A-545 A. Diagnosis

M. Krokidis | Monday, March 11, 08:30 – 10:00 / Room E1

Oesophageal cancer is the sixth leading cause of death from cancer worldwide. More than 90 % of oesophageal cancers are either squamous-cell carcinomas or adenocarcinomas. Approximately, three quarters of all adenocarcinomas are found in the distal oesophagus, whereas squamous cell carcinomas are more evenly distributed between the middle and lower third. The cervical oesophagus is an uncommon site of disease. The pathogenesis of oesophageal cancer remains unclear. At the time of the diagnosis of oesophageal cancer, more than 50 % of patients have either unresectable tumours or visible metastases on imaging. The most common symptom of presentation is dysphagia which is present in >70% of the cases; odynophagia may also be present in a smaller percentage of patients. The patients are usually presented also with significant weight loss which appears to be also an important prognostic factor of the outcome of the disease. Diagnosis is based on the findings of a contrast swallow- which is usually the first exam to be performed; oesophageal cancer may present as polypoid, infiltrative, varicoid, or ulcerative lesions. Endoscopy usually confirms the findings of the swallow study, revealing the presence of a mass and offering the possibility of taking biopsy samples. Endoscopic ultrasound is the imaging method that is used for local staging and CT and PET-CT are used to determine the presence of metastatic disease. In case of presence of enlarged lymphnodes, fine needle aspiration or even open biopsy may be performed.

B-0949 MRI-based selection of clinical complete and good responders after chemoradiation for rectal cancer allows for successful minimal invasive treatment

L. Heijnen, M. Maas, M.H. Martens, D.M.J. Lambregts, J.W.A. Leijtens, R.G.H. Beets-Tan, G.L. Beets | Monday, March 11, 14:00 – 15:30 / Room E2

Purpose: Patients with good or complete response after neoadjuvant chemoradiation have excellent long-term outcome. Minimal invasive treatment (i.e. transanal endoscopic microsurgery (TEM) and wait-and-see policy) are increasingly considered as an alternative to major surgery. With this prospective cohort study we aimed to evaluate long-term outcome of strictly MR-based selected patients who have been treated with minimal invasive treatment.
Methods and Materials: Eight weeks after chemoradiation, endoscopy and restaging MRI were performed (including diffusion-weighted MRI for yT-staging and gadofosveset-enhanced MRI for yN-staging). Complete responders were selected for wait-and-see policy and good responders with small tumour remnant for TEM. Both treatment groups underwent intensive 3-to-6 monthly follow-up, using MR imaging (DWI+gadofosveset), CEA, CT of thorax and abdomen and endoscopy was performed. Long-term outcome was estimated with Kaplan-Meier curves.
Results: Forty-one patients were included, thirty-three in the wait-and-see group and eight in the TEM-group. Mean follow-up was 26 months (range 6-91). For the TEM-group, 4 patients had ypT0 and 4 had ypT2. Two patients, both in the wait-and-see group, developed a local recurrence within two years and underwent surgery, leading to a 2-year local recurrence rate of 9 %. Both recurrences were detected on (DWI-)MRI in an early stage. The cumulative probabilities of 2-year disease-free survival and overall survival were 93 % and 100 %, respectively. No recurrences occurred in the TEM-group.
Conclusion: Both selection and follow-up of good and complete responders after chemoradiation for rectal cancer with MRI is feasible. Long-term outcome so far is excellent. (DWI-)MRI seems to be a reliable tool for early recurrence detection.

B-0948 Computed tomography of the bowel: a prospective comparison study between four techniques

M. Revelli, F. Paparo, L. Bacigalupo, A. Garlaschi, L. Cevasco, E. Biscaldi, G. Rollandi | Monday, March 11, 14:00 – 15:30 / Room E2

Purpose: Our purposes were to compare the grade of bowel distension obtained with four different CT techniques dedicated for examination of small intestine (CT-enteroclysis and CT-enterography), colon (CT with water enema), or both (CT-enterography with water enema) and to assess patient tolerance towards each protocol.
Methods and Materials: We recruited four groups of 30 patients. Each group corresponded to a specific CT technique, for a total of 120 consecutive patients (65 male, 55 female; mean age 51.09±13.36 years). CT studies were evaluated in consensus by two gastrointestinal-dedicated radiologists who performed quantitative and qualitative analysis of bowel distension. Presence and type of adverse effects were recorded.
Results: CT-enteroclysis provided the best distension of jejunal loops (median diameter 27mm; range 17-32mm) compared with all other techniques (p<0.0001). Frequency of patients with an adequate distension of the terminal ileum was not significantly different among the four groups (p=0.0608). At both quantitative and qualitative analysis CT with water enema and CT-enterography with water enema determined a greater and more consistent luminal filling of the large intestine compared with the one provided by both CT-enteroclysis and CT-enterography (p<0.0001 for all colonic segments). Adverse effects were more frequent in patients from the CT-enteroclysis group (p<0.0028).
Conclusion: CT-enteroclysis allows an optimal distension of jejunal loops, but it is the most uncomfortable CT protocol. When performing CT with water enema, an adequate retrograde distension of the terminal ileum was provided in a high percentage of patients. CT-enterography with water enema provides a simultaneous optimal distension of both small and large bowel.

