The study was a prospective observational design, carried out at a Tertiary care Gastroenterology center of the Armed forces. The study period was for two years (July 16 to July 18). The study population included consecutive adult patients undergoing UGIE at the Gastroenterology center, and who were on PPI therapy for at least two months or more. Using data from a previous study [9] and assuming a prevalence of SIBO in PPI users at 50%, along with a 10% margin of error and a 95% confidence interval, a sample size of 80 was arrived at.

The gold standard for SIBO diagnosis is considered aspiration and culture of duodenal content. The diagnosis of SIBO gets established when the bacterial count in proximal duodenal aspirate culture becomes ≥10⁵ bacterial CFU (colony forming units) per mL [5]. At endoscopy, duodenal aspirate was obtained via an aspiration catheter and at least 1 mL small bowel fluid was suctioned into a sterile container. The aspirate was sent to lab and cultured for aerobic bacteria. Growth of >10⁵ cfu/mL was considered positive. All the test reports were informed back to the patients and those with positive cultures were called to OPD for review of medications and a course of antibiotics as per the treating clinicians’ discretion.

The study was approved by the Institutional Ethics Committee (CHAFB/26/2016), and a written informed consent was obtained from all the study participants.

The inclusion criteria were:

The exclusion criteria were:

The data was entered in MS Excel before statistical analysis using Statistical Package for Social Sciences (SPSS ver 21.0). Duodenal aspirate culture results were correlated with the use of PPIs. The data on categorical variables were shown as n (% of cases) and the data on continuous variables were presented as Mean and Standard deviation (SD). The inter-group statistical comparison of categorical variables was done using Chi-square test and, p-values less than 0.05 were considered to be statistically significant.

A total of 80 cases were included in the study. The Mean±SD of age of the entire study group was 45.4±14.2 years. The male to female ratio was 2.6:1. Of the 80 cases who underwent UGIE, the most common finding was gastritis; seen in 30 cases (37.5%). H. pylori infection was detected in 28 (35%) cases. The most common PPI used was Pantoprazole, in 44 (55%) cases. Most patients were on PPI for 6-12 months. Details of the baseline data are given in [Table/Fig-1].

Of the 80 cases studied, 19 cases (23.8%) had positive duodenal aspirate culture examination and 61 cases (76.3%) had a negative duodenal aspirate culture examination. The distribution of prevalence of SIBO, as assessed by Duodenal aspirate culture did not differ significantly across various durations of PPI usage [Table/Fig-2]. The prevalence of SIBO also did not differ significantly between group of cases who had treatment for H.pylori infection and those who did not have it [Table/Fig-3]. The distribution of SIBO did not differ significantly according to history of use of NSAIDs [Table/Fig-4]. SIBO also did not differ significantly according to type of personal habits like smoking or alcohol usage [Table/Fig-5].

The aim of this study was to try and establish a relationship of SIBO with long term PPI therapy. This prospective observational study recruited (n=80) consecutive subjects who were on chronic PPI therapy (>2 months), and looked at positive duodenal aspirate cultures to establish SIBO, which was diagnosed in 23.7% (19/80) in this study population. SIBO was not associated with duration/dose/type of proton pump inhibitor used.

In a similar study by Lombardo L et al., which involved 200 patients on chronic PPI therapy (>2 months), the prevalence of SIBO was reported much higher at 50% [9]. This significantly higher prevalence could have been due to the method used to diagnose SIBO being the much more sensitive Glucose hydrogen breath test (GHBT), rather than more specific duodenal aspirate culture used in this study. Despite the higher prevalence of SIBO, PPI usage did not significantly correlate with duration or type of PPI used. The symptoms of SIBO were primarily bloating in about 50% of cases, diarrhea in 30%, followed by constipation and abdominal pain in 10%. This was similar to the pattern of symptoms noted in the index study.

In another large retrospective study published from Mayo clinic by Franco DL et al., the authors looked at duodenal aspirate cultures obtained from patients undergoing outpatient UGIE [10]. Duodenal aspirates were obtained in 1,263 patients, and the overall yield for positive (>100,000 cfu/mL) duodenal aspirates was 30%. PPIs were used in 37% (466/1263) of these patients. When looking at only PPI usage in the entire study cohort, its use was significantly higher (52.6%) in patients with culture positive duodenal aspirates than those with culture negative aspirates (30.2%; p<0.0001). However, there was no correlation between diagnosis of SIBO and duration of PPI therapy or the type of PPI used.

In a large prospective study from Athens by Giamarellos B et al., quantitative cultures of duodenal aspirates were performed for aerobic species in 897 consecutive patients undergoing UGIE [11]. The overall frequency of SIBO by culture was 10.6%, when the ≥10⁵ cfu/mL diagnostic cut-off was used. Only 184 (20.5%) of these patients had a history of chronic PPI intake. There was no difference between the two groups of patients with and without SIBO, regarding history of PPI intake. There was no association of SIBO with PPI exposure in this large study, similar to the findings in our study. Further, patients with SIBO were older, and had a greater frequency of IBS and Diabetes mellitus.

Using breath samples for hydrogen and methane after glucose challenge (GHBT) Ratuapli SK et al., studied a total of 1191 patients, of which 48% were on PPI. Breath test positivity did not differ significantly between PPI users and non-users [12]. In another study by Erdagon A et al., diagnostic yield of SIBO by duodenal aspirate culture was 45% vis-à-vis 27% by Breath test (GHBT) and both tests did not show significant association of SIBO with PPI usage [13]. Looking at the meta-analysis of data from previous studies [14], and comparing with our data from the present study, it is evident that PPI use is not associated with SIBO, irrespective of the diagnostic modality used and duration of PPI therapy.

We tabulated the data from various studies with similar study parameters [Table/Fig-6] [7,10,11,15,16], and found our data to compare favorably with most of the data on the subject, published over the last decade. Except for one study by Jacobs C et al., all other studies have noted no statistically significant association between chronic PPI usage and SIBO [15].

The major strength of our study is the diagnostic modality used for SIBO is duodenal aspirate culture analysis, which is considered to be the gold standard for diagnosing SIBO. The major weakness of our study is the lack of a control arm. Thus, patients on chronic PPI therapy and pre-existing SIBO could not be conclusively excluded from the study population. Also, the lack of anaerobic cultures may have falsely lowered the true prevalence of SIBO in the study population.

Based on data from this study, it was concluded that SIBO was not associated with dose, duration or type of PPI used. SIBO was also not associated with previous H. pylori infection, NSAID exposure in the past and smoking or alcohol usage. Since this was a hospital based single center observational study without a control arm, further studies with a case-control design will be useful in validating these results in the general population.