
==== Front
Clin Transl Gastroenterol
Clin Transl Gastroenterol
CLTG
CT9
Clinical and Translational Gastroenterology
2155-384X
Wolters Kluwer Philadelphia, PA

38624107
CTG-24-0016
10.14309/ctg.0000000000000703
00011
3
Article
Pediatrics
Longitudinal Microbiome Changes in Children Exposed to Proton Pump Inhibitors
Zhang Yanjia Jason MD, PhD 12jason.zhang@childrens.harvard.edu

Connearney Sarah RN, MSN, CPNP 1sarah.connearney@childrens.harvard.edu

Hester Lisa RN, BSN, CPNP 1lisa.hester@childrens.harvard.edu

Du Maritha BS 1marithadu.100@gmail.com

Catacora Andrea BS 1andreacatacora23@gmail.com

Akkara Anna BS 1annaakkara@gmail.com

Wen Anna BS 1anna.wen@childrens.harvard.edu

Bry Lynn MD, PhD 34lbry@bwh.harvard.edu

Alm Eric J. PhD 2ejalm@MIT.edu

https://orcid.org/0000-0003-3949-3060
Rosen Rachel MD 1
1 Gastroenterology/Nutrition, Boston Children's Hospital, Boston, Massachusetts, USA;
2 Biological Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA;
3 Massachusetts Host-Microbiome Center, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, USA;
4 Clinical Microbiology Laboratory, Department of Pathology, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, USA.
Correspondence: Rachel Rosen, MD. E-mail: rachel.rosen@childrens.harvard.edu.
9 2024
16 4 2024
15 9 e110 1 2024
30 3 2024
© 2024 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of The American College of Gastroenterology
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

INTRODUCTION:

Proton pump inhibitor (PPI) use has been associated with an increased risk of gastrointestinal and upper respiratory infections in children. There are limited longitudinal data on the effect of PPI in children. The goal of this prospective observational study was to compare the stool and oropharyngeal microbiome of children before and after starting PPIs.

METHODS:

We prospectively recruited participants from a gastroenterology clinic. Consented participants provided stool samples and oropharyngeal swabs at baseline and after 8 weeks of PPI therapy. Microbiome changes were measured by analyzing 16S sequencing from both body sites at both time points.

RESULTS:

Thirty-four participants completed the study and provided samples both at baseline and after 8 weeks on PPI therapy. Of those, 24 participants had sufficient sequencing from both stool and oropharyngeal samples at both time points. There were no differences between the pre-PPI and post-PPI samples using beta-diversity metrics in either the oropharynx or stool. There were, however, significant changes in specific taxa. There was an enrichment of Streptococcus in the stool after PPI use and a reduction in the relative abundance of Bifidobacterium, Peptostreptococcus, and Turicibacter (P-values < 0.01). Furthermore, there was an increase in the relative abundance of oropharyngeal bacteria in the stool after PPI therapy. This enrichment of oropharyngeal bacteria in the stool was most prominent in younger participants.

DISCUSSION:

Further investigation is needed to determine the clinical and microbial factors that predispose or protect against microbiome changes due to PPI use and why young children are more susceptible to this PPI effect.

KEYWORDS:

proton pump inhibitor
microbiome
gastroesophageal reflux
National Institute of Diabetes and Digestive and Kidney DiseasesR01 DK097112 Rachel RosenEunice Kennedy Shriver National Institute of Child Health and Human Development5K12HD052896 Jason ZhangOPEN-ACCESSTRUE
SDCT
==== Body
pmcINTRODUCTION

Proton pump inhibitors (PPIs) are commonly used medications in infants and children. They are prescribed for multiple pediatric conditions (colic, gastroesophageal reflux, feeding difficulties, abdominal pain, and cough). Prescribing rates in children have increased more than 500% over the last 2 decades despite more data showing a changing risk-benefit profile (1–6). One area of particular concern is the reported increased risk of clinical infections with PPI use including gastroenteritis, upper respiratory infections, pneumonia, pharyngitis, Clostridioides difficile infections, and necrotizing enterocolitis (1,7–15).

