Can severity of symptoms be used as an outcome measure in trials of non-ulcer dyspepsia and Helicobacter pylori associated gastritis?

Most trials of non-ulcer dyspepsia (NUD) and Helicobacter pylori associated gastritis (HPAG) have not used validated methods of measuring symptoms. Three attributes are necessary for use of symptom severity scoring systems as outcome measures in clinical trials: reproducibility, responsiveness to change and validity compared to corroborating measures. The objective of this study was to establish that selected gastrointestinal symptoms recorded as a series of 5-point Likert Scales meet the 3 criteria for use as outcome measures in clinical trials. Patients with NUD (Helicobacter pylori-negative) and HPAG were studied. A preliminary assessment of 24 patients was used to select the 8 most frequently occurring and most severe symptoms. These symptoms were then scored in a further 55 patients to assess their utility as outcome measures. Observations were made at 3 time points, enrollment (T1), after 1 week with no intervention (T2) and after 4 weeks of therapy for either disease (T3). The study took place in a university hospital outpatient gastroenterology service. Symptom scores were reproducible before treatment (symptom scores at T1 and T2 were correlated), responsive (symptom scores changed after treatment between T2 and T3) and valid (symptom score changes corresponded to changes in general health status). Scoring of gastrointestinal symptom severity using 5-point Likert Scales satisfies the 3 criteria for use as outcome measures in clinical trials of NUD and HPAG.

Pain following colonoscopy: elimination with carbon dioxide.

Fifty-six patients have been examined in a prospective randomized study on the effects of air and carbon dioxide on post-procedural discomfort following colonoscopy. A significant reduction in post-procedural pain was observed at 6 hours (p = < 0.0005) and was still present the next day (p = 0.01). This was associated with a difference in the grading of flatus at 6 and 24 hours (p = < 0.0001 and < 0.05, respectively). An abdominal radiograph 1 hour after the procedure showed minimal gas in the CO2 patients, while the patients who had air showed distention of large and small bowel (p = < 0.0001 and < 0.01, respectively). Seventeen of 29 patients who had air suffered post-procedural pain, compared with 2 of 27 of the CO2 patients. Fifty-seven percent of the patients who were given air had colonic diameters over 6 cm on a 1-hour post-colonoscopy radiograph and 18% over 10-cm diameter. Provision by equipment manufacturers of simple and safe devices for routine delivery of CO2 for lower gastrointestinal endoscopy is long overdue.

14C-urea breath test for the detection of Helicobacter pylori.

The high urease activity of Helicobacter pylori can be used to detect this bacterium by noninvasive breath tests. We have developed a 14C-urea breath test which uses 5 microCi 14C with 50 mg nonradioactive urea. Breath samples are collected at baseline and every 30 min for 2 h. Our study compared the outcome of the breath test to the results of histology and culture of endoscopically obtained gastric biopsies in 84 patients. The breath test discriminated well between the 50 positive patients and the 34 patients negative for Helicobacter pylori: the calculated sensitivity was 100%, specificity 88%, positive predictive value 93%, and negative predictive value 100%. Treatment with bismuth subsalicylate and/or ampicillin resulted in lower counts of exhaled 14CO2 which correlated with histological improvement in gastritis. The 14C-urea breath test is a better "gold standard" for the detection of Helicobacter pylori than histology and/or culture.

Can gastritis symptoms be evaluated in clinical trials? An overview of treatment of gastritis, nonulcer dyspepsia and Campylobacter-associated gastritis.

We carried out a review of the literature on Campylobacter pylori-associated gastritis and nonulcer dyspepsia (NUD) to determine whether or not symptoms related to these conditions can be measured reliably and whether or not any study to date has shown that treatment alters symptoms. Search strategies consisted of online Medline searching, a forward search of three articles using the Science Citation Index, a manual search of five gastroenterological journals, and a fully recursive search of cited references. Inclusion and quality criteria were applied to all retrieved studies. Nine of 23 studies did not fulfill the inclusion criteria. Of the 14 studies analyzed, two measured symptoms reliably. Neither showed a therapeutic benefit on symptoms. The difficulties encountered in conducting such studies and the methods of recording symptoms reliably are discussed. We conclude that to date, no treatment is of proven benefit in the relief of symptoms associated with C. pylori gastritis and NUD.

