Research Presentation Session: Musculoskeletal
RPS 1810 - Radiologic insights into skeletal ageing: osteoporosis and frailty
March 7, 09:30 - 11:00 CET
6 min
Ultrasound Assessment of Therapeutic Water-based Exercise in a Frail Population - A Randomized Controlled Trial
Rebeca Mirón Mombiela, Herlev / Denmark
Author Block: J. S. Cárdenas Herran1, J. Vucetic1, M. Inglés1, C. Borras1, R. Mirón Mombiela2; 1Valencia/ES, 2Herlev/DK
Purpose: Frailty, an age-related functional capacity loss, constitutes a significant health concern. High-impact exercises are effective in improving frailty, while the effect of therapeutic water-based exercise remains poorly studied. We aimed to investigate the impact of a therapeutic water-based exercise program on ultrasound parameters related to frailty and functional capacity in older adults.
Methods or Background: This was a randomized controlled trial set in a tertiary-care hospital and a private swimming pool. Thirty-five older adults, median age 72 (+/-10), were divided into two groups: i) therapeutic water-based exercise group (WEG) (n=19) and ii) control group (CG) (n=16). WEG underwent a twelve-week water-based exercise intervention supervised by a physiotherapist. CG did not receive any intervention. Frailty status, quantitative muscle ultrasound parameters, and functional outcomes were assessed before and after the intervention. Possible differences between quantitative variables were determined using a two-way ANOVA model with interaction. Minimum detectable change (MDC) between the quantitative variables was also determined.
Results or Findings: Frailty status, sex, physical function, and comorbidities were homogeneously distributed. There was a statistically significant effect on the ultrasound variables: Echo-Intensity (p = 0.013), Muscle-Thickness (p = 0.011), and Subcutaneous-Fat-Thickness (p = 0.010) with respect to treatment and time. We also observed differences in the values of Echo-Intensity in the treatment group after the intervention exceeded the MDC values, thus demonstrating a change that can be attributed to the intervention.
Conclusion: A therapeutic water-based exercise program improves parameters related to frailty and physical condition in older adults, likely by enhancing muscle quality or Echo-Intensity as measured by ultrasound. This suggests that ultrasound can be used as a tool to measure the effect of exercise-based treatments in frail older adults.
Limitations: The study was not double-blind, and the sample size was small.
Funding for this study: Not Applicable.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The protocol was approved by the Research Ethics Committee of the University of Valencia (num. H1520499833822) and by the Research Ethics Committee of the Consorcio Hospital General Universitario de Valencia. The study was also prospectively registered at www.clinicaltrials.gov as SWIMFRAIL (NCT03955302).
Purpose: Frailty, an age-related functional capacity loss, constitutes a significant health concern. High-impact exercises are effective in improving frailty, while the effect of therapeutic water-based exercise remains poorly studied. We aimed to investigate the impact of a therapeutic water-based exercise program on ultrasound parameters related to frailty and functional capacity in older adults.
Methods or Background: This was a randomized controlled trial set in a tertiary-care hospital and a private swimming pool. Thirty-five older adults, median age 72 (+/-10), were divided into two groups: i) therapeutic water-based exercise group (WEG) (n=19) and ii) control group (CG) (n=16). WEG underwent a twelve-week water-based exercise intervention supervised by a physiotherapist. CG did not receive any intervention. Frailty status, quantitative muscle ultrasound parameters, and functional outcomes were assessed before and after the intervention. Possible differences between quantitative variables were determined using a two-way ANOVA model with interaction. Minimum detectable change (MDC) between the quantitative variables was also determined.
Results or Findings: Frailty status, sex, physical function, and comorbidities were homogeneously distributed. There was a statistically significant effect on the ultrasound variables: Echo-Intensity (p = 0.013), Muscle-Thickness (p = 0.011), and Subcutaneous-Fat-Thickness (p = 0.010) with respect to treatment and time. We also observed differences in the values of Echo-Intensity in the treatment group after the intervention exceeded the MDC values, thus demonstrating a change that can be attributed to the intervention.
Conclusion: A therapeutic water-based exercise program improves parameters related to frailty and physical condition in older adults, likely by enhancing muscle quality or Echo-Intensity as measured by ultrasound. This suggests that ultrasound can be used as a tool to measure the effect of exercise-based treatments in frail older adults.
Limitations: The study was not double-blind, and the sample size was small.
Funding for this study: Not Applicable.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The protocol was approved by the Research Ethics Committee of the University of Valencia (num. H1520499833822) and by the Research Ethics Committee of the Consorcio Hospital General Universitario de Valencia. The study was also prospectively registered at www.clinicaltrials.gov as SWIMFRAIL (NCT03955302).
6 min
Association between spinal deformity and pulmonary function in patients with interstitial lung disease
Angela Ventura, Verona / Italy
Author Block: L. Pinali, A. Fassio, A. Ventura, L. Carobene, C. Micheletto, D. Gatti, M. Rossini, G. Puppini, S. Baltieri; Verona/IT
Purpose: Vertebral fractures and spinal deformities are common in patients with interstitial lung disease (ILD). We aimed to explore impact of vertebral damage on pulmonary restriction evaluating the association between the Spinal Deformity Index (SDI) and lung volumes in patients with ILD.
Methods or Background: This retrospective study included 200 patients with ILD: 76 idiopathic pulmonary fibrosis (IPF), 65 systemic sclerosis-associated ILD (SSc-ILD), 31 idiopathic inflammatory myopathy-associated ILD (IIM-ILD) and 28 other ILDs. SDI was assessed on multiplanar reconstructions of CT thoracic examinations. Pulmonary function tests included percent predicted forced vital capacity (ppFVC), total lung capacity (ppTLC), absolute FVC and TLC volumes. Multivariable linear regression models were used to evaluate the association between SDI and lung function parameters, adjusted for age, sex, ILD subtype and presence of diffuse idiopathic skeletal hyperostosis or osteoarthritis (DISH/OA).
Results or Findings: In adjusted models, higher SDI was significantly associated with lower ppFVC (2.9% per SDI unit, p<0.01) and ppTLC (2.7%, p<0.01). These associations persisted across ILD subtypes. SDI was inversely associated with absolute lung volumes: each unit increase in SDI corresponded to a 95.64 ml reduction in FVC (P = 0.02) and 199.46 ml in TLC (p<0.05 for both). No significant relationship was observed between SDI and ppDLCO.
