Radioloog van de toekomst programma & abstracts. Radiologendagen september Congrescentrum. Brabanthallen

Methods & materials: We included pairs of screening radiologists with a minimum of 7500 screening examinations per pair, obtained at a Dutch screening mammography region between January 1st 1997 and December 31st 2009. During 2-year follow-up, breast imaging reports, surgical reports and pathology results were collected of all screen detected cancers and interval cancers. Screening outcome parameters were calculated for each screening couple. 1 7 E

positive predictive value of referral and biopsy. Extra tumors detected at FFDM were mostly low-intermediate grade DCIS and smaller invasive tumors, of more favourable tumor characteristics. Mastectomy rates were not increased in the FFDM population, while increased overdiagnosis cannot be excluded.

Results: A minimum of 7500 screens for each pair of radiologists had been read by 27 unique combinations of two screening radiologists, totaling 313038 screens. The number of screening mammograms per pair of radiologists varied from 7500 screens to 20761 screens. Breast cancer incidence did not differ significantly among these 27 couples and ranged from 5.3 (95%CI 3.9-6.8) to 8.9 (95%CI 7.0-10.7) per 1000 screens. The referral rate among couples ranged from 0.9% (95%CI 0.7-1.1%) to 1.5% (95%CI 1.3%-1.8%), whereas the sensitivity for breast cancer detection and the positive predictive value of referral ranged from 55.9% (95%CI 45.6%-66.1%) to 82.8% (95%CI 73.3%-92.3%) and from 28.9% (95%CI 21.0%-36.9%) to 50.6% (95%CI 40.0%-61.2%), respectively. Conclusions: Screening outcome varied significantly among pairs of screening radiologists. We advise a continuous monitoring of screening outcome parameters for each screening couple that participates at screening mammography. Regular feedback of this outcome to each couple may avoid screening of women by pairs of radiologists with suboptimal screening performance.

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

29

4 programma & abstracts Abstracts O04.05 NON-INVASIVE NODAL STAGING IN BREAST CANCER PATIENT WITH GADOFOSVESETENHANCED MAGNETIC RESONANCE IMAGING M.B.I. Lobbes, R.J. Schipper, L. van Rozendaal, C. Castro, B. de Vries, L. Prompers, J. Wildberger, R.G.H. Beets-Tan, M. Smidt Maastricht Universitair Medisch Centrum, Maastricht

30

Histopathological examination was regard as golden standard. Sensitivity, specificity, PPV, and NPV values were calculated.

Aim: Assessing axillary lymph node status is breast cancer patients remains inaccurate. Past studies in rectal cancer have shown promising results in lymph node staging using gadofosveset-enhanced MRI The aim of this feasibility study was to determine the accuracy of gadofosveset enhanced MRI for axillary lymph node staging.

Results: A total of 152 lymph nodes were extracted during surgery of which 116 were matched with lymph nodes depicted on MRI. Histopathology showed 21 macro and 8 micro-metastases. With MRI, 20 lymph nodes were rated as true positive, 83 as true-negative, 4 as false-positive, and 9 as false-negative; resulting in an overall sensitivity, specificity, PPV and NPV (95 % confidence interval in parentheses) of respectively 67 % (49-80), 95 % (89-98), 83% (64-93) and 89 % (81-94). If ignoring micro-metastases, MRI showed a sensitivity of 87 % 68-96) and a specificity of 96 % (89-98). Calculated PPV and NPV was 83 % (64-93) and 97 % (90-99).

Materials & methods: Ten women with invasive breast cancer (> 2 cm) underwent a gadofosveset-enhanced 3D T1 weighted MRI. A radiologist evaluated the nodal status of all lymph nodes. All MRI depicted lymph nodes were matched with the lymph nodes removed during surgery.

Conclusion: This feasibility study concerning gadofosvesetenhanced MRI of axillary lymph nodes in breast cancer patients shows promising initial results. Nevertheless, more studies are warranted to further explore these promising results.

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

4

Mammaradiologie (deel 1) O04.06 DYNAMIC CONTRAST-ENHANCED MRI OF THE BREAST AT 7T AND 3T; INITIAL RESULTS OF AN INTRA-INDIVIDUAL COMPARISON OF BI-RADS-MRI LESION ASSESSMENT B.L. Stehouwer, D.W.J. Klomp, P.R. Luijten, W.P.Th.M. Mali, M.A.A.J. van den Bosch, W.B. Veldhuis Universitair Medisch Centrum Utrecht, Utrecht Purpose: To intra-individually compare BI-RADS-MRI classification results on 7T and 3T dynamic contrastenhanced Breast MRI. Materials & methods: Six breast cancer patients were submitted to an investigational 7T MRI examination and a clinical 3T MRI examination conducted according to hospital guidelines. 7T imaging was performed using a two-channel unilateral breast coil. The scan protocol included a dynamic series with the injection of 0.1mmol/kg Gadobutrol [voxel 1x1x2mm3, temporal resolution 63s.]. The 3T scan protocol included a dynamic series with injection of 0.1mmol/kg Gadobutrol [voxel 1.1x1.1x2.4mm3, temporal resolution

60s.]. Two radiologists (R1 and R2) independently scored all exams, without clinical information. Image quality was scored using four categories: excellent, good, sufficient, or insufficient. Lesion analysis was conducted using the BI-RADS-MRI lexicon. Results: 7T Image quality was scored as good in n=3 (R1) and n=4 (R2); sufficient in n=3 (R1) and n=2 (R2). Image quality at 3T was scored as excellent in n=2 (R1) and n=0 (R2); good in n=3 (R1) and n=5 (R2), and sufficient in n=1 (R1) n=1 (R2). At both field strengths, all biopsy-proven malignant lesions were identified and assessed using the BI-RADSMRI lexicon (Table 1). Conclusion: All six malignant lesions were prospectively identified at both field strengths by both observers. An inter-observer variability in the application of BI-RADS-MRI descriptors was seen. However, this did not affect the final assessment categories. This intra-individual validation of 7T versus 3T MRI in breast cancer patients paves the way for coming clinical trials to implement 7T MRI and benefit from ultra-high field breast imaging and spectroscopy.

Table 1: BI-RADS-MRI assessment of all lesions by two observers. Case Lesion type Shape Distribution Margin Enhancement Initial rise Delayed phase Category 1 7T mass irregular lobular spiculated irregular heterogeneous rapid washout 5 3T mass irregular spiculated heterogeneous rapid washout 5 2 7T mass irregular spiculated heterogeneous rapid washout 5 3T mass round irregular irregular heterogeneous rapid plateau 45 3 7T mass irregular irregular heterogeneous rapid washout 5 3T mass irregular spiculated heterogeneous rapid washout 5 7T non-mass ductal clumped rapid washout 5 3T mass lobular irregular heterogeneous rapid washout 4 3 7T non-mass segmental clumped rapid washout 5 3T non-mass segmental heterogeneous rapid washout 5 4 7T mass irregular spiculated heterogeneous rapid washout 5 3T mass irregular spiculated rim-enhancement medium rapid washout plateau 5 5 7T mass irregular oval irregular heterogeneous rapid washout 5 3T mass irregular lobular irregular heterogeneous rapid washout 5 washout 5 6 7T mass irregular irregular heterogeneous rim-enhancement rapid 3T mass irregular spiculated irregular rim-enhancement x x 5 Here the scorers for both observers are displayed. were they scored differently both results are shown one after the other . In case 3 one observer (R1) scored two lesions, namely a mass lesion and a nonmass lesion, while the other observer (R2) scored one non-mass lesion. Therefore this case is displayed seperately for both observers. In case 6 the 3T kinetic curve could not be assessed reliably due to patient movement and was therefore excluded from analysis.

1 7 E

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

31

5 programma & abstracts Abstracts Sessie 5

Neuroradiologie

Donderdag 27 september, 14:00 - 15:15 uur O05.01 INTRACRANIAL CAROTID ARTERY ATHEROSCLEROSIS RELATES TO A HIGHER RISK OF STROKE AND MORTALITY D. Bos, M.W. Vernooij, A. Hofman, J.C.M. Witteman, A. van der Lugt, M.A. Ikram Erasmus MC, Rotterdam Purpose: Intracranial atherosclerosis is one of the leading causes of stroke worldwide. Compared to other ethnic groups, little is known about the risk of stroke and mortality due to intracranial atherosclerosis in Caucasians. Hence, we investigated the relationship between intracranial carotid artery calcification, as proxy of intracranial atherosclerosis, and the risk of stroke and mortality in a Caucasian population.

Results: Intracranial carotid artery calcification was strongly associated with a history of stroke [multivariableadjusted odds ratio (OR) per standard deviation increase in calcification volume: 1.52 (95%CI: 1.11;2.07)]. It was also associated with a higher risk of stroke [multivariableadjusted hazard ratio (HR): 1.48 (95%CI: 1.03;2.11)], especially ischemic stroke [multivariable-adjusted HR: 1.59 (95%CI: 1.06;2.40)]. Finally, a larger intracranial carotid artery calcification volume was related to a higher mortality risk [HR: 1.34 (95%CI: 1.16;1.55)].

Methods: From the prospective, population-based Rotterdam Study, 2,496 participants (mean age: 69.6 years) underwent CT of the intracranial carotid arteries to quantify calcification volume. Information on history of stroke was assessed at the time of CT. Participants were followed until January 1, 2009 for incident strokes (n=53)

Conclusion: Intracranial carotid artery atherosclerosis in Caucasians is an important, independent risk factor for stroke. Furthermore, it is also associated with a higher risk of mortality. This suggests a considerable impact of intracranial carotid artery atherosclerosis in Caucasians.

O05.02 POOR COLLATERAL STATUS ON TIMINGINVARIANT CTA IS A STRONG PREDICTOR OF POOR CLINICAL OUTCOME IN ACUTE STROKE PATIENTS WITH LARGE VESSEL OCCLUSION E.J. Smit1, E.J. Vonken1, T. van Seeters1, J.W. Dankbaar1, I.C. van der Schaaf1, B. van Ginneken2, B. Velthuis1, M. Prokop2 1 Universitair Medisch Centrum Utrecht, Utrecht 2 Radboud University Nijmegen Medical Centre, Nijmegen

ischaemic stroke patients with ICA and/or MCA occlusion from our clinical database. CTA and CT perfusion (CTP) at admission and clinical outcome data (modified Rankin Scale, mRS) after three months were available. CTP source images were used to reconstruct timing-invariant CTA (TI-CTA), a technique that is insensitive to delayed contrast arrival. Four experienced observers assessed collateral status on CTA and TI-CTA in an independent, blinded, randomised manner. Collateral status was rated good if ≥50% of collaterals were present or poor if <50% were present. Clinical outcome was defined as good for mRS ≤2 or poor for mRS >2.

Purpose: To assess whether collateral status in acute stroke patients with large vessel occlusion predicts clinical outcome and to compare standard CT angiography (CTA) to a timinginvariant CTA for this purpose.

Results: Collateral status was rated significantly more often as good on TI-CTA (84%) than on CTA (49%; p<0.05). Inter- and intraobserver agreement were good for assessing collateral status on CTA and TI-CTA (mean kappa = 0.63 and 0.76, respectively). Poor collateral status on TI-CTA had a predictive value of 100 % for poor outcome. This number was

Methods & materials: We selected 40 consecutive 32

or death (n=137). Logistic and Cox regression-models were used to assess relationships between intracranial carotid artery calcification, history of stroke, the risk of stroke and mortality. All analyses were additionally adjusted for relevant confounders.

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

5

Neuroradiologie only 69 % for CTA. Good collateral status on CTA or TI-CTA predicted good outcome in only 64 % and 56 %, respectively. Conclusion: Poor collaterals status on timing-invariant CTA is a strong predictor of poor clinical outcome in acute stroke patients with large vessel occlusion. TI-CTA detects more collaterals than standard CTA due to its insensitivity to delayed contrast arrival.

Figure 1: Patient with right sided MCA occlusion and good clinical outcome showing poor collaterals on standard CTA but good collaterals on timing-invariant CTA.

O05.03 MULTISEQUENCE INTRACRANIAL VESSEL WALL IMAGING AT 7.0 TESLA MRI A.G. van der Kolk1, J.J.M. Zwanenburg1, M. Brundel1, G.J. Biessels1, F. Visser2, P.R. Luijten1, J. Hendrikse1 1 Universitair Medisch Centrum Utrecht, Utrecht 2 Philips Healthcare, Best

visibility of the major cerebral artery walls and their branches for five healthy volunteers, and compared the whole-braincoverage sequences with the MPIR-TSE sequence with 13mm coverage. Signal profiles were drawn for comparison of contrasts between sequences. For clinical applicability, two patients with cerebrovascular disease were imaged.

Background & purpose: Intracranial vessel wall imaging with the small-coverage 3D MPIR-TSE sequence may show intracranial atherosclerosis without luminal narrowing. However, for clinical implementation, larger coverage and multiple sequences for intracranial plaque characterization are warranted. In the current study we developed a multisequence MRI protocol with whole-brain coverage for vessel wall imaging at 7.0 Tesla MRI.

Results: Best conspicuity scores were given for the smallcoverage sequence, followed by the whole-brain-coverage T1-, T2 and PD-weighted sequences. Mean MRI signal intensity ratios of all vessel wall sequences compared to surrounding cerebrospinal fluid / blood were 2.6, 2.2, 2.2 and 2.0 for the small-coverage sequence, whole-brain-coverage PD-, T2-weighted, and T1-weighted sequence, respectively. These ratios increased in patients, due to age- or diseaserelated vessel wall thickening (Figure).

Methods: A modified MPIR-TSE sequence with nonselective short RF pulses and SENSE in two directions was used to obtain proton density (PD)-weighting (6167/1795/38ms), T1-weighting (3952/1375/38ms), and T2-weighting (8000/2200/287ms). All sequences had 190mm whole-brain coverage. Four observers independently scored

Discussion: With the whole-brain multisequence vessel wall protocol we have shown the possibility of assessing intracranial arterial vessel wall with multiple different contrast weightings. In future, these sequences may be able to characterize intracranial vessel wall pathology similar to assessment of carotid artery plaques.

Figure 1: 7.0 Tesla smallcoverage- (a), and whole-brain T1-weighted MPIR-TSE images before (b) and after (c-h) contrast administration of vasculitis patient.

1 7 E

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

33

5 programma & abstracts Abstracts O05.04 BLOOD-BRAIN-BARRIER LEAKAGE ON 7.0 TESLA MRI IN PATIENTS AFTER TRANSIENT ISCHEMIC ATTACK AND STROKE D. Sucha, A.G. van der Kolk, M. Brundel, G.J. Biessels, P.R. Luijten, J. Hendrikse Universitair Medisch Centrum Utrecht, Utrecht Purpose: Hemorrhagic transformation (HT) of ischemic stroke is associated with poor outcome and related to increased blood-brain-barrier(BBB)-permeability, which can be visualized indirectly by contrast-enhanced imaging, showing parenchymal enhancement (PE) on T1-weighted and hyperintense acute reperfusion marker (HARM) on FluidAttenuation Inversion Recovery (FLAIR)-images. The aim of this study was to evaluate frequency and distribution of contrast enhancement (CE) with high contrast-to-noise at 7.0 Tesla (7T). Methods: A retrospective analysis was performed on 7T images of patients with transient ischemic attack (TIA) or ischemic stroke and post-gadolinium FLAIR-imaging. All

patients were scanned within one week after symptom onset. Follow-up imaging was performed after one month. T1- and T2*-weighted images were evaluated for presence of PE; the latter were also used to assess presence of HT. FLAIR-images were evaluated for presence of HARM. Results: Thirty-five patients (14 TIA and 21 stroke) were included (mean age 57 years). Within one week, HARM was present in 2 (14%) TIA- and 14 (67%) stroke patients, whereas PE was present in 0 (0%) TIA- and 17 (81%) stroke patients. In total, 2 TIA- and 19 stroke patients showed BBBleakage. Subsequent HT only occurred in stroke patients; all HT-patients showed PE (11) and 8 (73%) HARM. Of 15 patients with follow-up, HARM remained in 1 TIA- and 3 stroke patients (80%), PE in 4 stroke patients (100%). Discussion: In this 7T study assessing BBB-leakage markers, PE showed to be a possible marker for increased risk of HT. In addition, BBB-leakage did not only occur in stroke- but also in TIA patients, and stayed visible in almost all follow-up patients. Longer follow-up studies may possibly clarify the duration of BBB-leakage after ischemic events.

Figure 1: FLAIR- (A), T2*weighted- (B) and T1-weighted (C) image of 27-year-old stroke patient, showing respectively HARM (A), HT (B) and PE (B+C).

O05.05 ARTERIAL SPIN LABELING PERFUSION MAGNETIC RESONANCE (MR) IMAGING CONTRIBUTES TO THE EARLY DIAGNOSIS OF DEMENTIA R.M.E. Steketee, S. Ooms, M. Luijten, G.C. Houston, J.C. van Swieten, M. Smits Erasmus MC, Rotterdam Purpose: Early diagnosis of dementia is challenging as brain atrophy may not yet be apparent. Changes in cerebral blood flow (CBF) preceding atrophy may be detected with arterial spin labeling (ASL). This study aims to assess ASL for diagnosing early dementia. Methods & materials: Fifteen patients with suspected diagnosis of dementia (mean age 63y, 7 male, mean Mini Mental State Examination 26) and 15 controls (mean age 59y, 34

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

8 male) underwent 3D PCASL and T1w MR scanning at 3T (GE Healthcare, US). Data were coregistered and normalised using SPM8 (London, UK). ROI analysis of the CBF maps, masked for grey matter only, was performed with MarsBaR (Marseilles, FR), in regions involved in dementia and regions initially spared. Mean CBF values were compared between groups using 2-sample t-tests. Results: Age and gender were not different between groups (p>0.05). Mean CBF values in dementia related brain regions were significantly lower in patients than controls (p<0.05, right/left medial temporal lobe: 30/30 versus 37/38; right/left precuneus: 29/29 versus 43/42; right/left posterior cingulum: 35/34 versus 56/58 ml/100cc GM/min). No difference was found in regions not implicated in dementia (right/left precentral gyrus: 29/30 versus 35/36; right/left occipital gyri: 24/25 versus 31/29 ml/100cc GM/min for patients and controls respectively).

