Physiological tests of small airways function in diagnosing asthma: a systematic review

Background Asthma is a common, heterogeneous disease that is characterised by chronic airway inflammation and variable expiratory airflow limitation. Current guidelines use spirometric measures for asthma assessment. This systematic review aimed to assess whether the most commonly reported tests of small airways function could contribute to the diagnosis of asthma. Methods Standard systematic review methodology was used, and a range of electronic databases was searched (Embase, MEDLINE, CINAHL, CENTRAL, Web of Science, DARE). Studies that included physiological tests of small airways function to diagnose asthma in adults were included, with no restrictions on language or date. The risk of bias and quality assessment tools used were Agency for Healthcare Research and Quality tool for cross-sectional studies and Quality Assessment of Diagnostic Accuracy Studies 2 for diagnostic test accuracy (DTA) studies. Results 7072 studies were identified and 10 studies met review criteria. 7 included oscillation techniques and 5 included maximal mid-expiratory flow (MMEF). Studies were small and of variable quality. In oscillometry, total resistance (R5) and reactance at 5 Hz (X5) was altered in asthma compared with healthy controls. The percentage predicted of MMEF was lower in patients with asthma compared with controls in all studies and lower than the % predicted forced expiratory volume in 1 s. In DTA of oscillometry, R5 showed a sensitivity between 69% and 72% and specificity between 61% and 86%. Conclusion There were differences in the results of physiological tests of small airway function in patients with asthma compared with controls. However, studies are small and heterogeneous. Further studies are needed to assess the effectiveness of these tests on a larger scale, including studies to determine which test methodology is the most useful in asthma.

ABSTRACT Background Asthma is a common, heterogeneous disease that is characterised by chronic airway inflammation and variable expiratory airflow limitation. Current guidelines use spirometric measures for asthma assessment. This systematic review aimed to assess whether the most commonly reported tests of small airways function could contribute to the diagnosis of asthma. Methods Standard systematic review methodology was used, and a range of electronic databases was searched (Embase, MEDLINE, CINAHL, CENTRAL, Web of Science, DARE). Studies that included physiological tests of small airways function to diagnose asthma in adults were included, with no restrictions on language or date. The risk of bias and quality assessment tools used were Agency for Healthcare Research and Quality tool for cross-sectional studies and Quality Assessment of Diagnostic Accuracy Studies 2 for diagnostic test accuracy (DTA) studies. Results 7072 studies were identified and 10 studies met review criteria. 7 included oscillation techniques and 5 included maximal mid-expiratory flow (MMEF). Studies were small and of variable quality. In oscillometry, total resistance (R5) and reactance at 5 Hz (X5) was altered in asthma compared with healthy controls. The percentage predicted of MMEF was lower in patients with asthma compared with controls in all studies and lower than the % predicted forced expiratory volume in 1 s. In DTA of oscillometry, R5 showed a sensitivity between 69% and 72% and specificity between 61% and 86%. Conclusion There were differences in the results of physiological tests of small airway function in patients with asthma compared with controls. However, studies are small and heterogeneous. Further studies are needed to assess the effectiveness of these tests on a larger scale, including studies to determine which test methodology is the most useful in asthma.

