Lung Function in Preschool Children in Low and Middle Income Countries: An Under-Represented Potential Tool to Strengthen Child Health

Background The burden of respiratory disease is high in low-middle income countries (LMIC). Pulmonary function tests are useful as an objective measure of lung health and to track progression. Spirometry is the commonest test, but its use is limited in preschool children. Other lung function methods have been developed but their use in LMIC has not been well described. Aim To review the use of preschool lung function testing in children in LMIC, with particular reference to feasibility and clinical applications. Methods Electronic databases “PubMed”, “Scopus”,” Web of Science”, and “EBSCO host” were searched for publications in low and middle income countries on preschool lung function testing, including spirometry, fractional exhaled nitric oxide (FeNO), oscillometry, interrupter technique, tidal breathing and multiple breath washout (MBW), from 1 January 2011 to 31 January 2022. Papers in English were included and those including only children ≥6 years were excluded. Result A total of 61 papers from LMIC in Asia, South America, Africa, Eurasia or the Middle East were included. Of these, 40 included spirometry, 7 FeNO, 15 oscillometry, 2 interrupter technique, and 2 tidal breathing. The papers covered test feasibility (19/61), clinical application (46/61) or epidemiological studies (13/61). Lung function testing was successful in preschool children from LMIC. Spirometry was the most technically demanding and success gradually increased with age. Conclusion Preschool lung function testing is under-represented in LMIC for the burden of respiratory disease. These tests have the potential to strengthen respiratory care in LMIC, however access needs to be improved.


INTRODUCTION
Childhood respiratory disease is a common cause of morbidity and mortality globally (1). The burden of acute and chronic respiratory disease is especially high in low-middle income countries (LMIC) (2), may result in impaired lung function and set a trajectory for chronic illness into adulthood (3,4). However, access to respiratory diagnostic and management tools such as lung function are limited in many LMIC (5).
Lung function attained in early life is important for respiratory health, with low lung function associated with subsequent risk of respiratory disease (6). Pulmonary function tests are an objective measure of lung health which can be used to diagnose and track lung disease and assess response to treatment. In recent years non-invasive tests have been developed and guidelines produced for preschool children, facilitating its use in assessing respiratory health in early life (7,8).
Lung function tests used in preschool children include spirometry, bronchial response testing, multiple breath washout (MBW), fractional exhaled nitric oxide (FeNO), oscillometry and other tests which measure resistance, including the interrupter technique (Rint). and plethysmography. With increased recognition of the importance of maximizing early life respiratory health and the growing availability of tools to do so, their use in LMIC is of particular interest.
Of the preschool lung function tests, spirometry is the most commonly used and most widely available. Even though use is limited in very young or uncooperative children, it is feasible in children as young as 3 years (9). Spirometry measures lung volumes at maximal expiration and is able to assess airflow obstruction, response to bronchodilator therapy and lung volumes on forced expiration. The commonly reported measures are the forced expiratory volume in the first second (FEV 1 ), forced vital capacity (FVC), FEV 1 /FVC and forced expiratory flow between 25 and 75% of FVC (FEF 25−75 ). (10). Spirometry predominantly reflects airflow in large and medium sized airways, and is a poor measure of peripheral small airways or early lung disease (11). Current international recommendations for spirometry collection and interpretation in young children are available (10,12).
Oscillometry (or the forced oscillation technique, FOT). is a simple, non-invasive technique which is performed during tidal breathing. Minimal co-operation is required thereby making this a popular measurement in preschool children. This measures the impedance of the respiratory system, which includes resistance and reactance across a range of frequencies reflecting the entire respiratory system including the small airways (13). The novel intra-breath measurement may be a more sensitive measure of small airway disease, thus allowing early detection of disease (14,15). Recent international recommendations for oscillometry have been published on methodology, technical standards and future developments for use in children (7,16,17).
The interrupter technique measures the resistance of the respiratory system requiring minimal cooperation. The technique involves a sudden interruption of flow during tidal breathing, this allows for alveolar and pressure at the mouth to equilibrate therefore alveolar pressure can be estimated (18,19). Different methods have been used to perform the test making the comparison of results difficult thus highlighting the need for test standardization. Furthermore, research is still required to determine the best algorithm to calculate pressure at the mouth during occlusion and the cut-off value for Rint post bronchodilator needs to be established (7).
Multiple breath washout (MBW) is used to assess ventilation homogeneity. It measures the functional residual capacity (FRC) and the lung clearance index (LCI) in preschool children (8). It is more sensitive than spirometry for the detection of peripheral airway disease and has been successfully used in the monitoring of children with cystic fibrosis (CF) and primary ciliary dyskinesia (PCD) (7,20). It correlates with high resolution CT scan in CF patients as well as in children with asthma, bronchiolitis obliterans or chronic lung disease of prematurity (8,21).
Fractional exhaled nitric oxide (FeNO) is a non-invasive marker of T-helper cell type 2 (TH2). eosinophilic airway inflammation. Tests can be performed with high repeatability and accuracy (22)(23)(24). Its main use is an adjunct to the diagnosis of TH2 type asthma and guiding the use of inhaled corticosteroid (ICS) treatment. In addition, FeNO can also assist in the differential diagnosis of other conditions such as cystic fibrosis (CF), PCD, scleroderma, obstructive sleep apnoea syndrome and hepatopulmonary syndrome. Fractional exhaled nitric oxide levels are low in PCD; however measurement of nasal NO improves the diagnostic accuracy and is a useful screening tool for PCD (23,25). Normal values for FeNO are published for children from 4 years of age (24), as well as international applications and the use of FeNO (23,24,26). However, there are limited data on the use and feasibility of such measures in LMIC, despite the high burden of lung disease and need for objective tools to diagnose and monitor these.
We aimed to review the use of preschool lung function testing in children in LMIC, with reference to feasibility and clinical applications, to identify opportunities for optimizing diagnosis and management of childhood respiratory disease in these settings.