Despite this higher risk of infection, there are few pediatric studies looking at the changes in the microbiome with PPI use (9,16–19). Our group and others have shown clear changes in the gastric microbiome after PPI use, including an increase in relative abundance of Streptococcus sp. in gastric fluid, but there are no longitudinal studies in patients beyond the neonatal period (18,20,21). In the single, small longitudinal study of 12 infants before and after starting PPIs, there were no differences in the alpha or beta diversity of the stool microbiome during the first 6 months of life (22). By contrast, several adult studies have demonstrated that PPI use results in measurable changes in the stool microbiome. One important pattern has emerged from the adult studies—many taxa that increase in the stool after PPI use are oropharyngeal or upper gastrointestinal (GI) microbes, such as Streptococcus and Prevotella (9,10,16–19). This suggests that suppressing gastric acid may result in the passage of upper GI microbes to the distal GI tract (9,17,18). Whether similar changes are seen in children is not known. We hypothesized that there would be identifiable changes in the oropharyngeal and stool microbiome in children using PPIs.

To address this hypothesis and overcome the limitations of the current literature, we recruited children and young adults (aged 1–18 years) who had been prescribed PPI in a pediatric gastroenterology clinic and profiled both the oropharyngeal and stool microbiome before and after 8 weeks of PPI use. This design allowed us to measure microbiome changes longitudinally and directly determine if PPI use results in a shift in the stool microbiome towards oropharyngeal microbes.

METHODS

Study design

In this prospective longitudinal cohort study, we recruited children and young adults aged 0–18 years who had been prescribed PPIs at the discretion of their primary gastroenterologist. Patients were recruited to participate if they were presenting to the gastroenterology office with symptoms of gastroesophageal reflux, feeding disorders, or abdominal pain and were going to be taking acid suppression for at least 8 weeks at the time of recruitment. Patients were excluded if they had antibiotic exposure or acid suppression exposure within 8 weeks of beginning proton pump inhibitors or if they had a history of fundoplication or other GI surgery. Before starting PPIs, we obtained oropharyngeal (posterior tongue swabs) and stool samples. Sample collection was repeated 8 weeks after starting PPIs.

Microbiome analysis

We extracted DNA using a Qiagen Powersoil DNA extraction kit. 16S rDNA libraries were prepared and sequenced by the Broad Institute Genomic Platform, using paired-end 250-bp reads on an Illumina HiSeq. We analyzed 16S data using Qiime2, DADA2, Phyloseq in R, and custom Python scripts. We assigned taxonomic labels to 16S sequences using the SILVA database. Differential abundance testing was performed using ANCOM-BC in the Qiime2 environment. Beta-diversity analysis was performed using Bray-Curtis dissimilarity within the Qiime2 environment. Linear discriminant analysis effect size (LEfSe) was performed using Phyloseq in R. To measure the abundance of oropharynx-associated microbes on a population level, we used the top 7 differentially abundant microbes in the oropharynx, as determined by ranking ANCOM W values. We then calculated the log ratio of relative abundance for these 7 microbes in the stool before and after PPI use. To determine an individualized metric for oropharynx-associated microbes, we defined participant-specific oropharynx-associated taxa as genera that were at a greater than 1% relative abundance in the oropharynx but less than 0.05% relative abundance in the stool. For every individual, we calculated the log ratio of relative abundance before and after PPI use for each microbe that met this criterion. For each individual, we then calculated the median of this ratio for all microbes that met this criterion. To determine the association between various clinical metadata metrics and this individualized measure of oral microbial expansion in the stool, we used the individual medians and calculated a correlation coefficient using the Pearson method. Only patients with complete sets of samples with identifiable microbial signatures were included in this analysis.

RESULTS

The patient flow diagram is shown in Figure 1. One hundred five patients were recruited for participation, and 47 had both baseline (T0) and follow-up (T1) sample sets. Patient characteristics are shown in Table 1.

Figure 1. Consort diagram detailing patient flow. PPI, proton pump inhibitor.

