Arthur J. McCullough

Design and validation of a histological scoring system for nonalcoholic fatty liver disease

Nonalcoholic fatty liver disease (NAFLD) is characterized by hepatic steatosis in the absence of a history of significant alcohol use or other known liver disease. Nonalcoholic steatohepatitis (NASH) is the progressive form of NAFLD. The Pathology Committee of the NASH Clinical Research Network designed and validated a histological feature scoring system that addresses the full spectrum of lesions of NAFLD and proposed a NAFLD activity score (NAS) for use in clinical trials. The scoring system comprised 14 histological features, 4 of which were evaluated semi‐quantitatively: steatosis (0‐3), lobular inflammation (0‐2), hepatocellular ballooning (0‐2), and fibrosis (0‐4). Another nine features were recorded as present or absent. An anonymized study set of 50 cases (32 from adult hepatology services, 18 from pediatric hepatology services) was assembled, coded, and circulated. For the validation study, agreement on scoring and a diagnostic categorization (“NASH,” “borderline,” or “not NASH”) were evaluated by using weighted kappa statistics. Inter‐rater agreement on adult cases was: 0.84 for fibrosis, 0.79 for steatosis, 0.56 for injury, and 0.45 for lobular inflammation. Agreement on diagnostic category was 0.61. Using multiple logistic regression, five features were independently associated with the diagnosis of NASH in adult biopsies: steatosis (P = .009), hepatocellular ballooning (P = .0001), lobular inflammation (P = .0001), fibrosis (P = .0001), and the absence of lipogranulomas (P = .001). The proposed NAS is the unweighted sum of steatosis, lobular inflammation, and hepatocellular ballooning scores. In conclusion, we present a strong scoring system and NAS for NAFLD and NASH with reasonable inter‐rater reproducibility that should be useful for studies of both adults and children with any degree of NAFLD. NAS of ≥5 correlated with a diagnosis of NASH, and biopsies with scores of less than 3 were diagnosed as “not NASH.” (HEPATOLOGY 2005;41:1313–1321.)

Nonalcoholic Fatty Liver Disease in Patients With Type 2 Diabetes

BACKGROUND & AIMS Nonalcoholic fatty liver disease (NAFLD) is reported commonly in patients with type 2 diabetes mellitus (DM), which has been suggested as a risk factor for the progressive form of NAFLD, or nonalcoholic steatohepatitis. The aim of this study was to assess the outcome of patients with NAFLD and DM. METHODS A cohort of patients with NAFLD was identified, and patients with other causes of liver disease (alcohol, medication, etc.) were excluded. Clinical, pathological, and mortality data were available for this cohort. Patients were categorized and compared according to the presence or absence of DM. RESULTS Of 132 patients with NAFLD, 44 patients (33%) had an established diagnosis of DM. Patients with DM were older and had greater serum glucose and triglyceride levels and a greater aspartate aminotransferase-alanine aminotransferase ratio. Liver biopsy specimens from patients with DM showed more vacuolated nuclei and acidophilic bodies. Cirrhosis (histological or clinical) occurred in 25% of patients with DM (11 of 44 patients) and NAFLD compared with only 10.2% (9 of 88 patients) of patients without DM with NAFLD (P = 0.04). After adjusting for potential confounders (age, body mass index, and the presence of cirrhosis), both overall mortality (risk ratio [RR], 3.30; 95% confidence interval [CI], 1.76-6.18; P = 0.002) and mortality related to liver disease (RR, 22.83; 95% CI, 2.97-175.03; P = 0.003) were greater in diabetic patients with NAFLD. Markers of hepatic dysfunction (low albumin level, high total bilirubin level, and prolonged prothrombin time) were the only independent predictors of increased mortality. CONCLUSIONS Patients with NAFLD and DM are at risk for the development of an aggressive outcome, such as cirrhosis and mortality. This study supports the potential role of insulin resistance in the development of poor clinical outcomes in patients with NAFLD.