METHODS
We reviewed published literature of preschool lung function testing in LMIC, which included children between the ages of 3 to 5 years. We included published papers from 1 January 2011 to 31 January 2022 that included lung function testing in the preschool age group from a World Bank defined LMIC.
The search was conducted on the following electronic databases: PubMed, Scopus, EBSCOhost (Cinahl, Africa wide information, Health source-Nursing/Academic Edition). and Web of Science including the search terms: Respiratory Function Test * " OR "Lung function test * " OR "pulmonary function test * " OR "respiratory function test * " OR "multiple-breath washout" OR "Forced oscillation technique" OR" tidal breathing" OR "fractional excretion of nitric oxide" OR Spirometry OR Oscillometry OR "impulse oscillometry" OR "interrupter technique" OR "interrupter resistance" AND "preschool child * ". Full search strategy can be seen in Supplementary Table S1. Reviews, editorials, case reports and conference proceedings were excluded. Any papers including only children ≥ 6 years were excluded. Abstracts of identified documents were reviewed and screened by SC, with second author DG assisting with inclusion queries. All included papers were reviewed by SC.

Spirometry
The majority (40, 68%) of papers included spirometry, ( Table 1). Spirometry was used to develop reference equations, diagnose and manage respiratory diseases including asthma or CF, assess the impact of air pollution, electronic exposures (e-exposures) or socio-economic status on lung function. Twenty (50%) studies were from Asia, 9 (22.5%) from South America, 7 (17.5%). from Africa and 4 (10%) from Europe and the Middle East.
A number of reference ranges were generated for individual population groups in LMIC, which were compared to either local reference ranges and/or The Global Lung Function Initiative (GLI) 2012 equation (32)(33)(34)(35)(36)(37)(38)(39). Many resulted in an over estimation/underestimation of lung function for the population assessed, highlighting the complexity of population differences in lung function. A Nigerian study determined reference equations for children with sickle cell anemia (40). Published international equations also used arm span to determine height for spirometry (41). The GLI 2012 "Caucasian" provided a reasonable fit for Jordanian children (29). When assessing spirometry in two groups of healthy children with Indian ancestry, one living in UK and the other in India, the GLI equations for best fit differed: the "GLI-Black" equation was most useful for interpreting the South-Asian data and "GLI-Other" for North Indian data (42). Similarly in a South Africa population the GLI 2012 "GLI-Caucasian" provided a good fit for the Caucasian population, "GLI-Black " and "GLI-Southeast Asian" was a good fit for the Indian population and "GLI-Other" fit the Black African and Mixed ancestry populations well (43). These findings highlight the importance of multiple factors, including environment and socioeconomic exposures that impact population differences (42,43).
Numerous environmental factors were reported to impact on lung function. Children living in rural areas or exposed to poorer socio-economic circumstances had lower lung function compared to those in urban areas (42,44). Exposures to volatile organic compounds, particulate matter 10 (PM 10 ) or carbon monoxide (CO) were associated with a decrease in FEV 1 and FVC in exposed preschool children (45)(46)(47).
In a study in Guatemala, where CO was used as a proxy for PM 2.5 , timing of chimney stove installation was compared to cooking over open fires, and showed a decrease in PEF of 173 ml/min/year (95% CI −341 to −7). with chimney stove installation at 18 months compared to installation at birth (48). A Chinese study failed to demonstrate any significant association between PM 2.5 exposure and any of the spirometry lung function measures, however an increase in oscillometry resistance was noted suggesting that oscillometry may be a more sensitive measure (49). In addition to air pollution, e-waste is a growing concern. Chinese children living in e-waste exposed areas had significantly lower birth weight, chest circumference and spirometry lung function compared to those in unexposed areas (27,28).
The majority of included papers used spirometry in the clinical diagnosis and management of pediatric obstructive lung disease including asthma and CF. The studies investigated genetic predisposition to asthma, and management of acute, poorly controlled and cough-variant asthma (CVA). and the impact of obesity (50)(51)(52)(53). Genes associated with increased susceptibility to asthma and lower spirometry indices were identified in Chinese and Egyptian children (54,55). One study assessed current definitions of BDR, suggesting that a BDR of >7.5% may be more valuable in young children rather than the adult defined 12% (56). Another study included assessing impact of ABPA on spirometry of asthmatic children (57). In a Chinese study acute asthma did not respond to adding montelukast to the regular regimen (58), however it was noted that in children with CVA, cough associated with chronic airway allergic inflammation without wheeze, had a significant improvement in FEV 1 , FVC, and PEF (p < 0.001). with montelukast and budesonide compared to budesonide alone, while another study noted that FEV 1 /FVC was normal in CVA compared to patients with asthma (p < 0.001) (51,52).
Spirometry was also used in South American, Turkish and South African children to monitor lung function in cystic fibrosis (59), and showed good correlation between CT scan Bhalla score and FEV 1 , FVC, FEF 25−75 (60).

