The Gastro & Hep Report

Gastroenterology & Hepatology

Spring 2015


Comprehensive Reports on the Latest Advances in Gastroenterology & Hepatology From:
American College of Gastroenterology 2014 Annual Scientific Meeting

October 17-22, 2014

Philadelphia, Pennsylvania
65th Annual Meeting of the American Association for the Study of Liver Diseases

November 7-11, 2014

Boston, Massachusetts
Advances in Inflammatory Bowel Diseases Conference

December 4-6, 2014

Orlando, Florida


Presentations in GERD

Implanted Device Restores Lower Esophageal Sphincter Function and Improves Quality of Life


The magnetic sphincter augmentation device is used to restore functioning of the lower esophageal sphincter and is an alternative to tissue fundoplication. At the American College of Gastroenterology (ACG) Annual Scientific Meeting, Philip Katz, MD, of the Einstein Medical Center Philadelphia in Philadelphia, Pennsylvania presented clinical findings in patients with gastroesophageal reflux disease (GERD) who were implanted with a magnetic sphincter augmentation device (Abstract 42).

The study included patients with chronic GERD who had an incomplete response to proton pump inhibitors (PPIs), abnormal acid exposure, small or no hiatal hernia, no Barrett esophagus, normal motility, and esophagitis of Los Angeles grade B or lower. All patients completed a GERD health-related quality of life (HRQL) questionnaire at baseline, while they were still taking PPIs. The questionnaire was again completed 4 years after implantation of a magnetic sphincter augmentation device, when patients were not taking PPIs. The GERD-HRQL score ranges from 0 to 50, with a higher number indicating worse symptoms.

Data were available for 86 patients. GERD-HRQL scores at baseline (when patients were receiving PPIs) were compared with scores 4 years after implantation of a magnetic sphincter augmentation device (when patients were not receiving PPIs). At 4 years postimplantation, the median total GERD-HRQL score improved from 11 to 4 (P<.0001; Table 1). For heartburn-related questions, the median GERD-HRQL score improved from 8 to 3 (P<.0001). The percentage of patients experiencing heartburn causing them to wake from sleep each night improved from 14% to 1% (P=.004). Many more patients reported being “satisfied” (84% vs 13% [P<.0001]). The percentage of patients with bothersome swallowing occurring daily increased from 2% to 7%, but this change was not significant (P=.10). Complaints of bothersome gas or bloating occurring at least daily decreased from 17% to 8%, but the difference was not significant (P=.11). The ability to belch or vomit was retained by 98% and 96% of patients, respectively.


Radiofrequency Ablation for GERD

Although the majority of GERD cases can be adequately controlled with medication, many patients require further treatment. Radiofrequency ablation with the Stretta system was recently recommended for GERD in guidelines issued by the Society of American Gastrointestinal Endoscopic Surgeons. At the ACG meeting, Seth Lipka, MD, of the Morsani College of Medicine, University of South Florida in Tampa, Florida presented results from a systematic review evaluating the efficacy of the Stretta device for managing GERD (Abstract 40).

Published studies were identified in a search of MEDLINE and the Cochrane Central Register of Controlled Trials from inception until February 28, 2014. Data were pooled using a random effects model. The primary outcomes were physiologic parameters of GERD, including normalization of esophageal pH and augmentation of lower esophageal sphincter pressure. Secondary outcomes included frequency of the use of PPIs and HRQL.

The analysis included 4 trials with a total of 168 patients. Three trials compared the Stretta system vs sham therapy, and 1 trial compared the Stretta system vs PPI therapy. Pooled results yielded no difference between Stretta treatment and sham or PPI therapy for all outcomes.

The studies had several methodologic shortcomings, including failure to provide details on blinding to treatment and failure to report outcomes data. None of the studies reported the proportion of patients who experienced a complete alleviation of GERD symptoms, normalization of esophageal pH, or augmentation of lower esophageal sphincter pressure. The authors concluded that future studies are needed to characterize the physiologic mechanisms invoked by the Stretta device, and that a high-quality randomized controlled trial is still needed to evaluate the efficacy of this treatment.


Transoral Esophagogastric Fundoplication Effectively Treats GERD Symptoms That Persist Despite PPI Therapy 

For some GERD patients with persistent symptoms despite PPI treatment, transoral esophagogastric fundoplication can decrease or eliminate symptoms. At the ACG meeting, Peter Kahrilas, MD, of Northwestern University in Chicago, Illinois presented findings from a study evaluating the efficacy of transoral esophagogastric fundoplication vs PPIs in controlling regurgitation in patients with well-documented GERD (Abstract 41).

The RESPECT (Randomized EsophyX Versus Sham, Placebo-Controlled Transoral Fundoplication) trial was conducted at 8 academic and community medical centers across the United States. After screening 696 patients for GERD symptoms, the trial enrolled 129 patients with proven GERD and a hiatal hernia of 2 cm or less. Patients were randomized 2:1 to receive either transoral esophagogastric fundoplication and 6 months of placebo, or sham surgery and 6 months of omeprazole (40 mg once or twice daily). Assessments were performed at 2, 12, and 26 weeks. The primary study endpoint was the elimination of troublesome regurgitation.

After 6 months, 54 of 80 patients (68%) receiving transoral esophagogastric fundoplication plus placebo reported elimination of troublesome regurgitation compared with 17 of 37 patients (46%) receiving the sham treatment plus PPI (P=.041). Transoral esophagogastric fundoplication was associated with decreased intraesophageal acid exposure by all parameters measured (P<.001). No improvement in pH was observed in the sham surgery patients. Dysphagia and bloating improved in both groups. Adverse events (AEs) were similar in both arms, with the exception of postoperative epigastric pain and early treatment failure, which were more common in patients randomized to sham surgery.


GERD Symptoms Are More Likely in Patients With Increased Abdominal Obesity Despite Normal Body Mass Index

At the ACG meeting, Shahid Karim, MBBS, of the Liaquat National Hospital and Medical College in Karachi, Sindh, Pakistan described results of a study evaluating whether GERD symptoms correlate to increased abdominal obesity despite a normal body mass index in a multiethnic South Asian population (Abstract P623). The prospective, cross-sectional, multicenter study was conducted from February 2009 to March 2010. The study enrolled nonsmoking, nonalcoholic patients with a normal body mass index, defined as 18.5 to 22.9 kg/m2 for Asians. Study subjects completed a validated questionnaire to assess the presence of GERD symptoms. Abdominal obesity was defined as waist circumference of at least 90 cm and waist:hip ratio greater than 0.90. Patient data were categorized based on a waist size of 79 cm or less (group A), 80 cm to 90 cm (group B), and 90 cm or greater (group C), and by a waist:hip ratio of 0.90 or less (group 1) vs greater than 0.90 (group 2).

