Effect of hyperglycemia on lung microbiota and treatment outcome in pulmonary tuberculosis: A scoping review

Protocol design

This scoping review adhered to the The Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) extension for scoping reviews (PRISMA-ScR) checklist (Kithinji et al., 2024). A scoping review methodology was appropriate since evidence on the relationships between PTB, hyperglycemia, and the lung microbiome is scarce and published in research articles with heterogeneous methods. A review protocol does not exist.

Literature search

Eligibility criteria

Peer-reviewed primary articles that explored the relationship between PTB, hyperglycemia, and lung dysbiosis were included in the study. The participants in the studies must have been aged over 18 years and diagnosed with PTB based on chest X-ray and bacteriological confirmation of the MTB infection. They must also have had hyperglycemia defined as fasting blood glucose >5.6 mmol/L, 2-hour oral glucose tolerance >7.8 mmol/L, or HbA1c >5.7%. The exclusion criteria eliminated studies whose outcomes were not objectively measured due to their susceptibility to bias. Studies published in non-English language and without English translations were also excluded because all the authors are English speakers.

Information sources

The MEDLINE, Scopus, and CINAHL Plus databases were searched to identify research studies that explored the relationship between MTB infection and hyperglycemia. The reference lists of the systematic reviews and meta-analyses retrieved from the search were screened to identify additional articles that would be included in this study. Google Scholar and Google were used to search and identify articles on the relationships between the significant changes in measured outcomes in the PTB - hyperglycemia publications with lung microbiota. Table 1 outlines the details of the searches conducted and the full search strategy for each source. The key search terms were tuberculosis, hyperglycemia, and lung microbiome. The key terms and their synonyms were combined using Boolean operators (AND, OR) to improve the sensitivity of the search ( Table 1).

Selection of sources of evidence

Titles and abstracts of the identified articles were screened to identify articles that met the inclusion and exclusion criteria. The full-texts of the relevant records were read to assess suitability for the study objectives. The study population, outcome measures, and results of the data analysis were examined to only select articles that explored the changes in physical health, immune cells, and cytokine levels in the interplay between pulmonary TB and hyperglycemia. Two reviewers, VMM and DK, a medical microbiologist and a research methodologist, respectively, scrutinized each article to identify relevant data regarding the effect of hyperglycemia on physical health, immune cell counts, and cytokine levels in patients with PTB. The remarkable changes in outcomes observed in the included TB-hyperglycemia studies informed the search for articles on the relationship between the changes and dysbiosis in the lung microbiota.

Data charting process

The researchers charted the data using a Microsoft excel data abstraction tool that captured information on the sample characteristics and the variables on the interaction between pulmonary TB and hyperglycemia. The two reviewers, VMM and DK, discussed any disagreements in their charting, which they resolved by engaging the co-authors to arrive at a majority decision.

Data items

The main data items included hyperglycemia in both pre-DM and DM, and PTB diagnosed through imaging or bacteriology as independent variables. The other data items included indicators of physical health in PTB and changes in immune cell counts and cytokine levels attributable to hyperglycemia. Changes in lung microbiota associated with changes in physical health, immune cell counts, and cytokine levels were the relevant data items in the second round of the search process. Details of the data items are provided in Table 2.

Synthesis of results

VMM and DK synthesized evidence on the effect of hyperglycemia on TB prognosis by grouping related evidence. The effects of hyperglycemia on physical indicators of health such as TB treatment failure and cavitation in the lungs upon chest X-ray; immune cells; and cytokines extracted from each of the sources were synthesized to elucidate the mechanism. The main strengths and weaknesses of the individual studies were considered during the synthesis to reduce the risk of bias.

Selection of sources of evidence

A total of 31 primary research articles were synthesized in this scoping review. Twenty-one were on the effects of hyperglycemia on various indicators of health in TB while 10 were on the effect of the changes in the indicators on the lung microbiota. Figure 1 shows the sources of evidence selected, screened, assessed for eligibility, and included in the review regarding the relationship between TB and hyperglycemia.

Figure 1. PRISMA flow diagram showing the selection of sources of evidence for the relationship between TB and hyperglycemia.

Characteristics and results of individual sources of evidence

Table 3 (Extended data; Kithinji et al., 2024) summarizes the details of the primary articles that were included in this scoping review. In instances where the study design was not identified in the article, it was deciphered from the description of the recruitment of participants and data collection processes.

