JBR 2010-6 by office

WETTEREN 1

P 702083

Volume 93 Page 285-336

November-December

Bimonthly

–

2010

DIAGNOSTIC AND INTERVENTIONAL IMAGING, RELATED IMAGING SCIENCES, AND CONTINUING EDUCATION

ORGANE DE LA SOCIETE ROYALE BELGE DE RADIOLOGIE (SRBR) ORGAAN VAN DE KONINKLIJKE BELGISCHE VERENIGING VOOR RADIOLOGIE (KBVR)

JBR-BTR ♦ 93/6 ♦ 2010 Journal Belge de ♦ Belgisch Tijdschrift voor ♦ RADIOLOGIE

Founded in 1907 A bimonthly journal devoted to diagnostic and interventional imaging, related imaging sciences, and continuing education Contents Cost-effectiveness of multi-detector CT angiography of the coronary arteries for the diagnosis of suspected non-ST elevation acute coronary syndrome (NSTE-ACS) in the emergency department. Mathematical analysis with a decision model. T. De Beule, P. Vanhoenacker, M. de Booij, L. Ardies, O. Bladt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Glomus tumour in the forearm: a case report and review of MRI findings. S. Lee, H. Le, P. Munk, D. Malfair, C.H. Lee, P. Clarkson . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Multifocal primary lymphoma of the cranial vault in a non-immunocomprimsed adolescent. K. Mohd, S.M. Azfar, Q.S.M. Danish, K. Saifullah, A. Mehtab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Breast liposarcoma. L. Fellah, M. Debehogne, J. El Mouedden, N. Akle, M. Berlière, C. Galant, I. Leconte . . . . . . . . . . . . . . . . . . . . . . . An elderly woman with obstructed obturator hernia: a less common variety of external abdominal hernia. L. De Clercq, K. Coenegrachts, T. Feryn, A. Van Couter, P. Vandevoorde, K. Verstraete, H. Rigauts . . . . . . . . . . . . . The association of tethered cord, syringomyelia, disatometamyelia, spinal epidermoid, spinal lipoma and dermal sinus tract in a child. S. Avcu, M. Necat Koseoglu, M. Deniz Bulut, O. Ozen, O. Unal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ectopic infraorbital nerve in a maxillary sinus septum: another potentially dangerous variant for sinus surgery. P. Mailleux, O. Desgain, M.I. Ingabire . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acute abdominal pain due to biliary cystadenoma. K. Van den Bergh, K. Op de beeck . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Prolapsed bilateral ureteroceles leading to intermittent outflow obstruction. H. Stunell, S. Barrett, N. Campbell, E. Colhoun, W.C. Torreggiani . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Antenatal diagnosis of a sacro-coccygeal teratoma. B. Mbumba, A. Massez, P. Lingier, C. Donner, D. Vermeylen, J.R. Makulo, F. Lepira, M. Cassart . . . . . . . . . . . . . .

285 292 296 299 302 305 308 310 312 314

LETTER TO THE EDITOR Epipericardic fat necrosis: CT diagnosis. B. Coulier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317

IMAGES IN CLINICAL RADIOLOGY Oriental cholangitis. S. Maeyaert, L. Bladt, D. Vanbeckevoort, W. Van Steenbergen, W. Van Vaerenbergh . . . . . . . . . . . . . . . . . . . . . . . . Hypertrophic abductor hallucis muscle with nerve compression syndrome. K. Boeren, Y. Vankan, A. Demeyere, D. Perdieus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Celiac artery compression syndrome. M. Eyselbergs, M. Camerlinck, W.J. Metsemakers, J. De Leersnyder, F.M. Vanhoenacker . . . . . . . . . . . . . . . . . . . . Homocystinuria, typical brain MRI findings. P. Iranpour, N. Karamifar, K. As’adi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Unilateral sight loss in a 4-year-old girl. K. De Smet, M. De Maeseneer, A.T. Yazdi, C. Ernst, J. De Mey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rigler’s triad in an 88-year-old woman. K. De Smet, M. De Maeseneer, A.T. Yazdi, I. Bart, J. De Mey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Articular tophaceous gout of the cervical spine: CT diagnosis. B. Coulier, M.H. Tancredi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Massive fatal simultaneous pulmonary and paradoxical embolism: MDCT diagnosis. B. Coulier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SRBR-KBVR General Assembly, Brussels, 20.11.2010 Presidential Address . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Honorary Membership Nominees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . President-Elect Address . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proceedings of papers presented at the RBRS Annual Symposium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Abstract of papers for full membership at the Royal Belgian Society of Radiology . . . . . . . . . . . . . . . . . . . . . . . . Forthcoming courses and meetings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Instructions to Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Subscribers information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Advertising index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

319 320 321 322 323 324 325 326 327 327 328 330 334 335 ii cii ciii

The terms used for indexation of subjects were developed by the Radiological Society of North America (RSNA) over a period of years. Their use here is by permission of the RSNA. The terms may not be used in any other index, print or electronic, except by specific permission of RSNA. ◆◆ Indexed in Index Medicus and in Zentralblatt Radiologie. Evaluated for Medline User, EMBASE and CANCERNET. Abstracted in Excerpta Medica Journals. ◆

JBR–BTR, 2010, 93: 285-291.

COST EFFECTIVENESS OF MULTI DETECTOR CT ANGIOGRAPHY OF THE CORONARY ARTERIES FOR THE DIAGNOSIS OF SUSPECTED NON-ST ELEVATION ACUTE CORONARY SYNDROME (NSTE-ACS) IN THE EMERGENCY DEPARTMENT. MATHEMATICAL ANALYSIS WITH A DECISION MODEL T. De Beule1, P. Vanhoenacker1, M. de Booij1, L. Ardies1, O. Bladt1 The purpose of our study is to model cost-effectiveness of MDCTA for the diagnosis of NSTE-ACS with initially negative enzymes, in the emergency department. In Belgium, the use of multi-detector computed tomography (MDCTA) is probably cost-effective in the diagnosis of NSTE-ACS in the acute setting A decision tree model was developed and a mathematical study was performed that included two hypothetical strategies: MDCTA and admission with classic clinical follow-up and treatment. Costeffectiveness for the Belgian situation was simulated with sensitivity analysis using known values for diagnostic performance and known costs for the different strategies or components of strategies. Key words: Cost-effectiveness – Diagnostic radiology.

Acute chest pain accounts for approximately 6.5 % of all emergency department visits in the US (1, 2) and is considered to lie in the same range in Belgium (oral communication). Failure to diagnose myocardial ischemia as a cause of acute chest pain has serious implications and the triage of patients with possible ischemia is often difficult. To reduce diagnostic error, many patients that present at the emergency department are admitted for observation, even when no initial ECG changes or elevated cardiac enzymes are present. Emergency departments have therefore developed chest pain units and diagnostic protocols commonly including serial cardiac enzyme evaluations and ECG’s, supplemented with some form of stress testing with or without imaging (3). Many of these patients are found to have no acute coronary syndrome (ACS) and in the US more than 2 million patients with acute chest pain are admitted to the hospital without developing an ACS (4, 5). Data from Germany reveal that the number of unnecessary hospital days is high, amounting to as much as 839 per 1000 patients admitted for acute chest pain (6). Non invasive access to coronary anatomy has become available with the emergence of multi-detector computed tomography (MDCTA) of the coronary arteries. Diagnostic performance of MDCTA has been

evaluated in many studies (7). Even though appropriate indications for MDCTA remain largely work in progress, the technique has been used as a tool to rule out non ST segment elevation acute coronary syndrome (NSTE-ACS) in the emergency department (8-15). Some of these studies suggest that MDCTA may be an attractive option to reduce costs when evaluating such patients in the emergency room. However, formal cost-effectiveness studies of MDCTA in this situation have not yet been performed. The purpose of our study was to model cost-effectiveness of MDCTA for the diagnosis of NSTE-ACS with initially negative enzymes, in the emergency department. Since it is know that determination of the exact degree of stenosis on MDCTA is cumbersome (5) we investigated to thresholds for considering MDCTA as positive: a classic approach wherein an obstructive lesion (> 50% diameter) is used a positivity criterion and a more prudent approach, wherein the presence of any degree of plaque is considered positive. A decision tree model was developed and a mathematical study was performed that included two hypothetical strategies: (1) MDCTA (CA), (2) admission with classic clinical follow-up and treatment. Cost-effectiveness for the Belgian situation was simulated with sensitivity analysis using known values for diagnos-

From: 1. Department of Radiology and Imaging, OLV Ziekenhuis, Aalst, Belgium. Address for correspondence: Dr T. De Beule MD, Dept of Radiology, OLV Ziekenhuis, Moorselbaan 164, B-9300 Aalst, Belgium. E-mail: tom.debeule@gmail.com

tic performance and known costs for the different strategies or components of strategies. Methods and materials Decision model Cost-effectiveness can be theoretically studied with a model to mathematically project costs for several strategies. In such a model a decision tree is developed that resembles a flow-chart or algorithm and consists of different nodes (16). To assess potential cost-effectiveness of initial MDCTA we developed such a decision model with two nodes, which compared costs of two model management strategies. The basic decision model used, based on a previous study (17) is displayed in Fig. 1. In the first strategy (MDCTA) the decision whether or not to admit the patient is based on the results of initial MDCTA, performed as fast as possible. Patients with “abnormal” scans are admitted. Patients with “normal” scans are discharged. The model is performed twice, using one out of two practical thresholds for judging a scan result as abnormal: 1/ a less prudent approach: at least one segment with a > 50% diameter stenosis, further referred to as “Obstructive” and 2/ a more conservative approach: at least one segment with plaque of any degree or type, further referred to as “Plaque” In the second strategy (OBSERVATION), all patients are admitted irrespective of their clinical risk assessment score and undergo conventional follow-up and/or treatment. (ECG, Serial enzymes, imaging…).

286

JBR–BTR, 2010, 93 (6)

Values and ranges for costs Values for costs and their ranges are tabulated in Table I and consist of data provided by the Belgian state institute for health insurance (RIZIV/INAMI) for all patient related diagnostic groups (APR-DRG) (20). Values used were the following costs (official terminology between quotation marks): (a) cost for “Scan procedure” (b), observation for “angina pectoris” (c), “acute coronary syndrome”(d). As a typical complication of the scanning procedure and its inherent iodine contrast injection, “acute renal insufficiency” and its related costs were used. The estimated cost for a short admission and quick release after a negative scan (excluding the cost for the scan procedure itself) was rated at the cost of a full observation for angina pectoris divided by an estimated arbitrary value of four (e) (local estimation of costs in our hospital). Except for the additional or altered costs generated by the scan (a), complications(d) or a shorter admission (e), caused by the MDCTA arm, all diagnostic and therapeutic costs for a usual observation or treatment were supposed to be included in the APR-DRG value for “angina pectoris” and “acute coronary syndrome”. This may include cumulated costs for ECG, serial enzymes, imaging (invasive/noninvasive), hospital admission and is an average for all patients in Belgium for the year 2004. All monetary values are expressed in Euro.

Fig. 1. — Decision tree.

Probabilities and test characteristics

Fig. 2. — Modified decision tree used in the decision analysis software

Table I. — Costs given by APR-DRG values. Cost (Euro) Cost “Scan procedure” including contrast medium “Angina pectoris” “Acute coronary syndrome” “Acute Renal Failure” (typical complication)

We choose to include the proportion of non-diagnostic scans and complications, since this may have an influence on the outcome of the model. The final model, used in the software (Tree-age Pro, Healthcare edition 2007) is shown in Fig 2. In our model the terminal nodes or outcomes were cumulated costs to obtain a diagnosis of ACS (Event). Cost-effectiveness was defined as

Coding

190.61 N260 2328.53 APR-DRG 202 4066.25 APR-DRG 190 5357.80 APR-DRG 410

the ratio of the costs to reach a diagnosis and the number of patients correctly diagnosed as having ACS (18, 19). A decrease in the cost per correct diagnosis indicates improved cost-effectiveness. Correct diagnosis of absence of disease was not considered a direct criterion of cost-effectiveness. We did not assess other utilities such as quality adjusted life years.

Values and ranges for probabilities and percentages with 95% confidence intervals (95% CI) are tabulated in Table II. Sensitivity (SE) and specificity (SP): For the first threshold -presence of an obstructive lesion- SE and SP and the number of non-diagnostic scans were taken from a previously published meta-analysis (18). In that study a pooled sensitivity/specificity of 0.95 and 0.90 respectively was found. For the second threshold presence of any degree of plaquedata from a study by Hoffmann (11) were taken. In that study the predictive value an SE/SP of any degree of plaque for ACS was evaluated (SE 1, SP 0.49). Complications Complications of MDCTA were rated at 0.004 (22).

COST EFFECTIVENESS OF CT CORONAROGRAPHY IN ACS — DE BEULE et al

Table II. — Proportions and diagnostic performance for sensitivity analysis. * Values from ref 11, ** from ref 21,*** from reference xx, § no single value was used.

MDCTA Sensitivity using obstructive lesion as threshold

0.95*

MDCTA Specificity using obstructive lesion as threshold MDCTA Sensitivity using presence of any plaque as threshold MDCTA Specificity using presence of any plaque as threshold Pre-test probability Non-diagnostic scans (ND) Complications of MDCTA Pre-test probability Non-diagnostic scans (ND) Complications of MDCTA

0.90* 1*** 0.49*** § 0.03* 0.004 ** § 0.03* 0.004 **

287

even analysis with calculation of crossover point of the two lines with equivalence of the two strategies was calculated A sensitivity analysis and plot of willingness-to-pay per diagnosis (WTP) versus pre-test probability was created for the two diagnostic thresholds, to be able to assess which was the preferred strategy for each combination of WTP and pretest probability. Software used was Tree-age Pro 2007, Healthcare, release 1.0 Results Cost-effectiveness

Fig. 3. — Cost-effectiveness for three hypothetical patients with different pre-test probability (patient 1: 0.13; patient 2: 0.30; patient 3: 0.63)

Calculations Cost-effectiveness Cost-effectiveness of the two strategies was calculated for three types of patients with varying pretest probability (PTP): (1) a 40 yearold female smoker with typical angina pectoris (13% PTP) (2) a 35 yearold male patient with typical angina pectoris but without risk factors (30% PTP) (3) a 60-year-old female diabetic patient with typical angina pectoris (63% PTP). Examples came from a study by Dewey (23). Sensitivity analysis One-way sensitivity analysis of cost-effectiveness on pre-test proba-

bility was performed. This was done by changing the PTP of a positive event in the model. Pre-test probability of NSTE-ACS was used ranging from 0.01 to 0.99 in 100 discrete steps. This was repeated for the two diagnostic thresholds (obstructive and plaque). Sensitivity analysis investigating the influence of different remuneration or reimbursement of MDCTA on cost-effectiveness was performed, in the three hypothetical patients quoted above, using a range of 190.65 Euro – the actual reimbursement in Belgium – to 1500 Euro. This was done again for the two diagnostic thresholds. Cost-effectiveness was plotted against the cost and break-

In Fig 3, a bar graph is displayed that shows the cost per diagnosis of an event for the two strategies in the three hypothetical patients with different pre-test probability. Three bars are provided per patient type, one displaying the observation strategy, the other two displaying the MDCTA strategy with the use of the two positivity thresholds. Highest cost is in the patient with PTP of 0.13 using the observation strategy (19649.79 Euro). Even when the presence of any plaque is used as positivity criterion, there is an important difference between the MDCTA strategy and the observation strategy, with a savings of 3901.51 Euro. The difference between the strategies becomes much less important when moving to the patients with high pre-test probability. Exact values are given in Table III. Sensitivity analysis One-way sensitivity analysis of cost-effectiveness on pre-test probability is displayed in Fig. 4. MDCTA is clearly the most cost-effective strategy in patients that have a pre-test probability that is lower than 0.71 when using as a threshold the presence of an obstructive lesion. For the other threshold MDCTA is most costeffective in patients that have a pretest probability below 0.85. In Fig. 5 A, B graphical examples of sensitivity analysis of cost-effectiveness for remuneration of MDCTA for one of the three hypothetical patients is displayed (low PTP of

Table III. — Cost-effectiveness for three hypothetical patients with varying pre-test probability (PTP), for observation and two thresholds (see also Fig. 3). Strategy Observation MDCTA obstructive MDCTA Plaque

Patient 1(PTP 0,13)

Patient 2 (PTP 0,30)

Patient 3 (PTP 0,63)

19649.79 11928.51 15748.06

9499.78 7148.813 8277.53

5434.1 5234.26 5305.91

288

JBR–BTR, 2010, 93 (6)

Fig. 4. — Cost effectiveness of the two strategies assuming estimates for SE and SP of 0.93 and 0.90 respectively (threshold = presence of obstructive lesion). MDCTA strategy is cost-effective when pre-test probability is below 0.71(A). Cost effectiveness of the two strategies assuming estimates for SE and SP of 1 and 0.49 (threshold= presence of any plaque) respectively. MDCTA strategy is costeffective when pre-test probability is below 0.85 (B). Note that the two curves are lying closer together in B, meaning that the two strategies differ much less in cost-effectiveness. Cost/eff: Cost-effectiveness. K: multiplied by 1000

Fig. 5. — Cost-effectiveness for a patient with pre-test probability of 0.13. MDCTA strategy is cost-effective up to 1133.37 Euro (threshold = presence of obstructive lesion (A). Cost-effectiveness for a patient with pre-test probability of 0.13. MDCTA strategy is cost-effective up to 701.72 Euro (threshold = presence of any plaque) (B).

Table IV. — Break-even values for three hypothetical patients with varying pre-test probability (PTP), for two thresholds. Strategy

Patient 1(PTP 0,13)

Patient 2 (PTP 0,30)

Patient 3 (PTP 0,63)

1133.71 701.72

845.30 557.24

308.46 282.26

MDCTA obstructive MDCTA Plaque

0.13). MDCTA strategy is most costeffective up to a cost of 1133.37 (break even point) when using as a threshold the presence of an obstructive lesion. For the other threshold MDCTA is most cost-effective when the cost for the scanning procedure is below 701.72. Exact val-

ues for break even values are given in Table IV. Break even values in the third patient type tend to approach the actual reimbursement value in Belgium (190.65 Euro) with values of 308.46 for the obstructive threshold and 282.26 for the plaque threshold.

In Fig. 6, sensitivity analysis and a plot of willingness-to-pay per diagnosis versus pre-test probability is displayed for the two thresholds. Combinations of willingness-to-pay and pre-test probability that lie to the right of the curve are in favour of the observation strategy whereas the

COST EFFECTIVENESS OF CT CORONAROGRAPHY IN ACS — DE BEULE et al

289

Table V. — Tabular values for willingness-to pay for the two thresholds and their corresponding pre-test probability (see also Fig. 6). MDCTA 50%

MDCTA Plaque

WTP

0.16 0.27 0.37 0.45 0.51 0.59 0.7 0.79 0.82

0.34 0.47 0.55 0.6 0.63 0.67 0.7 0.73 0.74

100000 50000 30000 20000 15000 10000 5000 2000 1000

Fig. 6. — Sensitivity analysis on pre-test probability and the willingness-to-pay threshold. The line is the division between the optimal test strategy for the base case. In situations with values of PTP and WTP lying to the right of the curve. observation strategy is more cost-effective. WTP: Willingness-to-pay threshold. PTP: Pre-test probability

reverse is true if lying to the left of the curve. From this curve we can deduct that below a WTP of 5000 using threshold “obstructive” is cost-effective up to patients with PTP of 0.70 . Above 5000 for WTP, threshold “plaque” is the most cost-effective for higher PTP. Tabular results can be found in Table V. Discussion With our decision model, we found that in patients presenting with a possible NSTE-ACS, a strategy that uses MDCTA of the coronary arteries as a quick first imaging step is potentially cost-effective in all patients with a pre-test probability lower than 0.71, using a cut-off value for positivity of MDCTA of 50% diameter stenosis. Depending on the pre-test probability of the patient costs are different for the diagnosis of a positive event, amounting to 19649.79 Euro for a typical low PTP patient (0.13) in the observation strategy. Cost for diagnosis of a positive event is much lower in the MDCTA arm: 11928.51 Euro. Costs for a positive diagnosis go down when considering patients with higher pre-test probability, and the difference between the two strategies diminishes to reach a break even at a PTP of 0.71. With respect to the inherent difficulties to exactly grade a stenosis on MDCTA (5), we also ran the same model with a different threshold or positivity criterion: the presence of any plaque. We observed the same trends. However

savings are substantially lower than with the use of the “obstructive” threshold: the cost for a typical low PTP patient was 15748.06 euro. We also found break-even values for the cost of the scanning procedure in the MDCTA strategy, that are substantially higher than the actual reimbursement, at least in low pre-test probability groups. Irrespective of the threshold used, MDCTA can probably be used as a reliable and cost-effective technique to rule NST-ACS in a acute setting and the test can be done very quickly and does not require complex patient preparation or logistic resources. The scan can be performed in the first hours after the onset of the clinical complaints and when negative, the patient can be discharged from the emergency department immediately. Although the purpose of this study was not to formally study the reliability of MDCTA to exclude all causes of chest pain, it is apparent from the individual data from most studies in this series that alternative diagnoses besides coronary disease can often be ruled out. These conclusions are important and can probably be an incentive for further randomised trials that should confirm these assumptions in reallife. Only one randomized study using an approach with MDCTA has been published and although the patient sample was rather small, the diagnostic efficacy was equal in the two arms, but the time to patient discharge was much shorter in the

MDCTA arm and this approach was probably cheaper than in the conventional arm (14). It can also be of value to regulatory offices and authorities and insurers. Cost-effectiveness studies for the diagnosis of coronary disease with non-invasive tests are found throughout the literature, and specifically it has been demonstrated that MDCTA is cost-effective in the diagnosis of stable coronary disease (2426). In one of these studies (25) a first triage of patients was done with myocardial perfusion scanning but this was only in stable patients. To our knowledge this is the first theoretical decision analysis that also suggests that in the acute setting, MDCTA can be cost-effective. In one other study (14), the patients in the MDCTA arm with equivocal results, or intermediate lesions were subjected to myocardial perfusion scanning, to avoid a simple bifurcated decision based on a simple threshold of 50% diameter stenosis. We tried to circumvent this problem by running the model on a lower threshold: “plaque”, that can be considered more prudent and less aggressive in discharging patients. We choose this threshold because it was the only one available in the literature: further data relating degree of stenosis on MDCTA to prediction of an acute coronary event is lacking. Other imaging techniques have been used to develop a quick and efficient triage strategy for acute chest pain in the emergency department. Nuclear medicine techniques and to a much lesser degree, echocardiography, have been extensively tested, and there is a large body of literature supporting its use in these circumstances (26). In 2003 a joint task force of the American College of Cardiology, the American Heart Association and the American Society for Nuclear Cardiology pub-

290

lished guidelines for the use of Myocardial Perfusion Imaging (MPI) (27). The task force gave a class 1 recommendation to the use of acute rest MPI for the assessment of patients presenting with a possible acute coronary syndrome in whom initial markers and the ECG are non diagnostic. A randomized study (37) showed that MPI improves triage and that unnecessary hospitalization was reduced among patients without acute ischaemia, without reducing appropriate admission for patients with acute ischemia. Wether MDCTA can be recommended as a substitute for MPI needs to be investigated further. Radiation dose may be a matter of debate. A clear disadvantage of MDCTA is the relatively high radiation dose that goes with the examination, with average doses ranging from 10 to 20 mSv (38). How this relates to the dose of MPI now and in the future is a matter that goes beyond this discussion, but it may be anticipated that radiation dose reducing protocols for MDCTA will have an important impact (38). Some important limitations have to be acknowledged. The amounts that have been used for the costs are the values that are used in Belgium, which biases the results towards local habits. It was assumed however that relative scaling of these amounts would be similar in other countries and that at least an identical decision tree could be used for different countries. The monetary values used for the APRDRG categories are only a rough and averaged approximation of all costs incurred and probably prone to inaccuracies. Although the pooled values for SE, SP and number of non-diagnostic scans were pooled from studies that specifically targeted the emergency situation of suspected NSTACS with initially negative troponins, it may be that these statistics are not as good in a general population of acute patients that have an irregular rhythm, high age, high body mass index, heavily calcified vessels or otherwise suboptimal patients (27-28). Third, this study was “theoretical”, meaning that the conclusions were not based on prospective clinical investigations but on a mathematical model. The model only looks at a very simple modification of the classic observation by adding only one test, early in the evaluation. Last, this simple model took not into account the possible downstream costs induced, due to unex-

JBR–BTR, 2010, 93 (6)

pected cardiac findings. We did not look at the cost of finding or ruling out significant or non significant extra-cardiac findings. All these limitations can be solved with prospective randomized studies that perform real-life accounting of all included patients. The conclusion of the study is that, for the Belgian situation, the use of MDCTA is probably cost-effective in the diagnosis of NSTE-ACS in the acute setting. The current reimbursement of MDCTA of the coronary arteries is cost-effective but real life prospective randomized studies have to be performed to confirm this and to investigate subpopulations with clinical characteristics that make them less or more amenable to MDCTA. An important consideration has to be made about the required level of expertise to accurately evaluate the coronary stenoses. Physicians inexperienced with coronary CT angiography have a clearly lower sensitivity and specificity compared to expert radiologists. Their ability to evaluate significant coronary stenoses can improve after one year of training with approximately 600 cases a year. An increase of sensitivity is seen after one year of training, ranging from 66%-75% (31). Published research about diagnostic performance was almost always carried out by physicians with 3 or more years of experience, resulting in a sensitivity of 90%. One of the suggested theories of this major difference in sensitivity is the approximately 10 years of preparation needed to attain a certain expertise in any domain (32). This period may overlap with normal medical training resulting in a required training period of about 3 years, a seen with expert radiologists.