Table 1. Demographical information

	N = 34	
Age (yr)		
 Mean	9.67	
 Median	11.08	
 Range	0–18	
Sex		
 Male	18 (52.94%)	
 Female	16 (47.06%)	
Ethnicity		
 Hispanic	2 (5.88%)	
 Non-Hispanic	29 (90.63%)	
 Unknown	3 (8.82%)	
Race		
 White	30 (88.23%)	
 Black	7 (20.59%)	
 More than 1 race	7 (20.59%)	
 Native American/Alaskan	0	
Dose PPI/weight (mg/kg)		
 Mean	0.89	
 SD	0.56	
PPI, proton pump inhibitor.

Distinct microbes in the oropharynx and stool at baseline

We first demonstrated that the oropharynx and the stool microbiomes in children are distinct (Bray-Curtis dissimilarity, permutational multivariate ANOVA P-value < 0.0001) (Figure 2a). We performed LEfSe and found that several taxa abundant in the oropharynx were rare in the stool. The oropharynx microbiome was defined by an increased abundance of multiple taxa including the phyla Proteobacteria, Fusobacteria, and Actinobacteria; the families Veillonellaceae, Streptococcaceae, and Prevotellaceae; and multiple genera including Prevotella, Veillonella, Streptococcus, and Rothia (linear discriminant analysis score [log10] > 4) (Figure 2b and see Supplementary Figure 1, Supplementary Digital Content 1, http://links.lww.com/CTG/B114). At the genus level, we performed ANCOM-BC to define the statistically significant genera associated with the oropharynx; the top 7 were Veillonella, Prevotella, a genus in the family Pasteurellaceae, Streptococcus, Fusobacterium, Neisseria, and Rothia (Figure 2c and see Supplementary Figure 2, Supplementary Digital Content 1, http://links.lww.com/CTG/B114). Distinctively stool-associated taxa were the phylum Firmicutes; the families Ruminococcaceae, Lachnospiraceae, and Bacteroidaceae; and the genera Faecalibacterium and Bacteroides (Figure 2b). The top 7 genera specifically associated with the stool identified by ANCOM-BC were Faecalibacterium, Bacteroides, Dialister, Roseburia, Akkermansia, Blautia, and Bifidobacterium (Figure 2c and see Supplementary Figure 2, Supplementary Digital Content 1, http://links.lww.com/CTG/B114).

Figure 2. At baseline, the oropharyngeal and stool microbiomes are distinct. (a) Beta-diversity analysis presented as a principal coordinate analysis (PCoA) plot using Bray-Curtis distances, permutational multivariate ANOVA P < 0.001 when comparing oropharynx and stool samples. (b) Cladogram of the taxonomic distribution of differentially abundant taxa in the stool vs oropharynx, obtained by LEfSe with a linear discriminant analysis score threshold of >4 (alpha < 0.01). (c) Top: Relative abundance in the oropharynx of the top 10 differentially abundant OP genera compared with stool, defined using ANCOM-BC rejection of the null hypothesis. Bottom: Relative abundance in the stool of the top 10 differentially abundant stool genera compared with the OP, defined using ANCOM-BC rejection of the null hypothesis. LEfSe, linear discriminant analysis effect size; PPI, proton pump inhibitor.

The oropharyngeal microbiome before and after PPI

We found no changes in alpha diversity (Shannon index) and beta diversity after PPI use in the oropharynx (Figure 3a). There were no significant differences in the relative abundance of bacteria by phylum, genus, or species between T0 and T1. We found an upward trend in relative abundance between T0 and T1 in an unnamed genus in the family Gemellaceae and a downward trend in the genus Campylobacter, but neither finding met our false discovery rate criteria (see Supplementary Figure 6, Supplementary Digital Content 1, http://links.lww.com/CTG/B114).

