Epidemiological and Diagnostic Profile of Peptic Ulcer Disease in Upper Gastrointestinal Endoscopy in Libreville ()
1. Introduction
Peptic ulcer is a localized loss of mucosa in the gastric and/or duodenal wall, reaching deep into the muscularis . It results from an imbalance between aggressive and protective factors of the gastroduodenal wall [1]-[3]. Of the several etiologies of PUD, the most frequent is chronic H. pylori infection, as shown by the study of Marshall and Warren [4]. However, H. pylori seems to decrease with the level of industrialization of the country [5] [6] and highlights the African paradox where the prevalence of H. pylori, estimated at 80% by the World Health Organization (WHO), is not associated with a higher frequency of peptic ulcer [7] [8]. It is in this light that we question the impact of H. pylori on the occurrence of PUD in Libreville.
The aim of this study was to establish the epidemiological and etiological profile of patients with PUD and investigate the relationship between H. pylori and PUD in Libreville.
2. Patients and Method
This is a prospective cross-sectional study carried out in the Hepato-Gastro-Enterology (HGE) Department of the Libreville University Teaching Hospital. The study was conducted between 1 January 2017 and 31 December 2019. The study population involved all patients undergoing oesogastroduodenal endoscopy (OGDE). We included patients with gastric and/or duodenal ulcers who gave their consent and had undergone at least two antral and two fundic biopsies for H. pylori, and for whom we had pathological results. We excluded patients who had previously received antibiotic therapy less than 1 month and those who had received a proton pump inhibitor (PPI) less than 2 weeks previously. Using a standardized form, we collected epidemiological data (age, sex, socio-economic level), clinical data (pain characteristics), indication for OGDE, personal history (PUD, notion of H. pylori gastritis, upper digestive hemorrhage, gastro-esophageal reflux disease (GORD), stroke, diabetes, sickle cell anemia, HIV infection, hepatitis B, hepatitis C, cirrhosis), family history (history of PUD, history of stomach cancer), history of taking gastrotoxic substances (anti-inflammatories, anti-platelet aggregates, anti-coagulants, traditional concoctions, tobacco, alcohol), notion of stress, results of OGDE (number of PUD, location and site, Forrest classification, associated endoscopic lesions), anatomopathological results of biopsies (presence or absence of H. pylori, acute gastritis, chronic gastritis, dysplasia, metaplasia, cancer). Data were entered and analyzed using Epi Info 7.2 statistical software. Qualitative variables were expressed as percentages and quantitative variables as means and standard deviation. Comparisons or associations between categorical variables were analyzed using Pearson’s chi-square test corrected according to Yates and Fisher to account for the number of patients. Confidence intervals were expressed at the alpha error threshold equal to 0.05. Variables whose association with the expected event had a p-value of less than 0.05 were considered statistically significant.
3. Results
3.1. Epidemiological Data on Gastric and/or Duodenal Ulcers
During the study period, 1407 patients underwent OGDE and 227 developed gastric and/or duodenal ulcers, representing a hospital frequency of 16.1%. Of the 227 patients, 60 were excluded because biopsies were absent, or they took antibiotics or PPIs (Figure 1).
Figure 1. Flow chart.
The analysis included 167 patients with PUD. There were 88 men and 79 women, giving a sex ratio of approximately 1.1. The mean age of the patients was 48.7 (±17.3) years. H. pylori testing was positive in 133 of the 167 patients included, representing a hospital frequency of 79.3%. The mean age of patients positive for H. pylori was 48.4 (±17.2) years compared to 50.0 (±18.0) years for patients negative for H. pylori. This difference was not statistically significant (p = 0.697). Both groups were predominantly male.
3.2. Diagnostic Data
Typical ulcer pain was present in 69.2% of patients in the H. pylori positive group and 79.4% in the H. pylori negative group. This difference was not statistically significant (p = 0.2919).
Cardiovascular and hepatic pathologies were present to a similar extent in both groups (p = 0.1984). Regarding drug consumption, 6.8% of the H. pylori positive group and 38.2% of the H. pylori negative group reported taking NSAIDs. This difference was statistically significant (p = 0.00014) (Table 1).
