Baseline Proteinuria and Serum Creatinine Concentration as Clinical Predictors of Complete Renal Response in Patients with Lupus Nephritis: A Single-Center Experience.

The aim of the study was to identify robust predictors of complete renal response (CRR), within 36 months, in a single-center cohort of lupus nephritis (LN) patients. Patients with biopsy-confirmed LN who underwent kidney biopsy between 1 January 2010 and 31 December 2020 were included and followed up for at least 6 months. CRR was defined as a reduction of urinary protein-to-creatinine ratio (UPCR) below 0.50 g/g. We evaluated baseline demographic, laboratory, and biopsy characteristics as potential predictors of CRR, and selected the variables further evaluated with Kaplan−Meier curves and log-rank tests. The traits with a p-value < 0.1 were later tested with both uni- and multivariable Cox proportional hazard models. Our sample consisted of 57 patients (84% females, median age 32 years), out of which 63.2% reached CRR within 36 months. The initial UPCR and estimated glomerular filtration rate (eGFR) were the only variables in multivariable Cox regression model, which were selected through backward elimination, with a significance threshold <0.05 (HR = 0.77, p = 0.01 and HR = 1.02, p = 0.001). Our results confirmed the role of initial UPCR and serum creatinine concentration (sCr) as predictors of CRR in LN.

Artificial neural network - an effective tool for predicting the lupus nephritis outcome.

BACKGROUND: Lupus nephropathy (LN) occurs in approximately 50% of patients with systemic lupus erythematosus (SLE), and 20% of them will eventually progress into end-stage renal disease (ESRD). A clinical tool predicting remission of proteinuria might be of utmost importance. In our work, we focused on predicting the chance of complete remission achievement in LN patients, using artificial intelligence models, especially an artificial neural network, called the multi-layer perceptron. METHODS: It was a single centre retrospective study, including 58 individuals, with diagnosed systemic lupus erythematous and biopsy proven lupus nephritis. Patients were assigned into the study cohort, between 1st January 2010 and 31st December 2020, and eventually randomly allocated either to the training set (N = 46) or testing set (N = 12). The end point was remission achievement. We have selected an array of variables, subsequently reduced to the optimal minimum set, providing the best performance. RESULTS: We have obtained satisfactory results creating predictive models allowing to assess, with accuracy of 91.67%, a chance of achieving a complete remission, with a high discriminant ability (AUROC 0.9375). CONCLUSION: Our solution allows an accurate assessment of complete remission achievement and monitoring of patients from the group with a lower probability of complete remission. The obtained models are scalable and can be improved by introducing new patient records.