Powered by
14:37 CET
MY 13 - Musculoskeletal
Musculoskeletal
Saturday, March 3, 08:30 - 10:00
Moderators: M. S. Posadzy (Poznan/PL), M. F. Reiser (Munich/DE)

19 presentations in this session:
B-1012
08:30
Spectral detector-computed tomography iodine density thresholds for the detection of vertebral metastases
N. Abdullayev, V. Neuhaus, M. Le Blanc, N. Grosse Hokamp, V. Maus, S. Lennartz, D. Maintz, J. Borggrefe; Cologne/DE
Abstract

Purpose: To evaluate quantitative iodine density mapping (IDM) from spectral detector computed tomography (SDCT) as a quantitative parameter for the separation of vertebral trabecular bone metastases (BM) and healthy trabecular bone (HTB).
Methods and Materials: This retrospective study includes portal venous SDCT datasets of cancer patients with and without known bone metastasis (n=43 and n=40, respectively). Target lesions as well as non-affected control vertebrae were defined by two radiologists in consensus using follow-up CT imaging, MRI and/or bone scintigraphy. IDM and its standard deviation (SD) were determined based on ROI measurements in BM and HTB of patients with and in HTB of patients without bone metastasis, and various reference tissues/vessels in both. Phantomless bone mineral density (vBMD) measurements of the lumbar spine were conducted.
Results: We found a significant difference between IDM of BM and HTB (mean 5.55±0.98 vs. 3.57±0.96) (p<0.0001); however, there was a considerable overlap and a vBMD bias. A trivariate analysis including IDM, the inhomogeneity of the IDM as determined by the SD, and the normalization to the vertebral venous sinus improved the statistical separation of metastasis to a specificity of 100.0% and a sensitivity of 95.4% (AUC 0.98, p<0.05 compared to univariate models).
Conclusion: SDCT allows for quantification of iodine in bone lesions which appears promising to serve as a parameter for the detection of bone metastases.

B-1013
08:34
Combining fractal- and entropy-based bone texture analysis for the prediction of osteoarthritis: data from the multicentre osteoarthritis study (MOST)
Z. Bertalan1, S. Nehrer2, R. Ljuhar1, A. Fahrleitner-Pammer3, D. Ljuhar1, H.-P. Dimai3; 1Vienna/AT 2Krems/AT 3Graz/AT
Abstract

Purpose: Osteoarthritis (OA) is the leading cause of long-term pain and disabilities associated with musculoskeletal disorders. Effective treatment depends on early detection and quantification of risk. Current disease parameters, like joint space width (JSW), have proven to be insufficient for the prediction of OA. The purpose of the present study was to investigate if combining bone texture analyses with JSW and joint space area (JSA) may improve prediction of OA.
Methods and Materials: Conventional posterior-anterior (PA) knee radiographs were obtained from the multicentre osteoarthritis study (MOST) database. Oriented texture algorithms were developed, using specific machine-learning algorithms. Selected areas used for textural analyses included 4 regions of interest (ROI) in the proximal tibia and one on each condyle of the distal femur. Furthermore, JSW/JSA were assessed using newly developed and fully automated software.
Results: 1092 radiographs obtained from one study centre were screened. 230 women, 344 men met the inclusion criteria, i.e. a Kellgren & Lawrence (KL) score of 0 at baseline. At month 84, 41 female, 79 male patients had developed KL≥1, and 189 female, 265 male patients remained at KL0. Area under the curve (AUC) for incident OA using JSW/JSA/clinical features was 0.67±0.08 for women, 0.61±0.1 for men. In contrast, combining texture, JSW/A and clinical features resulted in significantly improved AUC (0.80±0.07 for women and 0.69±0.1 for men, respectively).
Conclusion: This study provides strong evidence that a combination of textural analyses of plain radiographs together with JSW/A and clinical features is superior to JSW/A and clinical features alone in predicting incident OA.