Fatty liver disease: NASH and related disorders

Fibrosis is the most significant pathological con- sequence associated with non-alcoholic steatohepatitis (NASH). Activation of hepatic stellate cells (HSC) into extracellular matrix (ECM) producing myofibroblasts, the central event in hepatic fibrosis, is recognized in NASH. Hepatic fibrogenesis could represent the healing and tissue repair response to chronic necroinflammat- ory injury associated with NASH. However, there is Fatty Liver Disease: NASH and Related Disorders Edited by Geoffrey C. Farrell, Jacob George, Pauline de la M. Hall, Arthur J. McCullough Copyright

Afternoon Colonoscopies Have Higher Failure Rates than Morning Colonoscopies

BACKGROUND:There are several known predictors of an incomplete colonoscopy or difficult colonoscopy. In addition, inadequate bowel preparation has been reported in procedures scheduled later in the day. Operator fatigue, which tends to be higher as the day passes on, may also impact colonoscopy completion rate.AIMS:To determine the influence of performing outpatient colonoscopies in the afternoon versus morning on the completion rates of colonoscopy and adequacy of bowel preparation.METHODS:Retrospective chart review of all outpatient colonoscopies performed between November 2003 and October 2004 in the Division of Gastroenterology at MetroHealth Medical Center in Cleveland, Ohio. Patient demographics, indications for procedure, and colonoscopic findings were reviewed. Patients received polyethylene glycol electrolyte-based bowel preparation in the evening prior to the day of the scheduled colonoscopy.RESULTS:A total of 2,087 colonoscopies was performed, of which 1,084 were in the morning and 999 were in the afternoon. Patients in the morning and afternoon were similar in regards to the known risk factors predictive of an incomplete colonoscopy. The incompletion rate was significantly higher in the afternoon compared to the morning (6.5% vs 4.1%, P = 0.013, OR for incompletion was 1.64, CI 1.11–2.44). Inadequate bowel preparation was found in 167 out of 1,084 (15.4%) colonoscopies in the morning and 197 out of 999 (19.7%) colonoscopies in the afternoon (P = 0.011). Even after excluding incomplete colonoscopies due to poor bowel preparation precluding examination, the incompletion rate was still higher in the afternoon (5% vs 3.2%, P = 0.043, OR 1.60, CI 1.03–2.51).CONCLUSIONS:Scheduling of colonoscopies in the afternoon compared to the morning may be an independent predictor of an incomplete colonoscopy and inadequate bowel preparation. According to our study findings, scheduling of all outpatient colonoscopies preferentially in the morning would avoid suboptimal procedures in 5% of patients and the need for unnecessary repeat colonoscopy or an alternative imaging study in 2.4% of patients.