Figure 3. Specific taxa are affected by PPI use. (a) Each participants' change in alpha diversity, as measured by the Shannon index, before and after 2 months of PPI use in both the stool (left, red) and oropharynx (right, blue). The difference between pre-PPI and post-PPI was not statistically significant. (b) Beta-diversity analysis presented as a principal coordinate analysis (PCoA) plot using Bray-Curtis distances, comparing pre-PPI and post-PPI samples in both the stool (left) and oropharynx (right) samples. Permutational multivariate ANOVA P-values were nonsignificant in the pre-PPI vs post-PPI comparisons. (c) Four genera were differentially abundant in the stool after PPI use (Wilcoxon signed-rank test, false discovery rate <0.1), including Streptococcus (increased after PPI use) and Bifidobacterium, an unnamed genus in the family Peptostreptococcaceae, and Turicibacter (decreased after PPI use). The ratio (post/pre) of relative abundances is graphed as boxplots with individual dots representing each participant (left). Lines denoting the change in relative abundances are also presented with each line representing a single participant (right). PPI, proton pump inhibitor.

The stool microbiome before and after PPI

There were no significant differences in alpha diversity (Shannon index) and beta diversity (Bray-Curtis dissimilarity) after PPI use, but we did discover that several taxa changed in the stool after PPI use (Figure 3a and b). We found that a single phylum, Actinobacteria, was decreased in abundance after PPI use in the stool (Wilcoxon signed-rank test, false discovery rate [FDR] <0.1) (see Supplementary Figure 3, Supplementary Digital Content 1, http://links.lww.com/CTG/B114). At the genus level, we identified 4 taxa in the stool that were affected by PPI use (Wilcoxon signed-rank test, FDR <0.1); Streptococcus increased in abundance, whereas Bifidobacterium, an unnamed genus in the family Peptostreptococcaceae, and Turicibacter all decreased in abundance after PPI use (Figure 3c). We then used the 10 most common oropharynx-associated genera seen at baseline (T0) to see if those were increased in the stool at T1. Streptococcus had a statistically significant increase at T1 compared with T0; several other genera had relative abundances that trended upward between T0 and T1, although these were not statistically significant (see Supplementary Figure 4, Supplementary Digital Content 1, http://links.lww.com/CTG/B114). None of the 10 most common stool-associated genera at T0 increased at T1 (see Supplementary Figure 4, Supplementary Digital Content 1, http://links.lww.com/CTG/B114). Given the relatively large range of PPI dose-by-weight in our pediatric population, we postulated that we would find more PPI-affected taxa by finding the association of PPI dose and change in relative abundance. The relative abundance of Faecalibacterium and Ruminococcus decreased after PPI use in a dose-associated manner (FDR based on Pearson correlation <0.1) (see Supplementary Figure 5, Supplementary Digital Content 1, http://links.lww.com/CTG/B114).

Oropharynx-associated taxa in stool

We next focused on whether the oropharyngeal bacteria changed in abundance in the stool after starting PPIs. Streptococcus increased in the stool after PPI use, and several other oropharyngeal bacteria trended upwards as well (Figure 4a). We then looked specifically at the top 7 oropharynx-associated bacteria and the top 7 stool-associated bacteria and measured the change in relative abundance of these groups at T1 compared with T0 in the stool. We found a statistically significant increase in oropharynx-associated genera in the stool after PPI use compared with stool-associated genera (see Supplementary Figure 7, Supplementary Digital Content 1, http://links.lww.com/CTG/B114). We next sought to determine if this increase in oropharynx-associated taxa in the stool was occurring just on a population level or within individual patients. We defined participant-specific oropharynx-associated taxa as genera that were at a greater than 1% relative abundance in the oropharynx but less than 0.05% relative abundance in the stool. Participant-specific stool-associated taxa were defined vice versa. Using these participant-specific defined communities, we observed a statistically significant increase in oropharynx-associated taxa after PPI use, along with a concomitant decrease in stool-associated taxa (Figure 4b).

Figure 4. PPI use increases oropharynx microbes in the stool, and younger children are more susceptible to this effect. (a) Boxplots of the log-2 fold change in the stool after PPI use of the top 7 oropharynx-associated and stool-associated genera. (b) Log-2 fold change in the stool of participant-specific oropharynx-associated and stool-associated genera after PPI use. Oropharynx-associated genera increase in the stool after PPI use, stool-associated genera decrease (t test P < 0.01). (c) Distributions of log-2 fold changes of individually defined oropharynx-associated (blue) and stool-associated (orange) genera for each participant, ordered on the y-axis by age. (d) Linear association between age in months and the log-2 fold change of oropharynx-associated microbes in the stool after PPI use (Pearson R = 0.56, P = 0.01). PPI, proton pump inhibitor.