A-349 B. MDCT in a patient with small bowel ischaemia

S. Romano | Sunday, March 10, 08:30 – 10:00 / Room B

Intestinal ischaemia and infarction are important causes of acute abdominal disease which appropriate diagnosis has to be as more as accurate being of crucial importance for the surgical or medical managament of the affected patients. At the basis of an efficient report lays the deep knowledge of the patho-physiologic mechanism leading to arterial, venous or low-flow state vascular injury of the intestines. When the mechanism is clearly understood, from a radiological point of view, it is essential to know how to optimise the MDCT technique in case of suspected intestinal ischaemia and what are main findings and intestinal features of the injured intestine. A further step is represented from the differentiation of various stages of disease from early potential transient ischemia to late infarction. However, it is also strongly important to become familiar with the reperfusion damage of the intestine, frequently observed but as still as a challenge for an effective diagnosis. The accuracy of the various radiological signs of bowel ischaemia should be considered and discussed, whereas presentation of several clinical cases with the description of their respective reports and final diagnosis still represents the “key” to improve an effective diagnosis of intestinal ischemia in daily radiological practise.

B-0627 Additional value of diffusion-weighted (DWI) MRI for predicting complete tumour response (T0N0) in rectal cancer treated with neo-adjuvant chemoradiation therapy (CRT)

S. Sassen, M. de Booij, M.N. Sosef, G. Lammering, C.M.M. Bakker, R. Clarijs, R.C.M. Berendsen, J. Wals, R.F.A. Vliegen | Sunday, March 10, 10:30 – 12:00 / Room A

Purpose: Patients with complete response (CR) after CRT might be considered for less aggressive treatment like a wait-and-see strategy. Few studies investigated the value of DWI-MRI for predicting CR after CRT, but none included lymph nodes in the analysis (ypT0N0). The aim of the present study was to retrospectively determine the additional value of DWI-MRI to conventional (T2-weighted) MRI for predicting CR after CRT.
Methods and Materials: Eighty locally advanced rectal cancer patients underwent CRT followed by restaging MRI and operation. MRI consisted of conventional sequences and DWI. Two readers with different levels of experience independently scored conventional images for CR and, in a second reading, combined conventional and DWI-MRI images. A 5-point confidence level score was used to generate ROC curves. Differences in performance were calculated by comparing areas under the ROC curves (AUC). Interobserver agreement, sensitivity, specificity and positive predictive values (PPV) were calculated. Histology served as reference standard.
Results: Ten of 80 patients (13%) had a pathologic complete response (ypT0N0). Comparison of the ROC curves showed significant improvement of the AUC only for the experienced reader 1 from 0,77 to 0,88 (p=0,009). Sensitivity improved from 20-30% to 40-70%. Specificity and PPV improved only for reader 1 from 87 to 93 %, resp., 25 to 58 %. Interobserver agreement improved from 0,14 to 0,27.
Conclusion: Adding DWI to conventional MRI improves diagnostic performance of experienced readers and increases interobserver agreement for identification of CR. Sensitivity and PPV remain low, with a considerable risk of over- and undertreatment.

B-0789 CT colonography: accurate registration of prone and supine endoluminal surfaces of the colon

T.E. Hampshire, H.R. Roth, E. Helbren, A. Plumb, D. Boone, G. Slabaugh, S. Halligan, D.J. Hawkes | Monday, March 11, 10:30 – 12:00 / Room E2

Purpose: Computed tomographic (CT) colonography is a technique for detecting bowel cancer or potentially precancerous polyps. Because retained fluid and stool can mimic pathology, CT data are acquired with the patient in both prone and supine positions. Radiologists then match endoluminal locations between the two acquisitions to determine whether pathology is real. This process is hindered by the fact that the colon can undergo large deformations that often occur during repositioning of the patient. Automated registration between datasets could potentially improve efficiency and diagnostic accuracy.
Methods and Materials: We have developed software to establish correspondence between prone and supine endoluminal surfaces. An initialisation step generates image patches at the positions of haustral folds using depth map renderings and is optimised by virtual camera registration. Additional neighbourhood information is then included in a Markov Random Field model to establish landmark-based correspondences. Subsequently, the complexity of the registration task is reduced by mapping both prone and supine surfaces onto a cylindrical domain in which correspondence is established using non-rigid image registration.
Results: The registration was applied to 17 CTC cases including cases exhibiting luminal collapse, achieving fold matching accuracy of 96 %. Providing an accurate initialisation, the method significantly improved the cylindrical registration (p<0.001), achieving a mean error of 6.0mm measured at 1743 reference points.
Conclusion: The proposed method can successfully establish correspondence between prone-supine locations on the endoluminal surface derived from CT colonography. The ability to rapidly and automatically match polyps between acquisitions will facilitate CT colonography interpretation.