5

Neuroradiologie Conclusion: Patterns of hypoperfusion assessed with ASL in early dementia are concordant with those in established dementia and those assessed with positron emission

tomography, indicating the contribution of ASL in the early diagnosis of dementia. Future aims are to assess ASL for early differentiation between types of dementia.

Regions of interest

2

1

1 3 2

1

Figure 1: Mean CBF values were compared in ROIs (A) known to be associated with or (B) being initially spared in dementia.

Mean CBF in ROIs

1

1

1

2 2 3

Figure 2: Mean CBF values plotted per ROI for patients and controls.

CBF maps patient

CBF maps control

Figure 3: Representative examples of ASL-CBF maps, showing the typical parietotemporal hypoperfusion in the patient (left) and not in the control (right).

1 7 E

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

35

6 programma & abstracts Abstracts Sessie 6

Kinder- en Thoraxradiologie Donderdag 27 september, 14:00 - 15:15 uur O06.01 ARTERIAL SPIN LABELING MAGNETIC RESONANCE IMAGING FOR ASSESSMENT OF NEONATAL BRAIN PERFUSION J.B. de Vis, M.J.N.L. Benders, L.S. de Vries, F. Groenendaal, K.J. Kersbergen, T. Alderliesten, J. Hendrikse, E.T. Petersen Universitair Medisch Centrum Utrecht, Utrecht Purpose: Little is known about brain perfusion in neonates and its relation to brain development. The purpose of this study was to evaluate the relation between perfusion and brain development using Arterial Spin Labeling (ASL) MRI. Materials & methods: Pulsed ASL images (Philips-3T) were acquired in 31 infants; TR/TE/TI: 2500/20/1500 msec. Six infants were imaged at preterm age, 23 infants at term equivalent age (TEA) and 2 infants at 3-months equivalent age (3m). Serial MR imaging was performed in 4 of these infants. Total brain perfusion was measured for each infant. Regions of interest (ROIs) were drawn within the occipital cortex (OC), the frontal cortex (FC) and one ROI covered

Figure 1 36

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

Figure 2

6

Kinder- en Thoraxradiologie Discussion & conclusion: Arterial Spin Labeling MR images reflect the anatomical and functional maturation of the brain and are in agreement with previously obtained PET studies [1,2].

the basal ganglia and thalamus (BGT). Perfusion values measured in the ROIs were expressed relative to total brain perfusion. Results: Total brain perfusion increased from preterm age (7,1 ml/100g.min) to TEA (12.6 ml/100g.min) to 3m (30.2 ml/100g.min). A relative decrease in perfusion towards the BGT and a relative increase towards the OC and FC was measured with increasing postconceptional age (fig. 1). Perfusion images made at preterm age (A), at TEA (B) and at 3m (C) are shown in figure 2.

References: [1] Chugani et al. Prev Med 1998; 27(2): 184188. [2] Kinnala et al. Archives of disease in childhood fetal and neonatal edition 1996; 74.

O06.02 DE INVLOED VAN TOEGENOMEN BEELDVORMING MET INTRODUCTIE VAN MRI BIJ KINDEREN VERDACHT VOOR ACUTE APPENDICITIS MET EEN INCONCLUSIEVE OF NEGATIEVE ECHOGRAFIE E.R. Langedijk1, A. Meij-de Vries1, K.J. Ponsen1, J. Stoker2, B.M. Wiarda1 1 Medisch Centrum Alkmaar, Alkmaar 2 Academisch Medisch Centrum, Amsterdam Doel: Analyseren van gevolgen van toename in beeldvorming met introductie van MRI bij kinderen met verdenking appendicitis en een inconclusieve/negatieve echografie. Methoden: Retrospectief onderzoek met vergelijking van twee perioden kinderen (3-18 jaar) verdacht van acute appendicitis: 2005/2006 (voor introductie van MRI bij negatieve of inconclusieve echografie) en 2010/2011 (na introductie van MRI). De bevindingen bij beeldvorming, diagnose, aantal perforaties, aantal negatieve diagnostische laparoscopieën en opnameduur werden vergeleken. Referentiestandaard: diagnose bij operatie/pathologie en follow-up.

O06.03 PIXEL-BY-PIXEL ANALYSIS OF DCE-MRI CURVE SHAPES IN KNEES OF JUVENILE IDIOPATHIC ARTHRITIS PATIENTS R. Hemke1, C. Lavini1, M.A.J. van Rossum1, J.M. van den Berg1, K.M. Dolman2, D. Schonenberg1, C.M. Nusman1, T.W. Kuijpers1, M. Maas1 1 Academisch Medisch Centrum, Amsterdam 2 Reade, Amsterdam Purpose: To compare the relative number of time-intensity curve (TIC) shapes as derived from dynamic contrast enhanced (DCE) MRI in clinical active and clinical inactive juvenile idiopathic arthritis (JIA) patients.

1 7 E

Resultaten: In 2005/2006 werden 276 patiënten geïncludeerd (♀ 124, gemid. leeftijd 11.6 jaar SD 3.7),in 2010/2011265 (♀ 113, gemid. leeftijd 12.0 jaar SD 3.6). Groepen waren vergelijkbaar in aantal patiënten, leeftijd, geslacht, aanwezigheid van appendicitis (184 (66,7%) versus 189 (71,3%) P=0.211), aantal operaties( n=208 versus n=200, P=0.976). In 2010/2011 ondergingen 260 patiënten echografie tegenover 210 in2005/2006 ( P < 0.0001). Er was geen significantie verschil tussen aantal inconclusieve/ negatieve echografieën (n=98 versus n=117, P=0.718), wel in aanvullende MRI (n=1 versus n=64, P<0,0001). In periode 2005/2006 waren er significant meer perforaties (n=56 versus n=34; P=0.0217), negatieve diagnostische laparoscopieën (n=9 versus n=1, P=0.0302) en langere opnameduur (4,0 SD 4,6 versus 2,7 SD 2,3; P=0.001). Sensitiviteit en specificiteit van echografie in 2005/2006 was 79% (95%CI: 70.2-85.2) en 60% (95%CI: 27.4- 86.3), in 2010/2011 74% (95%CI: 67-80.1) and 67% (95%CI: 30.9-90.9). Die van MRI in 2010/2011 was 95% (95%CI: 80.9-99) en 88% (95%CI: 67.7-96.8). Conclusie: Toename in beeldvorming met introductie van MRI bij inconclusieve of negatieve echografie suggereert de verklaring voor afname van perforaties, negatieve diagnostische laparoscopieen en kortere opnameduur.

Methods: DCE-MRI datasets of knees of 56 JIA patients (64.3% female, mean age 13.3 years (SD 2.6)) were prospectively obtained using an open-bore magnet (1T). Patients were classified into two clinical subgroups: active arthritis (n=36) and inactive disease (n=20). Every voxel with its TIC was classified into one of seven predefined TIC shape categories, which resulted in a color-coded shape map. Spatial information of the synovial TIC shape distribution pattern and relative number of TIC shapes were calculated on a three-dimensional region of interest. Relative TIC shape numbers were compared by using a nonparametric MannWhitney U test. Results: No differences between active and inactive JIA patients were found regarding the relative number of type 2 R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

37

6 programma & abstracts Abstracts TIC shapes (slow enhancement) and type 3 TIC shapes (fast enhancement followed by plateau phase) (P=0.726, P=0.973 respectively). However a statistically significant higher relative number of type 4 TIC shapes (fast enhancement followed by early washout phase) was seen in clinical active patients compared with the inactive group (7.5% vs. 5.1%, P=0.034). Type 5 TIC shapes (fast enhancement followed by

slow enhancement increase) tended to be more present in clinical inactive patients compared with active JIA patients (16.1% vs. 13.4%, P=0.064).

O06.04 WHOLE-BODY MRI, INCLUDING DWI, COMPARED TO FDG-PET/CT FOR STAGING MALIGNANT LYMPHOMA IN CHILDREN M.A. Vermoolen1, T.C. Kwee1, F.J. Beek1, B. de Keizer1, I. Barber2, C. Granata3, A.S. Littooij4, J.A. Adam5, J. Zsiros5, M.B. Bierings1, R.A.J. Nievelstein1 1 Universitair Medisch Centrum Utrecht, Utrecht 2 Vall d’Hebron University Hospital, Barcelona, Spain 3 Giannina Gaslini Children’s Hospital, Genova, Italy 4 KK Women’s and Children’s Hospital, Singapore, Singapore 5 Academisch Medisch Centrum, Amsterdam

years; age range, 5-16 years) with malignant lymphoma prospectively underwent whole-body MRI (T1-weighted and T2-STIR [n=25], and DWI [n=24]) and FDG-PET/CT. Wholebody MRI and FDG-PET/CT were evaluated by different observers, who were blinded to the other imaging modality. Ann Arbor stages according to whole-body MRI (without and with DWI) were compared to those of FDG-PET/CT.

Purpose: Whole-body magnetic resonance imaging (MRI), including diffusion-weighted imaging (DWI), is emerging as a radiation-free method for oncological staging. This study aimed to compare whole-body MRI, including DWI, to 18 F-fluorodeoxyglucose positron emission tomography with low-dose computed tomography (FDG-PET/CT) for staging newly diagnosed malignant lymphoma. Methods: 25 children (male/female, 14/11; mean age, 12.9

O06.05 PREDICTION OF CARDIOVASCULAR RISK USING LUNG CANCER SCREENING CT O.M. Mets1, R. Vliegenthart2, M.J. Gondrie1, M.A. Viergever4, M. Oudkerk2, H.J. de Koning5, W.P.Th.M. Mali1, M. Prokop3, R.J. van Klaveren6, Y. van der Graaf7, C.F.M. Buckens7, P. Zanen8, J.W.J. Lammers8, H.J.M. Groen9, I. Isgum4, P.A. de Jong1 1 Universitair Medisch Centrum Utrecht, Utrecht 2 Universitair Medisch Centrum Groningen, Groningen 3 Universitair Medisch Centrum St Radboud, Nijmegen 4 Image Sciences Intstitute, Universitair Medisch Centrum Utrecht, Utrecht 5 Maatschappelijke gezondheidszorg, Erasmus MC, Rotterdam 6 Longziekte, Lievensberg Ziekenhuis, Bergen op Zoom 7 Julius Centrum voor Gezondheidszorg en Eerstelijnsgeneeskunde, Universitair Medisch Centrum Utrecht, Utrecht 8 Longziekte, Universitair Medisch Centrum Utrecht, Utrecht 9 Longziekte, Universitair Medisch Centrum Groningen, Groningen 38

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

Conclusion: Results obtained using the pixel-by-pixel DCE-MRI TIC-shape analysis differ between the active and inactive JIA patients with respect to TIC shape 4 and 5.

Results: Staging results of whole-body MRI without DWI were equal/higher/lower to those of FDG-PET/CT in 15, 8, and 2 of 25 patients, respectively. Staging results of wholebody MRI with DWI were equal/higher/lower to those of FDG-PET/CT in 16, 7 and 1 of 24 patients, respectively. Sites of discrepant staging between MRI and FDG-PET/CT were lymph node stations (n=6), spleen (n=2), lung (n=1) and pleura (n=1). Conclusion: Our results indicate that staging using wholebody MRI (without and with DWI) is equal to staging using FDG-PET/CT in little over half of patients. Overstaging by whole-body MRI relative to FDG-PET/CT was more frequent than understaging.

Background: Cardiovascular disease is a major smokingrelated disease, and Computed Tomography (CT)-based lung cancer screening may provide the opportunity to predict cardiovascular risk. Therefore, this study aims to derivate and validate a prediction model for cardiovascular events using coronary and aortic calcium in lung cancer screening chest CT. Material & methods: We collected smoking characteristics, patient demographics and physician diagnosed cardiovascular events in 3,559 subjects from two centers participating in a lung cancer screening trial. Data from 10 years before the CT (ie. cardiovascular history) until 3 years after the CT (ie. follow-up time) was obtained. Age, smoking status, smoking history, cardiovascular history, and automatically quantified coronary and aortic calcium volume were included as independent predictors. Cox proportional hazard analysis was used to derivate and validate a prediction model for cardiovascular events and re-events. Results: Incident cardiovascular events occurred in 145/1834

6

Kinder- en Thoraxradiologie (7.9%) and 118/1725 (6.8%) subjects of the derivation and validation cohort, respectively. After correction for over-optimism the model showed good calibration and discrimination (C-statistic 0.71, 95%CI, 0.67 – 0.76) in the validation cohort. When high-risk was defined as a 3-years risk ≥6%, 589 out of 1725 (34.1%) subjects were regarded as high-risk and 72 out of 118 (61.0%) of all events were correctly predicted by the model.

Conclusion: Cardiovascular risk can be predicted using automatically quantified coronary and aortic calcium volume in lung cancer screening CT. This additional information, which is readily available in the lung cancer screening test, may enhance the cost-effectiveness of CT screening in heavy smokers.

O06.06 BOTSUPPRESSIE IN THORAXFOTO’S VERBETERT DE DETECTIE VAN PULMONALE NODULES DOOR RADIOLOGEN S. Schalekamp1, B. van Ginneken1, E. Koedam1, M.W. Imhof-Tas1, L. Meiss2, L. Peters-Bax1, L.G.B.A. Quekel2, M.M. Snoeren1, A.M. Tiehuis2, R. Wittenberg2, N. Karssemeijer1, C.M. Schaefer-Prokop2 1 Universitair Medisch Centrum St Radboud, Nijmegen 2 Meander Medisch Centrum, Amersfoort

gemiddeld(n=32), moeilijk(n=28), zeer moeilijk(n=16). Commercieel verkrijgbare software (ClearRead Bone Suppression 2.4, Riverain Medical, Miamisburg, Ohio) werd gebruikt om botsuppressie beelden te construeren van de PA thoraxfoto. Lezers gaven verdachte gebieden aan met markers en een daarbij behorende verdachtheidsscore. Dit deden zij met en zonder de hulp van botsuppressie beelden. Statistische analyse bestond uit multi reader multi case receiver operating characteristics (ROC). Een klinisch relevant interval met een specificiteit tussen de 80 en 100% werd gebruikt om de detectie te meten.

Doel: Uit de literatuur is gebleken dat overprojectie van ossale structuren op thoraxfoto’s een onderliggende reden is voor het missen van longkanker in 22 tot 95% van de gevallen. In deze studie wilden we het effect meten van botsuppressie software op de detectie van long nodules in thoraxfotos. Methoden: Posteroanterior en laterale digitale thoraxfoto’s van 108 patienten met een solitaire nodule, en 192 controle patiënten werden beoordeeld door 5 radiologen en 3 artsassistenten. Zowel de nodules, als de afwezigheid van afwijkingen was geverifieerd middels CT. Nodules werden ingedeeld in 4 moeilijkheidscategorien: makkelijk(n=32),

Resultaten: Gemiddelde grootte van de nodules was 17,5mm (mediaan 17mm). ROC analyse liet een verbeterde detectie zien met behulp van botsuppressie (p=0.008). Bij een specificiteit van 90% steeg de detectie van long nodules van 67% zonder botsuppressie naar 72% met botsuppressie. Verbetering werd vooral gezien bij de gemiddeld tot moeilijk zichtbare nodules (van 66% naar 73%). Conclusie: Botsuppressie verbetert de detectie van long nodules in thoraxfoto’s door radiologen, vooral voor nodules met een gemiddelde tot moeilijke zichtbaarheid.

Figuur 1: Links thoraxfoto met nodule in rechter long. Rechts: botsuppressie beeld. Ribben worden onderdrukt, terwijl de nodule beter zichtbaar wordt.

1 7 E

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

39

7 programma & abstracts Abstracts Sessie 7

Educatieve sessie (deel 2) Vrijdag 28 september, 9:45 - 10:57 uur O07.01 PATHOLOGY OF THE THYMUS ON CT-IMAGING R. Peters, S. Braak, J. Verschakelen ZGT, Almelo

thymic hyperplasia, thymomas, thymic carcinoma, thymic non-hodgkin lymphoma, thymolipoma and thymic carcinoid. Radiological imaging is essential to avoid unnecessary invasive examinations such as mediastinoscopy with biopsy.

The thymus is located in the upper anterior mediastinum. A number of masses arise in relation to the thymus. The radiologist has an important role regarding the differential diagnosis between non-tumoral thymic pathology and malignant thymic tumors. In general, a benign hyperplasia of the thymus occurs in children and young adults, while in adults the thymoma is the most common tumor. Furthermore imaging is of great importance in the preoperative staging and oncological follow-up. To evaluate the thymus CT scan is used in the majority of the cases. We present an overview of thymic masses with typically imaging features: Figure 2: Thymoom.

40

Figure 1: Thymolipoom.

Figure 3: Thymuscarcinoom.

O07.02 DE ERVARINGEN MET DE EERSTE 100 CORONAIR CT ANGIOGRAFIE BIJ LAAGINTERMEDIAIRE RISICO PATIëNTEN BIJ EEN SAMENWERKINGSVERBAND TUSSEN DE AFDELINGEN RADIOLOGIE, NUCLEAIRE GENEESKUNDE EN CARDIOLOGIE P.A.C. van Rijn, R.W.F. Geenen, P.R. Algra, C.F. van Dijke, A.A.C.M. Heestermans, J.H. Cornel, R.J.J. Knol, F.M. van der Zant Medisch Centrum Alkmaar, Alkmaar

Doelstelling: Het beschrijven van een samenwerkingsverband tussen de radiologie, cardiologie en nucleaire geneeskunde ter bevordering van de cardiovasculaire diagnostiek.