A Cost Analysis of Alternative Treatments for Duodenal Ulcer

Duodenal ulcer disease is an important clinical problem primarily because of its frequency. Approximately 500 000 new cases and four million recurrences occur annually in the United States [1]. Although mortality due to duodenal ulcer disease is relatively low, this disease is associated with considerable morbidity and cost. The direct costs of diagnosis and treatment and the indirect costs of time lost from work total between three and four billion dollars per year [2]. Excess production of acid was long believed to cause this condition, but the exact pathogenesis and optimal treatment of duodenal ulcer disease remain areas of ongoing clinical interest and controversy. Helicobacter pylori, a gram-negative, urease-producing organism first isolated from gastric mucosa in 1982 by Marshall and Warren [3], is now considered the major cause of chronic active gastritis and is an integral part of the pathogenesis of peptic ulcer disease. Because H. pylori is present in 90% to 100% of patients with duodenal ulcer [4-6], it has become accepted management to detect and eradicate the organism, resulting in decreased ulcer recurrence. Currently, the most popular method of detection is with a test for urease in an antral biopsy specimen obtained during upper endoscopy. The urease test has a reported sensitivity of 89% to 98% and a specificity of 93% to 98% [7]. We did this cost analysis to determine the answers to several questions: Are there cost savings associated with the initial eradication of H. pylori as compared with traditional acid suppression? Given the high association of H. pylori with duodenal ulcer disease, is a urease test necessary, and what effect does this test have on cost? Do factors such as the prevalence of H. pylori infection in duodenal ulcer disease and the cost and side effects of treatment affect the optimal treatment strategy? Methods We used decision analysis, a quantitative method for determining the optimal management strategy under conditions of uncertainty. Decision analysis allows for the integration of information from the published literature and other sources. A cost-focused decision analysis consists of essentially four steps. The first is to create a decision tree that identifies decision alternatives and their clinical outcomes. The second step is to assign probabilities to each outcome and costs to each treatment strategy. The third step is to determine the best strategy under baseline assumptions of probability and cost (in this case, the lowest cost per cure). The fourth step is to test the stability of the baseline conclusions over a range of plausible probabilities and costs, a process known as sensitivity analysis. Design of the Decision Tree Decision analysis was used to compare the direct costs per symptomatic cure of an endoscopically documented duodenal ulcer for three initial treatment strategies: 1) therapy with an H2-receptor antagonist for 8 weeks; 2) therapy with antibiotics for H. pylori infection plus therapy with an H2-receptor antagonist [combination therapy]; and 3) urease test-based treatment with an H2-receptor antagonist alone or with antibiotics for H. pylori infection plus an H2-receptor antagonist for negative and positive results, respectively. Figure 1 is a decision tree showing the major choices that can be made and the chance events that can occur during the treatment of duodenal ulcer disease: the initial treatment choices as described above; the subsequent course, including initial healing and symptomatic recurrences; and the secondary strategies for evaluation and treatment of symptomatic ulcer recurrence. Figure 1. Decision tree for the treatment of duodenal ulcer. Helicobacter pylori Helicobacter pylori After completion of the initial treatment, patients with persistent symptoms at the end of 8 weeks receive a 4-week course of omeprazole and are assumed to be cured of symptoms after 12 weeks (Figure 1). Patients who later develop a symptomatic recurrence receive one of three secondary strategies: 1) endoscopy-guided treatment with either combination therapy for H. pylori infection; omeprazole; or antacids, based on the results of a diagnostic test [urease test or histologic stains], with an assumption of subsequent symptomatic cure; 2) empiric [that is, nonendoscopic] treatment with the same regimen; or 3) empiric treatment with the other regimen, with repeat endoscopy done only for persistent symptoms and then followed by definitive treatment. Of the three secondary strategies, repeat endoscopy and empiric treatment (or retreatment) with combination therapy are the two that are most likely to be used by practitioners. Empiric treatment (or retreatment) with an H2-receptor antagonist alone is much less likely to be used in practice today but was included in our analysis for completeness and for its former utility. A fundamental assumption of this cost analysis is that all members of the cohort are eventually cured. Patient Population The target population for this analysis is a patient cohort with endoscopically documented duodenal ulcer disease and no risk factors for serious underlying conditions such as cancer or inflammatory bowel disease. Such patients would be younger than age 60 years, would show no weight loss, would not have a history of continuous use of nonsteroidal anti-inflammatory drugs, and would have no evidence of gastrointestinal blood loss. The more recent literature suggests that the prevalence of H. pylori infection in such a cohort would be close to 100% [5]. We assumed that this low-risk patient population would be similar to populations in the cohort studies and clinical trials from which probability values were derived. Further Assumptions of the Model In addition to the assumptions already mentioned, other assumptions were incorporated into the model. Regarding compliance with the treatment for H. pylori, we assumed, in accordance with the published literature [8], that 50% of patients would take less than 60% of the prescribed treatment. Although it would be assumed that this would lower the eradication rate and subsequent long-term cure rate, recent data suggest that reducing the duration of treatment from 2 weeks to 1 week may not significantly affect the efficacy of treatment [9]. We also assumed that 10% of patients receiving antibiotic treatment for H. pylori infection plus an H2-receptor antagonist would require an office visit for adverse effects from the antibiotics. Half of these adverse effects would be rash, the other half would be diarrhea or upper gastrointestinal upset or both. The costs associated with such adverse effects were incorporated into the base cost for H. pylori treatment. The regimen used for H. pylori eradication consisted of ranitidine, 300 mg nightly for 8 weeks; amoxicillin, 750 mg three times daily; metronidazole, 500 mg three times daily; and bismuth subsalicylate, 120 mg four times daily, all for 12 days. This regimen was modified from that used by Hentschel and colleagues [10], who observed an eradication rate of 89%. We added bismuth subsalicylate to justify extrapolation of the published data on the efficacy and compliance with bismuth-containing regimens. Substitution of amoxicillin with tetracycline, 500 mg four times daily, results in no appreciable difference in the efficacy or cost of treatment. Finally, we assumed that any endoscopy done for recurrence of symptoms after attempted H. pylori eradication as the initial treatment would include the cost of histologic stains (hematoxylin-eosin and Giemsa) for the detection of H. pylori, provided that an ulcer is visualized during the repeat endoscopy. For repeat endoscopy for symptomatic recurrences after initial treatment with an H2-receptor antagonist alone, we assumed that only an antral biopsy specimen for urease testing would be obtained. Probabilities and Costs Baseline probability values and ranges used in sensitivity analysis were derived from both the medical literature and expert opinion (Table 1). Average healing and recurrence rates for duodenal ulcer were calculated from previous reports of either cohort studies or randomized clinical trials. Rates from each study were used to calculate an overall weighted average. For the H. pylori eradication rate, the calculated weighted average of 83% agrees with the results of a published meta-analysis [27]. Assuming that 55% of H. pylori-positive duodenal ulcers and 5% of H. pylori-negative ulcers recur symptomatically within 1 year [28], the recurrence rate would be 14% [(0.55 17%) + (0.05 83%)]. Therefore, our baseline rate of long-term cure of 85% is conservative. Table 1. Baseline Probabilities and Ranges Used in the Analysis* To provide cost estimates for economic analyses, the perspective simulated was that of a group practice model health maintenance organization. For all medications shown in Table 2, costs were estimated based on average wholesale prices from the 1994 Red Book [29]. Direct medical costs for diagnostic tests, for the technical component of upper endoscopy, and for an office consultation with a gastroenterologist were determined by using the costs for labor and materials assigned by the cost accounting system at a 742-bed teaching hospital. Costs for physician services were based on allowable Medicare reimbursement. Baseline cost estimates for diagnosis, treatment, and management of complications are summarized in Table 2, along with the ranges over which sensitivity analyses were conducted. All costs exclude the cost of the initial upper endoscopy but include all subsequent costs. Table 2. Baseline Direct Costs and Ranges Used in the Analysis* Costs along each branch of the decision tree were weighted by their probabilities and added, resulting in the expected total direct cost for that particular branch (Figure 1). For example, consider a patient who receives initial treatment with an H2-receptor antagonist followed by omeprazole for persistent symptoms and w

