Here we describe the technique of sputum induction. This protocol also explains the sputum processing to perform a differential cell count and to collect sputum supernatant and cells for further analysis.
The technique of sputum induction and processing is a recognized non-invasive method allowing the collection and analysis of cells from the airways, which is interesting in various respiratory diseases like asthma, chronic obstructive pulmonary disease (COPD), chronic cough, or idiopathic pulmonary fibrosis. This technique is well tolerated, safe and non-invasive, but is currently limited to research services and specialized centers in clinical practice because it is technically demanding, time-consuming, and requires trained staff. The success rate of sputum induction and analysis is about 80%.
Here, we describe the induction and laboratory processing of sputum samples. Sputum is induced by inhalation of hypertonic or isotonic saline with salbutamol. For the processing, we use the whole sputum technique. Dithiothreitol (DTT) is used to allow mucolysis of sputum samples. The primary aim of sputum processing is to obtain a differential cell count to study the cell types present in the airway lumen. Additional analyses may also be performed on sputum supernatant and sputum cells, which may allow further investigation into inflammatory processes and immune mechanisms. Examples include studying mediators in sputum supernatant and performing a large spectrum of analysis on sputum cells such as flow cytometry, genomics, or proteomics.
Finally, representative results of sputum analysis in healthy controls, asthmatics, and COPD patients are presented.
Several methods are used to investigate airway inflammation: direct measurements (like bronchial biopsies or bronchoalveolar lavages) and indirect methods (like symptom assessment, blood sample analysis and lung function tests)1. The direct techniques have the advantage of reliably assessing the airway inflammation, but they are invasive and not feasible at large scale because of patient discomfort and the risk incurred1. As for the indirect methods, they poorly correlate with the direct assessment of airway inflammation1.
Sputum collection is another way to sample cells from the airways and allows the direct assessment of airway inflammation. Nevertheless, producing sputum spontaneously may lead to samples of poor quality and is not possible for all patients2. This issue has been overcome by using ultrasonically nebulized hypertonic saline to induce the sputum production2. This method was initially used in patients infected with human immunodeficiency virus (HIV) for the diagnosis of Pneumocystis carinii pneumonia3 and was adapted for healthy subjects and asthmatics in 19924. Sputum induction is also feasible in more severe patients by using isotonic saline5. Although generally well-tolerated, the inhalation of saline may cause bronchospasm in patients with hyperresponsive airways5. Therefore, it is recommended to administer a short-acting beta agonist before the procedure1. Moreover, we have previously shown that adding salbutamol to the saline solution in the ultrasonic nebulizer further decreased this risk6. The advantages of sputum induction is that it is non-invasive2 and safe when appropriate precautions are taken5.
For the processing of sputum samples, two methods are currently used in the literature: the whole sputum technique and the plug selection7. In our laboratory, the whole sputum technique is performed. The primary aim of sputum processing is to obtain a differential cell count to study the type of inflammation present in the airway lumen. However, many additional analyses are also possible to further investigate inflammatory processes and immune mechanisms, either by studying mediators in the sputum supernatant8 or by performing detailed investigation on sputum cells (e.g., flow cytometry9, cell cultures10, genomics10, proteomics10, immunocytochemistry7, in situ hybridization7, etc.)
The technique of sputum induction and analysis is currently limited to research services and specialized centers in clinical practice because it is technically demanding, time-consuming, and requires trained staff1. Airway inflammation may be investigated with this method for various respiratory diseases like asthma, chronic obstructive pulmonary disease (COPD), chronic cough, or idiopathic pulmonary fibrosis11.
All methods described in this section have been approved by the Ethics Committee of the University Hospital of Liege and all healthy subjects gave written informed consent for participation.
1.Sputum Induction
2. Sputum Processing
NOTE: Process the sputum within 3 hours of sampling for optimum cell viability.
Caution: Wear a lab coat, protective gloves, and goggles.
