2pp. 360-367Improved CT-based Osteoporosis Assessment with a Fully Automated Deep Learning ToolRadiology: Artificial Intelligence2022Volume: 4Issue: 5Automated Screening for Future Osteoporotic Fractures on Abdominal CT: Opportunistic or an Outstanding Opportunity?Radiology2020Volume: 297Issue: 1pp. 73-74Opportunistic Screening at Abdominal CT: Use of Automated Body Composition Biomarkers for Added Cardiometabolic ValueRadioGraphics2021Volume: 41Issue: 2pp. 524-542Automated Abdominal CT Imaging Biomarkers for Opportunistic Prediction of Future Major Osteoporotic Fractures in Asymptomatic AdultsRadiology2020Volume: 297Issue: 1pp. 64-72See More RSNA Education Exhibits Value-Added "Opportunistic" CT: State-of-the-Art Insights into Bone, Muscle, and FatDigital Posters2019Automated CT Biomarkers for Opportunistic Cardiometabolic Screening: Adding Value Beyond the Clinical Indication Digital Posters2019Opportunistic CT Screening for Assessment of Vertebral Fracture RiskDigital Posters2020 RSNA Case Collection Sacral Insufficiency FracturesRSNA Case Collection2021Atypical insufficiency fracture RSNA Case Collection2020Postradiation sacral insufficiency fractureRSNA Case Collection2021 Vol. 291, No. 2 Metrics Altmetric Score PDF download
Referência(s)