Induced sputum: current progress and prospect

Abstract

Induced sputum is a noninvasive, safe, cost-effective, and reproducible method that is particularly advantageous for individuals who do not naturally produce sputum or provide inadequate samples. Its higher quality than spontaneously produced sputum makes it valuable in various respiratory conditions such as chronic obstructive pulmonary disease, asthma, lung cancer, and infectious respiratory diseases. Despite its potential, the clinical use of induced sputum is limited. In conditions like interstitial lung disease, where patients often have a dry cough, induced sputum shows promise, albeit with limited research. This review emphasizes the usefulness of induced sputum in such conditions, aiming for induced sputum to be established as a routine supplementary examination in clinical settings.

Introduction

Sputum, obtained directly from the airway, is a valuable indicator for diagnosing, assessing severity, exacerbations, and monitoring disease progression through its cellular, protein, and microbiological components [1, 2]. However, individuals without spontaneous sputum production, including healthy subjects, often lack this essential airway sample, requiring invasive procedures like bronchoscopy for sample collection. In 1958, Hylan et al. pioneered using hypertonic saline for sputum induction via nebulization, utilizing sputum cytology for lung tumor screening [3]. Advances in induced sputum (IS) collection techniques have broadened the range of amenable to IS analysis, particularly especially beneficial for patients with chronic obstructive pulmonary disease (COPD) [4] or asthma [5] without spontaneous sputum production. Even in patients with spontaneous sputum, IS exhibits superior cell viability and quality [6]. Despite its non-invasiveness, safety, effectiveness, affordability, and repeatability, IS is still underused, likely due to limited awareness of its advantages [7]. This review aims to delineate the current uses of IS and explore its future potential (Fig. 1), and briefly elaborates on the difficulties and challenges faced in clinical practice. We advocate for routine sputum induction in respiratory patients, especially those lacking spontaneous sputum production.

Fig. 1.

Brief procedure, applicable diseases and advantages of induced sputum

Induced sputum and respiratory conditions

Chronic obstructive pulmonary disease

The 2025 Global Initiative for Chronic Obstructive Lung Disease (GOLD) defines COPD as a heterogeneous lung condition characterized by persistent respiratory symptoms, including dyspnea, cough, expectoration, and exacerbations. These symptoms arise from abnormalities in the airways (bronchitis, bronchiolitis) and/or alveoli (emphysema), resulting in persistent and often progressive airflow obstruction [8]. Alongside local pulmonary manifestations, COPD is linked to systemic chronic inflammation [9]. Local inflammatory markers like interleukin-6 (IL-6) and C-reactive protein (CRP) may manifest earlier in COPD than systemic markers, with IS markers better reflecting airflow limitation severity [10]. The results showed that the application of IS in COPD had great advantages. Presently, induced sputum’s clinical applications in COPD patients primarily revolve around diagnosis (biomarkers), disease status assessment, exacerbation prediction, severity evaluation, and assessment, prediction, and guidance of treatment.

IS levels of matrix metalloproteinase-8 (MMP-8) can differentiate between asymptomatic and symptomatic smokers, aiding in the early detection of COPD [11]. N-acetyl-proline-glycine-proline (N-α-PGP) in sputum, a tripeptide that attracts neutrophils, shows promise as a novel diagnostic biomarker for COPD [12]. Integrating sputum analysis with omics techniques such as proteomics, transcriptomics [13], and lipidomics [14] allows for a comprehensive exploration of COPD biomarkers. The cellular composition of sputum, including macrophages, neutrophils, and T lymphocyte subsets, varies depending on the disease status, as do the soluble and intracellular components in sputum supernatant [15]. Inflammatory markers in sputum, including interleukin-8 (IL-8), myeloperoxidase (MPO) [16], interferon gamma-inducible protein-10 (IP-10), neopterin, tumor necrosis factor (TNF)-α [17], and oxidative stress products like intracellular nitrotyrosine and heme oxygenase (HO)-1 immunopositivity [16], increase during COPD exacerbations. Notably, IS can predict exacerbation frequency or risk. Elevated levels of human cathelicidin (hCAP18/LL-37), an antimicrobial peptide (AMPs), are associated with a higher risk of acute exacerbations in COPD [18]. Assessing lower airway bacterial colonization through IS establishes a link between bacterial presence in stable COPD and exacerbation frequency [19]. In individuals experiencing acute COPD exacerbations, IL-8 and reactive oxygen species in sputum neutrophils can differentiate between bacterial and non-bacterial exacerbations [20], providing essential guidance for treatment.

