You are here

Meeting Highlights

Digestive Disease Week 2011

Walter Alexander

At this year’s annual Digestive Disease Week (DDW) meeting, more than 15,000 physicians and 2,000 to 3,000 exhibitors gathered in Chicago from May 7 to 10, 2011, to hear the latest research from investigators representing the American Association for the Study of Liver Disease, the American Gastroenterological Association, the American Society for Gastrointestinal Endoscopy, and the Society for Surgery of the Alimentary Tract. Key sessions included studies of clopidogrel use in colonoscopy; rifaximin to prevent spontaneous bacterial peritonitis; effects of teduglutide, a novel analogue of glucagon-like peptide-2, on parenteral support volume; and the risk of Clostridium difficile infection with the use of specific antibiotics.

Low Rates of Postpolypectomy Bleeding With Clopidogrel (Plavix) During Colonoscopy

  • Fatema S. Uddin, MD, Fellow, University of Texas Southwestern Medical Center/Dallas Veterans Affairs Medical Center, Dallas
  • Linda A. Feagins, MD, Assistant Professor of Medicine, Division of Digestive and Liver Diseases, University of Texas Southwestern Medical Center/Dallas Veterans Affairs Medical Center, Dallas

Most adults 60 years of age and older have at least one type of cardiovascular disease, with 20% requiring anticoagulation or antiplatelet agents, such as aspirin, warfarin (Coumadin, Bristol-Myers Squibb) and clopidogrel (Plavix, Bristol-Myers Squibb/Sanofi-aventis). The American Society for Gastrointestinal Endoscopy guidelines recommend considering withholding clopidogrel for seven to 10 days before high-risk procedures such as polyp removal. Stopping these agents, however, entails a risk of myocardial infarction, cerebrovascular accidents, thromboembolic events, and stent thrombosis, which carries a particularly high mortality risk among patients who have received a drug-eluting stent. Data are lacking, however, on the risk of bleeding when clopidogrel treatment is continued during colonoscopy with polypectomy.

Dr. Uddin and colleagues evaluated the safety of continuing the medication during colonoscopy when polypectomy is performed. In the study, colonoscopy with polyp removal was performed in 118 patients who were taking clopidogrel and 1,849 patients who were not. A logistic regression analysis revealed no significant difference in the frequency of delayed postpolypectomy bleeding between clopidogrel users and non-users (0.8% vs. 0.3%, respectively; P = 0.37). The unadjusted odds ratio (OR) was 2.63, and the 95% confidence interval (CI) ranged from 0.31 to 22.

There were significant differences between clopidogrel users and non-users in the frequency of coronary artery disease (94.1% vs. 24.2%, respectively), aspirin use (78% vs. 27.9%, respectively), age (64.9 vs. 62.4 years, respectively), and lung disease (24.6% vs. 13.4%, respectively). With these statistics taken into account, a matched analysis also revealed no significant difference in rates of postpolypectomy bleeding among clopidogrel users and non-users (0.9% vs. 0%, respectively; P = 0.99). Importantly, the outcomes among patients who did experience bleeding were without major consequences.

Lead investigator Dr. Feagins, speaking at a DDW press conference, noted that the study was limited; it was retrospective, and most of the polyps removed were smaller than 1 cm. However, the largest polyp removed was 8 mm.

“We use hemoclips. That makes us feel better,” she said.

She further noted that at Dallas VA Hospital, because physicians have judged the cardiovascular risks of discontinuing clopidogrel before elective colonoscopy to exceed the risks of postpolypectomy bleeding, the policy is to not routinely discontinue clopidogrel.

Rifaximin (Xifaxan) in Prevention of Spontaneous Bacterial Peritonitis in Patients With Cirrhosis

  • Ibrahim A. Hanouneh, MD, Gastroenterology Fellow, Cleveland Clinic, Cleveland, Ohio

Spontaneous bacterial peritonitis (SBP) occurs in 8% to 25% of patients with liver cirrhosis and ascites. The in-hospital mortality rate among these patients ranges from 10% to 30%, and the illness recurs within a year in 70% of those who survive an episode.

Although the American Association for the Study of Liver Disease practice guidelines recommend the use of an antibiotic such as norfloxacin (Noroxin, Merck) or trimethoprim/sulfamethoxazole (Bactrim, AR Scientific) for the primary prophylaxis of SBP in high-risk cirrhotic patients with ascites, there is no current consensus on whether prophylactic antibiotics provide a long-term survival benefit. At the same time, a concern remains that prophylactic antibiotics can lead to the emergence of resistant gut flora—which, in turn, can cause spontaneous infection.

To determine whether rifaximin (Xifaxan, Salix) might prevent SBP in patients with cirrhosis and ascites and whether the drug would extend survival, Dr. Hanouneh reviewed all patients seen at the Cleveland Clinic liver transplantation unit between 2003 and 2007 (N = 664). Rifaximin is a poorly absorbed, well-tolerated antibiotic with low bacterial resistance risk. This broad-spectrum agent covers gram-positive and gram-negative organisms, aerobes, and anaerobes.