B-1014
08:38
Delayed gadolinium-enhanced MRI of menisci and cartilage (dGEMRIM/dGEMRIC) in overweight patients with knee osteoarthritis
S. Hangaard1, H. Gudbergsen1, C. L. Daugaard1, H. Bliddal2, J. D. Nybing1, V. Casula3, M. T. Nieminen3, C. J. Tiderius4, M. Boesen2; 1Frederiksberg/DK 2Copenhagen/DK 3Oulu/FI 4Lund/SE
Abstract

Purpose: Changes in the menisci are associated with development of knee osteoarthritis (KOA). The aim of this study was to examine the delayed-gadolinium-enhanced-MRI-of-menisci (dGEMRIM) and its relationship with Kellgren-Lawrence grade (KLG) and with articular cartilage dGEMRIC.
Methods and Materials: 86 overweight patients with KOA (mean KLG 3) from the CAROT study had an intra-articular injection of Gd contrast. Four invasion times (50, 350, 650, 1410 ms) dGEMRIC and dGEMRIM were performed in 1.5 T. T1 relaxation time values were calculated for posterior weight-bearing femoral cartilage in the lateral knee compartment, and for the posterior horn of the lateral and medial menisci.
Results: For femoral cartilage (N=86) the mean T1 value was 441ms. The mean T1 value in the lateral menisci (N=85) was 498ms and for the medial menisci (N=62) mean T1 value was 484ms. A positive correlation was found between posterior medial and lateral menisci R=0.62 (p<0.0001) and a similar trend was seen between lateral femoral cartilage and lateral menisci with R= 0.26 (p=0.02). Comparing the meniscus from the most affected knee compartment to KLG showed trends toward increasing T1 values for KLG 1-3 and a decreasing T1 value for KLG 4.
Conclusion: The positive correlation between lateral and medial menisci indicates parallel degeneration processes in both knee compartments. The correlation between menisci and cartilage suggests concomitant, but different, degeneration in the two tissues in OA. The interpretation of inverse U-shaped relation between meniscal T1 values and KLG is in accordance with findings comparing various degrees of meniscal degeneration with dGEMRIM and proteoglycan content.

B-1015
08:42
Ultrasound of the posterior interosseous nerve in the arcade of Frohse
C. Tulay, S. Aubry, A. Podda, J. Behr; Besancon/FR
Abstract

Purpose: To accurately assess the diameter of the posterior interosseous nerve (PIN) as it passes through the arcade of Frohse and to demonstrate that exist a physiological difference in anteroposterior diameter between the upstream, the entry point and downstream of the arcade.
Methods and Materials: Prospective monocentric study of the PIN in 30 healthy volunteers (sex ratio = 1) which have been examined bilaterally by ultrasound with a high frequency probe (18 MHz) in longitudinal section. The anteroposterior diameter of the PIN was measured at the entry point in the arcade of Frohse, 5mm upstream and 5mm downstream.
Results: The anteroposterior diameter of the NIP was 0.833 mm +/- 0.208, 0.6mm +/- 0.173 and 0.486mm +/- 0.131 respectively upstream, at the entry and downstream of the Arcade of Frohse. It significantly decreases by 27.2% between the upstream and entry point (-0.233mm; p<0.001), by 16.6% between the entry point and downstream (-0.114mm; p < 0.001), and by 39% between the upstream and downstream (0.357mm; p<0.001).
Conclusion: There is a physiological decrease in the anteroposterior diameter of the PIN as it passes through the arcade of Frohse, not to be wrongly considered as a compression of the nerve in the arcade.

B-1016
08:46
Safety and efficiency of treatment of partial supraspinatus tendon tear with injection of PRP and HA
C. Lupo1, A. Meli2, V. Incarbone2, N. Casamassima2, L. Callegari2; 1Palermo/IT 2Varese/IT
Abstract

Purpose: Evaluate safety and efficiency of treatment of partial supraspinatus tendon tears with intralesionals injection of autologous platelet-rich plasma (PRP) associated with low molecular weight hyaluronic acid (HA 0.8%) under ultrasound guidance.
Methods and Materials: Between January and September 2016, we selected 84 patients, mean age 57, with a ultrasound diagnosis of partial tear of supraspinatus tendon; 48/84 confirmed with MRI. Before the treatment (T0), all patients were evaluated the pain with the visual analogue scale (VAS). After local ultrasound guided anaesthesia with lidocaine, an injection of glenohumeral joint with a 22G needle with lateral access was given, followed by needling of supraspinatus tendon and subsequent intratendineus injection with 2.5cc of (PRP) and 2cc of (HA0.8%), 3 injections, one every three weeks. The safety of the procedure was considered by the analysis of complications; efficacy was considered by clinical follow-up at six weeks after the procedure and imaging follow-up at six months in 84/84 patients with US and in 35/84 with MRI.
Results: 84/84 patients improved pain with a statistically significant difference (p <0.001) of VAS scale before and after treatment without major complications. 55/84 patients showed some modification of the ultrasound and RM patterns.
Conclusion: Injection of (PRP) with (HA0.8%) is a well-tolerated procedure in treatment of partial the supraspinatus tendon tear with a marked improvement on pain.