Conclusion 9.

For the Belgian situation, the use of MDCTA is probably cost-effective in the diagnosis of NSTE-ACS in the acute setting. The current reimbursement of MDCTA of the coronary arteries is cost-effective but real life prospective randomized studies have to be performed. Clinical subgroups with less MDCTA-friendly characteristics have to be investigated. References 1. Lee T.H., Rouan G.W., Weisberg M.C., Brand D.A., Acampora D., Stasiulewicz C., Walshon J., Terranova G.,

10.

11.

Gottlieb L., Goldstein-Wayne B., Clinical characteristics and natural history of patients with acute myocardial infarction sent home from the emergency room. Am J Cardiol, 1987, 60: 219-224. McCaig L.F., Burt C.W. National Hospital Ambulatory Medical Care Survey: 2002 emergency department summary. Adv Data, 2004: 1-34. Pope J.H., Aufderheide T.P., Ruthazer R., Woolard R.H., Feldman J.A., Beshansky J.R., Griffith J.L., Selker H.P. Missed diagnoses of acute cardiac ischemia in the emergency department. N Engl J Med, 2000, 342: 1163-1170. Hoffmann M.H., Shi H., Schmitz B.L., Schmid F.T., Lieberknecht M., Schulze R., Ludwig B., Kroschel U,. Jahnke N., Haerer W., Brambs H.J., Aschoff A.J. Noninvasive coronary angiography with multislice computed tomography. JAMA, 2005, 293: 2471-2478. Leber A.W., Knez A., von Ziegler F., Becker A., Nikolaou K., Paul S., Wintersperger B., Reiser M., Becker C.R., Steinbeck G., Boekstegers P. Quantification of obstructive and nonobstructive coronary lesions by 64-slice computed tomography: a comparative study with quantitative coronary angiography and intravascular ultrasound. J Am Coll Cardiol, 2005, 46: 147-154. Raff G.L., Gallagher M.J., O'Neill W.W., Goldstein J.A. Diagnostic accuracy of noninvasive coronary angiography using 64-slice spiral computed tomography. J Am Coll Cardiol, 2005, 46: 552-557. Hamon M., Biondi-Zoccai G.G., Malagutti P., Agostoni P., Morello R., Valgimigli M., Hamon M. Diagnostic performance of multislice spiral computed tomography of coronary arteries as compared with conventional invasive coronary angiography: a meta-analysis. J Am Coll Cardiol, 2006, 48: 1896-1910. White C.S., Kuo D., Kelemen M., Jain V., Musk A., Zaidi E., Read K., Sliker C., Prasad R. Chest pain evaluation in the emergency department: can MDCT provide a comprehensive evaluation? AJR, 2005, 185: 533-540. Gallagher M.J., Ross M.A., Raff G.L., Goldstein J.A., O'Neill W.W., O'neil B. The diagnostic accuracy of 64-slice computed tomography coronary angiography compared with stress nuclear imaging in emergency department low-risk chest pain patients. Ann Emerg Med, 2007, 49: 125-136. Hoffmann U., Pena A.J., Moselewski F., Ferencik M., Abbara S., Cury R.C., Chae C.U., Nagurney J.T. MDCT in early triage of patients with acute chest pain. AJR, 2006, 187: 1240-1247. Hoffmann U., Nagurney J.T., Moselewski F., Pena A., Ferencik M., Chae CU., Cury R.C., Butler J., Abbara S., Brown D.F., Manini A., Nichols J.H., Achenbach S., Brady T.J. Coronary multidetector computed

COST EFFECTIVENESS OF CT CORONAROGRAPHY IN ACS — DE BEULE et al

12.

13.

14.

15.

16.

17.

18.

tomography in the assessment of patients with acute chest pain. Circulation, 2006, 114: 2251-2260. Olivetti L., Mazza G., Volpi D., Costa F., Ferrari O., Pirelli S. Multislice CT in emergency room management of patients with chest pain and mediumlow probability of acute coronary syndrome. Radiol Med (Torino), 2006, 111: 1054-1063. Sato Y., Matsumoto N., Ichikawa M., Kunimasa T., Iida K. ,Yoda S., Takayama T., Uchiyama T., Saito S., Nagao K., Tanaka H., Inoue F., Furuhashi S., Takahashi M., Koyama Y. Efficacy of multislice computed tomography for the detection of acute coronary syndrome in the emergency department. Circ J, 2005, 69: 10471051. Goldstein J.A., Gallagher M.J., O'Neill W.W., Ross M.A., O'Neil B.J., Raff G.L. A randomized controlled trial of multi-slice coronary computed tomography for evaluation of acute chest pain. J Am Coll Cardiol, 2007, 49: 863-871. Rubinshtein R., Halon D.A., Gaspar T., Jaffe R., Karkabi B., Flugelman M.Y., Kogan A., Shapira R., Peled N., Lewis B.S. Usefulness of 64-slice cardiac computed tomographic angiography for diagnosing acute coronary syndromes and predicting clinical outcome in emergency department patients with chest pain of uncertain origin. Circulation, 2007, 115: 17621768. Plevritis S.K. Decision analysis and simulation modeling for evaluating diagnostic tests on the basis of patient outcomes. AJR, 2005, 185: 581-590. Radensky P.W., Hilton T.C., Fulmer H., Mc Laughlin B.A., Stowers S.A. Potential cost effectiveness of initial myocardial perfusion imaging for assessment of emergency department patients with chest pain. Amer J Cardiol, 1997, 79: 595-599. Patterson R.E., Eng C., Horowitz S.F. Practical diagnosis of coronary artery disease: a Bayes' theorem nomogram to correlate clinical data with

19.

20. 21.

22.

23.

24.

25.

26.

27.

noninvasive exercise tests. Am J Cardiol, 1984, 53: 252-256. Patterson R.E., Eng C., Horowitz S.F., Gorlin R., Goldstein S.R. Bayesian comparison of cost-effectiveness of different clinical approaches to diagnose coronary artery disease. J Am Coll Cardiol, 1984, 4: 278-289. https://tct.fgov.be/etct/anonymous? lang=nl. 2007. Vanhoenacker P.K., Decramer I., Bladt O., Sarno G., Bevernage C., Wijns W. Detection of non-STElevation myocardial infarction and unstable angina in the acute setting: meta-anlaysis of diagnostic performance of multi-detector compute tomographic angiography. BMC Cardiovascular Disorders, 2007, 7: 39 Katayama H. ,Yamaguchi K., Kozuka T., Takashima T., Seez P., Matsuura K. Adverse reactions to ionic and nonionic contrast media. A report from the Japanese Committee on the Safety of Contrast Media. Radiology, 1990, 175: 621-628. Dewey M., Hamm B. Cost effectiveness of coronary angiography and calcium scoring using CT and stress MRI for diagnosis of coronary artery disease. Eur Radiol, 2007, 17: 13011309. Kuntz K.M., Fleischmann K.E., Hunink M.G., Douglas P.S. Cost-effectiveness of diagnostic strategies for patients with chest pain. Ann Intern Med, 1999, 130: 709-718. Cole J.H., Chunn V.M., Morrow J.A., Buckley R.S., Phillips G.M. Cost implications of initial computed tomography angiography as opposed to catheterization in patients with mildly abnormal or equivocal myocardial perfusion scans. J Cardiovasc CT, 2007, 1: 21-26 Ionnanidis J.P., Salm D., Chew P.W., Lau J. Accuracy of imaging technologies in the diagnosis of acute cardiac ischemia in the emergency department: a meta-analysis. Ann Emerg Med, 2001, 37: 471-477 Klocke F.J., Baird M.G., Lorell B.H., Bateman T.M., Messer J.V., Berman D.S., O’Gara PT, Carabello B.A., Russel R.O.,

28.

29.

30.

31.

32.

291

Verqueira M.D., John Sutton M.G., DeMaria A.N., Udelson J.E., Kennedy J.W., Verani M.S., Williams K.A., Antman E.M., Smith S.C., Alper J.S., Gregoratis G., Anderson J.L., Hiratzka L.F., Faxon D.P., Hunt S.A., Fuster V., Jacobs A.K., Gibbons R.J., Russel R.O. ACC/AHA/ASNC guidelines for the clinical use of cardiac radionuclide imaging – executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/ASNC Committee to Revise the 1995 Guidelines for the Clinical Use of Cardiac Radionuclide Imaging). J Am Coll Cardiol, 2003, 42: 1318-1333. Rumberger J.A., Behrenbeck T., Breen J.F., Sheedy P.F. Coronary calcification by electron beam computed tomography and obstructive coronary artery disease: a model for costs and effectiveness of diagnosis as compared with conventional cardiac testing methods. J Am Coll Cardiol, 1999, 33: 453-462 Shapiro M.D., Butler J., Rieber J., Sheth T.N., Cury R.C., Ferencik M., Nichols J.H., Goehler A., Abbara S., Pena A.J., Brady T.J., Hoffmann U. Analytic approaches to establish the diagnostic accuracy of coronary computed tomography angiography as a tool for clinical decision making. Am J Cardiol, 2007, 99: 1122-1127. Huber S., Huber M., Dees D., Redmond F.A., Wilson J.M., Flamm S.D. Usefulness of multislice spiral computed tomography coronary angiography in patients with acute chest pain in the emergency department. J Cardiovasc CT, 2007, 1: 29-37 Francesca Pugliese M., Hununk et al. Learning curve for coronary CT angiography: What constitutes sufficient training. Radiology, 2009, 251: 359-367 Ericsson K.A., Charness N. Expert performance: its structure and acquisition. American psychologist, 1994, 49: 725-747.

JBR–BTR, 2010, 93: 292-295.

GLOMUS TUMOUR IN THE FOREARM: A CASE REPORT AND REVIEW OF MRI FINDINGS S. Lee, H. Le, P. Munk , D. Malfair, Ch.H. Lee, P. Clarkson1 Glomus tumours are uncommon benign neoplasms characterised by the proliferation of modified smooth muscle cells known as glomus cells. Glomus tumours are well described in the extremities, particularly in the sub-ungual region and MRI is well established as the investigation of choice. However, a significant proportion of glomus tumours are extra-digital, but the discussion of MRI findings of extra-digital tumours is limited and restricted to case reports. We present a case of a solitary painful forearm lesion in an 81-year-old man, and review the English literature on extra-digital glomus tumours documenting MR imaging features. Radiologists should be aware of the existence of these lesions, particularly in the setting of chronic pain and focal tenderness. Key-words: Paraganglioma – Extremities, MR.

Glomus tumours are uncommon benign neoplasms characterised by the proliferation of modified smooth muscle cells known as glomus cells. Glomus cells play a role in thermoregulation. Glomus tumours are well described in the extremities, particularly in the sub-ungual region and MRI is well established as the investigation of choice (1). However, a significant proportion of glomus tumours are extra-digital, but the discussion of MRI findings of extradigital tumours is limited and restricted to case reports (2-13). We present a case of a solitary painful forearm lesion in an 81-year-old man, and review the English literature on extra-digital glomus tumours documenting MR imaging features. Radiologists should be aware of the existence of these lesions, particularly in the setting of chronic pain and focal tenderness. Case report

Fig. 1. — MRI left proximal forearm (GE echospeed 1.5T) demonstrating a subcutaneous lesion overlying the extensor compartment (arrow). A: TE 14 TR783, B: TE105 TR3700. The lesion is well defined with a hypointense capsule. It is hypointense on T1 and hyperintense on T2 relative to muscle.

An 81-year-old man presented with a sub-cutaneous nodule in the dorsal forearm that had been slowly growing over 10 years. It was slightly bluish in colour. It had become increasingly painful and was exquisitely tender on palpation. MRI demonstrated a 6 x 12 x 14 mm ovoid subcutaneous lesion which abutted the superficial muscular fascia on the extensor compartment (Fig. 1). It was hypointense on T1 and hyperintense on T2 and STIR. It was well marginated and no intralesional fat was seen. There was no blooming seen on gradient echo sequence. On ultrasound, the lesion was solid but uniformly hypoechoic. Fig. 2. — High power H&E stain of core biopsy demonstrating nests to sheets of uniform rounded cells with centrally located round nuclei and From: 1. Department of Radiology, Vancouver General Hospital, Vancouver, Canada. eosinophilic cytoplasm, set in a richly Address for correspondence: Dr S. Lee, MBBS, FRANZCR, Department of Radiology, vascular background of capillary-sized Vancouver General Hospital, 899 12th Avenue West, Vancouver, BC V5Z 1M9, Canada. vessels.

GLOMUS TUMOUR IN THE FOREARM — LEE et al

293

Table I. — Summary table of cases of extra-digital glomus tumours with MRI and/or ultrasound findings. Author

Age/Sex

Site & Presentation

Presentation

MRI

Ultrasound

Smith

62 F

Radial nerve

Pain

25 mm hypoechoic well defined lesion

Mabit

45 M

Patellar tendon

Pain

Ovoid 1 cm. ↑ T1 Only T1 weighted sequences performed

Patella tendon enlarged and hypoechoic.

Yoshikawa

35 M

Supraspinatus

Pain

Ovoid 4 x 2 cm ↔ T1 ↑ T2

Amillo

38 F

Vastus lateralis

Pain

Ovoid, 3 cm ↓ T1 ↑ T2 Enhanced, central nonenhancing area

Mohler

55 F

Plantar foot

Pain

Ovoid 1 cm ↓ T1 ↑ T2

Hardy

65 M

Hoffa fat pad

Pain

Ovoid 10 mm ↓ T1 ↑ T2

Ovoid hypoechoic mass

Enhanced McDonald

40 M

Buttock – subcutaneous

Pain

Ovoid, 6 mm ↓ T1 ↑ T2

Abela

52 M

Scapula region subcutaneous

Dull ache

Round 10 mm ↔ T1 ↑ STIR

Hyper-echoic, well defined, prominent vascularity [not shown]

[images suggest ↓T1] GonzalezLlanos

50 M

Periosteal, distal femoral diametaphysis

Pain

Ovoid, 12 mm ↓T1 ↑T2 Enhanced.

Waseem

73 M

Subcutaneous knee

Pain

Ovoid 50 mm ↓ T1 ↑ T2

Senol

21 M

Forearm ?superficial, adjacent to ulnar sensory nerve

Pain, neuralgia

Ovoid 20 x 10 mm ↔ T1 ↑ T2, ↑ TIRM, enhanced.

Ovoid, hypoechoic

Legend: ↑ Increased signal intensity compared to muscle ↓ Decreased signal intensity compared to muscle ↔ Signal intensity similar to muscle.

An ultrasound guided core biopsy was performed, which showed nests of uniformly appearing rounded cells with centrally located round

nuclei and eosinophilic cytoplasm, set in a richly vascular background of capillary-sized vessels with varying degree of perivascular hyalinization

(Fig. 2). There were no mitotic figures or cytologic atypia present. Tumour cells demonstrated strong Hcaldesmon and moderate smooth

294

muscle actin immunoreactivity and were negative for keratin and S-100. These findings confirm the diagnosis of a benign glomus tumour. The patient underwent uncomplicated excision of the lesion without any perioperative complications, resulting in resolution of his symptoms. Pathology of the specimen confirmed the diagnosis. Discussion Glomus tumours are thought to arise from glomus cells, which form part of glomus bodies. These are arteriovenous shunts typically found in the dermal-sub-dermal junction and involved in thermo-regulation. The majority of these tumours are solitary, and the multiple form, also known as glomangiomatosis, has distinct clinical and pathological characteristics. Glomus tumours are well described in the extremities, particularly in the sub-ungual region. Radiological findings such as chronic pressure erosion on phalangeal tuffs are well described. Extra-digital glomus tumours, although uncommon, comprise a significant proportion of glomus tumours. Heys et al reported 27 of their 43 cases of glomus tumours occurring outside the hand (2). These lesions tend to present in adults, and are often painful, presumably due to the proximity to the rich neural bed (14). They are small, typically smaller than 2 cm in diameter. Superficial lesions often have a bluish colouration, but deeper lesions are difficult to detect on physical examination (3). Most of the extra-digital glomus tumours are sub-cutaneous, but they can occur at any site including intraosseous, periosteal, intramuscular, intravenous and intraneural locations (4, 14). Radiologists should be aware of this entity, particularly in the setting of pain and tenderness at the affected site. MRI has a role in diagnosing extra-digital glomus tumours, especially when its small size makes clinical diagnosis difficult. The case reports in English literature on extradigital glomus tumours find that these lesions are ovoid, well defined and show hypointensity or isointensity on T1 and hyperintensity on T2 weighted sequences on MRI (Table I). The reported cases range from 6 mm to 50 mm in maximal dimension. Takei described a case of a 4 mm lesion adjacent to the superficial radial nerve which MRI did not detect (14). All reported cases which

JBRâ&#x20AC;&#x201C;BTR, 2010, 93 (6)

were administrated gadolinium demonstrated at least some contrast enhancement. Gonzalaz, Senol, and Hardy demonstrated homogenous enhancement (4-6). Amillo had a case of a 3 cm lesion within vastus lateralis that had a central nonenhancing area on the presented image; however, this finding was not discussed (7). Yoshikawa performed a non-contrast MR, but the post-contrast CT demonstrated peripheral enhancement of the lesion (8). Yoshikawa also documented multiple small foci of calcifications within the lesion. Our case demonstrated similar imaging characteristics with a well defined lesion showing T1 hypointensity and T2 hyperintensity. We did not administer gadolinium. The MR imaging features are nonspecific and the differential diagnosis will include nerve sheath tumours, epidermal cysts, venous malformations, nodular fasciitis, and angioleiomyoma. Cat-scratch disease may also have similar imaging characteristics but the lesions are typically located at nodal stations such as epitrochlear region of the elbow and axilla. There is some variability of the ultrasound appearance of extradigital glomus tumours. In our case, the lesion was well defined, solid, and uniformly hypoechoic. This is similar to that reported by Smith, Amillo, and Gonzalez (9, 7, 4). Abela showed a case of a subcutaneous lesion in the shoulder region which was hyperechoic (3). Mabit presented a case of a lesion within the patellar tendon which showed expansion of the tendon with hypoechogenicity, but it was not mentioned whether there was a discrete lesion seen on ultrasound (10). Pathologically, glomus tumours are usually small well-defined lesions. Microscopically, the tumor cells are uniform and round in shape with eosinophilic cytoplasm and centrally located round nuclei. The appearance can vary depending on the relative amount of glomus cells, vascular structures and smooth muscle (15). The current case shows the solid pattern with a predominance of glomus cells arranged in solid nests surrounded by capillary sized vessels. In contrast, lesions showing a predominance of dilated venous vessels and a predominance of smooth muscle cells are referred to as glomangiomas and glomangiomyomas respectively. Immunohistochemically, glomus cells are positive for vimentin, smooth muscle actin and H-caldesmon, and lack

the expression for epithelial, melanocytic and vascular markers. Treatment is surgical excision. Recurrence is rare, and usually related to incomplete excision. Malignant glomus tumours and metastases have been reported but are extremely rare (16, 17). These malignant glomus tumours typically occur in the subfascial or visceral locations, are larger (> 2 cm) in size and demonstrate malignant or atypical histologic features that include marked cytologic atypia and increased mitotic activity with atypical mitotic figures (15). Although rare, radiologists should be aware of extra-digital glomus tumours. Although the imaging features are non-specific, the clinical notion of pain is an important clue in the characterisation process. Prompt diagnosis and complete excision of the tumour often alleviates patientâ&#x20AC;&#x2122;s pain. References 1. Drape J., Idy-Peretti I., Goettmann S., Wolfram-Gabel R., Dion E., Grossin M.: Subungual glomus tumours: Evaluation with MR imaging. Radiology, 1995, 195: 507-515. 2. Heys S.D., Brittenden J., Atkinson P., Eremin O.: Glomus tumour: an analysis of 43 patients and review of the literature. Br J Surg, 1992, 79: 345-347. 3. Abela M., Cole A.S., Graham H.A., Carr A.J.: Glomus tumor of the scapular region. J of Shoulder Elb Surg, 2000, 9: 532-533. 4. Gonzalex-Llanos F., Lopez-Barea F., Isla A., Fernandez-Prieto A., Zubillaga A., Alvarez F.: Periosteal glomus tumor of the femur: a case report. Clin Orthop Relat R, 2000, 380: 199203. 5. Senol M.G., Oguz M., Haholu A., Saracoglu M.: Glomus tumor of forearm: a rare cause of neuralgia. Acta Neurol Belg, 2007, 107: 58-59. 6. Hardy P., Muller G.P., Got C., LortatJacob A., Benoit J.: Glomus tumor of the fat pad. Arthroscopy, 1998, 14: 325-328. 7. Amillo S., Arriola F., Munoz G.: Extradigital glomus tumour causing thigh pain: A case report. J Bone Joint Surg [Br], 1997, 79-B: 104-106. 8. Yoshikawa G., Murakami M., Ishizawa M., Matsumoto K., Hukuda S.: Glomus tumor of the musculotendinous junction of the rotator cuff: a case report. Clin Orthop Relat R, 1996, 326: 250-253. 9. Smith K., Mackinnon S., Macauley R., Mailis A.: Glomus tumor originating in the radial nerve: a case report. J of Hand Surg, 1992, 17A: 665-667. 10. Mabit C, Pecout C, Arnaud J.P.: Glomus tumor in the patellar ligament. A case report. J Bone Joint Surg Am, 1995, 77: 140-141.

GLOMUS TUMOUR IN THE FOREARM — LEE et al 11. Mohler D.G., Lim C.K., Martin B.M.: Glomus tumor of the plantar arch: a case report with magnetic resonance imaging findings. Foot Ankle Int, 1997, 18: 672-674. 12. McDonald J., Moonka R.: Extradigital glomus tumor as a cause of buttock pain. Orthopedics, 2000, 23: 851-852. 13. Waseem M., Jari S., Paton R.W.: Glomus tumour, a rare cause of knee

pain: a case report. Knee, 2002, 9: 161-163. 14. Takei T.R., Nalebuff E.A.: Extradigital glomus tumour. J Hand Surg – Brit Eur, 1995, 20B: 409-412. 15. Fletcher C.D., Unni K.K., Martens F.: Pathology and genetics of tumors of soft tissue and bone. WHO Classification of Tumors, 2002: 156177.

16. Kayal J.D., Hampton R.W., Sheehan D.J., Washington C.V.: Malignant glomus tumour: A case report and review of the literature. Dermatol Surg, 2001, 27: 837-840. 17. Folpe A., Fanburg-Smith J., Miettinen M., Weiss S.: Atypical and malignant glomus tumors. Am J Surg Pathol, 2001, 25: 1-12.

