1. Introduction

Non-cystic fibrosis bronchiectasis (NCFB) is defined as an irreversible and progressive dilatation of bronchi due to chronic bronchial inflammation [1,2].

Clinical presentations may range from no symptoms to chronic cough with frequent sputum production, hemoptysis, dyspnea, decreased exercise tolerance, frequent exacerbations, and respiratory failure requiring, in some cases, lung transplantation or causing death.

From a pathophysiological point of view, current literature supports a synergistic and amplifying involvement of infection, inflammation, and repair of the bronchial mucosa, which causes a vicious cycle similar to that described by Peter Cole [3].

Recently, Bush and Floto proposed a possible pathophysiological mechanism, which involves: (a) persistent or recurrent infection, (b) impairment of mucociliary clearance, and (c) airway obstruction [4]. Specifically, persistent or recurrent infections cause progressive neutrophilic inflammation [5,6]—favoring bronchial wall damage—Th17-biased adaptive immunity [7]—promoting enlargement of lymphoid follicles [8], neutrophil recruitment, and mucus hypersecretion [7]. The consequent loss of airway epithelium integrity impairs mucociliary clearance with subsequent airway occlusion. Finally, airway obstruction promotes further bronchial dilatation [9], retention of secretions, which in turn, attracts bacterial colonization and predisposes the patient to repeated infections.

Medical diagnosis begins by excluding patients with cystic fibrosis (CF); indeed, in some diagnostic and treatment guidelines, bronchiectasis is labeled as NCFB to capture all other conditions [1]. However, pathological findings of bronchiectasis associated with CF are indistinguishable from those found in NCFB. In general, CF is characterized by a worse clinical course with a high prevalence of gram-negative infection, especially with Pseudomonas aeruginosa [10].

The previously described pathophysiological process can be useful to systematically research the multiple causes of bronchiectasis.

Starting from the first culprit step of the process—persistent or recurrent infection—we can identify two main etiological groups: (1) childhood infections, including persistent bacterial bronchitis (PBB) [11], pneumonia, measles, whooping cough and tuberculosis; and (2) adult infections, including non-tuberculous mycobacteria (NTM) infections [12]. Instead, progressive inflammation is the main pathophysiological determinant in other two etiological groups of NCFB: (1) toxic damage to airways, as in case of inhalation, aspiration secondary to neuromuscular diseases and gastro-oesophageal reflux disease (GERD), and (2) systemic inflammatory diseases, including inflammatory bowel disease [13], connective tissue diseases [14] and yellow nail syndrome [15]. In contrast, defects in the immune system are responsible of another etiological group of NCFB including primary and secondary immune deficiencies. Primary immunodeficiency accounts for 12–34% of NCFB [16]. The most common forms of primary immune deficiencies include: common variable immune deficiency (CVID), X-linked agammaglobulinemia (XLA), chronic granulomatous disease (CGD), and antibody deficiency with normal IgG [17]. Bronchiectasis is also associated with HIV infection [18,19]. Other secondary causes of immunodeficiency include hematological malignancy, drug-induced immunosuppression, and post-allogenic bone marrow transplantation [20,21]. Finally, a mixed hypersensitivity reaction, including type I, III, and IV features, characterizes the inflammatory damage found in allergic bronchopulmonary aspergillosis, a condition frequently associated to bronchiectasis [22,23].

The impairment of mucociliary clearance is the cornerstone of another group of conditions, which characterizes the development of bronchiectasis, including primary and secondary ciliary dyskinesia [24] and channelopathies comprising CF transmembrane regulator (CFTR) dysfunction and epithelium sodium channel (ENaC) dysfunction [25].

Finally, airway obstruction is the oldest determinant of damages found in bronchiectasis. Laennec [26] identified for the first time this condition, which favors the onset of bronchiectasis in a wide group of conditions including: pulmonary structural alterations—typical of Williams-Campbell syndrome [27], Mounier-Kuhn syndrome [28] and Ehlers-Danlos syndrome [29]—single bronchial obstruction—as in case of neoplasms or foreign bodies—and, finally, obstructive diseases—such as asthma [30], chronic obstructive pulmonary disease (COPD) [31], and alpha-1-antitrypsin deficiency [32].

Despite etiological testing, it is not always possible to identify an underlying condition. Under these circumstances, bronchiectasis is labeled as idiopathic.

To date, respiratory infections play a key role in causing and worsening bronchiectasis. Furthermore, they are also considered an indicator for disease severity. Consequently, optimal management of infections is crucial in order to break the vicious circle described. For all these reasons, antibiotics are considered a treatment pillar.