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

Achtergrond: In 2011 werd er in ons ziekenhuis gestart met de Coronair CT Angiografie (CCTA) als voorbereiding van de cardio-straat. Hierbij wordt een CCTA verricht bij patiënten met een laag tot intermediair risicoprofiel volgens de Duke classificatie. CCTA in combinatie met calciumscore, worden uitgevoerd op een dual source Somatom Definition

7

educatieve sessie (deel 2) Flash scanner (Siemens). Bij een regulaire hartslag onder de 60 slagen/minuut wordt gekozen voor Flash protocol en standaard wordt nitroglycerine sublinguaal verstrekt. De CT-scans worden direct aansluitend beoordeeld. De arts-assistent radiologie maakt na gezamenlijk overleg met de radioloog, nucleair geneeskundige en cardioloog één verslag. Op grond van deze uitslag wordt het verdere beleid (o.a. al dan niet coronair interventie) direct aansluitend met de patiënt besproken.

volgens prospectief protocol. Bij 5 patiënten werd enkel calciumscore vervaardigd en bij 6 patiënten werd een pulmonaal venen protocol verricht.51 patiënten hadden een calciumscore hoger dan 0 (variërend van 0,3 tot 1580). In 39 patiënten werd coronairlijden uitgesloten. Van de 54 patiënten met coronairlijden betrof dit in 3 gevallen uitsluitend niet-gecalcificeerde plaques.De mediane, effectieve radiatiedosis was 1,95 mSv. Conclusie: De integrale benadering van verslaglegging door de arts-assistent radiologie garandeert een eenduidige tripartita verslaglegging.Het samenwerkingsverband tussen specialisten levert efficiencywinst op en is patiëntvriendelijk.

Beschrijving details: Honderd opeenvolgende scans werden beoordeeld bij 99 patiënten (36 mannen en 63 vrouwen, gemiddelde leeftijd 60±10.5 SD jaar). 58 CCTA’s werden uitgevoerd volgens Flash protocol, 31

O07.03 MRI-GUIDED HIGH-INTENSITY FOCUSED ULTRASOUND ABLATION OF BREAST CANCER L.G. Merckel1, K.G.A. Gilhuijs1, R. Deckers1, W.P.Th.M. Mali1, M.O. Köhler2, L.W. Bartels1, M.A.A.J. van den Bosch1 1 Universitair Medisch Centrum Utrecht, Utrecht 2 Philips Healthcare, Vantaa, Finland Learning objectives: To discuss challenges in non-invasive treatment of breast cancer patients with MRI-guided High Intensity Focused Ultrasound (MR-HIFU). Furthermore, a new and dedicated MR-HIFU breast platform is presented.

Figure 1: The MR-HIFU breast platform consists of a water-filled table top with breast cup, integrated in a 1.5 T clinical MRI scanner.

1 7 E

Background: Treatment of breast cancer remains a major topic of interest because of the growing incidence in the western world. Screening has resulted in an increased detection rate of small, nonpalpable breast tumors. Gold standard treatment for these lesions is breast-conserving therapy. The next step in the treatment evolution could be image-guided ablation. MR-HIFU is a completely noninvasive ablation technique. It combines MRI for tumor targeting, delineation and treatment monitoring, and focused ultrasound beams for ablation of the target lesion. Procedure details: MR-HIFU may be an interesting alternative for the treatment of breast cancer. However, many

Figure 2: Eight separate focused ultrasound modules with 32 transducers elements each are surrounding the breast cup for 270 degrees.

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

41

7 programma & abstracts Abstracts challenges have to be faced before MR-HIFU can be implemented in clinical practice. The first challenge is to improve the techniques used for HIFU ablation. A dedicated MR-HIFU breast platform has been developed which uses a lateral sonication technique. This reduces the risk of heating heart, lungs and ribs. A second challenge is to select patients with tumors best eligible for MR-HIFU ablation. These tumors are defined as breast cancers of limited extent, and nowadays treated with breast-conserving therapy. Additionally, a

margin should be defined for tissue ablation around the MRI visible index tumor. Furthermore, treatment and prognosis of breast cancer patients is dependent on the staging of the axilla with the sentinel node procedure. Currently, the impact of MR-HIFU ablation on this procedure is unknown. Conclusion: A novel and dedicated MR-HIFU breast platform is presented, which potentially allows safe and effective ablation of breast cancer.

Figure 3: A T1-weighted 3D gradient echo image without fat suppression of a healthy volunteer, with a schematic overlay of the ultrasound beam.

O07.04 DE DIAGNOSTISCHE WAARDE VAN ACOUSTIC RADIATION FORCE IMPULSE (ARFI) ELASTOGRAFIE IN HET ONDERSCHEIDEN VAN BENIGNE EN MALIGNE LAESIES IN DE MAMMAE A. van Lieshout, M.J.C.M. Rutten Jeroen Bosch Ziekenhuis, ‘s-Hertogenbosch Acoustic Radiation Force Impulse (ARFI) elastografie is een beeldvormende techniek welke de weefselelasticiteit kwantitatief en onderzoeker onafhankelijk kan meten. Het doel is het verhogen van de specificiteit van echografisch onderzoek bij laesies in de mammae. Hierdoor kan het aantal aanvullende biopsieën verminderd worden.Elastografie gaat uit van een verschil in weefselelasticiteit tussen maligne en benigne weefsels. ARFI is een relatief nieuwe techniek waarmee een kwantitatieve en objectieve elasticiteitsmeting wordt verricht. In tegenstelling tot andere vormen van elastografie is er geen manuale compressie nodig om de meting uit te voeren. Dit maakt dat de ARFI meting reproduceerbaar en betrouwbaar is. Middels echografie wordt een locatie bepaald waar de meting uitgevoerd moet worden. Vervolgens wordt in de laesie een meetvolume geplaatst. 42

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

Daarna zorgt een akoestische puls voor verplaatsing van weefsel. Hierbij ontstaat een zogenaamde shearwave. De akoestische puls wordt gevolgd door detectie pulsen die de shear wave velocity (SWV) berekenen (figuur 1). Deze waarde staat gelijk aan de weefselelasticiteit. Hoe stijver het weefsel des te groter de shear wave velocity. Het onderzoek van Berg et al. laat een toename in specificiteit zien, van 61.1% naar 77.4%, in classificeren van mamma laesies, wanneer shear wave elastografie wordt toegevoegd aan standaard echografisch onderzoek (1).Conclusie: Acoustic Radiation Force Impulse (ARFI) elastografie lijkt een

Figuur 1: Mechanisme ARFI meting.

7

educatieve sessie (deel 2) veelbelovende techniek waarmee op objectieve wijze gedifferentieerd kan worden tussen maligne en benigne laesies in de mammae.

Literatuur: 1. Berg WA, Cosgrove DO, Doré CJ, Schäfer FK, Svensson WE, Hooley RJ et al. Shear-wave elastography improves the specificity of breast US:BE1 multinational study of 939 masses. Radiology. 2012;262(2):435-49.

O07.05 SCHOUDERLUXATIES C.J. van Manen, R.E. Westerbeek, R.A.J.M. van Dijk Deventer Ziekenhuis, Deventer

case serie zal de toehoorder een opfriscursus krijgen over de verschillende schouderluxaties, de radiologische signs en de bijkomende letsels.

Leerdoel: Het structureel beoordelen van conventionele schouderopnamen met vooral het doel het herkennen van de verschillende zeldzamere luxaties. Tevens komt het beschrijven van geassocieerd letsels van de verschillende luxaties ter sprake. Achtergrond: Schouderluxaties zijn vaak voorkomende traumatische afwijkingen. In het merendeel van de gevallen zal het gaan om een anterieure luxatie (95%). De zeldzamere vormen, namelijk de posterieure luxatie (5%) en luxatio erecta (<0,5%) zijn ook meteen lastiger te herkennen en kunnen gemakkelijk gemist worden. Door middel van een

O07.06 WHAT IS THAT GLAND LESION? MR IMAGING OF SALIVARY GLAND TUMOURS, AN INTERACTIVE QUIZ A.I. Issa, B.M. Wiarda, M.A. Heitbrink Medisch Centrum Alkmaar, Alkmaar

Conclusie: Anterieure schouderluxaties zijn een veel voorkomende traumatische afwijking. De posterieure luxatie is een zeldzamere vorm en wordt gemakkelijk gemist en kan dan voor ernstige co-morbiditeit zorgen. Een specifieke vorm is de luxatio erecta, met typisch beeld van kliniek en radiologische afbeelding.

to discriminate between benign and malignant lesions and between benign mixed tumours (BMT) and Warthin tumours. Case based examples from a 1.5T MR scanner, before and after the administration of contrast, are illustrated. A database of pathology results from 2000 to 2010 of salivary gland cytology and histology from 611 patients will be used to illustrate the most common salivary gland lesions in our large teaching institute.

Purpose / Aim: To give the reader tools to differentiate salivary gland tumours using typical lesion characteristics in MR imaging. Content organization: In the format of a case based review exhibit, we describe by 10-15 cases the typical lesion characteristics of the most common salivary gland tumours in MR imaging. Using these lesion characteristics in MRI

Figure 1: Maligne lymfoom.

Beschrijving beelden: De verschillende luxaties worden getoond met eigen materiaal. De Radiologische signs en pittfalls worden getoond. Conventionele beelden van de Hills-sachs laesie, (reversed) Bankart fractuur en Tubeculum maius en minus fracturen passeren.

Summary: Knowledge of the typical lesion characteristics before and after contrast improves the discrimination between benign and malignant salivary tumours, as well as between BMT and Warthin tumours, when analyzing salivary gland tumours.

Figure 2: Adenoid cysteus carcinoom.

1 7 E

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

43

7 programma & abstracts Abstracts

Figure 3: Samenvatting.

O07.07 BEELDVORMING BIJ HET SPONTANE LIQUOR HYPOTENSIE SYNDROOM M. van der Vlies, A. Driessen-Waaijer, D.A.C. Duyndam Onze Lieve Vrouwe Gasthuis, Amsterdam Doelstelling: Herkennen van de vaak aspecifieke en subtiele beeldvormingskenmerken bij het spontane liquor hypotensie syndroom. Achtergrond: Spontane liquor hypotensie, ook wel bekend als Spontaneous Intracranial Hypotension Syndrome (SIH), wordt veroorzaakt door spontane liquor lekkage in de neuraxis. Deze aandoening kent een associatie met bindweefselaandoeningen, maar heeft vaak een onduidelijke etiologie. In de meeste gevallen presenteert het zich met orthostatische hoofdpijn op jonge en middelbare leeftijd. Verder lijkt SIH minder zeldzaam dan eerst gedacht en wordt het tegenwoordig klinisch vaker herkend. Bevestiging van de juiste diagnose en het specifiek lokaliseren van het liquorlek middels MRI en/of CT-Myelografie, kan de behandeling van

O07.08 LOCALIZING CRITICAL CORTICAL AREAS USING COMBINED FMRI AND DTI PRECEDING NEUROSURGERY A.I. Issa, P.R. Algra Medisch Centrum Alkmaar, Alkmaar Purpose / Aim: To describe and illustrate the emerging technique and application of combined functional magnetic resonance imaging (fMRI) and diffusion tensor imaging (DTI) fiber tracking in the evaluation of patients with brain tumours undergoing neurosurgery. Content organization: We describe the advantages of 44

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

deze aandoening gericht en effectief sturen. Therapeutische mogelijkheden zijn onder andere behandeling met bed rust, epidurale blood patching, percutane fibrine sealing of chirurgie. Beschrijving: Deze presentatie beoogt aan de hand van casuïstiek uit de eigen praktijk en uit de literatuur een overzicht te geven van het radiologische spectrum van SIH. Zo worden voorbeelden getoond zoals geassocieerde spinale vochtcollecties, versterkte durale aankleuring, epidurale veneuze engorgement, hypofyse hyperemie, sagging midbrain, het C1-C2 sign, en actieve contrast extravasaties. Tevens wordt er gewezen op de noodzaak van specifieke beeldvorming van de gehele neuraxis in het geval van aanvankelijk negatieve bevindingen bij een patiënt met orthostatische hoofdpijnklachten. Conclusie: SIH is een ondergediagnosticeerde aandoening. In deze presentatie bieden wij voor het stellen van de juiste diagnose een overzicht van de radiologische kenmerken op MRI en/of CT-Myelografie.

the combination of fMRI and DTI. Fifty patients underwent preoperative fMRI and DTI at 1.5T. During an fMRI experiment, paradigms specific to language, motor, auditory and/or visual functions of the brain were presented to the patient in order to create stimuli to help activate specific parts of the brain. The results of the combined examinations of fMRI and DTI influenced therapeutic management.Case based examples of fMRI and DTI are illustrated. Summary: fMRI combined with DTI fiber tracking is capable of localizing critical cortical areas more precisely than fMRI or DTI separately, changing therapeutic management in 40% of cases. Using fMRI and DTI combined is a more powerful tool than using them separately.

7

educatieve sessie (deel 2)

Figure 2: Anaplastisch glioom in zeer nauwe relatie tot de corticospinale banen.

Figure 1: Finger tapping beiderzijds.

O07.09 CLEARING THE PEDIATRIC C-SPINE: IS IT CLEAR? A. Slaar, N.W.L. Schep, L.F.M. Beenen, M. Maas Academisch Medisch Centrum, Amsterdam Purpose / Aim: Cervical spine injury in children is rare, accounting for only 1-2% in children presenting for trauma evaluation. Yet every attending radiologist is anxious not to miss these rare injuries, since devastating consequences may arise.The aim of our presentation is to provide evidence based guidelines and illustrate in an educational manner the optional approach of imaging, aiming to increase confidence in clearing the pediatricC-spine. Content organization: In a case based simulation we will illustrate the clinical relevance of decision rules, as NEXUS,

1 7 E

in the pediatric population. Age and maturation of the injured child in one of the important topics. When evaluating imaging in young children it is important to know that almost all cervical spine injuries occur in the upper cervical spine. Awareness of normal age related features, pathological entities as well as radiation exposure will be pointed out. The presentation will discuss the use of cervical radiographics and CT-scanning in children and provide clinical guidelines. The proper indications for MRI also are debated with clear examples of its use in various pathologies. Summary: Understanding of anatomic differences, fracture patterns and radiologic modality possibilties should give doctors more confidence in clearing the C-spine in children and likely to reduce radiation exposure and healthcare costs.

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

45

8 programma & abstracts Abstracts Sessie 8

Abdominale radiologie (deel 2) Vrijdag 28 september, 9:45 - 11:00 uur

O08.01 HIGH-TEMPORAL RESOLUTION LIVER PERFUSION MRI: COMPARISON OF GD-DTPA AND GD-EOB-DTPA. WHEN IS THE START OF THE HEPATOBILIARY PHASE? H.J. Schalkx1, M. van Stralen2, M.S. van Leeuwen1, M.A.A.J. van den Bosch1, W.B. Veldhuis1 1 Universitair Medisch Centrum Utrecht, Utrecht 2 Image Sciences Institute, University Medical Center utrecht, Utrecht Purpose: MRI plays an increasing role in diagnosis and characterization of focal liver lesions, with an important contribution of the arterial, portal and equilibrium phase of the contrast-enhanced (CE) series. When liver-specific Gd-EOB-DTPA is used, the dynamic series are assumed to be unaffected with additional diagnostic yield of the late hepatobiliary series at 10 and 20 min.However, the start of the hepatobiliary phase is hitherto undefined. With hightemporal resolution CE T1-weighted (4D-THRIVE) MRI, the enhancement of the liver parenchyma after Gd-DTPA and Gd-EOB-DTPA can be compared in order to assess the start of the hepatobiliary phase.

Materials & methods: Ten consecutive patients without diffuse liver disease were evaluated with MRI for focal liver lesions, underwent Gd-DTPA enhanced and Gd-EOB-DTPA enhanced series within 2 weeks. The 4D-THRIVE protocol included 16 timepoints up to 5 minutes for both contrast agents, with additional 10 and 20 min after Gd-EOB-DTPA. Parenchymal enhancement on all twenty scans was measured in 3 regions-of-interest and averaged over time. Results: In the early phase after Gd-DTPA injection, the high-temporal enhancement profile showed a faster upslope and higher peak enhancement compared to Gd-EOB-DTPA. The parenchymal enhancement of Gd-EOB-DTPA continued to increase, from 60 seconds onward, reaching a plateau at 10 min. Conclusion: Using Gd-EOB-DTPA, the continuous enhancement of the parenchyma after the early dynamic phases, indicates active uptake of contrast-agent starts from 1 minute onward. Apparently, the hepatobiliary phase already starts during the late portal and equilibrium phase. Consequently the clinical significance of the washout phenomenon is reduced for characterization of a focal liver lesions.