Hemocytometer
A typical image is shown in Figure 1 that is visualized with the microscope using the hemocytometer. Squamous cells are easily identified, as they are much larger than the non-squamous cells. These squamous cells are epithelial cells coming from the mouth. Both types of cells are stained by Trypan Blue when dead. Caution must be taken to avoid counting yeast, bacteria, and scrap.
Figure 1: Hemocytometer picture showing (A) a living non-squamous cell, (B) a dead non-squamous cell, and (C) a squamous cell. Scale bar = 50 µm. Please click here to view a larger version of this figure.
Cytospin slide: Figure 2 shows a representative image of a cytospin slide obtained after sputum processing. The different cell types (neutrophils, eosinophils, macrophages, lymphocytes, and epithelial cells) can be differentiated by means of their morphology and coloration. In some cases, contamination by squamous cells may be important and, if the percentage of squamous cells is greater than 80%, the sample is considered unsuccessful (Figure 3).
Figure 2: Cytospin slide picture showing (A) a neutrophil, (B) a macrophage, (C) an eosinophil, (D) a lymphocyte, and (E) an epithelial cell. Scale bar = 20 µm. Please click here to view a larger version of this figure.
Figure 3: Example of a poor quality cytospin slide with >80% of squamous cells. Scale bar = 20 µm. Please click here to view a larger version of this figure.
Success rate
In our department, the success rate of the procedure (combining a successful induction and a readable cytospin), based on a sample of 1,129 patients (healthy subjects, asthmatics, or COPD patients), is 82% (924/1,129). In a sub-analysis according to the type of patients, the success rate is 75% (57/76) in healthy subjects, 82% (827/1,004) in asthmatics, and 82% (40/49) in COPD patients.
Results in healthy subjects
In a retrospective analysis of a series of 289 healthy subjects from our department, the median (interquartile range) sputum weight was 3.72 g (interquartile range of 2.46 g – 5.54 g) and the median total non-squamous cell count/g of sputum was 0.59 x 106 (interquartile range of 0.37 x 106 – 1.29 x 106).
In those healthy subjects, the proportion of squamous cells is low at 19% (10%-34%) and the viability is high at 66% (54-78%). Regarding the percentage of the different cell types, results are summarized in Figure 4A. We can observe that the percentage of macrophages (49% [31-68%]) is higher than the percentage of neutrophils (34% [14%-60%]), while the percentages of lymphocytes (2% [1-3%]), eosinophils (0% [0%-0%]) and epithelial cells (4% [2-11%]) are low. These results are similar when data are expressed in absolute values (Figure 4B).
Figure 4: Representative results of the differential cell count observed in healthy subjects expressed as (A) percentages or (B) absolute values. Results are shown as median (interquartile range). Please click here to view a larger version of this figure.
It is also important to take the age of patients into consideration. Indeed, a strong correlation is present between a patient's age and the proportion of neutrophils in sputum samples (Figure 5). Likewise, when classifying patients according to 10 year age groups (Figure 6), we observed a significant increase in the neutrophil percentage with increasing age group. Therefore, this variable must be considered when comparing results from different cohorts, and caution must be taken in the matching of subjects.
Figure 5: Correlation between age and sputum neutrophil percentage. The correlation was calculated with the Spearman test. Please click here to view a larger version of this figure.
Figure 6: Evolution of the neutrophil percentage according to the age category. The p-value of ANOVA was <0.0001 for the comparison of neutrophil percentage between age classes. Multiple comparisons were made with Dunn's multiple comparisons test. The p-values are represented as follows: * p <0.05, ** p <0.01, and *** p <0.001. Results are shown as median (interquartile range). Please click here to view a larger version of this figure.
Results in patients suffering from respiratory diseases
The technique of induced sputum is commonly used to assess the inflammatory cell profile in asthmatic patients. This technique may also be applied to patients suffering from COPD, another inflammatory respiratory disease. When comparing healthy subjects, asthmatics, and COPD patients (the 3 groups being matched for age, gender, and tobacco habits), we observed that the inflammatory cell profile is quite different between these cohorts (Figure 7). Indeed, asthmatic patients are usually characterized by raised sputum eosinophils, while the proportion of sputum neutrophils is usually higher in COPD patients compared to healthy controls, which is linked to the disease severity.