Impact of age on microbial changes

For each individual, we measured the change in oropharynx-associated microbes and stool-associated microbes based on age younger than 12 and older than 12 years, the age when adult dosing is often used (Figure 4c). Overall, oropharynx-associated taxa increased after PPI use using this metric, but there was high interindividual variability. For most individuals, oropharynx-associated genera increased in the stool after 2 months on a PPI. However, there were 3 individuals who exhibited the opposite effect. We next searched for participant characteristics that explained this variability. We found no association between PPI dose-by-weight and oropharynx-microbe increases (see Supplementary Figure 8, Supplementary Digital Content 1, http://links.lww.com/CTG/B114). We tested this association using Pearson correlation with dose-by-weight as a continuous variable and by creating 2 categories (greater than and less than 1.0 mg/kg) based on participant PPI dosing; in both cases, PPI dose made no difference (Pearson R = 0.12, P = 0.61; t test for high vs low PPI dose, P = 0.43). We did however find that age was inversely associated with oropharynx microbe increases in the stool—younger children were more susceptible to the PPI-driven increase of oropharynx-associated microbes (Figure 4c and d).

DISCUSSION

We performed a prospective longitudinal study in children and young adults, measuring the changes in both the oropharyngeal and stool microbiomes before and after PPI use. We found no differences in alpha and beta-diversity metrics before and after PPI in either the oropharynx or the stool. After PPI use, however, we found that Streptococcus, a common oropharyngeal genus, increased in the stool. On both a population and individual level, we found that as a group, bacteria linked to the oropharynx increased in the stool after PPI use. Finally, we found that younger children were more susceptible to increases in oropharyngeal bacteria in the stool after PPI use.

Several studies have established that PPIs reliably alter the distal gut microbiome of adults. There are relatively few studies of the effects of PPI on the microbiome in children. A cross-sectional study of 116 children (59 on PPIs and 57 controls) showed an increased abundance of gastric and oropharyngeal Streptococcus in PPI-treated children (20). One prospective study followed infants (mean age 5.2 months, all younger than 11 months) before and after starting PPI for gastroesophageal reflux disease and noted no changes in stool microbiome alpha and beta diversity after PPI therapy, although there was a relative increase in Haemophilus and a relative decrease in Lactobacillus and Stenotrophomonas (22). There were more significant changes over time, although these changes correlated strongly with the introduction of solid foods, so teasing out the PPI effect is challenging (22). A recent pilot (n = 10) longitudinal study of the fecal microbiome in infants with esophageal atresia showed that longer duration of PPI use resulted in larger changes in stool microbial diversity (23). In older children, 2 small prospective studies, 1 with 5 children and another with 20 children, found no major changes in the distal gut microbiome after PPI use (24,25). All the prospective studies in children thus far have been small and/or restricted to a smaller subset of the pediatric population (i.e., infants or surgical patients), and all have sampled only the stool microbiome.

As far as we are aware, our study is the first longitudinal study in children and young adults on PPIs, wherein both the upper and lower GI tract microbiomes were analyzed. This design led us to one of the key findings of our study that oral microbes expand in the distal GI tract after PPI use. Many hypothesize that PPI use results in bacterial overgrowth and survival in the upper GI tract and this results in increased downstream transit of these bacteria. In our study, we directly measured the change in upper GI-associated microbes in the distal GI tract, both in our population as a whole and on an individual basis. We were able to do this because we longitudinally sampled both the oral cavity and the stool. Our analysis strongly suggests that the PPI-treated GI tract is more permissive of oral microbes moving to the distal gut and that this permissiveness may be a major driver of PPI-associated microbiome alterations. The consequences of oral microbes in the distal GI tract are still unclear, although several groups have pointed out that many of these oral microbes (Streptococcus, Proteobacteria) are associated with C. difficile infection. In fact, an increased abundance of oral microbes and decreased abundance of common stool microbes such as Bifidobacterium are shared features of the PPI-associated microbiome and the C. difficile infection-associated microbiome. The clinical consequences of oral microbe transit to the distal gut may also go beyond GI infections. As advanced sequencing technology now allows for tracking of oral microbe transit at the strain level, studies have postulated that oral microbes in the distal gut may contribute to irritable bowel syndrome, inflammatory bowel diseases, colorectal cancer, and several other GI diseases (26–28). These links are observational, and future research is needed to determine the consequences of oral microbes in the stool and how their presence predisposes children to disease.

