**4. Discussion**

To the best of our knowledge, this case-control study is the first conducted in this region of Najran, SA that considered the none-modifiable factors (age and gender) in modelling relationship between obesity and *H. pylori* positive infection. The key finding was that increased prevalence of HPPI was associated with increased BMI levels, both with and without adjustment for none modifiable factors (age and gender). The findings of this study added to the current debate about the relationship of obesity with HPPI. A recent review showed a significantly inverse correlation between *H. pylori* prevalence and rate of overweight/obesity in countries of the developed world. Thus, the obesity endemic observed in the Western world was attributed to the gradual decrease of the HPPI [6]. These laterresultswerealsosupportedwithstudiesincertainareasinAsianregione.g.,Taiwan,[29].

However, our study findings were in consensus with previous studies reported within SA [9] and elsewhere [11,30]. In Turkey a prevalence of 57.2% of HPPI in Turkish obese subjects compared to 27.0% in normal body weight was found and this further supported our finding [31].

The estimate of HPPI among obese patients who underwent sleeve gastrectomy in this study sample was 66% which was in mid-range (50–88) of previous studies conducted in in SA [13,32,33] and higher than a study among young medical students that showed a low prevalence of 35% [18]. Different study design, participants' recruitments and clinical investigation may reflect such variations in the estimate of HPPI among obese subjects. In addition, the role of the difference in geographic, economic or environmental factors should be considered. Notably, our study was comparable to one US study which found a higher prevalence rate of 61% in morbidly obese patients compared to 48% in the control group [34] though this later used *H. pylori* serologies, while our study used histopathology diagnosis. With respect to the role of age and gender in HPPI, they were found to be non-significantly associated with HPPI and these findings were in consensus with other studies reported in the region [15,33] and elsewhere [35].

As for the strengths of this study, firstly our study findings have substantiated the relationship of obesity with *H. pylori* among obese patients who were matched with none obese patients consecutively recruited in the same setting with a standardized investigation procedures i.e., internal validity plausible. Secondly, this was a hospital-based case-control as such more prone to bias, specifically when selecting participants. However, we were in line with what has been reported previously [30] patients underwent a gastroscopy and received a histological examination of gastric mucosal biopsies as a standard diagnostic examination before bariatric surgery. Thus, the majority of the examined patients were asymptomatic which excludes a strong selection bias toward HPPI. Our study results were of validity as the inclusion and exclusion criteria was adequate to rule out any confounding effects; e.g., use of antibiotics [34]. The enrolment of patients from the same region and in one Centre (Najran Hospital) is another strength of our study, the difference in the estimate of *H. pylori* between the two groups was less likely to represent a differences in geographic prevalence of *H. pylori* as reported elsewhere [34]. Finally, the feasibility of the study with respect of time and resources was additional strength. As for the limitations, the sampling method adopted precludes the generalizability of the study findings to individuals outside of this setting. The patients self-reported of neither tested nor treated for *H. pylori* was not objectively validated, therefore, recall bias was possible. The impact of the latter on the validity of the results might be inevitable. Importantly, in SA the role of race in HPPI should be investigated as elsewhere [29]. The use of the concept of nationality is not sufficient as many Saudi national of different ethnicity/origin backgrounds that presumably influence HPPI. The contribution of other environmental and host factors [36] which has not been controlled for in this study should be considered in future research.