Conclusion: Spinal deformity, as quantified by SDI, is independently associated with reduced lung volumes in ILD patients, irrespective of ILD subtype. Spinal integrity assessment may be a valuable addition to the interpretation of restrictive ventilatory defects, especially in elderly and osteoporotic populations.
Limitations: No limitations were identified.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:
Purpose: Vertebral fractures and spinal deformities are common in patients with interstitial lung disease (ILD). We aimed to explore impact of vertebral damage on pulmonary restriction evaluating the association between the Spinal Deformity Index (SDI) and lung volumes in patients with ILD.
Methods or Background: This retrospective study included 200 patients with ILD: 76 idiopathic pulmonary fibrosis (IPF), 65 systemic sclerosis-associated ILD (SSc-ILD), 31 idiopathic inflammatory myopathy-associated ILD (IIM-ILD) and 28 other ILDs. SDI was assessed on multiplanar reconstructions of CT thoracic examinations. Pulmonary function tests included percent predicted forced vital capacity (ppFVC), total lung capacity (ppTLC), absolute FVC and TLC volumes. Multivariable linear regression models were used to evaluate the association between SDI and lung function parameters, adjusted for age, sex, ILD subtype and presence of diffuse idiopathic skeletal hyperostosis or osteoarthritis (DISH/OA).
Results or Findings: In adjusted models, higher SDI was significantly associated with lower ppFVC (2.9% per SDI unit, p<0.01) and ppTLC (2.7%, p<0.01). These associations persisted across ILD subtypes. SDI was inversely associated with absolute lung volumes: each unit increase in SDI corresponded to a 95.64 ml reduction in FVC (P = 0.02) and 199.46 ml in TLC (p<0.05 for both). No significant relationship was observed between SDI and ppDLCO.
Conclusion: Spinal deformity, as quantified by SDI, is independently associated with reduced lung volumes in ILD patients, irrespective of ILD subtype. Spinal integrity assessment may be a valuable addition to the interpretation of restrictive ventilatory defects, especially in elderly and osteoporotic populations.
Limitations: No limitations were identified.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:
6 min
Intravenous Contrast Effects on Opportunistic Osteoporosis Assessment: Impact on Absolute Values, T-Scores, and Classifications in Single- and Dual-Energy CT
Simon Bernatz, Frankfurt am Main / Germany
Author Block: S. Bernatz, J. Gotta, V. Koch, S. Mahmoudi, R. Hammerstingl, T. Vogl, L. D. Grünewald; Frankfurt/DE
Purpose: This study evaluates the effect of contrast agents on Hounsfield measurements, T-scores, and Z-scores, assessing their impact on diagnostic accuracy to reduce misclassification and optimize CT-based BMD assessment.
Methods or Background: A retrospective analysis of 597 patients (median age:66 years, 157 females, 440 males) was performed using dual-energy CT (DECT) scans of the abdomen and chest. All patients underwent non-contrast, arterial, and venous phase CT. Automated segmentation (nnU-Net) delineated L1 and L1-L4 trabecular bone, validated by two radiologists. T-scores were calculated according to DEXA-equivalent guidelines.
Results or Findings: Based on non-contrast CT, 35% were diagnosed with osteoporosis, 46% with osteopenia, and 18% had normal bone status. Median T-score was -2.0 (L1) and -2.1 (L1-L4)(p<0.001). Contrast agents significantly altered BMD values, with median changes of 22.9% (arterial) and 20.1%(venous). The most pronounced changes occurred in patients under 50 years (+99% at L1, p<0.001). In older females, 21% were misclassified as osteopenic instead of osteoporotic (p<0.001).
Conclusion: Contrast agents significantly affect BMD measurements, leading to diagnostic misclassification. This effect should be considered when using CT for osteoporosis diagnosis and treatment planning.
Limitations: It was conducted at a single center, which may limit generalizability across different scanners. Slice thickness was not assessed, despite its known influence on body composition parameters. Bone density was evaluated using T-scores derived from routine CT rather than quantitative CT, while pragmatic and clinically familiar, this approach may lack the precision of QCT. The cohort was imbalanced, with a predominance of male patients and relatively few individuals under 50 years, reflecting the underlying clinical population but restricting subgroup analyses. Finally, only cases with complete agreement between two readers were included; although this ensured high-quality ground truth, a consensus process or third reviewer would have been more robust.
Funding for this study: No funding
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Ethics commitee of the University Hospital Frankfurt
Purpose: This study evaluates the effect of contrast agents on Hounsfield measurements, T-scores, and Z-scores, assessing their impact on diagnostic accuracy to reduce misclassification and optimize CT-based BMD assessment.
Methods or Background: A retrospective analysis of 597 patients (median age:66 years, 157 females, 440 males) was performed using dual-energy CT (DECT) scans of the abdomen and chest. All patients underwent non-contrast, arterial, and venous phase CT. Automated segmentation (nnU-Net) delineated L1 and L1-L4 trabecular bone, validated by two radiologists. T-scores were calculated according to DEXA-equivalent guidelines.
Results or Findings: Based on non-contrast CT, 35% were diagnosed with osteoporosis, 46% with osteopenia, and 18% had normal bone status. Median T-score was -2.0 (L1) and -2.1 (L1-L4)(p<0.001). Contrast agents significantly altered BMD values, with median changes of 22.9% (arterial) and 20.1%(venous). The most pronounced changes occurred in patients under 50 years (+99% at L1, p<0.001). In older females, 21% were misclassified as osteopenic instead of osteoporotic (p<0.001).
Conclusion: Contrast agents significantly affect BMD measurements, leading to diagnostic misclassification. This effect should be considered when using CT for osteoporosis diagnosis and treatment planning.
Limitations: It was conducted at a single center, which may limit generalizability across different scanners. Slice thickness was not assessed, despite its known influence on body composition parameters. Bone density was evaluated using T-scores derived from routine CT rather than quantitative CT, while pragmatic and clinically familiar, this approach may lack the precision of QCT. The cohort was imbalanced, with a predominance of male patients and relatively few individuals under 50 years, reflecting the underlying clinical population but restricting subgroup analyses. Finally, only cases with complete agreement between two readers were included; although this ensured high-quality ground truth, a consensus process or third reviewer would have been more robust.