Our selection of new radiology titles! Meyers' Dynamic Radiology of the Abdomen 6/e Normal and Pathologic Anatomy Springer Verlag – December 2010 € 213.95 Diagnostic Imaging Head & Neck 2e Harnsberger LWW / Amirsys – November 2010

€ 294.90

Echographie et imagerie pelvienne en pratique gynécologique 5/e Ardaens Y. Masson – Nov. 2010 – € 153.85 special price valid till 28/02/2011 (normal price: € 169.35) Head and Neck Radiology – 2 Volumes Mancuso Lippincott – November 2010

€ 421.10

Orthopedic Imaging 5/e - A Practical Approach Greenspan A. Lippincott – Nov. 2010 € 234.95 ACCO Leuven M-Theresiastraat 2 3000 Leuven Tel 016/29.11.00 Fax 016/20.73.89

ACCO Adrénaline 43, Rue Martin V 1200 Bruxelles Tel 02/763.16.86 Fax 02/772.10.04

ACCO Gent St-Pietersnieuwstr. 105 9000 Gent Tel 09/235.73.00 Fax 09/235.73.01

acco.medical@acco.be www.accomedical.be

295

JBR–BTR, 2010, 93: 296-298.

MULTIFOCAL PRIMARY LYMPHOMA OF THE CRANIAL VAULT IN A NONIMMNUNOCOMPROMISED ADOLESCENT M. Khalid, M.A. Siddiqui, S.M.D. Qaseem, M. Ahmad, S. Khalid1 Primary non-Hodgkin lymphoma (NHL) of the skull is extremely rare. The authors report a case in a 19-year-old boy who presented with a progressively increasing scalp mass on background of generalized headache of 6-months duration. Imaging showed multifocal skull vault lesion with intra- and extra-cranial soft tissue masses, causing permeative destruction of underlying skull bone. Further investigation failed to identify any other evidence of systemic lymphoma. Histopathology examination of superficial scalp mass showed low grade non-Hodgkin B cell lymphoma. The precise anatomical location and multifocal nature of the neoplasm occurring in young nonimmunocompromised individual makes this case clinically and radiologically unique. Key-word: Primary cranial vault – Non-Hodgkin lymphoma – Nonimmunocompromised.

Non Hodgkin lymphoma (NHL) is an uncommon neoplasm with an incidence of only 3% to 4% in general population. It occurs more frequently in patients with acquired immunodeficiency syndrome (AIDS). Primary involvement of the central nervous system (CNS) occurs only in 1% to 2% of patients with lymphoma (1). NHL originating primarily in bones is rare, occurring only in 3-4% of patients with lymphomas (1), with a predilection for the long bones of the upper and lower extremities, the pelvis and the spine. NHL presenting with primary involvement of cranial vault and dura mater is an unusual event and have rarely been reported in literature. Out of 16 cases reported, 7 cases were associated with systemic disease on presentation (2). In the other 9 cases, the disease was restricted to the cranial vault and these cases were therefore described as primary cranial vault lymphomas. Only 3 out of 9 cases were non-HIV positive. We report an unusual case of primary cranial vault lymphoma with multifocal involvement that occurs in a nonimmunocompromised adolescent boy. Case report A 19-year-old boy presented with generalized headache and progressively increasing swellings on the left side of scalp which had been developing over a period of 6months. His medical history was otherwise unremarkable. There was no history of head trauma.

Fig. 1. — A. Non contrast enhanced CT image shows extra-cranial mass lesion with intra-cranial extradural extension in left rolandic region. B. Post contrast enhanced image shows heterogeneous enhancement of mass lesion.

On initial examination the patient was alert and oriented; there were no focal neurological deficits. Physical examination revealed a 6x6 cm firm smooth, non tender swelling in left parietal region. There was no local rise of temperature and overlying skin was normal. There was no lymphadenopathy, hepatomegaly or splenomegaly. Results of hematological and blood chemistry tests were within normal limits. X ray chest was normal. Patient was HIV infection negative. Patient underwent computed tomography (CT) scan of the brain which confirmed the presence of the extra-cranial mass and showed its intra-cranial extradural extension in left fronto-parietal (rolandic) region. The extradural lesion exerted mass effect on the underlying brain in the form of displacement of gray-white

From: 1. Department of Radiodiagnosis, Jawaharlal Nehru Medical College, AMU, Aligarh, India. Address for correspondence: Dr Mohammed Azfar Siddiqui, M.D., Department of Radiodiagnosis, Jawaharlal Nehru Medical College, AMU, Aligarh-202002, India. E-mail: drazfarsiddiqui@gmail.com

matter interface without any evidence of cerebral edema (Fig. 1). Lesion was slightly hyperdense than brain parenchyma on pre contrast study. After contrast administration the lesion was heterogeneously enhancing and bone window revealed permeative destruction of both inner and outer table (Fig. 2B). In addition CT scan also revealed another intracranial extradural lesion overlying right occipitocerebellar region (Fig. 2B), however the underlying bone was normal. The superficial scalp mass was biopsied and diagnosis of low grade nonHodgkin’s B-cell lymphoma was made (Fig. 3A). Further work-up with CT examination of neck, thorax, abdomen and pelvis did not reveal presence of lymphoma at any other site. Ga-scintigraphy was negative. Whole-body fluorodeoxyglucose positron emission tomography (FDG-PET), bone marrow biopsy and cerebrospinal fluid examination also did not showed any other lesion. Patient was treated with cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP)

PRIMARY LYMPHOMA OF THE CRANIAL VAULT — KHALID et al

Fig. 2. — A. Post contrast CT image shows mass lesion in right occipito-cerebellar region. B. CT with bone window setting shows permeative destruction of left parietal bone without osteolysis.

Fig. 3. — A. Microscopy of histopathological specimen. Hematoxylin and eosin stained section shows small cells with round to oval small nuclei and clumped chromatin with no nucleoli consistent with low grade lymphoma (original magnification ×400). B. Post contrast CT image after two cycle of chemotherapy reveals significant reduction in size of lesion.

regimen. Superficial scalp swelling in left fronto-parietal region responded well and CT revealed resolution of intracranial lesion (Fig. 3B). Patient remains symptom free in 2-year follow up period. Discussion A variety of cases of primary CNS lymphomas has been reported in medical literature earlier. But only a few have characteristics similar to the one we have reported. The precise anatomical location and multifocal nature of the neoplasm occurring in nonimmunocompromised adolescent boy makes this case clinically and radiologically unique. Bony involvement is commonly seen with secondary NHL. However, primary bone NHL is extremely rare with reported incidence of only 34% (1). By definition primary bone lymphoma is a solitary mass lesion with no evidence of disease at other sites and no systemic dissemination

within 6 months of the detection of tumor (3). Clinical presentation of NHL primarily involving the skull vault is unusual. However in recent times incidence of CNS lymphoma has increased in both immunocompromised and immunocompetent individuals. Lymphomas of the cranial vault usually involve the pericranium, meninges, and subcutaneous tissues. Histologically all reported lymphomas of the cranial vault are of B-cell origin. Most of cases are reported in middle aged and elderly women. Its occurrence in an immunocompetent adolescent boy as reported by us is uncommon. Computed tomography (CT) features of primary bone NHL are non specific. Characteristic features are a permeative growth pattern with very little cortical destruction (4). Periosteal reaction is occasionally seen. Although lymphomas usually reveal homogenous appearance, most reported cases of the cranial vault lymphoma have demonstrated

297

heterogeneous hyperdensity (5). This finding is substantiated in our case and it may represent a characteristic feature of lymphomas of cranial vault. An associated large soft-tissue component is again very common with extension of mass in the extradural compartment and galeal compartment. Post contrast studies show mild contrast enhancement. Calcification is usually absent (5). Magnetic resonance (MR) imaging provides a better anatomic delineation in the evaluation of the soft tissue component and extradural extension of the lesion however it is also non specific. In our case MR imaging was not done because of financial constraint which is a commonly encountered problem in developing countries like us. Considering the age of patient with CT finding of soft tissue mass and permeative pattern of bone destruction, Ewing’s sarcoma is a likely differential diagnosis which is also one of the small round cell tumors. However it commonly involves long bone with femur being the most common site. Flat bones are commonly involved but pelvis and ribs are much more common sites than skull which is rarely involved. Patient usually presents with localized pain and swelling with systemic symptoms that simulate infection. Primary malignant lymphoma of the skull vault may also mimic metastases or less frequently osteomyelitis, solitary fibrous tumor and granulocytic sarcoma (1, 6, 7). A metastatic carcinoma to the skull may spread to extradural compartment and brain parenchyma, however it usually demonstrates massive osteolysis. Patient with chronic osteomyelitis also demonstrate lytic bone lesions but systemic signs of infection are usually evident before scalp swelling is evident (4). Granulocytic sarcomas (chloromas) are rare and seen most commonly in children with acute myelogenous lymphoma. They have a predilection for the epidural space and skull and revels osteolysis (7). Solitary fibrous tumor is an uncommon spindle cell tumor that usually occurs in the pleura. Extra pleural sites including meninges are rarely involved. Imaging features are nonspecific. Suggestive radiological features along with Clinical characteristic such as the patient’s age and the tumor site are important in diagnosing these lesions. Nevertheless definitive diagnosis of such rare lesion requires biopsy with immuno-

298

histochemistry. Treatment for extranodal Hodgkin lymphoma usually includes adriamycin containing regimen like CHOP protocol, followed by radiation therapy. Alternatively chemotherapy without subsequent radiation therapy and radiation therapy alone can be used in carefully selected cases (8). Clinical stage is the most important prognostic variable in predicting overall survival. Conclusion Although it is extremely uncommon, Primary malignant NHL of the skull vault should be considered in differential diagnosis of scalp masses. This case illustrates that this differential diagnosis should be considered not only with discrete and focal skull lesions in typical setting of imunocompromised middle age individual, but also with diffuse and multifocal vault infiltration in unfavorable clinical scenario like nonimmunocompromised individual and young age. Differentiation is important because their natural history and treatment outcome are

JBRâ&#x20AC;&#x201C;BTR, 2010, 93 (6)

much different from other more common diagnoses in the skull. A thorough search for systemic disease is an essential part of the management. Characteristic CT features of primary malignant NHL of cranial vault are a permeative growth pattern, a large soft tissue component, and very little cortical destruction. Biopsy of the superficial scalp mass was sufficient to obtain a diagnosis, avoiding the risks associated with open intracranial biopsy. Acknowledgement The authors acknowledge Dr Nazma Haider, M.D. [Department of Pathology, Jawaharlal Nehru Medical College, AMU, Aligarh] for her role as pathologist in this case. References 1.

Kantarci M., Erdem T., Alper F., Gundogdu C., Okur A., Aktas A.: Imaging characteristics of diffuse primary cutaneous B-cell lymphoma of the cranial vault with orbital and brain invasion. Am J Neuroradiol, 2003, 24: 1324-1326.

2. Gaitonde S., Patel R., AlagiozianAngelova V., Kadkol S., Peace D.: Primary low grade follicular lymphoma of cranial vault mimicking lipoma at presentation. Acta Oncol, 2008, 47: 326-329. 3. Shoji H., Miller T.R.: Primary reticulum cell sarcoma of bone. Significance of clinical features upon the prognosis. Cancer, 1971, 28: 1234-1244. 4. Holtas S., Monajati A., Utz R.: Computed tomography of malignant lymphoma involving the skull. J Comput Assist Tomogr, 1985, 9: 725727. 5. Galarza M., Gazzeri R., Elfeky H.A., Johnson R.R. 2nd.: Primary diffuse large B-cell lymphoma of the dura mater and cranial vault. Case report and literature review. Neurosurg Focus, 2006, 21: E10. 6. Kim H.J., Lee H.K., Seo J.J., Kim H.J., Shin J.H., Jeong A.K., Lee J.H., Cho K.J.: MR imaging of solitary fibrous tumors in the head and neck. Korean J Radiol, 2005, 6: 136-42. 7. Porto L., Kieslich M., Schwabe D., Zanella F.E., Lanfermann H.: Granulocytic sarcoma in children. Neuroradiology, 2004, 46: 374-347. 8. Abrey L.E., Yahalom J., DeAngelis L.M.: Treatment for primary CNS lymphoma: the next step. J Clin Oncol, 2000, 18: 3144-3150.

JBR–BTR, 2010, 93: 299-301.

BREAST LIPOSARCOMA L. Fellah1, M. Debehogne1, J. El Mouedden1, N. Akle1, M. Berlière2, C. Galant3, I. Leconte Liposarcoma of the breast constitutes 1% of all malignant breast tumors. We report the case of a 42-year-old woman presented with a 5 cm-mass in the left breast. Mammography, ultrasonography and MRI showed a non-specific appearance and the histological evaluation was necessary for definitive diagnosis. Key-words: Breast neoplasms, Diagnosis

Observation Une patiente de 42 ans est adressée dans le service de mammographie pour mise au point d’une masse du sein gauche apparue depuis 5 mois. Elle n’a pas d’antécédent personnel ni familial pour le cancer du sein. A l’examen clinique, on palpe une masse dure de 5 cm supéro-médiane gauche, mobile par rapport aux plans cutané et profond, sans anomalie cutanée, les aires ganglionnaires sont libres. Les clichés mammographiques montrent des seins de densité ACR 2 avec la présence d’une masse localisée dans le quadrant supéroexterne gauche, profonde, relativement bien circonscrite (Fig 1-2). L’échographie mammaire (Fig 3) retrouve une masse bien circonscrite mesurant 40 X 30 X 30 mm, hypoéchogène, hétérogène avec un renforcement postérieur, vascularisée au Doppler avec un index de résistance de 0,92. L’IRM mammaire (Fig 4) met en évidence une masse hétérogène en hypersignal T2 (A), en hyposignal T1 (B) et qui rehausse après injection de gadolinium (C). Une microbiopsie sous échographie a été réalisée à l’aide d’un pistolet Bard avec une aiguille de calibre 14 G. L’examen histologique montre des plages caractérisées par des cellules atypiques dont le cytoplasme comporte plusieurs vacuoles encochant les noyaux, le diagnostique de liposarcome myxoïde mammaire est posé (Fig 5). Discussion Les sarcomes mammaires sont rares, ils constituent moins de 1% des cancers du sein et la lésion la plus fréquente est le sarcome phyllode. Les tumeurs phyllodes sont

Fig. 1. — Cranio-caudal mammograms showing a relatively well-circumscribed mass in the left upper quadrant.

constituées d’éléments épithéliaux et de tissu conjonctif et 20 à 50% d’entre elles sont malignes. Le liposarcome mammaire est une lésion très rare et peu de cas ont été rapportés dans la littérature. Il représente 3 à 24% des sarcomes mammaires et il se développe au sein du tissu stromal interlobulaire (1-3). Il n’existe pas de terrain prédisposant, l’âge moyen des patientes est de 40-60 ans et il n’y aucune corrélation entre le liposarcome mammaire et les autres localisations extra-mammaires (1, 2, 4). La tumeur présente une croissance plus ou moins rapide en fonction du grade histologique. Elle habituellement unilatérale, bien circonscrite (5), mobile avec peu de signe cutané. Elle mesure en moyenne 8 cm (2-30 cm) et la taille n’a pas de valeur pronostique si elle est inférieure à 15 cm. L’atteinte ganglionnaire est rare (2, 5). La dissémination

From: Department of 1. Medical Imaging, 2. Gynecology, 3. Pathology, Cliniques Universitaires Saint-Luc, Brussels, Belgium. Address for correspondence: Dr L. Fellah, Département d’Imagerie Médicale, Avenue Hippocrate, 10, 1200 Bruxelles, Belgium. E-mail : latifa.fellah@rdgn.ucl.ac.be

se fait par voie hématogène avec des métastases viscérales, pulmonaires et osseuses (2, 3). A l’examen histologique, la tumeur est dépourvue de cellules épithéliales et les cellules tumorales étoilées et parfois vacuolisées ne présentent pas l’arrangement en faisceaux des sarcomes phyllodes (1, 5, 6). Les liposarcomes sont classés en 5 types: bien différentiée, à cellules rondes, myxoïde, pléomorphiques et indifférencié. Dans tous les types, des lipoblastes sont toujours présents, uniou multivacuolés avec des noyaux crantés (7, 8). La forme bien différentiée montre des cellules de forme irrégulière, hyperchromatiques légèrement plus grandes que les adipocytes normaux. Le type myxoïde comporte des lipoblastes, des capillaires et une matrice myxoïde. Le type à cellules rondes montre une prolifération uniforme de cellules rondes. Le type pleiomorphe montre une prolifération cellulaire anarchique

300

JBR–BTR, 2010, 93 (6)

Fig. 2. — Oblique views mammograms show mass relatively well circumscribed in the upper quadrant of the left breast.

C Fig. 3. - Ultrasonography showing a hypoechogenic and heterogeneous mass well circumscribed in the upper quadrant of the left breast.

avec un pléomorphisme cellulaire et des cellules géantes (7). Le bilan d’imagerie comprend une mammographie complétée d’une échographie. Dans quelques cas dans la littérature, une IRM a été réalisée (3, 4). Les aspects radiologiques dépendent du grade histologique, les formes bien différentiées étant plus riches en graisse contrairement aux formes mal différentiées plus vascularisées. La mammographie met en évidence une masse arrondie, ovale ou polycyclique et de densité variable en fonction de la composante graisseuse (3, 8). L’échographie retrouve une masse solide, hétérogène avec des contours irréguliers (1).

Fig. 4. – Breast MRI A. Axial T2-weighted image, B. Axial T1weighted image, C. Axial T1-weighted images after gadolinium, showing a heterogeneous enhancing mass.

Fig. 5. - Myxoid lesion with scattered tumor cells showing focally vacuolised cytoplasm (arrow). Hematoxyline and eosin coloration (x 225).

LIPOSARCOME MAMMAIRE — FELLAH et al

301

Table I. – Summary of case reported in the literature Case

Age

Clinical aspect

Pant I. et al.

Mobile 4 mm right upper mass Lymphnode : -

Charfi L. et al.

Left sided mobile mass lymphnode : -

Mazaki T. et al

10x5 cm irregular Well circum- Solid, heteroge- Heterogeneous left sided mass scribed oval neous well mass hypo on Lymphnode : shaped mass delineated vas- T1 and hyper cularized mass on T2

Diekmann F. et al.

Large growing tumor of the left breast

Well delin- Hypoechogenic Heterogeneous Malignant cyseated 25 x fat-containing fat-containing tosarcoma phyl13 mm mass soft mass mass with Gd lodes tumor uptaking with lipomatous metaplasia and diffuse transition into a dedifferentiated liposarcoma

Tunon de Lara C. et al.

15 mm soft, non adhering mass

Well circumscribed opacity

Pai R.R. et al.

Growing 6 cm mass

L’IRM permet de définir le caractère unifocal de la lésion, son extension et éventuellement sa richesse en tissu graisseux et de déterminer le type de rehaussement après injection de gadolinium. Ce rehaussement est variable en fonction du grade histologique, les liposarcomes de haut grade apparaîtront plus vascularisés avec un rehaussement hétérogène. Le diagnostique différentiel se fera avec les sarcomes phyllodes, les sarcomes du stroma, le fibroadénome. Des micro-biopsies sont nécessaires pour l’analyse histologique et pour établir le diagnostic. Le traitement consiste en une exérèse large et complète avec des marges saines (10, Table I). Le taux de survie à 5 ans et le taux de survie sans récidive dépendent des marges (9), le curage axillaire est inutile sauf si des ganglions sont palpés. La chimiothérapie n’a pas fait la preuve de son efficacité et le traitement hormonal n’a pas sa place (5). La radiothérapie est réalisée s’il persiste une tumeur résiduelle ou pour les lésions de haut grade. Le pronostic est mal établi, mal connu compte tenu du petit nombre de cas décrits. Le risque de récidive dépend du grade histologique de la tumeur avec un plus mauvais pronostic pour

X-ray mammo

MRI

Suspicion of fibroadenoma

Pathology Myxoid liposacoma

20 mm well circumscribed mass

Well differentiated liposarcoma

Solid nodule

les lésions mal différentiées et des marges chirurgicales (1-4, 9). La survie est la même chez les patients ayant bénéficiés d’un traitement néo-adjuvant (7, 10). Conclusion Le liposarcome est une tumeur rare et l’aspect peut être celui de tumeur bénigne avec une croissance rapide (2). La mise au point doit se faire par la mammographie, l’échographie complétée de microbiopsies afin d’établir le diagnostic. L’IRM permet de préciser l’extension. Le traitement est chirurgical radical ou conservateur. Le pronostic est mal connu du fait de la rareté du diagnostique.

Grade 1 well-differentiated liposarcoma Myxoid liposarcoma

Références Mazaki T., Tanak T., Suenaga Y., Tomioka K., Takayama T.: Liposarcoma of the breast : a case report and review of the literature. Int Surg. 2002; 87: 164-170. 2. Tunon de Lara C., Rousillon E., Rivel J., Maugey-Laulom B., Alfonso AL., Horovitz J.: Liposarcome du sein à propos d’un cas. J Gynecol Obstet Biol Reprod (Paris). 1998; 27: 201-204. 3. Diekmann F., Rudolph B., Winzer KJ., Bick U.: Liposarcoma of the breast

Myxoid liposarcoma

10.

arising within a phyllodes tumor. J Comput Assist Tomogr. 1999; 23: 764-746. Austin RM., Dupree WB.: Liposarcoma of the breast: a clinicopathologic study of 20 cases. Hum Pathol. 1986; 17: 906-913. Gallagher KE., Wu HH.: Pathologic quiz case: unilateral breast mass in 75-year-old woman. Arch Pathol Lab Med. 2001; 125: 1503-1504. Pai RR., Singh K., Kumar S., Kini H., Sahu K., Upadhyay K.: Myxoid liposarcoma of the breast-report of a rare case. Indian J Pathol Microbiol. 2006; 49: 387-389 Pant I., Kaur G., Joshi SC., Khalid IA.: Myxoid liposarcoma of the breast in a 25-year-old female as a diagnostic pitfall in fine needle aspiration cytology: report of a rare case. Diagn. Cytopathol. 2008;36:674-677 Charfi L., Driss M., Mrad K., Abbes I., Dhouib R., Sassi S., Ben Romdhane K.: Primary well differentiated liposarcoma: an unusual tumor in the breast. Breast J. 2009; 15: 206-207. Blanchard DK., Reynolds CA., Grant CS., Donohue JH.: Primary nonphylloides breast sarcomas. Am J Surg. 2003; 186: 359-361 McGowan TS., Cummings BJ., O’Sullivan B., Catton CN., Miller N., Panzarella T.: An analysis of 78 breast sarcoma patients without distant meta stases at presentation. Int J Radiat Oncol Biol Phys. 2000, 15; 46: 383-390

JBR–BTR, 2010, 93: 302-304.

