gastric acid suppression and the microbiome

it has been shown in several studies that gastric acid suppression puts people at risk for several infections, including pneumonia, c difficile, and gastroenteritis. In fact, one of the predisposing conditions to developing disseminated TB infection is post-gastrectomy, presumably because stomach acid inhibits TB dissemination (there are several reports in the literature, mostly case-controlled studies, finding that acid suppression also is associated with TB infections). well, not to shock you all, but it turns out that acid suppression leads to significant changes in the gut (gastric bacterial overgrowth) as well as the lung microbiome (see microbiome changes with acid suppression JAMA Peds 2014 in the dropbox, or doi:10.1001/jamapediatrics.2014.696​). this study was a 5-year prospective one in kids aged 1-18 yo who were getting bronchoscopy and endoscopy to evaluate chronic cough, with a questionnaire about used of acid-suppressant medications. the researchers assessed the difference in concentration and prevalence of bacteria in the stomach and lung, comparing those on acid suppressants (AS) to those not. a total of 99 patients were assessed -- 48 on acid suppression (45 with proton pump inhibitors, 3 with H₂-blockers), and 51 not. results:

--gastric bacterial growth: 46% in those on suppressants, 18% not.  specifically, more staphylococcus (prevalence ratio 12.75), streptococcus (prevalence ratio 6.91), Veillonella (prevalence ratio 9.56), Dermabacter (prevalence ratio 4.78) and Rothia (prevalence ratio 6.38)​. those on  AS had both higher prevalence and higher concentrations of the bacteria.
--bronchoalveolar lavage growth: no diff in % with bacterial growth (both 76%), but there was a trend to more propionibacterium, peptostreptococcus and fusobacterium in those on AS, and some more capnocytophaga in those not on AS, with higher concentrations of lactobacillus and bacilllus in those not on AS.
--overlap of gastric/lung bacteria: significant correlation between corynebacterium and propionibacterium in both sites in those on AS.
--specifically assessing kids with proximal nonacid reflux: there was a highly significant correlation between those with proximal nonacid reflux and lung concentrations of bacillus, dermabacter, lactobacillis, peptostreptococcus and capnocytophagia.

so, this study suggests a link between gastric bacteria and the lung microbiome in those with reflux and may offer an explanation for those studies finding an increase in pneumonia in those on AS. the increase in corynebacterium may be clinically significant, since its presence in the oropharynx has been associated with wheezing in kids. though several of the bacteria found are not, in themselves, pathogenic, it may be that alterations in commensal bacteria may alter the microbiota in a way to increase the likelihood that pathogens could invade tissues and cause infection.

part of the reason i bring up this study is to provide yet another example of medications which affect the microbiome and may lead to "adverse effects" (which, of course, are really just part and parcel of the the effects of the medication, just unwanted ones).  this study highlights the potential infectious complications (though other literature pretty strongly supports a relationship with community-acquired pneumonia, c difficile infections, and, with less robust support, gastroenteritis, especially from campylobacter and salmonella). other concerns are the dramatic hypergastrinemia created by proton pump inhibitors, or PPIs (carcinoid tumors in animals, though it is not clear what the human effects are, i am concerned about such high levels of gastrin for sustained periods of time), profound acid suppression (which in a couple of studies, but not all, have found atrophic gastritis, a presumptively premalignant condition, especially in the presence of h pylori infection), and malabsorption, with documented decreases in absorption of magnesium, vitamin B12, iron, and calcium (and several studies, but not all, finding increases in osteoporosis and fractures). so, the other reason I bring up this article is that my sense is that PPIs specifically are way-too-overused. and several studies corroborate my own experience: once someone with dyspepsia is doing well with a PPI, too commonly we do not think to step-down therapy to a weaker agent, either an H₂-blocker or simply antacids -- the path-of-least-resistance is simply to leave the patient on the PPI, so that we can attend to the multitude of other issues the patient has....