Funding for this study: No funding
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Ethics commitee of the University Hospital Frankfurt
6 min
Bone health assessment in patients with prostate cancer: the role of REMS technology
Manuela Montatore, Barletta / Italy
Author Block: M. Montatore1, S. Casciaro2, F. R. Contaldo2, F. Conversano2, F. A. Lombardi2, M. Muratore2, P. Pisani2, C. Stomaci2, G. Guglielmi1; 1Foggia/IT, 2Lecce/IT
Purpose: Prostate cancer (PCa) represents one of the most prevalent malignancies among men, and its management often leads to significant complications, particularly concerning bone health. The interaction between prostate cancer and osteoporosis is multifaceted, involving both the disease itself and the therapies employed in its treatment (such as Androgen Deprivation Therapy (ADT)), which accelerates bone loss and increases fracture risk. Given the intricate relationship between PCa, its treatments, and bone health, proactive monitoring and management of osteoporosis in these patients are essential. This study aims to evaluate the impact of PCa and ADT on bone mineral density (BMD) using the non-ionizing REMS (Radiofrequency Echographic Multi-Spectrometry), the ultrasound-based method for diagnosing osteoporosis and predicting fracture risk on spine and femur
Methods or Background: A cohort of 51 Caucasian men with PCa and a healthy control group matchedfor gender, ethnicity, age, and body mass index (BMI), underwent REMS scans on the femoral neck (FN). A t-test to evaluate the differences in the measured BMD values between the two groups was conducted.
Results or Findings: In subjects with PCa and healthy controls, the mean ± standard deviation for age, weight, height, and BMI were73.5±7.3years, 25.3 ± 2.9 kg/m2) and (73.4 ± 7.2 years, 25.3 ± 3.1 kg/m², respectively. Patients with PCa had a significantly reduced BMD, T- and Z-score at the femoral neck compared to the healthy control group, table 1. BMD percentage difference: -3.4% g/cm2
Conclusion: The obtained significant decrease in FN BMD in PCa patients, as measured by REMS, underlines the disease's negative impact on bone health. The accuracy and precision of REMS allows it to early detect a bone damage for better management of PCa patients, especially those at higher risk due to ADT.
Limitations: No limitation were identified
Funding for this study: No funding was received for this study
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:
Purpose: Prostate cancer (PCa) represents one of the most prevalent malignancies among men, and its management often leads to significant complications, particularly concerning bone health. The interaction between prostate cancer and osteoporosis is multifaceted, involving both the disease itself and the therapies employed in its treatment (such as Androgen Deprivation Therapy (ADT)), which accelerates bone loss and increases fracture risk. Given the intricate relationship between PCa, its treatments, and bone health, proactive monitoring and management of osteoporosis in these patients are essential. This study aims to evaluate the impact of PCa and ADT on bone mineral density (BMD) using the non-ionizing REMS (Radiofrequency Echographic Multi-Spectrometry), the ultrasound-based method for diagnosing osteoporosis and predicting fracture risk on spine and femur
Methods or Background: A cohort of 51 Caucasian men with PCa and a healthy control group matchedfor gender, ethnicity, age, and body mass index (BMI), underwent REMS scans on the femoral neck (FN). A t-test to evaluate the differences in the measured BMD values between the two groups was conducted.
Results or Findings: In subjects with PCa and healthy controls, the mean ± standard deviation for age, weight, height, and BMI were73.5±7.3years, 25.3 ± 2.9 kg/m2) and (73.4 ± 7.2 years, 25.3 ± 3.1 kg/m², respectively. Patients with PCa had a significantly reduced BMD, T- and Z-score at the femoral neck compared to the healthy control group, table 1. BMD percentage difference: -3.4% g/cm2
Conclusion: The obtained significant decrease in FN BMD in PCa patients, as measured by REMS, underlines the disease's negative impact on bone health. The accuracy and precision of REMS allows it to early detect a bone damage for better management of PCa patients, especially those at higher risk due to ADT.
Limitations: No limitation were identified
Funding for this study: No funding was received for this study
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:
6 min
Machine learning for the prediction of fragility fractures by bone and body composition parameters: the OsteoLaus 10 years populational cohort
Colin Vendrami, Lausanne / Switzerland
Author Block: C. Vendrami, G. Gatineau, E. Shevroja, E. Gonzalez Rodriguez, O. Lamy, D. Hans; Lausanne/CH
Purpose: Recent reviews have highlighted the potential value of machine learning (ML) in improving fragility fracture prediction. We aim to analyse the prediction of fragility fractures using regional and total body dual X-ray absorptiometry (DXA) assessments using a ML pipeline.
Methods or Background: A total of 1475 Swiss postmenopausal women (mean age 63.3±7.1 years, body mass index 25.4±4.3kg/m2) were followed for 10 years between 2010 and 2022. Parameters of bone health (from regional hip and spine DXA: 34 variables) and body composition (total body DXA: 65 variables) were assessed by DXA scans at baseline. Vertebral fractures were screened with lateral DXA during follow-up. Other risk factors (15 variables) and fragility fractures were collected from questionnaires at baseline and during follow up, respectively. All datasets were split for training (85%) and testing (15%) with a balanced fragility fractures proportion. Eight ML models were trained with hyperparameters tunning through 5-fold cross-validation to maximize the area under the curve (AUC).
Results or Findings: Using different combination of variables in the final complete case analysis, a total of 590 to 957 participants with 115-221 fragility fractures were included. The combination of all 114 variables in the same dataset (n=590) achieved the best AUC in both the training set (1.00-0.70) and test set (0.79-0.64). Logistic regression demonstrated the best balance in performance metrics: AUC 0.78, specificity 0.82, and sensitivity 0.71. ML models tend toward higher sensitivity and lower specificity.
Conclusion: These findings emphasize the combination of DXA-derived bone and body composition parameters for fragility fracture prediction. Logistic regression produced the most promising results. ML models remain at interest for further analysis in combination with image based analysis.
Limitations: Further studies including comparisons with FRAX®, larger sample sizes and external validation are needed.
Funding for this study: This study was funded by the Swiss National Science Foundation (SNSF 32473B_156978 and 320030_188886).
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The OsteoLaus study received approval from the Institutional Ethics Committee of the University of Lausanne, and all participants signed informed consent (reference 215/09).
Purpose: Recent reviews have highlighted the potential value of machine learning (ML) in improving fragility fracture prediction. We aim to analyse the prediction of fragility fractures using regional and total body dual X-ray absorptiometry (DXA) assessments using a ML pipeline.