The diagnosis of acute fatty liver of pregnancy by computed tomography.

We describe 2 patients with biopsy-proven acute fatty liver of pregnancy in whom the diagnosis was suggested by the finding of uniform reduced attenuation values on computed tomography (CT) of the liver. The attenuation values returned to normal with resolution of signs and symptoms following delivery. CT of the liver provides a useful method for the diagnosis of acute fatty liver of pregnancy.

Small bowel and colonic permeability to 51Cr-EDTA in patients with active inflammatory bowel disease.

51Cr-EDTA was administered both orally and per rectum via a catheter to controls and to patients with inflammatory bowel disease. The patients were divided into two groups, either with active inflammation of the small bowel or with active inflammation of the colon. Fifteen patients with Crohn's disease of the small bowel and 19 patients with either Crohn's disease of the colon or ulcerative colitis were investigated. After oral administration of the probe, controls showed a median excretion of 1.17%/24 h of the dose compared to 3.47%/24 h by patients with small bowel disease and 6.07%/24 h by patients with colonic disease. After rectal administration, controls showed a median excretion of 0.74%/24 h of the dose compared to 0.93%/24 h by patients with small bowel disease and 5.73%/24 h by patients with colonic disease. The rectal test differentiated small bowel disease from colonic disease with an accuracy of 85%. The results confirmed the inflamed colon as a site of increased intestinal permeation.

Reversibility of increased intestinal permeability to 51Cr-EDTA in patients with gastrointestinal inflammatory diseases.

Intestinal permeability in adults with inflammatory gastrointestinal diseases was investigated by measuring the 24-h urinary excretion of orally administered 51Cr-EDTA. Eighty controls along with 100 patients with Crohn's disease, 46 patients with ulcerative colitis, 20 patients with gluten-sensitive enteropathy, and 18 patients with other diseases were studied. In controls, the median 24-h excretion was 1.34%/24 h of the oral dose. Patients with Crohn's disease (median 2.96%/24 h), ulcerative colitis (median 2.12%/24 h), and untreated gluten-sensitive enteropathy (median 3.56%/24 h) had significantly elevated urinary excretion of the probe compared to controls (p less than 0.0001). Increased 24-h urinary excretion of 51Cr-EDTA had a high association with intestinal inflammation (p less than 0.0001). Test specificity and sensitivity were 96% and 57%, respectively. A positive test has a 96% probability of correctly diagnosing the presence of intestinal inflammation, whereas a negative test has a 50% probability of predicting the absence of disease.

Increased intestinal permeability in patients with rheumatoid arthritis: a side-effect of oral nonsteroidal anti-inflammatory drug therapy?

Intestinal permeability in controls and in patients with active rheumatoid arthritis was assessed by measuring the radioactivity in 24-h urine collections after oral administration of 51Cr-EDTA. Intestinal permeability in the patients was found to be significantly increased in comparison to controls. Of the patients, 6/7 (86%) not taking any prescribed nonsteroidal anti-inflammatory drug (NSAID) at the time of study had normal results, whereas 23/29 patients (79%) taking prescribed NSAIDs displayed increased intestinal permeability. It has not been determined beyond doubt whether this finding is due to disease process or therapy with oral NSAIDs.

Studies of intestinal permeability in inflammatory diseases using polyethylene glycol 400.

It has been proposed that increased bowel permeability might play a role in the pathogenesis of inflammatory disease. Intestinal permeation was investigated by measuring the 6-hour urinary excretion of polyethylene glycol (PEG) 400 in 40 adult volunteer controls and in patients with inflammatory disease. Of the patients, 15 had Crohn's disease; 7, ulcerative colitis; 2, celiac disease; and 7, rheumatoid arthritis. No significant difference in total urinary excretion over a 6-hour period was found between controls and patients with ulcerative colitis. Patients with Crohn's disease, celiac disease, or with rheumatoid arthritis were found to have significantly decreased urinary excretion of PEG 400. The results of this study indicate that there is no identifiable increase in intestinal permeation as measured by PEG 400 excretion during periods of active inflammatory disease.