B-1017
08:50
Improving the sensitivity of bone mineral density assessment using spectral detector CT
S. van Hedent, K.-H. Su, F. Liang, J.-W. Kuo, D. W. Jordan, B. Eck, P. R. Ros, R. Muzic; Cleveland/US
Abstract

Purpose: Bone mineral density (BMD) analysis by Dual-Energy x-ray Absorptiometry (DEXA) is complicated by false negatives due to overlapping densities in the projection area. Spectral Detector CT (SDCT) can overcome these limitations, while providing volumetric information. We investigated its performance for BMD assessment and compared it to DEXA and phantom-less Quantitative CT (QCT).
Methods and Materials: Ten uniform solutions of bone-like material, K2HPO4 ranging 0-600 mg/ml, were scanned in an acrylic phantom using the SDCT in four varying clinically relevant scan conditions. Images were processed to estimate the K2HPO4 concentrations. A model representing a human lumbar spine (European spine phantom) was scanned and used as calibration for linear regression analysis, so that our method could be retrospectively applied to abdominal SDCT scans of 20 patients who also had a QCT and DEXA. Performance of QCT, DEXA and our SDCT method were compared.
Results: There was excellent correlation (R2 >0.99, p = 0.00) between true and measured K2HPO4 concentrations for all scan conditions. Overall mean measurement error ranged from -11.49 ± 4.70 mg/ml (-2.78 ± 6.01%) to -12.34 ± 6.26 mg/ml (-4.77 ± 3.00%) depending on scan conditions. Using DEXA as gold standard, sensitivity and specificity for detecting decreased BMD in the scanned patients was 100 and 73.3%, and 100 and 40% for SDCT and QCT respectively.
Conclusion: SCDT has excellent sensitivity and high specificity for detecting decreased BMD, allowing its use for opportunistic screening or pre-operative planning.

B-1018
08:54
Virtual bone mineral density imaging with third-generation dual-energy CT for diagnosis of osteoporosis: a preliminary study
L. Wang, G. Shenchu, B. He, J. Chen; Nantong/CN
Abstract

Purpose: To quantitatively assess the diagnostic value of virtual bone mineral density (VBD) imaging at DECT for osteoporosis (OP).
Methods and Materials: Forty-five patients with vertebral trauma underwent DECT and DXA examinations of the lumbar 1-4. VBD images were derived from three-material decomposition algorithm using post-processing software (‘Liver VNC’, Syngo Via Dual Energy, Siemens), while the default CT values of fat, soft tissues, iodine and iodine slope were corrected to the yellow marrow, red marrow, calcium and calcium slope which were recommended in another software (‘Bone Marrow’, Syngo Via Dual Energy). CT attenuation values (VBD_Att), calcium density (VBD_CaD) and fat fraction (VBD_Fat) were measured for quantitative analyses. CT numbers on mixed images (LB_0.5) reconstructed with a merging coefficient 0.5 at DECT were recorded. Bone mineral density and T scores of lumbar measured with DXA served as the gold standard. Pearson correlation analysis was performed to compare DECT and DXA results. Diagnostic performance of VBD imaging was assessed by ROC analysis.
Results: A total of 166 lumbars were evaluated. LB_0.5 and VBD_Att were significantly different between OP and non-OP vertebral bodies (all P<.01). VBD_Att and VBD_CaD both had significant correlations with T scores (all P<.01). However, no significant correlation existed between VBD_Fat and T scores (P=.08). Furthermore, sensitivity and specificity for assessment of OP were 86.00%, 80.17% for VBD_Att and 84.00%, 81.03% for VBD_CaD. Diagnostic performance was significantly higher in VBD_Att and VBD_CaD compared to LB_0.5 (all P<.01).
Conclusion: Dual-energy VBD imaging shows acceptable diagnostic performance for OP in patients with vertebral trauma.