Methods or Background: A total of 1475 Swiss postmenopausal women (mean age 63.3±7.1 years, body mass index 25.4±4.3kg/m2) were followed for 10 years between 2010 and 2022. Parameters of bone health (from regional hip and spine DXA: 34 variables) and body composition (total body DXA: 65 variables) were assessed by DXA scans at baseline. Vertebral fractures were screened with lateral DXA during follow-up. Other risk factors (15 variables) and fragility fractures were collected from questionnaires at baseline and during follow up, respectively. All datasets were split for training (85%) and testing (15%) with a balanced fragility fractures proportion. Eight ML models were trained with hyperparameters tunning through 5-fold cross-validation to maximize the area under the curve (AUC).
Results or Findings: Using different combination of variables in the final complete case analysis, a total of 590 to 957 participants with 115-221 fragility fractures were included. The combination of all 114 variables in the same dataset (n=590) achieved the best AUC in both the training set (1.00-0.70) and test set (0.79-0.64). Logistic regression demonstrated the best balance in performance metrics: AUC 0.78, specificity 0.82, and sensitivity 0.71. ML models tend toward higher sensitivity and lower specificity.
Conclusion: These findings emphasize the combination of DXA-derived bone and body composition parameters for fragility fracture prediction. Logistic regression produced the most promising results. ML models remain at interest for further analysis in combination with image based analysis.
Limitations: Further studies including comparisons with FRAX®, larger sample sizes and external validation are needed.
Funding for this study: This study was funded by the Swiss National Science Foundation (SNSF 32473B_156978 and 320030_188886).
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The OsteoLaus study received approval from the Institutional Ethics Committee of the University of Lausanne, and all participants signed informed consent (reference 215/09).
6 min
Prospective assessment of bone mineral density in spectral localizer radiographs from photon counting detector CT: A prospective in-vivo study
Lukas Jakob Moser, Zürich / Switzerland
Author Block: L. J. Moser1, K. Klambauer1, V. Mergen1, M. Eberhard1, T. Nowak2, B. Schmidt2, T. Flohr2, O. Distler3, H. Alkadhi1; 1Zürich/CH, 2Forchheim/DE, 3Zurich/CH
Purpose: To determine in a prospective patient study the accuracy of areal bone mineral density (aBMD) measurements with spectral localizer radiographs obtained with a clinical photon-counting detector (PCD)-CT scanner in comparison with dual-energy x-ray absorptiometry (DXA).
Methods or Background: In this IRB-approved, prospective study, 41 patients (15 females, 26 males; mean age 61.3 years, age range 35 - 78 years) underwent PCD-CT of the abdomen with a spectral localizer radiograph (tube voltage 140 kVp, tube current 30 mA) and DXA within a median of 45 days. Areal bone mineral density (aBMD) values were derived for lumbar vertebrae L1 - L4 from both methods and were compared with linear regression, Pearson correlation, intraclass correlation coefficients (ICCs), and Bland-Altman plots. T-scores were calculated on a patient level and were compared between methods.
Results or Findings: DXA and spectral localizer radiographs showed strong correlation in aBMD measurements (R = 0.97, p < 0.001) and patient level T-scores (R = 0.99, p < 0.001). There was a strong agreement between aBMD from both methods (ICC, 0.96 (95% CI [0.94, 0.97])). Bland-Altman analysis revealed a very small mean difference in aBMD between methods (mean absolute error 0.019 g/cm2) with narrow limits of agreement (-0.083 g/cm2 to 0.121 g/cm2). Similarly, there were small differences in regard to the T-score (mean absolute error 0.156) with narrow limits of agreement (-0.422 to 0.734) between methods. ICCs indicated an excellent agreement between T-scores from DXA and spectral localizer radiographs (ICC, 0.98 (95% CI [0.95, 0.99])).
Conclusion: Our prospective patient study indicates that spectral localizer radiographs obtained with a clinical PCD-CT system enable accurate quantification of the lumbar bone areal mineral density.
Limitations: The limitations of this study are as follows: Small sample size. Manual segmentation process. Prototype software.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Approval by the local EC
Purpose: To determine in a prospective patient study the accuracy of areal bone mineral density (aBMD) measurements with spectral localizer radiographs obtained with a clinical photon-counting detector (PCD)-CT scanner in comparison with dual-energy x-ray absorptiometry (DXA).
Methods or Background: In this IRB-approved, prospective study, 41 patients (15 females, 26 males; mean age 61.3 years, age range 35 - 78 years) underwent PCD-CT of the abdomen with a spectral localizer radiograph (tube voltage 140 kVp, tube current 30 mA) and DXA within a median of 45 days. Areal bone mineral density (aBMD) values were derived for lumbar vertebrae L1 - L4 from both methods and were compared with linear regression, Pearson correlation, intraclass correlation coefficients (ICCs), and Bland-Altman plots. T-scores were calculated on a patient level and were compared between methods.
Results or Findings: DXA and spectral localizer radiographs showed strong correlation in aBMD measurements (R = 0.97, p < 0.001) and patient level T-scores (R = 0.99, p < 0.001). There was a strong agreement between aBMD from both methods (ICC, 0.96 (95% CI [0.94, 0.97])). Bland-Altman analysis revealed a very small mean difference in aBMD between methods (mean absolute error 0.019 g/cm2) with narrow limits of agreement (-0.083 g/cm2 to 0.121 g/cm2). Similarly, there were small differences in regard to the T-score (mean absolute error 0.156) with narrow limits of agreement (-0.422 to 0.734) between methods. ICCs indicated an excellent agreement between T-scores from DXA and spectral localizer radiographs (ICC, 0.98 (95% CI [0.95, 0.99])).
Conclusion: Our prospective patient study indicates that spectral localizer radiographs obtained with a clinical PCD-CT system enable accurate quantification of the lumbar bone areal mineral density.
Limitations: The limitations of this study are as follows: Small sample size. Manual segmentation process. Prototype software.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Approval by the local EC
6 min
Longitudinal Changes of Muscle and Bone Marrow Fat Fraction over 5 years Postpartum in Premenopausal Women with Prior Gestational Diabetes
Yannick Stohldreier, Munich / Germany
Author Block: Y. Stohldreier, U. Ferrari, S. Schläger, O. Dietrich, H. Hermann, E. Pappa, J. Seißler, A. Gersing, N. Hesse; Munich/DE
Purpose: This study investigates the longitudinal associations between transient insulin resistance during gestational diabetes mellitus (GDM), a prediabetic state, and the composition of vertebral bone marrow (VBM) and paraspinal muscles assessed using quantitative magnetic resonance imaging (MRI).
