W1914 Clinical Features of Eosinophilic Colitis – A Report of 7 Consecutive Cases

W1914 Clinical Features of Eosinophilic Colitis - A Report of 7 Consecutive Cases Abdulrahman A. Alfadda, Stefan J. Urbanski, Eldon A. Shaffer, Martin Storr Background: Primary eosinophilic gastrointestinal disease (EGID) is a rare spectrum of gastrointestinal disorders that primarily affects the gastrointestinal tract. EGID causes inflammation rich in eosinophilic cells in the absence of known causes for eosinophilia. Eosinophilic colitis (EC) represents the least frequent manifestation of EGID and it is unknown whether it represents an independent group of diseases. Methods: The pathology database of the Calgary Health Region was used to identify all cases of EGID and EC. The medical records of all seven patients diagnosed with EC in the Calgary Health Region from 1996-2008 were retrospectively reviewed and the diagnosis was again confirmed by an expert pathologist. Results: The median age of presentation was 42 years. Four patients (57.1%) were females and three (42.9%) were males. Abdominal pain was present in six patients (85.7%), nonbloody diarrhea in four patients (57.1%), bloody diarrhea in two (28.6%) and two patients (28.6%) had more than 10% significant weight loss. Four patients (57.1%) reported a history of drug allergies indicating a predisposition towards allergic diseases but in none of the patients a distinct allergy causing EC was identified. Two patients (28.6%) were on non steroidal anti-inflammatory drugs at diagnosis. One patient (14.3%) was found to have additional peripheral eosinophilia. On histology, all patients had eosinophilic infiltrate in the Lamina propria, two (28.6%) had additional Muscularis propria infiltrate, and four patients (57.1%) had cryptitis. Conclusion: EC is a rare medical condition with different patterns of presentations. A high index of suspicion is needed to diagnose this disease. Pathology plays a crucial rule in diagnosis and colonic biopsies are important to rule out EC for the indication diarrhea.

W1917 New Method of Manometric Measurement of Gastroduodenal Motility in Conscious Mice Mineko Fujimiya, Rumi Tanaka, Koji Ataka, Kaori Atsuchi, Akihiro Asakawa, Akio Inui

W1915 Further Experience of Optimising Treatment Outcome on Azathioprine by CoPrescription of Allopurinol in Patients With Inflammatory Bowel Disease Melissa A. Smith, Anthony M. Marinaki, Simon H. Anderson, Peter M. Irving, Jeremy D. Sanderson

Since no previous studies have reported dual measurements of stomach and duodenal motility in conscious mice, we developed a manometric method to measure the gastroduodenal motility in the physiological fed and fasted states of conscious mice. By this method we measured, for the first time, the gastroduodenal motility in Y2 knockout mice and analyzed the effects of ghrelin on the gastroduodenal motility in conscious mice. To evaluate this new method, we provide the comparison on the effects of CCK-8 examined by present and previous methods. In the fasted state of mice, phase III-like contractions with frequencies of 7.8 +/- 0.5/h in the antrum and 6.6 +/- 0.7/h in the duodenum were observed. This fasted pattern was disrupted and replaced by the fed pattern after feeding, with an increase of the motor index (MI) immediately after feeding. Intravenous injection of ghrelin induced the fasted pattern in the duodenum when injected in the fed state and increased %MI (114.3 +/- 9.8%) compared to saline-injected controls (64.4 +/- 9.6%) in the antrum. Intravenous injection of CCK-8 disrupted phase III-like contractions in both antrum and duodenum, which were replaced by fed-like motor patterns accompanied with the elevation of baseline pressure. In Y2 knockout mice, the frequency of phase III-like contractions was decreased in the antrum compared to wild-type mice and the immediate increase of MI after feeding seen in wild-type mice was disrupted in Y2 knockout mice. Our model provides a new method for studies of gastrointestinal motility in various mouse models, including transgenic and knockout ones.

Background: Patients with a relatively high thiopurine methyltransferase (TPMT) activity preferentially methylate thiopurines, resulting in high methylated metabolites and low thioguanine nucleotides(TGN)levels. This is associated with hepatotoxicity and non-response. Co-prescription of thiopurines (at 25-50% of standard dose) with allopurinol, appears to circumvent this, optimising metabolite profile and clinical outcome. We have previously reported our experience of this combination in thiopurine-related hepatotoxicity. In this study we report further experience for a broader indication, including toxicity (mainly hepatic) non-response with adverse metabolite profile, and very high TPMT activity. Methods: Patients prescribed combination treatment were identified from the IBD clinic, TGN monitoring results and pharmacy records. Data was collected retrospectively. We were particularly interested in whether combination treatment overcame the specific problem that had precluded thiopurine monotherapy. We also analysed changes in metabolite profiles and subsequent clinical outcome. Results: 35 patients have been co-prescribed allopurinol and thiopurines since our previous report. 20 had crohn's disease and 15 ulcerative colitis, age range 20-65 years. Their average TPMT activity was 40 pmol/h/mgHb, (vs 32.5 in our population overall). In those receiving co-prescription for adverse effects (14 hepatotoxicity and 3 other:rash, nausea, fatigue) 94% were able to tolerate combination treatment with complete resolution of LFT abnormalities where relevant, of these 88% achieved clinical remission. In 10 patients failing thiopurine monotherapy 50% achieved clinical response on combination treatment. In 8 started on combination treatment as primary therapy 6 (75%) achieved remission, comparing favourably with historical controls. Importantly no patients had adverse effects from the combination regime, with the exception of a transient lymphopaenia in a patient that had mistakenly continued full dose thiopurine. Thioguanine nucleotide levels increased from an average of 192 to 449 pmol/8x108RBC on co-treatment, whilst methylmercaptopurine levels decreased from an average of 503 pmol/8x108RBC to just 16. Conclusions: Combination treatment with reduced-dose thiopurine and allopurinol circumvents predominant methylation, optimising metabolite profiles, avoiding hepatic and other toxicity and improving clinical response. Using combination treatment as first line in those with high TPMT activity should lead to a faster remission. Whether combination therapy should be used first-line in those with high TPMT activity warrants further investigation.