B-1020
09:02
In vivo quantitative 4D-CT analysis of carpal kinematics with radio-scaphoid and luno-capitate angles during radio-ulnar deviation: feasibility and clinical interest
A. Rauch; Nancy/FR
Abstract

Purpose: To study the variation of radioscaphoid (RSA) and lunocapitate (LCA) angles during wrist radio-ulnar deviation (RUD) using semi-automatic quantitative 4D-CT in patients with and without scapholunate-ligament tears (SLT).
Methods and Materials: 37 patients with suspected scapholunate instability were prospectively evaluated with 4D-CT and CT-arthrography of the wrist. The study was approved by the local ethics committee and all patients signed an informed consent. Various dynamic parameters (mean, amplitude, maximal angle, standard deviation (SD), angular coefficient of variation (ACV)) describing RSA and LCA variation during wrist RUD were calculated and compared in patients with and without SLT by two readers.
Results: The mean values for RSA and LCA in the control group varied, respectively, from 102.98° ± 7.6° to 104.47° ± 8.89° and 85.92° ± 9.37° to 90.09° ± 10.56° with ACV from 10,78% to 11.03% and 13.49% to 13.52% for readers one and two, respectively. The inter-observer variability was excellent for RSA (ICC = 0.82) and substantial for LCA (ICC= 0.79). LCA amplitude, SD and maximal angle were significantly smaller in SLT group (p varied from 0.0001 to 0.0026), respectively, from 36% to 44%, 37% to 44% and 13% to 19%. RSA amplitude tended to be smaller in SLT group (18% to 27%, p varied from 0.0295 to 0.1328). LCA yielded the best sensitivity for SLT (71-91%) while RSA yielded the best specificity (87-100%).
Conclusion: 4D-CT can quantitatively assess carpal kinematics during RUD which could improve the diagnosis and prognostic evaluation of patients with SLT.

B-1021
09:06
Detection of osseous metastases using dual-energy CT with material decomposition algorithms: phantom development and preliminary clinical validation
H.-C. Huang1, B. M. Yeh2, R. Srinivasan2, Y. Sun2; 1Taipei City/TW 2San Francisco/US
Abstract

Purpose: Meta-analysis studies have shown that the conventional single-energy computed tomography (SECT) has limited ability in detecting subtle bone metastases. The purpose of this study was to use dual-energy CT (DECT) with bone-water material decomposition algorithms to detect nearly isodense bone metastases in phantoms, and to validate its use with retrospective patient DECT scans which were initially read as negative for bone metastases, but later detected by other imaging modalities or at follow-up CT.
Methods and Materials: We constructed 51 semi-anthropomorphic lumbar spine phantoms randomly embedded with 75 simulated tumours (25 mild lytic, 25 isodense, and 25 mild sclerotic). These phantoms were scanned in a rapid-kilovoltage-switching DECT scanner. Two radiologists independently reviewed images from 70 keV virtual monochromatic reconstruction and 4 material decomposition algorithms (hydroxyapatite-water, water-hydroxyapatite, cortical bone-water, water-cortical bone). We recorded reviewer’s response regarding the presence of tumours. The sensitivity and specificity of different reconstruction algorithms were evaluated with McNemar test. The tumour conspicuity scores were recorded in 3-point Likert scales and then evaluated by Wilcoxon signed rank test. Finally, we validated our testing algorithms by retrospectively reviewing patients’ images in our institution.
Results: Hydroxyapatite-water material decomposition algorithm achieved higher sensitivity in detecting isodense lesions (38% vs 18%, p=0.013) and possessed higher tumour conspicuity score (p<0.0001) as compared to the 70 keV reconstruction in the phantom study. Hydroxyapatite-water algorithm also detected more tumours in the retrospective patients’ studies.
Conclusion: DECT with hydroxyapatite-water material decomposition may help detect subtle spine metastases. Further study in prospective clinical scans is warranted.

B-1022
09:10
High diagnostic accuracy of kep on dynamic contrast-enhanced MRI perfusion in identifying vertebral malignancy
M. Verma, S. Sood, B. Singh, M. Thakur, S. Sharma, S. Sharma; Shimla/IN
Abstract