Methods or Background: We included 78 premenopausal women, 46 with a history of GDM (GDM group) and 32 healthy controls (HC) after normoglycemic pregnancy. At the 5-year-follow-up the glucose metabolism remained pathologic (pGM) in 10 women, based on oral glucose tolerance testing. 3T MRI including quantitative spine imaging was acquired at 1 year and 5 years postpartum, from which proton density fat fraction (PDFF) maps were calculated and autochthone muscles (AM), psoas, and VBM of the thoracic (T9-12) and lumbar (L1-L4) spine were assessed.
Results or Findings: At 1-year-follow-up, all women with prior GDM showed a significantly higher PDFF in the VBM (p=0.03) and AM (p=0.04) compared with controls, suggesting increased fat content and reduced quality. At 5-year-follow-up, no significant group differences were observed, indicating a potential reversal of the previous composition alterations. In contrast, after 5 years, the pGM group exhibited higher PDFF values in the PM and AM (both p=0.01), but not in VBM at 5-year-follow-up (p>0.05). Elevated muscle PDFF was associated with significantly increased odds of pGM at 5-year-follow-up after adjusting for age (PDFF AM OR 1.26 [95% CI 1.05, 1.56], p=0.02; PDFF psoas OR 1.49 [95% CI 1.07, 2.24], p=0.03).
Conclusion: Vertebral bone marrow fat fraction was elevated 1 year after GDM and normalized after 5 years postpartum, while persistent impairment of glucose metabolism was associated with increased PDFF, suggesting reduced skeletal muscle quality 5 years postpartum. Therefore, muscle PDFF may be a useful biomarker for monitoring muscle health in patients with metabolic diseases.
Limitations: Retrospective monocentric study.
Funding for this study: This work was funded by the Munich Clinician Scientist Program (MCSP) of the University of Munich (LMU; grant number ACS-10), LMU Klinikum, the German Center for Diabetes Research (DZD), and the Helmholtz Zentrum München
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study was approved by the local institutional review board (Ethics Commission of the Medical Faculty, Ludwig-Maximilians Universität München) and all study participants provided written informed consent prior to their participation in the study, which was conducted in accordance with the declaration of Helsinki.
Purpose: This study investigates the longitudinal associations between transient insulin resistance during gestational diabetes mellitus (GDM), a prediabetic state, and the composition of vertebral bone marrow (VBM) and paraspinal muscles assessed using quantitative magnetic resonance imaging (MRI).
Methods or Background: We included 78 premenopausal women, 46 with a history of GDM (GDM group) and 32 healthy controls (HC) after normoglycemic pregnancy. At the 5-year-follow-up the glucose metabolism remained pathologic (pGM) in 10 women, based on oral glucose tolerance testing. 3T MRI including quantitative spine imaging was acquired at 1 year and 5 years postpartum, from which proton density fat fraction (PDFF) maps were calculated and autochthone muscles (AM), psoas, and VBM of the thoracic (T9-12) and lumbar (L1-L4) spine were assessed.
Results or Findings: At 1-year-follow-up, all women with prior GDM showed a significantly higher PDFF in the VBM (p=0.03) and AM (p=0.04) compared with controls, suggesting increased fat content and reduced quality. At 5-year-follow-up, no significant group differences were observed, indicating a potential reversal of the previous composition alterations. In contrast, after 5 years, the pGM group exhibited higher PDFF values in the PM and AM (both p=0.01), but not in VBM at 5-year-follow-up (p>0.05). Elevated muscle PDFF was associated with significantly increased odds of pGM at 5-year-follow-up after adjusting for age (PDFF AM OR 1.26 [95% CI 1.05, 1.56], p=0.02; PDFF psoas OR 1.49 [95% CI 1.07, 2.24], p=0.03).
Conclusion: Vertebral bone marrow fat fraction was elevated 1 year after GDM and normalized after 5 years postpartum, while persistent impairment of glucose metabolism was associated with increased PDFF, suggesting reduced skeletal muscle quality 5 years postpartum. Therefore, muscle PDFF may be a useful biomarker for monitoring muscle health in patients with metabolic diseases.
Limitations: Retrospective monocentric study.
Funding for this study: This work was funded by the Munich Clinician Scientist Program (MCSP) of the University of Munich (LMU; grant number ACS-10), LMU Klinikum, the German Center for Diabetes Research (DZD), and the Helmholtz Zentrum München
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study was approved by the local institutional review board (Ethics Commission of the Medical Faculty, Ludwig-Maximilians Universität München) and all study participants provided written informed consent prior to their participation in the study, which was conducted in accordance with the declaration of Helsinki.
6 min
Assessment of bone microarchitecture and its correlation with trabecular fracture predictor using ImageView
Nagore Castro, Barcelona / Spain
Author Block: N. Castro, J. Catala March; Barcelona/ES
Purpose: This preliminary study evaluates ImgView, a novel software for high-resolution CT (HR-CT) analysis, in quantifying bone microarchitecture and predicting trabecular fracture risk. The relationship between microarchitectural parameters and predicted fracture risk is assessed in skeletal regions affected by morphological alterations or metabolic bone disease.
Methods or Background: Ten healthy subjects underwent 5G Cone Beam CT (CBCT) imaging. Scans were analysed using ImgView (RAR SRL, Italy) and ImageJ software to obtain bone volume fraction (BV/TV), trabecular thickness (Tb.Th), and trabecular spacing (Tb.Sp). ImgView automatically segments cortical and trabecular bone within a 9.5 mm distal radial ROI, computing densitometric and morphometric indices including bone mineral density, cortical thickness and porosity, and detailed trabecular parameters. A subtool predicts fracture risk by applying a uniform 1000N force to the reconstructed bone structure.
Statistical analysis used paired t-tests, Bland–Altman plots, and linear regression to evaluate bone parameters against fracture prediction.
Results or Findings: No significant differences were found for BV/TV between methods (p > 0.05). In contrast, minor differences were identified for Tb.Th (p=0.034) and Tb.Sp (p=0=0.028), Bland-Altman analysis confirmed these remained within acceptable clinical limits. Regression analysis demonstrated strong, significant correlations: BV/TV with fracture prediction (p=0.002; R²=0.764), and Tb.Sp with fracture prediction (p=0.001; R²=0.787). Tb.Th showed a weaker, marginally significant association (p=0.05; R²=0.444).