Figure 1: Mean liver parenchymal enhancement after Gd-DTPA and Gd-EOB-DTPA CE-MRI (N=10), plotted as the signal intensity relative to the pre-contrast series. 46

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

8

Abdominale radiologie (deel 2) O08.02 FOCAL CYSTIC LIVER LESIONS DIFFERENTIATED BASED ON RADIOLOGICAL IMAGING FINDINGS F.E.J.A. Willemssen, J. Liem, G.P. Krestin, R.S. Dwarkasing Erasmus MC, Rotterdam Purpose: Focal cystic liver lesions can be caused by many pathological processes, which require different management strategies. Can neoplastic cystic liver lesions be differentiated from benign and inflammatory cystic liver lesions, based on radiological imaging findings? Methods & materials: Between 2000 and 2011, 48 patients with 51 pathologically proven cystic liver lesions and prior imaging by CT or MRI were included. Thirty-nine patients were female, 9 were male. Mean age was 58.6 years (range 24-84years). Imaging features of 51 lesions were evaluated on either multiphase contrast enhanced CT (45 lesions) and/ or contrast enhanced MRI (19 lesions). Lesions characteristics such as size, shape, wall appearance, internal structure and enhancement characteristics were noted. Results: Pathological analysis revealed 16 (31%) benign lesions, all simple cysts. Eighteen lesions (35%) were neoplastic, 8 (16%) premalignant lesions, all biliary cystadenomas, 10 (20%) malignant lesions (2 cystadenocarcinoma, 1 sarcoma, 7 metastasis). Seventeen lesions (33%) were inflammatory (5 echinococcal cyst , 12 pyogenic abscess). Mean diameter was 11.9 cm (range 1.3-28.7 cm). Fifteen out of 16 (94%) simple cysts showed a thin regular wall (less than 2mm) without solid components. A small internal peripheral located septum was noted in 9 out of 16 (56%) benign lesions. Twelve (71%) inflammatory lesions showed a thickened irregular wall (2-9mm) with enhancement. Contrast enhancement of multiple internal septations (13 lesions) and/or solid components (8 lesions) were noted in all neoplastic lesions. Conclusion: Neoplastic cystic liver lesions can accurately be differentiated from benign and inflammatory cystic lesions. Neoplastic lesions all show multiple septations with internal enhancement, solid components or both. An irregular thickened wall is frequently present in inflammatory lesions.

O08.03 REPRODUCIBILITY OF MRI SCORING SYSTEMS IN DETERMINING DISEASE ACTIVITY IN PATIENTS WITH CROHN’S DISEASE J.A.W. Tielbeek1, S. Bipat1, C.Y. Nio1, C. Makanyanga2, D. Pendse2, S. Taylor2, J. Stoker1

Figure 1: Contrast-enhanced MRI of cystadenoma. Large cystic laesion with multiple internal septations with encencement.

Figure 2: Contrast-enhanced CT of echinococcal cyst. Large cystic lesion with multiple internal cystic lesions, typically daughter cysts.

Figure 3: Contrast enhanced CT of histiocytic sarcoma. Cystic lesion with internal solid components.

Academisch Medisch Centrum, Amsterdam University College London Hospital, London, United Kingdom

1

2

Purpose: Recently, two validated qualitative scoring systems for Crohn’s disease (CD) activity were introduced: the MaRIA index and Crohn’s Disease Activity (CDA) score. This study 1 7 E

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

47

8 programma & abstracts Abstracts aims to validate these scores with an external patient cohort. Methods: A dataset of 30 patients underwent standard 3T MR enterography with T2w, contrast enhanced (CE) and T1w sequences before and after Gadobutrol. The disease activity in each bowel segment (terminal ileum, left, transverse, right colon and rectum) was scored by 3 abdominal radiologists from 2 tertiary centers in different countries using both scores.For each segment an overall CDA score was determined (features: mural thickness, T1 enhancement, enhancement pattern, mural and perimural T2-signal) and an overall MaRIA index was calculated (features: mural thickness in mm, relative CE (RCE), presence of edema and ulcers).As all data were considered continuous we used the intra-class correlation coefficient (ICC) to study inter-observer variability.

O08.04 VIRAL HEPATITIS B AND C: NON-INVASIVE SELECTION OF PATIENTS WITH ADVANCED LIVER FIBROSIS USING MR ELASTOGRAPHY AND FIBROSCAN A.E. Bohte1, A. de Niet1, A.J. Nederveen1, S. Bipat1, J. Verheij1, V. Terpstra2, R. Sinkus3, C.M.J. van Nieuwkerk4, R.J. de Knegt5, L.C. Baak6, P.L.M. Jansen1, J. Stoker1 1 Academisch Medisch Centrum, Amsterdam 2 Bronovo Ziekenhuis, Den Haag 3 Hopital Beaujon, Paris, France 4 VU medisch centrum, Amsterdam 5 Erasmus MC, Rotterdam 6 Onze Lieve Vrouwe Gasthuis, Amsterdam Purpose: To define the cutoffs for MR Elastography (MRE) and Fibroscan for non-invasive selection of patients with advanced liver fibrosis. Methods: 100 patients who had a liver biopsy were consecutively included. All underwent MRE (3 Tesla) and Fibroscan. Liver elasticity values were expressed in kilopascals (kPa). Biopsies were scored by two pathologists according to the Metavir scoring system from F0 (no fibrosis) to F4 (cirrhosis). We defined for each imaging technique the following two cut-offs (in kPa) for distinguishing F0-F1 (no

Figure 1 48

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

Results: MRI features of 143 bowel-segments were scored. 7 segments were excluded due to resection or poor distension. The overall CDA score showed a good inter-observer variability (ICC of 0.70-0.66-0.76). The overall MaRIA index showed a very variable inter-observer variability (ICC of 0.230.79-0.14). However, the individual features mural thickness in mm and RCE, showed promising results. Mural thickness in mm showed good to excellent inter-observer variability (ICC of 0.93-0.74-0.73). The RCE inter-observer variability was fair to good (ICC of 0.71-0.36-0.61). Conclusion: The CDA score showed promising reproducible results and might be feasible for use in a clinical setting. Individual MRI features used in the MaRIA index showed high inter-observer variability, although the overall score of this index was not as reproducible as the CDA score.

or mild fibrosis) from F2-F4 (advanced fibrosis - cirrhosis):(1) the cut-off at which the negative predictive value (NPV) is 100% (no false negatives)(2) the ‘optimal’ cut-off at which NPV is lowered to 90%, but PPV increases (less false positives). Cut-offs were determined using receiver operating characteristics analysis and accuracy parameters with 95% confidence intervals were calculated. Results: Data of 83/100 patients (65 HBV/18 HCV) were analysed. Mean age was 44.5±23.6 years (range 19.073.9). 53/83 were male (64%). 56 patients had F0-F1 and 27 had F2-F4. (1) For MRE, at a cut-off of 1.9 kPa NPV was 100%(91-100%), PPV 64%(50-79%), sensitivity 100%(88100%) and specificity 73%(62-85%). At a cut-off of 5.3 kPa, Fibroscan performed comparably, with NPV 100%(91-100%), PPV 60%(46-74%), sensitivity 100%(88-100%) and specificity 68%(56-80%).(2) At a cut-off of 2.1 kPa for MRE, NPV was 90%(95% CI: 82-98%), PPV 71%(55-87%), sensitivity 82% (67-96%) and specificity 84%(74-94%). Using a cutoff of 6.8 kPa for Fibroscan, NPV was 91%(83-98%), PPV of 73%(58-89%), sensitivity 82%(67-96%) and specificity 86%(77-95%). Results are illustrated in fig.1 Conclusion: MR-elastography and Fibroscan can both be used to non-invasively select patients with advanced liver fibrosis, avoiding invasive liver biopsies.

8

Abdominale radiologie (deel 2) O08.05 TOEGEVOEGDE WAARDE VAN FDG-PET IN DE DIAGNOSTISCHE WORK-UP VOOR RADIOEMBOLISATIE BIJ PATIËNTEN MET COLORECTALE LEVERMETASTASEN C.E.N.M. Rosenbaum, M.A.A.J. van den Bosch, W.B. Veldhuis, J.E. Huijbregts, M. Koopman, M.G.E.H. Lam Universitair Medisch Centrum Utrecht, Utrecht

extrahepatische laesies werden gedocumenteerd. De toegevoegde waarde van FDG-PET werd bepaald voor 1) het opsporen van extrahepatische laesies en 2) wijzigingen in behandelbeslissing.

Doel: De standaard diagnostische work-up voor patiënten met levermetastasen die worden verwezen voor yttrium-90 radioembolisatie (Y90-RE) bestaat uit een 3-fasen CT abdomen. Gezien het locoregionale karakter van de behandeling, blijven eventuele extrahepatische laesies onbehandeld en zullen deze de prognose negatief beïnvloeden. Het doel van deze studie was het evalueren van de toegevoegde waarde van FDG-PET in de diagnostische work-up van patiënten met colorectale levermetastasen.

Resultaten: In totaal werden 42 patiënten (18 vrouwen, 24 mannen; gemiddelde leeftijd 59 jaar, range 34 - 82) opgenomen in de analyse. Bevindingen van CT en FDG-PET kwamen overeen in 20 patiënten (geen extrahepatische laesies n=15; identieke extrahepatische laesies, n=5). In 4 patiënten werden op CT laesies gezien die niet FDG-avide bleken, en in 18 patiënten toonde FDG-PET meer laesies dan CT (McNemar p<0.05). In 7/42 patiënten (17%) wijzigde de behandelbeslissing door de FDG-PET bevindingen, te weten, geen behandeling met Y90-RE (n=6) en behandeling van de hele lever in plaats van enkel een segment (n=1).

Methode: Alle patiënten met colorectale levermetastasen, die werden verwezen voor Y90-RE, en zowel een 3-fasen CT abdomen als whole-body FDG-PET hadden ondergaan, werden voor de analyse geselecteerd. Gedetecteerde

Conclusie: In 17% van de patiënten met colorectale levermetastasen, verwezen voor Y90-RE, leidde FDG-PET tot een gewijzigde behandelbeslissing, met name door de detectie van extrahepatische laesies. Op basis hiervan adviseren wij FDG-PET toe te voegen aan de diagnostische work-up voor Y90-RE.

O08.06 CAN INFLAMMATORY MYOFIBROBLASTIC TUMOR OF THE HEPATOBILIARY SYSTEM BE DIFFERENTIATED FROM CHOLANGIOCARCINOMA ON IMAGING? R. Elias, F. Willemssen, K. Biermann, G.P. Krestin, R.S. Dwarkasing Erasmus MC, Rotterdam

imaging revealed bile duct compression with peripheral dilatation (n=12, 63%), tumor extension in the main bile duct (n=14, 74%), periportal region (n=13, 68%) and main portal or hepatic veins (n=10, 52%). Focal liver capsule retraction (n=5, 26%) and extrahepatic enlarged lymph nodes (n=13, 68%) were only seen in patient with CC.CT-scan and MRI demonstrated no differences between IMT and CC in lesion density, signal intensity or contrast-enhancement characteristics.

Purpose: To investigate whether Inflammatory myofibroblastic (IMT) of the hepatobiliary system, which is a benign entity, can be differentiated from its malignant look-alike CholangioCarcinoma (CC) on CT scan and MRI.

Conclusion: IMT and CC can be recognized and differentiated in imaging by using morphological characteristics of the tumor. Radiologists should be aware of morphological characteristics of IMT in imaging. Subsequently,should be proven pathologically in order to avoid unnecessary liver resection in patient with suspected CC.

Method & materials: Thirty three patients with histopathological proven IMT and CC from 1995- 2011 were included in this study population. CT-scan was performed on 29 patients and MRI on 20 patients. All examinations were reviewed by a radiologist without knowledge of any clinical information. The following findings of the lesion; morphology, dynamic contrast-enhanced characteristics and signal intensity were registered by the observer using a standard form. Results: Of 33 patients, IMT was proven in 14 (42%) patients and CC in 19 patients (57%). The imaging of IMT patients often demonstrate multifocal lesions with sharp defined margin (n=5, 35%). The imaging of patients with CC showed a solitary lesion with ill-defined margin in the majority of cases (n= 16, 84%). Only patients with CC, the 1 7 E

Figure 1

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

49

9 programma & abstracts Abstracts Sessie 9

Cardiovasculaire radiologie (deel 2) Vrijdag 28 september, 9:45 - 11:00 uur

O09.01 MR-VENOGRAFIE TER BEOORDELING VAN HET POST-THROMBOTISCH SYNDROOM TYPE ONDERZOEK: PROSPECTIEVE STUDIE DISCIPLINE: CARDIOVASCULAIRE RADIOLOGIE C.W.K.P. Arnoldussen1, M.A.F. de Wolf1, R. de Graaf1, T. Leiner2, C.H.A. Wittens1 1 Maastricht Universitair Medisch Centrum, Maastricht 2 Universitair Medisch Centrum Utrecht, Utrecht Achtergrond: Het post-thrombotisch syndroom (PTS) is te definiëren als chronische veneuze symptomen secundair ontstaan na een diep veneuze thrombose (DVT). PTS ontstaat door beschadiging van de venen en kleppen in het diep veneuze systeem ten gevolge van een DVT met obstructie en insufficientie tot gevolg. Van patienten met PTS is bekend dat zij in hun daaglijks-leven veel hinder ondervinden van dit syndroom, waarbij frequent pijn, oedeem, een zwaar gevoel in de benen (soms met verminderde actie-radius) en zelfs veneuze ulcera worden beschreven. Klinische evaluatie van PTS met behulp van de beschikbare scorings systemen is echter moeizaam en aspecifiek. Indien middels beeldvorming accuraat chronisch veneuze pathologie kan worden vastgesteld zijn wellicht de ware PTS patienten te onderscheiden

van die patienten met vergelijkbare symptomen welke niet veroorzaakt worden door PTS. Methode: 75 patienten verwezen met klinische symptomen passend bij PTS werden prospectief geëvalueerd middels MR-Venografie en de ‘Lower-Extremity-Venous-Pathology’score. Klinisch werd gescoord middels VCSS en Villalta. De MR-onderzoeken werden vervaardigd op een 1.5-T scanner (Philips Medical, Best, The Netherlands). Anatomische beelden van het diep veneuze systeem, vanaf de proximale kuitvenen tot en met de vena cava inferior, werden verkregen middels post-contrast (Gadobutrol, Bayer Healthcare Pharmaceuticals) T1-gewogen opnamen met vetsuppressie (THRIVE). Reconstructies in 1 mm coupes in het coronale en axiale vlak. Resultaten: Bij 62 van de 75 patienten was er een duidelijke correlatie tussen de bevindingen gescoord op het MRV onderzoek en de klinische scores. Bij 13 patienten werden geen evidente afwijkingen gevisualiseerd. Conclusie: MR-Venografie onderzoek ter beoordeling van het Post-thrombotisch syndroom kan worden gebruikt om objectief vast te stellen of er sprake is van onderliggende diep veneuze pathologie bij een klinisch beeld suspect voor PTS.

Table 1: Lower Extremity Venous Pathology Score. 50

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

9

Cardiovasculaire radiologie (deel 2)

Figure 1: Villalta Scale.

1 7 E

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

51

9 programma & abstracts Abstracts

Figure 2: VCSS score (example).

O09.02 HIGH-FIELD 3T VERSUS 1.5T 3-STATION MOVING-TABLE MR ANGIOGRAPHY IN PERIPHERAL ARTERIAL OCCLUSIVE DISEASE : CONTRAST-TO-NOISE AND CLINICAL PERFORMANCE G.Y.M. The1, H.C.M. van den Bosch1, J.M. Westenberg2, A. Daniels1, W. Setz-Pels1, P.W.M. Cuypers1, A. de Roos1 1 Catharina Ziekenhuis, Eindhoven 2 Leids Universitair Medisch Centrum, Leiden Background / Aim: Contrast-enhanced MR angiography (CE-MRA) has evolved into a reliable imaging technique for peripheral arterial occlusive disease (PAOD). Larger homogeneous magnetic fields at 3T allow imaging of the complete runoff arterial tree by single-injection 3-station moving-table CE-MRA. Aim of this prospective study was to obtain diagnostic accuracy of 3T versus 1.5T CE-MRA in POAD, with Digital Subtraction Angiography as reference. Methods: Nineteen patients underwent single-injection 3-station moving-table CE-MRA at 3T and 1.5T MRI. At 1.5T, quadrature body coil (QBC) was used for imaging pelvic and thigh stations and 4-element phased-array coil 52

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

for calf station. At 3T, QBC for all stations.For each patient, the arterial tree was divided into 27 segments. Visual stenosis classification was performed as follows: class 1 (0%-stenosis), 2 (1-50%), 3 (51-75%), 4 (76-99%) and 5 (100%-occlusion). Quantitative analysis of contrast-to-noise ratio (CNR) was performed for the popliteal, external iliac and superficial femoral artery. Results: 500 arterial segments (97.5% of all available) were evaluated. 105 segments (21%) were appointed with relevant stenosis (≥class 2) on DSA. 3T and 1.5T CE-MRA showed equivalent excellent agreement (κ=0.96 versus 0.93) with DSA regarding stenosis classification. Sensitivity/specificity for stenosis scoring at 3T was >93%/99% and >90%/99% at 1.5T. For popliteal artery, CNR at 3T was 3.8±1.9 times higher than at 1.5T, 3.0±1.4 times higher for external iliac artery and 3.4±1.4 times higher for superficial femoral artery (all < 0.001). Conclusion: 3T CE-MRA in POAD showed excellent diagnostic performance with >3 times higher CNR compared to 1.5T CE-MRA with comparable scan protocol and identical contrast dosage.