COPD is a multifaceted syndrome consisting of distinct subtypes, primarily characterized by varying forms of airway inflammation: neutrophilic cell-related COPD and eosinophilic cell-related COPD. Neutrophilic inflammation, marked by an abundance of neutrophils in sputum cells, is the predominant subtype of COPD. Eosinophilic cell-related COPD is diagnosed when sputum eosinophils surpass 2–3% [21]. Patients with COPD can be categorized based on the presence of chronic bronchitis, with those manifesting this symptom displaying higher sputum eosinophil levels compared to those without it [22]. Imaging modalities can differentiate emphysema-type COPD, with sputum vascular endothelial growth factor (VEGF) levels elevated in the chronic bronchitis subtype and decreased in the emphysema subtype [23], suggesting a potential role for VEGF in distinguishing between these subgroups. The severity of COPD can be classified into four grades (GOLD grades 1, 2, 3, and 4, or mild, moderate, severe, and very severe) based on the FEV1% predicted lung function [24]. IS shows a strong correlation with COPD severity. Various cellular components found in sputum, such as neutrophils [25], lymphocyte subsets (CD8-IL4/CD8-IFNγ ratio) [26], and cytotoxic CD56 cells (1414 biological lytic activity) [27], inflammatory mediators including prostaglandin E2 (PGE2) [28], chemokines including C-X-C motif chemokine ligand (CXCL) 9, CXCL10, CXCL11, C–C motif chemokine (CCL) 5, and CCL3 [29, 30], proteases including matrix metalloproteinase-2 (MMP-2) [28], oxidative stress-related substances including peroxynitrite inhibitory activity and 8-iso-PGF2α [31, 32], as well as other molecules like VEGF [23], osteoprotegerin (OPG) [33], fatty acid-binding protein 4 (FABP4) [34], and microRNAs (hsa-let-7c, hsa-miR-34b, hsa-miR-34c, hsa-miR-125a-5p, hsa-miR-30a-3p, and hsa-miR-30e-3p) [35] have demonstrated significant correlations with the severity of COPD. In the assessment, prediction, and guidance of treatment, sputum induction is commonly employed to evaluate local inflammation changes post-treatment. Previous studies have shown that inhaled and oral glucocorticoids can reduce neutrophil counts in sputum [36, 37], with oral glucocorticoids also inhibiting airway eosinophilic inflammation [38]. Sputum induction testing is also utilized to evaluate new drugs, such as CHF6001, a novel inhaled phosphodiesterase-4 inhibitor [39]. Notably, sputum induction can predict treatment outcomes and guide therapy. Sputum eosinophilia is an indicator for predicting the improvement in airflow obstruction following glucocorticoid therapy [40]. Strategies aimed at minimizing eosinophilic airway inflammation have been linked to a decrease in severe exacerbations of COPD [41]. Additionally, Ditz et al. identified a gene signature through transcriptome analysis associated with the time to first exacerbation after discontinuing inhaled corticosteroids, showing higher predictive value than sputum eosinophils [42].

In summary, the extensive use of IS testing in COPD is demonstrated, especially in the notable clinical significance of sputum eosinophil count for distinguishing different types of airway inflammation and informing treatment decisions. It is essential to acknowledge that blood eosinophil count cannot entirely supplant sputum eosinophil count [43]. Additionally, the detection of individuals at elevated risk of recurrent or acute exacerbations through IS facilitates the implementation of focused preventive measures within this subgroup.

