Asthma-predictive-index, bronchial-challenge, sputum eosinophils in acutely wheezing preschoolers

Pediatric Pulmonology. 2014 Oct;49(10):952-9.



Dorit Ater MD, Bat-El Bar MD, Nir Fireman MD, Elizabeth Fireman PhD, Hanita Shai MD, Diana Tasher MD, Ilan Dalal MD and Avigdor Mandelberg MD.

Pediatric Pulmonary Unit, Wolfson Medical Centre, Holon, Israel and The Sackler School of   Medicine, Tel Aviv University, Israel.



Background: Most preschoolers with viral wheezing-exacerbations are not atopic.

AIM: To test in a prospective controlled trial whether wheezing preschoolers presenting to the ED are different from the above in three different domains defining asthma: the atopic characteristics based on stringent asthma predictive index (S-API), the characteristics of bronchial hyper-responsiveness (BHR) and airway inflammation.

Methods: The S-API was prospectively collected in forty-one preschoolers (age 31.9±17.4 months, range; 1-6years) presenting to the ED with acute wheezing and compared to healthy preschoolers (n=109) from our community (community-control-group). Thirty out of the 41 recruited preschoolers performed two sets of bronchial challenge-tests (BCT)-(methacholine and adenosine) within three weeks and following three months of the acute event and compared to 30 consecutive ambulatory preschoolers, who performed BCT for diagnostic workup in our laboratory (ambulatory- control-group). On presentation, induced sputum (IS) was obtained from 22 of the 41 children. Outcomes: primary: S-API, secondary: BCTs characteristics and percent eosinophils in IS. Results: significantly more wheezing preschoolers were S-API positive compared with the community-control- group: 20/41 (48.7%) versus 15/109 (13.7%, P<0.001). All methacholine-BCTs – 30/30 (100%) were positive compared with 13/14 (92.8%) in the ambulatory-control-group (p=0.32). However, 23/27 (85.2%) were adenosine-BCT positive versus 3/17 (17.5%) in the ambulatory-control-group (p<0.001). Diagnostic IS success rate was 18/22 (81.8%). Unexpectedly, 9/18 (50.0%) showed eosinophilia in the IS. Conclusions: wheezing preschoolers presenting to the ED is a unique population with significantly higher rate of positive S-API and adenosine-BCT compared with controls and frequently (50%) express eosinophilic airway inflammation.

PMID: 24166822



Wheezing preschoolers are heterogeneous in terms of atopic tendency but most are non- atopic and will not develop asthma during school years (1). In this study we tested in a prospective controlled trial, whether preschoolers presenting to the emergency department (ED) with acute moderate-severe wheezing episode vary from the general wheezing preschoolers in three different domains defining asthma: asthma predictive index (API), bronchial hyper responsiveness (BHR), and airway inflammation in induced sputum (IS).

  1. API: Castro-Rodriguez, based on data from the longitudinal Tucson Study (2), developed criteria to determine the risk of future asthma in school years in a child with recurrent wheezing in the first three years of life. The asthma predictive index (API) was adopted in the well-known asthma guidelines, the GINA and the NIH.
  2. BHR is the hallmark of asthma but is present also in other acute and chronic airway inflammatory diseases. There are two different types of challenges indirect and direct (3, 4). Indirect BCT (like adenosine) act through intermediary pathways, mimicking a clinical attack of asthma and considered more specific for asthma diagnosis. Direct BCTs (like methacholine) act directly on specific smooth muscle receptors are considered non-specific and are positive in acute and chronic airway inflammatory conditions beside asthma.
  3. Airway inflammatory phenotypes using induced sputum (IS): Inflammatory phenotypes are recognized in stable and acute adult asthma but are not well established in childhood asthma. Using IS, Wang demonstrated that in acute asthma, the inflammatory phenotype in school-age children (8-17 years) is frequently eosinophilic (50%) but is predominantly neutrophilic in adults (82%) (5). However, IS is considered impossible to obtain from preschoolers and the inflammatory phenotype in wheezing preschoolers is not known.


