Alarm features for predicting significant endoscopic findings among patients with dyspepsia in Cameroon

Abstract

Introduction: 

Dyspepsia is the major indication for gastroscopy in sub-Saharan Africa. However, endoscopic devices are not available in rural areas, and the examination is invasive and expensive for the rural populations. The aim of this study is to assess the predictive value of some alarm features in the prediction of endoscopic findings among adult patients with dyspepsia in Cameroon.

Materials and Methods: 

We carried out a retrospective study using all reports of patients who underwent gastroscopy for dyspeptic syndrome from July 1 to December 31, 2021 at the Yaoundé General Hospital, Cameroon. We documented the sociodemographic characteristics, clinical signs, the type of dyspepsia, and endoscopic findings. We also documented alarm features associated with significant pathology (presence of an ulcer and/or a malignant tumor) at the endoscopy.

Results: 

Of the 310 gastroscopies, dyspeptic symptoms represented 226 indications (72.9%). There were 90 males (39.8%). The mean age was 46.4 ± 15.9 years (range 16–86). Signs associated with dyspepsia were gastrointestinal (GI) bleeding in 8 patients (3.5%), weight loss in 6 patients (2.7%), and persistent vomiting in 13 patients (5.8%). A significant endoscopy finding was observed among 53 patients (23.5%). On the univariate analysis, alarm features associated with a significant lesion were male gender (P = 0.004), persistent vomiting (P = 0.047), and marginally for GI bleeding (P = 0.072). On the multivariate analysis, only the male gender was significantly associated with the presence of a significant lesion (P = 0.024).

Conclusions: 

Our results show that the male gender, GI bleeding, and persistent vomiting are alarm features associated with the presence of a significant lesion among patients with dyspepsia.

BACKGROUND

Dyspepsia is defined as one or more gastric symptoms such as epigastric pain, epigastric burning, postprandial fullness, and/or early satiety for at least 3 months.[¹] It is the major indication of gastroscopy in sub-Saharan Africa (SSA).[²³] It can be caused by a benign or a malignant affection of the upper gastrointestinal (GI) tract.[⁴] International recommendations with Rome IV criteria have suggested some alarm signs of dyspepsia.[⁵⁶] These signs should be able to predict the presence of significant pathology (ulcers and/or malignant tumors) on endoscopy.[⁷] These alarm signs are age (more than 45 years old), gender, unexplained weight loss, evidence of GI bleeding such as hematemesis or melena, anemia, persistent vomiting, dysphagia, personal history of cancer, family history of GI cancer, and lymphadenopathy.[^3458910] The high prevalence of Helicobacter pylori infection in an area could also affect the risk of malignancy.[¹¹] Studies carried out in America and Asia have revealed that the predictive value of alarm signs for significant pathology could be poor.[⁷¹²] In SSA, especially in Cameroon, the prevalence of the H. pylori infection among the general population and patients with dyspepsia is high.[¹³¹⁴] Therefore, the risk of malignancy could be high. But, endoscopic devices are not available in rural areas, and the examination is invasive and expensive for the population,[¹⁵¹⁶] thus justifying this study in order to define selection criteria of patients to refer for gastroscopy. This strategy could be useful in screening significant pathologies and to reduce the cost of the treatment of dyspepsia in rural areas.

Objective

This study aims to assess the predictive value of alarm features in endoscopic findings among adult patients with dyspepsia in Cameroon.

MATERIALS AND METHODS

This was a hospital-based retrospective study, which reviewed the data of patients who underwent gastroscopy from July 1 to December 31, 2021 at the Yaoundé General Hospital, Cameroon. It is a public referral hospital of the political capital of the country. Four competent gastroenterologists perform GI endoscopies daily. We retrieved and documented the data of gastroscopies carried out for dyspepsia. Incomplete gastroscopy reports and children of less than 15 years were excluded. Sociodemographic characteristics include age and sex, clinical features (such as loss of weight, bleeding, persistent vomiting), the type of dyspepsia (epigastric pain or postprandial distress syndrome), and endoscopic findings. The endoscopy findings were either normal or revealed a lesion. In case of any endoscopic lesion, it was described as inflammation, erosion, ulcer, and benign or malignant tumor. We also noted whether or not a biopsy was taken. Furthermore, alarm features associated with significant pathology at endoscopy was documented. A significant pathology was defined as the presence of an ulcer and/or a malignant tumor. Data were analyzed with SPSS version 21. The odds ratio and the corresponding 95% confidence intervals were used to summarize the strength of association between specific binary exposure and outcome variables. The level of statistical significance was set at P-value less than 0.05. A multivariate logistic regression analysis was done for potential risk factors with P-value less than 0.05. Approval for this study was obtained from the Yaoundé General Hospital and the Ethics Committee of the Faculty of Medicine and Biomedical Sciences, University of Yaoundé, Yaoundé, Cameroon.