Purpose: Vertebral lesions are not easily amenable to biopsy due to inherent risks associated with this invasive procedure. There is considerable overlapping of radiologic features of benign and malignant lesions on conventional MRI. We performed this study to evaluate the role of dynamic contrast-enhanced (DCE) MRI perfusion parameters in differentiating malignant from benign lesions.
Methods and Materials: We studied all patients presenting to our hospital with a radiologic evidence of vertebral lesions from July 2016 through June 2017. We performed DCE-MRI of spine on 1.5 T scanner using 3D VIBE sequence after intravenously injecting 0.1mmol/kg body weight of gadopentetate dimeglumine. We used Tofts model to calculate DCE parameters that included Ktrans, kep, Ve and iAUC. We compared the means of each perfusion parameters by type of lesion (benign/malignant) at 0.05 significance level and performed ROC curve analysis.
Results: We could confirm histologic/cytologic diagnosis in 35 of the 45 patients recruited. Of these 19 were tubercular and 16 malignant lesions. The mean (±sd) of kep (min-1) was significantly higher (2.89±3.3) in malignant compared to benign lesions (0.81±0.19), whereas Ve (ml/g) was significantly lower in malignant (0.27±0.13) compared to benign lesions (0.47±0.12) at 0.05 significance level. kep cut-off of ≥ 1.17 min-1 had a sensitivity of 93.8% and specificity of 100% with a diagnostic accuracy of 94.4% in detecting malignant disease.
Conclusion: High kep is highly suggestive of malignant lesions. We recommend kep cut-off of ≥ 1.17 min-1 that was found to have high diagnostic accuracy in identifying malignant lesions.

B-1023
09:14
Complementary metal artefact reduction of total hip replacements by monoenergetic and O-MAR reconstructions in spectral-detector computed tomography
K. R. Laukamp, S. Lennartz, V. Neuhaus, N. Grosse Hokamp, R. Rau, M. Le Blanc, N. Abdullayev, D. Maintz, J. Borggrefe; Cologne/DE
Abstract

Purpose: This study compares metal artefact (MA) reduction in imaging of total hip replacements (THR) using virtual monoenergetic images (VMI), for MA reduction specialised iterative reconstructions (O-MAR) and conventional images (CI) from spectral-detector computed tomography (SDCT).
Methods and Materials: 27 SDCT datasets of patients carrying THR were included. CI, O-MAR and VMI with different energy levels (60-200keV) were reconstructed from the same scans. MA width was measured. Attenuation (HU), noise (SD) and contrast-to-noise ratio (CNR) were determined in: extinction artefact, adjacent bone, muscle and bladder. Two radiologists assessed artefact reduction and image quality subjectively.
Results: In comparison to CI, VMI (200keV) and O-MAR showed a strong artefact reduction (MA width: CI 29.9±6.8mm, VMI 17.6±13.6mm, p<0.01; O-MAR 16.5±14.9mm, p<0.001; MA density: CI -412.1±204.5HU, VMI -279.7±283.7HU; p<0.01; O-MAR -116.74±105.6HU, p<0.001). In strong artefacts reduction was superior by O-MAR. In moderate artefacts VMI was more effective. O-MAR showed best noise reduction and CNR in bladder and muscle (p<0.05), whereas VMI were superior for depiction of bone (p<0.001). Subjective assessment confirmed that VMI and O-MAR improve artefact reduction and image quality (p<0.001).
Conclusion: O-MAR and VMI (200keV) yielded significant MA reduction. Each showed distinct advantages regarding effectiveness of artefact reduction and assessment of adjacent bone, muscle and pelvic organs.

B-1024
09:18
Ultra-high field 7 Tesla MRI and biomechanical investigation of vertebral bone microarchitecture
D. Guenoun, A. Foure, M. Pithioux, S. Guis, T. Le Corroller, V. Pauly, P. Chabrand, P. Champsaur, D. Bendahan; Marseille/FR
Abstract

Purpose: to investigate bone microarchitecture parameters of human ex-vivo vertebrae using ultra-high field 7Tesla MRI (7T MRI).
Methods and Materials: Twenty four vertebrae (L2, L3, and L4) from eight cadavers were studied using 7T MRI. Their Bone mineral densities (BMD) were investigated using dual energy x-ray absorptiometry. Then, all specimens underwent mechanical compression tests to failure. The failure load (in Newton) and constraint (in megapascal) were measured. Bone volume fraction (BV/TV), trabecular thickness (Tb.Th), and trabecular spacing (Tb.Sp) were measured in MRI using a digital topological analysis (Bone J). Measurements were recorded by two observators to characterize the interrater reliability. Statistical analyses were performed using SPSS. Correlations between variables were analyzed using Spearman correlations and stepwise regression. A p value of 0.05 was considered as significant.
Results: the inter-rater reliability for bone microarchitecture parameters quantification was very good. The failure load and constraint measured during the compression tests were significantly correlated with BV/TV, Tb.Sp and BMD while Tb.Th was correlated with constraint only (p<0.05). Stepwise regression with backward elimination demonstrated that combining BV/TV and BMD improved the relationship with the constraints from an adjusted r² =0.384 for BMD alone to an adjusted r²=0.414 for BMD+BV/TV.
Conclusion: we demonstrated for the first time that the variables characterizing the vertebral bone microarchitecture in 7T MRI were significantly correlated with biomechanical parameters. In addition, we illustrated that combining BMD and 7T MR, trabecular bone analysis provided additional information regarding vertebral bone strength in comparison with DXA alone.