AN ELDERLY WOMAN WITH OBSTRUCTED OBTURATOR HERNIA: A LESS COMMON VARIETY OF EXTERNAL ABDOMINAL HERNIA L. De Clercq1, K. Coenegrachts1, T. Feryn2, A. Van Couter3, P. Vandevoorde1, K. Verstraete4, H. Rigauts1 Obturator hernia is a type of external abdominal hernia. Site of herniation is the obturator canal, which is the superolateral aspect of the obturator foramen. The herniation sac exits the pelvis near the obturator nerve and vessels. Herniated loops of small bowel extend between the pectineus and obturator muscles. In this report, we present a case of obstructed obturator hernia. Key-word: Intestines, stenosis or obstruction

External abdominal hernias are usually found in the inguinal region, where most are inguinal or femoral hernias. Traditionally diagnosed clinically, hernias may be difficult to identify and even more difficult to classify. Ultrasonography is used to evaluate hernias in the inguinal region (1). The obturator canal however is hardly visible by sonography. Most patients with obstructed obturator hernia are thin elderly women presenting with acute or recurrent small bowel obstruction. Dilated fluid-filled loops of small bowel can be found on abdominal radiograph (2). CT images confirm the presence of obstructed small bowel, moreover they determine the site of obstruction and enable to identify the underlying cause of obstruction. Mortality rate in obturator hernia is highest among all abdominal wall hernias: 13-40% (3, 4). Rapid evaluation and early surgical intervention can reduce morbidity and mortality (3). Case report A 91-year-old woman was admitted to the geriatric ward with new onset pain in the right leg, radiating upwards to the right thigh, right groin and right fossa. She recently suffered from chronic obstipation. She had a medical history of deep venous thrombosis 4 years previously and appendectomy long before admission. On physical examination, abdominal tenderness was present but no rebound. Bowel sounds were normal on auscultation. Blood samples showed hyperglycemia. Ultrasonographic examination (Aplio, Toshiba, Japan) of the abdomen was non-specific. Doppler

B Fig.1. — A 91-year-old woman with obturator hernia. Axial CT (A) showing obturator hernia with small bowel protruded through the obturator canal (white arrow) between pectineus (white star) and obturator muscles (arrowhead). B shows obturator hernia with small bowel (white arrow) protruded between pectineus (white star) and obturator externus (arrowhead) muscle.

ultrasound of the legs excluded recurrent deep venous thrombosis. Abdominal X-ray was normal at the first day of admission. One day later pelvic X-ray revealed several pathological dilated small bowel loops up to 45 mm in the mesogastric region. No pathological air-fluid levels were visible. Consequently Computed Tomography (CT) (64-slice

From: Departments of 1. Radiology, 2. Surgery, 3. Internal Medicine, AZ St-Jan BruggeOostende, 4. Department of Radiology, Ghent University Hospital, Ghent, Belgium. Address for correspondence: Dr K. Coenegrachts, MD, PhD, Department of Radiology, AZ St-Jan Brugge-Oostende AV, Ruddershove 10, B-8000 Brugge, Belgium. E-mail: Kenneth.coenegrachts@azsintjan.be

CT HD 750 – General Electric Medical Systems, Milwaukee, USA) of the abdomen and pelvis was performed to investigate the cause of suspected small bowel obstruction. Acquired images were reconstructed in axial and coronal planes. On CT images dilated fluid-filled loops of small bowel were found up to a herniated loop of small bowel in the pelvic region, through the obturator canal. Small bowel was seen superficial to the obturator muscles and deep to the pectineus muscle. There were no signs of bowel strangulation but images suggested an incarcerated hernia. Diagnosis of an obstructed

RARE EXTERNAL ABDOMINAL HERNIA – OBTURATOR HERNIA — DE CLERCQ et al

303

Fig.2. — Reconstructed coronal CT image (A) shows multiple dilated small bowel loops and level of herniated small bowel through the obturator canal (white arrow). B shows multiple dilated small bowel loops and herniated small bowel (white arrow) inferior to the right ramus pubis inferior (arrowhead). C. shows multiple dilated small bowel loops and herniated small bowel (white arrow) inferior to the right ramus pubis inferior.

obturator hernia was made. Patient was taken into surgery room for laparoscopic repair.

obstruction with high risk of strangulation (5). Imaging findings

Discussion Definition Obturator hernia is a rare but lifethreatening cause of obstructed abdominal wall hernia. Rarity and non-specific signs contribute to late diagnosis and late operation (1). It accounts for up to 1.4% of all abdominal wall hernias and up to 1.6% of small bowel obstructions (3, 5). Obturator hernia is also called “little old ladies hernia”. It is more common in elderly females owing to the greater width of the pelvis, larger obturator canal after pregnancy and age-related increased tissue laxity. In thin ladies loss of protective fat in the obturator canal facilitates the formation of hernia (3, 5). There is no external lump on abdominal examination but a tender mass on rectal or vaginal examination may be found. Obturator nerve irritation by compression of the herniated sac causes pain in the medial aspect of the thigh, the Howship-Romberg sign (HRS), which is positive in 20-50% of cases (3, 5). Absent adductor reflex in the thigh, The HanningtonKiff sign, is more specific but less known (3). Obturator hernias affect the right side more frequently. Small bowel protruding through the obturator canal, which is 2-3 cm long and 1 cm wide, or lying between the pectineus and obturator muscles often incarcerates (1). This leads to acute or intermittent small bowel

Ultrasonography is used as a first step method to evaluate hernias in the inguinal region (2). Conventional radiography may detect small bowel obstruction and sometimes a fixed loop containing gas in the obturator region is seen (1). Ultrasonographic findings indicating hernial incarceration include free fluid in the hernial sac, bowel wall thickening in the hernia, fluid in the herniated bowel loop, and dilated bowel loops in the abdomen (6). Free fluid is an eye-catching finding on sonography. This single sign easily indicates that a hernial complication has a high probability of being present. Also, free fluid within the herniated bowel loops indicates incarceration of a hernia containing bowel with high specificity but limited sensitivity. The reason for the presence of fluid in an incarcerated bowel loop may be exudation into the bowel lumen, causing excess fluid in the bowel in cases of bowel obstruction (6). An indirect sign of an incarcerated hernia is the evidence of dilated, fluid-filled bowel loops in the abdomen. This sign indicates incarceration of the hernias containing bowel with excellent specificity but limited sensitivity. The absence of blood flow in the contents of a hernia should not be taken as a sign of incarceration because many incarcerated hernias have detectable blood flow on color Doppler sonography (6). Also, the absence of peri-

stalsis should not be considered a sign of incarceration because nonincarcerated hernias do not show peristalsis during the sonographic investigation in a relatively high percentage of patients. However, if peristalsis is present in an incarcerated hernia on sonography, bowel resection at surgery is probably not necessary. Gas in the bowel wall or free gas, either in the abdomen or the hernia sac, also is considered a sign of a complicated hernia (6). The best imaging tool in obturator hernia however is CT of the pelvis and upper thigh. CT evidence of bowel herniating through the obturator foramen and lying between the pectineus and obturator muscles is the best diagnostic clue (1). CT accurately identifies the anatomical definition of the hernia and its content and also differentiates it from other abdominal masses eg tumors, hematomas, abscesses. Postural manoeuvres or increasing intraabdominal wall pressure can help depict subtle hernias. Thin reformatted images of 2.5 mm or less may better delineate the size and shape of the hernia sac and associated complications (4). Intravenous administration of contrast material is necessary for characterization of the vascular supply of the bowel wall to detect complications and guide treatment planning. Dilation and/or fecalization of small bowel proximal to the hernia and tapering of the limps at the hernia defect are signs of obstructed hernia. Helical CT signs in the diagnosis of intestinal ischemia have been described before: in a study by Zalcman M et

304

al, reduced enhancement of the bowel wall had a sensitivity of 48% and specificity of 100%, mural thickening had a sensitivity of 38% and specificity of 78%, mesenteric fluid had a sensitivity of 88% and specificity of 90%, congestion of mesenteric veins had a sensitivity of 58% and specificity of 79%, and ascites had a sensitivity of 75% and specificity of 76% (7). Differential diagnosis Differential diagnosis of obturator hernia should be made with other hernias in the groin region: inguinal, femoral, perineal and sciatic hernias. Inguinal and femoral hernias are the most common type of abdominal wall hernia (4, 8). Indirect inguinal hernias originate lateral to the inferior epigastric artery through the deep inguinal ring, traversing the inguinal canal inferomedially, along the spermatic cord or round ligament. They may exit the superficial inguinal ring to enter the scrotum or labium majus. Direct inguinal hernias originate medial to the inferior epigastric artery and protrude anteriorly through Hesselbach’s triangle. Femoral hernias, like obturator hernias, originate inferior to the inguinal ligament but protrude medial to the femoral vein through the femoral ring (2, 8). Perineal hernias protrude anterior through the urogenital diaphragm and posterior between the levator ani and coccygeus muscles (1). Sciatic hernias protrude through the greater sciatic foramen and extend laterally into the subgluteal region.

JBR–BTR, 2010, 93 (6)

Treatment There is often a delay in diagnosis and surgical intervention of obturator hernia because of a hesitation to operate early due to vague symptoms, advanced age of patients and presence of comorbidities. Given the high risk of bowel incarceration and perforation, obturator hernias are always treated surgically (5). The majority of cases require resection of strangulated small bowel. The classical approach is abdominal exploratory laparotomy, although an inguinal approach is possible. In elective early surgery laparoscopic approach can be used (5). Repair is effected by closure of the peritoneum with stitches or by placement of a mesh. Contralateral side exploration is recommended.

References 1.

Conclusion Obturator hernia is a rare but lifethreatening type of external abdominal hernia. Signs and symptoms are often non-specific. Patients are usually thin elderly women presenting with intermittent or acute small bowel obstruction, which can be seen on abdominal radiographs. Small bowel herniating through the obturator foramen and lying between the pectineus and obturator muscles is a diagnostic clue on CT of the abdomen and pelvis. High risk of bowel incarceration requires rapid diagnosis and early surgical repair to reduce morbidity and mortality of patients with obstructed obturator hernia (3, 4).

Jamadar D., Jacobson J., Morag Y., Girish G., Ebrahim F., Gest T., Franz M. Sonography of Inguinal Region Hernias. AJR, 2006,187: 185-190. Clair L. Shadbolt, Stefan B. J. Heinze, Rosalind B. Dietrich. Imaging of Groin Masses: Inguinal Anatomy and Pathologic Conditions Revisited. RadioGraphics, 2001, 21 :S261-S271. Mantoo S.K., Mak K., Tan T.J. Obturator hernia: diagnosis and treatment in the modern era. Singapore Med J, 2009, 50: 866-870. Aguirre D., Santosa A., Casola G., Sirlin C. Abdominal Wall Hernias: Imaging Features, Complications, and Diagnostic Pitfalls at MultiDetector Row CT. RadioGraphics, 2005, 25: 1501-1520. Rodríguez-Hermosa J.I., CodinaCazador A., Maroto-Genover A., PuigAlcántara J., Sirvent-Calvera J.M., Garsot-Savall E., Roig-García J. Obturator hernia: clinical analysis of 16 cases and algorithm for its diagnosis and treatment. Hernia, 2008, 12: 289-297. Rettenbacher T., Hollerweger A., Macheiner P., Gritzmann N., Gotwald T., Frass R., Schneider B. Abdominal Wall Hernias: CrossSectional Imaging Signs of Incarceration Determined with Sonography. AJR, 2001, 177: 1061-1066. Zalcman M., Sy M., Donckier V., Closset J., Van Gansbeke D.. Helical CT Signs in the Diagnosis of Intestinal Ischemia in Small-Bowel Obstruction. AJR, 2000, 175: 1601-1607. Upponi S., Bungay H. Imaging of abdominal wall hernias. Imaging, 2006, 18: 268-277.

JBR–BTR, 2010, 93: 305-307.

THE ASSOCIATION OF TETHERED CORD, SYRINGOMYELIA, DIASTOMETAMYELIA, SPINAL EPIDERMOID, SPINAL LIPOMA AND DERMAL SINUS TRACT IN A CHILD S. Avcu, M. Necat Köseog˘lu, M. Deniz Bulut, Ö. Özen, Ö. Ünal1 We report on a 5-year-old boy presenting with tethered cord, diastometamyelia, spinal dysraphism, terminal lipoma, spinal epidermoid, and dermal sinus tract with CT, conventional MRI, and diffusion-weighted MRI findings. To the best of our knowledge, our case has the property to be the first case in the literature showing the association of these pathologies all together. Keyword: Spinal cord, developmental defect.

Tethered cord is a clinical syndrome which frequently is associated with a short and thick filum terminale. Orthopedic deformities are related to the lower settlement of the conus medullaris. While myelomeningocele, lipomyelomeningocele, and diastometamyelia may cause tense or short filum terminale, dermal sinus tract, tumor, hemangioma or scar tissue may cause tethered cord syndrome (1-3). In this case, diastometamyelia, spinal dysraphism, terminal lipoma, spinal epidermoid cyst and dermal sinus tract were all observed concomitant with tethered cord and we aimed to present the computerized tomography (CT), conventional magnetic resonance imaging (MRI) and diffusionweighted MRI findings. Case presentation A 5-year-old boy was referred to our hospital with the complaints of complete abasia and incontinence. On physical examination, motor strength in the upper extremity was complete, deep tendon reflex in the lower extremity was decreased, and he could not walk without any buttress. The patient had a sinus tract at sacral level. The anamnesis taken from the parents revealed that the patient was the ninth child of the 31year-old mother. His two siblings died just after they were born. No consanguineous marriage was reported in the family. A spinal MRI and a lumbosacral CT examination were carried out. The patient was examined with 1.5 Tesla (Siemens symphony, Erlangen, Germany) MRI T1-weighted turbo spin echo (TSE) sagittal, T2-weighted TSE sagittal-

axial, T2-weighted coronal images and sagittal and axial diffusionweighted images were obtained. Spinal MRI examination showed spinal cord ending at S1 level. At the level of L5-S1 vertebrae a 22 ⫻ 15 mm intracanalar cystic lesion showing restricted diffusion consistent with an epidermoid cyst was observed (Fig. 1A-C). On sagittal images syringohydromyelia all along the spinal cord (Fig.1A, B), and on coronal and axial images diastometamyelia (Fig. 1D, E) at lower thoracic and lumbar levels were demonstrated. Besides, at the level of S1-2 vertebrae, a 15mm sized intradural lipoma and an associated dorsal dermal sinus tract were visualised (Fig. 1A, B, F, G). CT examination demonstrated a spina bifida at the posterior elements of S1 vertebra (Fig. 1H). Discussion In 1953, Garceau first asserted that the short filum terminale causes spinal cord traction. In 1976, Hoffman et al. used the term tethered spinal cord (1-3). In 1981, Yamada et al defined the term tethered cord syndrome (4). In the 8th week of the gestation, spinal cord reaches the bottom length of the spinal channel. During the remaining time of the gestation, bone components develop faster than the spinal cord. Therefore the cord reaches T12-L2 level which is the normal position of conus medullaris (5). If an obstruction or a mechanical pull occurs due to any intradermal anomaly (e.g. terminal lipoma, diastematomyelia, dermal sinus, etc.), this rise is not completed (4).

From: 1. Department of Radiology, Faculty of Medicine, Yüzüncü Yıl University, Van, Turkey. Address for correspondence: Dr S. Avcu,.M.D., Yuzuncu Yil Universitesi Radyoloji Anabilim Dalı, Kazım Karabekir Cad., 65200, Van, Turkey E-mail: serhatavcu@hotmail.com

Surgical intervention is advocated even for the asymptomatic cases. Although the time and choice for the treatment of tethered cord remain controversial, treatments applied before 6 months reduce the probable deficits to a minimal level compared to treatments applied after 4 years. The therapy to be preferred is early surgical procedure. Aim of the therapy is to release the spinal cord by pulling it apart from its abnormal links. If no intervention is applied, irrecoverable neurological deficits will develop in 90% of the patients (5-8). Female/male ratio in tethered cord syndrome is 2/1. While it is observed more commonly in females, no data is found in the literature for the frequency of this rare syndrome. The most frequently encountered symptom is the motor deficit in the lower extremities and it is seen in 75% of the patients. Lower extremity pain is seen in 40% of the pediatric patients and urological problems are seen in 35% of all age groups (5, 9-12). Our patient had the complaint of abasia. Neurological deficits become more apparent as the child grows up especially between the ages of 5 and 15 (13). The most frequently reported causes of tethered cord are split spinal cord syndrome of 17.0%38.2% and lipomyelomeningocele with 19.1%-94.1%, short and thick phylum terminale with 5.8%-55.8% and the adherences which appear subsequent to the myelomeningocele operation with 5.4%-17.9%. A rare cause is the spinal epidermoid cysts. A few causes are asserted for the formation of spinal epidermoid cysts. Lumbar puncture, trauma and congenital causes are determined in the etiology (13-17). Epidermoid cysts result from the association of anarchic movement of ectodermal cells during the early fetal period and defect in the closure of the neural tube. In our literature review, we

306

JBRâ&#x20AC;&#x201C;BTR, 2010, 93 (6)

F came across a few spinal epidermoid cyst cases which accompany the tethered cord (17). Our case associates epidermoid cyst, diastometamyelia, syringohydromyelia, dermal sinus tract as well as terminal lipoma along with tethered cord. Referring to the literature, among

30% of the patients who had spinal lipoma and tethered cord, at least one of the supplementary pathologies (dermoid tumor, hydromyelia, thick phylum, and diastometamyelia) was observed. Since the epidermoid cysts associated with tethered cord are fre-

Fig. 1. â&#x20AC;&#x201D; On the sagittal T1 (A) weighted image, intradural lipoma (arrow) is observed in the sacral region. On the sagittal T1 (A) and T2 (B) weighted images, all along the spinal cord, syringohydromyelia (arrow), lower settled cord and thick phylum are prominent. On the sagittal diffusion-weighted image (C), an epidermoid cyst is observed showing diffusion restriction at S1 level (arrow). On the coronal (D) and axial (E) T2-weighted images, diastometamyelia is observed (arrow). On the sagittal (F) and axial (G) T2 weighted images, dorsal dermal sinus tract at the level of S2-3 and sacral spina bifida are observed (arrows). On the axial CT section (H), sacral spina bifida is observed (arrow).

quently isointense on conventional MRI, they can be missed easily. In these cases, additional diffusion-MRI should be done as epidermoid cysts characteristically exhibit restricted diffusion (17). In our case, the epidermoid cyst which cannot be clearly distinguished from the CSF on con-

TETHERED CORD SYNDROME — AVCU et al

ventional MRI sequences flared up on diffusion-weighted images and the fact the epidermoid cyst was detected by means of diffusion-MRI suggested that the spinal diffusionMRI would be beneficial as a routine examination method in the tethered cord cases. References 1. Yücesoy K., Özdemir N., Özer E., Mertol T., Arda M.N. Tethered cord syndrome in adulthood. J Neurol Sci Turk, 2001, 18: 1. 2. Barkovich A.J. Pediatric Neuroimaging. 3rd ed. Philadelphia, Lippincott Williams Wilkins, 2000, 621-681. 3. Hoffman J.H., Hendrick E.B., Humphreys R.P. The tethered spinal cord: Its protein manifestations, diagnosis and surgical correction. Child's Brain, 1976, 2: 145-155. 4. Yamada S., Zinke D.E., Sanders D. Pathophysiology of tethered cord syndrome. J Neurosurg, 1981, 54: 494-503.

5. Sherbum E.W., Park T.S. Occult spinal dysraphism. Contemp Neurosurg, 1997, 19: 1-9. 6. Yamada S., Iacona R.P. Tethered cord syndrome. In: Disorders of the Pediatric Spine, Pang D (editor). Raven Press, 1995, p. 159-173. 7. Yamada S., lacono R.P., Andrade T., Mandybur G., Yamada B.S. Pathophysiology of tethered cord syndrome. Neurosurg Clin N Am, 1995, 6: 311-323. 8. Mapstone T.B. Management of tethered spinal cord. Neurosurg Quart, 1994, 4: 82-91. 9. Hayward R. Tethered cord syndrome. Br J Neurosurg, 1998, 12: 75-76. 10. Chopra S., Gulati M.S., Paul S.B., Hatimota P., Jain R., Sawhney S. MR spectrum in spinal dysraphism. Eur Radiol, 2001, 11: 497-505. 11. Moore K.L. The nervous system. In: The Developing Human, Clinically Oriented Embryology, 6th ed, Persaud TVN (editor), WB Saunders Company, 1998, p. 452-465. 12. Raghavan N., Barkovich A.J., Edwards M., Norman D. MR imaging

307

13. 14.

15.

16.

17.

in the tethered spinal cord syndrome. AJR, 1989, 152: 843-852. Rauzzinno M.J., Iskandar B., Oakes W.J. Occult spinal dysraphism. Contemp Neurosurg, 1998, 20: 16. Juhl J.H., Crummy A.B. Essentials of radiologic imaging. 6th ed. LB. Lippincott Company, Philadelphia, 1993, p. 469-496. Roux A., Mercier C., Larbrisseau A., Dube L.J., Dupuis C., Del Carpio R. Intramedullary epidermoid cysts of the spinal cord. Case Report. J Neurosurg, 1992, 76: 528-533. Visdani A., Savoiardo M., Balestrini M.R., Solero L. Iatrogenic intraspinal epidermoid tumor: MyeloCT and MRI diagnosis. Neuroradiology, 1989, 31: 273-275. Severino M., Manara R., Faggin R., Dalle Nogare C., Gamba P., Midrio P. Anorectal malformation and spinal dysraphism: the value of diffusionweighted imaging in detecting associated intradural (epi)dermoid cyst. J Ped Surg, 2008, 43: 1935-1938.

JBR–BTR, 2010, 93: 308-309.

ECTOPIC INFRAORBITAL NERVE IN A MAXILLARY SINUS SEPTUM: ANOTHER POTENTIALLY DANGEROUS VARIANT FOR SINUS SURGERY P. Mailleux1, O. Desgain2, M.I. Ingabire1 We report on two patients with ectopic infraorbital nerve and canal located in a maxillary sinus septum. This very rare anatomic variation may possibly generate complications during sinus surgery if it passes unnoticed during the preoperative CT work-up. Key-word: Paranasal sinuses, Anatomy

Various paranasal sinuses anatomic variants have been described in the literature. Detailed knowledge of these variations is critical for surgeons performing functional endoscopic sinus surgery (FESS) and for the radiologist performing the preoperative work-up, in order to increase management outcome and to avoid potential complications. We describe two patients with a very rare undescribed yet anatomic variation that could potentially be dangerous during sinus – mainly FESS- surgery which consists in a bony septum crossing the upper part of the maxillary sinus and the infraorbital nerve being located within this septum. Case reports The first patient is a 61-year-old lady (Fig. 1,2) complaining of non specific symptoms of nasal obstruction. A CT scan of the sinus was performed, with no signs of chronic sinusitis but it showed in the right maxillary sinus a septum crossing its anterior-superior portion (Fig 1B,C, Fig 2). The infraobital nerve is located in this septum . The second patient is a 72-yearold male admitted in our emergency department with suspicion of cerebral infarction. The sinus findings are incidental but showed bilaterally a septum in the anterior and superior part of the maxillary sinus with an ectopic infraorbital nerve located in the septum (Fig.3). Discussion FESS is one of the most commonly performed procedures by otolaryngologists. However, the procedure has many potential com-

Fig.1. — Axial CT scan slices, from top to bottom. A. Upper part of the maxillary sinus.B,C, slightly below A: the septum (straight arrow ) starts from the lateral sinus wall. Within it the infraorbital nerve (curved arrow) .D: lower portion of the sinus , without septum. CT parameters in both cases were the following: 64 slices MDCT , 0625 mm thin slices, 100 kV, 50 mA resulting in patient 1 in a CTD/vol of 2,74 mGy, DLP 37,5 mGycm and 1,5 mSev

plications. It is essential that the anatomic sinonasal variations be clearly described in all sinonasal CT imaging. The preoperative work-out before FESS has three goals: to describe the signs of sinus inflammation and possible causes, such as odontogenic granulomas; to show the anatomic variations that can disturb

From: 1. Department of Medical Imaging, 2. Department of Otorhinolaryngology, Clinique St Luc, Bouge, Belgium Address for correspondence: Dr P. Mailleux, MD, Department of Medical Imaging, Clinique St Luc, rue St Luc 8 , B- 5004 Bouge, Belgium e-mail: p.mailleux@skynet.be

sinonasal air circulation; and finally to depict the “ dangerous “ anatomic variations, helping thus the surgery avoiding unnecessary surgical problems. Those variants are numerous, a few important ones being for example defect in the medial orbital wall, medial septum of the sphenoid sinus inserted on the carotid artery canal, abnormal location of the anterior ethmoidal artery in the ethmoid sinus, hypoplastic maxillary sinus, and septa inside the maxillary sinus with an abnormal ostium (1 and 2). Anatomic variation of the infraorbital nerve is usually minimal. Testut

ECTOPIC INFRAORBITAL NERVE IN A MAXILLARY SINUS — MALLEUX et al

Fig. 2. — Coronal and sagittal slices . A, B: coronal slices: the septum (straight arrows) crosses the lateral part of the sinus and within it the orbital nerve (curved arrow). C and D: sagittal slices through the right maxillary sinus.

describes a dehiscence of the bony infraorbital nerve canal containing the nerve, with no interposing bony structure between the nerve and the sinus mucosa (3). One case of right infraorbital nerve canal passing within the inferior lamella of an infraorbital ethmoid (Haller’s) cell was also described (4) , but meandering infraorbital nerve and canal in a maxillary sinus septum do not seem to have been described (5). Septa in the maxillary sinus are present in about 16% of the maxillary septa (6 ) , most often in atrophic maxillary sinuses (27%). They are complete in 0.5% , incomplete on the anterior wall in 5% and incomplete on the sinus floor for the other cases. Detailed information about location, morphology and height of the septa is clinically relevant to help avoiding complications during sinus surgery (7) . In the presented cases, the septum is incomplete with an oblique anterosuperior location (Fig. 2C, D).

