Upper gastrointestinal bleeding: what has changed during the last 20 years?

AIM OF THE STUDY: The purpose of this study was to assess the clinical, epidemiological, therapeutic and prognostic changes observed in patients with upper gastrointestinal bleeding (UGIB) during the last two decades. METHODS: Two cohorts of 200 patients with UGIB consecutively recruited during the period 1984-1987 (cohort A) then during the period 2004-2006 (cohort B) were compared. RESULTS: Median age was 61.5 years in the cohort A and 67 years in the cohort B. The main etiologies remained variceal bleeding and peptic ulcer, but esophagitis and Mallory-Weiss syndrome were more frequently observed recently. The intake of gastrotoxic drugs did not decreased despite the widely acknowledged harmful effects of these drugs. Regarding management practices, a therapeutic intervention during the first endoscopy was performed in 36% of the cases in the cohort B but only in 2% of the cases 20 years ago. The frequency and the volume of blood transfusion dramatically decreased during the last two decades. Regarding the outcome, the requirement for surgery and the rate of recurrent bleeding decreased by half, but mortality remained unchanged. CONCLUSION: The main changes observed in patients admitted for UGIB 20 years apart concerned epidemiological features, treatment and prognosis.

Epidemiology of coeliac disease.

In this article we review recent worldwide epidemiological data of coeliac disease (CD). An emphasis is made on adult figures as there is reduction of infants cases that is counterbalanced by the increase of CD in older children or adults. We review data from Europe, USA, South America, Australia, Asia and Africa. Studies in mixed population and in patients undergoing endoscopy are also mentioned. The prevalence of the disease varies between 1/100-500 in different continents. It is possible that the low incidence of CD in some regions is due to the decrease of the prevalence of classic form of the disease.

Impact of hepatitis C virus infection on the aetiology of cirrhosis and hepatocarcinoma in three affiliated hospitals in southern Belgium.

In a consecutive series of 411 patients with cirrhosis attending the outpatient liver clinics of 3 general hospitals located in the southern part of Belgium, hepatitis C virus infection accounted for 20% of the cases, far behind alcohol (63%). However, in a consecutive series of 57 hepatocarcinoma superimposed on cirrhosis, hepatitis C virus infection was the main aetiological factor accounting for 44% of the cases.

[lymphocytic colitis associated with lansoprazole treatment]. / Colite lymphocytaire associée à la prise de lansoprazole.

We report two cases of lymphocytic colitis with severe diarrhea in patients treated by lansoprazole for six weeks. Complete remission occurred rapidly after lansoprazole withdrawal.

Surveillance for hepatocellular carcinoma: compliance and results according to the aetiology of cirrhosis in a cohort of 141 patients.

Surveillance for early detection of hepatocarcinoma (HCC) in patients with cirrhosis is widely accepted. In a cohort of 141 patients with cirrhosis collected during the year 1995, we conducted a surveillance program by performing liver ultrasonography and blood alpha-foetoprotein measurement every 6 months. The median follow-up was 34 months. This study addressed to two questions: the compliance to the surveillance schedule according to the aetiology of cirrhosis and the results in terms of emergence of HCC and outcome. Aetiology of cirrhosis was alcohol-induced in 86 (61%), HCV-related in 30 (21%) and from other origins in 25 (18%). Compliance to the program schedule was good in patients with HCV-related cirrhosis (29/30--97%) and patients with cirrhosis of "other origins" (20/25--80%) but was poor in patients with alcoholic cirrhosis (45/86--52%). The lack of compliance was significantly linked to the failure to achieve alcohol abstinence. During follow-up, 6 HCC lesions were observed in 6 male patients with median age of 68 years. All 6 HCC were single nodule, less than 4 cm and accessible to percutaneous acetic acid injection. Nevertheless, the outcome was disappointing, four patients dying 3-15 months later (median: 8 months), two of them with extensive HCC. One of the two patients still alive developed extensive HCC, 36 months after percutaneous acetic acid injection.

Thromboembolism in inflammatory bowel disease.

Thromboembolism represents a severe complication of inflammatory bowel disease occurring in young patient, with active disease. Deep venous thrombosis and pulmonary embolism are the most frequent thromboembolism manifestations. Arterial complications and unusual sites for thromboembolism are more rare. Overall, inflammatory bowel disease is a real prothrombotic state as almost all parameters of coagulation are enhanced. Anticoagulation during the episode of thromboembolism is mandatory, and sometimes may ameliorate the course of inflammatory bowel disease.

Hypoxic hepatitis: a difficult diagnosis when the cardiomyopathy remains unrecognized and the course of liver enzymes follows an atypical pattern. A report of two cases.

In a clinical setting of cardiac or circulatory failure, the diagnosis of hypoxic (ischaemic) hepatitis is easy and can be elicited on mere clinical and biochemical features. We report two cases of hypoxic hepatitis where cardiomyopathy remained unrecognized at admission due to the lack of conventional signs of congestive heart failure and where the increase in liver enzymes activities followed an atypical pattern, characterized by only moderate elevation of serum aminotransferases activities, low ASAT/ALAT ratio and elevated ALAT/LDH ratio. This atypical pattern not suggestive of hypoxic hepatitis, could be explained by a delay between the onset of hypoxic injury of the liver and admission to hospital. Moreover one case was complicated by frank jaundice, an unusual feature in hypoxic hepatitis. Consequently, diagnosis and appropriate inotropic treatment were delayed resulting in progressive deterioration and eventually death of both patients. The report of these two cases and the review of other similar cases previously published, enlighten some atypical features of hypoxic hepatitis.