The benefit of emergency nasobiliary drain in cholelithiasis with ascending cholangitis, coagulopathy, and thrombocytopenia.

Severe coagulopathy in a male patient with septicemia, renal failure, and obstructive jaundice secondary to cholelithiasis precluded safe endoscopic sphincterotomy. A temporary nasobiliary drain, inserted at endoscopic retrograde cholangiopancreatography, decompressed the biliary tree, allowing eventual safe sphincterotomy and bile duct clearance after correction of coagulopathy and improvement in his clinical condition.

Reduced suppressor cell activity in intestinal lymphocytes from patients with Crohn's disease.

Suppressor cell activity in intestinal lymphocytes was investigated by a coculture assay with autologous peripheral blood lymphocytes. Greater than 45% suppression was found in 7 out of 10 patients without Crohn's disease. Reduced suppressor cell activity was found in intestinal lymphocytes isolated from 7 out of 8 patients with Crohn's disease. Intestinal lymphocytes from patients with Crohn's disease demonstrated a significantly greater response to phytohemagglutinin-P than lymphocytes isolated from non-Crohn's patients. However, this difference could not be explained by alterations in proportions of T lymphocytes as intestinal T lymphocytes from non-Crohn's disease patients (54%) were not significantly different from those found in patients with Crohn's disease (57%). Indomethacin, a prostaglandin synthetase inhibitor, decreased the proliferative response of intestinal lymphocytes to phytohemagglutinin-P whereas the response of peripheral blood lymphocytes was enhanced, thus providing further evidence that intestinal lymphocytes may represent a functionally distinct population from circulating lymphocytes. Cimetidine, a H2-receptor antagonist, had no effect on the response of intestinal lymphocytes to phytohemagglutinin-P. These results provide evidence to support the concept that disturbed immunoregulation at the mucosal level is found in Crohn's disease.

Hemolytic anemia in patients receiving sulfasalazine.

Hemolytic anemia is a well-recognized complication of sulfasalazine treatment. 17 of 40 (43%) patients with inflammatory bowel disease receiving sulfasalazine had evidence of hemolysis as detected by starch gel electrophoresis. Only 47% (8) of patients with hemolysis had Heinz body formation. The hemoglobin was significantly reduced in patients with hemolysis and 53% had a reticulocyte count of greater than 5%. A significant correlation was noted between hemolysis and serum sulfapyridine level, but no correlation was seen with serum sulfasalazine level. There was no significant difference in disease extent or activity in patients with hemolysis compared to those without hemolysis. Hemolysis is not a rare side-effect of sulfasalazine therapy. Heinz body formation is not invariably found in sulfasalazine-induced hemolysis.

Persistence of proctitis in 2 cases of clindamycin-associated colitis.

Two cases of persisting proctitis after an episode of clindamycin-associated colitis are reported. Both patients had fulminant colitis and required emergency operation. Severe proctitis persisted in the rectal stumps until their removal 8 months after the initial diagnosis of colitis was made. Persisting proctitis has not previously been described in clindamycin-associated colitis.

Cell mediated immunity to corn starch in starch-induced granulomatous peritonitis.

Two patients with histologically diagnosed starch induced granulomatous peritonitis (SGP) have been shown to have cell mediated immunity to corn starch using the techniques of macrophage migration inhibition and lymphocyte DNA synthesis. Control groups of normal subjects, patients with uncomplicated laparotomy, and patients with Crohn's disease were negative in both tests. Lymphocytes from two patients with band adhesions, one of whom had biopsy evidence of a granulomatous reaction to starch, were sensitized to starch. Cell mediated immunity to starch may contribute to the pathogenesis of SGP, and some band adhesions may be a chronic low grade manifestation of this disorder.