In this study, we have analysed the clinical features at the time of diagnosis of a large sample of children with GPA/WG whose data were retrieved from the PRINTO database and were collected on the occasion of the project that led to the validation of the EULAR/PRINTO/PRES criteria for the classification of childhood vasculitis [15–17].
Granulomatosis with polyangiits is very rare in children and, although it appears to be the same disease observed in adults, it is not known if childhood presentations are characterized by differences in the frequency of some clinical features well described in adults. Indeed, in other connective tissue diseases, such as systemic lupus erythematosus or juvenile dermatomyositis, the frequency of several clinical manifestations differed from those observed in adults [18, 19].
We have therefore investigated the clinical and laboratory features present at the time of diagnosis in our series of 56 children and compared them with those of the only other large series of paediatric patients (N = 65) reported in the literature as well as with 3 large (N > 50) reported series of adult patients with GPA/WG.
In agreement with previous studies in children, we have found a gender disproportion, as 68% of our patients were females,  while in adult populations females are less often affected ranging from 29 to 50% [7, 8, 10].
In the present paediatric series, the disease was characterised by high frequency of constitutional symptoms, upper and lower respiratory tract, mucosa, skin and musculoskeletal involvement. The frequencies of symptoms typically decreased in the first few months after diagnosis, presumably due to the effect of treatment although data on treatment modalities were not collected in the database. Our series confirms the previous observation by Cabral et al that patients with time to diagnosis longer than 12 months have a higher frequency of ear, nose and throat and cutaneous symptoms and less renal symptoms in comparison with earlier diagnosed patients . This likely reflects the non-specific nature of ENT and cutaneous features in GPA, that occur commonly in children and hence may not prompt clinicians to initially consider vasculitis as the cause. Out of 26 of our patients who underwent kidney biopsy, 82% had glomerulonephritis, similar to the finding by Cabral et al. (94.4% of children with biopsies available) . In adult patients, the data were similar with 73.5% of all patients found with histological evidence of glomerulonephritis .
Comparing the paediatric and adult cohorts, while organ involvement, signs and symptoms were similar, there were differences in their frequencies at disease presentation. Adults had a lower frequency of constitutional, ENT and respiratory involvement; while hearing loss was more frequent in the adult population.
The current work is limited by the lack of information on treatment and disease outcomes. In addition, comparison of clinical and laboratory findings with literature data might be hampered by the lack of standardised definition and by differences in the time of evaluation in the different series. Moreover our study was likely hampered by inconsistencies in laboratory techniques (in particular in relation to ANCA testing), and by the exclusion of cases in this series who did not fulfil the new paediatric classification criteria for c-GPA. Indeed, we recognise that patients may present with limited forms of the disease, without initially fulfilling all criteria for classification and clinicians must remain vigilant in relation to this presentation in children.
In conclusion, paediatric patients compared to adults with GPA/WG have a similar pattern of clinical manifestations but present different frequencies of organ involvement.
Children with GPA/WG are predominantly females/Caucasians with a median onset age of 11.7 years.
Adult showed lower frequencies of female, constitutional, ears/nose/throat, respiratory, laboratory involvement and higher frequency of hearing loss.
Paediatric patients compared to adults have similar clinical manifestations but different frequencies of organ involvement.