Bronchiectasis is a chronic lung disease characterised by irreversible dilatation of the airways and constitutes a heterogeneous nosological entity regarding its underlying causes, as a wide spectrum of other diseases eventually leads to its development. It is usually an underdiagnosed and inadequately treated condition with a high burden for both the patients and the healthcare system [1]. Eosinophilia was traditionally considered as a means for distinguishing bronchiectasis from asthma, whereas bronchiectasis was primarily associated with neutrophilic inflammation. Notably, in the 1960s, Strang [2] measured blood eosinophil counts (BEC) in 40 children with bronchiectasis and 43 with asthma and found that only the latter demonstrated peripheral eosinophilia [2].

In most cases of concomitant bronchiectasis and eosinophilia reported in the literature, bronchiectasis appeared as the result of another underlying disease involving eosinophilia. In the 1960s, Rab et al. [3] described two cases of tropical pulmonary eosinophilia that led to bronchiectasis as a direct complication [3]. Several decades later, Price et al. [4] reported the combination of eosinophilia and bronchiectasis in a patient with ulcerative colitis receiving mesalazine [4]. Bronchiectasis was also observed in the cases of two adolescents diagnosed with chronic eosinophilic pneumonia, suggesting bronchiectasis as a possible manifestation of the disease [5, 6]. Traction bronchiectasis was a common finding in four fatal cases of acute eosinophilic pneumonia in a small cohort of 41 patients, relating its presence to poor clinical outcomes [7]. Uncontrolled respiratory symptoms in the presence of bronchiectasis and airway microbial colonisation were also observed in a patient with eosinophilic granulomatosis with polyangiitis despite immunosuppressive therapy [8].

Ongoing research throughout the last three decades has questioned whether neutrophilic inflammation is a hallmark of bronchiectasis and highlighted that eosinophils have also a key role in the pathogenesis of the disease. Subsequently, eosinophilic inflammation in bronchiectasis remains a wide-open field for future research with potential clinical importance, as this distinct disease endotype seems to reflect clinical outcomes, such as clinical manifestations, disease severity, and response to particular treatment modalities, and thus it plays a critical role in decision-making by the physician during the management of the condition in the context of precision medicine.

Evidence of eosinophilic activation in bronchiectasis was reported as early as 1998 by Kroegel et al. [9], who investigated patients with bronchiectasis unrelated to cystic fibrosis (CF) or bronchopulmonary aspergillosis and compared them with healthy individuals and patients with allergic bronchial asthma or chronic obstructive pulmonary disease (COPD). Patients with bronchiectasis did not show increased BEC, in contrast to patients with asthma, but showed increased serum levels of eosinophil cationic protein, further highlighting it as a more sensitive marker to measure local eosinophil involvement [9]. Also, the presence of bronchiectasis in severe asthma was found to be associated with airway eosinophilia, more extensive eosinophil degranulation, and higher mRNA expression of Charcot-Leyden crystal galectin, indicating higher eosinophilic activity in the coexistence of asthma and bronchiectasis [10].

In patients with bronchiectasis with or without chronic rhinosinusitis (CRS), sputum eosinophilia is significantly higher in the presence of CRS and is correlated with BEC, whereas methacholine tests are negative in the vast majority of the patients, identifying bronchiectasis with CRS as an eosinophilic airway disease and distinguishing it from asthma. Eosinophilic bronchiectasis with CRS also encompasses an altered interactome [11]. In bronchiectasis-COPD overlap syndrome, peripheral eosinophilia is associated with higher exacerbation and hospitalisation rates, whereas peripheral eosinopenia (BEC less than 100 cells/μL) is associated with higher infectious outcomes, including pneumonia episodes and chronic airway infection. Furthermore, treatment with inhaled corticosteroids (ICS) in bronchiectasis-COPD overlap syndrome patients with increased BEC leads to improvement of exacerbation rate without any negative impact on infectious events [12]. Figure 1 illustrates the highlight features of the coexistence of eosinophilic bronchiectasis with other respiratory diseases based on the evidence in previous studies.

Nowadays, there is an emerging number of studies pointing towards the identification of eosinophilic bronchiectasis as a distinct disease endotype, demonstrating a constellation of specific clinical, laboratory, pulmonary functional, and chest imaging features, as well as response to specific treatment modalities and prognostic indices. However, although BEC has been associated with several clinical parameters, its baseline value significantly fluctuates in patients with bronchiectasis over time, indicating that its monitoring is crucial for its use in daily clinical practice [13].

