Prevalence of Helicobacter pylori in dyspeptic patients at a tertiary hospital in a low resource setting

We set out to establish the prevalence of H. pylori gastritis among patients presenting for endoscopic examination.

We found that over 36% had H. pylori gastritis and that it took 57 weeks on average for the participants to access endoscopic
services from the time of developing symptoms. H. pylori prevalence goes from less than 15% in some populations to virtually 100% depending
on socio economic status and country development. In High income Countries exposure
tends to occur later in life, which results in lower percentages in infected adults,
an average of 20–30% of adults are infected by age of 50. In this study those ?40 years
was 55%, twice as much as stated for some High Income Countries. It is anticipated
that the prevalence of H. pylori infection will decline as sanitary conditions improve and it is also a reflection
of wide spread use of antibiotics 10].

Helicobacter pylori infection has been reported by several studies to be high in developing countries,
and associated with low levels of education, low social economic status, and poor
sanitation 4], 11].

NSAID use and H. pylori infection have a significant impact on endoscopic findings while presence of H. pylori, smoking and alcohol consumption are all associated with increased risk of developing
chronic gastritis 11]. In this study the consumption of cigarettes was low however the consumption of alcohol
was moderate.

The overall prevalence of H. pylori associated gastritis may be an under estimate since the majority of the participants
had been on prior empirical treatment with antibiotics. However, this value is similar
to one found by Wabinga et al. 12] in his retrospective study in 2002. In contrast, studies in 2012 in neighboring Kenya
13] reported a 52% prevalence of H. pylori in adults 14].

In Nigeria, a 41% prevalence was reported in Lagos state in 2008 11]. Notably, all the above studies were carried out in an urban settings where there
is congestion, perhaps sub optimal sanitary conditions but easy access to antibiotics
compared to rural settings.

In the East Cape province of South Africa in 2008, 66.1% H. pylori prevalence was reported 15]. The East Cape Province is one of the poorest provinces of South Africa.

The majority of patients (close to 80%) had been on prior empirical treatment as opposed
to the test and treat approach practiced in some countries with a low prevalence of
H. pylori.

However close to half of the H. pylori infected adults had not been on antibiotics at the time of endoscopy. Also one patient
had H. pylori despite taking eradications antibiotics for over 4 weeks prior to endoscopic examination.
Perhaps a case of resistance to the regimen used.

The odds of H. pylori positivity were reduced in those who had taken antibiotics for more than 2 weeks.
OR = 0.62 (0.06–5.93). A possible explanation is that eradication is likely achieved
after 2 weeks of therapy.

The time from onset of symptoms to the endoscopic procedure was not normally distributed.
The minimum duration of symptoms before endoscopy was 3 weeks, and the maximum 1,248 weeks.
The average time to endoscopy was 125.6 weeks, and the median 57 weeks. This is the
first study that looked at the time duration from onset of dyspeptic symptoms to the
endoscopic procedure in Uganda. This reflects on the limitation to accessing endoscopic
services. The findings suggest that the average time from onset of symptoms to the
definitive diagnosis by endoscopy is about a year. This might be a long time, bearing
in mind that H. pylori is a potent risk factor for malignancy 12]. In this study, a 37 years old presented with dyspepsia (without alarm symptoms),
was found with moderate H. pylori gastritis of the antrum and body and adenocarcinoma in situ. Clinicians have to be
aware that dyspepsia in the young could be cancer 16]. In a systematic review of 4,018 patients the use of alarm symptoms to select dyspeptic
patients for endoscopy caused patients with early curable cancers to be overlooked
17].

The commonest macroscopic finding was gastritis similar to Kagimu et al. 12], Wabinga et al. in Uganda 12], Sang Thomas et al. 13] and Kimang et al. 14] In Kenya, Sang Thomas et al. found 4% normal findings at endoscopy and Kimangi et
al. found that 100% of the study participants with dyspepsia had abnormal findings
at endoscopy. In the Eastern Cape province S.A, 33.6% of the patients had functional
dyspepsia 15]. Similarly, Abioudun et al. in Nigeria found that gastritis was the commonest finding,
but also found a high incidence of H. pylori in the endoscopically normal study participants 18]. In contrast, in Italy, a case control study by Zagari et al. of 1,033 study participants
found that ¾ were normal endoscopically. In his study, 93.4% of PUD patients had H. pylori at histology 19]. In Korea, Jung et al. and colleagues found that 40% of the study participants were
normal endoscopically in 2012.

Despite the relatively high prevalence of H. pylori infection, only a few study participants had severe disease (PUD and malignancy).
Two 2] participants had H. pylori associated duodenal ulcers, and three patients had malignancy. Sang Thomas et al.
in Kenya found that 30% of his study participants had PUD with H. pylori and 73.6% of these had H. pylori13] of the 82 patients with Gastritis 7 (8.5%] had severe gastritis at endoscopy and
their histology confirmed H. pylori presence.

This study found that seven participants (6.3%) had hiatus hernia at endoscopy presented
with dyspepsia plus heart burn. This is in contrast to the majority of the studies
mentioned above that all had less than 2% of these cases and reflux oesophagitis.

Study limitations

One major limitation of this study is that it was conducted in a hospital setting
and this may not be a true representation of the prevalence of H. pylori among dyspeptics in the general population. A community study is therefore desirable.
Study involving asymptomatic controls was desirable but because endoscopy is expensive
and invasive, it was not possible to get study participants.

The study participants did not have an abdominal ultra sound done to exclude other
causes of undiagnosed dyspepsia, pancreatitis and hepatobiliary disease, especially
for those that had normal endoscopic findings.

There could have been recall bias as participants were required to remember when their
symptoms started.