ESPN 53rd Annual Meeting

ESPN 2021


 
The development and preliminary validation of a scoring tool for monitoring disease activity in patients with IgA vasculitis (HSP).
CHLOE WILLIAMS 3 JARED MURPHY 2 Tom Dowsett 1 LOUISE ONI 1

1- DEPARTMENT OF PAEDIATRIC NEPHROLOGY, ALDER HEY CHILDREN’S NHS FOUNDATION TRUST HOSPITAL, LIVERPOOL, UNITED KINGDOM
2- ROYAL LIVERPOOL AND BROADGREEN UNIVERSITY HOSPITALS, LIVERPOOL, UNITED KINGDOM
3- Department of Women’s and Children’s Health, Institute of Life Course and Medical Sciences, University of Liverpool, Liverpool, United Kingdom
 
Introduction:

IgA vasculitis (IgAV, HSP) is the most common form of childhood vasculitis. The Paediatric Vasculitis Activity Score (PVAS) is used for scoring vasculitis, however it may not be specific enough for IgAV. The aim is to develop and perform preliminarily validation of a vasculitis activity scoring tool for IgAV (IgA-VAS).

Material and methods:

The IgA-VAS was derived from existing tools and underwent face validity prior to this study. It consists of 40 manifestations, each scored 0-10, within 5 domains: cutaneous, gastrointestinal, musculoskeletal, renal and other. A retrospective cohort of paediatric patients were scored by two clinicians using the IgA-VAS and PVAS. Test validity, concurrent validity and inter-rater agreement were assessed. A subgroup were scored using a visual analogue scale.

Results:

For preliminary validation, retrospective scoring was performed (February 2021) from a single tertiary centre over a 5-year period. 196 children were identified; 153 met inclusion criteria. 54% were male, median age of 5.7 years (range 0.6-16.7). Median total IgA-VAS scores were 7/125 (range 2-31) and 5/125 (range 2-29) and median PVAS scores were 6/63 (range 2-25) and 5/63 (range 2-20) for rater 1 and 2 respectively. Inter-rater reliability was low for both tools (0.131, 0.225, p<0.001). For the renal domains, median IgA-VAS scores were 2/52 (range 0-24) and 0/52 (range 0-24), and renal PVAS domain score was 0/12 (range 0-12) for both raters. Inter-rater agreement for the renal domains of both tools was similar (0.237 and 0.304, p<0.001). Correlation between the renal domain scores of the tools was strong (rater 1=0.747, rater 2=0.825, p<0.001). The total IgA-VAS score moderately correlated with the visual analogue scale for both raters (0.482 and 0.362, p<0.05), the PVAS strongly correlated with rater 1 (0.504, p=0.004) and moderately correlated with rater 2 (0.372, p=0.043).

Conclusions:

The IgA-VAS continues to evolve following preliminary validation however refinement is needed before prospective validation.