Conclusion: This study confirms that favourable bone microarchitecture, with greater bone volume and thickness, correlates with lower fracture risk, while increased intertrabecular spacing indicates higher risk. Despite data variability, this study validates ImgView's indicators for fracture risk assessment. ImgView offers automated metrics to enhance diagnostic accuracy by detecting subtle trabecular changes in suspected osteoporosis, enabling radiologists to target high-risk fracture areas. This tool represents a significant advancement in objective bone health assessment, promising improved patient outcomes.
Limitations: Sample size.
Funding for this study: None
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:
Purpose: This preliminary study evaluates ImgView, a novel software for high-resolution CT (HR-CT) analysis, in quantifying bone microarchitecture and predicting trabecular fracture risk. The relationship between microarchitectural parameters and predicted fracture risk is assessed in skeletal regions affected by morphological alterations or metabolic bone disease.
Methods or Background: Ten healthy subjects underwent 5G Cone Beam CT (CBCT) imaging. Scans were analysed using ImgView (RAR SRL, Italy) and ImageJ software to obtain bone volume fraction (BV/TV), trabecular thickness (Tb.Th), and trabecular spacing (Tb.Sp). ImgView automatically segments cortical and trabecular bone within a 9.5 mm distal radial ROI, computing densitometric and morphometric indices including bone mineral density, cortical thickness and porosity, and detailed trabecular parameters. A subtool predicts fracture risk by applying a uniform 1000N force to the reconstructed bone structure.
Statistical analysis used paired t-tests, Bland–Altman plots, and linear regression to evaluate bone parameters against fracture prediction.
Results or Findings: No significant differences were found for BV/TV between methods (p > 0.05). In contrast, minor differences were identified for Tb.Th (p=0.034) and Tb.Sp (p=0=0.028), Bland-Altman analysis confirmed these remained within acceptable clinical limits. Regression analysis demonstrated strong, significant correlations: BV/TV with fracture prediction (p=0.002; R²=0.764), and Tb.Sp with fracture prediction (p=0.001; R²=0.787). Tb.Th showed a weaker, marginally significant association (p=0.05; R²=0.444).
Conclusion: This study confirms that favourable bone microarchitecture, with greater bone volume and thickness, correlates with lower fracture risk, while increased intertrabecular spacing indicates higher risk. Despite data variability, this study validates ImgView's indicators for fracture risk assessment. ImgView offers automated metrics to enhance diagnostic accuracy by detecting subtle trabecular changes in suspected osteoporosis, enabling radiologists to target high-risk fracture areas. This tool represents a significant advancement in objective bone health assessment, promising improved patient outcomes.
Limitations: Sample size.
Funding for this study: None
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:
6 min
Opportunistic screening of lumbar vertebrae in CT using TotalSegmentator to prevent osteoporotic vertebral fractures
Magdalena Seng, Basel / Switzerland
Author Block: M. Seng, J. Wasserthal, M. Bach, D. Harder, M. Obmann, C. Meier, E. M. Merkle, H-C. Breit, M. Segeroth; Basel/CH
Purpose: Osteoporotic vertebral fractures impair quality of life but are preventable. Routine CT examinations offer an opportunity for early risk stratification. This study assessed quantitative metrics of vertebral and paraspinal tissue extracted by TotalSegmentator to identify predictors of osteoporotic vertebral fractures.
Methods or Background: 3299 lumbar spine CT datasets were retrospectively identified in our PACS (2020-2025). 1469 were excluded using large language models (contrast agent, bone metastasis, severe artefacts, hematological diseases, high impact trauma) and 370 due to incompatible imaging parameters (image type, bone kernel, slice thickness). Automated segmentation of vertebrae and paraspinal tissue was performed using TotalSegmentator. Quantitative measures included mean attenuation values and volumes. Vertebral heights of L1 to L4 were assessed on three coronal slices, with fractures defined by a height reduction of >20%. Associations were assessed with logistic regression and ROC analyses.
Results or Findings: Automated vertebral segmentation was successful in 1185 of 1460 cases. The mean attenuation of the non-fractured lumbar vertebrae was lower in patients with an osteoporotic fracture (85.08 (CI: 81.25 - 89.17) n=612) compared to patients without any fracture (125.75 (CI: 120.50 - 131.13) n=573, p < 0.001). Attenuation alone showed an AUC of 0.70 (95% CI: 0.67–0.73) with an optimal cut-off at 90 Hounsfield units (sensitivity 0.71, specificity 0.59).
Volume and attenuation of the iliopsoas and autochthon muscles differed significantly between groups (p < 0.001), whereas subcutaneous and visceral fat did not. A combined logistic model (AUC 0.77 (95% CI: 0.71–0.82)) rendered the attenuation of the non-fractured vertebrae as most relevant followed by the attenuation of L1, L3, L2 and L4.
Conclusion: Vertebral attenuation, below 90 HU, is a robust predictor of osteoporotic fractures. Opportunistic CT-based screening could support early identification and timely intervention in patients at risk.
Limitations: None.
Funding for this study: None.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The requirement for informed consent was waived as this was a retrospective study.
Purpose: Osteoporotic vertebral fractures impair quality of life but are preventable. Routine CT examinations offer an opportunity for early risk stratification. This study assessed quantitative metrics of vertebral and paraspinal tissue extracted by TotalSegmentator to identify predictors of osteoporotic vertebral fractures.
Methods or Background: 3299 lumbar spine CT datasets were retrospectively identified in our PACS (2020-2025). 1469 were excluded using large language models (contrast agent, bone metastasis, severe artefacts, hematological diseases, high impact trauma) and 370 due to incompatible imaging parameters (image type, bone kernel, slice thickness). Automated segmentation of vertebrae and paraspinal tissue was performed using TotalSegmentator. Quantitative measures included mean attenuation values and volumes. Vertebral heights of L1 to L4 were assessed on three coronal slices, with fractures defined by a height reduction of >20%. Associations were assessed with logistic regression and ROC analyses.
Results or Findings: Automated vertebral segmentation was successful in 1185 of 1460 cases. The mean attenuation of the non-fractured lumbar vertebrae was lower in patients with an osteoporotic fracture (85.08 (CI: 81.25 - 89.17) n=612) compared to patients without any fracture (125.75 (CI: 120.50 - 131.13) n=573, p < 0.001). Attenuation alone showed an AUC of 0.70 (95% CI: 0.67–0.73) with an optimal cut-off at 90 Hounsfield units (sensitivity 0.71, specificity 0.59).