A large unicentre retrospective analysis in China showed that eosinophilic bronchiectasis was associated with a stronger inflammatory response, a longer duration of hospital stay, and a higher hospitalisation cost [14]. Also, in a study of 169 patients with non-CF bronchiectasis hospitalised for acute exacerbation, those with eosinophilic bronchiectasis based on BEC demonstrated sinusitis, worse airway obstruction, sputum eosinophilia, higher serum immunoglobulin E, higher BSI (bronchiectasis severity index) and E-FACED (exacerbations, forced expiratory volume in one second, age, chronic colonisation, extension, and dyspnoea) scores, as well as higher hospitalisation cost [15].

Importantly, in another cross-sectional study, the authors defined type 2 inflammation in patients with bronchiectasis as the presence of either BEC ≥ 300 cells/µL or fractional exhaled nitric oxide (FeNO) ≥ 25 ppb. While 88 patients were consistent with this definition, only 19 had both markers increased based on the cutoffs. BSI scores were higher in patients with type 2 inflammation, irrespective of whether both markers or one of them were elevated [16]. On the other hand, a study of 164 patients with non-CF bronchiectasis investigating the utility of type 2 inflammatory markers, i.e. BEC and FeNO, revealed that the subgroup of 41 patients with low BEC and high FeNO demonstrated better pulmonary function, less involvement of lung lobes, lower dyspnoea prevalence, and less disease severity [17]. In another retrospective cohort study of 318 patients with bronchiectasis, the patients with BEC below 250 cells/μL were at increased risk for an exacerbation that led to hospitalisation [18]. However, these observations should be treated with caution by taking into consideration the small numbers of patients in each subgroup, the large range of FeNO levels (from 27 ppb to 43 ppb), and the relatively high BEC used as cutoffs.

In a large Spanish study of 928 patients with bronchiectasis subdivided into four groups based on BEC, i.e. eosinopenic (< 100 cells/μL), low BEC (51–100 cells/μL), normal BEC (101–300 cells/μL), and eosinophilic (> 300 cells/μL) bronchiectasis, BEC, and disease severity, defined by exacerbation and hospitalisation rates, showed a significant U-shaped behavior. More specifically, at the extremes of the curve, eosinopenic and eosinophilic patients presented higher levels of the severity indices, while the patients with low and normal BEC presented lower levels of the severity indices and thus milder disease. Interestingly, when comparing patients receiving ICS with all patients, a response to treatment was only present in eosinophilic patients leading to a decrease in exacerbation and hospitalisation rates. This observation indicates the importance of recognising this subgroup of patients and taking eosinophilia into consideration when deciding the therapeutic approach in a patient with bronchiectasis [19].

In the study by Shoemark et al. [20] the authors extracted data from different bronchiectasis cohorts from five countries to correlate the BEC with the sputum eosinophil counts, suggesting that BEC could be used as a surrogate of lung eosinophil counts in the same way as in asthma and COPD, as blood and sputum eosinophils seemed to correlate in two cohorts. Patients with high BEC tended to have Streptococcus- and Pseudomonas-predominated microbiome profiles. Also, moderately or highly elevated BEC was associated with more frequent exacerbations [20]. Furthermore, stratification of the data from the FRIENDS cohort based on BEC (< 100 cells/μL, 100–299 cells/μL, and > 300 cells/μL) was associated with sputum microbiome, with Haemophilus- and Moraxella-predominant microbiomes observed in patients with BEC < 100 cells/μL and Streptococcus-predominant microbiome in patients with BEC > 300 cells/μL, whereas Pseudomonas airway infection was found in both groups [21].

Findings of most studies collectively indicate that neither too low nor too high BEC is in favour of the patients with bronchiectasis, but a state characterised by normal BEC shows better disease outcomes. Thus, defining the distinct “eosinophilic bronchiectasis” endotype as well as recognising specific therapeutic options beneficial for these patients is of clinical relevance, as described further below. Figure 2 summarises the major differences between bronchiectasis with low- and high-eosinophilic inflammation.

A significant degree of discrepancy between the studies can be observed. This is attributed: i) to the limited available data due to the small number of relevant studies and small sample sizes, ii) the non-harmonised protocols, including the diverse cutoffs of type 2 inflammatory markers, such as BEC and FeNO, between the studies, and iii) the possible confounding factors, including comorbidities or other conditions, such as severe eosinophilic asthma (SEA) and obesity, lifestyle habits, such as smoking, and treatment modalities, such as corticosteroids, affecting the same markers. Large-scale multicenter studies with harmonised definitions and cutoffs of type 2 inflammatory markers are needed to draw robust conclusions.

The way bronchiectasis is viewed by physicians has drastically changed over the last decades, with the topic of eosinophilia in the context of the disease being one of the substantially highlighted. Eosinophilic bronchiectasis is nowadays identified as a distinct disease endotype and is associated with a specific clinical course and response to treatment. Further research is needed to better characterise this endotype and improve its personalised investigation and management in daily clinical practice.