Table 1. Distribution of patients with gastric and/or duodenal ulcers according to H. pylori status and medication taken.
|
Number of cases |
H. pylori positive |
H. pylori negative |
p |
|
n |
% |
n |
% |
n |
% |
Non-steroidal anti-inflammatory drugs |
|
|
|
|
0.00014 |
Yes |
22 |
13.2 |
9 |
6.8 |
13 |
38.2 |
|
No |
145 |
86.8 |
124 |
93.2 |
21 |
61.8 |
|
Total |
167 |
100.0 |
133 |
100.0 |
34 |
100.0 |
|
Antiretrovirals |
|
|
|
|
|
|
1.0000 |
Yes |
6 |
3.6 |
5 |
3.8 |
1 |
2.9 |
|
No |
161 |
96.4 |
128 |
96.2 |
33 |
97.1 |
|
Total |
167 |
100.0 |
133 |
100.0 |
34 |
100.0 |
|
Oral hypoglycemics |
|
|
|
|
|
|
0.6889 |
Yes |
10 |
6.0 |
9 |
6.8 |
1 |
2.9 |
|
No |
157 |
94.0 |
124 |
93.2 |
33 |
97.1 |
|
Total |
167 |
100.0 |
133 |
100.0 |
34 |
100.0 |
|
Anti-hypertensive |
|
|
|
|
|
|
0.4683 |
Yes |
32 |
19.2 |
24 |
18.0 |
8 |
2.9 |
|
No |
135 |
80.8 |
109 |
82.0 |
26 |
97.1 |
|
Total |
167 |
100.0 |
133 |
100.0 |
34 |
100.0 |
|
Stress was found in 15.8% and 32.3% of patients with positive and negative H. pylori testing, respectively. This difference was statistically significant (p = 0.0108). Table 2 shows that work-related stress was the most frequent and had a higher frequency in the H. pylori negative group. This difference was statistically significant (p = 0.003). Among patients who did not experience stress, 84.2% were H. pylori positive and 64.7% were H. pylori negative. This difference was also statistically significant (p = 0.0108).
Table 2. Distribution of patients with gastric and/or duodenal ulcers according to stress and H. pylori status.
|
Number of cases |
H. pylori positive |
H. pylori negative |
p |
|
n |
% |
n |
% |
n |
% |
Professional |
12 |
7.2 |
5 |
3.8 |
7 |
20.6 |
0.003 |
Family |
10 |
6.0 |
6 |
4.5 |
4 |
11.8 |
0.121 |
Professional + Family |
4 |
2.4 |
4 |
3.0 |
0 |
0.0 |
0.583 |
Social |
3 |
1.8 |
3 |
2.3 |
0 |
0.0 |
1.000 |
Social + Family |
2 |
1.2 |
2 |
1.5 |
0 |
0.0 |
1.000 |
Professional + Social + Family |
2 |
1.2 |
1 |
0.8 |
1 |
2.9 |
0.367 |
No stress |
134 |
80.2 |
112 |
84.2 |
22 |
64.7 |
0.011 |
Total |
167 |
100.0 |
133 |
100.0 |
34 |
100.0 |
|
The proportion of patients reporting active alcohol consumption was 11.3% in the H. pylori positive group and 26.5% in the H. pylori negative group. This difference was statistically significant (p = 0.0242) as shown in Table 3.
Table 3. Distribution of patients with gastric and/or duodenal ulcers according to H. pylori status and alcohol and tobacco consumption.
|
Number of cases |
H. pylori positive |
H. pylori negative |
p |
|
n |
% |
n |
% |
n |
% |
Alcohol |
|
|
|
|
|
|
0.0242 |
Yes |
24 |
14.4 |
15 |
11.3 |
9 |
26.5 |
|
No |
143 |
85.6 |
118 |
88.7 |
25 |
73.5 |
|
Total |
167 |
100.0 |
133 |
100.0 |
34 |
100.0 |
|
Tobacco |
|
|
|
|
|
|
0.1844 |
Yes |
4 |
2.4 |
2 |
1.5 |
2 |
5.9 |
|
No |
163 |
97.6 |
131 |
98.5 |
32 |
94.1 |
|
Total |
167 |
100.0 |
133 |
100.0 |
34 |
100.0 |
|
Based on endoscopic evaluations, the location of the ulcer was gastric in 102 patients (61.1%), duodenal in 32 patients (19.2%), and gastroduodenal in 34 patients (20.3%). Gastric ulcers were present in 61.6% of patients positive for H. pylori and 55.9% of patients negative for H. pylori. This difference was not statistically significant (p = 0.489). In order of frequency, the site of the ulcer was antral (33.5%), fundic (21%), bulbar (19.2%), antral and bulbar (12%), antral and fundic (6%), fundic and antro-bulbar (4.8%), and fundic and bulbar (3.6%). According to H. pylori status, there was a statistically significant difference between the antral site (38.3%) and the fundal site (13.5%), with 22.6% and 13.5% respectively (p = 0.019 and p = 0.020). According to H. pylori status, the antral site with 38.3% and 13.5% in patients with positive and negative H. pylori, respectively, as well as the fundic site with 22.6% and 13.5%, showed a statistically significant difference (p = 0.019 and p = 0.020) (Table 4).
Forrest classification was, in order of frequency, stage III (65.3%), stage IIb (16.8%), stage IIc (15.0%), stage IIa (1.8%) and stage Ib (1.2%). There was no statistical difference according to H. pylori status.