The study included 1260 subjects with a mean age of 36.33 years. GERD symptoms were present in 42.2%. The waist circumference groups A, B, and C included 477, 349, and 434 patients, respectively. The proportion of patients with GERD symptoms rose with increasing waist circumference. GERD symptoms were reported by 30.2% of group A, 37.3% of group B, and 59.5% of group C. The increases were significant between group A vs group B (odds ratio [OR], 1.37; 95% CI, 1.03-1.84; P=.034) and group A vs group C (OR, 3.39; 95% CI, 2.56-4.45; P=.001). Study subjects with the higher waist:hip ratio had significantly more GERD symptoms (OR, 2.03; 95% CI, 1.61-2.56; P=.001). GERD symptoms were present in 32.9% of group 1 and 49.9% of group 2.


A Simple Scoring System to Distinguish Functional Dyspepsia From GERD 

The majority of patients who present with epigastric pain have functional dyspepsia and not GERD. However, a misdiagnosis of GERD is common among these patients. Currently, a GERD diagnosis requires the finding of esophagitis based on upper endoscopy or a positive result from a 24-hour pH study. A simpler means for distinguishing between functional dyspepsia and GERD is needed. At the ACG meeting, Neil Marya, MD, of the University of Massachusetts at Worcester described a new scoring system, known as GERDYS, which is based on clinical symptoms to help providers distinguish between functional dyspepsia and GERD (Abstract P1230).

The study retrospectively identified 34 consecutive patients who underwent 24-hour pH monitoring for evaluation of epigastric discomfort. The GERDYS score ranged from -3 to 3. Patient scores increased by 1 point each for ascending chest pain, intermittent symptoms, or nocturnal waking. Scores were reduced by 1 point each for the presence of continuous symptoms, nausea, or bloating. Patients were separated into cohorts based on DeMeester scores of 14.7 or higher for the GERD cohort and scores below 14.7 for the functional dyspepsia cohort.

Based on the DeMeester score, 12 patients were diagnosed with functional dyspepsia and 22 with GERD. GERDYS scores ranged from -1 to 0 for the functional dyspepsia group and from 0 to 3 for the GERD group. A 1-way analysis of variance revealed a significant difference in patient DeMeester scores by GERDYS scores (P=.04). Chi-square analysis demonstrated a significant difference in GERDYS scores for the functional dyspepsia cohort vs the GERD cohort (chi-square value, 14.82; P=.005).

The GERDYS score represents the first attempt to provide a quantitative scoring system to distinguish functional dyspepsia from GERD based on clinical criteria. Prospective validation of the GERDYS scoring system could lead to improved identification and management of functional dyspepsia and GERD.


Topical Corticosteroids Improve Histologic But Not Clinical Symptoms in Patients With Eosinophilic Esophagitis 

Eosinophilic esophagitis has gained recognition as a clinicopathologic condition characterized by esophageal dysfunction. It is associated with an eosinophil-predominant inflammation of the esophageal mucosa that does not respond to PPI treatment. Topical corticosteroids have shown efficacy in eosinophilic esophagitis and are recommended as a first-line therapy. At the ACG meeting, Ashutosh Gupta, MD, of the John H. Stroger Hospital of Cook County in Chicago, Illinois presented results of a systematic review and meta-analysis investigating the efficacy of topical corticosteroids in treating eosinophilic esophagitis (Abstract 37).

A systematic search of databases from MEDLINE, PubMed, SCOPUS, and the Cochrane library was conducted to identify studies investigating the efficacy of oral viscous budesonide or fluticasone in inducing histologic and clinical remission in children and adults with eosinophilic esophagitis. Only randomized, placebo-controlled trials were included.

The analysis identified 5 studies including 161 patients with a clinical and histologic diagnosis of eosinophilic esophagitis. Fluticasone was administered in 3 studies (n=101), and oral viscous budesonide was administered in 2 studies (n=60). Compared with placebo, topical corticosteroids were associated with a higher rate of complete histologic remission (OR, 20.81; 95% CI, 7.03-61.63) as well as partial histologic remission (OR, 32.20; 95% CI, 6.82-152.04). Corticosteroids were associated with a nonsignificant improvement over placebo for clinical symptoms—defined as an improvement in dysphagia or composite scores of upper gastrointestinal (GI) symptoms (OR, 2.72; 95% CI, 0.90-8.23).


Presentations in IBS

Rifaximin Treatment Can Be Successfully Repeated to Treat Patients With IBS-D Symptoms


Final data from the TARGET 3 (Targeted, Nonsystemic Antibiotic Rifaximin Gut-Selective Evaluation of Treatment for Non-C IBS) study demonstrated for the first time that rifaximin treatment can be repeated to successfully re-treat patients with recurrent diarrhea-predominant irritable bowel syndrome (IBS-D). Findings from the TARGET 3 trial were presented by Anthony Lembo, MD, of Beth Israel Deaconess Medical Center in Boston, Massachusetts at the ACG 2014 Annual Scientific Meeting (Abstract 45).

The study enrolled IBS-D patients who met the Rome III criteria. All patients were experiencing IBS-related abdominal pain and bloating. During the 7-day baseline period, the patients had at least 2 bowel movements that corresponded to type 6 or 7 on the Bristol Stool Scale. During the initial 14-day open-label phase, patients were treated with rifaximin (550 mg 3 times daily), followed by a 4-week treatment-free follow-up period to assess response. The composite primary endpoint, as required by the US Food and Drug Administration (FDA), was a decrease from baseline of at least 30% in the mean abdominal pain score and a decrease from baseline of at least 50% in the number of days per week that stool consistency can be categorized as type 6 or 7 on the Bristol Stool Scale.

Of the 2579 patients enrolled during the open-label phase, 1074 (42%) responded. Recurrent IBS-D symptoms were reported in 692 of these responding patients (64%); 636 were randomized to receive rifaximin retreatment (n=328) or placebo (n=308) for 14 days followed by a 4-week treatment-free period. Rifaximin induced significantly more responses compared with placebo (33% vs 25%; P=.02). During the second double-blind retreatment phase, rifaximin was again associated with a significantly greater proportion of responders (37% vs 29%; P=.04). The key secondary endpoint of prevention of recurrence also favored rifaximin (13.2% vs 7.1%; P=.0068). AEs were similar in the 2 treatment groups.


Plecanatide Demonstrates Efficacy at Tolerable Doses in Patients With IBS-C 

Plecanatide is a minimally absorbed peptide of uroguanylin, the ligand for the human intestinal guanylate cyclase-C receptor. At the ACG meeting, Philip Miner, Jr, MD, of the Oklahoma Foundation for Digestive Research in Oklahoma City, Oklahoma presented results from a dose-finding clinical trial that assessed the safety and efficacy of plecanatide (Abstract 14).

