Interpretation
It is well-established that atopy and allergic diseases are more frequent in prepubertal boys. In adolescence, in contrast to asthma and eczema prevalence that shifts toward females, AR continues to affect more males up to the age of 20 years(20). In our study, boys outnumbered the girls (69% vs 31%). However, male gender was not associated with asthma presence or AR severity. In agreement with previous childhood studies, allergic conjunctivitis was the most frequent (76.5%) co-morbidity amongst our patients with AR. Zicari et al.(3) have included 1200 Italian children with AR, and conjunctivitis were present in 51.7% of them, whereas Ibanez et al.(21) have reported a prevalence of 53.6% in 1275 Spanish children. However, the prevalence of conjunctivitis in our cohort was significantly higher. We note that the proportion of patients with moderate-severe persistent AR was substantially higher in our cohort compared to previous childhood studies(3, 21), and this may account for a higher proportion of those with ocular symptoms.
According to the united airway disease concept, AR and asthma are common diseases that frequently occur together, and patients share common immunopathological features including increased bronchial hyperresponsiveness and reactivity to a variety of stimuli(6). Consistent with previous data, asthma was present in 40% of our patients(3, 19, 21). Previous epidemiological studies have predominantly emphasized the worsening effect of AR on symptoms of asthma, and it has been shown that severe and uncontrolled rhinitis symptoms were associated with severe and uncontrolled asthma in children and adolescents(22, 23). However, the impact of asthma on the severity and persistence pattern of AR symptoms in children and adolescents has rarely been investigated. In our study, in contrast with the asthma-oriented studies, the rhinitis symptoms were more severe and persistent in children without asthma(24).
Results of several studies indicated aeroallergen sensitization, particularly to HDM and Aspergillus, as one of the independent risk factors for asthma in children(25, 26). In a recent study, Chiu et al. have found significant association between HDM sensitization and various urinary metabolites which were significantly associated with childhood asthma development(27). In our study, the prevalence of HDM and mold sensitizations was significantly higher in children with asthma whereas grass pollen sensitization was lower in this group. In the cross-sectional study by Bousquet et al.(24) which included 591 adults with AR, HDM allergy was associated with perennial and milder symptoms (consistent with our results), and - in contrast to our data - persistent rhinitis. In the study by Zicari et al. which included 2319 Spanish children with AR, no difference was found in terms of temporal pattern and severity between HDM and grass pollen allergic children(3). The results of these studies are based on investigator-led-assignments whilst our sensitization patterns are data-driven. Boulet et al. have concluded that patients with HDM sensitizations are more prone to have asthma as a co-morbidity than those with outdoor allergy(28). In accordance with their findings, lung function parameters were significantly reduced in this group, and there was a significant association with asthma. Bertelsen et al.(4) have speculated that perennial exposure to allergens can be more related to asthma, when compared to pollen exposure which occurs only during a limited time period.
Allergy to pets is regarded as a major risk factor for asthma(29). In the study of Konradsen et al.(30), children with severe asthma had higher levels of IgE antibodies towards furry animals including cat, dog and horse. Nevertheless, the number of the studies which investigated the features of pet allergy in children with AR is limited. Zicari et al.(3) reported the prevalence of pet sensitization only in 2.5% of patients, without any clinical relevance to AR symptoms. However, in the Environment and Childhood Asthma birth cohort study, pet allergens (dog and cat) were the second most prevalent sensitizing agent in children with AR and the investigators documented two major sensitization groups (grass pollen monosensitization vs. grass pollen/furry pets polysensitization)(4). When they compared the co-morbidities and clinical characteristics, in contrast to our findings, asthma prevalence and the severity of bronchial hyperreactivity were greater in children with pet sensitization. However, in accordance with our findings, pet sensitization was associated with milder and perennial AR symptoms.
Adenoid hypertrophy (AH) is one of the most frequent co-morbidities in children with allergic rhinitis(21, 31). Results of previous studies have reported mold sensitization as a risk factor for adenoid hypertrophy in children with AR(32, 33). In agreement with this finding, sensitization to molds was significantly higher in our cluster consisted of children who underwent tonsillectomy and adenoidectomy when compared with patients in the grass-sensitized patients with conjunctivitis. Dogru et al.(33) have found an association between AH and the severity of AR in children. However, this relationship was not demonstrated in our study. The plausible explanation for the lack of this association may be that the children in this cluster were already operated and the symptom severity could be relieved due to the surgery.
Several clinical implications can be extracted from our findings. Asthma symptoms should be queried and lung function tests can be performed especially in children with house dust mite and mold sensitizations. Severe and persistent symptoms should be expected in children with grass pollen monosensitization, therefore, physicians can consider early initiation of the allergen immunotherapy in such children. Pet sensitization can accompany grass pollen sensitization especially in boys. A referral to otorhinolaryngologist in order to exclude adenoid hypertrophy should be considered particularly in children with severe and treatment-resistant nasal obstruction and snoring.
In conclusion, we have identified four different subgroups of children with AR using LCA in this cross-sectional study, which were associated with different patterns of clinical symptoms and comorbidities.
Acknowledgements: We are extremely grateful to all the families who took part in this study. We would also like to acknowledge the hard work and dedication of the study team, which includes computer and laboratory technicians, and nurses.