Ascites that was clinically significant enough to justify paracentesis was found in 434 of the patients reviewed. Patients were stratified by their use of rifaximin (49 patients received the drug; 355 did not), which had been given primarily for hepatic encephalopathy at a dose of 400 mg three times daily. Patients who had received primary SBP prophylaxis with norfloxacin or trimethoprim/sulfamethoxazole (N = 8) and who had a history of SBP (N = 14), SBP with gastrointestinal (GI) bleeding (N = 2), or who had undergone prior transplantation (N = 6) were excluded from the study. Patients were observed until organ transplantation or death occurred.

The mean patient age was approximately 54 years (60% men, 83% Caucasian). Encephalopathy was more common in patients using rifaximin than in non-users (100% vs. 32%, respectively; P < 0.001), as were elevated Child–Pugh scores on a cirrhosis severity scale (11.5% vs. 10.4%, respectively; P < 0.001).

After about 60 months of follow-up evaluations, 118 patients (29%) had developed SBP—five in the rifaximin group and 113 in the norifaximin group (P = 0.001). Based on those percentages, the number needed to treat (NNT) to prevent one case of SBP with rifaximin treatment was 5.

A multivariable analysis revealed that rifaximin (hazard ratio [HR] = 0.28; P = 0.007) and the Mayo End-Stage Liver Disease (MELD) score (HR = 1.06; P < 0.001), another marker of disease severity and mortality in persons with chronic alcoholic liver disease, were factors associated with SBP.

The analysis showed a 72% reduction in the rate of SBP with rifaximin, after adjustments for the MELD score, the Child–Pugh score, serum sodium levels, and ascitic fluid total protein. Transplant-free survival was also higher in the rifaximin group (72% vs. 57% for non-users; P = 0.04).

A separate multivariable analysis of transplant-free survival factors showed significance for rifaximin (HR = 0.54; P = 0.039) and for the MELD score (HR = 1.09; P < 0.001).

“Intestinal decontamination with rifaximin is an effective measure to prevent SBP and improve transplant-free survival in cirrhotic patients with ascites,” Dr. Hanouneh concluded.

At the DDW press conference, he commented that because the study was retrospective, and until the findings are confirmed prospectively, he could not yet make recommendations.

Teduglutide (Gattex) and Short-Bowel Syndrome: The STEPS Trial

  • Palle B. Jeppesen, MD, Associate Professor of Gastroenterology at Rigshospitalet, University Hospital of Denmark, Copenhagen

Short-bowel syndrome with intestinal failure (SBS/IF) is generally the result of surgical resection, a congenital defect, or a disease-related loss of absorption. Patients with SBS/IF are unable to maintain protein, energy, fluid, and electrolyte levels and micronutrient balance. Compensatory fluid or nutritional support for this malabsorption is accomplished through hyperphagia (additional oral intake) and intravenous (IV) parenteral support. Intestinal rehabilitation may be addressed through intestinal growth factors and antisecretory agents.

Among the medical needs of SBS/IF patients are a decreasing dependence on IV parenteral support and fewer morbidity and life-threatening complications associated with parenteral support, including catheter-related bloodstream infections, venous thrombosis, and intestinal failure-associated liver disease. Improved quality of life and novel treatment options are additional needs.

Glucagon-like peptide-2 (GLP-2) repairs and restores the functional integrity of the intestinal mucosa, improving fluid and nutrient balance by enhancing absorption. GLP-2 increases villus height and diameter as well as crypt depth, promotes dilatation and thickening of mucosa, and increases RNA and DNA content, all leading to improved absorption. In some patients with SBS/IF, GLP-2 also promotes rapid gastric emptying and gastric acid hypersecretion, intestinal blood flow, and adaptation. Teduglutide (Gattex, TDG, NPS Pharmaceuticals) also increases wet-weight absorption (equivalent to 750 mL/day) and urine sodium excretion and reduces fecal weight and energy excretion in patients with SBS.

TDG, an orphan drug, is a dipeptidyl peptidase-IV degradation-resistant, recombinant analogue of human GLP-2. In order to evaluate TDG’s ability to reduce the volume of parenteral support (IV fluids, nutrients, or both), investigators enrolled 86 subjects with SBS/IF who were dependent on parenteral support for one year or more year in a randomized, double-blind, placebo-controlled, parallel-group, multinational multi-center trial called STEPS (Study of Teduglutide in Parenteral Nutrition–Dependent Short-Bowel Syndrome).

After an initial optimization and stabilization period of parenteral support volume (ensuring stable urine output), subjects were randomly assigned to receive subcutaneous TDG 0.05 mg/kg per day or placebo once daily for 24 weeks. Parenteral support volumes were reduced at scheduled visits if urine volumes exceeded the baseline value by 10% or more. Response was defined as a 20% to 100% reduction from baseline in weekly parenteral support volume at weeks 20 and 24. The primary efficacy endpoint was a comparison of the percentage of responders in the TDG and placebo groups.