The trial was a multicenter, double-blind, placebo-controlled, parallel-group study of 424 patients with constipation-predominant IBS (IBS-C). Patients were randomized to receive placebo or oral plecanatide dosed at 0.3 mg, 1.0 mg, 3.0 mg, or 9.0 mg once daily for 12 weeks. The primary efficacy endpoint was the change from baseline in complete spontaneous bowel movements. Key secondary efficacy endpoints included the change from baseline in worst abdominal pain intensity; the FDA’s overall responder endpoint for IBS-C; and the change in stool consistency using the Bristol Stool Scale.

All but the lowest dose of plecanatide yielded improvement in the weekly frequency of complete spontaneous bowel movements compared with placebo, with increases of 2.12, 2.74, and 2.44 for plecanatide doses of 1.0 mg, 3.0 mg, and 9.0 mg, respectively (P≤.05 for each comparison). Plecanatide dosed at 3.0 mg daily significantly improved the secondary endpoints of change from baseline of worst abdominal pain intensity, FDA overall responder endpoint, and stool consistency and straining. The most common AE was diarrhea, which was observed in 9.4%, 9.3%, and 11.8% of patients at the plecanatide doses of 1.0 mg, 3.0 mg, and 9.0 mg, respectively.


Urgency as a Measure of Eluxadoline Treatment Effect 

Eluxadoline is a locally acting µ-opioid receptor agonist and δ-opioid receptor antagonist that has been shown to improve symptoms associated with IBS-D in 2 randomized, double-blind, phase 3 clinical trials. These trials showed that eluxadoline yielded higher responder rates with concomitant improvements in stool consistency and pain at weeks 12 and 26 compared with placebo. At the ACG 2014 meeting, Anthony Lembo, MD, of Beth Israel Deaconess Medical Center in Boston, Massachusetts presented results of a post hoc study of phase 3 data, which evaluated urgency-free days as reported in patient diaries to assess treatment effect (Abstract 13).

The study pooled data from 2324 patients diagnosed with IBS-D from the phase 3 trials of eluxadoline. During the primary treatment period of 26 weeks, patients completed daily diaries to record IBS-D symptoms. Patients were considered responders if they were free of urgency on at least half of the days. Cumulative distribution functions for urgency-free days showed differentiation between eluxadoline (75 mg or 100 mg) vs placebo for weeks 1 to 12 and weeks 1 to 26. At the median and 75th percentile of the populations, treatment with eluxadoline showed a 16% to 18% increase in days without urgency. Moderate correlations were observed between the definition of responder used in the study and the FDA’s responder endpoint (for 50% of trial weeks, patients report ≥30% decrease in abdominal pain at its worst and, in the same week, an increase in complete spontaneous bowel movements of ≥1 from baseline). The measurement of urgency-free days may provide a valuable addition to the assessment of IBS-D severity and treatment outcome. Eluxadoline demonstrated a significant reduction in urgency over placebo.


Social Stress and Sex Differences in IBS 

Approximately two-thirds of IBS patients are female. At the ACG 2014 meeting, Elyse Thakur, MA, of Wayne State University in Detroit, Michigan presented results of a study examining whether personal stress differed in female vs male IBS patients (Abstract P506).

The study included 284 patients with Rome III IBS, 80% of whom were female. The patients’ median age was 41 years. All patients completed tests assessing social support (Interpersonal Support Evaluation List), interpersonal problems (Inventory of Interpersonal Problems), negative interactions (Negative Interaction Scale), and IBS symptom severity during the baseline phase of a trial conducted by the National Institutes of Health.

Overall, interpersonal distress was similar for male and female IBS patients. In the test assessing personal problems, men were more likely to report concerns with fighting with other people and keeping other people at a distance, which reflects a hostile-dominant interpersonal pattern. Male patients with vindictive and/or self-centered interpersonal problems reported less support and more negative interactions than female patients. Male patients generally reported less social support. The quality of relationship problems correlated with IBS symptom severity as measured by gastroenterologists but not patients. The increased interpersonal difficulties in male IBS patients may influence estimations of symptoms and impact the doctor-patient relationship.


Polyethylene Glycol Vs Antibiotics for Treating IBS-C Patients With a Positive Lactulose Breath Test

Although antibiotics are commonly prescribed for treating IBS-C, polyethylene glycol has demonstrated efficacy for treating IBS-C and represents an attractive alternative, due to its low cost and favorable side effect profile. At the ACG 2014 meeting, Bingru Xie, MD, of the University of Medicine and Dentistry in Newark, New Jersey presented findings of a study comparing polyethylene glycol vs antibiotics for the treatment of patients with IBS-C as confirmed by a lactulose breath test (Abstract P494).

A review of medical records from 2011 to 2013 at the authors’ GI treatment center yielded 36 patients with an IBS-C diagnosis based on the Rome III criteria and positive results on a lactulose breath test. Ten patients received polyethylene glycol, and 12 patients received antibiotics for 14 days. The lactulose breath test was used to evaluate levels of CH4, H2, and CO2 at baseline, and again at 60 minutes and 120 minutes after ingestion of lactulose. The test was administered before treatment and 2 weeks after.

Before polyethylene glycol treatment, the lactulose breath test showed a mean H2 increase of 17.6 ppm and a mean CH4 increase of 25.4 ppm at 60 minutes. After treatment with polyethylene glycol, the test showed a mean H2 increase of 9.2 ppm and a mean CH4 decrease of 24 ppm at 60 minutes. Lactulose breath testing before antibiotic treatment showed a mean H2 increase of 12 ppm and a mean CH4 increase of 33 ppm at 60 minutes. After antibiotic treatment, the test showed a mean H2 increase of 8.4 ppm and a mean CH4 decrease of 1.03 ppm at 60 minutes. Therefore, the mean H2 level and the mean CH4 level were reduced from baseline in both groups after treatment. When the differences in levels of H2 and CH4 before and after treatment were compared for the 2 treatment groups, polyethylene glycol yielded the greater improvement for CH4 levels (P<.05). Reductions in H2 levels, however, were similar (P>.05).


Presentations in Hepatology

Liver Function Improves in Cirrhotic HCV Patients Treated With a 3-Drug Antiviral Combination Plus Ribavirin

Patients infected with hepatitis C virus (HCV) are at increased risk for hepatocellular carcinoma and liver-related mortality. However, the risk is significantly reduced in patients who achieve a sustained virologic response (SVR) to treatment. At the ACG meeting,

Priyam Tripathi, MD, of Case Western Reserve University in Cleveland, Ohio presented results of a study examining hepatic function in HCV patients treated with antiviral agents (Abstract 5). The phase 3 TURQUOISE-II (A Study to Evaluate the Safety and Effect of ABT-450, Ritonavir and ABT-267 [ABT-450/r/ABT-267] and ABT-333 Coadministered With Ribavirin [RBV] in Hepatitis C Virus [HCV] Genotype 1-infected Adults With Compensated Cirrhosis) study evenly randomized 380 patients with Child-Turcotte-Pugh type A cirrhosis to receive the 3-drug combination of ABT-450/ritonavir, ombitasvir, and dasabuvir plus ribavirin (3D plus RBV) for 12 or 24 weeks of treatment. The study included both treatment-naive and treatment-experienced patients. Prior use of pegylated interferon and ribavirin was permitted. All patients had HCV genotype 1 infection and cirrhosis. Laboratory testing, including chemistry, hematology, and urinalysis, was conducted at each study visit during the treatment and afterward.

