MS NG1,2, M HARFIELD1, L FRANCIS3, AJ MALLETT1,2
1Kidney Health Service, Royal Brisbane and Women’s Hospital, Herston, Australia, 2Faculty of Medicine, The University of Queensland, Herston, Australia, 3Department of Anatomical Pathology, Pathology Queensland, Royal Brisbane and Women’s Hospital, Herston, Australia
Aim: To validate a web-based clinical decision support system (CDSS) for the prediction of end stage kidney disease (ESKD) in Australian Immunoglobulin A nephropathy (IgAN) patients.
Background: IgAN exhibits a highly variable clinical course. The patients who progress to ESKD may benefit from targeted treatment however, it is difficult to identify these patients at the time of diagnosis. A web-based CDSS is available to predict progression to ESKD in IGAN patients (http://www.igan.net). Notably, this CDSS has not been validated in an Australian population.
Methods: Electronic databases were queried for patients from the Royal Brisbane and Women’s Hospital with biopsy proven IgAN diagnosed between January 1, 2006 to December 31, 2015 (10 years). Demographic and outcome variables were collected from medical charts. Students t tests and chi-squared tests were conducted to compare patients who progressed to ESKD and those who did not progress to ESKD. Patient details were entered into the CDSS (http://www.igan.net) to predict progression to ESKD. Sensitivity and specificity of the CDSS were calculated.
Results: 82 patients were identified. 17 patients were excluded (2: charts not available, 13: follow-up time <1 year, 2: no proteinuria data available), resulting in a final sample size of 65 patients. Increased serum creatinine (p=0.0065), proteinuria (p=0.0037), haemodialysis (p=0.0032) and interstitial fibrosis (p=0.0017) at diagnosis were associated with ESKD. The CDSS had sensitivity of 82.61% and specificity of 66.67%. Specificity increased to 84.85% when patients who had follow-up duration less than the predicted time to ESKD were excluded.
Conclusions: The CDSS exhibited reasonable sensitivity and specificity in an Australian cohort – demonstrating its potential as a clinical tool to assist in decision-making for IgAN patients.
A/Prof Andrew Mallett is a Nephrologist with a special interest in inherited kidney disease and nephrogenetics. A/Prof Mallett has undertaken a Churchill Fellowship and been a recurrent Visiting Fellow at Addenbrooke’s Hospital (Cambridge, UK) and the Cambridge Institute for Medical Research. His PhD (2016, University of Queensland) in nephrogenetics involved extensive national and international collaboration. A/Prof Mallett is a Consultant Nephrologist at RBWH and co-lead of the statewide Queensland Conjoint Renal Genetics Service. He is the National Director of the KidGen Collaborative and the AGHA Renal Genetics Flagship.