© 2006 American Thoracic Society doi: 10.1164/rccm.200609-1393ED
To Wheeze or Not to WheezeProspective FENO-typing in Early InfancyErasmus University Medical Center–Sophia Children's Hospital, Rotterdam, The Netherlands Recurrent wheeze has a high prevalence in early childhood and is related to viral infections in the majority of infants. Most will not develop asthma but become symptom-free well within the first decade. Asthma can often be traced to early childhood, but unfortunately, the diagnosis can be made with confidence only in retrospect, when children are 6 to 8 years old. Airway eosinophilic inflammation and structural airway wall changes that are characteristic in asthma have been demonstrated in very young children (1), and it has been speculated that such abnormalities could precede clinical disease (2). Invasive techniques to assess airway inflammation in infants are not suitable for practice, and therefore, data are limited. The fractional concentration of exhaled nitric oxide (FENO) is a well-validated, noninvasive marker for eosinophilic airway inflammation (3). Clinical studies have shown that FENO performs well as a diagnostic and monitoring tool for asthma in adults and older children (4), but data on FENO in infants are limited. In this issue of the Journal (pp. 1292–1298), Latzin and colleagues describe a study in which they have measured FENO in healthy neonates, and examined its predictive value for respiratory symptoms in the first year (5). Essentially, this was a negative study, as no association was found between FENO and later development of respiratory symptoms. However, post hoc analysis of data stratified for maternal atopy and smoking showed that an elevated FENO indeed was a strong predictor of respiratory symptoms in infants with atopic and/or smoking mothers. This finding is new, and raises a number of questions about the interpretation of FENO in infants and about the mechanisms that predispose infants to early-onset asthma. Strong points of this study are the elegant technique to measure FENO in unsedated infants and the prospective collection of data on respiratory symptoms by frequent phone contacts; weak points include the small numbers in relevant subgroups—only eight children had an atopic, smoking mother—and the relatively short follow-up, which makes an asthma diagnosis impossible. An early diagnosis of asthma has always been regarded as important, as it could lead to secondary prevention of asthma and early effective treatment. Moreover, it might prevent overtreatment of young wheezers who will not develop asthma. Birth cohort studies have retrospectively identified risk factors in infancy that are associated with a later asthma diagnosis. Combinations of such factors can help to identify which wheezy children are at an increased risk to develop asthma (6). One would hope that early intervention would substantially alter the disease course. Disappointingly, recent studies have convincingly shown that early treatment with inhaled steroids in wheezy infants does not change the natural history of the disease (7–9). One could reason that preventive action in infants before the onset of first wheeze might be more effective. Identification of a "pre-asthmatic state" within the heterogeneous group of infants who will develop wheeze in their first year is therefore potentially important, and FENO measurements may serve this purpose. What does elevated FENO mean in infants and preschool children? First of all, normative values of FENO in infants have not yet been established, and the methodology has not been standardized (3). Latzin and colleagues used a previously validated method that does not require sedation and takes flow into account during tidal breathing (10), thus eliminating some of the variability of previous techniques that used tidal breathing without controlling for flow (11). One of the first studies on FENO in infants showed increased FENO during episodes of acute wheeze, and improvement with steroid treatment (11). Several other cross-sectional studies have reported marginally elevated FENO in wheezing infants, especially in those with eczema and/or an atopic parent (12). Maternal smoking affects FENO in infants, and the effect seems to be modified by the mothers' atopic status, with elevated FENO only in infants with atopic mothers (13). It is remarkable that high FENO levels, which are common in older children with steroid-naive atopic asthma, seem to be rare in infancy and up to the age of 4 years (14). This may be explained by a relatively high airflow in the small pediatric airway, which would reduce FENO due to limited time for diffusion of NO from the airway wall into the air stream or, alternatively, by less severe eosinophilic airway inflammation. How accurately does FENO reflect airway pathology? Another article in this issue, by Lex and colleagues, reports on the cross-sectional correlation of FENO, sputum and bronchoalveolar lavage eosinophils, and airway mucosal eosinophilia in selected, steroid-treated young children, and indicates that the agreement between any two of these markers is imperfect (15). Diagnosing eosinophilic inflammation on the basis of a single marker is therefore not necessarily accurate (15). Clearly, the interpretation of FENO in early infancy is still uncertain due to lack of insight into possible determinants of FENO in this age group. Only a limited number of potential confounders have specifically been examined (16). The range of FENO in infants is fairly narrow, and differences between children with or without an atopic or smoking mother are often within the range of reproducibility within individuals. This raises doubt as to whether FENO will become effective for individual prediction purposes in infants. Moreover, it is uncertain if the presence of eosinophilic inflammation in early-onset asthma is predictive of persisting symptoms and later asthma. A limitation of the present study is that FENO was only predictive in children with an atopic mother, whereas most children with an atopic asthma diagnosis have parents without atopy. Rather than being useful for individual prediction, the results by Latzin and colleagues are interesting because they suggest a pre-existing process in the airways of children who will become symptomatic—and perhaps asthmatic—later in life. The primary endpoint used by Latzin and colleagues was respiratory symptom severity derived from a diary questionnaire. Could this instrument identify future asthmatics? The authors argue that respiratory symptoms show tracking with age, that early persistent wheeze is associated with later wheeze, and that asthma often begins early in life. This has indeed been shown in epidemiologic studies, but unfortunately such data are of limited value for prediction in an individual case. Hence, it seems a bridge too far to interpret the present findings as relevant for early asthma recognition. The study by Latzin and coworkers is the first that prospectively links elevated FENO in early infancy to later development of respiratory symptoms. Their findings are of great interest and a major stimulus for the study of the interaction between early-life exposures and genetic background of the child with early-onset asthma, using exhaled markers of airway disease. Confirmation of the findings in larger groups and development of sensitive and robust techniques to assess FENO and other markers of airway inflammation in exhaled air or exhaled breath condensate are needed. Also, studies with a longer follow-up are necessary to ascertain the long-term outcome. Finally, the added value of FENO compared with other well-established risk factors for asthma should be addressed. Prospective FENO-typing of infants, followed by the development of well-targeted early treatment and prevention strategies before the onset of clinical disease, is an attractive perspective indeed. FOOTNOTES
Conflict of Interest Statement: J.C.D.J. participated in a scientific advisory board for GlaxoSmithKline in 2006 and received REFERENCES
Related articles in AJRCCM:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1,500. Speaker's fees for J.C.D.J. were paid to the Erasmus MC (AstraZeneca,