Dr YI ZHI CHEONG, EOIN D O’SULLIVAN1,2,3, DHARMENAAN PALAMUTHUSINGAM1,4, SHAUN CHANDLER1, SHARAD RATANJEE1, DWARAKANATHAN RANGANATHAN1,5, KEN WANG TAT CHAU1,5
1Kidney Health Service, Metro North Hospital and Health Service, Brisbane, Australia, 2QIMR Berghofer Medical Research Institute, Brisbane, Australia, 3Institute for Molecular Bioscience, University of Queensland, Brisbane, Australia, 4School of Medicine, University of Queensland, Brisbane, Australia, 5School of Medicine and Dentistry, Griffith University, Brisbane, Australia
Biography:
Yi is a 3rd year basic physician trainee in Metro North Hospital and Health Service. She graduated from Monash University Malaysia in year 2020. Her clinical and research interests are in kidney transplant and glomerulonephritis.
Aim:
To validate prognostication tools for ANCA associated glomerulonephritis (AAGN) in an Australian cohort.
Background:
AAGN can lead to substantial morbidity and mortality. The histological ANCA renal risk score (ARRS) had been recently updated and renamed to ANCA kidney risk score (ARKiS). However, these prognostication tools remain unvalidated in the Australian population.
Methods:
All kidney biopsies at a tertiary centre in Brisbane from 2005 to 2024 were reviewed. Hospital records were screened to extract information. A composite outcome was defined by end-stage kidney failure requiring dialysis or mortality.
Results:
The 98 patients included in this study were stratified into low-risk (54.1%, 53 patients), medium-risk (35.7%, 35 patients) and high-risk (10.2%, 10 patients) categories using ARRS scores. Using ARKiS, 65 patients (66.3%) were deemed low-risk, 21 patients (21.4%) moderate-risk, and 12 patients (12.2%) high-risk. Kaplan-Meier curves showed a statistically significant difference in the composite outcome at 3 years among ARRS risk groups – low-risk (6%), medium-risk (37%) and high-risk (70%) groups (p <0.001). Similar differences among ARKiS risk groups were seen – low-risk (8%), moderate-risk (43%) and high-risk (75%) (p <0.001). ARRS showed superior performance, with a lower Integrated Brier Score (0.162) compared to ARKiS (0.180), and both models outperformed the reference Kaplan-Meier model (0.210). At 3 years, more patients with MPO-AAGN reached the primary outcomes compared to PR3-AAGN (29% vs 13.6%, RR 2.13).
Conclusion:
In an Australian cohort of AAGN patients, both ARRS and ARKiS demonstrated prognostic value in predicting the composite outcome. However, ARRS showed better performance with a lower Integrated Brier Score, supporting its use for risk stratification and care planning in our region. External validation in larger Australian cohorts is warranted.