Clinical relevance of cluster analysis in phenotyping allergic rhinitis in the paediatric population of the Kuyavian-Pomeranian voivodeship, Poland

Introduction:
Allergic rhinitis (AR) is the most common allergic disease in the world, and additionally, its prevalence is successively increasing. Children with AR constitute a heterogeneous group of patients differing both in the course of AR and in the frequency of asthma coexistence. It is possible to identify children with AR at risk of a more severe course of the disease by analysing the potential family, environmental and clinical factors related to the development of the disease; the findings will help identify patients with a higher risk of developing asthma in the future, and who will benefit the most from early allergen-specific immunotherapy.

Aim:
Evaluation of clinical relevance of cluster analysis in phenotyping AR based on an analysis of selected clinical and environmental factors.

Material and methods:
The study sample was 80 children (7–17 y.o.) with AR, including 28 children with associated asthma. The effects of AR symptoms on the patients daily functioning, skin prick tests (Allergopharma), allergen-specific IgE for airborne allergens (Biocheck GmbH), total cholesterol, cholesterol high-density lipoprotein (HDL), cholesterol low-density lipoprotein (LDL), triglyceride levels in the blood (ARCHITECTcSystem), FeNO and nNO concentrations (HypAir FeNO Medisoft) and results of methacholine challenge test (Lungtest 1000Ispa) were analysed.

Results:
Four clusters of patients with AR were extracted, differing in the incidence and severity of AR symptoms and the coincidence of asthma. Most of the children from cluster 1 (n = 24; 85.71%) and cluster 2 (n = 15; 78.95%) had persistent AR, while most of the children from cluster 3 (n = 11; 73.33%) and cluster 4 (n = 14; 77.78%) had intermittent AR. The co-occurrence of asthma was significantly higher in cluster 1 than in other clusters (p = 0.0002). Children in clusters 3 and 4 reported a lower impact of AR symptoms on daily functioning (p = 0.0153). Children in cluster 1 had significantly more often an abnormally high total cholesterol level (p = 0.033) and in cluster 4 significantly more often abnormally high triglyceride levels (p = 0.009) were observed. Patients in cluster 2 were significantly less likely to have abnormal high LDL levels (p = 0.015).

Conclusions:
Children with AR from the Kuyavian-Pomeranian voivodeship differing in the course of AR, the frequency of coexistence of asthma, and occurrence of lipid parameter abnormalities.