2. The Role of Pseudomonas aeruginosa and Other Microorganisms in Non-CF Bronchiectasis

Diverse polymicrobial communities are present in the airways of patients with bronchiectasis and many microorganisms have been associated with bronchiectasis, as both a complication and a cause of the anatomic abnormalities (Table 1) [33,34,35].

Table 1. Bacteriology of bronchiectasis.

The most frequent microbiological findings are in bold. N/R: not reported. *: Studies in which the Authors report microbiological data only as percentages of the total sample.

Clinical and radiographic features in bronchiectasis do not allow the identification of involved pathogens, but can be investigated as possible markers for specific infections. For example, in an Israeli cohort of bronchiectasis patients, age less than or equal to 64 years was associated with a Haemophilus infection, while age over 64 was associated with an increased risk of Pseudomonas and Enterobacteriaceae [36]. However, this has not been confirmed in other large patient cohorts. Chronic Pseudomonas infection is known to be more common in more severe patients [37,38,39,40]. Nevertheless, other microbes can be found and cause patient’s symptoms. Therefore, it is mandatory to perform regular sputum cultures in all NCFB patients to optimize treatments and evaluate prognosis.

Pseudomonas aeruginosa, responsible for 20–40% of bacterial infections in bronchiectasis, is a gram-negative rod-shaped bacterium, which can grow rapidly under aerobic conditions. Furthermore, this facultative anaerobe can grow even in the absence of oxygen. It is a common human opportunistic pathogen capable of causing a wide range of infections. P. aeruginosa usually benefits from impaired defense mechanisms to cause both acute and chronic infections. The bacterium is inherently resistant to many antibiotics due to a difficult penetrable membrane and the presence of multiple efflux pumps. Furthermore, P. aeruginosa can develop resistance also through mutations or gene acquisition via horizontal gene transfer. Persistent P. aeruginosa infection has been associated to poorer outcomes, as airway inflammation, morbidity, hospitalization risk and premature mortality, in both CF and NCFB [41,42,43].

Many epidemiologic cohorts report Haemophilus influenzae as the most common bacterial organism. This pathogen is described in more than 30% of patients with chronic bronchiectasis [44,45,46,47]. Other gram-negative organisms isolated from patients with bronchiectasis include: Stenotrophomonas maltophilia [48], Klebsiella pneumonia [49], Moraxella catarrhalis [50], Achromobacter [51], Serratia marcescens [52], and Escherichia coli [53]. Instead, Staphylococcus aureus [54] and Streptococcus pneumoniae are gram-positive organisms most frequently seen in bronchiectasis patients [52,55,56].

Another group of pathogens found in NCFB patients is NTM. Members of this family, most frequently found in NCFB, are Mycobacterium avium complex and Mycobacterium abscessus [57]. Aspergillus [58], and Candida [59] are fungi commonly identified in the respiratory secretions of NCFB patients [59,60]. It can be difficult to determine if these organisms play a pathophysiological role in infectious symptoms or if they are simply bystander organisms in patients with other primary pathogens. In case of persistently positive cultures for Aspergillus, without other isolates, antifungal treatment should be considered. Instead, treatment for Candida is rarely needed as this microorganism is usually an oral contaminant.

The role of viral infections in bronchiectasis is unclear. To date, data regarding chronically infected patients and the trigger role of viruses are limited. In 2015, Gao et al. studied the incidence and clinical impacts of viral infection during bronchiectasis exacerbations in an adult population. Coronavirus, Rhinovirus, and Influenza A and B were the most commonly isolated viruses [61]. In the same year, Metaxas et al. explored the role of viruses in NCFB using polymerase chain reaction in bronchoalveolar lavage samples finding respiratory syncytial virus during stable visits, but not during bronchiectasis exacerbations [62].

Furthermore, it is common to grow “normal oropharyngeal” flora from lung secretions of patient with bronchiectasis [63]. To a certain degree, culturing only normal flora may be a limit of culture techniques routinely in use to identify pathogens in some bronchiectasis patients. Additionally, it is difficult to obtain sufficient sputum for laboratory processing.

