[Prévalence de l'infection à Helicobacter pylori et des lésions endoscopiques hautes chez les diabétiques: Etude cas/témoins]. / Prevalence of helicobacter pylori infection and gastroscopic finding in diabetic patients: a case-control study.

BACKGROUND: Current data on the prevalence of Helicobacter pylori infection in dyspeptic diabetic patients are contradictory in the literature. AIM: the aim was to assess the prevalence of Helicobacter pylori infection, gastroscopic lesions, and gastric histopathological lesions, in dyspeptic diabetic patients. METHODS: It was a case-control study collecting 394 dyspeptic patients (194 diabetic and 200 nondiabetic patients). RESULTS: The average age of patients was 47 years. 144 patients (47%) were male and 150 patients (53%) were female. The two patient groups were matched for age and sex. The prevalence of Helicobacter pylori infection was comparable between the two groups of patients (85% in diabetics versus 90% in the controls). The frequency of gastroscopic lesions was 50% in diabetics and 55% in controls with no significant difference between the two groups. At histology, the prevalence of chronic gastritis, intestinal metaplasia, and gastric atrophy was 85%, 13% and 39% respectively in the group of diabetic patients. These results were comparable to those found in patients without diabetes. CONCLUSION: Our work shows no difference between diabetics and non-diabetics on the prevalence of Helicobacter pylori infection, gastroscopic, and gastric histopathological lesions.

The natural history of hepatitis B virus cirrhosis after the first hepatic decompensation in Tunisia.

AIM: To define the natural long term course of viral B cirrhosis after the onset of hepatic decompensation and to determine the predictive factors of death. METHODS: Retrospective longitudinal study including 77 cases of viral B cirrhosis among 192 consecutive patients with cirrhosis, hospitalized between 1997 and 2005 for the first hepatic decompensation. All those patients were followed- up until death or until December 2006. The probability of survival after the first hepatic decompensation was calculated using the Kaplan Meier method. The predictive factors of death were determined through univariate and multivariate analyses with the Cox regression model. RESULTS: Fifty four men and 23 women with an average age of 54±14.9 years were hospitalized for the first decompensation of the viral B cirrhosis. The 77 patients had been under observation for an average period of 24.2 ±21.1 months. During that time 64% among them died. The probability of survival after decompensation was 47% in 2 years and 22 % in 5 years. During follow- up, ascites was the most frequent decompensation (85%) followed by hepatic encephalopathy (38 %), variceal hemorrhage (34 %), jaundice (30%), hepato renal syndrome (27%), hepatocellular carcinoma (21%), and spontaneous bacterial peritonitis (14%). At univariate analysis four factors were predictive of death: Child Pugh C score (p=0.009), hepatocellular carcinoma (p=0.01), rate of serum gammaglobulin superior to18g / l (p=0.008) and prothrombin time inferior to 50 % (p=0.02). According to the multivariate analysis only the rate of serum gammaglobulin superior to 18g /l was an independent predictive factor of mortality (p=0,001) with IC (95 %) [1.623 - 5.88]. CONCLUSION: In Tunisia, the prognosis of viral B cirrhosis after the first decompensation is bad, because a patient on 5 only was able to survive beyond 5 years. Ascites is the most frequent decompensation. Only the rate of serum gammaglobulin superior to 18g / l is an independent predictive factor of mortality.

[The particularity of Helicobacter pylori chronic gastritis in the west center of Tunisia]. / Les particularités de la gastrite chronique a Hélicobacter pylori au centre ouest de la Tunisie.

BACKGROUND: The Helicobacter pylori (HP) is strongly associated with chronic gastritis. The aim of our study was to determine prevalence of the HP chronic gastritis in the west center of Tunisia and to clarify the clinical and the histological particularities of this infection. METHODS: it is a prospectif study carried out from September 2002 to July 2005 and had included 352 patients. All the patients had an upper endoscopy in which five gastric biopsies were taken for a histological study. The biopsy specimens were fixed in 10% buffered formol and then included in paraffin. The specimens were cut at 4 microm and stained with Hematoxyline Eosine, modified Giemsa and Alcian bleu SAP. The histological examination revealed HP specimens, the lymphoplasmocytic infiltration, the gastritis activity, the gastric atrophy and the intestinal metaplasia according to Sydney system. RESULTS: They were 162 men and 190 women,with a mean age of 48.3 years. The erythematous antral gastritis was the most frequent at endoscopy (26.7%). The prevalence of HP chronic gastritis was 89% in patients with endoscopic lesions. The activity of the gastritis was 89.7% in the antrum and 52.2% in the fundus. The activity mean score in the antrum was 2, 1.8 and 0.3 respectively in the case of duodenal ulcer and gastric adenocarcinoma vs 0.68, 1.1 and 0.16 in gastric fundus. The prevalence of gastric atrophy was 35%. The atrophy was found in the antrum in 98% of cases. The prevalence of intestinal metaplasic was 11%. CONCLUSION: the HP chronic gastritis is very frequent in the west center of Tunisia. Its prevalence is found elevated since the adolescence. It predominate the gastric antrum. It is often associated with a duodenal ulcer rather than gastric cancer.

[Predictive factors of recurrent hemorrhage in duodenal ulcer: report of 208 cases]. / Facteurs prédictifs de récidive hémorragique de l'ulcère duodénal. A propos de 208 cas.

The aim of this study was to evaluate clinical, biological and endoscopic predictive factors of recurrent bleeding. Two hundred and eight patients (169 males/39 females) with a bleeding from duodenal ulcers, confirmed by an endoscopy, were included in a retrospective study performed over 3 years. Forrest classification was used to describe endoscopic stigmata of bleeding. Those who are Forrest la and lb were directly operated. Twenty two patients (10,5%) had rebled. Based on bivariate analysis, the predictive factors of recurrent bleeding were the site of ulcer, a shock at admission, ulcer size D 2.0 cm and the presence of stigmata of bleeding (Forrest II) p<0,05. Only endoscopic stigmata of bleeding and shock were predictive factors of recurrent bleeding in multivariate analysis. Mortality with 2.4% of rate, was strongly linked with recurrent bleeding in elderly and ill people.