309

Fig. 3. — CT scan. A: Axial slice with bilateral upper maxillary sinus septa (arrow) with ectopic nerve ( curved arrow ) .B: Oblique coronal view through the septum and nerve on the right sight ( curved arrow). C: sagittal slice through septum and nerve on the right side. D: same aspect on the left . CT scan parameters were the same as in patient 1 (Fig. 1, 2) with a resulting 1,4 mSev

It contains an ectopic infraorbital nerve and canal which do not follow the inferior orbital wall up to the orbital ridge but dive into the septum. Any surgical attempt to break down this septum would cause infraorbital palsy, with permanent soreness and paresthesias in the area, a complication relatively frequent after facial trauma with fracture of the orbital floor, or after malar augmentation procedures for esthetic purposes. It is thus important for the radiologist to mention it in any sinonasal preoperative report and for the surgeon to be aware of any anatomic variant.

during FESS or other maxillary sinus surgery. Bibliography 1. 2. 3. 4.

Conclusion We describe two patients with ectopic location of the infraorbital nerve in a maxillary sinus septum. This very rare anatomic variation should be described in the report of any sinonasal preoperative CT scan or MRI to avoid sectioning the nerve

Marsot-Dupuch K., Genty E., Les variantes anatomiques des sinus de la face. J Radiol, 2003, 84: 357-367. Earwaker J., Anatomic variants in sinonasal CT. Radiographics, 1993, 13: 381-415. Testut L., in Traité d’anatomie humaine , edited by Octave Doin et Fils, Paris, 1921, p. 1003. Chandra R.K., Kennedy P. Surgical implications of an unusual anomaly of the infraorbital nerve. Ear, Nose and Throat J, 2004, 83: 766-767. Jinkins J.R.in Atlas of neuroradiologic embryology, anatomy and variants edited by Lippincot, Williams & Wilkins, Philadelphia, 2000, p. 486-487. Krennmair G., Ulm C. and Lugmayr H., Maxillary sinus septa: incidence, morphology and clinical implications. J Cranio-maxillofacial Surg, 1997, 25: 261-265. Lambert F., Lecloux G., Rompen E., Sub-sinus bone augmentation: sinus lift concept update. Rev Odont Stomat, 2008, 37: 3-17.

JBRâ&#x20AC;&#x201C;BTR, 2010, 93: 310-311.

ACUTE ABDOMINAL PAIN DUE TO BILIARY CYSTADENOMA K. Van den Bergh, K. Op de beeck1 We present a case of a biliary cystadenoma, a rare benign cystic tumor arising in most cases of the intrahepatic bile ducts. A 30-year old woman presented with abdominal pain in the right upper quadrant for about 10 days with increasing severity. Radiological evaluation by means of abdominal ultrasound and CT-scan revealed a multilocular cystic lesion in the right liver lobe, not present on abdominal ultrasound performed 10 years before. The diagnosis of a biliary cystadenoma was proposed and complete surgical resection of the mass was performed. Key-word: Liver neoplasms, diagnosis.

Biliary cystadenomas are rare, usually slow growing, multilocular cystic tumors that represent less than 5% of intrahepatic cystic masses of biliary origin (1). They are generally intrahepatic (85%) and occur less commonly in the extrahepatic biliary system and gallbladder. Cystadenomas occur predominantly in middle-aged women (2). The tumor is well defined by a fibrous capsule and lined by cuboidal to columnar epithelium. Two histological variants are recognized: the more common mucinous type and the rare serous type (3). The clinical presentation is variable depending on the size and location of the cyst. Abdominal pain, obstructive jaundice, palpable mass, nausea and vomiting are common signs and symptoms. Occasionally, the tumor is incidentally discovered (2).

Fig. 1. â&#x20AC;&#x201D; Abdominal ultrasound reveals a well defined, septated and thin-walled cystic mass. The cyst fluid contains some low-level echoes.

Case report A 30-year-old woman was admitted to the emergency department with complaints of abdominal pain in the epigastric area and the right upper quadrant for 10 days with increasing severity. Physical examination revealed a tender upper abdomen. Her medical history was unremarkable. Laboratory data on admission were within normal limits. Abdominal ultrasound (US) showed a well defined 10-cm, septated and thin-walled cystic (anechoic) mass with enhanced through transmission, involving segment 4 of the liver (Fig. 1). The cyst fluid contained some low-level echoes. The gallbladder was normal. Contrast-enhanced CT-scan confirmed the presence of a multilocular cystic mass in segment

4 of the liver, with mean attenuation values of 10HU (Fig. 2 A). The internal septations and the thin regular wall showed slight enhancement (Fig. 2B). A fluid-fluid level was present within the lower portion of the cystic mass. No mural nodularity was identified. Intrahepatic bile ducts showed no dilatation. There was mass effect of the lesion on surrounding structures. The diagnosis of a biliary cystadenoma was proposed. The differential diagnosis included hydatid disease, complicated hepatic cyst and abscess. Because of the unclear etiology and severe pain an enucleation of the lesion was set up. Finally complete surgical resection of the mass was performed. During surgery it was remarkable that the

From: 1. Department of Radiology, University Hospitals Leuven, Leuven, Belgium. Address for correspondence: Dr K. Van den Bergh, Department of Radiology, KUL University Hospitals Leuven, Herestraat 49, B-3000 Leuven, Belgium. E-mail: katrien.vandenbergh@uzleuven.be

lesion was glued with the omentum and the stomach, due to an inflammatory reaction. Microscopic examination of the specimen confirmed the diagnosis of a biliary cystadenoma, there were no signs of malignancy. Discussion Biliary cystadenomas are benign cystic neoplasms lined by mucinsecreting epithelium. The fluid within the lesion can be proteinaceous, mucinous and occasionally gelatinous, purulent or hemorrhagic due to trauma (1). Biliary cystadenomas are rare, usually slow growing, multilocular cystic tumors that represent less than 5% of intrahepatic cystic masses of biliary origin. Eighty-five percent arise from the intra-hepatic biliary ducts, the remaining 15% are extrahepatic. They are usually solitary and located in the right lobe of the liver and measure from 1.5 cm to 25 cm in diameter. They occur predominantly in middle-aged women

ACUTE ABDOMINAL PAIN â&#x20AC;&#x201D; VAN DEN BERGH et al

311

Fig. 2. â&#x20AC;&#x201D; A. Contrast-enhanced CT-scan confirms the presence of a multilocular cystic mass in segment 4 of the liver, with mean attenuation values of 10HU. B. The internal septations and the thin regular wall show slight enhancement.

(mean age, 38 years). Although biliary cystadenomas are benign tumors, they may recur after excision and have potential to develop into biliary cystadenocarcinoma. Symptoms are usually related to the mass effect of the lesion and consist of intermittent pain or biliary obstruction (1, 2). At US, a biliary cystadenoma appears as a unilocular or multilocular cyst with enhanced through transmission. Acoustic shadowing may be present from septal or wall calcification. The cyst fluid may contain low level echoes from blood products, mucin, or proteinaceous fluid. Serous and bilious fluid is generally anechoic. Echogenic mural nodules and papillary projections may be present (2, 4). CT depicts a low-density, well defined cystic mass with internal septa and a thick fibrous capsule. The internal septa and capsule usually shows contrast enhancement. The CT attenuation of the fluid component varies depending on the fluid content. Higher attenuation may indicate recent hemorraghe. Tumor nodules and papillary excrescences can be present and appear as soft-tissue attenuation nodules that typically enhance with contrast material (2). These tumor nodules and papillary excrescences are seen more commonly in biliary cystadenocarcinoma than in cystadenoma, although they have also been reported in cystadenomas without frank malignancy.

Although the absence of mural nodularity is suggestive of cystadenoma (5). CT may also reveal septal and capsular calcifications (1, 4). Unfortunately there are no specific imaging features that permit reliable differentiation of biliary cystadenoma from cystadenocarcinoma (2). The major advantage of MRI is that it can accurately define the internal architecture of the lesion and specify the fluid content because of the better soft tissue contrast resolution compared with CT. The MR imaging appearance of a cystadenoma varies depending on the protein content of the fluid and the presence of an intracystic soft-tissue component. In general, the (mucinous) fluid within a cystadenoma appears hypointense on T1-weighted images and markedly hyperintense on T2weighted images. On T1-weighted images, the signal intensity may change from hypointense to hyperintense as protein concentration increases. Also, on T2-weighted images the signal intensity may decrease from markedly hyperintense to hypointense as protein concentration and viscosity increases. Similar changes may be provoked by internal hemorraghe, however, fluidfluid levels are usually found in these cases (1, 3). The differential diagnosis principally includes hepatic echinococcal cyst, hepatic abscess and a hemorrhagic bile duct cyst. In rare cases, other neoplasm such as mesenchy-

mal hamartoma, undifferentiated embryonal sarcoma, cystic hepatocellular carcinoma and cystic metastasis may have similar appearance (4). Correlation of imaging findings with patient age and clinical data may be helpful in the differential diagnosis. Because cystadenoma represents the premalignant form of biliary cystadenocarcinoma, resection is always recommended (5). References 1.

MortelĂŠ K.F., Ros P.R.: Cystic focal liver lesions in the adult: differential CT and MR imaging features. Radiographics, 2001, 21: 895-910. Levy A.D., Murakata L.A., Abbott R.M., Rohrmann C.A: Benign tumors and tumorlike lesions of the gallbladder and extrahepatic bile ducts: Radiologic-pathologic correlation.Radiographics, 2002, 22: 387-413. Frahm C., Zimmermann A., Heller M., Brossmann J.: Uncommon presentation of a giant biliary cystadenoma: correlation between MRI and pathologic findings. J Mag Res Imag, 2001, 14: 649-652. Buetow C.P., Buck J.L., PantongragBrown L. et al.: Biliary cystadenoma and cystadenocarcinoma: Clinicalimaging-pathologic correlation with emphasis on the importance of ovarian stroma. Radiology, 1995, 196: 805810. Anderson S. W., Kruskal J.B., Kane R. A.: Benign hepatic tumors and iatrogenic pseudotumors. Radiographics, 2009, 29: 211-229.

JBR–BTR, 2010, 93: 312-313.

PROLAPSED BILATERAL URETEROCELES LEADING TO INTERMITTENT OUTFLOW OBSTRUCTION H. Stunell, S. Barrett, N. Campbell, E. Colhoun, W.C.Torreggiani1 A ureterocele refers to a cystic dilatation of the distal ureter. It may be unilateral or bilateral and may be associated with a duplex system in some cases. We present an unusual case where a young patient was found to have large bilateral ureteroceles which prolapsed into the urethra, causing intermittent incontinence and obstruction. We discuss the case and review the literature concerning this rare anomaly. Key-word: Children, genitourinary system.

A 10-year old boy presented with a six-month history of difficulty passing urine as well as dribbling and incontinence. He had been treated by his family doctor for urinary tract infections however his symptoms had recently deteriorated. He was otherwise well with no prior medical history of note. An ultrasound of the kidneys demonstrated some scarring of the upper pole of the left kidney but was otherwise normal with no evidence of a duplicated system. Pelvic ultrasonography demonstrated bilateral large ureteroceles, which partially covered the bladder outlet (Fig. 1). A micturating cystourethrogram (MCUG) was then performed. This again demonstrated bilateral ureteroceles. It was noted that towards the end of micturition, both ureteroceles prolapsed into the urethra and caused some obstruction to flow leading to incomplete bladder emptying (Fig. 2). An isotope DMSA scan (not shown) demonstrated evidence of left-sided reflux nephropathy with percentage uptake of 32% on this side. Cystoscopy confirmed the presence of large ureteroceles bilaterally, with prolapse into the bladder neck and proximal urethra. Single ureteric orifices were noted bilaterally, which were located intravesically. Endoscopic puncture of bilateral ureteroceles was performed, with frank pus draining from both puncture sites. The patient was commenced on prophylactic I-V antibiotics and discharged home well 4 days post-operatively. At onemonth follow-up, there was complete resolution of symptoms with radiological evidence of collapse of both ureteroceles on ultrasound studies.

Fig. 1. — Ultrasound of the pelvis demonstrates bilateral large fluid filled ureteroceles (arrows) within the bladder, extending to overlie the bladder outlet.

Fig. 2. — MCUG demonstrates both ureteroceles to lie within the proximal urethra (arrows). It was noted during this study that there was partial obstruction to flow towards the end of micturition.

Discussion From: 1. Department of Radiology, Adelaide and Meath Hospital Incorporating the National Children’s Hospital, Dublin, Ireland. Address for correspondence: Dr S. Barrett, MB BCH, Department of Radiology, Adelaide and Meath Hospital Incorporating the National Children’s Hospital, Tallaght, Dublin 24, Ireland. E-mail: sarah.barrett@amnch.ie

A ureterocele refers to a cystic dilatation of the distal ureter protruding into the bladder lumen. This malformation is most commonly diagnosed in the paediatric popula-

PROLAPSED BILATERAL URETEROCELES â&#x20AC;&#x201D; STUNELL et al

tion for two reasons: firstly, due to prenatal ultrasound screening for congenital anomalies and secondly, due to the high incidence of postnatal urinary tract infections, resulting in ultrasonography of the renal tract as a first-line investigation. The incidence is reported to be between 1 in 5000 to 1 in 12000 of paediatric hospital admissions (1), however a number of autopsy studies have suggested a much higher incidence of up to 1 in 500 (2). It occurs 46 times more frequently in girls and is observed almost exclusively in the Caucasian population (3). Ureteroceles have no predilection for side and 10% of cases show bilateralism (4). A number of classification systems exist based on varied anatomical and pathological criteria. However, the most widely accepted and most frequently used system has been established by The American Academy of Paediatrics (5), which classifies ureteroceles contained entirely within the bladder (i.e. ureteric orifice and ureterocele itself) as intravesical or orthotopic, while those with some portion of the ureterocele permanently located at the bladder neck (or rarely urethra) as ectopic. Up to 80% of paediatric ureteroceles occur with ureteric duplication anomalies and are termed duplex-system ureteroceles, with 60-80% of these having an ectopic drainage orifice (6). In contrast to the majority of paediatric ureteroceles, those presenting in adulthood are typically singlesystem ureteroceles, which insert in a normal position in the trigone of the bladder, with only mild dilatation of the distal ureter. Although singlesystem ureteroceles usually manifest in adulthood, they have been reported with increasing frequency in recent years amongst urologists and radiologists in paediatric practice (7). Single-system ureteroceles are distinguishable clinically from the more common duplex-system ureterocele by their frequent occurrence in male children (in direct contrast to female predominance in duplex-system ureteroceles) and their association with multicystic dysplastic kidney (7). The aetiology of ureterocele has never been definitively established. A number of theories exist and most conclude that the underlying abnor-

mality is mucosal in origin and ureteroceles, in particular ectopic ones, involve a defect of the muscle coat of the affected ureter and often a defect in the bladder wall. Moreover, it is thought that the ureteroceles associated with single systems are not all congenital and may be acquired in adulthood (6). However no one theory satisfactorily explains the different types of ureterocele and it is thought unlikely that the underlying aetiology has any impact on clinical practice and management of ureterocele. The clinical presentation is varied. The advent of routine prenatal ultrasonography as part of obstetric care in the developed world has significantly affected the mode of presentation as well as the age at presentation of paediatric ureteroceles, with the number of neonates with prenatally detected ureteroceles having increased from 2% to 28% over the last two decades (8). Furthermore, prenatal diagnosis of ureterocele represents 15% of all prenatally diagnosed duplex kidneys (9). Prior to this, almost all children presented in early childhood with symptoms and signs of urinary tract infection, and overall, this is still the most common mode of presentation in children of both sexes (10). However, a number of other presentations are reported (11). Bladder outlet obstruction is extremely rare and may occur secondary to a prolapsing ureterocele at the bladder neck or proximal urethra. In this case, we have described outlet obstruction occurring secondary to both ureteroceles extending into the proximal urethra, likely resulting in a ball-valve effect. A prolapsed ectopic ureterocele in a female child may rarely present as an inter-labial mass. Rarely, children with ureterocele may present with haematuria after minimal trauma (9). Other symptoms may be more nonspecific and include failure to thrive or vague abdominal pain (4). Urinary incontinence may occur due to interference with normal sphincter function at or distal to the bladder neck by an ectopic ureterocele (11). In our patientâ&#x20AC;&#x2122;s case, it was felt that the large ureteroceles prolapsing into the bladder neck as demonstrated on ultrasound, MCUG and at cystoscopy, interfered with sphincter function in this way to cause a signif-

313

icant degree of urinary incontinence. The cystic distal ureter likely allowed stasis of urine and promoted the development of infection, resulting in the pus formation which was drained at endoscopic puncture. Endoscopic puncture has been found to be very successful at achieving decompression (91%) of a ureterocele (12). Some patients may require a second procedure, so follow up is required. References 1. Malek R.S., Kelalis P.P., Burke E.C., Stickler G.B.: Simple and ectopic ureterocele in infancy and childhood. Surg Gynecol Obst, 1972, 134: 611616. 2. Uson A.C., Lattimer J.K., Melicow M.M.: Ureteroceles in infants and children: a report based on 44 cases. Pediatrics, 1961, 27: 971-977. 3. Rickwood A.M.K., Reiner I., Jones M., Pournaras C.: Current management of duplex-system ureteroceles: experience with 41 patients. Br J Urol, 1992, 70:, 196-200. 4. Shokeir A.A., Nijman R.J.M.: Ureterocele: an ongoing challenge in infancy and childhood. BJU International, 2002, 90: 777-783. 5. Glassberg K.I., Braren V., Duckett J.W., et al.: Suggested terminology for duplex systems, ectopic ureters and ureteroceles. J Urol, 1984, 132: 1153-1154. 6. Coplen D.E., Duckett J.W.: The modern approach to ureteroceles. J Urol, 1995, 153: 166-171. 7. Zerin M., Baker D.R., Casale A.: Single-system ureteroceles in infants and children: imaging features. Pediatr Radiol, 2000 Mar, 30: 139-146. 8. Blyth B., Passerini-Glazel G., Camuffo C., Snyder H.M., Duckett J.W.: Endoscopic incision of ureteroceles: intravesical versus ectopic. J Urol, 1993, 149: 556-559. 9. Merlini E., Lelli Chiesa P.: Obstructive ureterocele-an ongoing challenge. World J Urol, 2004 Jun, 22: 107-114. 10. Mor Y., Ramon J., Raviv G., Jonas P., Goldwasser B.: A 20-year experience with treatment of ectopic ureteroceles. J Urol, 1992, 147: 1592-1594. 11. Schlussel R.N., Retik A.B.: Ectopic ureter, ureterocele, and other anomalies of the ureter. In: Campbellâ&#x20AC;&#x2122;s Urology, 8th ed, Vol 3, Chapter 58: 2022-2034. 12. Chertin B., de Caluwe D., Puri P.: Is primary endoscopic puncture of ureterocele a long-term effective procedure? J Pediatr Surg, 2003, 38: 116-119.

JBR–BTR, 2010, 93: 314-316.

ANTENATAL DIAGNOSIS OF A SACROCOCCYGEAL TERATOMA B. Mbumba1, A. Massez2, P. Lingier3, C. Donner4, D. Vermeylen5, J.R. Makulo1, F. Lepira1, M. Cassart2 Although rare, sacrococcygeal teratoma is the most common congenital tumor. We report a case of a precociously diagnosed and rapidly growing cystic lesion. These tumors may be associated with hemodynamic and hemorrhagic complications. Therefore, affected fetuses should be carefully followed during the entire pregnancy by ultrasound and MR imaging in order to evaluate the evolution of the mass, its consequences on the fetal organs and to appreciate the wellbeing of the fetus. Delivery and post natal work up can therefore be optimized. Key-word: Teratoma.

Case report A cystic extra pelvic mass was diagnosed on the routine first trimester ultrasound of a 37 year-old woman. It demonstrated a cystic lesion of 9 mm arising from the perineum of the 13 weeks old fetus, presenting a mostly extra pelvic extension. At 18 weeks of GA, the mass increased to 37 mm diameter (Fig. 1). The US hemodynamic findings (signs of hydrops, middle cerebral artery velocity) were regularly followed and stayed in normal ranges. The MR imaging performed at 30 weeks of GA confirmed the cystic nature of the lesion which presented multiple septas. Its greater dimension at that age was around 160 mm. MR imaging also confirmed the extra pelvic extension with a small pararectal cyst but without any involvement of the abdominal or retroperitoneal compartments (Fig. 2). The liquid nature of the mass and its extra pelvic location in close contact with the coccygeal bone were highly suggestive of the diagnosis of sacrococcygeal teratoma (1). The sudden increase in size of the lesion prompted the decision of performing a cesarian section at 38 weeks. A transuterine needle drainage of the lesion under US guidance was performed to facilitate delivery. Neonatal MR imaging confirmed antenatal data. It also demonstrated the integrity of the lumbar and sacral vertebras avoiding unnecessary radiation exposure (Fig. 3). At the second day after birth, a complete surgical resection of the mass was successfully performed. The histological data confirmed the

Fig. 1. — Coronal US scan of the fetus at 19 weeks of gestational age showing the cystic pelvic lesion.

diagnosis of mature teratoma with cells originating from the three germ cells layers. Fetal imaging has allowed to adopt an adequate obstetrical management and to optimize the neonatal work-up. Discussion The sacrococcygeal teratoma, the most common congenital tumor, is associated, in its antenatal presentation with a lot of complications. In this context, an early diagnosis and frequent ultrasound follow-up are advised in order to appreciate the

From: 1. University Clinics of Kinshasa, RD Congo, 2. Department of Medical Imaging, 3. Department of Pediatric Surgery, 4. Department of Fetal Medicine, 5. Department of Neonatal Intensive Care Unit, ULB, Erasme Hospital, Brussels, Belgium. Address for correspondence: Dr M. Cassart, Service de Radiologie, Département d’Imagerie Pédiatrique et Gynéco-obstétricale, Hôpital Erasme, Université Libre de Bruxelles, Route de Lennik, 808, B-1070 Brussels, Belgique. E-mail: mcassart@ulb.ac.be

fetal wellbeing, to decide about the best mode and period of delivery and to optimize the neonatal workup. The diagnosis, the size and the evaluation of the extension of the mass constitute the most important contribution of antenatal diagnosis. The differential diagnoses of cystic pelvic lesions include preferentially the cystic teratomas and lymphangiomas. Other cystic pelvic lesions (ovarian cysts, mesenteric cyst, duplication cysts…) were significantly less probable in this context because of the extra-pelvic extension of the mass. The close contact of the lesion with the coccygeal bone was highly suggestive of a sacrococcygeal teratoma. US can also depict potential associated anomalies classically encountered in 5 to 26% of cases. They mostly concern the urinary tract (bladder exstrophy), the spine (dysraphism) and associated clubbed feet which are accessible to

ANTENATAL DIAGNOSIS OF A SACROCOCCYGEAL TERATOMA — MBUMBA et al

315

B Fig. 2. — Prenatal MR imaging, coronal (A) and sagittal (B) T2 weighted images showing the liquid nature of the lesion and its expansive extra pelvic extension. The para rectal cyst is also well depicted (arrows).

Fig. 3. — Post natal MR imaging showing the mass and the integrity of the sacrum.