3. Antibiotic Resistance in Bronchiectasis

Lungs’ microbial ecology, antibiotic type, and administration consistency can strongly affect antibiotic resistance risks in patients with bronchiectasis. Past treatment errors caused a globally known health threat, known as MDR, with negative clinical practice impacts if inadequately identified and treated. In the following a list of pathogens with their MDR features: (1) Enterobacteriaceae: extended-spectrum-β-lactamase (ESBL) producing, resistance to most β-lactam antibiotics, such as cephalosporins, penicillin, and aztreonam; (2) P. aeruginosa: resistance against at least one agent in three or more antimicrobial categories, and; (3) S. aureus: oxacillin resistance with a MIC of ≥4 mcg/mL [100]. Chronic colonization with pathogens occurs more frequently in bronchiectasis patients causing recurrent exacerbations and infections for which patients take several broad-spectrum antibiotics. Those treatment cycles facilitate the development of MDR pathogens, which are frequently diagnosed in bronchiectasis patients, because of their specific treatment needs, restricted diversity of antibiotics and the fact that ESBL-producing Enterobacteriaceae and P. aeruginosa are likely MDR pathogens, which are very aggressive in patients with respiratory disorders necessitating specific antibiotic treatment, usually not recommended in guidelines [101].

During episodes of exacerbation, MDR pathogens are frequently isolated from patients with bronchiectasis, particularly if hospitalized. Pseudomonas, MRSA and ESBL + Enterobacteriaceae are the most frequently encountered MDR bacteria. Several risk factors are independently associated with the isolation of MDR bacteria—the most frequent being previous MDR isolation, hospitalization in the previous year and chronic kidney disease. Studies focusing on pneumonia recognize chronic kidney disease as a risk factor for MDR bacteria. Another widely recognized independent risk factor for MDR, particularly MRSA and Enterobacteriacea, is previous hospitalization for exacerbation, especially if there has been excessive or inadequate use of 3rd/4th generation cephalosporins or broad spectrum penicillins in this condition. MDRs are more frequently hospital related and not acquired in the community; their incidence and spectrum corresponds to the treatment attitude towards these challenging pathogens considering preceding antibiotic therapy and patient characteristics [42]. Furthermore, it is known that elderly patients, especially those suffering from multiple comorbidities, are at greater risk of exacerbations from MDR bacteria, indicating possible associations between risk of MDR infection and more debilitating diseases, previous use of inhaled antibiotics, and long-term oxygen therapy [101].

These risk factors are common in patients over 65 years with bronchiectasis, making the impact of MDR bacteria significant, especially during flare-ups. Hence, performing sputum cultures during the flare-up phases of bronchiectasis is important to optimize the choice of antibiotic therapy and to avoid excessive use of broad-spectrum antibiotics. It is essential to modify empirical therapy based on microbiological results and antibiotic sensitivity tests as soon as the results of the sputum culture are available. Nevertheless, empirical antibiotic therapy should be based on preceding microbiological findings until new results are available.

The use of broad-spectrum antibiotics against MDR pathogens must be based on the presence or absence of risk factors. In fact, the use of these antibiotics is indicated if two or more risk factors are present. Whether or not a broad-spectrum antibiotic is confirmed will depend on a microbiological work-up. This strategy allows minimizing broad-spectrum coverage for MDR in episodes of exacerbation for which such antibiotics are not needed and reducing the future appearance of resistant microorganisms. In general, MDR pathogens are related to longer hospital stays, increased need for antibiotics and utilization of healthcare resources, and may adversely affect patient outcomes [101].

In the treatment of chronic infection, long-term use of macrolides may lead to an increased risk of antimicrobial resistanc. In fact, the BAT study showed that after 12 months of treatment with azithromycin, patients had 88% macrolide resistance rate compared to 26% observed in the placebo group. Interestingly, over 80% of the overall pathogens (H. influenzae, S. pneumoniae, M. catarrhalis, and H. parainfluenzae), considered to be chronically present in the airways of recruited patients, were potentially sensitive to azithromycin, while only 11% of all enrolled patients were infected with P. aeruginosa, a bacterium naturally resistant to macrolides [80]. Hence, attention needs to be paid to chronic treatment with macrolides, which should be reserved only for patients with bronchiectasis with more than three exacerbations per year. The results of the BLESS study also show that treatment with macrolide (erythromycin, 400 mg, administered twice daily consecutively over 12 months) increases the percentage of macrolide-resistant oropharyngeal streptococci [79]. The clinical relevance of the increased risk of antibiotic resistance resulting from the long-term use of macrolides must be further investigated considering also characteristics and microbiological data of patients with bronchiectasis. To date, the major concern about the long-term use of macrolides is the possible impact on antibiotic resistance of NTM. Indeed, newer macrolides such as azithromycin and clarithromycin are the first-line therapy for NTM infection due to their direct antimicobacterial action [102]. Resistance to macrolides by ubiquitous microorganisms such as staphylococci, streptococci and Haemophilus can also be found in patients with community-acquired pneumonia [103]. Furthermore, the intake of antibiotics can change the composition of the microbiome. Indeed, Wang et al. [104] demonstrated a higher prevalence of gram-negative organisms in patients who have had a COPD exacerbation, with a subsequent shift towards a predominance of gram-positive organisms after antibiotic administration. Finally, long-term administration of antibiotics could potentially induce substantial changes in the respiratory microbiome [104]. This was further confirmed by Rogers et al. [105]. Indeed, in a post hoc analysis of the BLESS study, the authors analyzed changes in microbiome composition after 48 weeks of erythromycin administration. The authors demonstrated that long-term treatment with erythromycin increased the positivity rate of P. aeruginosa in patients initially colonized by microorganisms other than P. aeruginosa. In contrast, patients who were already colonized by P. aeruginosa did not have any changes in the respiratory microbiome. Hence, the choice of long-term treatment with erythromycin in patients with bronchiectasis not colonized by P. aeruginosa needs to be carefully considered [105].