BACKGROUND
Asthma is a common but heterogeneous disease characterised by chronic airway inflammation and clinically defined by the presence of respiratory symptoms that vary over time and in intensity. Physiologically, asthma is characterised by variable expiratory airflow limitation which may become persistent over time. 1 Symptoms and airflow limitation can be extremely variable, including the age of onset, triggers for symptoms, the decline in lung function and therapeutic response.
It is estimated that 339 million people are affected by asthma globally 1 but diagnosing asthma is often challenging as there is no gold standard test. This has led to a high burden of undiagnosed disease, especially in children and older adults. 2 3 According to current guidelines, 1 4 a diagnosis of asthma should be objectively supported with an assessment of forced expiratory volume in 1 s (FEV 1 ) reversibility. However, some patients with asthma have no evidence of reversibility or airflow obstruction 5 6 and airflow obstruction and reversibility are seen in patients with alternative diagnoses such as chronic obstructive pulmonary disease (COPD). 7 8 Furthermore, the forced manoeuvres required for spirometry requires effort and coordination, which can be difficult for some individuals. 9 In the past, asthma was thought to only affect larger airways 10 but current evidence suggests that small airways (defined as airways of ≤2 mm in diameter) are affected as well. The small airways may form a site of active disease, even in the absence of airflow Key messages ► Is there evidence to support the use of physiological tests of small airways function in the diagnosis of asthma? ► There is evidence of small airways dysfunction in asthma, which some physiological tests can identify. However, studies are small and heterogeneous and more studies are needed to understand the clinical utility of these tests. ► This systematic review provides a summary of the current evidence around physiological tests of small airways and asthma. It includes recommendations for the future work required to before the adoption of physiological small airways tests in the diagnosis of asthma. copyright.

Open access
obstruction. 11 If the small airways are the first to be affected in asthma, identifying small airways dysfunction (SAD) may help identify asthma earlier, enabling treatment. However, there are a large number of tests that report small airways function. Some of these are being used as secondary outcomes in experimental studies of asthma, to determine asthma phenotype and assess the response to new therapies. 12 The evidence to support the use of any physiological test of small airways function in the diagnosis of asthma is unclear. This systematic review aimed to assess the evidence to support the use of commonly reported physiological tests of small airways function to diagnose asthma in adults, and assess if the selected tests should be included in future clinical studies of the disease.

METHODS
The protocol was prospectively registered in the international registry of systematic reviews (PROSPERO) with registration number CRD42019133239. The review was prepared in accordance to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines 13 and the PRISMA checklist is provided in online supplemental material file S1. Meta-analysis was considered where homogenous results were provided, otherwise data were pooled for graphical presentational purposes.
Through both scoping searches and discussion with experts, the following test were selected to be included in the search, forced oscillation technique (FOT), impulse oscillometry (IOS) and maximal mid-expiratory flow (MMEF) also known as forced expiratory flow between 25% and 75% of forced vital capacity (FVC) (FEF 25%-75% ), and multiple breath washout test (MBW). These tests were selected as they represented some of the most commonly reported physiological tests of small airways function in obstructive lung disease in adults. Online supplemental figure S2 shows the Population, Intervention, Comparison and Outcome (PICO) chart with the studies selection criteria.

Eligibility criteria
Studies were considered for inclusion if they used one of the proposed physiological small airways function tests (FOT/IOS, MBW, MMEF) in diagnosing asthma in adults aged >18 years old. Patients with either a physician diagnosis or a suspected diagnosis of asthma were considered for inclusion. FEV 1 was used as the comparator as it is the current standard in physiological airway assessment. Studies were excluded if they included only children (<18 years), patients with respiratory infections within 2 months of the assessment, did not assess FEV 1 , included patients with asthma-COPD overlap, were laboratorybased studies, animal-based studies or case series of less than 10 participants. There were no language or publication date restrictions.
Search queries were carried out in May 2019 (and the detailed search strategy is found in online supplemental material file S3) on the following electronic databases: Embase, MEDLINE, CINAHL, Cochrane Central Register of Controlled Trials (CENTRAL), Web of Science (Abstracts and Proceedings) up to 5 years and DARE database for grey literature. Clinicaltrials. gov and EudraCT were also searched for active trials or published data. Hand searching of references listed in the selected articles was included. Search terms contained subject heading and terms for the selected test (IOS/FOT, MBW and MMEF) combined with terms of asthma and small airways function.

Study selection
Search results were imported into EndNote 9.1 (Clarivate Analytics) where duplicates were removed and data was uploaded to Rayyan 14 (a webapp tool used for screening titles and abstracts). Abstracts were screened blindly and independently by the authors MA and NYA using the predefined inclusion and exclusion criteria. Disagreements were resolved by discussion, otherwise by the third reviewer, whose initials were RGE. Full-text articles were acquired and imported into EndNote 9.1 by author MA and similar abstract screening methodology was used in screening full texts for eligibility.