Volume and attenuation of the iliopsoas and autochthon muscles differed significantly between groups (p < 0.001), whereas subcutaneous and visceral fat did not. A combined logistic model (AUC 0.77 (95% CI: 0.71–0.82)) rendered the attenuation of the non-fractured vertebrae as most relevant followed by the attenuation of L1, L3, L2 and L4.
Conclusion: Vertebral attenuation, below 90 HU, is a robust predictor of osteoporotic fractures. Opportunistic CT-based screening could support early identification and timely intervention in patients at risk.
Limitations: None.
Funding for this study: None.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The requirement for informed consent was waived as this was a retrospective study.
6 min
Novel and Compact Digital Tomosynthesis DXA Device for Bone Mineral Density
Karen Knapp, Newton Abbot / United Kingdom
Author Block: J. Hu1, J. Chen1, S. Wells2, V. Soloviev2, S. Coulson2, S. Phillips2, J. Evans1, A. Forbes-Brown1, K. Knapp1; 1Exeter/UK, 2Oxford/UK
Purpose: Bone mineral density (BMD) measurement is central to the diagnosis and management of osteoporosis [1]. Dual-energy X-ray absorptiometry (DXA) remains the most widely used clinical tool, but is limited by large, immobile scanners restricted to specialist centres, reducing accessibility [2] and two-dimensional (2D), low-resolution measurements of areal BMD (aBMD) that do not capture the true volumetric density of local bone variation [3].
Methods or Background: We present early feasibility results from an ongoing collaboration with Adaptix Ltd. to develop a digital tomosynthesis–based DXA (DT-DXA) system. It employs a compact x-ray source that can sequentially fire X-rays from an array of positions with a short source–detector distance, enabling portable deployment and three-dimensional tomographic capability with areal or volumetric BMD (vBMD) as an add-on to extremity orthopaedic imaging. DT-DXA can use synthetic 2D to calculate aBMD with the potential advantage of being able to exclude slices containing only soft tissue from the calculation.
3D-printed trabecular bone phantoms [4] and 30 surgically replaced human femoral heads, representing a broad spectrum of densities and microarchitectures, were used to compare DT-DXA performance against iDXA (GELunar) and microCT and to evaluate its accuracy and robustness across clinically relevant bone conditions.
Results or Findings: DT-DXA aBMD and vBMD results were benchmarked against GE iDXA. Phantom and femoral head studies showed close agreement.
Conclusion: DT-DXA offers BMD measurement in a compact, mobile platform. These results highlight its potential for offering point-of-care access to bone density assessment, with the potential to expedite diagnosis and treatment of osteoporosis.
Limitations: In vivo patient imaging has not yet been performed. The radiation dose of DT-DXA was higher than the iDXA, but far lower than medical CT, with scope for further optimisation.
Funding for this study: UK Research and Innovation
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Exeter Clinical Research Facility
Purpose: Bone mineral density (BMD) measurement is central to the diagnosis and management of osteoporosis [1]. Dual-energy X-ray absorptiometry (DXA) remains the most widely used clinical tool, but is limited by large, immobile scanners restricted to specialist centres, reducing accessibility [2] and two-dimensional (2D), low-resolution measurements of areal BMD (aBMD) that do not capture the true volumetric density of local bone variation [3].
Methods or Background: We present early feasibility results from an ongoing collaboration with Adaptix Ltd. to develop a digital tomosynthesis–based DXA (DT-DXA) system. It employs a compact x-ray source that can sequentially fire X-rays from an array of positions with a short source–detector distance, enabling portable deployment and three-dimensional tomographic capability with areal or volumetric BMD (vBMD) as an add-on to extremity orthopaedic imaging. DT-DXA can use synthetic 2D to calculate aBMD with the potential advantage of being able to exclude slices containing only soft tissue from the calculation.
3D-printed trabecular bone phantoms [4] and 30 surgically replaced human femoral heads, representing a broad spectrum of densities and microarchitectures, were used to compare DT-DXA performance against iDXA (GELunar) and microCT and to evaluate its accuracy and robustness across clinically relevant bone conditions.
Results or Findings: DT-DXA aBMD and vBMD results were benchmarked against GE iDXA. Phantom and femoral head studies showed close agreement.
Conclusion: DT-DXA offers BMD measurement in a compact, mobile platform. These results highlight its potential for offering point-of-care access to bone density assessment, with the potential to expedite diagnosis and treatment of osteoporosis.
Limitations: In vivo patient imaging has not yet been performed. The radiation dose of DT-DXA was higher than the iDXA, but far lower than medical CT, with scope for further optimisation.
Funding for this study: UK Research and Innovation
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Exeter Clinical Research Facility
6 min
Reproducibility of DXA BMD measurements: effect of soft-tissue thickness and acquisition mode
Francesco Cicchetti, Milan / Italy
Author Block: F. Cicchetti, G. De Padova, L. Macrì, M. Costa, K. Caringal, D. Albano, A. Zagarella, C. Messina; Milan/IT
Purpose: To evaluate how acquisition mode and simulated soft-tissue thickness affect the stability of dual-energy X-ray absorptiometry (DXA) measurements.
Methods or Background: DXA scans were performed with Stratos DR densitometer (DMS, France), software version 5.3.3.3. Three acquisition modes (Fast, Normal, Precision) were tested under four simulated adipose tissue thicknesses (0, 1, 3, 6 cm, applied with pork rind). Each condition had 25 repeated acquisitions, totalling 300 measurements. Data were analyzed with two-way ANOVA and Tukey’s HSD test (p < 0.05).
Results or Findings: ANOVA demonstrated a highly significant effect of tissue thickness (F = 27.0, p < 0.0001), while acquisition mode alone was not significant (F = 0.72, p = 0.49). A significant thickness × mode interaction was observed (F = 3.26, p = 0.004). Post-hoc analysis revealed that at the same thickness no differences were observed between Fast, Normal, and Precision modes. Within modes, Normal and Precision showed significant changes between 0 cm and 1 cm (p ≤ 0.04), and between 1 cm and thicker layers (p ≈ 0.001–0.002). The Fast mode showed significance only between 0 cm and 1 cm (p = 0.0002). Although the 0 cm vs 6 cm comparison was not statistically significant (p = 0.53), reproducibility indices indicated a reduction in measurement precision from ~98.5% at 0 cm to ~94% at 6 cm, suggesting a relevant downward trend.
