Dysmotility and Proton Pump Inhibitor Use Are Independent Risk Factors for Small Intestinal Bacterial and/or Fungal Overgrowth

C. Jacobs; E. Coss Adame; A. Attaluri; J. Valestin; S. S. C. Rao


Aliment Pharmacol Ther. 2013;37(11):1103-1111. 

In This Article


In this study, we evaluated the effects of dysmotility and PPI use on small bowel bacterial and fungal overgrowth by performing a comprehensive assessment in a cohort of patients with chronic gastrointestinal symptoms without clear aetiology and whose overlapping symptom profiles could not be categorised based on traditional Rome criteria.

We found evidence for small intestinal bacterial and/or fungal overgrowth as defined by the presence of increased numbers of bacteria or fungi in 63% of this patient population. Furthermore, the majority of patients (76%) with dysmotility had SIBO, SIFO or Mixed SIBO/SIFO. Similarly, 75% of patients on PPI therapy had cultures that were positive for bacteria and/or fungal organisms.

We found a higher prevalence of overgrowth in PPI users than those recently reported [100/200 (50%)] by Lombardo et al.[7] The higher yield in our study is most likely due to the use of small intestinal aspirate and aerobic/anaerobic/fungal culture (generally considered a gold standard, although invasive) for a diagnosis of overgrowth as opposed to the use of a less sensitive glucose hydrogen breath test (sensitivity and specificity ranges are 20–93% and 30–86% respectively) in the previous studies.[14–17] Also glucose breath test cannot diagnose SIFO.

Choung et al. found that patients taking PPIs more commonly had an abnormal culture result than patients who were not taking PPIs, although there was no significant relationship at bacterial concentrations ≥105 CFU/mL.[17] There was, however, a significant association between PPI use and 'indeterminate' bacterial concentrations (0 through less than 105 CFU/mL).[17] They suggested that PPI use may be associated with a 'low grade' form of SIBO. In the present study, we used the bacterial concentration ≥103 CFU/mL, which is consistent with the findings of Choung et al.[17] However, unlike their study, we did find significant differences in the yield of positive test for SIBO between the two bacterial concentrations. Furthermore, the bacterial concentration at which patients may experience symptoms, in the context of SIBO is not known.

Regarding the cut offs for bacterial counts, we have provided data for both, the conventional ≥105 counts and for our proposed ≥103 counts. Although we found a significant difference between the two bacterial concentrations for the prevalence of SIBO, even with the higher bacterial concentration (>105 CFU/mL), we found a significant association for the presence of SIBO/SIFO with dysmotility and PPI use. This finding further attests to the validity of these proposed risk factors. However, for most of the key data analyses and discussion, we used the lower cut off values since we felt that this count may be more representative of the normal bacterial concentration in the duodenum because of its close proximity to the acid environment in the stomach than the higher cut off that is typically used for any microbial infection. Furthermore, we wish to acknowledge that whether otherwise healthy subjects have any bacteria or fungus in this duodenal segment and at count of 103 is not known. We believe that further research regarding bacterial concentrations in different gut segments in healthy subjects is needed.

We hypothesised that the combined presence of two factors such as dysmotility and PPI use may predispose to a higher prevalence of SIBO, but this was not the case, either because the individual prevalence was moderately high or because of the heterogeneity of our patient groups or that these two factors are independent and not related.

One novel finding of our study was the detection of fungal organisms in the 24 patients who only had a positive culture for candida. Unlike SIBO, where there is some recent information, there is virtually no data regarding normal concentrations of fungi in the proximal small bowel. This may in part be due to the slow-growing nature of the fungal organisms and also a lack of knowledge of this possibility. Although the concentration of fungal overgrowth in the proximal bowel is considered to be very low,[13] we identified that 27% of our patients had positive fungal culture. This observation merits further study and confirmation.

Although candida infections are usually seen in the neonatal, elderly or immunocompromised[13,18,19] individuals or those on steroids, or repeated antibiotic use, our findings suggest that fungal organisms are not uncommonly present in patients with chronic GI complaints. In one study, candida was the most common organism identified in nasogastric aspirates from the proximal GI tract of preoperative patients with GI disorders (malignancy, inflammatory bowel disease, and benign conditions).[20] Another study identified fungal growth in stool cultures of six patients with diarrhoea and abdominal pain.[21] Apart from these anecdotal case reports, there has been no systematic study of the prevalence and clinical presentation of small intestinal fungal organisms in patients with chronic, unexplained GI symptoms.

The clinical manifestations of SIBO are nonspecific and include symptoms such as gas, bloating, abdominal pain and diarrhoea. More serious manifestations of bacterial overgrowth of the small bowel include malabsorption syndromes, weight loss, malnutrition, vitamin deficiency and anaemia.[22] Symptoms of SIFO are, however, not known but based on our study we believe that SIFO shares the same set of symptoms as SIBO. Our study was unable to identify a single symptom or cluster of symptoms that can clinically recognise patients with either SIBO or SIFO. Thus, symptoms were generally poor predictors of bacterial and/or fungal overgrowth. However, SIBO and/or SIFO were prevalent in over 50% of this patient population with dysmotility and chronic use of PPI and should be considered in the differential diagnosis of patients with nonspecific chronic GI complaints.

We would advocate screening for SIBO in symptomatic patients who have known dysmotility or taking PPIs. The glucose breath test is widely used as a non-invasive method of diagnosing SIBO.[23] Screening for SIBO can start with a non-invasive glucose hydrogen breath test and, if test results are negative and if clinical suspicion remains high, aspiration and culture of small bowel contents may be considered. At present, culture of small bowel aspirate appears to be the only method of identifying fungal organisms in the small bowel.

There are some limitations to our study that should be considered when interpreting these findings. Because our documentation relied on prospective patient questionnaire and medical records, we were unable to accurately determine the duration of PPI treatment in all our subjects. It has been suggested that longer durations of PPI therapy are associated with an increased risk of SIBO.[7] Another limitation may involve collection and culture of organisms. Although sterile techniques were consistently used, and by a single operator, it is possible that collecting specimens at precisely the same point along the small intestine is difficult. However, the high prevalence of bacteria and candida in duodenal aspirates attests to the presence of these organisms in a significant proportion of this population. The symptom profiles were based solely on the bowel symptom questionnaire and are prone to subject bias, and may account for a lack of difference, especially as these patients had long standing refractory symptoms and were referred to a tertiary care centre. Also, other factors such as visceral hypersensitivity and IBS may have been present in these patients leading to reports of greater symptom severity. Also, our results from a selected population presenting to a tertiary care specialist centre should be interpreted with caution and may not be generalisable to the population at large.

We conclude that dysmotility and PPI use appear to be important and independent risk factors associated with an overgrowth of small intestinal bacteria and/or fungal organisms. Symptom profiles by themselves are poor predictors of overgrowth in this patient population.