B-1025
09:22
Postoperative evaluation of a arthroscopic coracoid bone block surgery with CT-Scan.
O. Andreani1, P. Gendre2, C. Dekimpe1, A. Rudel1, N. Amoretti1, P. Boileau1; 1Nice/FR 2Cagnes sur Mer/FR
Abstract

Purpose: To know and use the criteria of good positioning and bone consolidation of a coracoid bone block. To evaluate the results in terms of positioning and bone consolidation of an arthroscopic technique of Latarjet coracoid bone block.
Methods and Materials: A retrospective monocentric study of 161 patients who underwent a Latarjet coracoid bone block surgery under arthroscopy by a senior operator for management of anterior chronic shoulder instability. All patients were monitored at 15 days and 6 months by CT-scan. The positioning of the stop was evaluated in the horizontal and vertical plane. Bone consolidation was evaluated at 6 months.
Results: The bone block was sub-equatorial in 151 (93%) patients, equatorial in 9 (5%) and supra-equatorial in 2 (1%). The bone block was flush in 153 patients (94%), overflowing in 8 (4.5%) and medial in 1 patient (0.5%). At 6 months postoperative, complete bone consolidation was obtained in 147 patients (90.5%). 14 (9%) had a bone block pseudarthrosis, one a fractured of the bone block, one a complete lysis and one a bone block migration. The clinical prognostic factor of lack of consolidation was smoking. Five patients had recurrent instability.
Conclusion: CT-scan is an easy and reproducible tool for postoperative osseous evaluation of Latarjet arthroscopic surgery.This technique gives good results in terms of positioning and consolidation of the coracoid bone block.

B-1026
09:26
Volumetric bone mineral density assessment of the lumbar spine using a novel phantomless dual-energy CT postprocessing algorithm in comparison to dual x-ray absorptiometry
C. Booz1, P. C. Hofmann2, M. Sedlmair2, T. Flohr2, B. Schmidt2, S. S. Martin1, D. Leithner1, T. J. Vogl1, J. L. Wichmann1; 1Frankfurt a. Main/DE 2Forchheim/DE
Abstract

Purpose: Current techniques for the evaluation of bone mineral density (BMD) commonly require phantom calibration. The purpose of this study was to evaluate a novel algorithm for phantomless in vivo dual-energy computed tomography (DECT)-based volumetric assessment of BMD of the lumbar spine in comparison to dual x-ray absorptiometry (DXA).
Methods and Materials: This retrospective study was approved by the institutional review board. The requirement to obtain informed consent was waived. Data from clinically indicated DECT and DXA examinations within 2 months of 47 patients were evaluated. Using a novel automated dedicated postprocessing algorithm for DECT, the trabecular bone of lumbar vertebrae L1-L4 was analysed after five volumes of interest on different slices had been manually defined for each vertebral body. Pearson product-moment correlation, Bland-Altman plot and further regression analyses were calculated for statistical analysis.
Results: A total of 186 lumbar vertebrae in 47 patients (24 male, 23 female) were analysed. Mean BMD of L1-L4 determined by DXA was 0.986 g/cm², and 20 patients (42.6%) showed an osteoporotic BMD according to guidelines of the World Health Organisation. Average DECT-based BMD of L1-L4 was 88.68 mg/cm³. According to CT guidelines of the American College of Radiology DECT measurements identified seventeen patients (36%) with an osteoporotic BMD. Statistical analysis showed a moderate correlation between volumetric and areal BMD values based on DECT and DXA with a Pearson's product-moment correlation coefficient of r=0.421.
Conclusion: A novel DECT-based postprocessing algorithm allows for phantomless volumetric BMD assessment of the trabecular bone of lumbar vertebrae.