Table 4. Distribution of patients according to site of gastric and/or duodenal ulcer and H. pylori status.
|
Number of cases |
H. pylori positive |
H. pylori negative |
p |
|
n |
% |
n |
% |
n |
% |
Antrum |
56 |
33.5 |
51 |
38.3 |
5 |
13.5 |
0.019 |
Fundus |
35 |
21.0 |
30 |
22.6 |
5 |
13.5 |
0.020 |
Bulb |
32 |
19.2 |
20 |
15.0 |
12 |
32.4 |
0.779 |
Antrum + fundus |
10 |
6.0 |
8 |
6.0 |
2 |
5.4 |
1.000 |
Antrum + bulb |
20 |
12.0 |
14 |
10.5 |
6 |
16.2 |
1.000 |
Fundus + bulb |
6 |
3.6 |
2 |
1.5 |
4 |
10.8 |
0.575 |
Fundus + antrum + bulb |
8 |
4.8 |
8 |
6.0 |
0 |
0.0 |
0.115 |
Total |
167 |
100.0 |
133 |
100.0 |
37 |
100.0 |
|
In terms of pathology, all patients had a histological abnormality. The most frequent histological lesion was chronic gastritis with 90.2% in the H. pylori positive group and 79.4% in the H. pylori negative group (p = 0.371). Acute gastritis lesions were 6.8% in the H. pylori positive group and 17.6% in the H. pylori negative group. This difference was statistically significant (p = 0.047). For precancerous lesions, dysplasia represented 1.8% and metaplasia 9.6% with no statistically significant difference between the two groups (p = 0.384 and p = 0.023). We observed one gastric adenocarcinoma in the H. pylori positive group and none in the other group.
4. Discussion
Study limitations
The fact that immunohistochemistry is not routinely used to detect H. pylori infection means that the incidence of this condition may be underestimated. However, pathological analysis remains the gold standard for diagnosing H. pylori infection [9].
Epidemiology
The hospital incidence of PUD was 16.1% in upper GI endoscopy at the Libreville University Teaching Hospital. This frequency, similar to that found in Togo (15.53%) and Cameroon (17.1%), was higher than that reported by Perret et al. in Gabon (8.75%) and Diouf et al. in Senegal [10]-[13]. These differences support the observation by Archampong et al. in Ghana that the development of digestive endoscopy in Africa has led to a significant increase in the number of cases in recent publications [14]. The mean age was 48.7 ± 17.3 years, similar to that found in all African countries [10]-[15]. However, it remains lower than in studies conducted in the Western countries, where it is 15 years to 20 years higher [16] [17]. The male predominance of patients with PUD (52.7%) is consistent with the literature [10]-[18]. Analysis of the socioeconomic profile reveals a predominance of the disease in participants with low socioeconomic status (65.9%). This same observation was made in Cameroon, including 64.9% of patients with PUD by Ankouane et al. . Diagnosis Clinically, the typical ulcer syndrome was present in 71.3% of patients. It was also observed in 64.3% of patients in Congo by Ontsira et al. and in 64.52% of patients in Togo by Lawson et al. [10] [19]. The most common location of the ulcer was gastric (55.1%), whereas it was duodenal in the other African studies [10] [11] [13]-[15]. This gastric location of PUD in Gabon was already highlighted by Perret et al. more than 30 years ago [12].
In terms of etiology, H. pylori infection was present in 79.6% of patients. Although this prevalence is close to that found in Cameroon (72.5%), Congo (72%), Togo (70.41%), and Senegal (83.6%), it contrasts with previous Gabonese data which found a prevalence of H pylori of 37.5% in endoscopy and 62% in pediatrics [10]-[13] [15] [21]-[23]. This difference with previous Gabonese data is probably a consequence of methodological differences. We excluded patients who had taken antibiotics and PPIs, which was not the case in previous studies. In addition, comparative analysis of other known risk factors for PUD revealed a statistically significant difference between the absence of H. pylori infection and the use of NSAIDs (p = 0.00014), as well as alcohol consumption (p = 0.0242) and occupational stress (p = 0.003). These other risk factors, although poorly identified in African studies, are among the most common causes of PUD in the West [10]-[20]. Similarly, the predominant gastric location of positive H. pylori PUD and the predominant duodenal location of PUD without H. pylori, reveals a decrease in gastric protective factors during H. pylori infection and an increase in aggressive factors when taking NSAIDs, alcohol, and occupational stress [16]-[18].
5. Conclusion
PUD is a frequent pathology in upper digestive endoscopy in Libreville, preferentially affecting men of low socioeconomic status aged under 50. It is revealed by a typical PUD syndrome with a common endoscopic site in the stomach. H. pylori infection is the main etiology, although the use of NSAIDs, alcohol, and occupational stress are risk factors that need to be clarified in our context.