9

Cardiovasculaire radiologie (deel 2) O09.03 ASSOCIATION BETWEEN AORTIC STIFFNESS, CAROTID VESSEL WALL THICKNESS AND STENOSIS SEVERITY IN PERIPHERAL ARTERIAL OCCLUSIVE DISEASE: A COMPREHENSIVE 3 TESLA MRI STUDY S.L. Wolters1, H.C.M. van den Bosch1, J.J.M. Westenberg2, L.E.M. Duijm1, W. Setz-Pels1, J.A.W. Teijink1, A. de Roos2 1 Catharina Ziekenhuis, Eindhoven 2 Leids Universitair Medisch Centrum, Leiden Introduction: Arterial wall remodeling is an early atherosclerotic manifestation. Vessel wall (VW) thickening and stiffening precede plaque deposition. MRI is wellvalidated for imaging VW thickness and stiffness expressed in pulse wave velocity (PWV). Purpose of this study was to use 3T MRI comparing stenosis severity on CE-MRA in peripheral arterial occlusive disease (PAOD) with VW thickness (common carotid artery) and PWV (descending aorta). Methods: Forty-two patients with suspected POAD were included. Black-blood carotid VW imaging, phase-contrast PWV-assessment and CE-MRA were performed at 3T. PAOD

stenosis severity was visually scored: class 1 (0%-stenosis), 2 (1-50%), 3 (51-75%), 4 (76-99%) and 5 (100%-occlusion). Mean cross-sectional VW area of common carotid artery was obtained. PWV was assessed in descending aorta and obtained from systolic wave propagation analysis based on transit-time method. PWV was compared with maximal stenosis severity class and carotid VW area indexed for body surface area (BSA). Results: PWV in the descending aorta was well-correlated with maximal stenosis class (Spearman correlation 0.63 (p<0.001)). Carotid VW area/BSA and PWV (Pearson correlation 0.48 (p=0.002)) and carotid VW area/BSA and maximal stenosis class (Spearman correlation 0.43 (p=0.005)) were correlated to lesser extent. Correlation between stenosis severity and PWV remained significant, independent from age. Correlation between stenosis severity and carotid VW area/BSA was not significant when corrected for age. Conclusion: Stiffness in descending aorta expressed by PWV is significantly associated with PAOD stenosis severity, independent from age. Low association between carotid VW thickness and PAOD severity suggests site-specific coupling of vessel wall stiffening and atherosclerosis.

Figure 1: MRA, positioning of carotid vessel wall sampling and positioning for PWV of the aorta. PWV is determined from wave propagation analysis.

1 7 E

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

53

9 programma & abstracts Abstracts O09.04 ASSOCIATION BETWEEN ARTERIAL CALCIFICATIONS AND LACUNAR AND NONLACUNAR INFARCTS A.C. van Dijk, S. Fonville, T. Zadi, A.M.G. van Hattem, P.J. Koudstaal, A. van der Lugt Erasmus MC, Rotterdam Introduction: Lacunar infarcts (LACI) are presumed to be caused by small vessel disease; nonlacunar infarcts (nLACI) are generally caused by large vessel atherosclerosis or cardiac embolism. We investigated if calcification volume, as measure of amount of atherosclerosis, differs between LACI and nLACI. Methods: We categorized 740 consecutive patients with TIA or ischemic stroke in the anterior circulation into LACI and nLACI based on the clinical OCSP classification or multi-detector CT-scan (MDCT) of the brain. Patients with a cardiac or rare cause of stroke were excluded. We used CT-angiography to semi-automatically score calcification volumes in the aortic arch and carotid bifurcations. Multivariable logistic regression analysis was used to

O09.05 LOW IODINE CONCENTRATION - HOW LOW CAN WE GO? A FEASIBILITY STUDY USING A CIRCULATION PHANTOM C. Mihl1, T. Jurencak1, F.F. Behrendt2, J.E. Wildberger1, M. Das1 1 Maastricht Universitair Medisch Centrum, Maastricht 2 Universitätsklinikum RWTH Aachen, Germany Objectives: Iodine concentrations between 300 and 400 mg/ml are usually applied in CT. To test feasibility of lower concentrated contrast material (CM), injection protocols with low iodine concentration were compared to a standard injection protocol for CT angiography in a circulation phantom. Materials & methods: A circulation phantom was used which consists of a low-pressure venous system for injection, a pulmonary circulation, a high-pressure body circulation with a replica of the aorta and the coronary arteries simulating physiological conditions (HR 60 bmp, SV 60 ml). Reference standard injection protocol: 7.2 ml/s, iodine delivery rate (IDR) 2.16 g/s, total amount of iodine

54

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

assess if calcifications differ between LACI and nLACI. Adjustments were made for degree of stenosis and cardiovascular risk factors. Results: Based on clinical criteria only, 221(30%) of the patients had LACI and 519(70%) nLACI. Calcification volume in the carotid arteries did not differ significantly between LACI and nLACI (median volume 2.9 versus 5.6 mm3;p=0.14). Calcification volume in the aortic arch was significantly higher in the nLACI compared with LACI (median volume 43.2 versus 9.7 mm3;p=0.001), even after adjusting (aOR[95CI]; 1.14[1.01-1.28]). In 212(29%) patients, relevant infarct was visible on MDCT: 71(33%) LACI and 141(67%) nLACI. Calcification volume in the carotid arteries did not differ significantly between LACI and nLACI (median volume 3.3 versus 6.9 mm3;p=0.24). Calcification volume in the aortic arch tended to be higher in nLACI compared with LACI (median volume 40.5 versus 15.7 mm3;p=0.06), although not significantly after adjusting (aOR[95CI]; 1.11[0.85-1.44]). Conclusion: The notion of a different pathophysiology in LACI and nLACI is supported only by the higher calcification volume in the aortic arch in nLACI.

21g, 300 mg/ml Iopromide. Maintaining constant IDR, CM was injected with different iodine concentrations and flow rates (150mg/mL-14,0mL/s, 200mg/mL-11,0mL/s, 250mg/ mL-8,6mL/s and 300mg/mL-7,2mL/s).Serial CT-scans at the level of the ascending aorta (AA), descending aorta (DA) and the coronary arteries (CA) were obtained. Time-enhancement curves were calculated for AA, DA and CA. Time to peak and peak enhancement were determined for all vessels. Linear regression and T-test were used for statistical analysis. Results: The shapes of the time enhancement curves were comparable. Peak enhancement with usage of 150mg iodine showed moderate attenuation (AA;401, DA;400, CA;189). Peak enhancement with usage of respectively 200mg iodine (AA;607, DA;583, CA;343), 250mg iodine (AA;697, DA;658, CA;397) and 300mg iodine (AA;760, DA;704, CA;434) showed moderately higher attenuation. Time to peak did not significantly differ in all enhancement curves. Conclusion: Injection of low iodine concentration is feasible and diagnostic enhancement values for CTA can be achieved. To test enhancement curves in patients would be the next step to validate feasibility of low concentration contrast material for CTA.

9

Cardiovasculaire radiologie (deel 2) O09.06 PULSE PRESSURE AS RISK FACTOR FOR MRI DETECTED INTRAPLAQUE HEMORRHAGE IN THE CAROTID ARTERIES: THE ROTTERDAM STUDY M. Selwaness, Q. van den Bouwhuijsen, G.C. Verwoert, A. Dehghan, M. Vernooij, F. Mattace-Raso, O.H. Franco, A. Hofman, J.J. Wentzel, J.C.M. Witteman, A. van der Lugt Erasmus MC, Rotterdam Purpose: Intraplaque hemorrhage (IPH) is a characteristic of the vulnerable atherosclerotic plaque that has been associated with ischemic stroke. Potential determinants of plaque vulnerability and in specific IPH remain unclear. We studied whether blood pressure parameters are associated with presence of IPH. Methods & materials: Within the framework of a prospective population-based cohort study, The Rotterdam Study, the carotid arteries of 1,006 healthy participants of 45 years and older and with intima-media thickness (>2.5mm) on ultrasound were imaged with a 1.5-T MRI scanner. IPH was defined as a hyperintense signal on a 3D-T1w-GRE

1 7 E

MR sequence. Generalized estimation equation analysis, adjusted for sex, age, wall thickness and cardiovascular risk factors including smoking, BMI, total cholesterol and diabetes was used to assess the association between blood pressure parameters and IPH. Results: MR imaging of the carotid arteries revealed presence of IPH in 444 of 1866 (24%) plaques. Systolic blood pressure (SBP) and pulse pressure (PP) were significantly associated with IPH after adjustment for age and sex. After further adjustment for carotid wall thickness and cardiovascular risk factors, PP yielded the strongest association, with an odds ratio (OR) per SD of 1.22 (95%CI 1.07-1.40). The OR per SD for SBP was 1.13 (95%CI 0.99-1.28). Only PP remained significant after additional adjustment for other blood pressure components. Conclusion: Pulse pressure was the strongest determinant of IPH independent of cardiovascular risk factors and other blood pressure components. The association between pulsatile flow and IPH may provide novel insights into the development of the vulnerable plaque.

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

55

10 programma & abstracts Abstracts Sessie 10

Mammaradiologie (deel 2) Vrijdag 28 september, 9:45 - 11:00 uur O10.01 FOUT POSITIEVE RATIO DRIE DIMENSIONALE ECHOGRAFIE MAMMAE M.D.F. de Jong, G.J. Jager, M.J.C.M. Rutten Jeroen Bosch Ziekenhuis, ‘s-Hertogenbosch Inleiding: Vanwege de lage sensitiviteit van mammografie bij dicht borstklierweefsel is aanvullende beeldvorming belangrijk en nodig. Mogelijk kan drie dimensionale echografie (3DUS) hiertoe bijdragen, vanwege de betrekkelijke eenvoud van het onderzoek, de minimale patiëntenbelasting en de relatief lage kosten. Een van de hindernissen voor implementatie hiervan zijn het aantal fout positieve uitslagen dat met conventioneel echografisch onderzoek gegenereerd wordt. Er is echter nog geen prospectief onderzoek, waarbij het aantal fout positieve uitslagen bij 3DUS is onderzocht.

Resultaten: In totaal zijn 885 vrouwen geïncludeerd, waar in 23 gevallen het echografisch onderzoek werd geclassificeerd als BI-RADS 0, 3 of hoger. In 7 gevallen bleek het daadwerkelijk borstkanker te betreffen, waarbij er dus in 16 gevallen een onterechte verwijzing zou zijn, hetgeen 1,8% is. Eén geval van borstkanker werd gemist, welke retrospectief wel zichtbaar was.

Materiaal & methode: Aan vrouwen met een afspraak voor mammografie werd een 3DUS mammae aangeboden na informed consent en na het reguliere onderzoek, zodat deelname geen invloed had op de gangbare werkwijze.

Conclusie: Gezien het lage aantal onterechte verwijzingen kan implementatie van 3DUS mammae in het bevolkingsonderzoek naar borstkanker in onderzoeksverband worden overwogen.

O10.02 VERGELIJKING VAN DRIEDIMENSIONALE ECHOGRAFIE MAMMAE MET MRI EN HISTOLOGIE M.D.F. de Jong, I.J.M. Dubelaar, T.A. Fassaert, G.J. Jager, M.J.C.M. Rutten Jeroen Bosch Ziekenhuis, ‘s-Hertogenbosch

geblindeerd en onafhankelijk van overige beeldvorming beoordeeld volgens de BI-RADS classificatie. Na sluiting van de studie werd de uitkomsten van 3DUS en MRI met elkaar vergeleken en waar mogelijk gecorrigeerd met histologie.

Inleiding: Een veelbelovende techniek in echografische beeldvorming is de drie dimensionale echografie (3D-US) van de mammae. Er zijn echter nog geen grote studies waarin deze techniek wordt onderzocht in vergelijking met de huidige gouden standaard in beeldvorming: MRI. In gepresenteerd onderzoek worden beide technieken met elkaar vergeleken. Materiaal & methode: Geïncludeerd werden alle vrouwen ouder dan 18 jaar, welke ingepland waren voor MRI. Na informed consent werd een 3D-US vervaardigd binnen tien dagen voor of na het MRI-onderzoek. De 3D-US-data werd 56

Geëxcludeerd werden vrouwen met een genetische of familiaire belasting, verwijzing via het bevolkingsonderzoek en/of een voorgeschiedenis van borstkanker. De echografische beeldvorming werd onafhankelijk en geblindeerd voor overige beeldvorming beoordeeld volgens de BI-RADSclassificatie. Onderzoeken met een BI-RADS classificatie van 0, 3 of hoger werden na gemiddeld vier weken vergeleken met het initiële en eventueel aanvullende onderzoek, hetgeen kon bestaan uit MRI, tomosynthese en/of histologie.

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

Resultaten: Van 375 aangeschreven patiënten gaven 222 informed consent. Eén patiënt werd geëxcludeerd vanwege een foutieve datatransfer. De leeftijd van de patiënten was gemiddeld 48 jaar (18-78). De sensitiviteit en specificiteit van 3D-US in vergelijking met MRI was respectievelijk 80% en 98%. Wanneer de uitslagen gecorrigeerd werden met histologie bedroeg de sensitiviteit en specificiteit 88% en 96%. Conclusie: 3D-US van de mammae heeft een hoge sensitiviteit en specificiteit in vergelijking met MRI en histologie.

10

Mammaradiologie (deel 2) O10.03 INTRA-OPERATIVE ULTRASOUND IN BREASTCONSERVING SURGERY FOR PALPABLE BREAST CANCER: AN UNDENIABLY EFFECTIVE TECHNIQUE RESULTING IN COST SAVINGS M.H. Haloua1, N.M.A. Krekel1, V.M.H. Coupé1, E. Bergers1, A.M.F. Lopes Cardozo2, S. Muller3, H. van der Veen3, A.M. Bosch4, S. Meijer1, M.P. van den Tol1 1 VU medisch centrum, Amsterdam 2 Medisch Centrum Alkmaar, Alkmaar 3 Rode Kruis Ziekenhuis, Beverwijk 4 Ziekenhuis Gelderse Vallei, Ede Background: Breast-conserving surgery for palpable breast cancer is worldwide associated with a high rate of tumour-involved margins. A randomised controlled trial was initiated to compare ultrasound-guided surgery (USS) with the standard palpation-guided surgery (PGS) for palpable breast cancer.

O10.04 OPMERKELIJKE VERBETERING VAN RADICALITEIT EN EXCISIEVOLUME DOOR PEROPERATIEVE ECHOGRAFIE BIJ MAMMASPARENDE CHIRURGIE: RESULTATEN VAN EEN MULTICENTRISCHE, PROSPECTIEF GERANDOMISEERDE STUDIE N.M.A. Krekel1, M.H. Haloua1, E. Bergers1, A.M.F. Lopes Cardozo2, R.H. de Wit2, A.M. Bosch3, S. Muller4, H. van der Veen4, S. Meijer1, M.P. van den Tol1 1 VU medisch centrum, Amsterdam 2 Medisch Centrum Alkmaar, Almaar 3 Ziekenhuis Gelderse Vallei, Ede 4 Rode Kruis Ziekenhuis, Beverwijk

1 7 E

Results: In the USS-group, 3.3% of margins were involved with invasive carcinoma, compared with 15.9% in the PGSgroup (p<0.05)). The use of intra-operative US resulted in a significant reduction in additional therapies; in the PGSgroup re-excisions were necessary in 2 patients (3.2%) and in 1 in the USS-group (1.6%), mastectomies were performed in 5 patients (7.9%) of the PGS-group and in none of the USS-group, additional radiotherapy boosts were also taken into account. Mean extra costs per patient due to margin involvement in the PGS-group were 516€ compared with 161€ in the USS-group. Yearly costs of a US-system are 8,286€. Therefore, USS for palpable breast cancer can save a breast unit 355€ per patient after the first 24 operated patients (on a yearly basis). Conclusion: USS can prevent the unacceptably high rate of tumour-involved resection margins in palpable breast cancer excision, thus not only improving oncological and cosmetic outcomes by avoiding subsequent surgery or radiotherapy, but also considerably reducing treatment costs.

Methods: A total of 124 eligible patients with palpable T1-T2 invasive breast cancer were randomised to either USS (n = 61) or PGS (n=63). Outcome measures included resection margin status, re-excision rates, mastectomy rates

Achtergrond: De mammasparende chirurgie voor het palpabel mammacarcinoom komt wereldwijd in aanmerking voor verbetering. Naast een hoog percentage irradicaliteit (>20%) van deze ‘blinde’ palpatiegeleide excisies zijn ook de volumina verwijderd gezond mammaweefsel te groot (factor 2.5). Het gebruik van peroperatieve echografie, is bewezen effectief bij de excisie van het niet-palpabele mammacarcinoom. Peroperatieve echografie zou ook tot optimalisering van de radicaliteit en excisievolumina bij de palpabele tumoren kunnen leiden. Deze multicentrische, prospectief gerandomiseerde studie werd opgezet om de resultaten van echogeleide chirurgie te vergelijken met de standaard palpatiegeleide chirurgie van het palpabel mammacarcinoom.

and additional radiotherapy. The costs of purchasing the US-system were included in the USS-group.

Methoden: In deze studie zijn 124 patiënten met een T1-T2 invasief mammacarcinoom gerandomiseerd voor echogeleide chirurgie (n=61) of palpatiegeleide chirurgie (n=63). Primaire eindpunten waren radicaliteit en excisievolume. Secundaire eindpunten waren tijdsduur van de excisie, postoperatieve complicaties, cosmetisch resultaat en kwaliteit van leven. Resultaten: In de echogeleide groep was 96,2 % radicaal en 3.8% focaal irradicaal, terwijl in de palpatiegeleide groep 83,6% radicaal, 5,5% irradicaal en 10,9% focaal irradicaal was (p<0,05). Als gevolg hiervan werden in de palpatiegeleide groep meer reëxcisies, mastectomieën en additionele radiotherapeutische boosts voor irradicale marges verricht. De excisievolumina waren gemiddeld aanzienlijk kleiner met echogeleide chirurgie ten opzichte van palpatiegeleide chirurgie (respectievelijk 40 en 58 cc (p<0,05)). Conclusie: Het gebruik van peroperatieve echografie voor het palpabel mammacarcinoom voorkomt een hoog percentage irradicaliteit en daarmee de benodigde re-interventies. Echogeleide chirurgie reduceert daarnaast de hoeveelheid overtollig gereseceerd gezond mammaweefsel en verbetert daarmee hoogstwaarschijnlijk het postoperatieve cosmetisch resultaat en kwaliteit van leven van de patiënt.

