Does Inhaled Corticosteroids Increase the Risk of Nontuberculous Mycobacteria Pulmonary Disease: A Meta-Analysis

Li. Does Inhaled Corticosteroids Increase the Risk of Nontuberculous Mycobacteria Pulmonary Disease: A Meta-Analysis. Biomed of NTM-PD with the usage of ICS among chronic respiratory patients. Methods: The PubMed, EMBASE and Web of Science databases were applied to identify eligible studies until Nov. 18, 2019. Studies focusing on the correlation between ICS and NTM-PD among chronic respiratory patients were concluded. Odds Ratio (ORs) with 95% confidence intervals (95% CIs) were extracted and synthesized to evaluate the risk of NTM-PD with the usage of ICS. Heterogeneity was assessed by using Cochran Q and I-square statistics. Sensitivity and subgroup analysis were performed to trace the source of heterogeneity. Results: A total of 5 studies with 20968 chronic respiratory patients were enrolled in current meta-analysis. The pooled OR for the risk of NTM-PD with the usage of ICS was 1.97 (95% CI: 1.45—2.67, I2=77.3%, P=0.001). In sensitivity analysis, Bual’s study was the origin of heterogeneity. In the subgroup analysis, there was a trend towards the increased risk of NTM-PD with the usage of ICS among asthma and COPD patients. In addition, high-dose and current usage of ICS could increase the risk of NTM-PD. A positive relationship was the of ICS and the risk of NTM-PD. In the future, the selection of ICS should be more cautious.


Introduction
Inhaled Corticosteroids (ICS) has been widely used for the treatment of chronic respiratory diseases, especially Chronic Obstructive Pulmonary Disease (COPD) and asthma, in order to improve symptoms, reduce airway inflammation, limit further disease progression. ICS is delivered directly to lungs and therefore ICS may increase the susceptibility of local (lung) infections. Some studies have confirmed that long-term usage of ICS might result in some adverse effects, including increases in the incidence of pneumonia [1,2], oropharyngeal candidiasis [3] and Tuberculosis (TB) [4]. Corticosteroids can attenuate cellular immunity against intracellular pathogens, including mycobacteria, and raise the potential of predisposing patients to the incidence and persistence of Nontuberculous Mycobacteria (NTM) infections. Recent studies suggested that the presence of chronic respiratory diseases were associated with an increased likelihood of NTM Pulmonary Disease (NTM-PD) [5,6]. Some studies also reported that there was a significantly increased odds of NTM-PD among chronic respiratory disease patients who were exposed to ICS, especially fluticasone [7][8][9].
On one hand, current usage of ICS was reported to be associated with significantly increased risk of NTM-PD compared with nonusage, and this risk was showed to have a strong dose response relationship [10,11]. On the other hand, NTM-PD could make the original chronic respiratory diseases more difficult to treat, especially in older individuals with more severe airflow obstruction and greater exposure to inhaled corticosteroids [7]. However, the sample sizes of the above studies were relatively modest, and the results were inconclusive. Therefore, it seemed worthy to conduct a meta-analysis to assess the risk of NTM-PD with the usage of ICS among chronic respiratory disease patients. This study also further explored the relationship between the dosage/exposure of ICS and the risk of NTM-PD.

Search Strategy
The present study was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guideline. Comprehensive literature search was conducted in the following databases, including PubMed, EMBASE and Web of Science. No language or data restrictions were imposed and the searching was updated to Nov. 18, 2019. The keywords of searching included "Mycobacterium" or "Nontuberculous" and "Inhaled corticosteroids" or "Inhaled steroids". Meanwhile, reference lists of retrieved articles were examined manually to further identify the potentially relevant publications. If analysis was based on the same patients' origin, only the latest research was considered.

Selection Criteria
The inclusion criteria for eligible studies include:

Data Extraction and Quality Assessment
The full texts of the selected studies were carefully reviewed, and data were extracted by two independent researchers. A third investigator was consulted to resolve inconsistencies. The following data were extracted: the first author's name, year of publication, country of research, sample size, study type, diseases, NTM pathogens, ICS types, ICS exposure, ICS dosage and OR. The Newcastle-Ottawa Scale (NOS) was used to assess the study quality.
The selection, comparability and exposure were evaluated. The scores according to NOS varied from 0 to 9. A score of 6 or more was recognized as high quality. The NOS was used to assess the equality of each selected study by two independent authors.

