Dhavan A. Parikh

Lemierre syndrome: a pediatric case series and review of literature.

BACKGROUND Lemierre syndrome is a rare disease of the head and neck often affecting adolescents and young adults. Classically, infection begins in the oropharynx with thrombosis of the tonsillar veins followed by involvement of the parapharyngeal space and the internal jugular vein. Septicemia and pulmonary lesions develop as infection spreads via septic emboli. Although a rare entity in modern times, Lemierre syndrome remains a disease of considerable morbidity and potential mortality. METHODS This was a retrospective review of 3 cases and associated literature. RESULTS A common 1- to 2-week history of fever, sore throat, neck pain, and fatigue was observed in all patients. Patient 1 developed right facial swelling, neck tenderness, trismus, and tonsillar exudate. Patient 2 displayed right tonsillar erythema and enlargement with right neck tenderness. Patient 3 revealed bilateral tonsillar enlargement with exudate and left neck tenderness. Subsequent studies included blood cultures and computed tomography, after which empiric antibiotic therapy was started. Patient 1 underwent drainage of a right peritonsillar abscess, right pressure equalization tube placement, and ligation of the right external jugular vein. He subsequently developed subdural empyemas, cavernous sinus thrombosis, and carotid artery narrowing and required 9 weeks of antibiotic therapy. Patients 2 and 3 developed pulmonary lesions and received 6 weeks of antibiotic therapy. Timing was crucial in all cases. CONCLUSIONS Lemierre syndrome is a rare but severe opportunistic infection with poor prognostic outcomes if left untreated. Early diagnosis and treatment is essential. Aggressive antibiotic therapy coupled with surgical intervention, when necessary, provides excellent outcomes.

216 Improved Diagnostic Yield With Early Video Capsule Endoscopy After Diagnosis of Obscure Gastrointestinal Bleeding

asymptomatic individuals to quantify the amount of injury and attempt to determine baseline scores. Methods: Healthy volunteers were recruited for this IRB-approved study. Subjects were excluded if they had taken any NSAIDs in the two weeks prior to enrollment . Other exclusion criteria included pregnancy, implanted medical devices, a history of Crohns disease, a previous small bowel resection, a history of anklyosing spondylitis, and gastrointestinal symptoms of diarrhea, abdominal pain or nausea. CE exams were performed after fasting and the administration of simethicone prior to capsule ingestion. All studies were read at a fixed frame rate of 7fps in a quad view. Mucosal breaks were defined as lesions with white or yellow centers and a red border. Scores were determined according to the Lewis Score. Results: 34 volunteers participated in the study (17/34 male, mean age 27 years [range 18-43]). The capsule reached the colon in 31/34 (91%) of exams. All studies were of good or excellent quality. The mean small bowel transit time was 249 minutes (range 78-404 minutes). 31/34 (91%) volunteers received a Lewis Score of zero. Three CE studies had a Lewis Score of 450 for small mucosal breaks in the first and middle tertiles of the small intestine. This score is consistent with mild mucosal injury. There were no adverse events in any of the participants. Conclusion: These results demonstrate that 9% of healthy individuals who do not use NSAIDs will have mucosal breaks on CE with Lewis Scores that do not surpass 450. The percentage of healthy volunteers with mucosal breaks is consistent with prior studies that included healthy individuals. Parameters and relative weight values used to analyze each tertile of small bowel

Keloid banding using suture ligature: a novel technique and review of literature.

Introduction: Successful wound healing represents the coordinated response of cellular, cytokine, and growth factor mechanisms involved in tissue recovery. Disruptions in transforming growth factor‐β signaling, senescence/apoptosis, keratinocyte‐fibroblast interactions, and other regulatory cascades can lead to the production of hypertrophic scar or keloid tissue formation. Current clinical investigations support surgical excision, meticulous closure, postoperative steroid injections, and postprocedural pressure dressings in the treatment of keloid tissue formation. Unfortunately, a universal approach in keloid therapy has yet to be identified. Here we offer a novel banding technique using suture ligature for the removal of these tissues.

Sa1385 Potential for New-Onset Diabetes Mellitus As a Marker of Resectable Pancreatic Cancer From EUS Registry