Fig. 4. — Neonatal aspect of the lesion

316

US diagnosis. The association with ano-rectal malformation is described in Currarino triad and is more difficult to diagnose by US. MR imaging may be helpful in this context. In the present case, the lesion was an isolated finding. The fetal prognosis relies on the compressive mass effect that such huge lesions can exert on the abdominal visceral organs (i.e major hydronephrosis) but also on the potential development of hemodynamic anomalies (in cases with important solid components) (2). Increased blood flow in the mass may lead to blood shunting and fetal cardiac failure. Intra tumoral hemorrhage can also occur and induce severe secondary fetal anemia. All these complications can be detected by US (3). An intra tumoral hemorrhage presents as an acute enlargement of the mass which appears more heterogeneous. An increase in systolic velocity of the median cerebral artery signifies about fetal anemia. In the present case no associated malformations or hemodynamic complications have been encountered. The antenatal prediction capacity about potential malignant transformation of the lesion is poor. It mostly relies on the volume of the solid component of the mass; the more solid tissue, the higher risk of malignant transformation.

JBR–BTR, 2010, 93 (6)

MR imaging is a complementary imaging technique to US performed in order to better characterize the tissular components of the mass (haemorrhage) and helps in precisely depicting the extension of the lesion particularly in very large lesions (4). The delivery is therefore adapted, some authors advise cesarian section after a percutaneus drainage of the mass under US guidance (whenever the fetus presents a high risk of haemorrhage (solid tumor, tumor size > 10 cm)) (5). This attitude has been adopted in the present case because of the huge size of the lesion (Fig. 4). The necessity of neonatal surgical resection is presently admitted considering the mass effect of the lesion but also its potential malignant transformation (6). Tumoral markers ( foetoprotein, HCG…) are currently measured out after surgery to have a baseline value allowing biological follow-up. An increase after surgery is suggestive of possible recurrence and leads to imaging investigations.

management of the fetus in utero, of the delivery and of the post natal work-up. References 1.

Conclusion The contribution of imaging techniques (US and MR imaging) is mandatory in the diagnosis and prognosis of sacrococcygeal teratomas. It allows a better

Altman R.P., Randolph J.G., Lilly J.R.: Sacrococcygeal teratoma: American Academy of Pediatrics Surgical Section Survey-1973. J Pediatr Surg, 1974, 9: 389-398. Benachi A., Durin L., Maurer S.V., Aubry M.C., Parat S., Herlicoviez M., et al.: Prenatally diagnosed sacrococcygeal teratoma: a prognostic classification. J Pediatr Surg, 2006, 41: 1517-1521. Westerburg B., Feldtein V.A., Sandberg P.L., Loppo J.B., Harisson M.R., Albanese C.T.: Sonographic prognostic factors in fetuses with sacrococcygeal teratoma. J Pediatr Surg, 2000, 35: 322-325. Avni E.F., Guibaud L., Robert Y., Segers V., Ziereisen F., Delaet M.H., Metens T. MR Imaging of fetal sacrococcygeal teratoma: diagnosis and assessment. AJR, 2002, 178: 179-183. Lo Curto M., D’angelo A., Cechetto G., Klersy C., Dall’Igna P., Frederico A., et al.: Mature and immature teratoma: results of the first paediatric Italian study. Pedriatr Surg Int, 2007, 23: 315322. Mekki M., Jallouli M., Krichene I., Belghith M., Jouini R., Sahmoun L., et al.: Les tératomes sacro-coccygiens. Expérience d’un service de chirurgie nord-africain. Oncologie, 2007, 9: 864868.

JBRâ&#x20AC;&#x201C;BTR, 2010, 93: 317-318.

LETTER TO THE EDITOR EPIPERICARDIC FAT NECROSIS: CT DIAGNOSIS B. Coulier1 Dear Editor,

In a recent large pictorial review we have revisited the spectrum of intraperitoneal focal fat infarction IFFI - a term including various acute abdominal conditions in which focal fatty tissue necrosis represents the common pathologic denominator (1). The option of conservative treatment after a very specific imaging diagnosis - today mostly obtained through ultrasound but merely through CT evaluation - now represents the other common denominator of IFFI. IFFI is a benign disease and informed radiologists can avoid surgery for most of their patients. Most cases of focal fat infarction are confined within the peritoneal cavity but there is a rare exception: epipericardic fat necrosis in the chest. We recently experienced the case of a 53-year-old patient presenting to the emergency department with a three days history of continuous increasing left precordial pain. This pain was breath dependent and the patient was forced to breathe superficially. There was neither dyspnea nor temperature. The patient was apyretic and its physical examination remained normal. Chest plain films were normal. Laboratory tests were normal except a mild elevation of CRP at 83 mg/l and a discrete elevation of D-dimer at 0,57 mg/l. A chest angio-CT was performed to exclude pulmonary embolism. The pulmonary arteries appeared normal but an inflammatory fatty mass was clearly demonstrated in the left anterior cardiophrenic space (Fig. 1). This inflammatory fat was separated from the normal pericardic fat by a focally thickened pericardium and thus was corresponding to epipericardic fat. Except for the intrathoracic location this CT findings were extremely similar to many previous experienced cases of intraperitoneal focal fatty

Fig. 1. â&#x20AC;&#x201D; Axial (A) , sagital oblique (B) and coronal oblique (C & D) MPR views of the left cardiac area show an inflammatory fatty mass in the left anterior cardiophrenic space (white arrow). This inflammatory fatty mass remains separated from the normal epicardiac fat by a focally thickened pericardium (small white arrowhead). Minimal sub-segmental collapse is found in the neighboring lung (small white arrow).

infarction (IFFI). The final diagnosis was that of benign epipericardic fat necrosis and the patient was successfully treated conservatively with analgesics and anti-inflammatory drugs. Epipericardial fat necrosis is a rare but important benign disease (2). Only about 25 cases have been

From: 1. Dr B. Coulier, MD, Department of Diagnostic Radiology , Clinique St Luc, Bouge, Belgium. Address for correspondence: Dr B. Coulier, MD, Department of Diagnostic Radiology, Clinique St Luc, Rue St Luc 8, B-5004 Bouge, Belgium. E-mail: bcoulier@skynet.be

documented in the English-language medical literature since first description in 1957 (3). At the acute phase it is characteristically mistaken for a serious disease particularly myocardial infarction or pulmonary embolism (3, 4). The pathogenesis of epipericardial fat necrosis is unknown (2, 5) and its pathologic features are very similar to those of fat necrosis in epiploic appendagitis and omental infarction (5). The entity has commonly been designated as pericardial fat necrosis but on the basis of surgical results of previously reported cases this term appears as a misno-

318

mer because the lesion is located anterior to the pericardium and thus concerns the epipericardial fat rather than the pericardial fat itself (6). Pericardial fat necrosis classically strikes suddenly, without warning (2). Severe chest pain, typically pleuritic, is the initial manifestation. It is left-sided in most patients but right-sided pain is also reported. Pain is located anteriorly to the diaphragm and may radiate to the neck, shoulder, upper arm, axilla or back. It lasts from several days to one week or so, but can recur with less intensity for up to one year. Soon after the onset of pain the patient may be dyspneic, with tachypnea, tachycardia but electrocardiogram characteristically remains normal (2). Chest plain films obtained during the first days may show no abnormality but thereafter an anteriorly

JBR–BTR, 2010, 93 (6)

located mass invariably appears in or near the affected cardiophrenic angle. The mass is always contiguous with the cardiac silhouette (2). CT helps determine the nature and exact location of the chest mass. The main CT features are an encapsulated fatty lesion with inflammatory change such as dense strands and/or thickening of the adjacent pericardium (5). Radiologic follow-up shows spontaneous improvement or resolution of findings. Given the unique clinicoradiologic picture and benignity of pericardial fat necrosis coupled with the capability of CT to show the fatty nature of the chest mass, a clinical diagnosis and symptomatic care should suffice in most instances (2,6). Exactly the same conclusion was recently drawn for intraperitoneal focal fatty infarction.

References 1.

2. 3.

Coulier B.: Contribution of US and CT for diagnosis of intra-peritoneal focal fat infarction (IFFI): a pictorial review. JBR-BTR 2010; 93: 171-184. Fred H.L.: Pericardial fat necrosis: a review and update. Tex Heart Inst J. 2010; 37: 82-84 Jackson R.C., Clagett O.T. and McDonald J.R.: Pericardial fat necrosis; report of three cases. J Thorac Surg. 1957; 33: 723-729. Bensard D.D., St Cyr J.A., Johnston M.R.: Acute pleuritic chest pain and lung mass in an elderly woman. Chest. 1990; 97: 1473-1474 Pineda V., Andreu J., Cáceres J. et al. Lesions of the cardiophrenic space: findings at cross-sectional imaging. Radiographics. 2007; 27: 19-32. Pineda V., Cáceres J., Andreu J., Vilar J., Domingo M.L.: Epipericardial fat necrosis: radiologic diagnosis and follow-up. AJR. 2005; 185: 1234-1236.

JBRâ&#x20AC;&#x201C;BTR, 2010, 93: 319.

IMAGES IN CLINICAL RADIOLOGY Oriental cholangitis S. Maeyaert1, L. Bladt1, D. Vanbeckevoort1, W. Van Steenbergen2, W. Van Vaerenbergh3

A 51-year-old Filipino woman was referred to our department for diagnostic and therapeutic work-up of mildly abnormal liver tests. Two years before admission, she presented with the same biochemical findings in another hospital. At that time, CT scan has been performed (Fig A) as well as an ERCP with sphincterotomy and extraction of common bile duct stones. There was no history of abdominal pain, biliary colic, jaundice or fever. The clinical examination did not reveal any specific findings. The ERCP performed at the referring hospital showed a dilatation of the intra- and extrahepatic bile ducts, and a large intraluminal filling defect in the left hepatic duct, compatible either with a stone or with a tumor in the left hepatic duct. To further investigate the bile duct system, MRCP was performed first. It showed a high-grade stricture of the left hepatic duct and a mildly irregular, saccular dilatation of the left intrahepatic ducts with sludge and multiple stones. These findings already suggested the diagnosis of oriental cholangitis. The MRI also showed a dilatation of the right intrahepatic ducts and of the common bile duct down to the papilla of Vater. A mechanical fibrotic stenosis at the level of the papilla, as it can be seen with oriental cholangitis, was suspected. ERCP confirmed the presence of stones in the left hepatic duct, as well as the presence of a stricture by demonstrating a â&#x20AC;&#x153;missing-duct signâ&#x20AC;? of the left hepatic duct; it also showed a marked decrease in arborization of the right hepatic ducts (Fig B and C). Due to the severe strictures of the left hepatic duct we thought that endoscopic removal of these stones would be impossible, so it was decided to perform a percutaneous transhepatic cholangiography (PTC) and percutaneous cholangioscopy (PTCS) with electrohydraulic lithotripsy via the right approach. After three PTCS procedures, a complete removal of all intrahepatic stones could be obtained. The patient could be discharged from the hospital and a control MRCP was planned six months later. Comment

Oriental cholangiohepatitis or oriental cholangitis is characterized by the presence of hepatolithiasis. This is defined as the presence of gallstones in the bile ducts peripheral to the confluence of the left and right hepatic ducts, irrespective of the co-existence of gallstones in the common duct and/or gallbladder. The disease is endemic in South-East-Asian countries but is now being encountered more often in Western societies due to an increased. The highest incidence of hepatolithiasis occurs in the fifth to sixth decades. The presentation of oriental cholangitis includes abdominal pain, jaundice, attacks of fever and chills. Longstanding intrahepatic gallstones frequently lead to cholangitis and sepsis, but in addition they are associated with an increased risk of cholangiocarcinoma. In contrast to gallbladder stone-disease, oriental cholangitis is relatively intractable and is characterized by frequent recurrences, demanding multiple operative interventions. The radiologic workup can consist of ultrasonography, CT scan and MRI. These non-invasive imaging modalities should not only be used to diagnose the presence of the intrahepatic stones, but more importantly to evaluate the location of the stones, the presence of biliary strictures and of a concurrent cholangiocarcinoma. Nevertheless, even with progression of modern imaging techniques, invasive techniques such as ERCP and PTC are indispensable for detailed diagnosis and, especially, for accurate treatment.

1. Department of Radiology, 2. Department of Internal Medicine, Hepatology Unit, University hospital Leuven, Catholic University Leuven, 3. Department of Internal Medicine, Gastro-enterology Unit, Hospital Heilig Hart Lier, Belgium.

JBRâ&#x20AC;&#x201C;BTR, 2010, 93: 320.

IMAGES IN CLINICAL RADIOLOGY Hypertrophic abductor hallucis muscle with nerve compression syndrome K. Boeren1, Y. Vankan1, A. Demeyere1, D. Perdieus1

A 13-year-old girl presented with a soft tissue swelling at the posteromedial side of the foot. This swelling had been present since birth but recently started to cause discomfort and radiating pain to the plantar aspect of the first and second toes. A magnetic resonance imaging (MRI) showed a soft tissue mass (2,6 x 3,7 x 13 centimetres) at the posteromedial side of the foot with signal-intensity similar to the intensity of muscle tissue on all the different sequences. There was no pathological contrast enhancement. Anatomically this mass was located at the abductor hallucis muscle (AH), representing a prominence or hypertrophy of the abductor hallucis muscle (Fig A). The diagnosis was made of a hypertrofic abductor hallucis muscle with narrowing the space for the medial plantar branches of the tibial nerve (circle) at the crossover between the flexor hallucis longus and flexor digitorum longus tendons (arrow), the quadratus plantae muscle (QP) and flexor digitorum brevis muscle (FDB) (Fig B). Example of a normal fibromuscular tunnel (Fig. C). Our patient was treated conservatively first and referred to a foot surgeon. Comment

Muscle hypertrophy is a relative rare anatomic entity which can be responsible for nerve entrapment syndromes, as in this case. For anatomical description purpose, the abductor hallucis lies along the medial side of the foot and covers the origins of the plantar vessels and nerves. It arises from the medial process of the tuberosity of the calcaneus and the flexor retinaculum and inserts, together with the medial tendon of the flexor hallucis brevis, into the tibial side of the base of the first phalanx of the big toe. It flexes and abducts the big toe and supports the medial longitudinal arch. Muscle bulks can best be demonstrated on axial MR images which shows the relation between the fibro-osseous tunnel and the adjacent soft tissue. The signal characteristics are similar to muscle tissue on T1 and T2 images. In some cases they can have an edema-type signal best appreciated on STIR-images. For treatment there are a number of nonsurgical options like adapted footwear but when a patient does not respond to conservative treatment surgery with nerve release is indicated. More typical causes of nerve entrapment are acute or repetitive minor trauma, cysts or bone and joint abnormalities. Awareness of less common causes and knowledge of the complex anatomic structures of fibro-osseous and fibro-muscular tunnels can aid early diagnosis and treatment.

1. Department of Radiology, Imelda Hospital, Imeldalaan 9, 2820 Bonheiden, Belgium

JBR–BTR, 2010, 93: 321.

IMAGES IN CLINICAL RADIOLOGY Celiac artery compression syndrome M. Eyselbergs1,2, M. Camerlinck1,2, W-J. Metsemakers3, J. De Leersnyder3, F.M. Vanhoenacker1,2

A 41-year-old woman was referred to our institution for evaluation of recurrent cramping pain in the epigastrium since a couple of months. Contrast-enhanced computed tomography revealed a high grade stenosis with a hook like configuration of the celiac trunk with poststenotic dilatation (Fig. A-B, arrowheads at the celiac trunk, single arrowhead:origin, double arrowhead: stenosis and triple arrowhead: poststenotic dilatation; asterisk in aorta) . Ultrasound of the abdomen confirmed the stenosis which was more pronounced during expiration. Doppler ultrasound showed peak velocities up to 278 cm/s during inspiration and 334 cm/s during expiration (Fig. C) at the level of the stenosis and poststenotic turbulent flow was clearly visible. The peak velocities in the aorta and superior mesenteric artery were all within normal range. The diagnosis of celiac artery compression syndrome was made. After multidisciplinary consultation, a conservative approach was chosen. Comment

The celiac artery compression syndrome (CACS), also known as the median arcuate ligament syndrome (MALS) or Dunbar’s syndrome, consists of a focal stenosis of the celiac artery arising as a consequence of compression by the median arcuate ligament. This fibrous structure is bow shaped and connects the left and right diaphragmatic crura on both sides of the aortic hiatus. Occasionally, this ligament may have an intimate relationship with the celiac trunk and may cause focal indentation of the celiac trunk. Some patients show a higher than normal origin of the celiac trunk, which may also predispose to CACS. This syndrome occurs typically in young slim woman (20 to 40 years old). The celiac stenosis may present as an incidental finding on imaging or may be symptomatic. Symptoms include nausea, vomiting, weight loss, pain in the epigastrium occurring usually postprandial, or after exercise. The clinical examination sometimes reveals an epigastric bruit that varies during respiration. The gold standard for diagnosis is catheter angiography in inspiration and expiration, although this examination is currently replaced by multi-detector computed tomography (MDCT) and CTangiography. Typically, a ‘hook like’ appearance of the celiac artery is seen. In addition, color Doppler sonography is the imaging modality of choice for functional evaluation of the severity of the stenosis which is more pronounced during expiration. This is explained by a cranial movement of the celiac trunk during expiration rendering the stenosis more significant. A peak systolic velocity greater than 200 cm/s is believed to be diagnostic. Differential diagnosis includes stenosis of the celiac trunk due to atheromatosis which is more frequently seen in the elderly patient. Furthermore, calcifications are absent in CACS. Treatment is still controversial. Decompression of the celiac trunk can be achieved by cleavage of the median arcuate ligament. Alternative methods such as percutaneous transluminal angioplasty and stenting of the celiac trunk can also be performed but with variable results. 1. Department of Radiology , AZ Sint-Maarten Duffel-Mechelen, 2. Department of Radiology, Antwerp University Hospital, Edegem, 3. Department of Vascular Surgery, AZ Sint-Maarten Duffel-Mechelen, Duffel, Belgium

JBRâ&#x20AC;&#x201C;BTR, 2010, 93: 322.

IMAGES IN CLINICAL RADIOLOGY Homocystinuria, typical brain MRI findings P. Iranpour1, N. Karamifar2, K. Asâ&#x20AC;&#x2122;adi

An 8-year-old male presented with recent history of headache, several episodes of convulsion and visual field deficits. Brain MRI revealed multiple T2 hypersignal intensity foci within the left cerebral peduncle, basal ganglia, thalami and left parietal cortex, believed to represent ischemia. There was also isoâ&#x20AC;&#x201C;slightly hyperintense signal at the site of superior sagittal sinus, replacing the normal flow void pattern, suggestive for dural sinus thrombosis. After gadolinium injection thrombosis of superior sagittal sinus was confirmed (Fig A). Evaluation of orbits disclosed posterior dislocation of the right lens (Fig B). These findings were compatible with the diagnosis of homocystinuria which was later confirmed by laboratory data. Comment

Homocystinuria is an inborn error of methionine metabolism caused mainly by low activity or a deficiency of the enzyme cystathionine- -synthase. This defect results in intimal irregularities that cause arteriosclerosis, arterial thromboembolism ,and venous thrombosis in children and young adults. This appears to be the result of excess homocysteine, which changes coagulation factor levels, prevents small arteries from dilating, and causes proliferation of smooth muscle cells in arterial walls. Homocystinuric infants appear healthy at birth, and their early development is unremarkable until seizures, developmental slowing, or cerebrovascular accidents occur between 5 and 9 months of age. Ectopia lentis is seen in more than 90% of affected individuals. Lenticular dislocation has been recognized as early as 18 months of age, but it generally occurs between 3 and 10 years.

1. Medical Imaging Research Center, Department of Radiology, 2. Department of Pathology, Shiraz Medical University, Shiraz, Iran

JBRâ&#x20AC;&#x201C;BTR, 2010, 93: 323.

IMAGES IN CLINICAL RADIOLOGY Unilateral sight loss in a 4-year-old girl K. De Smet, M. De Maeseneer, A. T. Yazdi, C. Ernst, J. De Mey

For 6 weeks the parents of a 4-year-old girl had noticed a difference between the two eyes of their child. Ophtalmological examination revealed leukocoria. This finding raised clinical suspicion of retinoblastoma. MRI was performed. On T2 weighted images a hypointense mass relative to the vitreous humor was evident. There were several hypointensities compatible with calcifications. An area of retinal detachment was also seen. On T1 weighted images a mildly hyperintense mass relative to vitreous humor was seen. It showed marked contrast enhancement. On diffusion weighted images there was restricted diffusion suggestive for a tumoral lesion. The imaging findings were compatible with a retinoblastoma without transscleral or optic nerve extension. The tumor was complicated by retinal detachment reducing visual potential. The patient was treated with enucleation of the affected eye. Comment

Retinoblastoma is the most common intraocular malignant tumor of childhood. It accounts for 5% of childhood blindness and 1% of cancer deaths.The incidence is 1/15000 to 1/30000 live births. There are two forms. A sporadic nonfamilial form due to spontaneous mutation of both copies of the Rb 1 gene in a retinoblast. This form accounts for most unilateral disease. Another from consists of a familial heriditary form. In this form bilateral disease is present in 25%. Bilateral disease always points to the heriditary form. Trilateral disease includes bilateral disease and a pineal or suprasellar tumor. Quadrilateral disease consists of bilateral disease and a pineal and suprasellar tumor. Imaging findings include an intraocular mass with multiple calcifications. In the absence of calcifications other mass lesions responsible for leukocoria should be suspected such as persistent hyperplastic primary vitrious, retrolental fibroplasia, toxocariasis, and Coatâ&#x20AC;&#x2122;s disease. The primary role of imaging is to determine tumor extension. Important findings are optic nerve extension, scleral break through and metastasis. On T1 weighted images the tumor is mildly hyperintense relative to vitrious. Contrast enhancement is moderate to marked. On T2 weighted images the tumor is hypointense relative to vitrious. Retinal detachment is also best demonstrated on T2 weighted images. Treatment consists of enucleation when vision can not be preserved. Other treatments consist of chemotherapy, external beam radiation therapy, plaque radiotherapy, cryotherapy, and photocoagulation. The rate of cure for noninvasive intraocular retinoblastoma is 90%. Extraocular disease has a more dismal prognosis.

1. Department of Radiology, UZ Brussel, Brussel, Belgium

JBR–BTR, 2010, 93: 324.

IMAGES IN CLINICAL RADIOLOGY Rigler’s triad in an 88-year-old woman K. De Smet, M. De Maeseneer, Y.A. Talebian, B. Ilsen, P. Dewachter, J. De Mey1 An 88-year-old woman was seen at the emergency department with sudden onset of abdominal pain and nausea. She had been constipated for a week. Medical history was positive for arterial hypertension, diabetes mellutis type II, and heart failure. Laboratory findings showed an elevation of cholestatic liver enzymes. CRP was within normal limits. An upright radiograph of the abdomen showed multiple air fluid levels (F) in the small bowel consistent with obstruction. Pneumobilia (short arrow) was also seen as well as an obstructing gallstone (long arrow). A CT of the abdomen was performed. This showed pneumobilia (long arrow), a small bowell faeces sign (white asterisk) in the jejenum consistent with obstruction, and a non radioopaque gallstone in the ileum (dark asterisk). A CT scan from a few years earlier shows signs of cholecystitis and the non radioopaque gall stone (dark asterisk).The patient was treated with laparoscopic enterotomy with removal of the gallstone and cholecystectomy. The cholecystoduodenal fistula was demonstrated. Comment

B C

Gallstone ileus is an unusual complication of biliary stone disease. It consists of a mechanical intestinal obstruction due to impaction of one or more large gall stones (larger than 2 cm). Gallstones usually enter the intestinal lumen through a cholecystenteric fistula. Most of these are between the gallbladder and the duodenum (cholecystoduodenal fistula). A history of prior biliary tract disease is present in 50 % of patients. The most frequent site of stone impaction is the ileum because this is the smallest part of the small bowel. Other sites of obstruction include the jejnum, stomach, colon, and duodenum. Gastric outlet obstruction or Bouveret’s syndrome occurs when the gallstone is located in the duodenal bulb. Gallstone ileus accounts for only 2% of intestinal obstructions. It occurs more frequently in the elderly and in women. Clinical diagnosis is often delayed due to a variable clinical presentation. Clinical symptoms include abdominal pain and nausea. Characteristic features of intestinal obstruction are found in only 50-70% of patients. There may be alternate phases of improvement and relapse due to progression of the gallstone along the digestive tract. Laboratory findings may show elevated levels of alkaline phosphatase and bilirubin. A high mortality rate exists (20%). Rigler’s triad consists of pneumobilia, small bowell obstruction and a visualized gall stone. It is considered pathognomonic for gallstone ileus, but not frequently seen on the initial abdominal radiographs. Findings are easily identifiable on MDCT even if the gallstone is not calcified. MDCT in our case was not necessary for diagnosis but for determination of treatment. In conclusion gallstone ileus is an unusual cause of intestinal obstruction. It has a high mortality rate if not diagnosed early. MDCT plays a crucial role in diagnosis and radiologists should be aware of Rigler’s triad. It allows for an earlier diagnosis reducing morbidity and mortality.