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

57

10 programma & abstracts Abstracts O10.05 IMAGING FEATURES IN MAMMOGRAPHY AND BREAST ULTRASOUND ARE RELATED TO HER-2/NEU RECEPTOR OVER-EXPRESSION OF PRIMARY INVASIVE BREAST CANCER A. Adams1, S.G. Elias1, K.E. Pengel2, C. Loo2, W.P.Th.M. Mali1, K.G.A. Gilhuijs1 1 Universitair Medisch Centrum Utrecht, Utrecht 2 Nederlands Kanker Instituut - Antoni van Leeuwenhoek Ziekenhuis, Amsterdam Purpose: To investigate what mammography and ultrasound imaging features are related to over-expression of the HER-2/neu receptor in invasive breast cancer. These features could give insight in the molecular basis of imaging phenotype, facilitate pre-biopsy patient recruitment for early phase HER-2/neu targeting molecular imaging trials, and could ultimately aid in predicting therapy response and prognosis. Methods: Mammography and breast ultrasound imaging features of 606 early invasive breast cancers were scored according to the 4th edition of the BI-RADS lexicon. HER-2/ neu over-expression status was determined on surgical specimens. Multivariable logistic regression methods were applied to identify imaging features predicting HER-2/neu over-expressing cancer.

58

Results: HER-2/neu over-expression was found in 80 cancers (13%). On multivariable analysis, presence of microcalcifications (OR 3.4, 95%-CI 2.1-5.8, p<0.001) on mammography, and posterior attenuation of a mass (OR 0.5, 95%-CI 0.3-0.9, p=0.014) on ultrasound, were statistically significant independent predictors. Non-significant predictors were breast density ACR category 3/4 (OR 0.9, 95%-CI 0.5-1.6) and mass presence (OR 0.8, 95%-CI 0.4-1.4) on mammography, and circumscribed mass-margin (OR 0.8, 95%-CI 0.3-2.1) and irregular mass-shape (OR 1.3, 95%-CI 0.8-2.1) on ultrasound. The area under the ROC curve (AUC) was 0.69 (95%-CI 0.63-0.75, p<0.001). After adjustment for over-optimism, post-test probabilities of HER-2/neu overexpression were 23-37% for cancers with calcifications, without posterior attenuation, and 5-10% for cancers without calcifications, with posterior attenuation (overoptimism adjusted AUC: 0.67.). Conclusion: Microcalcifications on mammography, and posterior acoustic attenuation of masses on ultrasound were predictors of HER-2/neu over-expression, indicating that imaging characteristics reflect molecular expression patterns in breast cancer.

O10.06 BELASTE FAMILIE ANAMNESE: MAMMOGRAM VOLGENS DE RICHTLIJN? VOORLOPIGE DATA D. van der Vlies, K. Droogh-de Greve, J. de Win, H.J. van der Zaag-Loonen, J.W. Gratama Gelre ziekenhuizen, Apeldoorn

vanaf begin 2012. Een vragenlijst werd afgenomen door de radiologisch laborant waarbij de familieanamnese werd nagevraagd met aandacht voor 1e,2e of 3e graads verwantschap bij mamma-, prostaat- of ovariumcarcinoom in de familie. Op basis van de tabel in de nieuwe richtlijn werden patiënten ingedeeld in diverse risicogroepen (geen verwijzing, mammogram of consult klinisch geneticus).

Doel: In de nieuwe richtlijn van 2012 worden de indicaties besproken voor het bevolkingsonderzoek borstkanker (BOB). Vrouwen met een belaste familieanamnese voor mammacarcinoom worden regelmatig via de huisarts verwezen voor screening buiten het BOB. Voor deze groep vrouwen bestaat een aparte beslisboom in de nieuwe richtlijn. Het is echter onduidelijk hoe vaak de richtlijn op de juiste wijze wordt opgevolgd.

Resultaten: Tot medio april werden 56 patiënten geïncludeerd. In 7 (13%) van de 56 patiënten werd de richtlijn juist gevolgd. Negen patiënten (16%) werden ten onrechte verwezen voor mammografie (volgens de richtlijn was hier een indicatie om te verwijzen naar de klinisch geneticus). In 40 (71%) patiënten was volgens de richtlijn geen verwijzing of screening noodzakelijk anders dan het bevolkingsonderzoek.

Methoden: In een prospectief cohort werden alle patiënten die door de huisarts verwezen werden voor mammogram met als indicatie “belaste familie anamnese”geincludeerd

Conclusie: In 13% is de verwijzing voor een mammogram conform de richtlijn mammacarcinoom.

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

11

interventieradiologie

Sessie 11 Interventieradiologie Vrijdag 28 september, 9:45 - 11:00 uur O11.01 SUBINTIMAL ARTERIAL FLOSSING WITH ANTEGRADE-RETROGRADE INTERVENTION (SAFARI) FOR LOWER EXTREMITY TOTAL OCCLUSIONS M.C. Burgmans1, K.D. Zhuang2, M Taneja2, F.G. Irani2, T.K. Teo2, S.G. Tan2, B.S. Tan2, K.H. Tay2 1 Leids Universitair Medisch Centrum, Leiden 2 Singapore General Hospital, Singapore, Singapore

obtained from the following arteries: popliteal (n=7), dorsal pedal (n=7), anterior tibial (n=4), posterior tibial (n=5) and peroneal (n=2). Balloon-assisted hemostasis was performed for distal retrograde puncture sites in 18 procedures. Mean follow-up was 7.2 months. Results: Technical success was achieved in 22/25 (88%) of procedures. Hard calcified plaques prevented crossing of occlusion in 3 cases. A total of 27 limb segments were successfully recanalized (SFA n=11; popliteal artery n=5; anterior tibial n=6; posterior tibial n=3; peroneal n=2). The lenght of treated segments were: <10cm: n=9; 10-20cm: n=6; >20cm: n=12). Limb salvage rates at 6 month was 83.3%. There was a significant complication of distal embolization in 1 patient which was successfully treated with clot aspiration. There were no proximal or distal puncture site complications or procedure-related mortality.

Purpose: Retrospective review of SAFARI technique procedures after failed antegrade recanalization in patients with lower limb chronic total occlusions (CTOs) Material & methods: Between January 2009 and August 2010, 25 SAFARI procedures were performed in 24 patients (mean age: 68 years (range 51-86 yrs); M:F 1:1) after failed antegrade recanalization of lower extremity CTOs. All patients were treated for critical limb ischemia (Fontaine III (n=2) or IV (n=22). Proximal access was obtained from the ipsilateral and contralateral common femoral artery in 19 and 6 procedures respectively. Distal retrograde access was

Conclusion: The SAFARI technique is effective and safe for recanalization of lower limb CTOs when antegrade recanalization fails.

Materials & methods: Patients with dysfunctional, stenotic dialysis AVFs/AVGs without central venous stenoses were enrolled and randomized to receive HPBA or CBA if conventional PTA was suboptimal, defined as residual stenosis of >30%. Primary end point was angiographic patency at 6 months. Secondary end points included technical success, procedural complication rate, 30 day mortality and 6 months secondary patency.

O11.02 PROSPECTIVE, RANDOMISED TRIAL OF CUTTING BALLOON ANGIOPLASTY VS HIGH PRESSURE BALLOON ANGIOPLASTY IN DIALYSIS ARTERIO-VENOUS GRAFT AND FISTULA STENOSIS RESISTENT TO CONVENTIONAL PERCUTANEOUS TRANSLUMINAL ANGIOPLASTY M.C. Burgmans1, S.A. Aftab2, M. Taneja2, F.G. Irani2, T.K. Teo2, H.H. Win2, S.G. Tan2, B.S. Tan2, K.H. Tay2 1 Leids Universitair Medisch Centrum, Leiden 2 Singapore General Hospital, Singapore, Singapore Purpose: To compare the efficacy and safety of high pressure balloon angioplasty (HPBA) versus cutting balloon angioplasty (CBA) in patients with dialysis arteriovenous graft (AVG) and arteriovenous fistula (AVF) stenoses resistant to conventional balloon angioplasty.

1 7 E

Results: 516 patients were enrolled from October 2008 till September 2011. 439 (85%) patients had good results with conventional PTA. Seventy-seven patients (mean age: 60 years, M:F 52:25) with suboptimal conventional PTA results were randomised to CBA (n=39) or HPBA (n=38). Primary and secondary patency at 6 months were 60%, 64%, 40% and 84%, 91%, 81% for cPTA, CBA and HPBA respectively. The difference in primary and secondary patency between CBA and HPBA were statistically significant (p<0.013 and p<0.024 respectively). There was a significant complication of venous

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

59

11 programma & abstracts Abstracts perforation following CBA which was successfully managed with prolonged balloon inflation. Thirty day mortality was 0.01%.

Conclusion: Conventional PTA is an effective treatment in most patients with stenotic AVF/AVG. For stenoses resistant to conventional PTA, CBA is superior to HPBA as the second line treatment.

O11.03 RESULTATEN VAN EEN FASE 1, DOSIS ESCALATIE STUDIE NAAR HOLMIUMRADIOEMBOLISATIE IN PATIëNTEN MET IRRESECTABELE, CHEMOREFRACTAIRE LEVER METASTASEN: DE HEPAR TRIAL M.L.J. Smits, J.F.W. Nijsen, M.A.A.J. van den Bosch, M.G.E.H. Lam, M.A.D. Vente, W.P.Th.M. Mali, A.D. van het Schip, B.A. Zonnenberg Universitair Medisch Centrum Utrecht, Utrecht

Gy (n=3), en 80 Gy (n=3)). De cohorten werden uitgebreid tot maximaal zes patiënten in geval van dosis limiterende toxiciteit. Het primaire eindpunt van deze studie was de maximaal getolereerde radiatie dosis. Deze studie staat geregistreerd bij ClinicalTrials.gov, nummer NCT01031784.

Introductie: Holmium-166 (166Ho) poly(L-lactic acid) microsferen zijn ontwikkeld voor radioembolisatie van lever tumoren. 166 Ho-microspheren zijn uniek vanwege het feit dat ze in vivo kunnen worden afgebeeld met zowel SPECT als MRI. In deze fase 1 klinische studie werd de maximaal getolereerde radiatie dosis (MGRD) en de veiligheid van 166Ho-radioembolisatie (RE) onderzocht in patiënten met lever metastasen. Methoden: In deze prospectieve studie werden 15 patiënten met irresectabele, chemorefractaire lever metastasen geïncludeerd en behandeld met een oplopende lever geabsorbeerde doses 166Ho-microsferen (20 Gy (n=6), 40 Gy (n=3), 60

Conclusie: 166Ho-radioembolisatie kan veilig worden uitgevoerd in patiënten met lever metastasen. Voor verder klinisch gebruik is een geabsorbeerde lever dosis van 60 Gy het streven bij 166Ho-RE. Een fase 2 studie start op korte termijn.

O11.04 ENDOLEAK DETECTION IN PATIENTS AFTER ENDOVASCULAR ABDOMINAL ANEURYSM REPAIR WITH AN ALBUMIN BINDING CONTRAST AGENT J. Habets, H.J.A. Zandvoort, L.W. Bartels, F.L. Moll, J.A. van Herwaarden, T. Leiner Universitair Medisch Centrum Utrecht, Utrecht

FOV 450x345mm, acquisition matrix 2x2x2mm, and acquisition time 18 seconds. Post-contrast images were required 5 and 15 minutes after contrast administration. For the CTA examinations, the presence of endoleaks was evaluated on both the arterial and delayed enhanced images. Two observers evaluated all MRI examinations in consensus for the presence of endoleaks comparing pre-contrast and post-contrast T1 fat suppressed spin echo images.

Purpose: To determine the diagnostic value of an albumin binding contrast agent in the detection of endoleaks in patients with aneurysm growth and no or uncertain endoleak on CTA.

Results: Eight patients (median aneurysm diameter was 74mm (IQR 69-94) were evaluated. CTA assessment resulted in the following scores for the presence of endoleaks: absent (2/8; 25%), very uncertain 3/8; 38%), somewhat uncertain (1/8; 13%), certain (1/8; 13%) and very certain (1/8; 13%). Endoleaks were present in 8/8 patients (100%) on the postcontrast MRI images. Endoleaks were more pronounced at MR imaging on late post-contrast images in 6/8 patients (75%) (Figure 1).

Methods: Patients after endovascular repair for abdominal aortic aneurysm (EVAR) were included if (1) there was continued aneurysm growth; and (2) no or uncertain endoleak at CTA. Multihance®, an albumin binding contrast agent with a weak protein interaction, was administered at a dose of 0.15mmol/kg and a flow rate of 1.0ml/sec. All MRI scans were performed on a 1.5T clinical scanner with a surface body coil for signal reception. The following scan parameters were used: TR/TE1/TE2/α 5.9 ms/1.8ms/4.0ms/15°, slice thickness 2mm, 60

Resultaten: 15 patiënten hebben 166Ho-radioembolisatie (RE) ondergaan met een gemiddelde totaal toegediende activiteit van 5085 MBq ± 2876 MBq en een gemiddelde geschatte geabsorbeerde dosis op de lever van 40 Gy ± 23 Gy. Het 20-Gy cohort werd uitgebreid naar zes patiënten vanwege het optreden van een serious adverse event (longembolie). In het 80-Gy cohort trad dosis limiterende toxiciteit op in twee patiënten: graad 4 trombocytopenie, graad 4 hyperbilirubinemie, graad 3 leukopenie, en graad 3 hypoalbuminemie in één patiënt, en graad 3 buikpijn in de andere patiënt. De MGRD werd derhalve vastgesteld op 60 Gy.

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

Conclusions: MRI with an albumin binding contrast agent is a promising additional imaging technique to CTA for the detection of endoleaks in patients with aneurysm growth after EVAR.

11

interventieradiologie

Figure 1

O11.05 EFFICACY OF A NOVEL VOLUMETRIC MAGNETIC RESONANCE-GUIDED HIGH INTENSITY FOCUSED ULTRASOUND TECHNIQUE FOR THE TREATMENT OF SYMPTOMATIC UTERINE FIBROIDS M.E. Ikink, M.J. Voogt, H.M. Verkooijen, K.J. Schweitzer, W.P.Th.M. Mali, L.W. Bartels, M.A.A.J. van den Bosch Universitair Medisch Centrum Utrecht, Utrecht Purpose: To assess the clinical efficacy of Magnetic Resonance-guided High Intensity Focused Ultrasound (MR-HIFU) using a novel volumetric ablation technique for treatment of symptomatic uterine fibroids. Materials & methods: Premenopausal women (n=169) with symptomatic uterine fibroids were prospectively included. Forty-four patients (26%) were considered eligible for MR-HIFU based on diagnostic MRI. The remaining 125 (74%) women were excluded because of e.g. fibroid size, number, intensity on T2-weighted MRI, or lack of contrast enhancement. The primary endpoint was a significant clinical improvement in the transformed Symptom Severity Score (SSS) of the Uterine Fibroid Symptom and Quality 1 7 E

of Life (UFS-QoL) questionnaire, defined as a 10-point improvement at 6 months. Secondary endpoint was fibroid volume reduction during follow-up determined on T2-weighted MR Images. Results: Forty-four patients were treated with volumetric MR-HIFU. To date, 37 patients (84%) have completed the targeted 6 months follow-up. In two cases the treatment was terminated preliminary due to insufficient tissue temperature rise and/or patient physical discomfort. Five patients sought alternative treatment shortly after MR-HIFU treatment because of unsatisfactory symptom relief and withdrew from further follow-up. Nineteen patients (60%) reported a clinical improvement of more than 10-points in the SSS at 6 months. The median SSS improved from 53.1 (range 6.3-90.6) at baseline to 36.0 (range 0-75.0) after 6 months (p<0.0001). Median fibroid volume was 328 cm3 (range 23-1028) at baseline, which reduced to 223 cm3 (range 4.5-1021) at 6 months, corresponding to a significant shrinkage of 27%. Conclusion: Volumetric MR-HIFU treatment for symptomatic uterine fibroids resulted in clinically relevant symptom improvement in 60% of the patients at 6 months post-treatment. R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

61

11 programma & abstracts Abstracts O11.06 PERCUTANEOUS CHOLECYSTOSTOMY: SINGLE CENTRE EXPERIENCE IN 111 PATIENTS WITH AN ACUTE CHOLECYSTITIS R. Peters, S. Braak, S. Rakic ZGT Almelo, Almelo Objectives: To evaluate the safety and long-term outcome of percutaneous cholecystostomy (PC) under radiologic guidance for acute calculeous and acalculeous cholecystitis (ACC and AAC) in all patients undergoing the procedure at our institution. Methods: We performed a retrospective analysis of 111 patients undergoing PC from 2005 to 2011. The mean follow-up was 55 months. Patients were divided into two groups AAC and ACC. Of these 111 patients, comorbidity and American Society of Anesthesiologists (ASA) classification were determined. The outcome of technical success, indication, complications, recurrence rate and long-term outcome for both groups were determined.

62

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

Results: 24 patients with AAC and 87 patients with ACC underwent PC. 12/24 (50%) patients with AAC were hospitalized on the Intensive Care. Drain dislodgment was found without sequelae in 8 (7,2%) patients. The procedure failed in 2 (1,8%) patients. There were 4 (3,6%) abscess and 2 (1,8%) fistulas post PC. Elective cholecystectomy was performed in 33/111 (29,7%). 49/87 (56,3%) patients with gallstones underwent cholecystectomy, 38/87 (43,7%) didn’t need surgery because the follow-up was too short to make conclusions or they died of non-biliary disease. In the AAC group no recurrent cholecystitis occurred in 17/24 (70,8%) patients, 3/24 (12,5%) needed surgery later on and 4/24 (16,6%) patients died on the Intensive Care. Conclusion: PC is a low-invasive treatment with low complication rate for patients with acute cholecystits whom considered being at high-risk for cholecystectomy. Good selection (ASA III and IV) and indication is needed in patients with ACC before PC because the majority will be operated later on. AAC can be managed non-operatively and further treatment might not be needed.