To our knowledge this study is the first pediatric study looking simultaneously at the different domains defining asthma:

  1. API: In our study almost half (49%) of the children had positive Stringent-API compared to only 13.7% in the community-control-group (p<0.001). Thus our study revealed that children presenting to the ED with acute moderate-severe wheezing episode might differ from the general preschool population, unexpectedly expressing atopic tendency according to Stringent-API and thus possibly are at a much higher risk to develop asthma at school years.
  2. BCTs: Methacholine acts directly on specific airway smooth muscle receptors causing bronchoconstriction in asthma and non-asthmatic airway inflammatory conditions. Therefore we expected a high prevalence of positive methacholine BCT three weeks following the acute event. Indeed, all our studied children demonstrated positive methacholine BCT three weeks and even more than 3 months following the acute event. For the same reason, our ambulatory controls who were all non-healthy symptomatic preschoolers showed a high incidence of (93%) positive methacholine BCT. Adenosine releases endogenous mediators from eosinophils and mast cells mimicking clinical-asthma, thus we expected a lower incidence of positive adenosine BCTs in our preschoolers with mainly viral triggered wheezing. Unexpectedly we found high percentage of positive adenosine BCTs in our studied children: 23/27 (85%) versus 3/17 (17.5%) in the ambulatory-controls (p<0.05). These findings can be explained by a much higher incidence of atopic characteristics as reflected also by a high percentage of positive Stringent-API in these children.
  3. Airway inflammatory phenotypes in IS: Although our numbers of IS samples are small, to our knowledge this is the first study trying to investigate IS inflammatory phenotypes in wheezing preschoolers. In contrast to our hypothesis that preschoolers with viral triggered wheezing which are generally non atopic, would exhibit predominantly a neutrophilic non-eosinophilic phenotype, we found unexpectedly, that 50% of the 18 adequate IS samples showed eosinophilia. These findings may reflect the high tendency for atopic characteristics in these patients. The studied children had evidence for viral infection as documented by 75.0 % positive PCR assay results for one or more respiratory viruses in IS samples studied. This is in concordance with the presence of neutrophilia (more than 30% neutrophils) in 17 out of 18 (94.4%) IS samples. This high level of neutrophils, independent of the percent eosinophils, is expected in viral infection and is substantially higher as compared with normal (<19.5%(.These results can be explained by a reaction to acute infection in a predisposed population with atopic tendency.


Our study has few weaknesses: of the 41 recruited children whose parents consented for the evaluation of API only 30 consented for BCTs and 22 for IS production. Thus for the latter two parts of the study the numbers, although revealing statistically significant results, are relatively small. However, all three outcomes: 49% children with positive Stringent-API, 85% with positive adenosine BCTs and 50% with IS eosinophilia point to the same direction of atopic tendency which further strengthen the validity of our conclusions.

In conclusion: preschoolers presenting to the ED with viral triggered moderate to severe wheezing episode have significantly higher positive API and have significantly more adenosine (asthma specific) positive BCTs than their respective controls. In addition, these preschoolers frequently express eosinophilic airway inflammation. This may suggest a possible atopic predisposition in this population as opposed to the typical non-atopic ambulatory preschoolers with viral triggered wheezing. We also showed that obtaining IS in wheezing preschoolers is possible with a similar success rate (82%) as in adults (80-90%).


  1. L.P. Brand.ERS Task Force. Definition, assessment and treatment of wheezing disorders in preschool children: an evidence-based approach. Eur Respir J 2008; 32: 1096–1110
  2. Castro-Rodriguez JA, Holberg CJ, Wright AL and. Martinez FD. A Clinical Index to Define Risk of Asthma in Young Children with Recurrent Wheezing. Am J Respir Crit Care Med. 2000; 162: 1403–1406
  3. Avital A, Springer C, Bar-Yishay E and Godfrey S. Adenosine, methacholine, and exercise challenges in children with asthma or paediatric chronic obstructive pulmonary disease. Thorax 1995; 50: 511-516
  4. Bakirtas A, Turktas I. Airway hyper-responsiveness to adenosine 5-monophosphate in preschool-age children with asthma. Pediatr Allergy Immunol 2006; 17: 428–434
  5. Wang F, He XY, Baines KJ, Gunawardhana LP, Simpson JL, Li F and Gibson PG. Different inflammatory phenotypes in adults and children with acute asthma. Eur Respir J 2011; 38: 567–574