SAT0183 LUPUS NEPHRITIS: A RETROSPECTIVE LONGITUDINAL STUDY LOOKING FOR CHRONIC RENAL DISEASE ASSOCIATED FACTORS, FROM THREE SOUTH-EUROPEAN COHORTS OF PATIENTS IN FOLLOW-UP SINCE 2000

Abstract

Background: Renal involvement in systemic lupus erythematosus (SLE) is the most frequent severe manifestation and carries a bad prognosis. Objective(s): To analyze the outcome of lupus nephritis (LN) in terms of chronic kidney disease development (CKD). Method(s): Design: multicentre restrospective observational study. Patient(s): SLE patients (ACR97) with biopsy proven LN attending to three South European Rheumatology departments. Variables: demographics, SLE related variables, including global activity (SLEDAI-2K), renal flares, therapies, ACR response criteria and CKD. Statistical analysis: descriptive bivariate and multivariate analysis exploring factors associated to CKD. Result(s): Seventy-six patients with biopsy-proven LN were included, 90, 7% female; mean age: 33 years; mean disease: duration 14 years; mean follow-up time (since LN diagnosis): 8, 5 years. LN class III, IV and V were present in 22%, 75% and 3% of the cases, respectively. At LN diagnosis 68 (89%) patients had a severe renal flare. Forty-one (56.1%) and 49 (64.4%) had HTA and nephrotic syndrome, respectively. The mean 24h proteinuria levels at LN diagnosis was 4, 6g. Mean SLEDAI-2K at the time of flare was 20.3, with 69 (65.7%) patients having an extrarenal flare. The treatments used to induce remission were: glucocorticoids (100%): pulses M-prednisolone in 49 (64%) and oral prednisone (mean starting dose): 43 mg/day (+/-20.6); intravenous cyclophosphamide in 42 (55%) patients; mycophenolate mofetil (MMF) in 21 (27%) patients; calcineurin inhibitors in 5 (11%) patients; rituximab in 4 (5.2%) patients; and oral cyclophosphamide in 4 (5, 2%) patients. Forty-eight (63%) patients were receiving hydroxychloroquine. MMF was the immunosuppressant (IS) more frequently used (52%) as maintenance therapy. At 3, 6 and 12th months, the mean proteinuria was 2.3g/24h, 1.53g/24h, 1.1g/24h, respectively (p< 0.001). Fifty-five (77, 5%) of patients achieved complete response and 61 (84.7%) presented complete or partial response. Median time to renal remission: 12.5 months (6, 17.5). In 32 patients (42%) it was possible to discontinue IS. Sixteen (21.9%) patients developed CKD, 4 (5.3%) needing dialysis and 1 (0.76%) renal transplantation. Serious infection was recorded in 23 (34.8%) patients. Five (6.6%) patients died (2 cardiovascular cause). In the bivariate study, the following variables were significantly associated with CKD: male sex, hypertension, ACEI drugs, severe infection after LN, temporal dyalisis, non ACR renal response, non use of hydroxychloro-quine, time to achieve 10mg/day of prednisone, previous creatinine to LN, maximum creatinine at LN, hyperlipidemia at 3 months of LN, active urinary sediment at 12 months, creatinine at 6 and 12 months, proteinu-ria at 6 and 12 months. In the logistic regression model, using genetic algorithms, we found that proteinuria at 6 months was significantly associated with CKD (OR:2.95; 95%CI 1.19, 9.29, p= 0.03). Hypertension and male sex were marginally associated (p=0.06, both). Conclusion(s): A considerable percentage of LN patients developed renal chronic failure (21, 9%). A high percentage of ACR response was achieved using medium dose (40mg) of glucocorticoids for induction. A significative reduction of proteinuria was achieved at 3 months, but protei-nuria at 6 months was the only factor finally associated with CKD.