B-1027
09:30
Learning curve of an ultrasound-guided percutaneous release of carpal tunnel: a cadaveric study
O. Andreani, C. Dekimpe, C.-P. Raffaelli, N. Amoretti; Nice/FR
Abstract

Purpose: The aim of our study was to evaluate the feasibility and the learning curve of an ultrasound-guided percutaneous carpal tunnel release.
Methods and Materials: 14 carpal tunnel releases were carried out, on un-embalmed cadavers, 7 by a senior and 7 by a junior radiologist. Sonographic evaluation, with 18 MHz probe, was first performed to detect the anatomical variant and length of the retinaculum. After hydrodissection by Xylocaine of the carpal tunnel, we used an Acufex knife (2,5 mm; Smith and Nephew®) to cut the retinaculum, under ultrasonographic guidance. We evaluated the time (from the first injection of Xylocaine until the knife was taken off); success of the procedure: complete release of retinaculum and absence of complication (lesion of the median nerve and its branch, especially the thenar motor branch, or vascular structure, especially the arterial palmar arch), checked by dissection after procedure; and size of cutaneous incision.
Results: The mean time of the procedure was 14 minutes. (11 min senior versus 17 min junior). The section was complete at the 4th procedure for the senior and at the 5th procedure for the junior radiologist. No lesion of the median nerve or vascular structure was observed. The release was incomplete in 7 firsts’ wrists, due to an incomplete section of the distal retinaculum (lack of 10 mm on average). Several passages were associated with more frequent complete section. The mean size of the incision was 3 mm.
Conclusion: We observed a rapid learning curve of carpal tunnel release with ultrasound guidance, for a senior or junior radiologist, without difference in the term of success. This procedure seems to be safe, reproducible and minimally invasive.

B-1028
09:34
The accuracy of ultrasound with and without sonoelastography in the diagnosis of partial-thickness rotator cuff tears with MDCT arthrography and arthroscopic verification
V. Gazhonova, M. Emelianenko, M. Onishchenko; Moscow/RU
Abstract

Purpose: To compare the diagnostic accuracy of conventional US and sonoelastography (SE) for the diagnosis of partial-thickness rotator cuff tears (RCTs) with MDCT-arthrography and arthroscopic correlation.
Methods and Materials: 50 shoulders with suspected partial RCT based on MRI or US findings were assessed with conventional US and SE. SE images were evaluated by reviewers using experimentally proven colour grading system (grade 2, red colour - RCT, grade 1, green - tendonitis, grade 0, blue - normal). Strain ratios (SR) in the affected region of the tendon were calculated in comparison to the deltoid muscle. MDCT arthrography and arthroscopy were used as reference methods for verification.
Results: Using conventional US, 75% (24 of 32) of the partial RCTs were found correctly. Using SE, 28 RC tendons were categorised as grade 2, 15 as grade 1, and 7 as grade 0. SR in RCT regions <1.2 was in 30 shoulders, and SR >1.2 in 20 patients. Evaluation of interobserver reliability of the SE findings showed “almost perfect agreement” with a weighted kappa coefficient of 0.83 for colour SE grades. By comparing SE with the reference methods, colour grading system had a positive correlation (r=0.84, p=<0.001), and SR values <1.2 were characteristic for RCT with positive correlation (r=0.93, p=<0.001). In diagnosing a partial RCT, SE was more sensitive than conventional US (p<0.05).
Conclusion: SE using colour grading scale and quantification of SR values in the affected tendon’s regions is a more sensitive and specific technique for diagnosing partial-thickness RCT than conventional US.

B-1029
09:38
Iterative reconstruction improves image quality in virtual non-calcium images of the spine for the detection of bone marrow edema in patients with vertebral compression fractures
N. Engelhard1, K.-G. Hermann1, M. Fuchs1, M. Pumberger1, M. Putzier1, J. Mews2, B. Hamm1, T. Diekhoff1; 1Berlin/DE 2Neuss/DE
Abstract

Purpose: Dual-energy computed tomography (DECT) allows detection of bone marrow edema (BME) after vertebral compression fractures using virtual non-calcium imaging (VNCa). The aim of this study was to evaluate, which reconstruction offers the best image quality.
Methods and Materials: 18 patients over 50 years of age with available CT raw-data were retrospectively included. DECT was performed on a single-source machine (80/135 kVp sequential volume acquisition) and reconstructed using filtered back projection (FBP) and iterative reconstructions (IR) with three different iteration levels (IR1-IR3). VNCa images were generated using a raw-data (RD) and an image-data (ID) based software. Target vertebras (TV) with fracture and BME in STIR sequence and normal appearing reference vertebras (RV) were defined. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated. Intervertebral ratio (IVR) served as measurement for the detectability of BME. Friedmann-Test with multiple comparisons was applied.
Results: 39 TV and 18 RV were included. RD showed significantly inferior SNR (p = 0.0001-0.0003) compared to ID. CNR differences were only significant for FBP vs IR3 (p = 0.002). In ID, IR gradually improved SNR values, however, differences were only significant for FBP vs IR3 (p = 0.005). For CNR there was no significant difference. IVR improved with increasing iteration levels, however, differences were only significant for the comparison to FBP (p = 0.03-0.0008). We found no differences of IVR comparing RD and ID.
Conclusion: IR improved the image quality and detectability of BME in DECT. The ID was superior to the RD based approach.