Asthma

Asthma, characterized by coughing or chest tightness, is a chronic inflammatory airway disease, and has long been associated with eosinophilic inflammation [44]. In the early twenty-first century, bronchoscopy and small sample studies revealed heterogeneity in asthma airway inflammation [45, 46]. The introduction of IS expanded sample sizes and led to the initial classification of asthma airway inflammation into eosinophilic and non-eosinophilic subtypes [47]. Eosinophils, key players in asthma pathogenesis [48], have been extensively investigated in IS, showing implications for disease severity, activity, inflammation identification, response, and treatment guidance. Variations in sputum eosinophil levels may indicate different asthma severities, with elevated levels suggesting severe asthma possibly due to reduced eosinophil apoptosis [49, 50]. Increased sputum eosinophils also signify poor asthma control, as evidenced by significantly higher levels of poorly controlled asthma compared to controlled asthma [51]. Research by P.G. Gibson et al. demonstrated that in pediatric asthma, exacerbation frequency in the previous 12 months rose with higher sputum eosinophil counts [52]. Furthermore, sputum eosinophil percentage has been utilized as an indicator for assessing the type of asthma airway inflammation, with levels above 2–3% defining eosinophilic asthma [53]. Surprisingly, a large global study uncovered that only approximately half of asthma patients have eosinophilic asthma, a proportion decreasing to around one-third in low to middle-income countries [54]. Regarding assessment and treatment guidance, oral and inhaled corticosteroids can decrease sputum eosinophils, with inhaled steroids exhibiting rapid onset but short duration, suggesting differing effects on airway inflammation based on the administration route [55]. Additionally, sputum eosinophilia is a valuable predictor of response to oral steroid trials in asthma [56], a metric not entirely captured by blood eosinophils [57]. Monitoring eosinophil levels in IS is crucial for determining the appropriate dosage of inhaled steroids to manage inflammation during treatment [58]. Normalization of eosinophil counts in sputum has been shown to reduce asthma exacerbations and hospitalizations effectively [59]. Utilizing eosinophil counts in sputum can also inform decisions regarding medication discontinuation, helping to decrease the reliance on inhaled corticosteroids without increasing the risk of exacerbations [59]. Neutrophils in sputum are commonly assessed alongside eosinophils to differentiate between various types of airway inflammation. Different subtypes of airway inflammation can be identified by evaluating the ratio of eosinophils to neutrophils in sputum, including eosinophilic, neutrophilic, mixed granulocytic, and paucigranulocytic [60]. Elevated levels of neutrophils in sputum indicate severe asthma, with studies demonstrating a significant increase in sputum neutrophils compared to mild cases [61]. Notably, a rise in sputum neutrophils is associated with a diminished response to inhaled corticosteroid therapy [62]. Additionally, gene expression in sputum cells, akin to sputum cell counts, can provide insights into disease activity, differentiate inflammatory subtypes, and predict responses to treatment. Notably, gene expression in sputum cells is more accurate in predicting reactions to oral corticosteroid therapy compared to assessing sputum eosinophils [63–65].

Soluble components found in sputum provide valuable insights for assessing and managing asthma in patients. Levels of IL-8 and MPO in sputum demonstrate fluctuations in response to asthma activity, increasing during exacerbations and decreasing during stable periods [66]. Complement factor H (CFH) in sputum has been associated with asthma severity and control, contrasting with CFH levels in peripheral blood [67]. Previous studies have analyzed 75 proteins in sputum, revealing distinct changes in severe asthma cases [68]. Biomarkers related to airway remodeling, such as matrix metalloproteinase-9 (MMP-9) and tissue inhibitors of matrix metalloproteinases-1 (TIMP-1), can be influenced by inhaled corticosteroid (ICS) therapy to alleviate airway remodeling in asthma patients [69]. The integration of IS analysis with omics approaches allows for a comprehensive exploration of local asthma pathology, incorporating lipidomics, metabolomics, and proteomics [70–72].