Synthesis of results

Effects of hyperglycemia on overall TB treatment outcomes

The results on the effects of hyperglycemia on overall TB treatment outcomes are shown in Table 4 (Extended data; Kithinji et al., 2024). The findings show that hyperglycemia contributed to poor treatment outcomes among patients with TB (Lin et al., 2020; Liu et al., 2021; Meng et al., 2023; Pardeshi et al. 2024; Faurholt-Jepsen et al., 2013). A prospective cohort study in six clinics in Jilin province, China reported a statistically significant association between persistent hyperglycemia of any duration and unfavorable treatment outcomes among patients with TB (Lin et al., 2020). However, the study relied on clinical presentation of patients to classify them as TB-negative, thus giving rise to a risk of overdiagnosis. Pardeshi et al. (2024) and Liu et al. (2021) also found that patients with TB and transient hyperglycemia had 2-4 times the odds of treatment failure compared to euglycemic patients. Although the estimates of the incidence rates of events were imprecise based on the wide confidence intervals, they were significant and showed a temporal relationship between hyperglycemia and poor TB treatment outcomes. Similarly, Faurholt-Jepsen et al. (2013) reported that the risk of mortality among patients with DM-TB in Mwanza, Tanzania was five times the risk of mortality among patients with TB only within the first 100 days after initiation of TB treatment. Considering the large sample size and the cohort study design, the evidence presented by Faurholt-Jepsen et al. (2013) cements the relationship observed between hyperglycemia and poor treatment outcomes in the various studies.

Poor treatment outcomes co-occured with dysbiosis of the lung microbiota (Ticlla et al., 2021; Wu et al., 2013). Wu et al. (2013) found that dysbiosis in the lung microbiota is associated with poor TB treatment outcomes, treatment failure, and recurrence. Although Wu et al. (2013) analyzed sputum rather than bronchoalveolar lavage to characterize the lung microbiota, their findings suggest dysbiosis as a significant determinant of TB treatment outcomes. Ticlla et al. (2021), who sought to identify the bacteria that constitute lung microbiota in TB by analyzing the bronchoalveolar lavage fluid obtained from both lungs of patients with unilateral TB lesions, found that Cupriavidus spp. rather than Streptococcus spp. was the most dominant bacterial genus in the lower respiratory tract of patients with TB.

The form of dysbiosis in the lung microbiota is dependent on the pulmonary TB status. Wu et al. (2013) compared sputum microbiota across patients with three types of TB: new, recurrent and treatment-failure TB among 25, 30, and 20 patients, respectively. Higher abundance of Bulleidia spp. and Atopobium spp. was observed in patients with recurrent TB compared to newly diagnosed patients. The Pseudomonas / Mycobacterium and Treponema / Mycobacterium ratios were higher and lower, respectively, in patients with recurrent TB than in newly diagnosed TB patients. Similarly, in treatment failure TB, the Pseudomonas / Mycobacterium ratio was higher than in newly diagnosed cases (Wu et al., 2013). Patients with recurrent TB had more abundant Corynebacterium spp. compared to patients with treatment failure TB. On the other hand, the more abundant genera in the patients with treatment-failure TB compared to patients with recurrent TB included Campylobacter, Prevotella, Atopobium, Blastobacter, and Treponema (Wu et al., 2013).

Studies on lung microbiome have reported different compositions of the microbiota concurrent with variations in TB treatment outcomes; hence the TB’s pathogenesis could be due to the interplay of both hyperglycemia and lung dysbiosis. Species in the phylum firmicutes including Staphylococcus and Selenomonas were more abundant in the lung microbiota of patients with TB compared to patients without TB (Ding et al., 2021; Ticlla et al., 2021). Abundance of members of phylum Actinobacteria such as Mycobacteria was also evident in the lung microbiota of patients with TB compared to those without it (Ticlla et al., 2021). Porphyromonas in the phylum Bacteroidetes and Fusobacterium in the phylum Fusobacteria were more common compared to Streptococcus in the phylum Firmicutes in the sputum of patients with TB (Ding et al., 2021).

Poor physical health outcomes in TB were more common in hyperglycemic patients than in euglycemic patients (Barreda et al., 2020; Meng et al., 2023; Meng et al., 2022). Both the retrospective cross-sectional study by Barreda et al. (2020) and the cohort study by Meng et al. (2023) reported that hyperglycemic patients with TB were more likely to have various types of pulmonary lesions such as cavities, alveolar infiltrates, and fibrous tracts compared to euglycemic patients. The two-year follow-up period in the study by Meng et al. (2023) allowed the identification of hyperglycemia as a risk factor of poor health outcomes. The poorer the glycemic control, the higher the rates of pulmonary lesions. Byashalira et al. (2022) improved health outcomes of patients with DM-TB by integrating glycemic control into a TB and HIV care programme.

