The Potential Role of Gastric Microbiology in Respiratory Disease

To the Editor:

We read with interest the study from Narayana and colleagues documenting a dysregulated gut-lung axis in bronchiectasis in sputum and stool sample microbiota (1). This described a novel clinical phenotype, featuring a “high gut-lung interaction” cluster characterized by lung Pseudomonas (Pa), and gut Bacteroides and Saccharomyces, with associated increased exacerbations. Both the paper and the accompanying editorial highlighted that coordinated sampling of gastrointestinal and lung compartments provided novel complementary information. The editorial cited this as “the first human study of any respiratory condition to integrate analysis of microbiota across both anatomic compartments” (p. 812) (2). We congratulate the authors for undertaking this complex adult study. Previous work by Gallacher and colleagues (3), which showed dissimilarity of the gut-lung axis and dysbiosis of the lower airways in ventilated preterm infants, emphasizes the broad ranging scope of gut-lung studies (1, 3–9).

We feel that a discussion of sampling strategies may be timely for future research, and our previous work including longitudinal microbiome analyses of gastric juice and lung samples (7) could be relevant. We showed strains of Pa and nontuberculosis mycobacteria in gastric and sputum samples from people with cystic fibrosis with symptoms of extraesophageal reflux. Molecular techniques, including sequencing, confirmed that gastric and lung organisms were clonally related, and Pa and nontuberculosis mycobacteria were sometimes identified in gastric samples, but not in lung samples, indicating a possible gastric reservoir of microorganisms. Our work is potentially consistent with a complex, direct, bidirectional interplay, from lung to gut and gut to lung, in chronic suppurative airway disease.

It is underappreciated that a degree of microaspiration, including metabolically active microbiota, may be physiological and homeostatic, with disruption in lung disease (7, 8). In bronchiectasis, gastroesophageal reflux disease, a potential precursor to aspiration, ranges from 34% to 74%, and asymptomatic reflux is documented in 42–73% of patients (5). Three large prospective studies associate gastroesophageal reflux disease with increased bronchiectasis symptoms, exacerbations, and hospitalizations. Increased mortality has been described in a single-center, 212-patient study and a multicenter study of 986 patients (5, 6).

We postulate that increased microaspiration of gastric content may be consistent with airway and lung injury. In patients with hypochlorhydria who were treated for symptoms of reflux with acid-suppressing medications (4–7), gastric content can include metabolically active, viable microbiota (7). Our proposed interactions of the gastric and lung compartment potentially fit ecology-model explanations of microbial species present in the respiratory system, with diverse immigration and extinction factors at play, including direct microaspiration (8, 9).

Narayana and colleagues (1) discussed that the “low percentage of bacterial versus fungal overlap between the gut and lung does suggest that direct microbial interaction between sites is less likely” (p. 915). We wonder whether one potential explanation of our varied interpretations is that gut microbiome analyses are commonly predicated on the use of stool samples, influenced by the large intestine, whereas few data exist from gastric samples? Fecal samples are valuable and accessible, and they have informed studies of the microbiome. We have also found that paired, contemporaneous gastric secretion and lung samples provide insights. Gastric secretions are routinely discarded during endoscopy and ethics review boards, and patients are broadly supportive of interdisciplinary translational research. We suggest that future research could benefit from a discussion of a recommended panel of biological samples that could inform future studies of gut-lung interaction in chronic lung disease, including bronchiectasis.