The Human Respiratory Microbiome: Implications and Impact

Alicia B. Mitchell, BMedSci (Hons); Allan R. Glanville, MBBS, MD, FRACP

Disclosures

Semin Respir Crit Care Med. 2018;39(2):199-212. 

In This Article

Lung Transplantation

Many host-specific and environmental factors have been shown to influence the human microbiota; however, in this review, the focus will be on the impact of lung transplantation on respiratory microbiome dynamics and in contrast, how these variations in the lung microbiome affect transplant outcomes. Lung transplantation provides us with a unique opportunity to separate out host-specific, immune-related and extrinsic factors in shaping the microbiome after donor lungs have been transplanted into a new recipient. In effect, this is transplantation of the human respiratory virome.

Lung transplantation is often the only viable option remaining for individuals with end-stage lung disease, providing patients with hope for an improved longevity and quality of life. However, compared with other solid organ transplantation, lung transplantation has the lowest long-term survival rates. Worldwide, the median survival time post bilateral lung transplant is 7.4 years as reported in the International Society for Heart and Lung Transplantation (ISHLT) 2017 Registry report compared with a median of 12.4 years for both living and deceased donor kidney transplantation.[78]

Lung transplantation is associated with multiple opportunities for lower respiratory tract sampling with both surveillance and clinically indicated bronchoscopies. Some of the first lung microbiome studies incorporating longitudinal analysis were performed in the transplant cohort, due to the ability to conduct serial invasive sampling for surveillance of the allograft. Borewicz et al[79] showed in a small cohort that the lung microbiome in transplant patients is not very stable in the early post-transplant period with less than 10% of species retained over the three time points. Another cohort of lung transplant recipients underwent longitudinal sampling of their lower respiratory tract for up to 12 months, showing that bacterial diversity increased during the first 9 months post-transplant. This likely reflects development of a new stable state in the donor lungs. After 9 to 12 months post lung transplant, bacterial diversity began to decrease.[80] Finally, Willner et al[81] surveyed 16 patients who had been transplanted for CF and found dynamic changes in the lung microbiome over time which was heavily influenced by antibiotic use. These preliminary data indicate that there are important fluctuations in both microbiota burden and diversity, especially in the early transplant period. Some associations between bacterial diversity and negative outcomes[81,82] have been suggested; however, limited data at this point restrict conclusions.

It has been previously thought that unanticipated donor transmission of infectious organisms is a rare and dangerous event.[83] Much work has focused on reducing the occurrence of transmission events, specifically of blood-borne viruses such as HIV and hepatitis B (HBV) and C (HCV) viruses. However, emerging evidence shows that the microbiome is transplanted from the donor into the recipient at the time of lung transplantation and that the composition of the donor microbiome may have important implications for transplant outcomes. The post-transplant lung microbiome may also be modified through a range of factors. These include nosocomial acquisition of microbes from the intensive care unit (ICU) while intubated and from contact with other infected individuals while in hospital, transmission from the native lung in recipients who receive a single lung transplant, aspiration of gastric organisms due to the increased burden of gastroesophageal reflux disease (GERD) in transplant recipients,[84] and the impact of the upper respiratory tract microbiota (Figure 2). Little research currently addresses the origin of the microbes that induce the longitudinal changes in the microbiome that have been previously described.[79–81] Further research studies and a greater understanding are needed to characterize implications for clinical management.

Figure 2.

Dysbiosis of the lung microbiome after lung transplantation. After lung transplantation, the healthy lung equilibrium between resident and transient species may be challenged by the donor microbiome transplanted within the new lungs which may be qualitatively and quantitatively different from the microbiome of the explanted native lungs. Other external events can also lead to dysbiosis including community acquired respiratory virus infection, the ex-vivo stage of lung procurement and the impact of immune suppression.

The vagus and other sympathetic nerves are transected at the hilum during lung transplantation causing loss of neural supply and complete denervation of the lungs in the early transplant period. This leads to decreased cough reflex, impaired ciliary beat frequency leading to impaired mucociliary clearance and implications for gastroesophageal motility.[85] Reduction in mechanical clearance mechanisms such as cough and the mucociliary escalator has important implications in the ability of the recipient to clear inhaled pathogens and allergens and therefore reduces the rate of elimination of species which is important in maintaining the "steady state" airway microbiota.[86] Furthermore, recipients receive high-level induction and maintenance immunosuppression[87] which reduces the ability of the immune system to respond and clear microbial species, further impacting the elimination aspect in determining the microbial balance.

The immigration of bacterial species into the lower respiratory tract is accelerated with gastroesophageal dysfunction and reflux, which lead to increased aspiration events. The incidence of GERD is significantly increased after lung transplantation,[84] which has been shown to impact the composition of the lower respiratory tract microbiota.[32] Furthermore, the use of acid-suppression medications such as proton-pump inhibitors has been associated with an increased risk of overgrowth of acid-sensitive gastric flora such as Streptococcus and Staphylococcus, which were also found at higher concentrations in the lungs following reflux events, indicating the likely exchange of microflora between these two sites.[88] These bacteria are often associated with symptomatic upper respiratory tract infections and pneumonia.[89] Laryngeal dysfunction and oral hygiene may also have an impact on lower respiratory tract microbiota due to increased pathogenic species in the oral cavity spreading to the airways, causing variations in the lung microbiome and through the increased risk of aspiration and inhalation events when there are alterations to the larynx.[90,91] Furthermore, altered growth conditions within the lungs may have significant effects on microbiome dynamics with outgrowth of certain species and increased spatial heterogeneity within the lung field. These altered conditions may be caused by several factors including modifications to the airway terrain due to changes in ciliary function and mucus production, changes in regional growth conditions due to pneumocystis and general bacterial prophylaxis, and finally, effects of reperfusion after the transplant has occurred (Figure 2).[37]