Interstitial lung disease

Interstitial lung disease (ILD) comprises over 200 diseases characterized by diffuse pathological changes in the interalveolar interstitial tissue, resulting in nonspecific clinical manifestations, such as dyspnea and dry cough [73]. Diagnosis and evaluation of ILD currently rely heavily on high-resolution computed tomography (HRCT) and various forms of lung biopsy (bronchoscopic, percutaneous, or surgical) [74]. While HRCT is commonly used for diagnosis, its ability to differentiate the heterogeneity and pathology of ILD is limited. Lung biopsies are constrained by invasiveness and a lack of extensive research. IS shows promise as an alternative approach. Despite its widespread use in COPD and asthma, research on IS in ILD is notably deficient. Investigations have revealed significant variations in macrophages and neutrophils in sputum from idiopathic pulmonary fibrosis (IPF) patients, along with correlations between eosinophils and CD4/CD8 + cells [75]. In ILD related to systemic sclerosis (SSc), strength of correlation between IS and BALF was significant for macrophages and neutrophils [76]. Studies in sarcoidosis patients have demonstrated that the CD4/CD8 ratio in sputum correlates with BALF, aiding in the differentiation of sarcoidosis from non-granulomatous ILD, with IS showing superiority in advanced cases [77–79]. It follows that IS and BALF have many similarities, albeit with slight differences.

Previous research has identified a local imbalance of proteases/antiproteases in IPF through IS, specifically an MMP-9/TIMP-1 imbalance [80]. Furthermore, increased expression of insulin-like growth factor binding protein (IGFBP)-2, IL-8, transforming growth factor β (TGF-β), and matrix metalloproteinase-7 (MMP-7) has been confirmed in sputum from IPF patients [81]. A notable reduction in glutathione (GSH) levels in the sputum of individuals with IPF suggests the presence of a localized imbalance between oxidants and antioxidants in the development of IPF [82]. These findings underscore the potential of IS in assessing the local pathological processes in IPF. Limited research has explored IS in connective tissue disease-associated ILD (CTD-ILD). Airway inflammation in rheumatoid arthritis (RA) and systemic sclerosis (SSc)-related ILD is primarily characterized by neutrophilic inflammation [83]. Nevertheless, in SSc-ILD, no significant differences in ILD biomarkers (IGFBP-1, IGFBP-2, IGFBP-3, TGF-β, IL-8, TNF-α, YKL-40, MMP-7, and MMP-9) were observed in sputum between SSc-ILD and SSc-nonILD patients [84]. This could be attributed to small sample sizes or the pulmonary origin of the immune response in SSc. Elevated levels of sputum Krebs von den Lungen-6 antigen (KL-6) and YKL-40 were detected in individuals exhibiting a fibrotic pattern of hypersensitivity pneumonitis (HP) [85]. These investigations offer initial support for the feasibility of IS in CTD-ILD and HP.

Njock MS et al. conducted a study on the microRNA composition of exosomes derived from sputum in individuals diagnosed with IPF. The study identified three specific microRNAs (miR-142-3p, miR-33a-5p, let-7d-5p) that show promise as biomarkers for IPF, particularly noting a correlation between miR-142-3p and lung function parameters [86]. Subsequent investigations revealed that miR-142-3p can be transferred to lung epithelial cells and fibroblasts via exosomes originating from macrophages. This transfer mechanism induces anti-fibrotic effects by targeting transforming growth factor-β receptor 1 (TGFβ-R1) and inhibiting the proliferation of lung fibroblasts [87]. Interestingly, prior research on immune responses in individuals with IPF has indicated that sputum-induced autoantibodies may better reflect lung function parameters [88]. Additionally, multiple studies have detected relevant autoantibodies in the sputum of individuals at elevated risk for RA despite negative serum tests and in RA patients. These findings suggest a local origin of lung antibodies that may subsequently enter the bloodstream [89–91]. This evidence supports the hypothesis that early RA-related autoimmunity involves the lungs, with locally sourced antibodies eventually circulating in the blood. This process may provide insights into the intricate relationship between CTD and ILD.