Figure 7: Sputum inflammatory cell profile of healthy subjects (n = 45), asthmatic patients (n = 108), and COPD patients (n = 54). The three groups were matched for gender, age, and tobacco habits. The p-values of ANOVA were <0.05, <0.0001, and <0.0001 for the comparison of neutrophils, eosinophils, and macrophages between groups, respectively. Multiple comparisons were made with Dunn's multiple comparisons test. The p-values are represented as follows: * p <0.05 and *** p <0.001. Results are shown as median (interquartile range). Please click here to view a larger version of this figure.
The induced sputum method is a useful tool to study the airway compartment. There are several possible applications of this technique. First, it may improve knowledge about immune cells and mechanisms involved in various respiratory diseases. For example, this technique has allowed the investigation of airway inflammation in large cohorts of patients, and it has been shown that about half of asthmatic patients are characterized by an abnormal airway eosinophilic inflammation15,16,17, while COPD patients usually exhibit a raised sputum neutrophil count18. This technique has also contributed to better characterization of airway inflammation in patients with idiopathic pulmonary fibrosis and to evidence mediators that may contribute to this disease19. Second, the technique of induced sputum may be useful to predict response to treatment. For example, the presence of an abnormal percentage of sputum eosinophils has been shown to be a predictive marker of corticosteroid responsiveness11,18. In asthma and COPD, tailoring the dose of corticosteroids to normalize the percentage of sputum eosinophils was shown to be more effective in decreasing the number of exacerbations than adjusting treatment according to the current clinical guidelines11,20,21. Third, sputum analysis may help to develop targeted therapies. For example, the presence of an abnormal number of sputum neutrophils in COPD and some asthmatic patients has led to the development of antineutrophilic treatments11. Fourth, the technique may play a role in diagnosis. For example, the presence of sputum eosinophilia is necessary to make a diagnosis of non-asthmatic eosinophilic bronchitis11.
Besides obtaining a differential cell count, the technique of inducing sputum also allows the performance of many additional analyses, by studying either sputum supernatant or sputum cells. Examples with sputum supernatant include the analysis of mediators8,19,22 and the assessment of the chemotactic activity of the sample for eosinophils22. From sputum cells, RNA can be extracted and used for microRNA or gene expression analyses19,23 . The sputum cells can also be analyzed by flow cytometry9,23 which allows, among others, immunophenotyping and sorting cells. Furthermore, sputum cells can be cultured10, and their mediator production can be measured in vitro24. It is worth noting that in this case, the mediator content is different from what can be found in sputum supernatant. Indeed, mediators in sputum supernatant may be influenced by secretions from airway resident cells and by plasma exudation, contrary to the sputum cell culture model24. Finally, immunocytochemistry and in situ hybridization can also be performed using sputum cells7.
The induced sputum technique has some limitations. The induction must be performed under medical supervision. It is also essential for the operator to give thorough instructions to patients. Other limitations include the cooperation and medical condition of patients, as respiratory efforts are needed. Regarding the feasibility of this technique in children, several studies have reported that it was successful and safe in children older than 6 years old25. Data on children of <6 years of age are lacking25, but it is likely that sputum induction is difficult to perform in those children14. The technique is currently restricted to research services and specialized centers because it is technically demanding, time-consuming, and it requires trained staff1. Another limitation is that the technique of sputum induction and analysis is not always successful, meaning that a readable cytospin is not always obtained26. However, the success rate is usually around 80% in different cohorts of patients4,26,27,28,29. Finally, caution must be taken when comparing different cohorts of patients regarding the matching of age as it was shown to impact the sputum cell count30,31. Likewise, other parameters such as gender and tobacco habits have to be matched as they may also interfere with the sputum cell count29,32. When patient matching is not possible, another way to take age, gender or smoking status into account is to adjust the statistical analysis for these characteristics.