1. Department of Radiology, Univeristeit Ziekenhuis Brussel, Brussels, Belgium.

JBRâ&#x20AC;&#x201C;BTR, 2010, 93: 325.

IMAGES IN CLINICAL RADIOLOGY Articular tophaceous gout of the cervical spine: CT diagnosis B. Coulier, M.H. Tancredi

A 62-year-old female was referred to the department of medical imaging with complaints of right cervico brachialgia. MDCT examination of the cervical spine was performed. An exophytic pseudotumoral enlargement of the right posterior articular process of C6 was first suspected on the axial native scan (A, white arrow) but multiplanar reconstructions (B) clearly identified the articular nature of the process. Some relatively intact segments of the articular space were still present and vacuum phenomenon was clearly visible (black arrowhead) excluding an associated synovial effusion. Similar lesions were also found at the level of the costovertebral joints of T1 (black arrows). Both sides of the joint space were symmetrically affected by deep massive well circumscribed erosions which where filled by spontaneously very dense material. Nodular deposits of the same dense material were also found outside the natural limits of bones and some exophytic bone projections had developed around theses nodular deposits. These findings, and particularly the dense material, were recognized as gouty arthritis with massive tophaceous deposits. Laboratory tests revealed an elevated acid uric level at 104 mg/l (n.v 2065 mg/l). The patient was re-called for complementary X-ray. Plain films of the cervical spine (not illustrated) were rather uncontributive because the destructive lesion of the C6-C7 articular process was very difficult to characterize; moreover the erosions of the costovertebral joints were not detectable. Nevertheless the diagnosis of gout was reinforced by the presence of multiple typical erosions (black arrows) and tophi (white arrows) on the plain films of the feet (Fig. C). Comment

Gout is a common metabolic disorder with well-defined clinical, biochemical and radiologic features. Gouty arthritis typically affects the distal joints of the appendicular skeleton. Involvement of the axial skeleton is uncommon and tophaceous gout in the spine is rare. Most of reported cases presented with symptomatic cord or root compression. All segments of the spine can be affected but lumbar spine is the commonest region involved followed by cervical and thoracic spine. The site of involvement can be the vertebral corpus, the intervertebral disk and discovertebral junction, the ligamentum flavum, the posterior intervertebral ligament but also the epidural space, intradural space, the pedicles, facet joint, filum terminale, and neural foramen. Bone erosions by the urate crystal deposits and secondary proliferative bone changes are the prominent but no specific features of spinal gout on plain films. In the spine, the differential diagnosis include discovertebral infection, epidural abscess, rheumatoid arthritis, metastatic disease, amyloid spondyloarthropathy, facet joint infection, synovial cysts, and calcified intradural tumour. Most recent cases of tophaceous gout of the spine have been especially investigated with MR which represents now the usual imaging of spine, but numerous MR manifestations also lack of specificity. For example, on MR, gout tophi may exhibit isointense-to-hypointense signal on T1weighted images but variable signal intensity on T2-weighted images has been reported. On the contrary, though actually generally less often performed in first intension in spine disorders, CT appears very specific for the diagnosis of tophaceous gout; thophi appear indeed as very typical intra-articular and juxta-articular hyperdense massa of about 160-170 HU causing bony erosions with well-defined sclerotic margins.

Department of Diagnostic Radiology, Clinique St Luc, Bouge, Belgium.

JBRâ&#x20AC;&#x201C;BTR, 2010, 93: 326.

IMAGES IN CLINICAL RADIOLOGY Massive fatal simultaneous pulmonary and paradoxical embolism: MDCT diagnosis B. Coulier

An 81-year-old female was admitted to the emergency room with complaints of nausea, vomiting, vertigo and episodes of paresia of her right inferior face, right arm and both legs. There was neither confusion nor aphasia. Unenhanced CT of brain (not illustrated) demonstrated multiple hypodensities evocating multiple embolic infarctions. Dyspnea was also present at admission and laboratory tests revealed D-Dimers elevation. Thoracic angio-MDCT was immediately performed (A, B, C) and demonstrated bilateral massive pulmonary embolism with large serpiginous clots in both main pulmonary arteries (black arrowheads). The patient had also an aberrant right aortic arch in which a massive packet of fresh serpiginous clots was floating (black arrow). The fixation of this floating packet was assured by the deep impaction of a long portion of clot in the proximal third of the right subclavian artery (white arrowhead). The diagnosis of massive pulmonary embolism with massive synchronous paradoxical embolism was made and the patient was immediately admitted in the intensive care unit to manage intravenous fibrinolysis. Unfortunately, the patient died a few hours later from cardiac arrest. Comment

Paradoxical embolism is defined as a systemic arterial embolism requiring the passage of a venous thrombus into the arterial circulatory system through a right-to-left shunt. It is a relatively rare condition accounting for nearly 2% of systemic arterial emboli. Nevertheless it can lead to severe prognosis, with a reported rate of mortality in more than 20% of cases. This condition can be related to an abnormal intracardiac communication. The most common cardiac defect associated with paradoxical embolism is the patent foramen ovale (PFO) which reaches prevalence from 27 to 35% in the normal population. Under physiological conditions, PFO only determines a small amount of left-to-right shunt without any hemodynamic significant changes. In case of increased right atrial pressure the inversion of shunt from right-to-left can occur and lead to paradoxical embolism. The transient right atrial hypertension due to acute pulmonary embolism can be the presumed mechanism in many cases. It has been estimated that 67% of cases of paradoxical embolism are associated with pulmonary embolism. Usual sites of paradoxical embolism are the inferior limbs (49%), the brain (37%) and more rarely coronary, renal or splancnic arteries and retinal artery. The immediate recognition of paradoxical embolism is of primordial importance and treatment mainly consists of thrombectomy or thrombolysis. With concomitant pulmonary embolism a permanent IVC filter and anticoagulation are often needed.

Department of Diagnostic Radiology, Clinique St Luc, Bouge, Belgium.

JBR–BTR, 2010, 93: 327-329.

ANNUAL GENERAL ASSEMBLY OF THE ROYAL BELGIAN SOCIETY OF RADIOLOGY (RBRS), Brussels, 20.11.2010 Presidential address

Dear Colleagues, I am happy to welcome you to this session of the Annual Symposium of the RBRS devoted to « New Trends in Oncological Imaging ». There are two underlying statements in this title: first a specificity of oncologic imaging that goes beyond organ specificity and second a clear evolution in the expectations from the imaging techniques on part of the oncologists, which is following a paradigm change in cancer treatment. Response assessment in cancer treatment has become a major need and challenge, at a time when oncologists are facing a multitude of new clinical research protocols.

Reliability and standardisation are fundamentals when multi-centric, multinational studies are to be set in place. New imaging biomarkers are needed for early response assessment in highly individualized therapies. Integration or morphologic, metabolic and physiologic imaging requires more than ever a close cooperation between Radiology and Nuclear Medicine departments. These are the new trends and challenges for oncologic imaging today, and the framework on which the program of the Symposium is built. I am very honoured and pleased to welcome as the first orator about therapy assessment Professor Martine Piccart, Head of Department of Internal Medicine at Jules Bordet Institute Chair of the Breast International Group (BIG), and President-elect of the European Society for Medical Oncology (ESMO). She will bring us the perspective of the oncologist involved in clinical management and therapeutic research. Other distinguished speakers will present you multiple facets of the new methods and tools that we as radiologists could have to operate in the future to fulfil the needs of modern oncology. My colleague of the Nuclear Medicine department at Jules Bordet Institute, Patrick Flamen, will close the session with the important point of the needed cooperation between radiologists and nuclear medicine specialists. Honorary Membership of our Society will be presented to

Professor John Griffiths, MD, PhD, Co-Director of imaging at the Cancer Research UK Cambridge Research Institute, Honorary Professor of Magnetic Resonance at Cambridge University. He is a long standing researcher in MR of cancer, in both clinical and pre-clinical research. This particular way of doing will be of increasing importance in the future of oncologic imaging, participating in the translational trend of the oncologic research. The other foreign speaker to which we will present Honorary Membership is not a radiologist but a physicist, Tom Scheenen, MSc, PhD, from the University of Nijmegen, Radboud University Medical Centre. He is a specialist in MR spectroscopy of prostate cancer. His dynamism and the close multidisciplinary integration of the Nijmegen team illustrate another unavoidable trend in oncologic imaging, which is the necessity of a close association of physicians, basic scientists and engineers to achieve efficient and reliable results in tumour imaging. I thank the industrial sponsors who helped us to organize this Annual Symposium. Their support is greatly appreciated. I hope each of you will enjoy the stimulating and sometimes unusual topics of this Symposium. Have a pleasant Saturday morning, Pr. Dr. Marc Lemort, RBRS President 2010

Honorary Membership Nominee of the RBRS in 2010 The two scientists to which we present today the honorary membership of the RBRS are not of the same generation nor coming from the same

branch of science but I felt important to underline the essential contribution of scientists from basic and applied sciences in our radiological

practice today particularly in oncology . More than ever, we have to work in multidisciplinary teams to address the new challenges of imaging.

328

JBR–BTR, 2009, 93 (6)

Professor John Griffiths qualified in medicine from the St Bartholomew’s hospital in London before studying for his Ph D at Oxford University. He then was involved from the start in research in magnetic resonance and made the first magnetic resonance spectroscopy readings from a cancer patient and the first use of MRS to detect an anticancer drug in a tumour, in the early eighties. He is not a clinical radiologist by his initial training but a biochemist and worked at St George’s Hospital Medical School where he became Professor of Biochemistry as applied to Medicine. His research was since

the beginning funded by Cancer Research UK , a large charity organisation in Great Britain. He is now Senior Group Leader and co-director of imaging at the new cancer research UK Cambridge research institute, where he works on fundamental, preclinical and clinical topics in a translational way. He was awarded an honorary chair at Cambridge Univesity as Professor of Magnetic Resonance as applied to cancer. He has published more than 270 research papers, is Editor in Chief of the Journal NMR in Biomedicine which he founded in 1988. He was received in 2010 the European Magnetic Resonance Award.

Tom Scheenen is a physicist and is presently the MR physicist of the Radiology Department of the Radboud University Medical Center at Nijmegen in the Netherlands. He defended his PhD thesis ten years ago on NMR imaging of water motion in plants. Since then he is working full time on advanced MR application in medicine en was the coordinator of a multicentric study of MR spectroscopic imaging of the prostate, in which I had the pleasure to participate. He is now principal investigator of another multicentre study on MR multiparametric analysis of

prostate cancer for which he was awarded an European starting grant from the European Research Council. Tom is a physicist but has a clear vision of what is necessary for the patients and physicians. He shares his office with an interventional radiologist and is part of a truly multidisciplinary team. He has 39 scientific publications, is assistant-editor of Investigative Radiology, member of the board of the Erwin Hahn Institute for Magnetic Resonance Imaging in Essen, Germany where he is involved in 7 Tesla research work.

Mr President, dear colleagues,

I believe that all of us should give a big ovation to Marc’s outstanding performance. Dear colleagues, I do hope that you will grant me your support and cooperation during the next twelve months. Indeed, as you are well aware of, the economic environment is becoming increasingly rougher, and this, coupled with the fusion of the RBSR and the National Union of Radiologists, represents an important challenge to face. We all have to make sure that everybody’s interests shall be maintained. The bare fact that our field is subjected to more and more financial

President-elect address

I would like to start by presenting my sincere congratulations to Marc and the organizers for the excellent scientific level and organization of this symposium. I feel highly honored to have been elected as Chairperson of the Royal Belgium Society of Radiology and to take over from Marc. Dear Marc, you have set the level of this symposium and your work as President of the RBRS so high that it will be a major challenge for me to achieve such an excellence the forthcoming year.

ANNUAL GENERAL ASSEMBLY â&#x20AC;&#x201C; BRUSSELS, 20.11.2010

pressure nevertheless constitutes an opportunity to position us and to make use of the funds in the most effective manner. We all experienced during the last years severe cuts in the reimbursement of diagnostic and therapeutic medical imaging. It is up to us to provide the health care system with enough data to decrease these cuts, as well as to introduce new diagnostic

approaches and fields that will benefit to the patients and to the health care system. Moreover, I would like to insist on the importance of reinforcing the cooperation between our universities and all the other referring hospitals, in order to keep the Belgium diagnostic imaging at its high care and scientific level. I strongly believe that it can only be accomplished by a team effort,

329

and I would again like to use this opportunity to thank the board for the confidence conceded to my person by electing me. I ensure you that I will work to the best of my abilities to avoid disappointing you. I impatiently look forward to working with you all. Thank you. Dr Brigitte Desprechins President RBRS 2011

Members of the Symposium Faculty and of the Board of RBRS in 2010

From left to right: J. Verschakelen, J. Griffiths, M. Lemort, B. Desprechins, G. Villeirs, T. Scheenen, J.F. De Wispelaere, J. Pringot.

JBR–BTR, 2010, 93: 330-334.

ABSTRACTS OF PAPERS presented at the Annual Symposium of the SRBR - KBVR, on November 20, 2010 SAMENVATTINGEN VAN DE UITEENZETTINGEN voorgesteld aan het Jaarlijks Symposium van de KBVR, op 20 november 2010 RESUMES DES COMMUNICATIONS présentées lors du Symposium Annuel de la SRBR, le 20 novembre 2010 The specificity of oncological imaging M. Lemort1 It has now become common sense to say that treatment paradigms in oncology are changing. From large range, toxic, deleterious treatments that had only a very small margin of benefit we are now going to targeted, more specific, more effective therapies. This change challenges research mechanisms, treatment habits and health economics. Even if cancer is a widespread disease and a major cause of chronic illness and mortality, costs of large-scale research and development of a competitive business model begin hampering innovation. Diagnostic imaging being a tool of primary importance to assess the progress of disease and to give objective data on efficacy of a given treatment, it unavoidably has to take into account this paradigmatic change in therapy. Till now, radiology has had as first objective to detect and diagnose lesions based on morphological images and clinical data in the organ or area of concern. Primary lesions were first sighted, and then if it was going wrong, individual metastases. Then, if a classical treatment was applied, response criteria were disappearance of the lesion(s), or a (large) reduction in size (often coarsely measured). It is evident that clinicians and researchers on one side, industry on the other side are now demanding that we adapt our methods to design new imaging biomarkers that could be surrogate markers for therapy monitoring, giving earlier indications of treatment success and outcome. This implies not to rely only on anatomy and size measurements, but on other characteristics of the tumour such as metabolism or microenvironment. Physiological imaging or metabolic imaging (using or not specific tracers) is to be implemented. In-depth knowledge of both the common characteristics of cancer cells or tumours and the tools to explore these are then required. In addition, since cancer is by nature not a disease limited to an organ but has to be considered from the start as a potentially generalized disease, oncological radiology does not comply very well with the organ-oriented subspecialisation which is very common in our profession. There is a so-called transversal component complementary to the vertical,

organ-oriented, subspecialisation that forces oncological radiologists (which are not a legion) to have a more holistic approach of the patient. The same is true for the multidisciplinary interaction in the oncology centre. Even for such an – apparently – simple problem as size measurement of lesions, mastering of response criteria such as RECIST 1.0 or 1.1 that could imply measurements of multiple lesions in multiple organs is no easy task. Physiological and metabolic imaging often overshoots the limits of an organ or system and may require a whole-body approach. Development of novel imaging biomarkers imply a global careful thought from imaging specialists on standardisation and quality assurance, before large clinical trials can validate the approach. This is what we will try to do notably in the Imaging Group of the EORTC. This is why I think oncological imaging has a specificity which has to go beyond the traditional subspecialties of diagnostic radiology and also beyond the anachronistic distinction between radiology and nuclear medicine which is for me more and more nonsense. 1. Institut Jules Bordet, Cancer Centre of the Université Libre de Bruxelles, Brussels, Belgium. New trends in cancer therapy, new needs for imaging markers M.J. Piccart-Gebhart, K. Saini1 This is an exciting time in oncology. The past decade has seen a vast expansion in our knowledge about the basic biology of cancer. This has been made possible by advances on several fronts, including sequencing of the whole genome, the use of microarray technology, development of gene signatures, unravelling the signal transduction pathways, modulation of apoptosis and angiogenesis, gene silencing using RNAi, laser capture microdissection, use of targeted therapy, advanced bioinformatic algorithms, the use of nanotechnology in drug delivery, and advanced imaging techniques. Imaging technology has made remarkable progress in the recent past, and multi-slice spiral CT, high Tesla MRI, digital mammography, and PET are everyday

tools in many cancer hospitals. Multimodality imaging, especially the combination of PET and CT has matured into a valuable tool, and FDG-PET-CT has been widely adopted for diagnosis, initial staging, and response assessment in various malignant tumors with high diagnostic accuracy. While the development of integrated PET/MRI is technologically more challenging, advanced prototypes are already available. Monoclonal antibodies labelled with positron-, gamma-emitting molecules, optical dyes, or paramagnetic contrast agents are used for PET, SPECT, optical, and magnetic resonance imaging, respectively. Non-FDG tracers under development for oncology applications include markers of hypoxia, tumor proliferation; and amino acid metabolism. In addition, new tracers that bind to specific intra- or extra-cellular targets (eg. HER2 receptor) are being developed. Overexpression of the HER2 protein is seen in about 20% of invasive breast cancers. Such cancers are associated with high recurrence rate, shorter disease free interval, a propensity for cerebral metastases and decreased survival. Molecular imaging of HER2 can therefore facilitate a tailored approach based on specific tumor characteristics of individual patients. Because of the inherent heterogeneity of breast cancer and possible discordance in HER2 status between primary tumors and distant metastases, a non-invasive method like molecular imaging is needed to assess HER2 expression. Molecular PET imaging using radionuclide-labelled trastuzumab is currently undergoing human trials, and is expected to help select patients for appropriate anti-HER2 therapy. More importantly, serial molecular imaging will help differentiate between responders and nonresponders. This early readout of drug efficacy will thus modulate future treatment and help deliver truly “personalized medicine”. 1. Department of Internal Medicine, Jules Bordet Institute, Belgium, Brussels. How to scan a patient in a clinical trial? D. De Becker1 In order to accelerate validation of new drugs in oncology and reduce costs,

ABSTRACTS OF PAPERS clinical trials endpoints are often based on response rate (observable and measurable tumor shrinkage directly linked to drug effect) or time to progression as surrogate of overall survival, meaning that radiology is a cornerstone in clinical trial evaluation. Imaging plays 3 roles in clinical trials: detection, characterization and quantitative assessment. The first 2 receive most attention in radiologists training, the latter less. Quantitative assessment is nevertheless of utmost importance as a decrease in variability of measurements permits to reduce cohort to be investigated to study drug effect. CT is the most routinely used technique in oncological trials evaluating effects of tested new drugs as it can explore quickly all body, as it is an easily reproducible examination, widely available and not too expensive to use. Parameters of acquisition of the CT system as slice thickness, reconstruction intervals, KV and mAs, field of view, IV injection of contrast protocols, iodine contrast load, use of oral contrast, coverage of body to be scanned must be clearly defined to insure reproducibility of examinations and should be applied throughout the study for each enrolled patient. In case of contra-indication of iodine contrast media, abdomen and pelvis can be studied by MRI. Sequences must be clearly specified. Examination will be completed by a plain chest CT to screen and follow lung metastasis. At this point in time tumor evaluation remains mostly based on anatomical measurements as functional imaging is still neither sufficiently standardized nor widely available, with the exception of FDG-PET in specific situations. To document objective response to treatment, radiological examinations are assessed utilizing specific criteria fitted to the tumor studied. The oldest and less codified is WHO bidimensional measurement. RECIST 1.0 and now new version 1.1 are tools to follow solid tumors. Cheson criteria are adapted to follow lymphoma. More specific criteria are needed as some new medications do not provoke tumor shrinkage: Choi’s criteria for the follow-up of GIST are not only based on tumor dimension but also on internal density modifications due to treatment. New criteria are needed to assess effect of non cytotoxic drugs. Specific training must be organized for each study to reduce variability between readers, as this is a FDA requirement. Most clinical trials are multicenter and for phases 2, 3 and 4 enroll a large number of patients. Care must be taken to assure a constant imaging quality through accreditation of medical devices. Most studies impose the follow-up of patient on same equipment. For new drug approval by regulatory agencies like FDA or EMEA, centralized review is organized in addition to local radiologist work, with interpretation performed by 2 independent reviewers and

an adjudicator in case of disagreement. Transmission of anonymized data to imaging core lab is now facilitates by PACS electronic archiving with DICOM standard. When coping with clinical trials examinations, care must be taken not to compress data to be compliant with FDA rules. In conclusion, it may be said that radiologists play a key role in oncologic clinical trials as tumor response is actually mostly based on anatomic measurements. Reproducibility in the data acquisition parameters, lesion measurements and assessment in accordance with specific criteria adopted in clinical trials are of utmost importance. In the future new imaging techniques based on physiologic effects of drug showing earlier changes than tumor size regression will be utilized but they are at this time considered not widely enough accessible or not sufficiently reproducible for multicenter large scale clinical trials and they must still be validated by regulatory agencies. 1. Department of Radiology, Institut Jules Bordet, Brussels, Belgium.

Automated diagnosis from brain cancer metabolism J.R. Griffiths1 Magnetic resonance spectroscopy methods can be employed to study medical problems both ex vivo and in vivo. NMR spectroscopy ex vivo by classical, vertical-bore instruments is often used to obtain metabolic profiles of cancer cells and tissues. Solid biopsies can be studied by HR-MAS NMR or by perchloric acid extraction. A typical problem would be comparison of a tumour cell line with a single gene knocked out (or overexpressed) against the normal or “wildtype” cells in which that gene was functional. The data obtained a set of concentrations of tissue metabolites, is a subset of the metabolome – the totality of smallmolecule metabolites in an organism, cell or disease state – and is termed a “metabolic profile (Griffiths & Stubbs, 2005). Data analysis methods such as principle component analysis (Raamsdonk et al., 2003), and pattern recognition (Tate et al., 2006) are often necessary to visualise these complex datasets. I will show the use of use of a relatively new method: metabolite-metabolite correlation analysis (Kose et al, 2001). These correlations maps are helpful in understanding the perturbed metabolic pathways in the cells due to gene modifications, enzymatic modulations (inhibition/over-expression), toxic and/or drug effects and nutrient supply. However, although NMR of the metabolome is proving to be a very useful scientific technique it seems unlikely that studies of patient biopsies by this method will have a routine clinical role as there are many established methods for studying biopsies ex vivo.