Data extraction
Data were extracted by author MA and checked by author NYA for consistency and accuracy using a custom, piloted data extraction form. Diagnostic criteria used to identify asthma, tests used to aid the diagnosis such as airway reversibility, asthma severity, phenotype, medications, the device used and comorbidities were extracted to aid narrative review and provide clinical context. Studies were categorised based on the small airways test used. In diagnostic test accuracy (DTA) studies, sensitivity and specificity values were extracted and a 2×2 contingency table was calculated.
Quality and risk of bias assessment Quality and risk of bias were assessed using validated tools based on study design. Cross-sectional studies were assessed using the Agency for Healthcare Research and Quality (AHRQ) checklist tool. 15 The Quality Assessment of Diagnostic Accuracy Studies 2 16 (QUADAS-2) was used in DTA. The QUADAS-2 tool assesses the risk of bias of studies over four domains: flow and timing, reference standard, index test and patient selection. The tool also assesses for applicability concerns under three domains: reference standard, index test and patient selection.

Descriptions of the tests of small airways function included in the reported studies
Here, only tests included in the analysed studies are described.

Open access
Oscillometry Oscillometry can be assessed using either the FOT or IOS. Oscillometry transmits oscillating sound signals of various frequencies along the bronchial tree, providing a measure of the total airway resistance (resistance at 5 Hz (R5)) and the proximal airway resistance (resistance at 20 Hz (R20)), which allows for the derivation of small airways resistance (R5-R20). Reactance at 5 Hz (X5) relates to physical properties of the lung parenchyma and its ability to expand and facilitate alveolar filling. Resonant frequency (Fres) is the point at which reactance is zero (when forces of inertia and capacitance are equal). The area of reactance (AX) is the sum of area under the reactance curve between X5 and Fres. 17 18 Limitations with this technique include the lack of universal normal ranges for all populations and variance of results between different devices, which can impede interpretation. 19 Maximum mid-expiratory flow The MMEF is the mean forced expiratory flow between 25% and 75% of the FVC (FEF 25%-75% ) and is taken from the spirometric blow with the largest sum of FEV1 and FVC. The MMEF is highly dependent on the validity of the FVC measurement and the level of expiratory effort. 20 21 MMEF is commonly reported in studies of small airways as it is readily accessible from spirometry reports.

Patient and public involvement
Due to the nature of the study design, patients and public were not involved in this systematic review.

Study selection
Initial searches identified 7072 abstracts. After the removal of duplicates, 5764 abstracts were screened of which 469 abstracts included for full text screening. Ten articles ultimately met the inclusion criteria ( figure 1 shows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram). Articles excluded in the full-text screening phase are described in online supplemental material file S4 with reasons given. All included studies were cross-sectional in design and 3/10 of the included studies were DTA studies.

Study characteristics
Seven of the included articles reported oscillometry (IOS/FOT) [22][23][24][25][26][27][28] and five reported MMEF. 23 29-31 None of the included studies reported MBW use in diagnosing asthma. Seven studies were not designed as DTA studies and are presented in table 1. Three studies were designed as DTA studies and these are presented in table 2. The diagnostic criteria used to confirm a diagnosis of asthma  Open access differed among studies. Four studies used Global Initiative for Asthma guidelines, 24 25 28 30 one American Thoracic Society guidelines 27 and one the global strategy: Joint Report of the National Institute for Heart, Lungs and Blood and WHO. 29 Three studies recruited patients based on a previous diagnosis of asthma, without reporting the diagnosis criteria used. 22 23 26 One study reported that patients with symptoms of asthma were included without any formal diagnosis. 31 All included studies were based in different countries (the USA, UK, Japan, Korea, Turkey, Egypt, Russia, Serbia, Iran and China) from four different continents (North America, Europe, Asia and Africa) making the ethnicity of participants heterogeneous. Body mass index (BMI) was only reported in three of the included studies. [23][24][25] Meta-analysis of the data were inappropriate due to the variety and scope of methodological design. Where appropriate, data were displayed graphically to aid the representation of results. No MMEF studies explicitly corrected for FVC, which can potentially affect interpretation.