RESULTS

Three hundred and twenty gastroscopies were performed during the period. Ten incomplete gastroscopy reports were excluded. Of the 310 gastroscopies, 226 (72.9%) were on account of dyspeptic syndrome. There were 90 (39.8%) males and 136 (60.2%) females. The mean age was 46.4 ± 15.9 years (range 16–86) [Table 1]. Among these patients, 115 (50.9%) were at least 45 years old. About 213 out of 226 complained of epigastric pains (94.2%), 19 patients had a postprandial distress syndrome (8.4%), whereas 6 of them had both symptoms. Alarm features associated with dyspepsia were GI bleeding in 8 patients (3.5%), weight loss in 6 patients (2.7%), and persistent vomiting in 13 patients (5.8%). The gastroscopies revealed at least one lesion 216/226 (95.6%), whereas 10/226 gastroscopies were normal (4.4%) [Table 2]. A significant endoscopy finding was observed among 53 (23.5%) patients. The spectrum of lesions was inflammation among 113 (50.0%) patients, erosion among 45 (19.9%) patients, an ulcer in 49 (21.7%) patients, a malignant tumor in 4 (1.8%) patients, and a polyp in 2 (0.9%) patients [Table 2]. Malignant tumors were found in 3 men of 34, 44, and 46 years and in a woman of 53 years. Biopsies were done in 19/226 (8.4%) patients. On the univariate analysis, alarm features associated with a significant lesion were male gender (P = 0.004) and persistent vomiting (P = 0.047) [Table 3]. GI bleeding was marginally correlated with a significant lesion (P = 0.072) [Table 3]. On the multivariate analysis, only male gender was significantly associated with the presence of a significant lesion (P = 0.024) [Table 3].

Table 1: Characteristics of the patients with dyspepsia

Table 2: Alarm features and endoscopic findings

Table 3: Relationship between alarm features and significant endoscopic lesions

DISCUSSION

In our study, there was a female predominance (60.2%). This finding is similar to that reported in other endoscopic studies in SSA.[³⁹¹⁵] This is probably because women take care of their health more than men. In this study, the mean age was 46.4 years, meaning that we have many young adults. Regarding the life expectancy in SSA, the mean age is around 40 years in most studies of this area.[³⁹¹⁵]

The main presentation of dyspepsia was epigastric pain. It is also the main indication for gastroscopy among adults and children.[⁸¹⁶¹⁷] Other presentations of dyspepsia are less common: postprandial fullness and/or early satiety.

Only 10 (4.4%) gastroscopies were normal. These represent functional dyspepsia.[¹⁸] But significant lesions have been described in 53 patients (23.5%). Inflammation such as gastritis (50%) was the most frequent lesion. A predominance of gastritis has also been described in other studies.[³¹⁵] This lesion is non-malignant. The role of endoscopy in dyspepsia is to exclude malignancy. However, upper GI endoscopy could be useful to perform a biopsy to assess the severity of gastritis or to look for H. pylori infection.[¹¹]H. pylori infection is significantly associated with dyspepsia and correlates with its symptoms.[¹⁹] But its diagnosis could be non-invasive.[²⁰]

In this study, using the univariate analysis, male gender, GI bleeding, and persistent vomiting were alarm features which predicted significant lesion on gastroscopy. This result has been described in Nigeria where GI bleeding and persistent vomiting have a good predictive value of malignancy[⁹] as well as in China where male gender was also associated with more risk of malignancy.[⁷] The age above or equal to 45 years and an unexplained loss of weight were not associated with a significant lesion in our study. Unlike, in Asia and Europa, age was not associated with a high risk of malignancy in our study.[⁷] This difference has already been observed in another African country.[⁹] Some studies have shown that GI malignancy could occur in young patients, particularly in Africa.[²¹] Unlike, in some other studies, an explained weight loss is not associated with a high risk of significant lesion. This is not pathognomonic of a GI injury.

The limitation of this study was that data collection was retrospective. For this reason, we have some missing data.

To conclude, our study shows that male gender, digestive bleeding, and persistent vomiting are warning features associated with the presence of a significant endoscopic lesion on patients with dyspepsia.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.