Fractional Exhaled Nitric Oxide
Of the 7 papers using FeNO, 4 (57%) clinical studies measured FeNO to assess risk and treatment response of asthma or recurrent wheeze; and 3 epidemiological studies assessed the impact of IAP on airway inflammation in children. These included studies mainly from Asia (5 from China, 1 from Thailand) and 1 from Ecuador, South America. There were no Frontiers in Pediatrics | www.frontiersin.org   published preschool studies from Africa. Included studies are summarized in Table 2.
The success rate for performing FeNO in preschool children ranged between 86-99%. Thai children living in a metropolitan area attending day care, average age of 50.1 months (range 29-72 months), achieved a success rate between 86-93% with data collected over 3 time points (61). A cohort of 507 Chinese children aged 5 years achieved a success rate of 99% (62). Over half of studies (57%) used a Niox Mino analyzer (Aerocrine, Solna, Sweden). to measure FeNO using the single breath technique. Most studies used the recommended normal standards (26). No studies have explored population differences in FeNO.
All 3 studies assessing impact of air quality on respiratory health found that environmental pollution, including benzene and PM 2.5 , were associated with high FeNO (49,61,62).
Studies from China and one from South America, used FeNO measurement in the management of preschool asthma (63)(64)(65), noting the strong association with IgE mediated inflammation and ICS efficacy (63). However, it was also suggested that FeNO may be less useful in the preschool age group for detecting ICS response, as the mean FeNO level was significantly higher only in the "older" age group than the cut off values reported in other studies for the diagnosis of asthma (63). FeNO was not able to predict recurrence (64).

Oscillometry and Interrupter Technique
Oscillometry has been successfully used in LMIC to assess the impact of early life exposures and for clinical management of children with respiratory disease. These include: 6 studies from Asia, 6 from South America, 2 from West Africa and 1 from the Middle East summarized in Table 2. The studies used a range of commercially available equipment including both impulse oscillometry (IOS). and airwave oscillometry (AOS).
Oscillometry has proven to be effective in assessing the impact of early life exposures on preschool lung function (49,66). An increase in household PM 2.5 exposure increased airway reactance in Nigerian children, similarly high personal exposure to PM 2.5 was associated with an increase in small airway resistance (R 5−20 ), total airway resistance (R 5 ). and resistance frequency dependence (R 5−20 ), representing small airway disease, in asthmatic Chinese children living outside Shanghai (49,66).
In a Ghanaian longitudinal cohort, infants exposed to a less diverse nasopharyngeal microbiome had a higher small airway resistance compared to a more diverse microbiome at 4 years of age (67). Further, in a Mexican birth cohort study, prenatal omega 3 fatty acid supplementation in pregnancy did not influence preschool lung function at 36, 48 or 60 months (68).
Oscillometry was easily performed by preschool children with the overall success rate ranging between 74-98% in studies making it a particularly attractive option (69)(70)(71)(72)(73). As a clinical tool oscillometry for preschool children is supported by studies that are able to detect differences in prebronchodilator lung function of preschool children with respiratory symptoms compared to those without respiratory symptoms (74).
Reference data for Mexican, Thai, Turkish and Colombian children have been collected (70)(71)(72)(75)(76)(77), facilitating the use in diagnosis of respiratory disease. It is also a useful tool to assess airway reversibility in asthmatic children and cut off values for bronchodilator response were proposed (69,73). A Chinese study in children with obstructive sleep apnea hypopnea syndrome (OSAHS). demonstrated an increase in total airway resistance in children with 2 | Detail of included studies that assessed fractional exhaled nitric oxide (n = 7), oscillometry (n = 15), interrupter technique (n = 2). and tidal breathing (n = 2).   OSAHS compared to children with snoring but without OSAHS (78). The interrupter technique was used in 2 South American studies, one of which looked at the development of reference values for newborn, infants and preschool children, while the other measured Rint in children with cystic fibrosis and found that Rint correlated well to spirometry FeV 1 and FEF 25−75 , but not accurate in determining bronchodilator response (79,80).