Risk of bias
Two risk of bias and quality assessment tools were used in this systematic review, based on the design of the included studies. Seven studies were assessed using the AHRQ tool for cross-sectional studies 15 (see figure 2A). This highlighted potential methodological issues around subject selection and quality assurance concerns, which may have impacted on the reliability of results and the

Open access
reporting of study follow-up. There was an overall low risk of bias around patient recruitment (including the source of subjects), the inclusion/exclusion criteria and time periods when patients were identified. Response rates and completeness of results were all reported. A summary of all included studies using both tools is available in online supplemental material file S5. Three studies were assessed using QUADAS-2 tool for DTA studies (see figure 2B). One study had a high risk of bias and applicability concern in the patient selection phase. Two studies had a high risk of bias in the index test.

Results of individual studies
Oscillometry Seven studies used oscillometry. One study used FOT, 24 five studies used IOS 22 25-28 and one study did not reported which type of oscillometry was used. 23 Five studies reported R5, [22][23][24][25][26] which represent the total lung resistance. Only one study reported R5-R20. 24 The values of the test were reported in different units with Mori et al 24  Mori et al 24 reported R5, R5-R20, X5 and MMEF in 49 asthmatic patients,13 controls and 51 COPD patients. They described differences in MMEF, R5-R20 and X5 when comparing asthma to control subjects but not R5. In addition, they reported that the coloured threedimensional model provided by the FOT device could differentiate between asthma, COPD and healthy subjects, with a higher resistance and lower reactance observed in asthma. Asthma severity was not reported and 24 of the asthmatic subjects were ex-smokers.
Mendonça et al 23 studied 35 asthmatic and 34 nonasthmatic participants but used a different technique and frequency than the commonly reported value. Oscillometry values were reported in cmH 2 O/L/s. The whole breath resistance at 8 Hz (R8) and the minimum resistance at maximum inhalation R min were both different when comparing the asthmatic (R8=2.91±0.99) and non-asthmatic group (R8=2.21±0.48). In the asthmatic patients, both the MMEF % predicted value was lower (69%±20) than healthy controls (93%±20) and a higher R min was observed. They also conducted a methacholine challenge test (MCT) in all participants and found that (31/35) asthmatic subject and (8/34) non-asthmatic group had a positive result. A subgroup analysis was reported including asthmatics with positive MCT (31/35) and non-asthmatic with a negative MCT (26/34) and similar results were reported to the overall analysis with a higher R min in positive MCT (1.41±0.42) compared with (1.02±0.24) in negative MCT. Moreover, MMEF was lower in the MCT positive group (68%±18) compared with (99%±18) in negative MCT group. The mean FEV 1 was 88%±11 predicted in the asthmatic group and 95%±10 predicted in the non-asthmatic group, both Mousa and Kamal 25 recruited 25 asthmatic patients and 20 healthy controls (with differences in the mean ages of the groups: asthmatic=45 years and the controls=34 years). Mean BMI did not differ between groups. The severity of the asthmatic group was not reported. IOS was used to assess asthma, with X5 and R5 being reported. R5 was reported in % predicted, but the X5 was reported in absolute values, but did not indicate the unit used. X5 and R5 were different between the two groups (asthma: mean X5 −2.87±1.84 and R5% 245.24±109.18. Healthy controls mean X5 −0.28±0.10 and R5% 109.25±19.40). FEV 1 was lower in the asthmatic group with a mean of 59.68%±23.73 predicted compared with the healthy controls mean of 89.75%±8.70 predicted.