Effects of hyperglycemia on TB immunity

Results on the effects of hyperglycemia on the immunological markers of health in TB (Table 5 (Extended data; Kithinji et al., 2024)) show that hyperglycemia reduced the population and functioning of dendritic cells and macrophages in patients with DM-TB (Gomez et al., 2013; Kumar et al., 2017). Kumar et al. (2017) observed a negative correlation between hyperglycemia and the frequencies of monocytes, plasmacytoid, and myeloid dendritic cells, as well as intermediate and classical monocytes, upon phenotyping monocytes and dendritic cells ex vivo from patients with DM-TB comorbidity and others with only TB under standard TB treatment. Although the correlations were borderline, perhaps due to the small sample size and the cross-sectional study design, the results pointed to hyperglycemia’s downregulation of the functioning of monocytes and dendritic cells in patients with TB (Kumar et al., 2017). Another ex vivo study revealed lower attachment and phagocytosis of MTB in monocytes from patients with DM, especially upon heat inactivation of serum in the case of poorly controlled hyperglycemia (Gomez et al., 2013). This suggests that hyperglycemia could impair the functioning of monocytes against MTB by inhibiting their activation through the complement system. Similarly, multiple ex-vivo experiments showed that patients with DM-TB have significantly low complement receptor 3 or CD11b and MARCO levels compared to patients with TB, which translated to low capacity for phagocytosis in macrophages (Panda et al., 2022).

Hyperglycemia variedly affected the populations of different types of T-cells. A prospective comparative study demonstrated that within two months of TB treatment and follow-up, patients with DM-TB had lower CD4+ T cell counts compared to patients with TB only, DM only, and healthy controls (Rendón Ramírez et al., 2023). Although the study could have been underpowered and ungeneralizable due to a small sample recruited from a single hospital, its findings indicate the possibility of DM contributing to a decrease in CD4+ T cells count. On the other hand, a prospective observational cohort study reported that increased HbA1c in the course of TB treatment is associated with elevated Th1 and Th17 CD4+ responses (Krause et al., 2023). In an experimental study, the numbers of Th17 CD4+ T cells and a regulatory T cell subset, RORγt⁺ Tregs, changed depending on the presence of Lactobacillus murinus (Lagilactobacillus murinus) strain (CNCM I-5314), which is a pulmonary Lactobacillus strain (Bernard-Raichon et al., 2021). Intranasal introduction of the CNCM I-5314 and not intragastrically in mice infected with MTB increased the expansion of lung leukocytes, particularly Th17 CD4+ T cells and RORγt⁺ Tregs locally in the lungs and airways (Bernard-Raichon et al., 2021). Upon administering the CNCM I-5314, the proportion of RORγt⁺ Tregs increased most prominently since they changed from 14% to 50% of all the lung Foxp3⁺ CD41 T cells in MTB-infected mice.

In an experiment that tested the activation of immune cells from patients with active and latent TB by stimulating the peripheral blood mononuclear cells with Mycobacterium bovis BCG, an inverse correlation was observed between the level of glycemia and IFN-γ- and TNF-α- CD4+ T cell productions (Boillat-Blanco et al., 2018). This suggested that hyperglycemia could be downregulating antigen processing and presentation of the whole MTB, thus stalling the production of cytokines that promote TB immunity. Although the study did not include patients with transient hyperglycemia, the findings among patients with DM empirically suggest the effect of hyperglycemia on TB immunity.

Findings from different studies show that hyperglycemia could be disrupting cytokine responses in favor of TB immunopathology. An ex-vivo experimental study on the mRNA expression of pro-inflammatory and anti-inflammatory cytokines in monocyte-derived macrophages reported a positive correlation between IL-1β expression and HbA1c levels (Panda et al., 2024). Levels of IL-1β were also elevated among patients with DM-TB compared to patients with DM only or TB only (Kumar et al., 2020). In contrast, lower levels of IL-1β were reported in patients with DM- TB comorbidity compared to patients with TB only (Chao et al., 2015).