Omeprazole is superior to ranitidine plus metoclopramide in the short-term treatment of erosive oesophagitis.

Histamine H2‐receptor antagonists are moderately effective in symptomatic treatment and healing of erosive oesophagitis, but they are not as effective as the proton pump inhibitor omeprazole. In some studies prokinetic agents seem to increase the effectiveness of H2‐antagonists, but no study comparing the efficacy of omeprazole to H2‐antagonists plus prokinetic agents has been performed. The purpose of this study was to compare the efficacy and tolerability of 20 mg omeprazole daily with 150 mg ranitidine b.d.s. plus the prokinetic agent 10 mg metoclopramide q.d.s. in patients with erosive oesophagitis. After both 4 and 8 weeks of treatment, omeprazole healed the mucosa in significantly more patients than did ranitidine plus metoclopramide. Omeprazole also provided significantly greater relief from daytime heartburn, nighttime heartburn, and acid regurgitation, and was associated with decreased concomitant antacid use. Although the overall incidence of adverse events was similar in the two treatment groups, a significantly higher number of treatment‐related adverse events and more treatment‐related withdrawals from the study occurred in the ranitidine plus metoclopramide treatment group. Omeprazole is more effective and better tolerated than the combination of standard dose ranitidine plus metoclopramide for patients with erosive oesophagitis.