Mortality within the first-year post-transplant is associated with primary graft failure and acute rejection events. Within the first year of transplantation, up to 55% of recipients are treated for acute allograft rejection.[92] However, there has been limited research in acute rejection compared with chronic rejection and bronchiolitis obliterans syndrome (BOS), but there is some evidence to suggest that both bacterial and viral infections, and possibly perturbations to the underlying microbiota, may have an important role in mediating early allograft dysfunction. Glanville et al showed that persistent Chlamydia pneumoniae infection is associated with early mortality and rejection events.[93] Parainfluenza virus (PIV) infection also has a demonstrated association with the development of acute allograft dysfunction, with one study showing that 82% of patients with parainfluenza viral infection in their cohort also had evidence of acute allograft dysfunction. Furthermore, 32% of these PIV-positive patients developed bronchiolitis obliterans within 18 months follow-up.[94] Conversely, Ahya et al showed that detection of Epstein–Barr virus (EBV) DNA was not correlated with the development of acute allograft rejection. Instead, the authors suggested that increased viral load may be a surrogate marker of effective immunosuppression thus providing protective allograft effects.[95]

The principal factor limiting long-term survival in lung transplant recipients is chronic lung allograft dysfunction (CLAD) mostly due to BOS. According to the ISHLT Registry Report, 50% of lung transplant recipients develop BOS within 5 years of transplant, while 76% of recipients develop BOS within 10 years.[96] The primary pathological feature is bronchiolitis obliterans, a form of intraluminal airway fibrosis located mainly within the terminal bronchioles.[97] It is now recognized that the causes of BOS are heterogeneous, and have differing clinical courses. Both viral and bacterial infections have been suggested to be risk factors for the development of BOS.[98,99] New evidence emerging with the use of nonculture-based techniques indicates that the underlying lung microbiome may play a greater role.[8] Early studies investigating the respiratory microbiome post lung transplantation indicate that lung transplant recipients have variations in their lung microbiome compared with healthy controls,[100] and furthermore, that the microbiome changes over the post-transplant period[79] as the lung microbiota becomes established within its new host and as an equilibrium is reached between the host immune response and immunosuppression levels. In these studies that have shown differences in transplant microbiome compared with healthy controls, one study has indicated increased bacterial burden in the BAL of lung transplant recipients,[79] while in others, decreased bacterial burden was observed in transplanted lungs.[81,82,100] Some suggestions of decreased community diversity were seen. However, this appears to more commonly be the case in those with active respiratory symptoms and is not necessarily observed in asymptomatic controls.[82] Therefore, further information about the differences between symptomatic and asymptomatic transplant patients is required including elucidating how these may correlate with both microbiome dysbiosis and negative outcomes in this cohort.

Several studies have investigated the relationship between the presence of specific members of the microbiome and the risk of BOS. This relationship was first demonstrated using traditional culture-based techniques and serum antibody detection. More recently, the advent of metagenomics techniques has allowed the dynamics of the greater lung microbiome to be examined in more detail. Some bacterial species, such as Pseudomonas, appear to play a greater role in mediating the development of BOS. Multiple culture-based studies have consistently shown that airway colonization with P. aeruginosa is predictive of subsequent development of chronic rejection."[99–103] However, one study has shown a negative correlation between presence of Pseudomonas spp. and development of BOS.[81] In those studies where a positive relationship between Pseudomonas colonization and development of BOS was seen, de novo colonization post-transplant rather than persistence of pretransplant Pseudomonas colonization had an increased likelihood of BOS development.[81,103]

Patients with suppurative lung disease, including CF, who had chronic colonization prior to transplant are at a greater risk of reinfecting the allograft with persistent bacterial species from the upper airway and the sinuses (Figure 2). Vital et al attempted to mitigate this risk by performing endoscopic frontal-spheno-ethmoidectomy procedures immediately post-transplant and using multiple Pseudomonas-active antibiotics. Despite this treatment, 65% of patients had persistent colonization post-transplantation. This study also demonstrated greater BOS-free survival in those patients who did not demonstrate detection of Pseudomonas.[104] Dickson et al identified two distinct Pseudomonas species which were both prominent in the lung microbiome samples of lung transplant recipients, but were associated with distinct clinical phenotypes. P. aeruginosa, when detected, is usually seen in high abundance and is accompanied by acute clinical infection, whereas P. fluorescens is usually detected at a moderate abundance and is rarely correlated with development of acute symptomatic infection.[82] If the bacteria are only identified at the family level, the distinction between these species may be missed. P. fluorescens has been shown to be rarely detected using culture-based methodologies; thus, the authors suggest this may account for the lack of correlation between detection of Pseudomonas and development of BOS seen in the study by Willner et al, who utilized culture-independent techniques.[81] Therefore, characterization of the microbiome and subsequent monitoring may provide further information on risk of BOS. It also highlights the future possibility of modifying the microbiome in attempts to reduce the development of BOS.

The increased access to culture independent techniques has played an important part in defining the role of the respiratory microbiome in lung transplantation. Variations in the lung microbiome post-transplant appear to modify the risk of both acute and chronic rejection events. Lung transplantation provides us with a unique scenario in which to investigate the lung microbiome and the changes that may occur in response to a new host, surgery-related damage and donor-transmission events. The substantial follow-up opportunities presented have allowed the lower respiratory tract microbiome to be monitored longitudinally and for the dynamics and shift of microbial species to begin to be elucidated.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....