IFN-γ levels were low in newly diagnosed DM and TB comorbidity cases. Patients with newly diagnosed DM (NDM) were characterized by significantly diminished unstimulated levels of INF-γ (Kumar et al., 2020; Panda et al., 2024). Additionally, INF-γ levels were positively correlated with IL-β levels (Panda et al., 2024). Earlier, Kumar et al. (2014) reported that latent TB in treatment-naïve patients with DM or pre-DM was associated with diminished circulating levels of INF-γ.

IFN-γ levels were elevated in patients with known DM-TB comorbidity compared to patients with DM only (Chen et al., 2022; Kumar et al., 2020). Although the power of the study by Chen et al. (2022) was low due to a small sample size, the observed differences between the compared age- and sex-matched groups suggest a possible effect of the hyperglycemia on the cytokine levels. Similar findings have been reported from a cross-sectional study indicating that INF-γ levels were more raised in patients with DM but with controlled hyperglycemia and latent TB than in healthy endemic controls upon purified protein derivative antigen (PPD) stimulation (Masood et al., 2022).

Alteration of the lung microbiota affected the production of INF-γ (Bernard-Raichon et al., 2021; Palma Albornoz et al., 2021; Xia et al., 2022). The introduction of CNCM I-5314 into the lung microbiota significantly decreased the production of INF-γ (Bernard-Raichon et al., 2021). Dysbiosis characterized by increased Bacteroides and Firmicutes in the lungs decreased the number of IFN-γ-producing CD4+ cells and consequently IFN-γ levels (Palma Albornoz et al., 2021). Xia et al. (2022) reported a positive correlation between IFN-γ and Leptotrichia and Gemella, which are members of the lung microbiota in healthy control populations.

IL-10 levels in DM-TB are inconsistent across various studies. For instance, IL-10 levels were significantly low in patients with increased HbA1c (Krause et al., 2023). In contrast, findings from a study by Chen and colleagues showed that the levels in DM-TB patients were double those in patients with DM only or TB only (Chen et al., 2022). In another in vitro study, IL-10 levels were significantly enhanced in the mice infected with the MS1987 protein (the virulence protein for MTB) especially after intranasal infection with anaerobic Escherichia coli and Delftia acidovorans (Fraser-pitt et al., 2023).

Significantly high IL-6 levels were reported among patients with DM-TB compared to patients with TB only or DM only, even in cases with latent TB (Chen et al., 2022; Kumar et al., 2020; Masood et al., 2022). This upregulation was mainly in patients with HbA1c levels greater than 8 g/dL. Contrary to these findings, an in vitro study reported a significant decrease in IL-6 in mice at day 9 post-infection with a model containing the MS1987 protein (Fraser-pitt et al., 2023). Intranassally introducing E. coli into mice increased the production of IL-6 while exposure to D. acidovorans inhibited IL-6 production. Xia et al. (2022) also reported a negative correlation between IL-6 levels and Corynebacterium in patients with TB.

IL-17 levels were elevated among DM-TB patients (Chen et al., 2022). Similarly, IL-17A levels were high in patients with DM-TB but diminished in patients with newly diagnosed DM and TB comorbidity (Kumar et al., 2020). Kumar et al. (2014) also reported diminished levels of IL-17F in patients with latent TB and DM or pre-DM. Fraser-pitt et al. (2023) found that models of MTB with the pathogenic protein MS1987 caused a decrease in levels of 1L-17 in the lungs at day nine post-infection. Administering E. coli intranassaly into mice led to increased production of IL-17. Administering CNCM-I-5314 caused the lung CD4+ T cells to produce more IL-17 and express more IL-17a mRNA (Bernard-Raichon et al., 2021). Xia et al. (2022) reported that IL-17 was positively correlated with Gemella and Streptococcus of the lung microbiota.

TNF-α levels varied depending on the status of the DM and TB. They were elevated in DM-latent TB cases upon PPD stimulation (Masood et al., 2022). Kumar et al. (2020) also reported increased levels of TNF-α in patients with DM-TB but the unstimulated levels were diminished in patients with newly diagnosed DM and TB comorbidity. Similarly, the circulating levels of TNF-α were diminished in patients with DM and pre-DM, and latent TB (Masood et al., 2022; Chen et al., 2022). Bernard-Raichon et al. (2021) reported significantly decreased production of TNF-α in lung exudates upon CMCM I-5314 treatment. Intranasal administration of D. acidovorans was shown to inhibit TNF-α production (Fraser-pitt et al., 2023).