Among 78 patients completing the study, 39 received TDG and 39 received placebo. In the intent-to-treat (ITT) population, responses were reported in 27 of 43 SBS/IF subjects receiving TDG (63%) and in 13 of 43 patients receiving placebo (30%) (P = 0.002). At week 24, mean weekly parenteral support volume was reduced by 4.4 L/week with TDG (12.9 L/week at baseline), compared with a volume reduction of 2.3 L/week with placebo (13.2 L/week at baseline) (P ≤ 0.001). Significantly more TDG subjects (54%) than placebo subjects (23%) were able to achieve reductions in the number of infusion days by one or more days (P = 0.0047). Despite the significant reduction in parenteral support volume, however, body weight remained constant.

TDG was well tolerated. Of the eight dropouts, two of four patients in the TDG group and three of four patients in the placebo group discontinued therapy because of adverse drug events (AEs). AEs were documented in 35 of 42 patients (83%) receiving TDG, compared with 34 of 43 subjects (79%) receiving placebo. The most common AEs were abdominal pain, nausea, GI stoma complications, and abdominal distention.

More than 97% of the subjects who completed the study elected to enroll in an open-label 24-month continuation study.

Dr. Jeppesen characterized the quality of life of these patients, with their nightly infusions and alternation between dehydration and fluid overload, as comparable to that experienced by patients on dialysis. The 30% reduction in IV parenteral support volume with TGD treatment might allow patients an additional two days off from this modality.

“They cherish days off so they can lead a more normal life,” he said.

Antibiotic Use and Clostridium difficile Infection In the U.S.: A Population-Based Study

  • Narayan Dharel, MD, Internist, Beth Israel Deaconess Medical Center, Brockton, Mass.

Clostridium difficile infection is a major cause of morbidity and mortality among hospitalized patients, particularly in the elderly and in those who receive antimicrobial therapy. An increased incidence of C. difficile has recently been reported in populations previously thought to be at low risk, including among healthy persons in the community. However, the status of this infection in the general U.S. population and the associated risk factors remain largely unclear. Dr. Dharel’s analysis estimated the incidence of C. difficile infection in the general U.S. population and assessed the risk associated with exposure to antibiotics in common use.

An analysis of claims data from among 11.2 million commercially insured individuals identified 7,023 C. difficile cases over a 7.9-year period ending in 2007. Among these claims, the investigators identified 3,527 cases of individuals who had been enrolled for at least two years prior to the onset of the infection.

C. difficile was defined as any inpatient or outpatient claim coded as International Classification of Diseases, 9th revision (ICD-9) code 008.45 or as the use of oral vancomycin (Vancocin, Eli Lilly) to treat a presumptive C. difficile infection. These claims were compared with those of up to 10 controls, who were matched on the basis of sex, age within two years, geographical region, type of insurer, subscriber status (individual, spouse, or dependent), and date of enrollment within one year in a nested case–control study. Conditional logistic regression analyses were used to evaluate exposure to various antibiotics within the preceding 30, 60, and 90 days.

The average incidence of C. difficile infection for the U.S. population in this period was 16.4 cases per 100,000 person-years. The incidence increased exponentially from 0.3 per 100,000 person-years in 2000 to 39.7 per 100,000 person-years in 2006. In addition, C. difficile risk increased not only with antibiotic use but also in relation to the number of antibiotics used. For patients receiving one antibiotic within the previous 30 days, the adjusted odds ratio (OR) was 5.83 (5.05–6.74, P < 0.0001 vs. none). For those receiving three or more antibiotics, the OR was 11.44 (8.27–15.84; P < 0.0001 vs. none).

An evaluation of C. difficile risk with exposure to specific antibiotics within 30 days showed that the risk was greatest among individuals who had received linezolid (Zyvox, Pfizer), with an adjusted risk ratio (RR) of 33.04 (95% CI, 5.38–202.83). Linezolid, an oral agent, Dr. Dharel commented, is commonly used in the community to treat methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant enterococci (VRE) infections.

“People have to be aware of the increased risk,” he said.

The risk was also raised with clindamycin (Cleocin, Pfizer) (RR = 14.9; 95% CI, 10.8–20.6), fluoroquinolones (RR = 6.43; 95% CI, 5.43–7.6), and cephalosporins (RR = 5.22; 95% CI, 4.28–6.36).

Macrolides showed the lowest elevation in risk (RR = 1.42; 95% CI, 1.18–1.72), and among them, the increase for azithromycin (Zithromax, Pfizer) was found to be nonsignificant (RR = 1.11; 95% CI, 0.89–1.39). Of note, only 53% of the patients had a history of antimicrobial exposure within the preceding 90 days.

In an interview, Dr. Dharel noted that C. difficile infections flourish when antibiotics knock out normal gut flora. The use of probiotics, he said, may be protective. He said further that C. difficile spores are very resistant to chemical hand sanitizers and that handwashing with soap and water is probably necessary for prophylaxis against contamination.

A search for other potential factors contributing to the increase in C. difficile infections is needed, Dr. Dharel said. Among them might be changes in medication use in recent years, including the increased use of proton pump inhibitors, cardiac drugs, and chemotherapeutic agents.

Dr. Dharel concluded: “C. difficile incidence is rapidly increasing in the U.S. population. Antibiotic-associated C. difficile risk was highest with linezolid and clindamycin, and lowest with macrolides. Azithromycin conferred no significant risk.”