Treatment for 12 or 24 weeks resulted in SVR rates at week 12 of 92% and 96%, respectively. In most patients, liver enzymes normalized by the end of the 12- or 24-week treatment period, with normalization of alanine transaminase, aspartate transaminase, and gamma-glutamyl transpeptidase in 93.1%, 87.8%, and 92.5% of patients, respectively. Mean liver enzyme values normalized by week 4. Liver function was largely restored after treatment, as demonstrated by the normalization of conjugated bilirubin levels, albumin levels, and prothrombin time. Platelet counts increased in the overall population, consistent with a possible improvement in portal hypertension.


Ledipasvir/Sofosbuvir Plus Ribavirin for HCV Patients With Decompensated Cirrhosis

At the American Association for the Study of Liver Diseases (AASLD) Liver Meeting, Steven Flamm, MD, of the Northwestern University Feinberg School of Medicine in Chicago, Illinois presented preliminary results of a prospective, multicenter study evaluating ledipasvir/sofosbuvir plus ribavirin in HCV patients with decompensated cirrhosis (Abstract 239). The combination of ledipasvir (90 mg/day), sofosbuvir (400 mg/day), and ribavirin (initially 600 mg/day, then escalated) was administered for 12 or 24 weeks. The study included 108 adults with HCV genotype 1 or 4. Patients had a Child-Turcotte-Pugh score of B (n=59) or C (n=49). Six patients were excluded from the analysis because they underwent liver transplantation during the course of the study. SVR12 rates were 87% for 12 weeks of treatment and 89% for 24 weeks of treatment. Among patients with a Child-Turcotte-Pugh score of B, there were 4 relapses and 3 deaths. The group with a Child-Turcotte-Pugh score of C had 2 relapses and 2 deaths (plus 1 patient who was lost to follow-up). The relapse rates were similar to those reported in patients with compensated cirrhosis. The longer treatment duration of 24 weeks did not appear to confer any additional benefit. A virologic response was associated with improvements in bilirubin, albumin, Model for End-Stage Liver Disease scores, and Child-Turcotte-Pugh scores, regardless of the patient’s Child-Turcotte-Pugh score at baseline. The 12-week and 24-week regimens were generally well tolerated, with 4 serious AEs attributed to study treatment and 3 patients discontinuing treatment because of an AE.


An All-Oral Regimen of 3 Direct-Acting Antiviral Agents in HCV Genotype 1 Patients With Cirrhosis

The phase 3 TURQUOISE-II trial evaluated a combination regimen of 3 direct-acting antiviral agents in HCV genotype 1 patients with cirrhosis. Results were reported at the AASLD Liver Meeting by Michael Fried, MD, of the University of North Carolina at Chapel Hill School of Medicine (Abstract 81). The study randomized 380 patients to receive a combination of paritaprevir, ombitasvir, and dasabuvir plus ribavirin for 12 or 24 weeks. Follow-up lasted for 48 weeks after the cessation of treatment. Enrolled patients could be treatment-naive or treatment-experienced; a key exclusion criterion, however, was prior therapy with a direct-acting antiviral agent. The overall SVR12 rates were 91.8% after 12 weeks of treatment and 96.5% after 24 weeks of treatment. Analysis of baseline demographic, clinical, and virologic factors did not identify significant differences in SVR12 rates for most comparisons (Figure 1). SVR12 rates were higher with the 24-week regimen than the 12-week regimen in patients who were prior null responders (95.2% vs 86.7%), treatment-naive (95.9% vs 94.2%), genotype 1a (95.0% vs 88.6%), or genotype 1b (100% vs 98.5%).

The study also examined the demographic and disease characteristics of the patients who did not achieve SVR12. In the 12-week treatment arm, 8% of patients failed to achieve SVR12, and 6% of patients relapsed. In the 24-week arm, 3.5% of patients failed to achieve SVR12, and 0.6% of patients relapsed. All but 1 of the patients who relapsed had HCV genotype 1a, and nearly all of the relapsed patients had HCV RNA levels of 800,000 IU/mL or higher. Three factors emerged that were significantly associated with reduced rates of SVR12: interleukin 28B genotype TT (P=.021), prior null response (P=.038), and HCV genotype 1a infection (P=.046).


Sofosbuvir and Simeprevir for the Treatment of Recurrent HCV After Liver Transplantation

Recurrence of HCV after liver transplantation has been associated with increased rates of fibrosis. Among these patients, traditional therapies have been associated with low SVR rates and significant AEs. A study reported at the AASLD Liver Meeting by Heather O’Dell, ANP-BC, from the Vanderbilt University Medical Center in Nashville, Tennessee evaluated the use of sofosbuvir and simeprevir to treat recurrent HCV after liver transplantation (Abstract LB-8).

The 18 patients in this study had undergone transplantation at least 3 months before the study start and had documented recurrence of HCV infection. The patient’s mean age was 61 years, 78% of patients were male, and 3 patients had cirrhosis. Patients received treatment with sofosbuvir and simeprevir while continuing to receive standard immunosuppressant therapy, which consisted of tacrolimus in 89% and cyclosporine in 11%. All patients completed 12 continuous weeks of therapy. No patients required adjustments to the immunosuppressant dose or experienced transplant rejection.

The rapid virologic response rate was 72%, and all patients achieved an end-of-treatment response. The only reported AEs were mild headache and nausea. There were no interactions that required adjustment of the treatment dosage. Among the 15 patients with HCV RNA at 4 weeks posttreatment, the SVR4 rate was 100%. For the 7 evaluable patients with data at 12 weeks posttreatment, the SVR12 rate was also 100%. The results from this small study support further investigation of new direct-acting antiviral therapies in the liver transplant population.


Entecavir and Tenofovir Combination Therapy for Chronic Hepatitis B

At the AASLD Liver Meeting, Fabien Zoulim, MD, PhD, from Lyon University in Lyon, France described results from the ENTEBE (Safety and Efficacy of Entecavir Plus Tenofovir in Adults With Chronic Hepatitis B and Previous Nucleos(t)ide Treatment Failure) study, a single-arm, open-label, multicenter study evaluating a combination of entecavir and tenofovir as rescue therapy in patients with chronic hepatitis B who had failed prior treatment with a nucleos(t)ide therapy (Abstract 230). Entecavir and tenofovir are both potent agents with resistance profiles that do not overlap.