Koruga et al 22 included 31 male military recruits in Serbia with a previous diagnosis of asthma. Histamine was used to assess bronchial hyperreactivity, recording the dose that decreased FEV 1 by 20% predicted value (PD20). X5, R5 and Ax was reported at baseline and after PD20. They found that the overall change in FEV 1 after PD20 was 25 patients with previous diagnosis of asthma and 61 healthy subjects. The asthma group was older (mean age 49 years vs mean age 28 years in the control group). Weight was not reported in either group. Nineteen per cent of the asthma patients were current or previous smokers, but the controls were all never smokers. All inhaled drugs such as short acting beta-agonists and long acting betaagonists were withheld before reversibility testing except inhaled corticosteroids. Asthma severity and comorbidities were not reported. Airways reversibility was assessed using 400 μg of salbutamol via a metered dose inhaler and spacer and reported a mean change of 6.34% of FEV 1 in the asthma group and 2.25% in the healthy controls. The mean percentage of change after administering salbutamol was −33.78±4.43 and −72.93±88.73 in R5 and X5, respectively in the asthma group. In the control group, the mean change was −14.91±2.48 in R5 and 40.09±65.64 in X5.
Maximal mid-expiratory flow Guldent Pasaoglu et al 30 recruited 433 asthmatic patients (mean age 37 years) and 152 patients with COPD (mean age 54 years), aiming to assess differences in clinical and spirometric features of asthma and COPD. 29% of the asthma group and 64% of the COPD group were current smokers. Reversibility was assessed in both groups using 200 μg of salbutamol and defined by an increase of more than 12% and 200 mL of the FEV 1 value. 62.1% of the asthma group met criteria for reversibility compared with 39.5% in the COPD group. MMEF was the only parameter that was below the normal range in non-smoking asthmatic patients with normal auscultation, suggesting that MMEF was a physiological marker of asthma in Open access non-smoking asymptomatic patients. Although bronchodilator responses were measured, these were not reported.
Son et al 31 conducted a retrospective study of 125 patients with a clinical suspicion of asthma who had undergone an MCT. Patients were stratified into three groups based on their FEV 1 and MMEF response to MCT. The positive response to MCT was considered if there was a decline of 20% in FEV 1 and for MMEF, as well. Group 1 included patients with negative MCT tests for both parameters. Group 2 included patients with a negative FEV 1 and a positive MMEF. Group 3 included patients with positive test to both parameters. The mean ages of the included subjects were 45 years in group 1, 39 years in group 2 and 43 years in group 3. Eight subjects had a previous diagnosis of asthma, three in group 1 and five in group 3. Allergic rhinitis was reported in 34 subjects, 16 of them were positive to MCT in both spirometric indices, therefore included in group 3. Mean baseline MMEF in groups 1 and 2 was 97.67%±3.48 predicted and 95.08%±5.74 predicted, respectively. In group 3, mean MMEF was 70.16%±4.64 predicted. The authors suggested that MMEF may be a more sensitive marker of asthma than FEV 1 in patients with otherwise normal spirometry results. Li et al did not report the FEV 1 results, although they performed bronchodilator response tests on all participants. Figure 3A shows the pooled data of R5 of the two studies.

Maximal mid-expiratory flow
There was only one DTA study which used MMEF in asthma compared with controls (Yartsev). 29 The asthma group was older than the control group. Both the asthma and control groups had a majority of female participants. Asthma severity and comorbidities of participants were not reported. The author stratified asthma patients into three groups based on the baseline FEV 1 . Group 1 included participants with FEV 1 of >80% predicted value, group 2 with FEV 1 60%-80% predicted and group 3 with FEV 1 25%-60% predicted. In MMEF tests, the cutoff used was 90% predicted in group 1, 70% predicted in group 2 and 50% predicted in group 3. The DTA of MMEF in group 1 was a sensitivity of 66% and specificity of 91%. Identical results were found in groups 2 and 3 with a sensitivity of 99% and specificity of 100%. The accuracy of MMEF was assessed on all groups with a cutoff value of 70% showed a sensitivity of 88% and specificity 97%. Using FEV 1 , cut-off was set at 120% predicted in group 1, 90% predicted in group 2, 70% predicted in group 3. The DTA of FEV 1 in group 1 was a sensitivity of 77% and specificity of 65%. In groups 2 and 3, identical sensitivity of 100% and specificity of 100% was reported. All groups were assessed for accuracy using FEV 1 , with a 70% predicted cut-off, showing a sensitivity of 92% and specificity of 88%. DTA data were pooled into the forest plot shown in figure 3B.

