Gastroduodenal lesions in upper gastrointestinal endoscopy associated with positive Helicobacter pylori histology in cirrhotic patients at Sikasso Hospital: prevalence study

¹Service of Gastro-Enterology, Regional Hospital of Sikasso, Sikasso, Mali, ²Reproductive Health Division, Regional Health Direction, Mopti, Mali, ³Service of Laboratory, Blood Bank, Regional Hospital of Sikasso, Sikasso, Mali, ⁴Department of Urology, Hassan II University Hospital Center, Fez, Morocco, ⁵Service of Internal Medicine, Regional Hospital of Sikasso, Sikasso, Mali, ⁶Administration Division, Regional Health Direction, Mopti, Mali, ⁷Service of Gastro-Enterology, Gabriel Touré University Hospital, Bamako, Mali

^&Corresponding author
Oumar Traoré, Service of Gastro-Enterology, Regional Hospital of Sikasso, Sikasso, Mali

Introduction    

Cirrhosis is the ultimate stage of all chronic liver diseases and as a result it causes enormous complications, some of which are life-threatening in the short and medium term. Upper gastrointestinal bleeding is one of the most dangerous complications in cirrhotic patients. Among the main causes of these hemorrhages in cirrhotic patients are: rupture of esophageal varices which is the most frequent (57.4%); followed by gastric ulcers (13.9%) and duodenal ulcers (11.1%) [1]. The prevention of these hemorrhages involves searching for all causes such as digestive lesions, in particular gastroesophageal varices and gastric ulcers likely to cause bleeding. Hemodynamic changes in cirrhotic patients lead to significant weakening of the gastroduodenal mucosa, which causes certain disorders such as those of its vascularization. The presence of Helicobacter pylori is a major contributor to the genesis of peptic ulcer disease, although its role in the pathogenesis of ulcer in cirrhotic patients has not yet been well established. The objective of this study was to determine the prevalence of gastroduodenal lesions associated with positive histological Helicobacter pylori in cirrhotic patients at the Sikasso Regional Hospital.

Methods     

It was a retrospective study of cirrhotic patients due to various etiologies, collected from January 2017 to May 2018 at Sikasso Hospital. All patients included in this study underwent esophagogastroduodenoscopy (EGD) during which two antral biopsies, two fundic biopsies and one angulus biopsy were performed. A histological study of the samples taken was performed at the pathology department of Bamako to confirm the presence of Helicobacter pylori. The gastroduodenal ulcers objectified by the EGD were classified according to the FORREST classification. Non-cirrhotic patients, those who did not perform EGD, as well as patients whose histological examination of the biopsies were negative for Helicobacter pylori, were not included in this study. The data were collected from the endoscopy and results of the anatomy-pathology examinations requested registry. The data was analyzed using Epi Info software version 7.0. All quantitative variables were described in terms of mean plus standard deviation and qualitative variables in terms of percentage.

Results     

A total of 34 patients were used in our study. The average age was 38 ± 17 with extremes of 16-81 years and a male predominance (67.6%). In our case study patients, there were post hepatitis viral B cirrhosis in 21 patients, cirrhosis due to viral hepatitis C in 8 patients, post-ethyl cirrhosis in 1 patient, primary biliary cirrhosis in 1 patient, and cryptogenic cirrhosis in 3 patients. Clinically, a total of 17 patients (50%) had gastrointestinal symptoms; epigastralgia (26.47%), nausea (8.82%), early postprandial vomiting (5.88%) and hematemesis (8.82%). The esophagogastroduodenoscopy revealed gastroesophageal varices in 47% (16 cases), of which 11.76% (4 cases) were esophageal varices grade III without red signs, 1 case of esophageal varices grade III with red signs, 5 cases (14 cases), 70% of esophageal varices grade II without red signs, 2 cases (5.88%) of esophageal varices grade II with red signs, 3 cases (8.82%) of esophageal varices grade I without red signs and 1 case of oesogastric varices (GOV) grade 1. Portal hypertension gastropathy was noted in 12 patients (35.29%), gastroduodenal lesions in 33% of which 17.64% were erythematous antro-fundic gastritis, 2 cases (5.88%) of antral gastric ulcers and 3 cases (8.82%) of antral ulcerations. Peptic oesophagitis was present in 20% of cases, a simple bulbitis in 13% of cases, an uncomplicated hiatal hernia in 5.88% of cases. Using the FORREST classification, stage III gastroduodenal ulcers were the most recorded. Anatomopathological examination of the biopsies carried out revealed Helicobacter pylori in 57% of cases, active chronic gastritis in 44.11% (15 cases) and chronic gastritis with intestinal metaplasia in 1 case (2.94%). Management of all lesions observed was done at Sikasso Hospital.