A regimen of entecavir (1 mg) plus tenofovir (300 mg) was administered for 96 weeks to 92 patients (6 discontinued). Prior nucleos(t)ide therapy included monotherapy with entecavir (53%), lamivudine (22%), tenofovir (12%), adefovir (4%), and telbivudine (2%), or combinations of these agents (7%). More than half of patients (58%) had evidence of single-drug or multidrug resistance mutations.

The primary endpoint—hepatitis B DNA level of less than 50 IU/mL—was achieved by 76% of patients at week 48 and 85% at week 96 (Table 2). The primary endpoint at week 96 was achieved by 100% of patients who had failed adefovir or telbivudine, 88% who had failed entecavir, 82% who had failed tenofovir, 80% who had failed lamivudine, and 83% who had failed combination therapy. Six patients experienced on-treatment serious AEs, none of which were considered related to study treatment. One patient died from hepatocellular carcinoma.


Cognitive Changes in Hepatic Encephalopathy May Not Be Reversible

Results from a multicenter study evaluating the persistence of cognitive impairment in hepatic encephalopathy patients were reported at the AASLD Liver Meeting by Jasmohan Bajaj, MD, of Virginia Commonwealth University in Richmond, Virginia (Abstract 94). This international study included 187 outpatients with cirrhosis from 3 different medical centers (in Virginia, Ohio, and Rome). Cognitive testing included assessment of the psychometric hepatic encephalopathy score and the inhibitory control test. The psychometric hepatic encephalopathy score is based on 6 subtests. The inhibitory control test consists of 2 identical halves that are given one after the other (with a short break in between). Subjects with an intact learning ability should show improvement in the second half as compared with the first half.

At baseline, results of all cognitive tests were worse among patients with hepatic encephalopathy. Patients without hepatic encephalopathy showed significant improvement on the second half of the inhibitory control test as compared with the first half. Patients with hepatic encephalopathy did not improve on the second half. These results were replicated when the patients underwent subsequent testing (performed a median of 20 days later). In addition, subsequent assessment of the psychometric hepatic encephalopathy score showed that patients without hepatic encephalopathy improved on 4 subtests compared with the first assessment, whereas patients with hepatic encephalopathy improved on 2 subtests. Despite adequate medical therapy, patients with prior hepatic encephalopathy showed persistent significant learning impairment compared to those without. The authors concluded that the recognition of these continued cognitive deficits should increase efforts to avoid an initial episode of hepatic encephalopathy and perhaps increase the transplant listing priority for patients with this condition.


Idiosyncratic Drug-Induced Liver Injury

At the ACG meeting, Naga Chalasani, MD, of Indiana University in Indianapolis, Indiana presented results of a prospective study enrolling patients with idiosyncratic drug-induced liver injury in the United States (Abstract 31). Patients with suspected drug-induced liver injury were enrolled prospectively and followed for at least 6 months. Among the 1257 subjects enrolled, causality adjudication was completed in 1091. A diagnosis of drug-induced liver injury was deemed definite in 235 patients, highly likely in 466 patients, and probable in 198 patients. The liver injury was hepatocellular in 54% of these patients, cholestatic in 23%, and mixed in 23%. Approximately 10% of patients died or underwent liver transplantation, and 17.5% developed chronic drug-induced liver injury.

Most cases (86%) of drug-induced liver injury could be attributed to 5 classes of agents: antimicrobials (n=408), herbal and dietary supplements (n=145), cardiovascular agents (n=88), central nervous system agents (n=82), and antineoplastic agents (n=49). The individual agents associated with the most cases were amoxicillin-clavulanate (n=91), isoniazid (n=48), nitrofurantoin (n=42), trimethoprim/sulfamethoxazole (n=31), minocycline (n=28), cefazolin (n=20), azithromycin (n=18), ciprofloxacin (n=16), levofloxacin (n=13), and diclofenac (n=13). The duration of disease latency did not impact outcome. Patients ages 65 years or older had higher rates of drug-induced liver injury than younger patients. Rates of mortality and liver transplantation did not differ according to age. Nine patients developed severe cutaneous reactions (Stevens-Johnson syndrome), which were associated with lamotrigine (n=2), azithromycin (n=2), moxifloxacin (n=1), nitrofurantoin (n=1), diclofenac (n=1), carbamazepine (n=1), and cephalexin/lamotrigine (n=1). Four of these patients died.


Grazoprevir and Elbasvir With or Without Ribavirin in HCV Genotype 1

The randomized, open-label, phase 2 C-WORTHY trial was designed to examine the efficacy and safety of grazoprevir and elbasvir with or without ribavirin in patients with HCV genotype 1 infection. In part A of the C-WORTHY trial, this regimen achieved SVR12 rates of 89% to 100% in 65 treatment-naive, noncirrhotic, HCV genotype 1–infected patients. At the AASLD Liver Meeting, Eric Lawitz, MD, of the University of Texas Health Science Center in San Antonio reported the final results from the subset of treatment-naive, cirrhotic patients and prior null responders enrolled in part B of the study (Abstract 196).

Patients received 12 or 18 weeks of grazoprevir (100 mg/day) plus elbasvir (50 mg/day) with or without ribavirin (dosed according to body weight). In the treatment-naive patients who received 12 weeks of therapy, intent-to-treat analysis yielded SVR12 rates of 90% with ribavirin and 97% without. Among patients who received 18 weeks of treatment, SVR12 rates were 97% with ribavirin vs 94% without. Among the prior null responders, 12 weeks of treatment yielded SVR12 rates of 94% with ribavirin vs 91% without. Eighteen weeks of treatment yielded rates of 100% with ribavirin vs 97% without ribavirin in null responders. When data for the 12-week and 18-week treatment groups were pooled, SVR12 rates were 95% with ribavirin vs 94% without; 93% for HCV genotype 1a patients vs 99% for genotype 1b; 94% for treatment-naive patients vs 95% for null responders; and 95% for patients with or without cirrhosis. Among the subset of 25 patients who were prior null responders with cirrhosis, 12 weeks of treatment with or without ribavirin yielded an SVR12 rate of 92%. Seven serious AEs were reported, but 6 were considered unrelated to the study treatment. All treatment-emergent AEs were mild to moderate. The most common events were fatigue (26%), headache (23%), and asthenia (14%).


Presentations in IBD

Low Rates of Infection With Vedolizumab Alone or With Corticosteroids and/or Immunosuppressants in Ulcerative Colitis and Crohn’s Disease Patients

The GEMINI 1 and 2 studies examined the efficacy and safety of vedolizumab for ulcerative colitis and Crohn’s disease, respectively. In both studies, rates of some infections were higher with vedolizumab therapy than with placebo. At the ACG meeting, Edward Loftus, MD, of the Mayo Clinic in Rochester, Minnesota presented results from an analysis of infection rates in patients treated with vedolizumab alone or with concomitant corticosteroids and/or immunosuppressants vs placebo in GEMINI 1 and 2 (Abstract 16).