The influence of portacaval anastomosis on gonadal and anterior pituitary hormones in a rat model standardized for gender, food intake 9 and time after surgery

Portacaval anastomosis causes delayed growth, decreased testes and liver weights, and elevated estradiol serum levels in male rats compared with sham-operated controls. Female rats treated with portacaval anastomosis grow at a normal rate despite changes in liver weight and estradiol levels similar to those observed in the male rats. This study examined the pituitary gonadal axis in both genders in this animal model. The rats receiving portacaval anastomosis were compared with both pair-fed and sham-operated control groups. Portacaval anastomosis decreased serum testosterone and increased estradiol in the male animals, while both testosterone and estradiol were increased in the females compared with gender-matched pair-fed and sham controls. Because pair feeding lowers male testosterone to a lesser extent, impaired nutrition may partially account for the decrease in the males treated with portacaval anastomosis. The ratio of estradiol to testosterone increased following anastomosis in male rats, but it was decreased in similarly treated females. Portacaval and anastomosis decreased luteinizing hormone without changing follicle-stimulating hormone in both male and female rats compared with sham-operated controls. Growth hormone was significantly decreased in male portacaval-treated rats compared with sham- and pair-fed animals. Increased insulin levels were found in both male and female pair-fed and portacaval anastomosis-treated animals. These data suggest that following portacaval anastomosis in rats, growth, serum testosterone, estradiol to testosterone ratios, and growth hormone are altered in a gender-specific manner with gender-independent changes in insulin and luteinizing hormone levels. These gender-specific effects may protect the portacaval anastomosis-treated female rat from growth retardation.

Use of tegaserod along with polyethylene glycol electrolyte solution for colonoscopy bowel preparation: a prospective, randomized, double‐blind, placebo‐controlled study

Background  Polyethylene glycol electrolyte solution (PEG‐EL) used for colonoscopy preparation is not well tolerated by several patients. A significant number of patients have inadequate bowel preparation despite taking PEG‐EL.

A continuous quality improvement initiative reduces inappropriate prescribing of prophylactic antibiotics for endoscopic procedures.

OBJECTIVE:Despite the availability of guidelines, most gastroenterologists do not administer prophylactic antibiotics appropriately to patients having endoscopic procedures. In 1994 we recognized that in our endoscopy unit, many patients were receiving antibiotics without proper indication. We devised a continuous quality improvement initiative to analyze and improve this problem.METHODS:Divisional guidelines for the appropriate administration of prophylactic antibiotics for endoscopy were drawn up in 1995. By retrospective analysis of our comprehensive endoscopy database we compared the rate of prophylactic antibiotic administration, and the proportion of antibiotics that were indicated before and after adoption of the divisional guidelines.RESULTS:A total of 1427 endoscopic procedures were done during a 6-month period in 1994 (before adoption of guidelines). Of these, 55 (3.85%) received antibiotics. In a 6-month period in 1996 after adoption of guidelines, 1452 procedures were performed and 29 of these (1.99%) received antibiotics. The odds ratio for receiving antibiotics appropriately in 1996, compared with 1994, was 3.4 (χ2p= 0.016). Given an annual volume of 2900 procedures in our endoscopy unit, approximately 54 patients will avoid unnecessary antibiotics, yielding a cost saving of

402 ADRENERGIC BLOCKADE HEIGHTENS THE EXERCISE INDUCED INCREASE IN LEUCINE OXIDATION

1128 per year.CONCLUSIONS:A divisional continuous quality improvement initiative on antibiotic prophylaxis for endoscopy significantly reduced the proportion of patients receiving antibiotics unnecessarily. This quality improvement initiative enhanced the quality of care for patients having endoscopy and yielded a small cost saving. These improvements were achieved with minimal effort and cost to the division.