331 A routine clinical use for MRS is more likely to be found in vivo, although attempts to develop one have been fraught with difficulties. The history of MRS (i.e. NMR spectroscopy in vivo) is almost as long as that of MRI. The first rat muscle spectra were produced in 1974 (Hoult et al., 1974), the year after Lauterbur and Mansfield independently demonstrated MRI. At first it was by no means clear that MRI would be clinically useful, and the unique ability of MRS to detect and quantify metabolites noninvasively in living tissue seemed at least as promising as the low-resolution soft tissue images from the early MRI instruments. However, during the 1980s and early 1990s MRI rapidly became an enormously important clinical imaging method, revolutionising the diagnosis and management of many diseases. In contrast, MRS is still looking for a routine clinical role more than 30 years after it was introduced, even though it has become well-established in many areas of medical research. The number of metabolites detected by MRS in vivo is much smaller than with the ex vivo methods. By 1H MRS of tumours in vivo it is usually possible to detect lactate, lipids, a “total choline” peak, creatine, myo-inositol and N-acetyl aspartate. Despite this limitation many studies have shown that 1H MRS can give useful information about brain cancer (Howe et al. 2003), and it would be easy to integrate it into a conventional radiological MRI examination (it adds around 20 minutes to a session). Also, the current diagnostic procedures leave much to be desired. Although MRI gives exquisitely detailed images and the radiologist can often give a fairly confident diagnosis, the definitive diagnosis and the grading of malignancy are still made by the histopathologist from a biopsy of the living brain, a very unpleasant procedure with significant risks of haemorrhage causing stroke-like paralysis, and even death. However, even though the existing method is unsatisfactory and even though addition of 1H MRS to routine radiological examinations, can substantially improve diagnosis and grading without requiring a brain biopsy, there has been no widespread clinical adoption of MRS for brain tumour diagnosis. One possible reason is that few radiologists are skilled at MRS interpretation, a method that is very different from the subjective image analysis in which they are trained. To try and eliminate that bottleneck, two multi-centre EU programmes (INTERPRET and eTUMOUR) have developed an automated pattern recognition method that requires no biochemical knowledge (Tate et al., 2006), and combined it with a large database of spectra obtained from brain tumour cases with validated diagnoses. An anticipated problem that was overcome by this method was the subtle differences between spectra produced by different manufacturers’ instruments or different pulse sequences (e.g. STEAM and PRESS): the program was successfully trained to ignore them.

332 The pattern recognition computer programs developed by INTERPRET and eTumour use the “raw” spectrum (i.e. without any manual quantification of the peaks). The INTERPRET prototype presents the results in a Decision Support System (DSS) that allows the radiologist (or other clinician) to integrate the new information into a conventional clinical diagnostic procedure. This procedure is designed to give the physician useful help in situations in which there are inadequate numbers of spectra of a rare type of cancer in the database, or even none at all. The eTumour method (CADS) is more automated, and the results are presented with a numerical probability for each alternative diagnosis. Conventional radiology classifies tumours with a specificity of 85-100%, though sensitivity is much more variable and some tumour types (high-grade meningiomas and low-grade oligodendrogliomas, for instance) are poorly diagnosed at present (Julià-Sapé et al, 2006). An unpublished blinded trial on the use of the INTERPRET DSS on 50 consecutive brain tumour patients (Margarida JuliàSapé, Carles Arús and colleagues) has shown that diagnosis of gliomas by radiologists is significantly improved when they are given the results from the DSS. A trial of the eTumour method (Monleon et al., 2010) gave roughly similar results. There is clearly scope for improving radiological diagnoses of brain tumours by the use of these decision support systems and MRS databases, but optimisation of the way they are used by radiologists is still required. Large trials will be necessary to demonstrate improved diagnosis of individual tumour types. If the these decision support systems go into regular clinical use we will have to find ways to constantly update the database of spectra on which they depend, as MRS technology will continue to change and histopathological classification will doubtless improve. In the long term it is not impossible that a non-invasive radiological method utilising MRS data could give a better prediction of disease outcome than conventional histopathology, but we will need to acquire extensive long-term response data to test that idea. 1. Cancer Research UK Cambridge Research Institute, Cambridge CB2 0RE, UK. From tissue to the cellular level: diffusion-weighted MRI and MR spectroscopy V. Vandecaveye1 At the time of diagnosis of malignancy, determination of locoregional disease extent and detection of potential distant metastases is important for treatment planning and assessment of patient prognosis. In the post-treatment phase, accurate and, preferably, timely assessment of the patients response to treatment is necessary in order to detect tumour recurrence at an early stage and

JBR–BTR, 2010, 93 (6) thus enable salvage treatment with curative intent. Additionally, non-surgical treatment such as RT, chemotherapy and the novel generation of targeted cytotoxic and antivascular agents do not necessarily lead to a decrease of tumour volume as a result to treatment, especially in the early treatment phase requiring. As such, specific imaging biomarkers are required. The reliance of morphological and size-related criteria confront conventional MRI to a certain extent with similar limitations as computed tomography (CT) for lesion characterization, and differentiation of tumoral recurrence as treatment induced inflammation and fibrosis. Similarly, in the non-surgical treatment of malignancy the response evaluation criteria in solid tumours (RECIST) may fall short to accurately separate responders from non-responders. Early detection of treatment effects and thus the early separation or responding from non-responding lesions may help to guide treatment escalation or change of treatment strategy in case of non-favourable response, while toxic side-effects may be avoided in non-effective treatment. In this setting, diagnostic imaging modalities that probe the tumoral microstructure or metabolism may be useful as they do not depend on lesion morphology, size or volumetric changes for lesion characterization. Mainly due to the technical challenges, the clinical application of magnetic resonance spectroscopy (MRS) remains limited to a number of anatomical subregions and strict clinical indications. Development of MRS has been focussed on applications in brain tumours, prostate cancer and breast tumours. MRS in combination with anatomical sequences may provide more detailed information about a tumour’s location and extent of its infiltration in the surrounding tissues. Furthermore, MRS may be useful for treatment monitoring and to guide biopsy procedures. Although the technique has already been researched for nodal staging in breast cancer and the head and neck, further technique optimization and especially improvement of spatial resolution is required in order to show additional value for characterization of subcentimetric lymph nodes. Diffusion-weighted magnetic resonance imaging (DWI) is a non-invasive imaging technique that measures differences in water mobility between different tissue microstructures. The water mobility is influenced by cell size, density and cellular membrane integrity, and is quantified by means of the apparent diffusion coefficient (ADC). The ability to measure small differences in tissue microstructure enables DWI to differentiate tumoral tissue from normal tissue and inflammatory or necrotic tissue. Currently, DWI is often used for imaging of body-regions (for instance: liver, head and neck, pelvis,….) as a problem solving technique complementary to CT or fluorodeoxyglucose positron-emission tomography (FDG-PET); for instance in the detection of small liver metastases. In

addition, DWI shows value for tumour differentiation and primary tumour detection, locoregional staging (primary tumour and regional lymph nodes), characterization of metastases and early assessment of treatment effects. One of the major advantages of DWI in the recent past years has been the rapid technological development which makes the technique suitable for whole-body imaging. This may be of use for tumour screening and staging and may show value in the treatment follow-up or prediction of patients with systemic cancer spread like, for example, lymphoma or metastatic solid tumours like neuroendocrine tumours, breast cancer and colorectal cancer, where the treatment with antivascular and targeted cytotoxic compounds may lead to heterogeneous tumour response. In the presentation, advantages and disadvantages in the application of MRI and DWI will be outlined as well as their potential complementary value to anatomical and metabolical imaging modalities. In addition development for clinical applications will be discussed. Department of Radiology, University Hospitals KUL, Leuven, Belgium. Chemical Shift MR imaging in pre-operative evaluation of prostate cancer: ready? validated? challenging? T. Scheenen1 In order to improve the non-invasive detection of cancer in the prostate, its location, grade and stage, several MR techniques have been and are being explored. Before prostatectomy, it is of great importance to know the location and extent of the tumour to assess the risk of positive resection margins and to assess the possibilities of a nerve-sparing operation. Among the different MR techniques, chemical shift MR imaging (CSI) has shown promising results. With the CSI technique MR signals of molecules other than water and fat are studied. From every voxel of an imaginary threedimensional box around the prostate (Figure 1), spectra are constructed with signals from protons from metabolites at characteristic positions (the so-called ‘chemical shift’). Multiple studies have shown significant differences in the metabolic state of different tissue types as depicted by CSI. Prostate cancer tissue is characterized by reduced levels of citrate (Ci) and increased levels of cholinecontaining compounds (Cho), which both are detectable in vivo with CSI together with signals from creatine (Cre) and polyamines. In the multi-centre IMAPS (International Multi-Centre Assessment of Prostate MR Spectroscopy) study at 1.5T, CSI was validated for prostate cancer localization throughout the gland, and its reproducibility was assessed. At 1.5T, CSI of the prostate is usually performed with an endorectal coil for reception of the MR signal. With the steady increase in availability of 3T MR systems, CSI has also proven its value at this high-

ABSTRACTS OF PAPERS

333 results across different institutions. Once post-processing is fully automated and reliable (currently a shared challenge among vendors and clinical institutions), the technique is ready for introduction into clinical routine. The greatest challenges of the CSI acquisition technique can currently be found in ultra-high magnetic field strengths. 1. Department of Radiology, Radboud University Nijmegen Medical Centre, the Netherlands. Whole-body MRI: achievements and expectations F.E. Lecouvet, P. Omoumi, J. Malghem, B. Tombal, B. Vande Berg1

Fig. 1. — Axial T2-weighted MR image of the prostate of a patient with prostate cancer at 3 Tesla. From two voxels, indicated with a sphere, the spectra are shown: one in healthy peripheral zone (left) and one in a cancer focus (right).

er magnetic field strength in single institution studies, and possibilities for measurements without endorectal coils have been investigated. Recently, new possibilities at 7T were explored for CSI of the prostate at unprecedented spatial resolution. Within the IMAPS study, 99 untreated patients with proven prostate cancer from 8 different institutions underwent T2-weighted MR imaging and 3D CSI with an endorectal coil at 1.5T. After the MR exam, patients were treated with a prostatectomy, and the histopathological analysis of the resected prostates was the gold standard for the presence and location of cancer. Ten other subjects underwent the MR examination twice, to assess the reproducibility of the technique. Based on the gold standard, blinded from the CSI spectra, a selection of voxels was assigned to peripheral zone (PZ), central gland (CG), peri-urethral area (U), and cancer (PCa) tissue. After fitting the spectra and visual quality control values for the (Cho+Cr)/Ci ratio (CC/C) were evaluated as a marker for tumor tissue. Mean values of this metabolic ratio for PZ, CG and PCa were significantly different. Within non-cancer tissues, but between patients of different institutions, no statistically significant differences were found, and the within subject differences of the reproducibility study showed smaller standard deviations within a subject than across subjects. The area under the ROC curve, discriminating between cancer and non-cancer tissue was 0.88 for PZ and 0.76 for the combined CG & U. In a different, prospective

multi-center study in the United States no clear additional benefit of CSI over T2weighted MRI alone was found in the localization of cancer within the prostate. As this study was undertaken without validation of the at that point relatively immature measurement technique, without a clear definition of tumor sizes to be detected, and with – still existing – matching difficulties between histopathology and MRI, its results were disappointing. In both trials, a visual quality control after semi-automatic fitting of all data is still a necessity, as this process has not been automated in a robust and trustworthy way, which is the current hurdle to overcome before easy introduction of the technique into clinical routine. At 3 Tesla it has been shown that in a validation setting the CSI technique with only an array of external coils can produce almost equal results to CSI with an endorectal coil at 1.5 Tesla. Currently, more efforts are underway in a broader setting to not only validate this technique in a combination with other MRI techniques, but also to assess its diagnostic value in the localization and grading of prostate cancer. The current greatest technical challenge can be found at the ultrahigh field strength of 7 Tesla, at which the first volunteers have been measured with CSI techniques of not only protons, but also of phosphorus nuclei, offering potential for a more fundamental insight in the disease. In short, it can be stated that CSI of the prostate in the localization of prostate cancer is validated: it is a robust and quantitative technique producing similar

Whole body-MRI (WB-MRI) has emerged for several years as a new imaging technique. It enables the study of the complete skeleton, but extends its indications to the study of almost all organs of the body. The development of the sequences and coils, as well as the major input of diffusion-weighted imaging, contribute to establish WB-MRI as one of the leading imaging approaches of cancer, challenging and outperforming other techniques and current multimodality diagnostic algorithms used to stage cancer. The technical bases of image acquisition and reconstruction, recommendations for image reading and interpretation, limits and pitfalls of WB-MRI can be illustrated by practical examples. The challenges for the radiologists include adequate protocol design, highly dependent on the targeted cancer, disease-oriented whole organ screening, huge information management, and knowledge of limits and some confusing findings at WB-MRI. The next challenges are the use of WBMRI surveys to quantify metastatic disease and evaluate the response of this disease to treatment. The fundamental clinical and therapeutic issues and the role of WB-MRI in the modern management of cancer patients will be illustrated. 1. Department of Radiology, Cliniques Universitaires Saint Luc, Université de Louvain, Brussels, Belgium. Working together: data integration of CT and MR with PET and SPECT P. Flamen Medical imaging has undergone a tremendous development over the last three decades. Diagnostic information provided by the structural imaging modalities, essentially based on alterations of morphology, structure and anatomy, was complemented by functional (ultrasound, dynamic CT, diffusion and perfusion MRI) and metabolic or molecular information provided by nuclear imaging techniques (PET and SPECT). The success of PET is based upon the high avidity of most malignant

334 tumors for FDG, a surogate marker of glycolysis, its whole body field of view, and, most importantly, on its integration with CT scan, leading to the currently standard PET-CT technology. Integrated PET-CT resulted in a significantly higher diagnostic confidence of both modalities through continuous cross checking of structural and metabolic aspects of diagnosis. This proofed invaluable for restag-

JBR–BTR, 2010, 93 (6) ing disease after surgery or ablatif techniques, or after chemo or radiotherapy, because in these conditions accurate diagnosis of residual or recurrent tumor activity, utmost important from a prognostic point of view, was often difficult using stand alone radiological techniques because of the treatment induced nonspecific structural changes. The lecture will deal with the future developments of

multimodality imaging (PET-SPECT-CTMRI) and how further image integration will impact on the personalized cancer patient care of tomorrow. 1. Department of Nuclear Medicine, Free University of Brussels, ULB, Brussels, Belgium.

JBR–BTR, 2010, 93: 334.

ABSTRACTS OF PAPERS FOR FULL MEMBERSHIP RESUMES DES TRAVAUX DE TITULARIAT SAMENVATTING VAN DE TITULARIAATSWERKEN

CHEST 18F-FDG PET-CT and Fever of unknown origin B. Arys1 Fever of unknown origin (FUO) is frequently a challenging diagnostic problem. The present retrospective study evaluates the role of PET-CT using 18F-FDG in the investigation of FUO. Methods: A total of 97 patients (42 men, 55 women, age range 1384 years) underwent a 18F-FDG PET-CT scan for the clinical problem of ‘febris e causa ignota’ over a time period of 5 years. The performance of PET-CT in identifying the etiology of the FUO was assessed. Final diagnosis was based on histopathology, microbiology, clinical follow-up and imaging follow-up. Results:PET-CT detected suggestive foci of increased 18F-FDG uptake in

50 patients (51.5%). In 82% of these patients a final diagnosis was made. We found an infectious or inflammatory pathology in 35 patients, a responsible benign neoplastic cause in 2 patients and a malignancy in 4 patients. In 14% of the patients with a suggestive focus on PET-CT the result was false positive, as confirmed by other diagnostic means. 2 patients in this group were lost for follow up. 18-FDG PET-CT was negative in 47 patients (48.4%). 42 patients in this group were true negative and no cause for the FUO was found in follow up (follow-up period 12-48 months). In 5 patients a final diagnosis was reached by other means of investigation. 3 of these patients had a systemic infection, 1 patient had an erysipelas and 1 patient had a poorly differentiated gastric carcinoid tumor. Conclusion:In an overall patient population with FUO an infectious or inflammatory cause is found in 41%, in 5.3%

malignancy is found, in 2.1% a benign neoplastic cause and in 51.6% no cause is found. 18F-FDG PET-CT identified the underlying infectious of inflammatory pathology in 89.7% in the current study population. Underlying neoplastic processes were identified in 85.7%. The overall usefulness of PET-CT for diagnosis or exclusion of pathology was in this study 87% with a negative predictive value of 89% rising to near 100% for diagnosis or exclusion of focal pathology. Thanks to the high sensitivity, high specificity and the high negative predictive value PET-CT is a very useful diagnostic modality for assessment of FUO. It should however find its place as a second line diagnostic approach after a previous accurate workup. Further prospective studies are needed to confirm these findings. 1. Department of Radiology, Ghent University Hospital, Ghent, Belgium

JBR–BTR, 2010, 93: 335-336.

FORTHCOMING COURSES AND MEETINGS CALENDRIER NATIONAL – NATIONAAL KALENDER 19.11.11 Annual RBRS symposium 2011 Imaging of the spine and the spinal cord : state-of-the-art Organization: Dr B. Desprechins 02.02.11 SRBR – Section Cardiaque KBVR – Hart Sectie Information:rodrigo.salgado@scarlet.be 18.03.11 SRBR - Section Radiologie Cardiovasculaire et Interventionnelle KBVR – Sectie Cardiovasculaire en Interventionele Radiologie Antwerpen, UZA Information: Dr O. d’Archambeau olivier.archambeau@uza.be 25.03.11 SRBR - Section Ostéo-articulaire KBVR – Sectie Osteo-articulaire Beeldvorming In cooperation with Dutch Section Annual Symposium 2011: 19.11.11 SRBR – Section Radiologie Cardiovasculaire et Interventionelle/ KBVR – Sectie Cardiovasculaire en Interventionnele Radiologie 18.03.11, 24.06.11, 30.09.11, 16.12.11

Leiden, The Netherlands Theme: Imaging of foot and ankle Information: pieter.van.dyck@uza.be 21.05.11 SRBR - Section Ostéo-articulaire KBVR – Sectie Osteo-articulaire Beeldvorming Antwerp, UZA Information: pieter.van.dyck@uza.be 10.06.11 SRBR – Section Radiologie Tête et Cou KBVR - Sectie Hoofd en Hals In cooperation with Dutch section Rotterdam, Erasmus MC Information: Pr Dr R. Hermans, robert.hermans@uzleuven.be

30.09.11 SRBR - Section Radiologie Cardiovasculaire et Interventionnelle KBVR – Sectie Cardiovasculaire en Interventionnele Radiologie 16.12.11 SRBR - Section Radiologie Cardiovasculaire et Interventionnelle KBVR – Sectie Cardiovasculaire en Interventionnele Radiologie Information: Dr M. Laureys, marc.laureys@chu-brugmann.be

24.06.11 SRBR - Section Radiologie Cardiovasculaire et Interventionnelle KBVR – Sectie Cardiovasculaire en Interventionnele Radiologie Mont-Godinne, UCL Information: Dr J.F. De Wispelaere, jean-francois.dewispelaere@uclouvain.be SRBR-KBVR – Section Imagerie Cardiaque/ Sectie Hart Radiologie 02.02.11

SRBR - Section Ostéo-articulaire KBVR – Sectie Osteo-articulaire Beeldvorming 25.03.11, 21.05.11

SRBR-SRBR - Radiologie Section Radiologie Tête et Cou / Sectie Hoofd en Hals 10.06.11

Detailed and real time information is available on RBRS website at www.rbrs.org

CALENDRIER INTERNATIONAL – INTERNATIONAAL KALENDER 16.01.11 EUROPEAN SOCIETY FOR MOLECULAR IMAGING: EMERGING MEDICINE 2011 Les Houches, Frances Information: http://www.e-smi.eu 16-01-2011 ISET 2011: International Symposium on Endovascular Therapy Venue: Fontainebleau Miami Beach Miami, USA Information: www.iset.org 24-28.01.11 ERASMUS COURSE HEAD & NECK MRI Lisbon, Portugal Information: www.erasmus2011.com.pt

03-02-2011 SOCIETY FOR CARDIOVASCULAR MAGNETIC RESONANCE JOINT SCIENTIFIC MEETING 2011 Nice, France Information: http://www.scmreurocmr2011.org/ 10.02.11 JOINT INTERVENTIONAL MEETING Rome, Italy Information: www.jim-vascular.com 23.02.11 ANNUAL SYMPOSIUM OF THE AMERICAN SOCIETY OF SPINE RADIOLOGY 2011 USA, Hawaii Information: http://theassr.org

24-01-2011 ERASMUS COURSE ON MAGNETIC RESONANCE IMAGING (EMRI) Leiden, The Netherlands Information: www.emricourse.org

03-04-2011 SRES 11 SURGICAL AND RADIOLOGICAL ENDOVASCULAR SYMPOSIUM Venue: Hotel Carayou, Martinique

24-01-2011 SCANNER VOLUMETRIQUE VE SYMPOSIUM Nancy, France Information: www.imagerieguilloz.com

24-02-2011 6 TH CONGRESS OF ECCO: INFLAMMATORY BOWEL DISEASES 2011 Dublin, Ireland Information: www.ecco11.ecco-ibd.eu/

03-03-2011 ECR 2011 Information: www.myesr.org 14-03-2011 NEONATAL ULTRASOUND COURSE. WHY, HOW AND WHEN AN ULTRASOUND IMAGE? Florence, Italy Information: www.aimgroup.eu 26-03-2011 SIR 2011 SOCIETY OF INTERVENTIONAL RADIOLOGY ANNUAL SCIENTIFIC MEETING 2011 Chicago USA Information: http://www.sirmeeting.org/ 03-04-2011 SRES 11 SURGICAL AND RADIOLOGICAL ENDOVASCULAR SYMPOSIUM Martinique Information: www.rbrs.org 03-04-2011 43RD INTERNATIONAL DIAGNOSTIC COURSE DAVOS Information: www.idkd.org

336

JBR–BTR, 2010, 93 (6)

13-04-2011 14TH EUROPEAN SOCIETY OF GASTROINTESTINAL AND ABDOMINAL RADIOLOGY HANDS-ON WORKSHOP ON CT- COLONOGRAPHY 2011 Gothenburg, Sweden Information: http://www.esgar.org/ index.php?pid=285&lang=1

15-05-2011 INTERNATIONAL CONFERENCE OF NON-INVASIVE CARDIOVASCULAR IMAGING 2011 Amsterdam, The Netherlands Information: http://www.escardio.org/congresses/ ICNC10/Pages/welcome.aspx

18-04-2011 SHORT COURSE ON ABDOMINAL ULTRASOUND IN INFECTIOUS DISEASES AND TROPICAL MEDICINE 2011 Pavia, Italy Information: http://www.tropicalultrasound.org/

21-05-2011 ESGAR 2011 European Society of Gastrointestinal and Abdominal Radiology Venice, Italy Information: www.esgar.org

18-04-2011 ERASMUS COURSE Cardiovascular Bologna,Italy Information: www.emricourse.org 27-04-2011 GEST 2011 EUROPE - Global Embolization Symposium and Technologies Paris, France Information: www.gest2011.eu 28-04-2011 ESOR GALEN ADVANCED COURSE: Urogenital Cross-Sectional Imaging Barcelona, Spain Information: www.myESR.org/esor

08-06-2011 CONGRÈS 2011 MEETING CIRA-SFICV 10e réunion annuelle de CIRA conjointement avec la SFICV Montreal, Canada Information & Registration For Canadian and International participants: Registration (starting January 2011): www.ciraweb.org Email: info@ciraweb.org Tel.: 514-282-2744 For participants from France: Registration (starting January 2011): agence@coralys.fr Tel: 04 67 79 24 95

09-11.06.11 18TH EUROPEAN SOCIETY OF MUSCULOSKELETAL RADIOLOGY MEETING (ESSR) Crete, Greece Main theme refresher course: Bone marrow. Additional refresher courses on different MSK topics. Scientific papers on all aspects of MSK Radiology Information: http://www.essr.org/cms/ website.php?id=/en/essr_home.htm 22-06-2011 COMPUTER ASSISTED RADIOLOGY AND SURGERY - 25TH INTERNATIONAL CONGRESS AND EXHIBITION 2011 Berlin, Germany Information: http://www.cars-int.org/ 21-24.09.11 ISS MUSCULOSKELETAL IMAGING COURSE FUNDAMENTALS TO ADVANCED CONCEPT USA, San Diego Information: www.internationalskeletalsociety.com 08.11.11 11TH WORLD FEDERATION OF INTERVENTIONAL AND THERAPEUTIC NEURORADIOLOGY CONGRESS South Africa, Cape Town Information: www.wfitn2011.org