Lung cancer

IS was initially proposed for cytological diagnosis of lung cancer [3], with a primary focus on the diagnostic aspect of bronchogenic carcinoma. The noninvasive nature of IS cytology makes it particularly suitable for patients suspected of having lung cancer but who do not spontaneously produce sputum, have insufficient sputum production, or have failed or cannot tolerate other diagnostic measures. In such cases, IS can be considered a routine diagnostic investigation [92, 93]. Even when spontaneous sputum is adequate, IS can significantly enhance the diagnostic rate of lung cancer, comparable to bronchoscopy [94]. It can also diagnose bronchogenic carcinoma that is not visible under endoscopy [93]. Moreover, abnormal methylation of tumor suppressor genes (TSGs) in IS has been identified as an early diagnostic biomarker for lung cancer [95].

Infectious respiratory disease

Sputum induction is primarily employed in infectious diseases for diagnostic purposes, specifically for identifying pathogens. It is particularly beneficial for patients with infectious respiratory diseases who do not naturally produce sputum, as traditional sputum cultures are not feasible in such instances. IS and BALF have similar sensitivity and specificity regardless of the test methods. Especially when PCR diagnosis is applied, IS has a negative diagnostic rate of 99% excluding pneumocystis pneumonia [96]. However, in other contexts, sputum induction does not provide substantial diagnostic advantages over spontaneous sputum production [97]. For individuals with pulmonary tuberculosis who do not spontaneously produce sputum, the sensitivity of sputum induction testing three times is similar to bronchoscopy [98, 99], with increased diagnostic accuracy when combined with Xpert MTB/RIF Ultra [100]. Even in cases of pulmonary tuberculosis where patients do produce spontaneous sputum, particularly those with negative sputum smears, sputum induction can enhance diagnostic rates [99]. In community-acquired pneumonia, sputum induction generates high-quality specimens with significant microbiological yield, regardless of spontaneous sputum production [101, 102]. Notably, during the recent global coronavirus disease 2019 (COVID-19) outbreak, sputum induction has demonstrated superior sensitivity and reliability compared to the commonly used throat swab, justifying its recommendation as the standard for discontinuing isolation [103, 104]. It is also expected to provide reliable samples for future pandemics or new infectious diseases.

Limitations

Although induced sputum offers significant advantages in the assessment of respiratory system diseases, several deficiencies remain that require further attention.

1. Lack of unified procedures and standards

   Firstly, there is considerable variability in the concentration of nebulized saline, which ranges from 0.9 to 7%. Although no significant differences have been observed in the cell composition and differential cell count of induced sputum, this inconsistency may still pose challenges. Secondly, discrepancies exist regarding the analyzed components of sputum; specifically, two methods are employed: whole sputum detection and sputum plug detection. These variations can directly influence the results obtained from analyses [105]. Finally, a consistent diagnostic framework for interpreting induced sputum results has yet to be established.

2. Variability of samples

   Induced sputum is believed to sample different regions of the airway at various times—in sequence: central airways, peripheral airways, and alveoli [105]. However, distinguishing between these areas within induced sputum is challenging due to its typically mixed nature. This amalgamation may result in substantial discrepancies during sputum analysis.

3. The impact of dithiothreitol or dithioerythritol on detected substances

   Dithiothreitol and dithioerythritol serves as potent reducing agents utilized during the induced sputum procedure to liquefy samples by breaking disulfide bonds [106]. However, this property may interfere with the quantification and characterization of proteins present in the sample.