Compared to other techniques that allow collecting cells from airways, like biopsies and bronchoalveolar lavages, the induced sputum method has the advantages of being simple, well-tolerated, safe, reproducible, cost-effective, and non-invasive. These advantages allow the induced sputum technique to be performed at a large scale and repeatedly over time, and make it an alternative of choice for airway sampling. Bronchosorption is another technique allowing the collection of the mucosal lining fluid in order to assess airway mediators33. While this technique (requiring bronchoscopy) is more invasive than induced sputum, it has the advantages of avoiding any contamination with saliva and obtaining higher concentrations of mediators than in bronchoalveolar lavage33.
Critical steps need attention in the protocol. First, caution must be taken concerning the saline concentration and the time of induction, because these two parameters can influence the results and therefore have to be standardized34,35. Second, the mucolysis with DTT is an important step of the processing. As compared with PBS, DTT was shown to better disperse sputum cells7, which improves the cytopsin slide quality and is essential to have a reproducible cell count2. However, the mucolytic agent may interfere with the measurement of biochemical components. Spiking experiments should then be performed when measuring a new biochemical compound36. Finally, another critical step of the procedure is the cell counting step, which must be performed by a trained technician to ensure reliable and interpretable results.
An alternative method using sputum plugs (viscid or denser parts), instead of the whole sputum, is also described in the literature7. Both techniques have advantages and disadvantages. The whole sputum technique allows more rapid processing7, but is frequently contaminated with saliva, which dilutes the sample and can reduce the quality of cytospins2,7. With the plug selection technique, the saliva contamination is reduced, which means that samples are often of better quality for the dosage of fluid phase mediators and for cytospin slides7. The processing is however longer7 and selected plugs may not be representative of the whole sample37. It is also worth noting that not all samples contain plugs, which can be limiting. Both methods are validated and reproducible, and there is currently no evidence that the differential cell counts obtained from both these methods are different14,38.
In the future, induced sputum could be used as a clinical tool to provide biomarkers for the investigation, diagnosis, and management of all kinds of inflammatory respiratory diseases.
The authors have nothing to disclose.
This work was supported by the University Hospital of Liege (CHU Liege). We acknowledge “Instants productions” for the video production. We also acknowledge Cedric François, the patient who appeared in the film.
Spirometer – Spirobank | MIR France | ||
Salbutamol MDI – Ventolin | GLAXOSMITHKLINE | CNK: 0135-913 | See drug available in the country |
Nebulizer UltraNeb | DeVilbiss Healthcare USA | UltraNeb U3000 | |
Devilbiss UltraNeb Disposable Cups & Lids | DeVilbiss Healthcare USA | 100HD-647 | |
DeVilbiss Bacterial Filter For UltraNeb Nebulisers | DeVilbiss Healthcare USA | 1001005879 | |
One-way Valves | Hudson RCI USA | 41664 | |
One-way Valves | Hudson RCI USA | 41665 | |
Aerosol T-connector | Hudson RCI USA | 41077 | |
Ventilation circuit monobranch corrugated | Int'Air Medical France | TA30A | |
NaCl 5 % | / | / | Produced by our hospital pharmacy |
Mini-Plasco NaCl 0.9% | B.Braun Medical Diegem Belgium | ||
Salbutamol sulfate 5 mg/mL – Ventolin | GLAXOSMITHKLINE | CNK: 0094-987 | See drug available in the country |
Ipratropium bromide 0.25 mg/2 mL – Atrovent | Boehringer Ingelheim Germany | CNK: 1543-305 | See drug available in the country |
Dulbecco's Phosphate-Buffered Saline | LONZA Belgium | BE17512F | |
Sputolysin reagent | Calbiochem | 56000 | |
Trypan blue 0.4% solution 100 mL | LONZA Bio Whittaker Belgium | 17-942E | |
Hemocytometer – Thoma chamber | Labor Optik Lancing UK | 1500000 | |
Hemacolor Stainig set | Merck Belgium | 1116610001 | Alternative product : Diff Quik Staining Set Medion Diagnostic |
Mounting medium – Entellan | Merck Belgium | 107960 | |
Statspin Cytofuge 12 | Beckman Coulter Belgium | X00-003066-001 |