12

Musculoskeletale radiologie / Onderwijs

Sessie 12 Musculoskeletale radiologie / Onderwijs Vrijdag 28 september, 9:45 - 11:00 uur O12.01 DE RADIOLOOG VAN DE TOEKOMST: WERKZAAMHEDEN VAN JONGE RADIOLOGEN IN ORGAANGEBIEDEN NU EN IN DE TOEKOMST D.R. Rutgers1, P.A.M. Kint2, B.A.R. Tonino3, D.S. te Boekhorst4, J.P.J. van Schaik1 1 Universitair Medisch Centrum Utrecht, Utrecht 2 Amphia Ziekenhuis, Breda 3 Rijnstate Ziekenhuis, Arnhem 4 Maasziekenhuis Pantein, Boxmeer Doel: Onderzoeken in hoeveel en welke orgaangebieden jonge radiologen nu werken en later willen werken. Methoden: Descriptieve pilot-studie onder 40 achtereenvolgende radiologen die tussen 2006 en 2012 radioloog werden en hun radiologie-opleiding gedeeltelijk of geheel doorliepen in het UMC van één van de OOR-regio’s. Digitaal werd geënquêteerd naar leeftijd, geslacht, werkomgeving, recente verrichtingen in de 8 orgaangebieden (abdomen/thorax/MSK/neuro-hoofdhals/mammo/kinder/cardio-vasculair/ angio-interventie), expertise-terreinen (gedefinieerd als orgaangebieden waarin de radioloog zijn/haar kennis en ervaring lokaal als bovengemiddeld beschouwt), en gewenste toekomstige werkzaamheden.

O12.02 TECHNICAL AND CLINICAL IMAGE QUALITY COMPARISON OF 3-MEGAPIXEL AND 6-MEGAPIXEL LIQUID CRYSTAL DISPLAY (LCD) MONITORS FOR RADIOLOGY F.E.M. Dams, K.Y.E. Leung, S.P. Niehof, P.H.M. van der Valk, J. Bosman Albert Schweitzer ziekenhuis, Dordrecht Purpose: To be able to make a cost-effective decision in purchasing new medical displays we compared displays of three major manufacturers to evaluate if there is a significant difference in image quality. Materials & methods: We compared the image quality of 1 7 E

Resultaten: De respondenten (n=32, responspercentage 80%; man/vrouw=19/13 (59%/41%)) waren gemiddeld 37 jaar (± 3 SD) en sinds gemiddeld 2.7 jaar (± 1.6 SD) radioloog. Zij werken nu in academische ziekenhuizen (n=12, 38%), perifere ziekenhuizen met arts-assistenten radiologie (n=12, 38%), en perifere ziekenhuizen zonder arts-assistenten radiologie (n=8, 24%).In de voorafgaande 6 maanden hadden 3 respondenten (9%) verrichtingen uitgevoerd in 1-2 orgaangebieden, 10 (31%) in 3-4, 12 (38%) in 5-6, en 7 (22%) in 7-8 orgaangebieden, meestal abdomen (84% van de respondenten), thorax (81%) en neuro-hoofdhals (78%). Zeven respondenten (22%) beschouwden 3 orgaangebieden als hun expertise-terrein, 18 (56%) 2 orgaangebieden, en 7 (22%) 0-1 orgaangebied. In de toekomst willen 6 respondenten (19%) het liefst in 1-2 orgaangebieden werken, 16 (50%) in 3-4, 7 (22%) in 5-6, en 3 (9%) in 7-8 orgaangebieden. Negentien respondenten (59%) prefereren dan hetzelfde aantal expertise-terreinen als nu, 12 (38%) meer en 1 respondent (3%) minder. Conclusie: De meeste geënquêteerde jonge radiologen werken in 3-4 of 5-6 orgaangebieden en beschouwen 2 of 3 orgaangebieden als expertise-terrein. De meesten wensen dit ook in toekomstige werkzaamheden. Gezien deze bevindingen is te overwegen om in de radiologie-opleiding differentiatie in tenminste twee orgaangebieden mogelijk te maken.

three three-megapixel (3 MP) displays, Barco, Eizo and NEC, and a Barco six-megapixel (6 MP) display. The names of the manufactures were shielded. Technical assessment was performed following acceptance criteria and test patterns proposed by the American Association for Physicists in Medicine (AAPM). For clinical assessment twelve radiologists and seven residents scored AAPM Task Group (TG) 18 X-ray images of thorax, knee and breast, thorax CT and brain MR images. AAPM image quality criteria, such as sharpness and contrast, were scored on an analogue scale from 0 to 10. Statistical analysis was performed using repeated measures analysis of variance (ANOVA). Results: The Barco 3MP display passed all acceptance criteria. The Eizo and NEC displays passed the acceptance

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

63

12 programma & abstracts Abstracts criteria, except for the darkest pixel-value in the grayscale display function. The Barco 6 MP failed criteria for maximum luminance response and veiling glare. Clinical assessment outcomes were 7.8 ± 1.1 (Barco 3 MP), 7.8 ± 1.2 (Eizo), 8.1 ± 1.0 (NEC) and 8.1 ± 1.0 (Barco 6 MP). No statistically significant differences were found.

Conclusion: According to the tested criteria the displays have comparable technical and clinical image quality, if calibrated properly. In the purchase of new displays financial aspects should also be taken into consideration.

O12.03 3D DELAYED GADOLINIUM-ENHANCED MRI OF CARTILAGE AT 3.0 TESLA USED TO EVALUATE THE EFFECT OF HYALURONIC ACID ON CARTILAGE QUALITY IN KNEE OSTEOARTHRITIS PATIENTS J. van Tiel, M. Reijman, P.K. Bos, J. Hermans, G.M. van Buul, J.A.N. Verhaar, G.P. Krestin, S.M. Bierma-Zeinstra, H. Weinans, G. Kotek, E.H.G. Oei Erasmus MC, Rotterdam

Methods: In 15 patients with knee OA, dGEMRIC was acquired at baseline and four months after HA at 3T using a 3D FSPGR sequence. To evaluate patient symptoms, the knee injury and osteoarthritis outcome score (KOOS) questionnaire was recorded at baseline and follow-up. Cartilage quality was assessed in eight cartilage regions of interest (ROIs) in the medial and lateral tibiofemoral knee compartment by calculating the T1GD relaxation times in each ROI. Outcomes of dGEMRIC and KOOS before and four months after HA were compared using paired testing to evaluate symptomatic and potential structural effects of HA.

Purpose: Intra-articular viscosupplementation with hyaluronic acid (HA) of osteoarthritic (OA) knees has a well-established positive effect on patient symptoms. Although the working mechanism is not clear, it has been suggested that HA might improve cartilage quality because of its potentially beneficial effect on glycosaminoglycan (sGAG) content of cartilage. Recently, we showed that dGEMRIC is a highly reproducible measure of cartilage quality in longitudinal OA research. This study assessed if improvement in knee cartilage quality can be detected with dGEMRIC in OA knees four months after HA.

Results: Outcomes of dGEMRIC four months after HA did not improve significantly compared to baseline in any of the analyzed cartilage ROIs (figure 1A). However, except for the subscale “Symptoms”, all KOOS subscales improved significantly after HA (figure 1B). Discussion: Outcomes of dGEMRIC indicate that no improvement in cartilage quality is detectable in OA knees four months after HA. However, similar to previous research, patient complaints decreased significantly after HA. These results suggest that the working mechanism of HA is not acting through an improvement of sGAG content in cartilage.

Figure 1: A: Mean dGEMRIC outcome in search ROI pre (white box) and post (grey box) viscosupplementation with hyaluronic acid. B: Mean KOOS outcomes per subscale pre (white box) and post (grey box) viscosupplementation with hyaluronic acid. ADL: daily activities, QOL: quality of life. Boxes range from 25th to 75th percentile, whiskers run from min to max, the horizontal line in the box represents the median, the plus sign shows the mean and **: p < 0.001.

64

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

12

Musculoskeletale radiologie / Onderwijs O12.04 DIFFERENTIAL DIAGNOSTIC VALUE OF 1.5 TESLA EXTREMITY MRI IN EARLY ARTHRITIS W. Stomp, A. Krabben, D.M.F.M. van der Heijde, J.L. Bloem, T.W.J. Huizinga, A.H.M. van der Helm-van Mil, M. Reijnierse Leids Universitair Medisch Centrum, Leiden Purpose: Many studies have described the use of MRI in established RA and other rheumatic diseases. However there is a lack of studies determining its value in patients with early arthritis. The objective of this study is to evaluate the differential diagnostic value in early arthritis, with an emphasis on the early diagnosis of rheumatoid arthritis.

diagnostic groups, except for tenosynovitis score which was lower in osteoarthritis (1,00 vs 5,83, p<0.01). No positive determinants were found for RA. Discussion: None of the individual parameters evaluated nor the total score adequately discriminated between RA and other diagnostic groups in early arthritis. Further longitudinal studies are required that combine MRI data with clinical data and determine the value of MRI specifically in the group of undifferentiated arthritis patients, in whom the need for supplemental diagnostic tools is the highest.This research was supported by the Center for Translational Molecular Medicine.

Methods: In a prospective study, patients with recent onset arthritis received an MRI of the wrist, metacarpophalyngeal and metatarsophalyngeal joints. Acquired sequences were Coronal T1, coronal T2 fatsat and coronal and axial T1 fatsat after IV gadolinium contrast administration. Synovitis, tenosynovitis, bone marrow edema and erosions were scored according to OMERACT RAMRIS score and diagnoses were established by the treating rheumatologists, blinded to MRI findings. Results: 58 early arthritis patients were studied. Initial diagnosis was rheumatoid arthritis according to ACR87 criteria in 12 patients, undifferentiated arthritis in 27, psoriatric arthritis (PsA) in 8, osteoarthritis (OA) in 5 patients and other diagnoses were made in 6 patients. MRI parameters did not differ between RA and any of the other

Table 1: Mean RAMRIS scores grouped by diagnosis. RA = rheumatoid arthritis, UA = undifferentiated arthritis, OA = osteoarthritis, PsA = psoriatric arthritis.

O12.05 COMPUTED TOMOGRAPHY FOR THE DETECTION OF THUMB BASE OSTEOARTHRITIS, COMPARISON WITH DIGITAL RADIOGRAPHY M.S. Saltzherr, J.W. van Neck, G.S.R. Muradin, R. Ouwendijk, J.J. Luime, J.H. Coert, S.E.R. Hovius, R.W. Selles Erasmus MC, Rotterdam Objective: To compare Computed Tomography (CT) with digital radiography for the evaluation of osteoarthritis (OA) of the first carpometacarpal (CMC1) and scapho-trapezotrapezoidal (STT) joint. Design: We retrospectively identified patients clinically suspected for CMC1 OA or STT OA between January 2008 and March 2011 at our outpatient clinic who had both a digital radiograph and a CT-scan of the hand within a threemonth period. CT and radiographic images were scored

1 7 E

independently by two musculoskeletal radiologists for joint space narrowing (JSN), osteophytes, subchondral sclerosis, bone cysts, and erosions in the CMC1 and STT joint. Results: Thirty patients were identified. The inter-reader reliability of CT for the detection of CMC1 OA (ICC 0.87) and STT OA (ICC 0.80) was higher than radiography (ICC’s 0.51 and 0.45). In comparison with their own radiographical scoring, both readers detected with CT more cases with CMC1 OA (8 and 4 respectively) and STT OA (13 and 4 respectively). Conclusions: CT had a higher inter-reader reliability and detection rate for both CMC1 and STT OA, compared to radiography. Since treatment selection of thumb base OA depends on the pathology presence in both the CMC1 and STT joint, CT may improve treatment selection and thus patient outcome.

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

65

12 programma & abstracts Abstracts O12.06 OSTEOPOROTIC VERTEBRAL FRACTURE PREVALENCES VARY WIDELY BETWEEN RADIOLOGICAL SCORING METHODS: A PROSPECTIVE COHORT STUDY S.J. Breda, L. Oei, F. Ly, E. van Meel, E. Dogterom, L.G.C. de Kok, J.B.J. van Meurs, A. Hofman, H.A.P. Pols, A.G. Uitterlinden, M.C. Zillikens, G.P. Krestin, F. Rivadeneira, E.H.G. Oei Erasmus MC, Rotterdam Purpose: Ageing of populations implicate increasing numbers of osteoporotic fractures, which are associated with morbidity and mortality. Vertebral fractures (VFx) are most prevalent and are often a first presentation of osteoporosis. Several methods for radiological assessment of VFx exist, but a gold standard is lacking. Methods: We applied two methods for scoring osteoporotic VFx. The algorithm based qualitative (ABQ) method primarily focuses on depression of the central endplate, while quantitative morphometry (QM) evaluates vertebral height loss. Trained assistants, supervised by a musculoskeletal radiologist, scored lateral spine radiographs using either

Figure 1: Fracture prevalence by vertebral level.

66

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

ABQ or SpineAnalyzer® software-assisted QM according to Genant. Results: Radiographs were scored for 2475 participants. With QM, 516 fractures were identified (20.8%, CI: 19.2%-22.4%), compared to 83 fractures (3.4%, CI: 2.6%4.1%) with ABQ. Of all individuals, 77.9% (n=1930) were identified as having no fractures, while 2.2 % (n=54) were scored fractured according to both methods. QM scored 462 fractures (18.7%) which were scored as no fracture with ABQ and 29 fractures (1.2%) were exclusively scored by ABQ. With ABQ, most fractures were found at the thoraco-lumbar junction (T11-L1), at which superior endplate fractures were most frequent. Lumbar and mid-thoracic (T7) fractures were also common. With QM, the majority of fractures were mid-thoracic (T6-T8) and low thoracic (T11, T12). Most of them were wedge-shaped (95.6%). Conclusions: Osteoporotic VFx prevalence rates are significantly different when applying either QM or ABQ. Because of the clinical relevance of osteoporotic VFx, a golden standard is needed to achieve consensus on fracture definition.

auteursindex

Auteursindex AUTEUR

abstract

Adam, J.A. O06.04 Adams, A. O10.05 Aftab, S.A. O11.02 Alderliesten, T. O06.01 Algra, P.R. O07.02, O07.08 Arnoldussen, C.W.K.P. O09.01 Baak, L.C. O08.04 Bakers, F.C.H. O02.02, O02.06 Bakker, C. O02.03 Barber, I. O06.04 Bartels, L.W. O01.08, O07.03 O11.04, O11.05 Beek, F.J.A. O06.04 Beenen, L.F.M. O07.09 Beets-Tan, R.G.H. O02.02, O02.05 O02.06, O04.01, O04.05 Beets, G.L. O02.02, O02.05, O02.06 Behrendt, F.F. O09.05 Benders, M.J.N.L. O06.01 Berendsen, R. O02.03 Berg, J.M. van den O06.03 Bergers, E. O04.03, O10.03, O10.04 Bierings, M.B. O06.04 Bierma-Zeinstra, S.M. O12.03 Biermann, K. O01.03, O08.06 Biessels, G.J. O05.03, O05.04 Bipat, S. O08.03, O08.04 Bloem, J.L. O12.04 Boekhorst, D.S. te O12.01 Boellaard, T.N. O02.04 Boetes, C. O04.01 Bohte, A.E. O08.04 Booij de, M. O02.03 Bos, D. O05.01 Bos, P.K. O12.03 Bosch, A.M. O10.03, O10.04 Bosch, H.C.M. van den O09.02, O09.03 Bosch, M.A.A.J. van den O01.08 O04.06, O07.03, O08.01 O08.05, O11.03, O11.05 Bosman, J. O12.02 Bouwhuijsen, Q. van den O09.06 Braak, S.B. O07.01, O11.06 Breda, S.J. O12.06 Brundel, M. O05.03, O05.04 Buckens, C.F.M. O06.05 Budde, R.P.J. O01.09, O03.03, O03.04 Burggraaf, K. O03.01

Burgmans, M.C. O01.01, O01.02 O11.01, O11.02 Buul, G.M. van O12.03 Cappendijk, V.C. O02.02, O02.06 Castro, C. O04.05 Clarijs, R. O02.03 Coert, J.H. O12.05 Cornel, J.H. O07.02 Coupé, V.M.H. O10.03 Cuypers, P.W.M. O09.02 Dams, F.E.M. O12.02 Daniels, A. O09.02 Dankbaar, J.W. O05.02 Das, M. O03.02, O03.05 O03.06, O09.05 Deckers, R. O07.03 Dehghan, A. O09.06 Deserno, W. O02.06 Devos, A. O01.04 Dijk, A.C. van O09.04 Dijk, R.A.J.M. van O07.05 Dijke, C.F. van O07.02 Dikkers, R. O01.04 Dogterom, E. O12.06 Dohle, G.R. O01.06 Dolman, K.M. O06.03 Donkers-van Rossum, A.B. O04.02 Driessen-Waaijer, A. O07.07 Droogh-de Greve, K. O10.06 Dubelaar, I.J.M. O10.02 Duijm, L.E.M. O04.02, O04.04, O09.03 Duyndam, D.A.C. O07.07 Dwarkasing, R.S. O01.03, O01.04 O01.05, O01.06, O08.02, O08.06 Elderen, S.G.C. van O03.01 Elias, R. O01.03, O08.06 Elias, S.G. O10.05 Erkel, A.R. van O01.01, O01.02 Fassaert, T.A. O10.02 Fonville, S. O09.04 Franco, O.H. O09.06 Garsse, L. van O03.05, O03.06 Geelkerken, R.H. O02.01 Geenen, R.W.F. O07.02 Gilhuijs, K.G.A. O07.03, O10.05 Ginneken, B. van O05.02, O06.06 Gondrie, M.J. O06.05 Graaf, R. de O09.01 Graaf, Y. van der O06.05 Granata, C. O06.04 1 7 E