Our finding that Streptococcus increased in the stool after PPI use is consistent with several longitudinal studies of the PPI effect on the fecal microbiome (9,18,29–32). Interestingly, in a systematic review of PPI-associated fecal microbiome alterations, all 4 adult studies that measured differences at the genus or species level found an increase in the relative abundance of Streptococcus, while none of the longitudinal pediatric studies found such changes (18,23–25). This may be due to an overrepresentation of infants and neonates in the pediatric studies to date, as the average abundance of Streptococcus is higher in infants and neonates than in older children and adults (33). Of the remaining fecal microbial changes we found in this study at the genus level (decreases in Bifidobacterium or Turicibacter), none have been found to significantly decrease in relative abundance in longitudinal studies, although both genera were found to be in lower relative abundance in at least 1 cross-sectional study of PPI users than in controls (16,18,34). As a possible explanation, Bifidobacterium is more abundant in children than adults; most longitudinal studies to date have been conducted in adults and may therefore have underappreciated the PPI effect on Bifidobacterium. Overall, taxa-level changes in longitudinal studies are highly variable and study-dependent, likely influenced by the small size and different populations of most longitudinal studies. However, our study does recapitulate the most common finding in the literature to date (increase in Streptococcus) and formally measures an increase in oral microbes in the stool.

We also found that not all children were equally susceptible to PPI-driven microbiome changes. By tracking each individual's oropharyngeal bacteria, we found that younger children had more oropharyngeal bacteria in the stool after PPI use than in older children. This has several implications. Despite guideline recommendations limiting use, pediatric PPI use has increased over time, and overprescribing PPI seems to be most prevalent for young children (3). Our findings add to an already growing set of reasons to avoid PPI use in young children in the absence of a clear indication. In addition, PPIs uniquely increase the risk of infections such as necrotizing enterocolitis, gastroenteritis, upper respiratory infections and pharyngitis, so this finding may have significant implications for understanding the mechanism behind these infections.

There are several limitations to our study. First, the size of our study was relatively small. A larger pediatric cohort, or meta-analyses as more pediatric cohorts are published, would be informative. That said, to our knowledge, this is the largest prospective pediatric study of the PPI effect on the microbiome to date. Second, we recruited patients from a gastroenterology clinic, so our findings may be specific to children with gastroenterological conditions. Third, our study is limited to 2 months of exposure, although this is consistent with other published PPI trials and gastroesophageal reflux disease guidelines that recommend 8 weeks of therapy for esophageal healing. However, a longer study would be useful to understand the temporal dynamics of longer courses of PPI use, which some children still experience. Relatedly, but beyond the scope of our study, is the question of how persistent PPI-associated changes are long term. In the future, longer studies extending into post-PPI time points will be important. Another possible limitation is the uncertainty of how children metabolize proton pump inhibitors; poor metabolizers may be at increased risk of infectious complications, and it is possible that these patients may have the greatest microbial disturbances. Future studies should include CYP2C19 genetic testing to assess PPI metabolizer status. Another limitation is that there could be other confounders beyond PPI that might have affected the microbiome, although the biggest confounder, antibiotics, was addressed by excluding all patients taking them. Finally, it was not possible to assess for the development of clinical infections in this sample size, but the next important step will be to match the microbial changes with the development of clinical infections.

In summary, this prospective study of the oral and stool microbiome in children and young adults prescribed PPIs provides evidence that PPI use increases the presence of oral microbes in the stool. We measured this microbial alteration in most pediatric subjects and observed that younger children are more susceptible to the PPI-associated microbial shifts. Our data suggest that PPIs drive a mixing of the microbial populations in the upper and lower GI tracts, which may explain a mechanism by which PPIs put children at risk for a myriad of GI and non-GI diseases. Altogether, our study gives another reason for caution against PPI use in the absence of a clear indication, especially in young children whose microbiomes are still developing.