Severe nontyphoidal salmonellosis probably in relation with omeprazole treatment: report of 2 cases.

We report 2 cases of severe nontyphoidal salmonellosis (Salmonella enteritidis), occurring 4 and 5 weeks after starting a treatment with Omeprazole (20 mg a day). No other member of the families was sick, and none of the 2 patients took any meals outside home during the two weeks preceding the first symptoms. Gastric hypochlorhydria is a major risk factor for Salmonella enteritidis, and several cases of severe infection have been described with this condition. After a review of the literature we conclude that patients with diminished gastric acid run an increased risk of developing Salmonella infection and often with a more serious clinical course.

The effects of lintopride, a 5HT-4 antagonist, on oesophageal motility.

AIM: To evaluate the effects of various doses of lintopride, a new 5HT-4 antagonist with moderate 5HT-3 antagonist properties, on oesophageal body and lower oesophageal sphincter (LOS) motility in humans. METHODS: Eight healthy male volunteers, mean age 22 (19-28) years, without any history of digestive disease or chest pain, were randomized to three doses of lintopride (0.1, 0.3 and 0.5 mg/kg i.v.) and a placebo at 1-week intervals in a double-blind, crossover study. Oesophageal motility was recorded continuously for 4 h after each dose, using perfused catheters inserted at the level of the LOS and in the body of the oesophagus, at 5, 10 and 15 cm from the LOS. Peristalsis was studied during 10 wet swallows, at 30-min intervals (T0-T240 min). RESULTS: The LOS basal pressure (23.3 +/- 2.0 cmH2O; mean +/- s.d.) remained stable after dosing with placebo to T240. After lintopride, LOS basal pressure increased significantly vs. placebo (AUC comparison: 0.1 mg/kg, P = 0.036; 0.3 mg/kg, P = 0.027; 0.5 mg/kg, P = 0.052). In contrast, the duration and extent of LOS relaxation after swallowing was not affected by any of the three doses of lintopride. The amplitude of peristaltic waves in the oesophagus increased significantly at T30 after lintopride 0.3 mg/kg (34.5 cmH2O, P = 0.020) and 0.5 mg/kg (32.5 cmH2O, P = 0.027), at T60 after 0.3 mg/kg (48.8 cmH2O, P = 0.0009) and 0.5 mg/kg (29.1 cmH2O, P = 0.029) and at T90 after 0.3 mg/kg (34.5 cmH2O, P = 0.0018). CONCLUSIONS: Lintopride significantly increased the LOS basal tone without affecting LOS physiological relaxation after swallowing. Peristaltic waves were also enhanced by lintopride.

[Can the L/A ratio identify acute alcoholic pancreatitis?]. / Le rapport L/A permet-il d'identifier les pancréatites aiguës alcooliques?

Early distinction between acute alcoholic pancreatitis is important, because of possible emergency endoscopic sphincterotomy in case of biliary pancreatitis. The aim of this study was to evaluate the value of L/A ratio in the diagnosis of acute alcoholic pancreatitis. From 1990 to end 1993, 133 patients with acute pancreatitis were reviewed. Inclusion criteria were: 1) abdominal pain, 2) pathological serum amylase or serum lipase on admission or within 24 hours after beginning or abdominal pain, 3) acute pancreatitis at the echography or CT scan within 48 hours after admission. 60 patients met the inclusion criteria (31 alcoholic pancreatitis, 19 biliary pancreatitis and 10 pancreatitis of other causes). L/A ratio was studied in terms of delay from beginning of abdominal pain. There was no statistical difference between alcoholic and biliary pancreatitis at any time of the study, with the exception of admission. AST, ALT and alkaline phosphatase were higher in biliary pancreatitis than in alcoholic pancreatitis. AST and ALT were the best biochemical tests to diagnose biliary pancreatitis. Blamey's criteria can also contribute to diagnose biliary pancreatitis. These biochemical tests are the most helpful if they are collected very soon in the evolution of acute pancreatitis. It is concluded that L/A ratio is not helpful in the diagnosis of alcoholic acute pancreatitis.

Assessment of the new immunological test Hemoblot for detecting occult blood in faeces.

Hemoblot, a new immunological faecal occult blood test, produced by Gamma, Angleur, Belgium, was characterized and compared with another immunological test (HemeSelect, SmithKline Diagnostics, USA) and with a guaiac test (Hemoccult II, SmithKline Diagnostics). The analytical sensitivity of Hemoblot is 0.15 mg haemoglobin/g faeces and the test is specific for human haemoglobin. In addition, 135 symptomatic patients who had to undergo a colonoscopy were tested using the three tests. Two criteria were considered for the analysis: (1) the blood criterion: any pathology likely to cause colorectal or other bleeding; and (2) the precancerous-cancerous criterion: the pathology being either a colorectal polyp > 0.5 cm or a colorectal cancer. Considering both criteria, the sensitivity of Hemoblot was significantly higher than the sensitivity of Hemoccult: 38% and 23%, respectively, for the blood criterion; and 54% and 29% for the precancerous-cancerous criterion. Sensitivity and specificity did not differ statistically between Hemoblot and HemeSelect but Hemoblot was faster and simpler to perform. It could be widely used in mass screening.