POS1151 OPPORTUNISTIC BONE MINERAL DENSITY SCREENING IN PATIENTS UNDERGOING CARDIAC CT SCANS: EFFECT OF USING IMAGES CONTAINING INTRAVENOUS CONTRAST

Abstract

BackgroundOsteoporosis is under-diagnosed worldwide causing increased risk of fractures and death (1). Computed tomography (CT) scans performed on other indications such as coronary artery disease harbor the potential for automatic detection of low volumetric bone mineral density (vBMD) of the vertebrae using quantitative CT (QCT); hence allowing estimation of future fracture risk (2). CT is often performed with intravenous (iv) contrast administration. In 2015, the International Society of Clinical Densitometry stated: “There is insufficient evidence to judge the effect of contrast agents on a classification of low BMD” (3); this position remains. Thus, it is important to assess the effect of contrast enhancement in order to broaden the potential of vBMD screening using routine CT scans.ObjectivesWe aimed to compare thoracic vBMD measurements from CT scans with and without iv contrast enhancement.MethodsThis cross-sectional multicenter sub-study is based on a larger clinical trial, Dan-NICAD-1, from which we randomly selected a cohort of 136 participants. First, a non-contrast scan was performed followed by a contrast-enhanced scan during which 60-90mL of iv contrast was administered (Iomeron, 350 mgI/mL). Mindways QCT Pro software was used to measure BMD values (mg/cm3) and the mean estimate was calculated for each participant (4). American College of Radiology quantitative CT cut-off values for lumbar spine were used to categorize patients into very low (<80mg/cm3), low (80-120mg/cm3), or normal BMD (>120mg/cm3).ResultsIn 136 participants undergoing cardiac CT (Table 1), we found a different mean vBMD before vs. after contrast; 117.5 mg/cm3 [95%CI: 111.6–123.4] vs. 132.1 mg/cm3 [95%CI: 125.1–139.1], p<.0001. The absolute difference was 14.7mg/cm3 [95%CI: 12.3–17.0]; the relative difference, was 12.5% [95% CI: 10.5–14.5]. In total, 8/15 (53%) participants changed from very low BMD to low BMD after contrast administration, and 21 participants (21/63, 33%) changed from low to normal BMD (Figure 1). No participants changed from very low BMD to normal BMD.Table 1.Demographics by vBMD*CharacteristicsAll (n=136)Very low BMD (n=15)Low BMD (n=63)Normal BMD (n=58)Gender M:F89:478:749:1432:26Age, yrs (range)57±9 (40-73)64±6 (48-72)59±8 (44-73)54±8 (40-72)Mean vBMD before contrast, mg/cm92.2±16.168.0±10.598.0±10.1151.5±21.9Mean vBMD after contrast, mg/cm95.2±16.068.6±10.5100.8±10.3159.6±31.6Risk factorsDiabetes mellitus**16079Smoking status**Never6541744Former181710Active5252423Bone dataDXA performed previously**7232Osteoporosis diagnosed previously**1010Family history of osteoporosis**22589Anti-osteoporotic medication**152310* Classifications defined by American College of Radiology and grouped using the non-enhanced CT. Data: number of participants, (range) and mean with standard deviations.** Self-reportedFigure 1.Participants with change in BMD category after contrast administration Figure 1. vBMD measurements before and after contrast administration. 33/136 participants changed BMD category illustrated by the dotted lines (80 mg/cm3; 120 mg/cm3). Black lines: increase in vBMD after contrast (n=29); blue lines: decrease in vBMD after contrast (n=4).ConclusionOur data suggest a significant effect of contrast on clinical vBMD measurements; thus, this should be adjusted for before using contrast-enhanced cardiac CT for opportunistic vBMD screening. This urges further studies on the effect of scan protocols on the contrast-enhanced increase in BMD.References[1]M. S. Nanes et al., Seminars in nuclear medicine44, 439-450 (2014).[2]J. Therkildsen et al., Radiology296, 499-508 (2020).[3]K. Engelke et al., Journal of clinical densitometry18, 393-407 (2015).[4]J. Therkildsen et al., Journal of Clinical Densitometry23, (2018).AcknowledgementsThe Danish Osteoporosis Foundation, The Danish Council for Independent Research (DFF–7025–00103), the Danish Heart foundation (15-R99-A5837–22920), the Hede Nielsen Foundation, Acarix A/S (unrestricted grant) and Mrs. Lily Benthine Lunds Foundation of 1.6. 1978 supported this project. The authors would like to thank all study participants and the clinical staff involved in this project.Disclosure of InterestsAndia Cheneymann: None declared, Josephine Therkildsen: None declared, Simon Winther Grant/research support from: Disclosed an unrestricted grant from Acarix A/S., Louise Nissen: None declared, Jesper Thygesen: None declared, Bente Langdahl Consultant of: Worked as a consultant for Amgen, UCB, Gedeon-Richter, Eli Lilly and Gedeon., Grant/research support from: Received honorariums from Amgen, UCB, Eli Lilly, Gedeon-Richter and Astellas. Received financial grants from Amgen and the Novo Nordic Foundation., Ellen-Margrethe Hauge Consultant of: Received honorariums and/or consulting fees from AbbVie, Sanofi, Sobi, and SynACT Pharma., Grant/research support from: Research grants to Aarhus University Hospital from Danish Regions Medicine Grants, Danish Rheumatism Association, Roche, Novartis, and Novo Nordic Foundation., Morten Böttcher Consultant of: Advisory board participation for NOVO Nordisk, Astra-Zeneca, Pfizer, Boeringer Ingelheim, Bayer, Sanofi, Novartis and Acarix.