Conclusion: Simulated soft tissue thickness is the main determinant of DXA measurement variability. Although the three acquisition modes provide comparable results under the same thickness, Normal and Precision modes are more sensitive to minimal increments in tissue layers, whereas the Fast mode appears less sensitive but more variable.
Limitations: The limitations of the study are the phantom design and use of simulated soft tissue.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:
Purpose: To evaluate how acquisition mode and simulated soft-tissue thickness affect the stability of dual-energy X-ray absorptiometry (DXA) measurements.
Methods or Background: DXA scans were performed with Stratos DR densitometer (DMS, France), software version 5.3.3.3. Three acquisition modes (Fast, Normal, Precision) were tested under four simulated adipose tissue thicknesses (0, 1, 3, 6 cm, applied with pork rind). Each condition had 25 repeated acquisitions, totalling 300 measurements. Data were analyzed with two-way ANOVA and Tukey’s HSD test (p < 0.05).
Results or Findings: ANOVA demonstrated a highly significant effect of tissue thickness (F = 27.0, p < 0.0001), while acquisition mode alone was not significant (F = 0.72, p = 0.49). A significant thickness × mode interaction was observed (F = 3.26, p = 0.004). Post-hoc analysis revealed that at the same thickness no differences were observed between Fast, Normal, and Precision modes. Within modes, Normal and Precision showed significant changes between 0 cm and 1 cm (p ≤ 0.04), and between 1 cm and thicker layers (p ≈ 0.001–0.002). The Fast mode showed significance only between 0 cm and 1 cm (p = 0.0002). Although the 0 cm vs 6 cm comparison was not statistically significant (p = 0.53), reproducibility indices indicated a reduction in measurement precision from ~98.5% at 0 cm to ~94% at 6 cm, suggesting a relevant downward trend.
Conclusion: Simulated soft tissue thickness is the main determinant of DXA measurement variability. Although the three acquisition modes provide comparable results under the same thickness, Normal and Precision modes are more sensitive to minimal increments in tissue layers, whereas the Fast mode appears less sensitive but more variable.
Limitations: The limitations of the study are the phantom design and use of simulated soft tissue.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:
6 min
HR-pQCT versus DXA for longitudinal monitoring of patients with osteogenesis imperfecta: results at 1-year follow-up
Silvia Gazzotti, Bologna / Italy
Author Block: S. Gazzotti1, G. Fraterrigo1, E. Schileo1, R. Sassi1, M. P. Aparisi Gomez2, A. Moroni1, L. Sangiorgi1, F. Taddei1, A. Bazzocchi1; 1Bologna/IT, 2Auckland/NZ
Purpose: Osteogenesis imperfecta (OI) is a rare genetic disease characterised by frequent fractures. Affected patients undergo periodic dual-energy X-ray absorptiometry (DXA), but this has limitations. High-resolution peripheral quantitative computed tomography (HR-pQCT) is an alternative low-dose technique providing information on volumetric density and microstructure, but there are scarce data on its use for follow-up. The aim of this work is to compare the performance of DXA and HR-pQCT in the longitudinal monitoring of OI.
Methods or Background: This preliminary analysis includes 19 adult patients (median age: 35.5 years) with OI type I who underwent DXA scans at lumbar spine, hip, total body, and forearm and HR-pQCT at distal radius and tibia at baseline and at 1-year follow-up.
Results or Findings: Median Z-scores at the femoral neck and lumbar spine were -1.3 and -1.9 at baseline versus -1.4 and -1.9 at 1-year, respectively. No significant changes in DXA-areal bone mineral density (aBMD) could be detected in 1-year at central sites. HR-pQCT parameters overall showed minimal variations (median change: +1.5% to -6.4% at radius; +1.75% to -0.4% at tibia). DXA-aBMD at ultradistal radius was correlated with total volumetric BMD by HR-pQCT at radius both at baseline (Spearman’s rho = 0.70; p < 0.05) and at 1-year (Spearman’s rho = 0.88; p < 0.001). However, when the respective longitudinal changes were directly compared, the results were not statistically significant.
Conclusion: While data indicate some site-specific agreement between DXA and HR-pQCT, the two modalities are likely to play a differential role in the longitudinal monitoring of OI. A 1-year time frame may be too short to observe significant changes in measured parameters, which could have important implications for selecting the most appropriate interval for follow-up in clinical practice.
Limitations: Ongoing study, small sample size.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: CE-AVEC 890/2022/Sper/IOR
Purpose: Osteogenesis imperfecta (OI) is a rare genetic disease characterised by frequent fractures. Affected patients undergo periodic dual-energy X-ray absorptiometry (DXA), but this has limitations. High-resolution peripheral quantitative computed tomography (HR-pQCT) is an alternative low-dose technique providing information on volumetric density and microstructure, but there are scarce data on its use for follow-up. The aim of this work is to compare the performance of DXA and HR-pQCT in the longitudinal monitoring of OI.
Methods or Background: This preliminary analysis includes 19 adult patients (median age: 35.5 years) with OI type I who underwent DXA scans at lumbar spine, hip, total body, and forearm and HR-pQCT at distal radius and tibia at baseline and at 1-year follow-up.
Results or Findings: Median Z-scores at the femoral neck and lumbar spine were -1.3 and -1.9 at baseline versus -1.4 and -1.9 at 1-year, respectively. No significant changes in DXA-areal bone mineral density (aBMD) could be detected in 1-year at central sites. HR-pQCT parameters overall showed minimal variations (median change: +1.5% to -6.4% at radius; +1.75% to -0.4% at tibia). DXA-aBMD at ultradistal radius was correlated with total volumetric BMD by HR-pQCT at radius both at baseline (Spearman’s rho = 0.70; p < 0.05) and at 1-year (Spearman’s rho = 0.88; p < 0.001). However, when the respective longitudinal changes were directly compared, the results were not statistically significant.
Conclusion: While data indicate some site-specific agreement between DXA and HR-pQCT, the two modalities are likely to play a differential role in the longitudinal monitoring of OI. A 1-year time frame may be too short to observe significant changes in measured parameters, which could have important implications for selecting the most appropriate interval for follow-up in clinical practice.
Limitations: Ongoing study, small sample size.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: CE-AVEC 890/2022/Sper/IOR