Data from the 2 studies were pooled from the 6-week induction phase, in which patients received vedolizumab (300 mg) or placebo at weeks 0 and 2, and from the 46-week maintenance phase, in which patients received vedolizumab (300 mg) or placebo every 4 or 8 weeks. The proportions of patients with AEs and serious AEs were determined for those who received the antibody or placebo continuously through the induction and maintenance portions of the study.

For the 1434 pooled GEMINI 1 and 2 patients who received vedolizumab, the rates of infection considered AEs or serious AEs were similar among the subgroups, regardless of the use of concomitant corticosteroids or immunosuppressants. In general, rates of infection-related AEs were similar between the placebo and vedolizumab subgroups, with rates of any infectious AE ranging from 32% to 44% for the 4 placebo subgroups vs 42% to 45% for the 4 vedolizumab subgroups. Nasopharyngitis was more common with vedolizumab, occurring in 10% to 14% of the subgroups (vs 4% to 12% of the subgroups receiving placebo). Rates of infections considered serious AEs were also similar or nominally lower for the placebo subgroups vs the vedolizumab subgroups, but the infrequency of these events limited interpretation of the data.


Meta-Analysis of Cyclosporine Vs Infliximab for Patients With Acute Severe Corticosteroid-Refractory Ulcerative Colitis

Treatment with intravenous corticosteroids fails in up to 40% of patients who present with acute severe ulcerative colitis. The prognosis for these patients is poor. Cyclosporine and infliximab can be used as salvage therapies for these corticosteroid-refractory patients. At the ACG meeting, Edward Loftus, Jr, MD, of the Mount Sinai Hospital in New York, New York, presented results from a systematic review and meta-analysis conducted to assess cyclosporine and infliximab as rescue agents in patients with corticosteroid-refractory ulcerative colitis (Abstract 15).

A literature search identified studies that investigated cyclosporine and infliximab in corticosteroid-refractory ulcerative colitis patients. The primary outcome was short-term response to treatment. Secondary outcomes included the rates of colectomy at 3 months and 12 months, adverse drug reactions, postoperative complications in those who received rescue therapy but subsequently underwent colectomy, and mortality.

Eleven studies with 988 participants were eligible for inclusion. For the 2 randomized controlled trials with 145 patients, no significant difference was seen for infliximab vs cyclosporine based on treatment response, colectomy at 3 months, or colectomy at 12 months. For the 9 nonrandomized studies with 843 eligible participants, infliximab treatment was associated with significantly higher rates of treatment response (OR, 2.99; 95% CI, 2.99-4.30) and lower rates of colectomy at 12 months (OR, 0.38; 95% CI, 0.17-0.85). There was no significant difference between infliximab and cyclosporine in the rates of colectomy at 3 months (OR, 0.71; 95% CI, 0.26-1.89). No significant differences emerged between cyclosporine and infliximab in terms of drug-related AEs, postoperative complications, or mortality.


Increased Vedolizumab Dosing Frequency in Ulcerative Colitis and Crohn’s Disease Patients Who Have Lost Response

The GEMINI 1 and 2 trials demonstrated the efficacy and safety of vedolizumab in patients with ulcerative colitis and Crohn’s disease, respectively. At the 2014 Advances in Inflammatory Bowel Diseases conference, Bruce Sands, MD, of the Icahn School of Medicine at Mount Sinai in New York, New York presented results of GEMINI-LTS, an open-label, long-term extension study evaluating an increased frequency of vedolizumab in patients who lost response to the drug during the maintenance phase of GEMINI 1 or 2 (Abstract P098).

The current extension study included patients from GEMINI 1 and 2 who responded to vedolizumab during the 6-week induction phase (300 mg at weeks 0 and 2) but subsequently lost response during the 46-week maintenance phase (300 mg every 8 weeks) and discontinued treatment. These patients accounted for 26% of the population in GEMINI 1 and 37% in GEMINI 2.

For the extension study, the frequency of vedolizumab dosing was increased to 300 mg every 4 weeks. This dosage improved the mean disease activity scores for both ulcerative colitis and Crohn’s disease patients. Rates of clinical remission and response also improved. For the GEMINI 1 cohort, the proportion of patients in clinical remission was 6.3% at baseline, 25% at week 28, and 25% at week 52. The rate of clinical response increased from 18.8% at baseline to 53.1% at week 28, but then decreased to 37.5% at week 52. For the GEMINI 2 cohort, clinical remission rates were 3.5% at baseline, 22.8% at week 28, and 31.6% at week 52. The clinical response rates were 38.6% at baseline, 54.4% at week 28, and 47.4% at week 52. AE profiles were similar for vedolizumab dosed at 300 mg every 8 weeks vs every 4 weeks.


Immunosuppressant Treatment Does Not Increase Cancer Incidence in IBD Patients With a History of Cancer

Nearly 30% of inflammatory bowel disease (IBD) patients with a history of cancer develop a secondary or recurrent cancer. At the 2014 Advances in Inflammatory Bowel Diseases conference, Jordan Axelrad, MD, of the Icahn School of Medicine at Mount Sinai in New York, New York described findings from a retrospective analysis showing no correlation between exposure to IBD treatments and the likelihood of developing cancer (Abstract O005).

The retrospective study included 185 patients with IBD and a history of cancer from 3 different institutions. Among these patients, 65 (35%) had received antitumor necrosis factor α (TNFα) therapy, 46 (25%) had received antimetabolites, including thiopurines or methotrexate, and 74 (40%) had not received immunosuppressants. The primary outcome was the development of new or recurrent cancer.

No significant differences emerged in the development of cancer for the 3 cohorts. Approximately 14% of patients developed a new cancer, 12% developed a recurrent cancer, and 3% developed both a new and a recurrent cancer. More skin cancers were observed in the anti-TNFα group and more GI cancers occurred in the control group, but neither difference was statistically significant. Incident cancer rates were 3.9 with 361 person-years of follow-up for the control group; 6.6 with 181 person-years of follow-up for the antimetabolites group; and 8.8 with 306 person-years of follow-up for the anti-TNFα group. After 5 years of follow-up, there were no significant differences in cancer-free survival rates among the 3 treatment groups. The study was limited by the population size and the lack of data on dose-related effects and periods of cancer remission.


Baseline 5-Aminosalicylic Acid Use Is Compatible With Budesonide Foam Treatment

Budesonide foam is a rectally administered, second-generation corticosteroid. It was developed to treat distal forms of ulcerative colitis with optimal drug retention and uniform drug delivery. At the ACG meeting, William Sandborn, MD, of the University of California San Diego in La Jolla, California described results from 2 phase 3 studies that evaluated the impact of baseline oral 5-aminosalicylic acid (5-ASA) on the safety and efficacy of budesonide foam (Abstract P470).