Statistical Analysis
Statistical analyses of this meta-analysis were fulfilled using Stata version 12.0 (Stata Corporation, College Station, TX, USA).
Pooled ORs and 95% CIs were used to evaluate the risk of NTM-PD caused by ICS among chronic respiratory disease patients. The heterogeneity across studies was tested by Cochran's Q and Higgins I2 statistics. P < 0.05 and I2>50% were considered to be significant heterogeneity, while I2 < 25% and 25% < I2 < 50% indicated no heterogeneity and moderate heterogeneity, respectively. A random effects model was used when statistical heterogeneity was detected (P < 0.05, I2 > 50%); otherwise, the fixed effects model was applied.
Subgroup analysis was performed to detect the origin of the heterogeneity when the data was too heterogeneous. Additionally, sensitivity analysis was carried out by sequentially omitting individual studies at each step. If the results did not substantially alter when one study was excluded, this meant that the pooled results were stable. Publication bias was evaluated by Egger's test, and P < 0.05 indicated that a statistically significant publication bias may exist.

Literature Search
The search and selection strategy were shown in Figure 1

Characteristics of Eligible Studies
The main characteristics of included studies are shown in Table 1 (Figure 3). We also found Bual's study was the origin of heterogeneity through sensitivity analysis (Figure 4).

Subgroup Meta-Analysis According to Potential Confounding Factors
Because a substantial heterogeneity existed, subgroup analyses were carried out to explore the potential sources of heterogeneity.  Figure 8).

Publication Bias
Publication bias was assessed by Egger's test. The result suggested that no asymmetry was existed (Pr>|z|=0.854).

Discussion
Worldwide, pulmonary infections caused by nontuberculous mycobacteria are gaining increasing attention. NTM-PD showed relatively poor response to current multidrug therapy [13] and resulted inthe universal structural lung injury [14]. NTM-PD is more common in individuals with chronic respiratory diseases such as asthma [15], COPD [16], cystic fibrosis [17,18], bronchiectasis [19,20], which may predispose to NTM infection by causing chronic epithelial cell inflammation and impaired mucociliary clearance.
Recent published studies illustrated that the increasing prevalence and budesonide was likely due to differences in pharmacokinetic and pharmacodynamic properties [22]. Fluticasone has a greater effect on glucocorticoid receptors, is more lipophilic and has a longer half-life than budesonide [23]. The pathophysiological mechanisms that contribute to an increased susceptibility to NTM-PD in patients treated with ICS are unclear. The usage of ICS could diminish cytokine production and CD8T cell activation among severe COPD patients [24], while CD8T cells were essential for protective immunity in nontuberculous mycobacteria [25].

Mucosal-associated invariant T (MAIT) cells were numerically and
functionally deficient in NTM patients and these deficiencies could contribute to immune system dysregulation in NTM infections [26]. MAIT cells were also deficient in blood and bronchial tissue in ICS-treated, but not ICS-untreated, COPD patients [27]. Besides, previous studies showed that IFN-γ production by NK cells played an important role in activating and enhancing innate and adaptive immune responses at early stage of pulmonary NTM infections [28]. Furthermore, the earlier reports had demonstrated that in vitro treatment with inhaled fluticasone propionate significantly decreased NK cell activities [29].
There are some limitations in our study. First, most studies only followed less stringent microbiological criteria, which might lead to overestimation of their risk of NTM pulmonary disease, despite all of studies based on ATS/IDSA criteria. Second, in subgroup analysis, there were different definitions regarding current/prior ICS exposure and high-dose/low-dose ICS, which might reduce the effectiveness of polled ORs. Third, crude OR was provided in most studies, but not adjusted OR, which might be more susceptible to confounding factors. Fourth, the eligible studies were relatively small and retrospective in design. Finally, Egger's test has relatively lower power when the number of studies included in meta-analysis less than 10. Hence, the publication bias might not be detected by Egger's test.

Conclusion
Based on this study, it can be concluded that patients with chronic respiratory diseases, particularly asthma and COPD treated with ICS, are at highly increased risk of NTM-PD. Therefore, we believe the risk of NTM-PD should be considered when prescribing ICS, especially since NTM-PD is a chronic infection that is extremely difficult to treat. Clinicians should consider this risk when prescribing ICS to patients and if they are needed, strive to start from the lowest effective dose.

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflicts of Interest
The authors declare no conflict of interests.