CONFLICTS OF INTEREST

Guarantor of the article: Rachel Rosen, MD.

Specific author contributions: R.R., M.D., A.C., A.A., and A.W. designed and carried out the study. L.B., E.J.A., and Y.J.Z. designed and performed the microbiome analysis. S.C., L.H., and R.R. collected the microbiome samples. R.R. and Y.J.Z. prepared the manuscript, with editing contributions from all authors.

Financial support: Y.J.Z. and this research were supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (award 5K12HD052896). R.R. was supported by NIDDK R01 DK 097112.

Potential competing interests: None to report.Study Highlights

WHAT IS KNOWN

✓ Proton pump inhibitors (PPIs) have been associated with an increased risk of infections in children.

✓ PPI use has been associated, in cross-sectional studies, with increased growth of Gram-positive organisms in the gastric fluid of children.

WHAT IS NEW HERE

✓ PPIs alter both the oral and fecal microbiome in children.

✓ PPI treatment increases the likelihood of finding oral microbes in the stool.

✓ Young children were more likely to have oral microbes in the stool after PPI treatment.

Supplementary Material

SUPPLEMENTARY MATERIAL accompanies this paper at http://links.lww.com/CTG/B114
==== Refs
REFERENCES

1. De Bruyne P Ito S . Toxicity of long-term use of proton pump inhibitors in children. Arch Dis Child 2018;103 (1 ):78–82.29237614
2. Tolia V Boyer K . Long-Term proton pump inhibitor use in children: A retrospective review of safety. Dig Dis Sci 2008;53 (2 ):385–93.17676398
3. Slaughter JL Stenger MR Reagan PB . Neonatal histamine-2 receptor antagonist and proton pump inhibitor treatment at United States children's hospitals. J Pediatr 2016;174 :63–70.e3.27131401
4. Blank M-L Parkin L . National study of off-label proton pump inhibitor use among New Zealand infants in the first year of life (2005–2012). J Pediatr Gastroenterol Nutr 2017;65 (2 ):179–84.28403034
5. Closs ER Mårild K Nielsen RG . Use of proton pump inhibitors in scandinavian children and adolescents: An observational study. Front Pediatr 2023;11 :1052978.36873656
6. Lyamouri M Mårild K Nielsen RG . Proton pump inhibitors for infants in three Scandinavian countries increased from 2007 to 2020 despite international recommendations. Acta Paediatr 2022;111 (11 ):2222–8.35860963
7. Wang YH Wintzell V Ludvigsson JF . Association between proton pump inhibitor use and risk of asthma in children. JAMA Pediatr 2021;175 (4 ):394–403.33555324
8. Pasman EA Ong B Witmer CP . Proton pump inhibitors in children: The good, the bad, and the ugly. Curr Allergy Asthm Rep 2020;20 (8 ):39.
9. Freedberg DE Toussaint NC Chen SP . Proton pump inhibitors alter specific taxa in the human gastrointestinal microbiome: A crossover trial. Gastroenterology 2015;149 (4 ):883–5.e9.26164495
10. Freedberg DE Lamousé-Smith ES Lightdale JR . Use of acid suppression medication is associated with risk for C. difficile infection in infants and children: A population-based study. Clin Infect Dis 2015;61 (6 ):912–7.26060292
11. Turco R Martinelli M Miele E . Proton pump inhibitors as a risk factor for paediatric Clostridium difficile infection. Aliment Pharmacol Ther 2010;31 (7 ):754–9.20047577
12. Dipasquale V Cicala G Spina E . A narrative review on efficacy and safety of proton pump inhibitors in children. Front Pharmacol 2022;13 :839972.35222047
13. More K Athalye-Jape G Rao S . Association of inhibitors of gastric acid secretion and higher incidence of necrotizing enterocolitis in preterm very low-birth-weight infants. Am J Perinat 2013;30 (10 ):849–56.