B-0625 Features on MRI after transanal endoscopic microsurgery in patients with rectal cancer

L.A. Heijnen, M. Maas, M.H. Martens, D.M.J. Lambregts, J.W.A. Leijtens, W. Deserno, G.L. Beets, R.G.H. Beets-Tan | Sunday, March 10, 10:30 – 12:00 / Room A

Purpose: Transanal endoscopic microsurgery (TEM) is a minimal invasive technique for local resection of T1 and selected T2 tumours and is also an emerging option for good-responders after chemoradiation. In most centres follow-up includes regular MRI. This study aimed to describe the MR morphology of the rectal wall during follow-up in patients that received TEM.
Methods and Materials: Forty-nine patients underwent a post-TEM MRI in our centre. For 21 patients only one post-operative MRI was available. For 28 patients >1 MRIs were available. The MR morphology of the TEM-site was studied on the consecutive MR examinations. 32 patients were primary treated with TEM, 17 patients underwent chemoradiation followed by TEM.
Results: We identified three morphological patterns: (1) rectal wall thickening with or without fibrosis, (2) a notch at the TEM-location, and (3) irregular delineation of the rectal wall. Multiple patterns could occur within one patient. 32 patients (65%) had rectal wall thickening, 17 patients (35%) a notch, and 28 patients (57%) irregular delineation of the rectal wall. In addition to these patterns, oedema (due to chemoradiation) persisted in post-chemoradiation TEM-patients. Ten patients had dehiscence after TEM post-chemoradiation (n=58%). Six luminal recurrences occurred; 3 had rectal wall thickening, 3 a notch, and 5 an irregular rectal wall.
Conclusion: Three patterns were identified on MRI after TEM. This enables radiologists to monitor this group of patients more accurately. Since minimal invasive techniques are gaining, it is very important for radiologists and surgeons to have knowledge about the normal follow-up findings after TEM.

B-0198 Crohn’s disease activity: correlation of inflammatory mediators with overall small-bowel motility

S. Bickelhaupt, S. Pazahr, J.M. Froehlich, R. Cattin, H. Bouquet, G. Rogler, P. Frei, A. Boss, M. Patak | Thursday, March 7, 14:00 – 15:30 / Room E2

Purpose: Active Crohn’s disease (CD) increases the level of inflammatory markers of which C-reactive protein (CRP) and calprotectin are commonly used to monitor disease activity. The aim was to evaluate the correlation between CRP and calprotectin levels and overall small bowel motility in patients with Crohn’s disease assessed with MRI.

Methods and Materials: 13 patients with Crohn’s disease (4f/9m, mean 42y) were included in this IRB-approved prospective study. MRI (1.5-T, Philips Achieva) was performed after a 1-h preparation of 1000 ml Mannitol-Solution (3%). Cine T2w-2D-SSFP motility acquisitions (TR 2.47/TE 1.23/250ms slice repetition time) were performed in free breathing over 69-84sec. Randomly chosen small-bowel segments were analysed in two abdominal quadrants using dedicated MR-motility assessment software (Motasso). Contraction frequency, amplitude, luminal diameter and amplitude diameter ratio (occlusion ratio, ADR) were evaluated as well as CRP (ngl/ul) and Calprotectin (ug/g) levels. Pearson’s correlation was calculated.

Results: Calprotectin was determined in mean 12 days (SEM±10.09) before, CRP 15 days (SD±28.80) before MRI. A significant inverse linear correlation was found between the contraction frequency and both the level of CRP (r=-0.701, p=0.008) and calprotectin (r=-0.805, p=0.001). Expansion of the mean small bowel diameter significantly correlated with calprotectin levels (r=0.857, p=<0.001) but not with CRP (r=0.447, p=0.126). The absolute amplitude of the contractions did not correlate neither with the level of CRP (r=-0.527, p=0.064) nor with calprotectin (r=-0.612, p=0.026). The ratio describing relative luminal occlusion during contraction (ADR) significantly correlated with calprotectin (r=0.736, p=0.004) and with CRP (r=0.577, p=0.039).

Conclusion: Alterations of overall small bowel motility during active phases of CD significantly correlate with the level of calprotectin and CRP.