Major Finding: Evidence of small intestinal bacterial overgrowth was found in 50% of patients on proton pump inhibitors, 24.5% of patients with irritable bowel syndrome, and only 6% of controls.
Data Source: Two hundred gastroesophageal reflux disease patients who were taking PPIs, 200 IBS patients not taking PPIs, and 50 healthy controls.
Disclosures: The study was not funded by outside grants, and neither Dr. Lombardo nor his colleagues reported any competing interests.
Fifty percent of the people taking proton pump inhibitors to treat gastroesophageal reflux disease develop small intestinal bacterial overgrowth, compared with a quarter of the patients with irritable bowel syndrome who are not taking PPIs, according to a study conducted by Dr. Lucio Lombardo and his colleagues.
In the study, 450 consecutive patients underwent glucose hydrogen breath tests, which measure the metabolic activity of enteric bacteria (Clin. Gastroenterol. Hepatol. 2010 [doi:10.1016/j.cgh.2009. 12.022]).
Two hundred of the patients had been diagnosed with gastroesophageal reflux disease and had been taking one of several PPIs for a median of 36 months, although some had taken the medication for as little as 2 months.
The investigators recruited an additional 200 study subjects who had been diagnosed with irritable bowel syndrome (IBS) and were not taking PPIs. Dr. Lombardo and his colleagues noted that the symptoms of IBS—including bloating, diarrhea, and constipation—frequently overlap with the symptoms of small intestinal bacterial overgrowth (SIBO). They also recruited 50 healthy controls who did not have symptoms of either IBS or SIBO and had not taken a PPI for at least 3 years.
Patients with other gastric diseases, who had recent gastric surgery, who were taking antibiotics, or had other potentially confounding factors were excluded.
Evidence of SIBO was found in 50% of the patients on PPIs, 24.5% of patients with IBS, and only 6% of healthy controls, wrote Dr. Lombardo, of the gastroenterology department of Mauriziano U.I. Hospital, Turin, and his colleagues.
Moreover, the researchers found a correlation between the duration of PPI treatment and the detection of SIBO, with more than 70% of the PPI group testing positive for SIBO after 13 months of PPI use—more than triple the proportion of positives among those taking PPIs for a year or less.
Although several studies have used breath-based tests to assess the prevalence of SIBO in patients with IBS, few have been designed to assess the independent influence of PPIs, the investigators wrote. This, they said, is as an “important oversight,” as PPI use is widespread in patients with IBS.
The authors cited a recent study reporting that IBS patients not taking PPIs and GERD patients on PPIs have roughly equal rates of SIBO, as assessed by lactulose breath tests. However, the researchers wrote that no mention was made in that study of the duration of PPI treatment, while their own study showed that PPI treatment increased the incidence of SIBO drastically after the first year.
Dr. Lombardo and his colleagues speculated that PPI-related SIBO might be frequently underdiagnosed or misdiagnosed as IBS because of the overlap of common symptoms.
Patients in the investigation with SIBO were treated with the antibiotic rifaximin 400 mg three times daily for 14 days. Eradication of SIBO, as confirmed by a glucose hydrogen breath test, occurred in 87% of the PPI group and 91% in the IBS group.
In the PPI arm of the study, eradication was more successful among subjects who had taken PPIs for less than a year, which suggested “a more profound or qualitatively different alteration in enteric microflora after a year of treatment,” according to the investigators.
In both the PPI and IBS groups, symptom severity was reduced by more than 90% in subjects whose SIBO eradication had been confirmed by breath testing and to a lesser but measurable degree in those subjects whose SIBO had not been eradicated after the same course of rifaximin.
The investigators did not seek to learn whether SIBO returned after eradication in patients who continued PPI therapy but cited another study suggesting that such an outcome was likely.
The authors noted a few limitations of the study, which included a lack of distinction between specific PPIs taken by the patients, the observational open-label study design, and that fact that Helicobacter pylori was not investigated as an independent contributor (although all 450 patients were tested, with 68% found to be negative).