4. High time consumption and need for well-trained staff

   The current protocol for obtaining induced sputum is complex and time-consuming; it necessitates skilled personnel who can effectively guide patients through successfully producing high-quality specimens while also monitoring for adverse reactions and managing nebulization cessation when necessary. Such requirements significantly limit its applicability within clinical practice.

Conclusion

IS is a noninvasive, safe, cost-effective, and reproducible method for obtaining localized lung samples for diagnostic purposes. The presence of sputum pathogens, cells, and soluble mediators plays a certain role in diagnosing respiratory conditions, assessing disease severity and activity, differentiating between subtypes, and guiding treatment strategies. We hope that IS can be widely used as a routine supplementary sputum examination in clinical settings in future, especially suitable for patients without spontaneous sputum. Due to the current high time consumption and need for well-trained staff limitations in clinical practice, IS microbiological detection may be more frequently utilized for patients with unclear diagnoses who cannot tolerate invasive examinations, while cytological examinations may be more suitable for the assessment before and after treatment of patients with COPD or asthma. Essentially, while IS holds considerable promise for managing respiratory conditions, it is imperative that the scientific community refines and streamlines its application.

Abbreviations

IS

Induced sputum

COPD

Chronic obstructive pulmonary disease

GOLD

Global Initiative for Chronic Obstructive Lung Disease

IL-6

Interleukin-6

CRP

C-reactive protein

MMP-8

Matrix metalloproteinase-8

N-α-PGP

N-Acetyl-proline-glycine-proline

IL-8

Interleukin-8

MPO

Myeloperoxidase

IP-10

Interferon gamma-inducible protein-10

TNF

Neopterin and tumor necrosis factor

HO

Heme oxygenase

AMPs

Antimicrobial peptides

AECOPD

Acute exacerbations of COPD

ROS

Reactive oxygen species

VEGF

Vascular endothelial growth factor

PGE2

Product prostaglandin E2

CXCL

C-X-C motif chemokine ligand

CCL

C–C motif chemokine

MMP-2

Matrix metalloproteinase-2

8-iso-PGF2α

8-Iso-prostaglandin F2α

OPG

Osteoprotegerin

FABP4

Fatty acid-binding protein 4

ICS

Inhaled corticosteroid

CFH

Complement factor H

MMP-9

Matrix metalloproteinase-9

TIMP

The tissue inhibitors of matrix metalloproteinases

ILD

Interstitial lung disease

HRCT

High-resolution CT

BALF

Bronchoalveolar lavage fluid

IPF

Idiopathic pulmonary fibrosis

SSc

Systemic sclerosis

MMP-9

Metalloproteinase-9

IGFBP

Insulin-like growth factor binding protein

MMP-7

Matrix metalloproteinase-7

TGF-β

Transforming growth factor β

GSH

Glutathione

CTD-ILD

Connective tissue disease-related interstitial lung disease

RA

Rheumatoid arthritis

YKL-40

Chitinase-3-like-protein 1

KL-6

Krebs von den Lungen-6 antigen

HP

Hypersensitivity pneumonitis

TGFβ-R1

Targeting transforming growth factor-β receptor 1

TSGs

Tumor suppressor genes

PCR

Polymerase chain reaction

COVID-19

Novel coronavirus disease 2019

Author contributions

FZ, PG and HW: manuscript draft writing and revision; WW: supervision and revision. FZ, HW, WW, YZ, YM, TW and PG critically reviewed the manuscript for important intellectual content and approved the final version to be published. FZ, HW, WW, YZ, YM, TW and PG agree to be accountable for all aspects of the work, in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Funding

This work was supported by Natural Science Foundation of Henan Province (No. 232300421304), Henan Provincial Science and Technology Research Project (No. 222102310533, 222103810052), Henan Provincial Medical Science and Technology Research Project (No. LHGJ20220671) and Young Elite Scientists Sponsorship Program by Luoyang Association for Science and Technology (No. 2022HLTJ18).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

I am hereby giving my consent for publication.

Competing interests

The authors declare no competing interests.