W1918 Mesenteric Hypoperfusion Following Acute Experimental Spinal Cord Injury Emily Qualls-Creekmore, Melissa Tong, Gregory M. Holmes Background & Aims: Apart from the visible sensory and motor dysfunctions, spinal cord injury (SCI) produces dramatic autonomic nervous system impairment, including gastrointestinal (GI) and vascular regulation. Many patients, especially those with high SCI (i.e. spinal level T3 or above), present reduced gastric motility and vascular hypotension that results in presumptive visceral hypoperfusion. Visceral hypoperfusion in the critically ill patient leads to ischemia and the initiation of an inflammatory cascade with consequent endorgan dysfunctions. Despite the dramatic clinical scenario, there has been no systematic investigation of GI blood flow after experimental SCI. We hypothesized that superior mesenteric artery (SMA) blood flow is reduced in T3-SCI rats 3d post-SCI. Insufficient SMA flow may lead to reduced post-prandial hyperemia and failure to meet the metabolic demands of the intestine during digestion. Methods: In Vivo SMA blood flow was measured in fed and overnight fasted laboratory rats that were deeply anesthetized with Inactin (100-150 mg/kg ip). Following a midline laparotomy, the SMA was isolated and a transonic flow probe was implanted so as to surround the SMA without impinging blood flow. Following instrumentation, the wound was closed and animals were placed on a feedback controlled thermal pad. In a separate study, placement of the flowprobe was preceded by implanting a feeding tube which terminated in the proximal duodenum for infusion of a mixed nutrient liquid meal (Ensure™). The resulting SMA flow rates were normalized for body weight before analysis. Results: In contrast to fasted controls during the acute (3d) phase, fasted T3-SCI rats displayed a 56±7% reduction in SMA blood flow. Ad libitum fed T3-SCI rats did not display an elevation in SMA blood flow (0%) while blood flow in fed control animals increased (140±7%). Unlike controls, duodenal infusion of Ensure in T3-SCI rats did not trigger an elevation in SMA flow. Furthermore, pathology of the intestinal lumen in T3-SCI

W1916 Reduced Visceromotor Response to Colorectal Distension in Serotonin Transporter Knockout (Ko) Mice Daniel P. Holschneider, Zhuo Wang, Raina Pang, Jean-Michel I. Maarek, Sylvie Bradesi, Emeran A. Mayer Altered serotonin (5-HT) signaling in the gut and brain has been implicated in the pathophysiology of irritable bowel syndrome (IBS). The serotonin transporter (SERT) mediates reuptake of 5-HT and thereby critically regulates 5-HT signaling within the brain gut axis. SERT-KO mice which exhibit altered bowel habits and anxiety-like behavior have been proposed as an animal model for IBS, even though alterations in visceral sensitivity have never been reported in this model. Our aim was to test the hypothesis that SERT-KO mice express

S-765

AGA Abstracts

AGA Abstracts

visceral hypersensitivity compared to wild-type (WT) controls. Adult, male homozygous (-/, n=8), heterozygous (+/-, n=7) and WT mice (+/+, n=9) were implanted with telemetry transmitters for abdominal electromyography (EMG). Visceromotor response (VMR) was evaluated as abdominal EMG signals evoked by CRD (20-s distensions delivered with a barostat at 10, 20, 30, 40, 50 and 60 mmHg in a pseudorandom sequence, repeated three times for each pressure level, with 3-min intertrial intervals). VMR was quantified as area under the curve of rectified EMG traces during CRD, normalized to the before CRD baseline. Similar to the biphasic EMG response previously reported in mice, average EMG activity was characterized by a peak during the first 5 s of CRD, followed by a much lower plateau thereafter. Compared to the WT, SERT-/- mice showed significantly lower VMR during the first 5 s of CRD (P=0.014). VMR in the SERT+/- mice fell between that of the WT and -/, showing a gene dose effect. No genotypic differences were noted during the plateau phase (6-20 s). Contrary to the hypothesized visceral hypersensitivity, SERT-KO mice showed visceral hyposensitivity compared to the WT. These results extend findings from earlier work demonstrating that short-term pharmacological blockade of SERT reduces colonic sensitivity in adult WT mice. These findings greatly reduce the face validity of the SERT KO mouse as a rodent model of IBS. Support by R24 AT002681 and NARSAD.

Satistical Ananlysis