Synthesis of results
Small airways function in asthma were found to be different when compared with healthy controls. The % predicted MMEF value appeared consistently lower than the % predicted FEV 1 , as shown in figure 4. In oscillometry, R5 was also found to be consistently higher in asthmatic when compared with healthy controls as shown in figure 5. These results highlight the presence of small airways limitation in asthmatic patients with

Risk of bias across studies
There were some concerns of bias in regard to reporting outcomes. Kamal and Mousa 25 reported R5 as the % predicted value while X5 was reported without a unit of measurement. Iartsev 29 did not report how the cut-offs were determined or how subjects were recruited.

DISCUSSION
To the authors' knowledge, this is the first systematic review to assess the use of physiological tests of small airways function in the diagnosis of asthma. Previous work has suggested that SAD is associated with asthma and that the prevalence of SAD increases with the severity of asthma. 32 33 This review suggests that most published studies of small airways function tests in asthma are heterogeneous, of varying methodological quality and have primarily identified SAD rather than using measures of small airways to diagnose asthma. No studies reported the severity of asthma in the participants and participants groups were often poorly matched or characterised in terms of other comorbidities and weight.
This review focuses on MMEF and oscillometry and does not explore all potential measures to assess small airways function. MMEF and oscillometry were chosen as these represented the most commonly cited small airways measures. The clinical utility of oscillometry techniques has been described in asthma and other lung conditions such as interstitial lung diseases and COPD. 19 Oscillometry has been suggested as a useful tool in diagnosing asthma in children. 34 However, there remains a lack of universal reference ranges, especially in adults. Height 23 and sex 35   Open access conducted a multicentre study on healthy subjects in an effort to produce reference ranges for oscillometry in adults, but only one ethnicity was studied. Another study was also conducted in Japan to establish reference ranges for Japanese adults. 36 Understanding and interpreting oscillometry remains challenging. In this review, it was unclear if oscillometry studies provided the most robust measure of small airways function. The R5-R20 (often referred to as resistance of the small airways) was only reported by Mori et al. 24 Airways reversibility, a hallmark of asthma, was only assessed using oscillometry by Nair et al 26 and, here, the mean percentage change in the FEV 1 in the asthma group was 6.34%, which is less than the standard reversibility change of 12%.
The MMEF is an effort-dependent test and guidelines for reproducibility of the manoeuvre is based on FVC and FEV 1 . 37 In all the included articles that studies MMEF, the % predicted of MMEF was found to be lower in asthmatic groups compared with control groups. Moreover, the % predicted value of MMEF was lower than the % predicted FEV 1 in the asthmatic group, suggesting that small airways limitation might be an early marker of airways obstruction. The potential utility of MMEF in early disease was described in one study of patients with alpha-1 antitrypsin deficiency, where an MMEF less than 80% predicted, with a normal FEV 1 /FVC ratio, was associated with increased respiratory symptoms and a faster decline in FEV 1 compared with those with an MMEF of 80% or greater and normal spirometry, suggesting a role for MMEF in early disease monitoring. 38 There are significant limitations to the evidence base described in this review including study heterogeneity, poor patient characterisation and differences in reported values. Not all tests of small airways function have been assessed in asthma (eg, MBW). There are no universally accepted predicted values for oscillometry, especially in adults, making the interpretation of the results more difficult. Oscillation techniques produce many parameters in both inspiratory and expiratory phases and the differences in reported values limits comparisons between studies. MMEF was not corrected for FVC in any study, and this is a limitation as MMEF is a timed/flow measurement and FVC exhalation curve changes may affect the results. 39 Nevertheless, most studies provide at least some signal of SAD in asthma suggesting these indices could be helpful in diagnosing and monitoring asthma. To take this field forward, further research is needed. This should include standardising the assessment of small airways tests (although different tests may have greater or lesser utility in different diseases) and forming normal reference ranges to aid interpretation. Studies in asthma need to predefine how asthma was diagnosed, and report clearly which small airways tests have been measured, by what device, what units are reported and what would be considered an abnormal result or clinically meaningful change in a specified value.

