#4360 FACTORS ASSOCIATED TO RENAL REMISSION IN CRESCENTIC GLOMERULONEPHRITIS

Abstract

Abstract Background and Aims Impaired renal function is one of the most relevant factors associated to crescentic glomerulonephritis (CGN) prognosis. However, renal remission has not yet been defined and clinicians are required to use systemic scores (such as BVAS) to predict severity. In this retrospective study, we aim to describe the associated factors to long-term chronic dialysis incidence and survival in CGN. Method We included all biopsy-proven CGN of our center between 2004 and 2022. At baseline, demographics, treatments and comorbidities were collected. Renal function was assessed by glomerular filtration rate (GFR) using CKD-EPI equation, quantification of proteinuria and demonstration of hematuria. During follow-up (median 1486, interquartile range [25-3082] days) renal and vital status was evaluated. Factors associated to dialysis requirement were assessed using Cox regression models. A combined endpoint of death and dialysis was established. Factors associated to the combined endpoint were assessed. Results We included 47 CGN (77% female, 67±15 years). Of them, 35 (75%) presented positive ANCA antibodies, 3 (6%) positive glomerular basement membrane (GBM) antibodies, 8 (17%) ANCA and GBM antibodies and one (2%) presented negative autoimmunity. Induction treatment was based on prednisone and cyclophosphamide in 43 patients (91%), prednisone and rituximab in 3 (6%) and prednisone alone in one (2%). At admission, median CKD-EPI was 11 (11-21) ml/min/1.73 m2, proteinuria was 1030 (552-1872) mg/g and 43 (91%) patients presented hematuria. Nineteen patients (40%) required dialysis at admission. Following the definition of KDIGO guidelines for renal remission, 28 (64%) patients achieve it at 6 months. During follow-up, fifteen patients (36%) started chronic dialysis. Factors associated to chronic dialysis were the type on CGN (dual ANCA and GBM and GBM (+) vs ANCA (+), p = 0.003), CKD-EPI at admission (p = 0.001), AKIN (p = 0.050), requirement of dialysis at admission (p<0.001), percentage of crescents (p = 0.037), proteinuria at admission (p = 0.046) and remission after induction treatment (p<0.001). Twenty-four patients (53%) died or needed dialysis during follow-up. Factors associated to this combined endpoint were CKD-EPI at admission (p = 0.017), type on CGN (dual ANCA and GBM vs others, p = 0.003) (Figure 1), debut in dialysis (p<0.001), remission after induction treatment (p<0.001). An adjusted Cox regression model demonstrated that the need for dialysis or death during follow-up was independently associated to not achieving remission after induction (i.e. 6 months) (HR 8.78, 95%CI (2.88-26.7), p<0.001) and requirement of dialysis at debut (HR 3.96, 95%CI [1.15-13.6], p = 0.029). Conclusion The requirement of dialysis at debut and not achieving remission after induction are independent predictors of death or need for chronic dialysis in CGN.