Background: New onset diabetes has been reported frequently with patients who are diagnosed of pancreatic adenocarcinoma (PDAC). Studies have suggested that 30-80 % of patients diagnosed with pancreatic cancer have new onset diabetes mellitus. However, whether new onset diabetes mellitus is associated with resectable pancreatic cancer stage has not been demonstrated. For consideration of potential for this phenotypic characteristic as initial surrogate/selection marker of PDAC for possible intervention, we have investigated the pattern of developing diabetes in patient population with resectable diagnosis of PDAC in our EUS registry data. Methods: We reviewed our IRB-approved UC Davis Pancreas Registry (11/2002 11/2013) retrospectively for identification of patients with PDAC diagnosis with potential resectability (AJCC stages I & II) with diabetes diagnosis (per ADA definition) and the duration of the disease; onset of diabetes less than 3 yrs prior to index PDAC diagnosis was considered as new-onset diabetes. Other associated variables including age, gender, BMI were also reviewed. Comparsion of the categorical groups (new-onset vs long-history of diabetes with age, gender, BMI) was evaluated with Fisher exact test (p<0.05 as significant). Results: A total of 181 patients were noted to have stage I-II disease from our database. 144 patients had data for new onset diabetes (within 3 years), diabetes more than 3 years or no diabetes. We found 56.9% (n=82) of resectable cancer stage had new onset diabetes mellitus (within 3 years), 11.8 % (n=17) had diabetes mellitus for more than 3 years and 31.25% (n=45) had no diabetes. Among new onset diabetes group, 79.2 % (n=65) of patients had new onset diabetes mellitus within one year of the pancreatic cancer diagnosis. There were Stage IA (n=9), stage IB (n=17), Stage IIA (n=12), and stage IIB (n=44). On comparing the different variables within the resectable group including age, BMI, gender, there were no significant differences found on the new onset diabetes group. Among the new onset diabetes mellitus group 5% of patients were under age 50, 44% were male and 61% of patients had BMI< 30 kg/m2. Amongst the diabetes mellitus for more than 3 years group, all the patients were above age 50, 65% were male and 67% of patients had BMI <30 kg/m2. There was no statistically significant difference in the age (p-value=0.58), gender (pvalue=0.18) and BMI (p-value=1) amongst new onset diabetes and late diabetes group. Conclusion: The new onset diabetes mellitus can be considered as an initial surrogate marker of resectable pancreatic cancer staging, allowing to capture a significant number of PDAC patients still at resectable stage. Further studies with additional variables including new biomarker(s) associated with the resectable pancreatic cancer stage are needed.

Mo1530 Utility of Serum CA19-9 in the Detection of Pancreatic Ductal Adenocarcinoma (PDAC) Among EUS-Referred Patients: the Clinical Features of Patients With Pdac With Normal CA 19-9

lesion (p 0.003), lesion size (p 0.01), dilation of the main pancreatic duct (MPD, p 0.01), total numbers of lesions (p 0.01) and presence of cystic component (p 0.03). Multivariate analysis (table 1) demonstrated that patients with PAC when compared to patients with pNETs are older [odds ratio (OR) 1.07], present with weight loss (OR 3.9), have elevated CA 19-9 (OR 6.36), CT-AP evidence of arterial invasion (OR 6.69), and SPL size 2 cm (OR 1.05). Further, patients with PAC have a dilated MPD when compared to patients with pNETs and METs (ORs 2.73 and 1.82 respectively). Patients with pNET when compared to those with MET have history of diabetes (OR 6.33). The sensitivity, specificity, positive predictive value (PPV) and accuracy of EUSFNA for all malignant lesions, PAC, pNET and MET is listed in table 4. While the highest sensitivity was for diagnosis of PACs, the specificity and positive predictive value were 100% for pNETs and METs. Conclusion: While pre-test findings can be suggestive of SPL pathology, EUS findings further improve diagnostic prediction by differentiating SPLs. Addition of EUS-guided FNA provides high PPV and accuracy in the diagnosis of SPLs.

W1177 Utilization and Efficacy of Video Capsule Endoscopy in the Evaluation of Anemia: Influence of the Severity of Anemia on Diagnostic Yield

G A A b st ra ct s of lesions in patients with OGIB. Methods: Forty CE examinations for OGIB (Pillcam SB2; Given Imaging Ltd, Israel) were randomly chosen. Two GI fellows (observers) analyzed the capsule endoscopy results with and without FICE enhancement using Rapid Reader (Version 6: Given Imaging Ltd, Israel). Randomization was such that a fellow did not assess the same examination with and without FICE enhancement. Both fellows were blinded to the findings of a senior consultant which were taken as reference. The senior consultant analyzed findings as P0, P1 and P2 lesions (non pathological, less likely to bleed, more likely to bleed respectively) based on double balloon enteroscopy findings. Inter-observer correlation was calculated by using kappa statistics. Sensitivity and specificity for P2 lesions was calculated for FICE and white light CE. Results: In 40 patients, observers 1 & 2 detected 102 and 80 lesions respectively as compared to 86 by the senior consultant. Intra-class correlation for P0, P1 and P2 lesions was calculated as -0.02, 0.75, 0.86 respectively. Intra class kappa correlation between the observers and reference findings was 0.85 and 0.92(P2), 0.69 and 0.82 (P1), 0.0 and 0.0 (P0) for CE using FICE and white light respectively. Overall 102 lesions were diagnosed by CE using FICE as compared to 80 by CE with white light (p = 0.8). For P2 lesions, the sensitivity was 90% vs. 86% and specificity was 55% vs. 78% for FICE and white light respectively. Significantly more P0 lesions were diagnosed when FICE was used as compared to white light (24, 4, p < 0.01). Conclusions: FICE was not better than white light for diagnosing and characterizing lesions on capsule endoscopy for OGIB. FICE detected significantly more non pathologic lesions.