IL-2 levels also varied depending on the status of the DM-TB. They were elevated in DM-latent TB cases compared to healthy endemic controls upon PPD-stimulation (Masood et al., 2022; Kumar et al., 2020). Unstimulated levels were significantly low in patients with newly diagnosed DM and TB comorbidity (Kumar et al., 2020). Similarly, diminished circulating levels of IL-2 were associated with latent TB infection in DM or pre-DM patients (Kumar et al., 2014).

Effects of hyperglycemia on immune responses of patients with DM-TB

Dendritic cells and monocytes are central in the presentation of MTB antigens to the naïve T-cells in order to activate the adaptive immune response but their numbers are decreased in DM (Comberiati et al., 2021). T-cell-mediated adaptive immune responses are integral in controlling MTB infections in humans (Jasenosky et al., 2015). For instance, the availability of polyfunctional T cells, which are desirable in the fight against TB due to their capacity for effector and proliferative abilities considering their trifunctional production of TNF, IFN-γ, and IL-2 determines TB progression (Jasenosky et al., 2015). Therefore, a decrease in CD4+ T cells during DM-TB comorbidity may impede the induction and maintenance of the adaptive immunity against TB.

The contrasting findings on IL-1β levels in similar studies among patients with DM-TB suggests that the production of IL-1β may be influenced by other factors such as the status of DM or TB. MTB decreases the activation of pyrin and AIM2- domain containing 3 (NLRP3)-inflammasomes, which induces the production of INF-β that subsequently reduces the secretion of IL-1β (Etna et al., 2014). A prospective cohort study that follows up patients with TB and hyperglycemia in the course of disease and treatment is essential to define the factors behind varying patterns of IL-1β levels among patients with DM-TB.

IFN-γ levels are low in newly diagnosed DM-TB and high in known DM-TB cases. When diabetes status is known, treatment is often initiated, hence hyperglycemia is controlled (Li et al., 2020). It is therefore important to control hyperglycemia in DM-TB to reduce the blunting of IFN-γ production. However, this is a double edged sword as the reduction of the pro-inflammatory IFN-γ in hyperglycemia not only slows down the inflammation-based pathology of TB but also hinders killing of the bacteria, induction of nitric oxide synthesis, and macrophage activation that depend on IFN-γ (Etna et al., 2014).

In newly diagnosed DM-TB, IL-2, IL-17, and TNF-α levels are low, while in known DM-TB, IL-2, IL-17, and TNF-α levels are high. Chronic hyperglycemia rather than acute hyperglycemia is important in inducing inflammation in the lungs through cytokines such as IL-2, IL-17, and TNF-α (Ngo, Bartlett and Ronacher, 2021). IL-17 is involved in the immune defense against bacteria, including MTB. As a pluripotent pro-inflammatory cytokine, it recruits neutrophils, induces chemokines’ and cytokines’ production, modifies the differentiation of T-helper cells, and prompts the release of antimicrobial proteins that mediate inflammation in TB (Chung, Ye and Iwakura, 2021). Similarly, TNF-α, which induces bactericidal effects in macrophages and has pro-inflammatory effects (Yuk et al., 2024), is low in newly diagnosed DM-TB but high in known DM-TB. Its levels are also high in active TB (Suzukawa et al., 2020), which may explain why they are elevated in DM-TB but suppressed in newly diagnosed DM and TB comorbidity. IL-2 is central in adaptive immune responses and immune tolerance (Damoiseaux, 2020), hence alterations of its levels by hyperglycemia may have adverse health implications.

Effects of hyperglycemia-instigated changes on the lung microbiome in TB

The lungs are not sterile since bacteria and fungi dominate the lower airways. Mycobacteria spp. are present in the lower airways of both individuals with and without TB, although they are dominant among patients with TB (Hu et al., 2020). This scoping review demonstrates that patients with TB have different compositions of lung microbiota with varying patterns. Pseudomonas, Acinetobacter, Porphyromonas, Bulleidia, and Selemonas were abundant while Treponema, Streptococcus, and Prevotella were scanty in patients with TB compared to healthy controls. Other studies have reported the abundance of Gramulicatella and Pseudomonas in the lung microbiota of patients with TB (Wu et al., 2013). Sharpea and Bergeyella were distinctively detected in the sputum of patients with TB. The dominant genera in healthy subjects include Streptococcus, Selenomonas, Prevotella, Bifidobacterium, and Neisseria. Others include Leptotrichia, Catonella, Treponema, and Coprococcus (Wu et al., 2013). The fungal members of the lung microbiome belong to the phyla Ascomycota and Basidiomycota (Hu et al., 2020). Patients with treatment failure TB had more abundant Pseudomonas than patients with newly diagnosed TB while Corynebacterium was more common in recurrent TB compared to treatment failure TB (Wu et al., 2013). Therefore, there is varying lung dysbiosis in patients with different TB status, which implies that patients with DM-TB comorbidity may have a distinct composition of lung microbiome.