14. Terrin G Passariello A De Curtis M . Ranitidine is associated with infections, necrotizing enterocolitis, and fatal outcome in newborns. Pediatrics 2012;129 (1 ):e40–e45.22157140
15. Levy EI Hoang DM Vandenplas Y . The effects of proton pump inhibitors on the microbiome in young children. Acta Paediatr 2020;109 (8 ):1531–8.32027402
16. Reveles KR Ryan CN Chan L . Proton pump inhibitor use associated with changes in gut microbiota composition. Gut 2018;67 (7 ):1369–70.28993417
17. Imhann F Bonder MJ Vich Vila A . Proton pump inhibitors affect the gut microbiome. Gut 2016;65 (5 ):740–8.26657899
18. Macke L Schulz C Koletzko L . Systematic review: The effects of proton pump inhibitors on the microbiome of the digestive tract—Evidence from next-generation sequencing studies. Aliment Pharmacol Ther 2020;51 (5 ):505–26.31990420
19. Shi YC Cai ST Tian YP . Effects of proton pump inhibitors on the gastrointestinal microbiota in gastroesophageal reflux disease. Genom Proteom Bioinform 2019;17 (1 ):52–63.
20. Rosen R Hu L Amirault J . 16S community profiling identifies proton pump inhibitor related differences in gastric, lung, and oropharyngeal microflora. J Pediatr 2015;166 (4 ):917–23.25661411
21. Tsuda A Suda W Morita H . Influence of proton-pump inhibitors on the luminal microbiota in the gastrointestinal tract. Clin Transl Gastroenterol 2015;6 :e89.26065717
22. Castellani C Singer G Kashofer K . The influence of proton pump inhibitors on the fecal microbiome of infants with gastroesophageal reflux—A prospective longitudinal interventional study. Front Cell Infect Microbiol 2017;7 :444.29075620
23. Brusselaers N Pereira M Alm J . Effect of proton pump inhibitors in infants with esophageal atresia on the gut microbiome: A pilot cohort. Gut Pathog 2022;14 (1 ):47.36527125
24. Simakachorn L Tanpowpong P Chanprasertyothin S . Gut microbiota characteristics in children after the use of proton pump inhibitors. Turk J Gastroenterol 2021;32 (1 ):70–5.33893768
25. Andrew S Wei Z Wendy SWW . The impact of gastric acid suppression on the developing intestinal microbiome of a child. medRxiv 2021;2021.12.21.21268064.
26. Tan X Wang Y Gong T . The interplay between oral microbiota, gut microbiota and systematic diseases. J Oral Microbiol 2023;15 (1 ):2213112.37200866
27. Schmidt TS Hayward MR Coelho LP . Extensive transmission of microbes along the gastrointestinal tract. eLife 2019;8 :e42693.30747106
28. Kitamoto S Nagao-Kitamoto H Hein R . The bacterial connection between the oral cavity and the gut diseases. J Dent Res 2020;99 (9 ):1021–9.32464078
29. Jackson MA Goodrich JK Maxan ME . Proton pump inhibitors alter the composition of the gut microbiota. Gut 2016;65 (5 ):749–56.26719299
30. Koo S Deng J Ang D . Effects of proton pump inhibitor on the human gut microbiome profile in multi-ethnic groups in Singapore. Singapore Med J 2019;60 (10 ):512–21.30488079
31. Otsuka T Sugimoto M Inoue R . Influence of potassium-competitive acid blocker on the gut microbiome of Helicobacter pylori-negative healthy individuals. Gut 2017;66 (9 ):1723–5.27965281
32. Mishiro T Oka K Kuroki Y . Oral microbiome alterations of healthy volunteers with proton pump inhibitor. J Gastroenterol Hepatol 2018;33 (5 ):1059–66.29105152
33. Moore RE Townsend SD . Temporal development of the infant gut microbiome. Open Biol 2019;9 :190128.31506017
34. Takagi T Naito Y Inoue R . The influence of long-term use of proton pump inhibitors on the gut microbiota: An age-sex-matched case-control study. J Clin Biochem Nutr 2018;62 (1 ):100–5.29371761