Discussion     

The principal esogastroduodenal lesions observed during digestive endoscopy in cirrhotic patients constitute a deleterious prognostic factor because they can cause digestive hemorrhages that can lead to short life terms in these patients. Helicobacter pylori, known as the most incriminated bacterium in digestive tract lesions, requires special attention in cirrhotic patients. It is with the aim of providing early, adapted and effective management in cirrhotic patients, we conducted this study to determine the frequency of gastroduodenal lesions associated with positive histology for Helicobacter pylori in Sikasso Hospital. A total of 34 patients were collated over a 17-month period. The average age was 38 ± 17 years and the male/female sex ratio was 2.09. In the study reported by Abdoul et al. and Kim et al. average ages were much higher than ours [2,3]. In our study, this apparently high average age could be explained, on one hand, by the insufficiency or even the absence of early follow-up of patients who are unaware of their cirrhotic status, most often revealed at the decompensated stage, but also on anther hand, by the low level of education of the population in general. The sex ratio between men and women of 2.09 is close to those reported by Calvet et al. [4] and Kim et al. [3] which were 2 and 3.9 respectively but significantly higher than that found by Abdoul et al. [2]. The gastroduodenal lesions represented the most frequent lesions of our endoscopic observations (33%). These results corroborate most of the studies carried out in Africa and the West [5,6], but the peculiarity in Africa is that Helicobacter pylori infection is very frequently associated with these digestive lesions in cirrhotic patients as evidenced by our data (57%).

In our context, this high prevalence of Helicobacter pylori in gastroduodenal lesions in cirrhotic can be explained by the low income of the population (poverty) in general, which is a factor favoring infection with Helicobacter pylori. Literature also reports that, in general, this prevalence of Helicobacter pylori in gastroduodenal lesions in cirrhotic patients varies according to socioeconomic level [7]. However, our results on the Helicobacter pylori infection rate (57%) are lower than those found in a Malagasy study [8]. In clinical terms, epigastralgia was the most common reason for consultation as reported by some authors [2,9]. Upper gastrointestinal bleeding was observed in three out of thirty-four patients and is one of the most common complications in cirrhotic patients. Literature reports that these hemorrhages are 4 to 10 times more frequent in patients infected with Helicobacter pylori compared to uninfected patients, whereas in the western countries, non-steroidal anti-inflammatory drugs (NSAIDs) are the most common causes of gastroduodenal ulcers (UGD) [10,11] because they are prescribed for low-dose long-term use in the majority of the population because of advanced age-related factors [12]. In our series, the gastrointestinal hemorrhage observed in the patients is explained on the one hand by the ulcerous lesions associated with portal hypertension gastropathy but also by the gastric infection with Helicobacter pylori, an aggression factor in the pathogenesis of the gastric lesions in an already weakened terrain by cirrhosis. Esophageal varices did not show signs of obvious bleeding apart from the red signs observed in two cases.

Conclusion     

This study reveals a fairly high incidence of Helicobacter pylori in the digestive lesions observed in cirrhotic patients. Helicobacter pylori infection in cirrhotic terrain requires urgent therapeutic management to prevent hemorrhagic complications due to the risk of compromising the vital prognosis in the short and medium term. It also reveals a high frequency of gastroduodenal lesions in cirrhotic patients since one in three patients included suffered from these lesions.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Oumar Traore: study design, data collection, edition of the final report and edition of this manuscript; Abdoul Salam Diarra: study design, data statistical analysis, edition of the final report and edition of this manuscript; Oumar Kassogue: edition of final report and reviewing the manuscript; Tawfiq Abu: reviewing and translating the manuscript in English; Mamadou Mariko: edition of final report and edition of this manuscript; Aguissa Maiga: edition of the final report and edition of this manuscript; Moussa Diarra: edition of this manuscript. All the authors have read and agreed to the final manuscript.

References