Data were pooled from 2 identical multicenter, double-blind, placebo-controlled, phase 3 studies. Both studies included patients with mild-to-moderate active ulcerative proctitis or ulcerative proctosigmoiditis. Patients were randomized to receive budesonide foam or placebo. The dose of budesonide foam was 2 mg in 25 mL; it was given twice daily for 2 weeks followed by once daily for 4 weeks. Patients could receive concomitant treatment with oral 5-ASA (up to 4.8 g daily). Use of rectal 5-ASA was not permitted.

Efficacy was evaluated at week 6. The pooled studies yielded 267 patients in the budesonide foam treatment group and 279 patients in the placebo group. Baseline use of 5-ASA was reported in 147 patients (55.1%) receiving budesonide foam and 154 patients (55.2%) receiving placebo. The percentages of patients achieving remission and a rectal bleed score of 0 were significantly higher with budesonide foam as compared with placebo, regardless of the use of 5-ASA (Table 3). No significant differences in AEs emerged for patients who did or did not report 5-ASA use at baseline.


Presentations in Endoscopy

Retroflexion Vs Forward View for Detecting Adenomas in the Right Colon

Colonoscopy is effectively used to screen for distal colon cancers; however, its ability to detect right-sided colon cancers is unclear. Some studies have suggested that retroflexion in the right colon may improve adenoma detection. At the ACG meeting, Vladimir Kushnir, MD, of the Washington University School of Medicine in

St. Louis, Missouri presented results from a study evaluating whether a second withdrawal from the right colon in retroflexion vs a forward view can improve detection of colonic adenomas (Abstract 8).

The randomized controlled trial enrolled patients at 2 centers. All patients underwent cecal intubation; the colonoscope was then withdrawn to the hepatic flexure, and all visible polyps were removed. Endoscopist confidence in the quality of the first examination of the right colon was recorded on a 5-point Likert scale. After reintubation of the cecum, patients underwent a second examination of the proximal colon and were randomized to either forward or retroflexion view. The primary outcome was the rate of adenoma detection per patient.

The 850 patients had a mean age of 59.1±8.3 years, and 59% were female. Randomization assigned 400 patients (47%) to forward view and 450 (53%) to retroflexion view. Retroflexion was successfully performed in 421 patients (93.5%) in the latter arm. In both groups, 46% to 47% of patients had at least 1 adenoma (P=.69), and the mean number of adenomas per patient was 1 (P=.69). The proportion of patients with at least 1 additional adenoma detected on the second withdrawal from the proximal colon was also similar in the forward view and retroflexion view groups (10.5% and 7.5%, respectively; P=.13). Logistic regression analysis revealed the following significant predictors that adenomas would be identified on the second withdrawal from the right colon: older age (OR, 1.04; 95% CI, 1.01-1.08), adenomas seen on initial withdrawal (OR, 2.8; 95% CI, 1.7-4.7), and low endoscopist confidence in quality of the first examination of the right colon (OR, 4.8; 95% CI, 1.9-12.1). No AEs were observed.


Optimal Timing of Endoscopy After Acute Caustic Ingestion

At the ACG meeting, Munish Ashat, MD, of the Postgraduate Institute of Medical Education and Research in Panchkula, Haryana, India presented results from a study assessing the utility of upper GI endoscopy on day 1 vs day 5 in predicting cicatrization and other outcomes after caustic ingestion (Abstract 20). The study included consecutive patients admitted for ingesting a caustic substance. Upper GI mucosal changes were graded according to the Zargar classification, and these changes were classified as mild for grade IIa or less and severe for grade IIb or higher. Endoscopy changes on day 1 vs day 5 were evaluated for the development of cicatrization and complications.

Among the 63 consecutive patients who presented within 24 hours of caustic ingestion, 51 underwent upper GI endoscopy on both day 1 and day 5 and were included in the study (Table 4). Patients had a mean age of 32±13.3 years, and 61% were male. The caustic substance was acid in 43 patients (84.3%), alkali in 6 (11.8%), and unknown in 2 (3.9%). Antropyloric stricture developed in 18 patients (35.3%), and esophageal stricture developed in 12 (23.5%). One patient (2%) died, 1 (2%) required emergency surgery, and 7 (13.7%) required definitive surgery for cicatrization. Conservative management consisting of only dilatation in those with cicatrization led to recovery in 42 patients (82.3%).

Esophageal grading on day 1 overestimated severity by 23.5% compared with day 5 (P=.008), and stomach grading on day 1 overestimated severity by 29.4% compared with day 5 (P=.006). Stricture formation correlated with endoscopic grading of esophageal injury on day 5 (P=.019) but not day 1 (P=.287). Gastric cicatrization correlated with endoscopic grading on both day 1 and day 5 (P=.005 and P=.000, respectively). Day 5 endoscopic grading correlated with the need for surgery and recovery (P<.05).


Improving Endoscopic Techniques for Complete Resection of Subepithelial Tumors

Medical centers in Asia have pioneered the development of 2 natural orifice endoscopic techniques that achieve R0 en bloc resection of subepithelial tumors originating in the muscularis propria. Submucosal tunnel endoscopic resection uses the submucosal tunnel method to ensure secure closure of the full-thickness defect in the wall of the GI tract. Endoscopic full-thickness resection involves direct resection with closure of the defect by clips or sutures. At the ACG meeting, Stavros Stavropoulos, MD, of the Winthrop University Hospital in Mineola, New York described results using these 2 techniques as an alternative to laparoscopic wedge resection for subepithelial tumors originating in the muscularis propria (Abstract 22).

Procedures were performed between April 2012 and June 2014 at Winthrop University Hospital. This report is the first to describe use of these procedures in the United States. Data were retrieved from a prospectively maintained database. Among the study group, there were 26 endoscopic full-thickness resection procedures and 7 submucosal tunnel endoscopic resection procedures, all performed by a gastroenterologist with extensive experience in similar procedures.

Patients had a mean age of 58 years (range, 18-84 years). According to criteria from the American Society of Anesthesiologists, 12% were class I, 70% were class II, and 18% were class III. The 33 subepithelial tumors were located in the esophagus (6), stomach (22), colon (2), and rectum (3) and included 17 GI stromal tumors, 8 leiomyomas, 2 pancreatic rests, 1 schwannoma, and 1 leiomyosarcoma. Tumors had a mean size of 22 mm (range, 10-55 mm).

Mean resection time was 72 minutes (range, 21-220 minutes). Means of closure included endoclips (30%), endoscopic suturing (49%), and both (21%). Complete en bloc resection was achieved in 91% of procedures. Piecemeal resection was required in 3 patients, 2 with pancreatic rests and 1 with a GI stromal tumor of 5 cm. Notable AEs included 2 cases of needle decompression of the capnoperitoneum and 3 cases of bleeding requiring prolonged endoscopic hemostasis, with 2 of the 3 patients requiring blood transfusion. The mean length of the hospital stay was 1.5 days (range, 1 to 3 days).