CONCLUSION
Physiological tests of small airways function are feasible in diagnosing asthma and have been shown to be altered in asthma when compared with healthy adults. However, a lack of robust reference ranges and the heterogeneity of approach complicate their use.
Further studies are needed to assess small airways function in asthma, especially in early disease. Larger studies are needed to assess the impact of demographic characteristics and comorbidities such as obesity or allergic rhinitis. This systematic review of current literature suggests these tests may have promise as part of the future diagnostic criteria of asthma, but more work is needed before they can be embedded into clinical care.
Correction notice The license type of the paper has changed from CC BY-NC to CC BY.
Contributors MA and ES conceived, planned and analysed the data, and made a major contribution in writing the manuscript. MA, NYA and RGE performed abstract screening, full-text screening and quality assessment. MA and RGE planned and performed searching strategy and data synthesis. RGE and JS contributed to data analysis and writing the manuscript. All authors have read and approved the manuscript.
Funding This systematic review was part of a funded PhD by King Saud University, Riyadh, Saudi Arabia under the aegis of the Government of Saudi Arabia.
Competing interests ES reports grants from MRC, grants from Wellcome Trust, grants from NIHR, grants from British Lung Foundation, grants from HDR-UK, outside the submitted work. RGE reports grants from NIHR, grants from Chest Foundation, grants from Alpha 1 Foundation, outside the submitted work. All other authors report no conflict of interest.

Patient consent for publication Not required.
Data availability statement All data relevant to the study are included in the manuscript or uploaded as supplementary information. All the included data has been obtained from the included peer-reviewed articles.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
Open access This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https:// creativecommons. org/ licenses/ by/ 4. 0/.

4
Objectives 4 Provide an explicit statement of questions being addressed with reference to participants, interventions, comparisons, outcomes, and study design (PICOS).

METHODS
Protocol and registration 5 Indicate if a review protocol exists, if and where it can be accessed (e.g., Web address), and, if available, provide registration information including registration number.

5
Eligibility criteria 6 Specify study characteristics (e.g., PICOS, length of follow-up) and report characteristics (e.g., years considered, language, publication status) used as criteria for eligibility, giving rationale.

6
Information sources 7 Describe all information sources (e.g., databases with dates of coverage, contact with study authors to identify additional studies) in the search and date last searched.

6
Search 8 Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated.

S.3
Study selection 9 State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if applicable, included in the meta-analysis).

6
Data collection process 10 Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes for obtaining and confirming data from investigators.

6
Data items 11 List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and simplifications made.

Risk of bias in individual studies
12 Describe methods used for assessing risk of bias of individual studies (including specification of whether this was done at the study or outcome level), and how this information is to be used in any data synthesis.

7
Additional analyses 16 Describe methods of additional analyses (e.g., sensitivity or subgroup analyses, meta-regression), if done, indicating which were pre-specified.

Study selection
17 Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally with a flow diagram.

8
Study characteristics 18 For each study, present characteristics for which data were extracted (e.g., study size, PICOS, follow-up period) and provide the citations.

8
Risk of bias within studies 19 Present data on risk of bias of each study and, if available, any outcome level assessment (see item 12).

9
Results of individual studies 20 For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for each intervention group (b) effect estimates and confidence intervals, ideally with a forest plot.  14 Additional analysis 23 Give results of additional analyses, if done (e.g., sensitivity or subgroup analyses, meta-regression [see Item 16]).

Summary of evidence 24
Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (e.g., healthcare providers, users, and policy makers).