Lower risks of MDR and systemic adverse reactions frequently favored nebulized antibiotics over systemic antibiotics in CF and NCFB patients.

The higher drug concentrations in the airways, which are reached by inhaled antibiotics, compared with systemic administration, and the minimal systemic absorption of the drug through the alveolar-capillary barrier support this therapeutic approach. Furthermore, the administration of inhaled antibiotics may help to contain the increase of antimicrobial resistance in bronchiectasis patients with chronic infections.

The AIR-BX1 and AIR-BX2 trials demonstrated that tobramycin-resistant P. aeruginosa strains developed in 11% of tobramycin-treated patients compared with 3% of placebo-treated patients (p = 0.36) [91]. However, in a double blind, placebo-controlled crossover study focusing on the safety and efficacy of tobramycin, Drobnic et al. found no differences in antibiotic resistance [106]. More recently, the ORBIT-3 and ORBIT-4 studies, which focused on the safety and efficacy of liposomal ciprofloxacin, did not document any significant reduction in antibiotic activity over 48 weeks in over 1000 patients enrolled worldwide [89]. The RESPIRE 1 and RESPIRE 2 studies, which involved the use of dry powder ciprofloxacin, documented a significant increase in the MIC of the pathogens in both treatment arms (14 days on/off and 28 days on/off). In the RESPIRE 1 trial, 24.5% of patients had a pathogen with an elevated MIC for ciprofloxacin at baseline. The number of patients with elevated MIC pathogens was 20.4%, 26.2%, and 12.3% for ciprofloxacin DPI 14 days on/off, ciprofloxacin DPI 28 days on/off, and pooled placebo, respectively. In the RESPIRE 2 trial, 18.8% of patients had a pathogen with an elevated minimal MIC for ciprofloxacin at baseline. The number of patients with one isolate from sputum with an elevated MIC from pre-treatment at any time point during the study was 21.0% for ciprofloxacin DPI 14 days on/off, 16.5% for ciprofloxacin DPI 28 days on/off, and 9.8% for pooled placebo [87,88]. The AIR-BX 1 and AIR-BX 2 studies, focusing on the use of nebulized aztreonam, described an increase in MIC in 15–23% of enrolled patients [91]. Finally, Haworth et al. found no significant increase in resistant P. aeruginosa strains in a study on inhaled colistin [107].

To date, studies exploring the use of gentamicin, colistin, and tobramycin have not demonstrated a significant emergence of antimicrobial resistant isolates in sputum. In addition, any increase in MIC was transient with return to baseline after discontinuation of treatment. Finally, few data support a possible causal link between the reduction of the bacterial load or the apparent eradication of the dominant pathogen and the establishment of treatment-emergent pathogens. Reasons that may explain the absence of resistance development are diverse and multifactorial. First, it is likely that traditional MIC, determined in combination with parenteral breakpoints, is not applicable to inhaled antibiotics, because there would be significantly increased concentrations in the sputum to allow for safe parenteral administration without toxic effects. Further research is needed to define adjusted airway MICs, which more directly reflect concentrations achieved by inhaled antibiotics. Furthermore, a transient increase in MIC, especially if clinically insignificant, could be abundantly offset by the benefits of reducing exacerbations and therefore by exposure to systemic antimicrobial therapies.

Indeed, unlike the significant antimicrobial resistance observed with macrolide therapy, the administration of inhaled antibiotics resulted in only a modest increase in resistant strains in a variable proportion of patients based on study drug and formulation (solution for inhalation, dry or liposomal powder).

Ultimately, multiple factors must be considered when choosing between oral and inhaled antibiotics: in addition to the risk of antimicrobial resistance, patient characteristics, comorbidities, concomitant drugs, and expected benefits in terms of exacerbations and QoL must be considered.