The Role of Esophageal Biopsy in Screening At-Risk Patients for Eosinophilic Esophagitis

Eosinophilic esophagitis patients commonly present with dysphagia and food impaction. However, variation in clinical symptoms and endoscopic findings can confound the diagnosis. At the ACG meeting, Kristina Ross, MD, of the University of Colorado Anschutz Medical Campus in Denver, Colorado presented findings from a study that examined the role of esophageal biopsies in screening patients for eosinophilic esophagitis (Abstract P616).

A retrospective chart review from 2001 to 2012 was conducted in a tertiary care, university-affiliated hospital. Included patients were referred for esophagogastroduodenoscopy for either food impaction or dysphagia with documented stricture. Patients with esophageal malignancy, a history of radiation therapy, or esophageal dysmotility disorders were excluded. The study calculated the proportion of patients with esophageal biopsies taken at the time of the initial endoscopy or on subsequent esophagogastroduodenoscopy, and the number of patients with follow-up esophagogastroduodenoscopies.

Ninety-one patients with food impaction were included. Thirteen patients (14%) underwent initial biopsy, and 54 (58%) did not have a follow-up esophagogastroduodenoscopy. Among the 85 patients with esophageal stricture, 33 (39%) underwent an initial biopsy; 29 patients (34%) did not undergo subsequent biopsy. Eosinophilic esophagitis was diagnosed in 13 of the food impaction patients (14%) and 17 of the stricture patients (20%). This rate of eosinophilic esophagitis diagnosis is considerably higher compared with that in the general population, in which prevalence estimates range from 0.05% to 6.5%. These results suggest that esophageal biopsies are currently underutilized, and that consistent use of esophageal biopsy in at-risk populations could improve diagnosis of this condition.


Endoscopic Management of Esophageal Anastomotic Leaks Following Surgery for Cancer

Esophageal anastomotic leaks constitute a major cause of morbidity and mortality after surgery for gastric or esophageal cancer. Endoscopic interventions such as stenting, clipping, and percutaneous endoscopic jejunostomy have been increasingly used to limit the risk of anastomotic leaks in these patients. At the ACG meeting, Eugene Licht, MD, of the Memorial Sloan Kettering Cancer Center in New York, New York presented results of a study that evaluated the efficacy of endoscopic management of esophageal leaks following cancer surgery (Abstract P636).

The study identified 107 patients with anastomotic leaks, and 51 underwent endoscopic management. Patients had a mean age of 61 years, and 78% were male. Procedures included 42 (82%) esophagectomies, 6 (12%) partial or total gastrectomies, and 3 (6%) esophagogastrectomies. Fully covered esophageal stents were placed in 32 patients. The stents remained in place for a mean of 59 days (range, 12-170 days). Stent migration was reported in 17 patients (53%) and was managed with endoscopic revision. Twenty-six patients (81%) treated with stents healed.

Direct percutaneous endoscopic jejunostomy tubes were placed in 41 patients, for a mean duration of 106 days (range, 18-358 days). No complications from percutaneous endoscopic jejunostomy tube placement were observed. The tube placement wound healed in 39 patients (95%). Three patients who received stents also received endoscopic clips. No complications related to clip placement were observed, and all 3 patients healed. Among the 22 patients who received both a stent and a percutaneous endoscopic jejunostomy tube, 21 (95%) healed, by a mean of 105 days (range, 31-337 days).

Among the 51 patients who underwent endoscopic management, 44 (86%) achieved documented anastomotic healing, with a mean healing time of 92 days (range, 3-337 days). Of the remaining 7 patients, 3 required anastomotic revision, 1 required esophageal exclusion, 2 died of multiple surgical complications, and 1 died at another facility with a stent in place.


An Ultrathin Endoscope for Diagnosing Barrett Esophagus

The GIF-XP290N is a new, ultrathin endoscope that supplies a resolving power similar to that of the GIF-H260 at a distance of 3 mm. At the ACG meeting, Takashi Kawai, MD, of the Endoscopy Center, Tokyo Medical University Hospital in Japan described findings from a study that evaluated patients with Barrett esophagus using the ultrathin endoscope (Abstract P1221).

Diagnosis of Barrett esophagus was made according to Japanese guidelines. The lower margin of the lower esophageal palisade vessels was defined as the gastroesophageal junction when the diaphragm was lowered in deep inspiration. A diagnosis of Barrett esophagus was made if the columnar epithelium was present on the oral side of the gastroesophageal junction.

Upper GI screening using an ultrathin endoscope was performed in 135 patients. The patients’ mean age was 63.5±9.7 years, and most patients were male. Both white light and narrow band imaging were used for all examinations. Barrett esophagus was classified as long (>30 mm), short (10-30 mm), or ultrashort (<10 mm). The Goda classification system was used to categorize the mucosal pattern.

Barrett esophagus was confirmed in 116 of 135 patients (86%) and included 17 cases (15%) of short segments and 99 cases (85%) of ultrashort segments. Narrow band imaging to assess the mucosal structural pattern revealed cases that were villous (41%), oval or round (25%), long straight (25%), cerebriform (7%), and irregular (5%). Histologic examination showed that 8 patients (7%) had intestinal metaplasia. Analysis of the relationship between mucosal patterns and background factors revealed a significant correlation between the presence of intestinal metaplasia and the combined “open type” cohort of villous, cerebriform, and irregular patterns.


Comparison of Cold Biopsy and Other Techniques for Removal of Diminutive Colonic Polyps

Colonoscopic polypectomy is an effective technique for preventing colon cancer. Cold biopsy is the current standard practice for removal of diminutive polyps, the type most often found during colonoscopy. At the ACG meeting, Priyam Tripathi, MD, of Case Western Reserve University in Cleveland, Ohio performed a systematic review and meta-analysis to evaluate the efficacy of cold biopsy vs other techniques for eradicating diminutive polyps (Abstract 24).

The study identified published reports of randomized controlled trials available through MEDLINE, Web of Science, and EMBASE and through abstracts presented at meetings of the ACG, the American Gastroenterological Association, and Digestive Disease Week. The primary outcome was the complete eradication rate of diminutive polyps, and the secondary outcome was procedural time.

The analysis included 5 randomized controlled trials (N=610). Mean polyp size was 4.5 mm (range, 2-10 mm). Removal techniques included cold biopsy, jumbo forceps biopsy, and cold snare polypectomy. The rate of incomplete polyp eradication was significantly lower with cold snare or jumbo forceps biopsy compared with cold biopsy (relative risk, 0.48; 95% CI, 0.30-0.77), with little heterogeneity (I2, 9%). The procedure time for the cold snare or jumbo forceps biopsy was on average 4.1 minutes shorter compared with that of cold biopsy (95% CI, -8 to -2). The authors concluded that randomized controlled trials comparing the efficacy of cold snare, jumbo biopsy, and cold biopsy techniques in the eradication of diminutive polyps are warranted.

Millennium Medical Publishing, Inc