engleski

SP188PLASMA EXCHANGE TREATMENT AS PREDICTOR OF CLINICAL OUTCOMES IN ANCA ASSOCIATED VASCULITIS IN CROATIAN REFERRAL CENTER

INTRODUCTION: Plasma exchange treatment (PLEX) in ANCA associated vasculitis (AAV) is still debated in terms of indications, number of treatments needed and its effects on overall patient and renal survival. We present data from Croatian referral center. METHODS: 108 patients with ANCA associated vasculitis have been analyzed, 60 females (55, 6%) with median age of 61 (Interquartile ratio - IQR= 51-70) years. Patients were categorized according to clinical (microscopic polyangiitis = MPA, granulomatosis with polyangiitis = GPA and renal limited vasculitis = RLV), serological (MPO-ANCA, PR3-ANCA, MPO and PR3- ANCA, ANCA negative) and histopathological phenotypes (according to Berden et al. classification). All the patients had renal biopsy performed. We analyzed PLEX as predictor for combined outcome end-stage renal disease and death (ESRDD), ESRD alone, death alone and relapse rate. Survival univariate analysis was performed using Kaplan-Meier analysis and log-rank (Mantel-Cox) test. Variables that had p<0, 1 in univariate analysis were alongside age and gender included in multivariate Cox proportional hazard model. RESULTS: According to clinical phenotype, there were 66 (61, 1%) patients with MPA, 20 (18, 5%) with GPA, 20 (18, 5%) with RLV and 2(1, 9%) with EGPA which were due to small number excluded from analysis. Median serum creatinine levels (SCr) were 316.5 umol/l (IQR= 207-548.5), minimum SCr was 68 and maximum 1402 umol/l. In clinical phenotypes PLEX treated mainly included MPA, while non-PLEX treated group included RLV (p=0.002). In serological phenotypes MPO-ANCA AAVs were dominant in PLEX treated group (p=0.042) and trend in ANCA positive AAVs being more often PLEX treated compared to ANCA negative AAVs (p=0.07). In pathohistological phenotypes, PLEX treated patients were predominantly crescentic class (p=<0.001). When analyzing all the patients together, univariant Cox regression analysis found surprisingly PLEX to be significant positive predictor for ESRDD (p=0.068, HR 1.85 ; CI 0.956-3.592), ESRD (p=0.039, HR 2.3 ; CI 1.04.-5.114) and relapse rate (p=0.038 ; HR 4.2 ; CI 1.082-16.353) Multivariant analysis was then preformed in which PLEX was not found to be significant predictor for all the outcomes. We furthermore divided patients into 4 groups based on SCr quartiles (Q1 SCr min-208 umol/l ; Q2 SCr 209-319 umol/l ; Q3 320-563 umol/l ; Q4 SCr 564-max). Patient distribution according to PLEX treatment per quartile was: Q1 N(PLEX)= 3, N(nonPLEX)= 23 ; Q2 N(PLEX)= 6, N(nonPLEX) =21 ; Q3 N(PLEX)= 14, N(nonPLEX)= 14 and Q4 N(PLEX)= 24, N(nonPLEX)= 1. When analyzing third quartile alone or in combination with fourth or first and second ones, Kaplan Mayer analysis found PLEX not to be significant predictor for all outcomes in our patient population. The only significance was found in Q4 group, but there was only one patient in that group not being treated with PLEX. CONCLUSIONS: The significance of PLEX remains variable depending on the indications and disease severity as well as probably standard of care in individual centers. Our results probably stem from the fact that PLEX was more often used for patients with more severe disease at presentation and it is therefore expected that these patients would had less favorable outcomes. The important question is weather PLEX could have more effect if used independently of severity of renal or lung involvement in AAV patients.

plasma exchangepatient referraltreatment outcomeanca-associated vasculitiscroatian

nije evidentirano