B-1030
09:42
Quantitative MRI for knee osteoarthritis: a rising star
S. Eijgenraam1, A. Chaudhari2, G. Gold2, M. Reijman1, E. H. G. Oei1, B. Hargreaves2; 1Rotterdam/NL 2Stanford/US
Abstract

Purpose: Quantitative MRI measures such as T2 relaxation times of cartilage and meniscus, and semi-quantitative assessments such as MRI Osteoarthritis (OA) Knee Score (MOAKS), are commonly used to track spatial and temporal changes of knee-OA. Acquiring these biomarkers currently requires multiple MR sequences, resulting in long acquisition-times. We evaluated the double-echo steady-state (DESS) sequence with an acquisition-time of only five minutes for simultaneous T2 relaxometry of cartilage and meniscus, and MOAKS scoring, in patients with no, mild and moderate knee-OA.
Methods and Materials: 54 patients (20, 18 and 16 with Kellgren-Lawrence (KL) grades 0, 2, and 3 resp.) were scanned using DESS and a clinical multisequence knee-protocol. MOAKS was performed with only DESS (with multiplanar reformatting). Mean T2-values of cartilage and meniscus were calculated on single slices. Statistical testing between KL-groups was performed using Students T-Tests. Correlation between T2-values and MOAKS was assessed using linear regression analysis.
Results: Mean T2-values in cartilage, obtained with DESS, were 37.7 ± 4.1, 44 ± 8 and 50.8 ± 8.4 ms for KL0, KL2 and KL3 resp. (p < 0.001). In menisci, mean T2-values of 15.4 ± 3.3, 18 ± 3.3 and 21 ± 6.4 ms were found for KL0, KL2 and KL3 resp. (p < 0.001). T2-values showed a good correlation with corresponding MOAKS-cartilage findings (r = 0.61, p = 0.002).
Conclusion: Simultaneous quantitative T2 and morphological assessment of cartilage and meniscus with a 5-minute DESS-sequence show consistent outcomes with increasing stages of degeneration, making this sequence a promising tool for OA research.

B-1031
09:46
Correlation of body mass index with paraspinal muscle fatty degeneration in non-diabetic patients with lumbar spinal canal stenosis: results from 685 patients
S. Winklhofer1, J. Burgstaller1, U. Held1, T. Finkenstaedt1, F. Del Grande2, G. Andreisek3, J. Steurer1, N. Bolog4; 1Zurich/CH 2Lugano/CH 3Münsterlingen/CH 4Munchenstein/CH
Abstract

Purpose: Purpose of this study was to investigate the correlation between lumbar muscle fatty degeneration assessed by magnetic resonance imaging (MRI) and body mass index (BMI) in a large non-diabetic patient group with lumbar spinal stenosis (LSS).
Methods and Materials: MRI images of 685 patients (52% female, mean age 74 years, interquartile range 67 - 80 years) with LSS were analysed in this institutional review board-approved study. Muscle degeneration was assessed by two independent readers on axial T2 weighted images using a modified 5-grades Goutallier classification. Th correlation between muscle degeneration and BMI was calculated using the non-parametric Spearman coefficient test.
Results: The following distribution of muscle degeneration was found in the 685 patients: grade 0: 136 (20%), grade 1: 265 (39%), grade 2: 220 (32%), grade 3: 41 (6%), grade 4: 23 (3%). The mean BMI was 26.6 kg/m2, interquartile range 23.8 to 29.7. Correlation analysis between BMI and the muscle stage degeneration in 5 ordinal categories was 0.060 for female patients (p = 0.26), 0.148 for male patients (p = 0.007), and 0.067 for all patients (p = 0.08).
Conclusion: The weak correlation between BMI and the degree of muscle degeneration indicates that overweight might not be an important potential source of paraspinal muscle degeneration in non-diabetic patients with LSS.

This website uses cookies. Learn more