Tidal Breathing
Tidal breathing was used in 2 studies, one an Indian birth cohort and the other a Chinese retrospective cohort study (81,82), ( Table 2). Tidal breathing measurements were in keeping with an obstructive pattern in patients with bronchiolitis obliterans (82). Indian children who had an acute respiratory tract infection in infancy had increased ratios of tidal expiratory flow (TEF) at 25 or 50% of tidal expiratory volume to peak TEF (TEF50 or TEF25/peak TEF) at 3 years suggesting increased airway resistance (81).

CONCLUSION
Preschool lung function tests used in LMIC were feasible with high success rates. Success with spirometry increased with age (29), and oscillometry had higher feasibility compared to spirometry. All tests were useful for clinical application and epidemiological studies. Comprehensive preschool testing including spirometry, FeNO, oscillometry and tidal breathing were only reported from China, and only 19 countries of 137 registered LMIC (14%) were represented in the review, a large discrepancy as the majority (80%) of children live in LMIC, where the burden of early life respiratory disease and exposures known to cause illness are high. These include air pollution, maternal smoking, a high infectious load including tuberculosis, and high rates of preterm birth (2,(83)(84)(85)(86). Lung functions tests assessing these vulnerable groups are lacking. A number of studies explored the effects of air pollution; and the studies suggest that air pollutants increase airway inflammation and may in part explain the association between household air pollution and recurrent wheeze (83). Given the burden of early life exposures and the need to identify preventative measure, priority should be given to strengthening access to lung function assessment tools.
The majority of clinical studies focussed on asthma diagnosis and management, including defining BDR in young children. The prevalence of asthma in LMIC is high, with a temporal increase in severe asthma (87). Of particular concern is that up to 40% of children in LMIC with severe symptoms are undiagnosed (87,88). Oscillometry was useful, and more sensitive than spirometry, in measuring airway resistance and cut off values for BDR response was also determined. FeNO has been identified as a useful adjunct in diagnosis and informing treatment in asthmatic patients (89). Access to these tests to improve diagnosis and management is needed for the many children living in LMIC.
Other clinical uses have included management of chronic lung conditions like CF, bronchiolitis obliterans, OSAHS, obesity and systemic sclerosis all of which are associated with respiratory complications. Lung function tests assessing vulnerable groups such as children born preterm, those living with HIV or exposed in utero and those with a history of pulmonary TB and other early life lower respiratory tract infections are lacking in children from LMIC.
The importance of reference range equations were highlighted in these studies. A healthy standard is needed for lung function tests to distinguish between health and disease in different populations. The GLI reference equations attempt to address this, however unique environmental exposures including inutero exposures influence lung development and this needs to be considered when developing reference equations (90). Research studies assessing impact of exposures require appropriate patient samples and statistical modeling and should include both exposed and unexposed control groups.
There are currently no published data for MBW in preschool children in LMIC settings, however data on infant testing have been published in South Africa (3,91). There is limited data on tidal breathing and the interrupter technique. Impulse and airway oscillometry use different input signals and this may impact results especially at high impedance, which is associated with disease. Furthermore, FeNO single breath and tidal breathing measures have different interpretation of normal ranges, and the results are not interchangeable. These factors may affect the interpretation and comparison of data.
Challenges to accessing lung function testing need to be addressed, these include lack of trained personnel as each lung function test requires specialized training skills and lack of financial resources to support development and implementation of these tests. Spirometry and oscillometry are relatively inexpensive, whereas FeNO and MBW are currently more expensive further limiting access. Equipment maintenance and poor access locally to consumables and technical support incur further costs. This reduces research outputs and lack of robust data to better diagnose, prevent and manage respiratory disease in these settings (5).
In conclusion, preschool testing in LMIC is feasible, both epidemiologically and clinically. It has the potential to be useful in strengthening the diagnosis, management and prevention of respiratory diseases in younger children but is underutilized. Spirometry still remains a key clinical and epidemiological tool in LMIC, however has limitations especially in young children. Understanding and addressing the challenges for improving access to these tools is needed in order to strengthen the prevention, early diagnosis and management of childhood respiratory disease in LMIC.

AUTHOR CONTRIBUTIONS
SC, DG, and HZ conceived the idea. SC and DG reviewed the literature and drafted the manuscript. HZ provided further input into the manuscript. All authors reviewed, contributed, and approved the final manuscript.