The lung microbiome remains vital in determining pulmonary health and disease. The poor treatment outcomes evident in patients with both TB and hyperglycemia indicate the possibility of hyperglycemia contributing to the lung dysbiosis. Since studies on lung microbiome have also reported variations in composition of the microbiota based on TB treatment outcomes, TB pathogenesis could be due to the interplay of both hyperglycemia and lung microbial dysbiosis. Hyperglycemia is commonly associated with dysbiosis in the gut (Darra et al., 2023), it is therefore likely to be associated with dysbiosis in the lungs given the gut-lung axis. Thus, there is a need to determine the relationship between hyperglycemia and dysbiosis in the lung microbiota among patients with TB.

Changes in T-cells populations such as regulatory T cells, which the composition of lung microbiota licenses, are central in the control of pathogenesis of TB (Mori, Morrison and Blumenthal, 2021). As a result, in patients with DM-TB, hyperglycemia could influence lung dysbiosis, subsequently reducing the production of polyfunctional T cells and compromising the adaptive immune response against TB.

The present review of literature did not find evidence on the interactions between lung microbiota and dendritic cells in patients with TB. However, some members of the microbiota may interact with the dendritic cells to produce cytokines that promote type 1 immunity (Koizumi et al., 2008). Similarly, the interaction between commensal bacteria through their metabolites and the immune system can contribute to the differentiation of monocytes to macrophages and boost the host innate defense against MTB (Schulthess et al., 2019). Thus, dendritic cells and monocytes may mediate the interactions between the lung microbiome and hyperglycemia, thus influencing the course of TB in patients with DM-TB comorbidity.

Changes in the levels of IFN-γ, IL-17, and TNF-α, which are associated with both hyperglycemia and lung dysbiosis, are key players in the inflammatory circuits that control the progression of TB (Mori, Morrison and Blumenthal, 2021). For instance, less IFN-γ is produced in TB disease states when Bacteroidetes and Firmicutes are dominant in the lung microbiome. On the other hand, more IFN-γ is produced in health state when Gemella and Leptotrichia dominate the lung microbiota in healthy individuals. IL-10, which is produced to modulate the pathogenesis of TB, acts by dampening excess inflammation (Etna et al., 2014). The increase in IL-10 levels with the intranasal administration of E. coli and D. acidovorans implies that lung dysbiosis contributes to IL-10 levels variation in enhancing or dampening inflammation. Similarly, lung dysbiosis affects IL-17 levels as evident in upsurge in IL-17 when dysbiosis was altered via intranasal exposure to E. coli, Gemella, Lactobacillus, and Streptococcus. This scoping review also shows that TNF-α levels were diminished upon intranasal introduction of Lactobacillus and D. acidovans in the lungs. Thus, hyperglycemia may contribute to the pathogenesis of TB by altering the lung microbiota, which then disorients the immune response to TB through changes in cytokine levels.

Increase in IL-6 levels, common in patients with DM-TB, could be due to alterations in the composition of the lung microbiome. In this scoping review, we found that, IL-6 levels in mice infected with TB could be increased by intranasally introducing E. coli or reduced by administering D. acidovorans. Corynebacterium, whose low population in the lung microbiota is associated with increased IL-6 levels, is a major immunomodulatory bacterium in the lung microbiota (Liu et al., 2022). IL-6 is in the pro-inflammatory cytokine family and a major player in the human cytokine network (Uciechowski and Dempke, 2020), hence variations in its levels should be explored when studying the relationship between hyperglycemia and lung microbiota in TB pathogenesis.

The main weakness of this scoping review is that majority of the included studies analyzed sputum samples to assess lung microbiome. There was, therefore, a possibility of specimen contamination with bacterial genera that are common in the oropharynx such as Prevotella, Atopobium, and Bulleidia. As a result, it is likely that the oropharynx microbiota would be misclassified for lung microbiota (Mori, Morrison and Blumenthal, 2021; Goddard et al., 2012). This underlines the need to consider bronchoalveolar lavage (BAL) specimen for an accurate depiction of the lower respiratory system microbiota.

Ethics and consent

Ethical approval and consent were not required.