R A D I O L O G E N D A G E N

Gratama, J.W. O10.06 Groen, H.J.M. O06.05 Groenendaal, F. O06.01 Groenewoud, J.H. O04.02 Haan, M.C. de O02.04 Habets, J. O03.04, O11.04 Haloua, M.H. O04.03, O10.03, O10.04 Hattem, A.M.G. van O09.04 Hausen, A. zur O02.05 Heer, L.M. de O01.09 Heestermans, A.A.C.M. O07.02 Heijde, D.M.F.M. van der O12.04 Heijnen, L.A. O02.02, O02.05, O02.06 Heitbrink, M.A. O07.06 Helm-van Mil, A.H.M. van der O12.04 Hemke, R. O06.03 Hendrikse, J. O05.03, O05.04, O06.01 Hermans, J. O12.03 Herwaarden, J.A. van O11.04 Hofman, A. O05.01, O09.06, O12.06 Houston, G.C. O05.05 Hovius, S.E.R. O12.05 Huijbregts, J.E. O08.05 Huisman, A.B. O02.01 Huizinga, T.W.J. O12.04 Ikink, M.E. O01.08, O11.05 Ikram, M.A. O05.01 Imhof-Tas, M.W. O06.06 Irani, F.G. O11.01, O11.02 Isgum, I. O06.05 Issa, A.I. O07.06, O07.08 Jager, G.J. O10.01, O10.02 Jansen, P.L.M. O08.04 Jong, M.D.F. de O10.01, O10.02 Jong, P.A. de O01.09, O03.04, O06.05 Jurencak, T. O03.05, O03.06, O09.05 Karssemeijer, N. O06.06 Keizer, B. de O06.04 Kersbergen, K.J. O06.01 Kesselring, S.C. O04.04 Keymeulen, K. O04.01 Kietselear, B. O03.05, O03.06 Kint, P.A.M. O12.01 Klaveren, R.J. van O06.05 Knegt, R.J. de O08.04 Klomp, D.W.J. O04.06 Knol, R.J.J. O07.02 Koedam, E. O06.06 Köhler, M.O. O07.03 Kok, L.G.C. de O12.06 -

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

67

programma & abstracts auteursindex Kolk, A.G. van der O05.03, O05.04 Kolkman, J.J. O02.01 Koning, H.J. de O06.05 Koopman, M. O08.05 Kotek, G. O12.03 Koudstaal, P.J. O09.04 Krabben, A. O12.04 Krekel, N.M.A. O04.03, O10.03, O10.04 Krestin, G.P. O01.03, O01.04, O01.05, O01.06, O08.02, O08.06, O12.03, O12.06 Kuijpers, T.W. O06.03 Kwee, T.C. O06.04 Lam, M.G.E.H. O08.05, O11.03 Lamb, H.J. O03.01 Lambregts, D.M.J. O02.02, O02.05 O02.06 Lammering, G. O02.03 Lammers, J.W.J. O06.05 Lange, E.S.M. de O04.03 Langedijk, E.R. O06.02 Lavini, C. O06.03 Leenders, G.J.L.H. van O01.05, O01.06 Leeuw, L.A. de O02.01 Leeuwen, M.S. van O08.01 Leijtens, J.W.A. O02.06 Leiner, T. O01.09, O03.03, O03.04, O09.01, O11.04 Leung, K.Y.E. O12.02 Liem, J. O08.02 Lieshout, A. van O07.04 Littooij, A.S. O01.07, O06.04 Lobbes, M.B.I. O04.01, O04.05 Loeffen, D. O03.02 Loo, C. O10.05 Lopes Cardozo, A.M.F. O10.03, O10.04 Louwman, M.W.J. O04.02 Lugt, A. van der O05.01, O09.04, O09.06 Luijten, M. O05.05 Luijten, P.R. O04.06, O05.03, O05.04 Luime, J.J. O12.05 Ly, F. O12.06 Maas, M. O02.02, O02.05, O02.06 Maas, M. O06.03, O07.09 Makanyanga, C. O08.03 Mali, W.P.Th.M. O01.08, O03.04, O04.06, O06.05, O07.03, O10.05, O11.03, O11.05 68

k i j k

o o k

o p o f

Manen, C.J. van O07.05 Martens, M.H. O02.02, O02.05, O02.06 Mattace-Raso, F. O09.06 Meel, E. van O12.06 Meer, R.W. van der O03.01 Meij-de Vries, A. O06.02 Meijer, S. O04.03, O10.03, O10.04 Meiss, L. O06.06 Merckel, L.G. O07.03 Mets, O.M. O06.05 Meurs, J.B.J. van O12.06 Mihl, C. O03.02, O03.06, O09.05 Moll, F.L. O11.04 Muller, S. O10.03, O10.04 Muradin, G.S.R. O12.05 Nae, Y. O03.03 Neck, J.W. van O12.05 Nederend, J. O04.02, O04.04 Nederveen, A.J. O08.04 Niehof, S.P. O12.02 Niet, A. de O08.04 Nieuwkerk, C.M.J. van O08.04 Nievelstein, R.A.J. O01.09, O06.04 Nijsen, J.F.W. O11.03 Nio, C.Y. O08.03 Nusman, C.M. O06.03 Oei, E.H.G. O12.03, O12.06 Oei, L. O12.06 Ommen, V. van O03.05, O03.06 Ong, C.L. O01.07 Ooms, S. O05.05 Oosterhof-Berktas, R. O02.01 Osanto, S. O03.01 Oudkerk, M. O06.05 Ouwendijk, R. O12.05 Paardt, M.P. van der O02.04 Palen, J. van der O02.01 Papanikolaou, N. O02.05 Pendse, D. O08.03 Pengel, K.E. O10.05 Peters-Bax, L. O06.06 Peters, R.P. O07.01, O11.06 Petersen, E.T. O06.01 Pols, H.A.P. O12.06 Ponsen, K.J. O06.02 Prokop, M. O05.02, O06.05 Prompers, L. O04.05 Quekel, L.G.B.A. O06.06 Raaijmakers, R. O03.03

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

Rakic, S.R. O11.06 Ravesteijn, V.F. van O02.04 Reijman, M. O12.03 Reijnierse, M. O12.04 Riedl, R.G. O02.05 Rijn, P.A.C. van O07.02 Rijswijk, C.S.P. van O01.01, O01.02 Rivadeneira, F. O12.06 Roos, A. de O03.01, O09.02, O09.03 Rosenbaum, C.E.N.M. O08.05 Rossum, M.A.J. van O06.03 Rozendaal, L. van O04.05 Rutgers, D.R. O12.01 Rutten, M.J.C.M. O07.04, O10.01, O10.02 Saltzherr, M.S. O12.05 Sassen, S. O02.03 Schaaf, I.C. van der O05.02 Schaefer-Prokop, C.M. O06.06 Schaik, J.P.J. van O12.01 Schalekamp, S. O06.06 Schalkx, H.J. O08.01 Schep, N.W.L O07.09 Schilham, A.M.R. O01.09 Schinkel, L.D. van O03.01 Schip, A.D. van het O11.03 Schipper, R.J. O04.05 Schonenberg, D. O06.03 Schouten, W.R. O01.05 Schreurs, W.H. O04.03 Schuttevaer, H.M. O01.01, O01.02 Schweitzer, K.J. O11.05 Seeters, T. van O05.02 Selles, R.W. O12.05 Selwaness, M. O09.06 Serlie, I.W.O. O02.04 Setz-Pels, W. O09.02, O09.03 Sinkus, R. O08.04 Slaar, A. O07.09 Smidt, M. O04.01, O04.05 Smit, E.J. O05.02 Smits, M. O05.05 Smits, M.L.J. O11.03 Snoeren, M.M. O06.06 Sosef, M. O02.03 Stehouwer, B.L. O04.06 Steketee, R.M.E. O05.05 Stockmann, H.BA.C. O04.03 Stoker, J. O02.04, O06.02, O08.03, O08.04

auteursindex

Stomp, W. O12.04 Stralen, M. van O08.01 Sucha, D. O05.04 Swieten, J.C. van O05.05 Takx, R.A.P. O03.02 Tan, B.S. O11.01, O11.02 Tan, S.G. O11.01, O11.02 Taneja, M. O11.01, O11.02 Tay, K.H. O11.01, O11.02 Taylor, S. O08.03 Teijink, J.A.W. O09.03 Teo, T.K. O11.01, O11.02 Terpstra, V. O08.04 The, G.Y.M. O09.02 Tiehuis, A.M. O06.06 Tiel, J. van O12.03 Tielbeek, J.A.W. O08.03 Tol, M.P. van den O04.03, O10.03, O10.04 Tonino, B.A.R. O12.01 Uitterlinden, A.G. O12.06 Valk, P.H.M. van der O12.02 Veen, H. van der O04.03, O10.03, O10.04 Veldhuis, W.B O04.06, O08.01, O08.05 Velthuis, B. O05.02 Vente, M.A.D. O11.03 Verhaar, J.A.N. O12.03 Verheij, J. O08.04 Verkooijen, H.M. O11.05 Vermoolen, M.A. O06.04 Vernooij, M.W. O05.01, O09.06 Verschakelen, J.V. O07.01 Verschuuren, S.I. O01.05, O01.06 Versteylen, M. O03.02 Verwoert, G.C. O09.06 Viergever, M.A. O06.05 Vis, J.B. de O06.01 Visser, F. O05.03 Vlassenbroek, A. O03.03 Vliegen, R. O02.03 Vliegenthart, R. O06.05 Vlies, D. van der O10.06 Vlies, M. van der O07.07 Vliet, L.J. van O02.04 Vonken, E.J. O05.02 Voogd, A.C. O04.02 Voogt, M.J. O01.08, O11.05 Vos, F.M. O02.04

Vries, B. de O04.05 Vries, L.S. de O06.01 Wals, J. O02.03 Weinans, H. O12.03 Wentzel, J.J. O09.06 Westenberg, J.J.M. O09.02, O09.03 Westerbeek, R.E. O07.05 Wiarda, B.M. O06.02, O07.06 Wildberger, J.E. O03.05, O03.06, O04.01, O04.05, O09.05 Willemink, M.J. O01.09, O03.03, O03.04 Willemse, P.M. O03.01 Willemssen, F.E.J.A. O01.03, O08.02, O08.06 Win, J. de O10.06 Win, H.H. O11.02 Wit, R.H. de O10.04 Witteman, J.C.M. O05.01, O09.06 Wittenberg, R. O06.06 Wittens, C.H.A. O09.01 Wolf, M.A.F. de O09.01 Wolters, S.L. O09.03 Zaag-Loonen, H.J. van der O10.06 Zadi, T. O09.04 Zandvoort, H.J.A. O11.04 Zanen, P. O06.05 Zant, F.M. van der O07.02 Zhuang, K.D. O11.01 Zillikens, M.C. O12.06 Zonnenberg, B.A. O11.03 Zsiros, J. O06.04 Zwanenburg, J.J.M. O05.03

1 7 E

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

69

programma & abstracts aantekeningen

70

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

aantekeningen

1 7 E

R A D I O L O G E N D A G E N

-

2 7

e n

2 8

S E P T E M B E R

2 0 1 2

71

programma & abstracts PLATTEGROND En programma

LIMOUSIN 1

LIMOUSIN 2

LIMOUSIN 3

DEXTER 19

DEXTER 29

FOYER

DEXTER 11-14

DEXTER 21-24

REGISTRATIE

GARDEROBE

ENTREE

72

k i j k

o o k

o p o f

w w w . c o n g r e s s c o m p a n y . c o m w w w . r a d i o l o g e n . n l

Met vriendelijke groet, Vincent Cappendijk Voorzitter Organiserend comité Radiologendagen 2012

Het uitgebreide programma en de omschrijving en de leerdoelen van de Refresher courses en de Interactieve workshops vindt u op pagina 2 t/m 14.

Voorwoord

Geachte collega’s,

Welkom in ‘s Hertogenbosch! Ik hoop van harte dat U de komende 2 dagen als zeer leuk zult ervaren. En trots naar huis zult gaan... Ja! Radiologen, dat zijn wij! Loop in de wereld een willekeurig museum in met een tentoonstelling over de geneeskunde. Altijd zult U 2 essentiële attributen zien die grote stappen voorwaarts in de gezondheidszorg mogelijk hebben gemaakt: De microscoop en het röntgentoestel. Het is voor de radioloog niet bij een röntgentoestel gebleven. Met al deze apparatuur heeft de radioloog een natuurlijke voorsprong om zich te ontplooien. En dat kan... tijdens de Radiologendagen 2012!

De radioloog van de toekomst. De man / vrouw met veel meer competenties dan alleen kennis van zijn apparaten en beoordeling van beelden. Een goede communicator, handige organisator en expert in sociale – en digitale netwerken. Oog voor de wereld om ons heen.

Zeer veel dank aan de sponsoren. Uw investering in onze toekomst is hopelijk ook een waardevolle investering in uw bedrijf. Collega radiologen: Neem graag de moeite een moment van gedachte te wisselen met onze industriële partners.

Graag ook uw blik in het programmaboek op het lijstje OC en WECO leden. Zij hebben met heel veel plezier een mooi programma voor U samengesteld. Speciale dank voor Marion Smits, als voorzitter van het wetenschappelijk comité heeft zij met veel energie het programma tot een gestroomlijnd geheel gemaakt.

In deze introductie wijs ik U niet op potentiële hoogtepunten in het programma. Dat doet U zelf. De radioloog van de toekomst... dat bent U!

Foyer Limousin 2 Limousin 1 Limousin 3 Dexter 11-14 Dexter 21-24 Dexter 19 08:30 – 09:00 Ontvangst & registratie 09:00 – 09:40 Richtlijnensessie 09:45 – 11:00 Parallelsessie 7: Parallelsessie 8: Parallelsessie 9: Parallelsessie 10: Parallelsessie 11: Parallelsessie 12: Educatieve sessie Abdominale radiologie Cardiovasculaire radiologie Mammaradiologie Interventieradiologie Musculoskeletale (deel 2) (deel 2) (deel 2) (deel 2) radiologie / Onderwijs 11:00 – 11:30 Koffiepauze 11:30 – 12:30 Ronde tafel discussie: kapers op de kust 12:30 – 13:00 Prijzensessie Scientific paper award Travel Grant Philipsprijs 13:00 – 14:00 Lunch 14:00 – 15:15 Refresher course: Refresher course: Refresher course: Refresher course: Longafwijkingen na therapie: Nieuwe ontwikkelingen Traumatologie Cases: Tinnitus: oor-zaken en wat is normaal en wat niet? in de interventieradiologie verschil tussen leven en dood diagnostische protocollen 15:20 – 16:05 Interactieve workshop: Interactieve workshop: ‘IK HEB NOOIT EEN CONFLICT’ Het nieuwe opleiden: Bord voor je kop, ethiek voor beginners vermijdend of jokkebrok?

Vrijdag 28 september 2012

Interactieve workshop: RSNA’s teaching file

Dexter 29

Foyer Limousin 2 Limousin 1 Limousin 3 Dexter 11-14 Dexter 21-24 Dexter 19 Dexter 29 09:15 – 09:55 Ontvangst & registratie 09:55 – 10:00 Opening door de voorzitter Radiologendagen 2012 10:00 – 12:00 Openingssessie: Radiologie verleden, heden en toekomst 12:00 – 13:10 Lunch 13:10 – 13:55 Specialist van de toekomst 14:00 – 15:15 Parallelsessie 1: Parallelsessie 2: Parallelsessie 3: Parallelsessie 4: Parallelsessie 5: Parallelsessie 6: Educatieve sessie Abdominale radiologie Cardiovasculaire radiologie Mammaradiologie Neuroradiologie Kinder- en thoraxradiologie (deel 1) (deel 1) (deel 1) (deel 1) 15:20 – 16:35 Refresher course: Refresher course: Refresher course: Refresher course: MRI beeldvorming van Cardiale perfusie: Sports or no sports: Dagelijkse valkuilen in het prostaatcarcinoom which flow to follow? that’s the question de (kinder)neuroradiologie 16:35 – 17:05 Theepauze 17:05 – 17:20 Errors in radiology due to irrationality: hidden traps in day-to-day practice 17:20 – 17:30 Kwaliteitsnotitie NVvR 17:30 – 18:10 Quiz 18:10 – 18:15 Sluiting door de voorzitter 18:15 – 19:15 Borrel 20:00 Aanvang diner & feest in de Orangerie

Donderdag 27 september 2012

programma & abstracts voorwoord

17

sponsoren en exposanten

AGFA HealthCare Alphatron Medical Systems AngioDynamics Bard Benelux Bayer Schering Pharma Bracco Imaging Europe GE Healthcare Guerbet Nederland ICT Automatisering Nederland MML Medical Oldelft Benelux Radiologienetwerk Siemens Nederland Tromp Medical W.L. Gore & Associates

programma & abstracts - 17E nederlandse radiologendagen - 27 en 28 september 2012

Hoofdsponsoren radiologendagen 2012

17

Radioloog van de toekomst programma & abstracts

Radiologendagen 27 - 28 september 2012 1931 Congrescentrum

Brabanthallen

’s-Hertogenbosch

www.radiologen.nl

www.congresscompany.com