#6399 ROLE OF THERAPEUTIC PLASMAPHERESIS IN PAUCI-IMMUNE VASCULITIS – A SINGLE CENTRE EXPERIENCE

Abstract

Abstract Background and Aims The indication for plasmapheresis (PLF) in pauci-immune vasculitis (PIV) is controversial. This study aimed to characterize the population and renal outcome of patients with PIV undergoing PLF in a hospital in the north of Portugal. Method We conduct a retrospective cohort study of patients followed by a multidisciplinary team, from diagnosis to death or May/22. Two groups were defined: patients undergoing PLF (group A, GA) and not undergoing PLF (group B, GB). Demographic variables, comorbidities, serological subtype, clinical presentation, histological classification and therapeutic were evaluated. Recorded outcomes were death and ESRD (end-stage renal disease) (eGFR,15 ml/min per 1.73 m2, dialysis, or transplantation). Renal survival was estimated using the Kaplan-Meier analysis and differences between curves were evaluated using the log-rank test. Results A total of 72 patients, 26 (36.1%) in group A and 46 (63.9%) in group B were included. In both groups the most frequent serotype was myeloperoxidase ((MPO) (n = 23, GA; n = 38, GB)). In both groups, most patients were male (61.5% in GA and 56.5% in GB), with a mean age at admission of 64 (SD±11.5) and 66.5 years (SD±12.2) in GA and GB, respectively (p = 0.362). Most GA patients had pulmonary and renal involvement on admission (80.7%, n = 21), versus GB (36.9%, n = 17), p<0.05. Mean serum creatinine was superior in GA (6.74 mg/dl, SD ± 2.75) versus GB (3.82 mg/dl, SD±2.95), p<0.05. Twenty patients from GA (76.9%) and 7 from GB (15.2%) required dialysis upon admission, p<0.05. In both groups, almost all patients underwent induction immunosuppression with cyclophosphamide (mean cumulative dose of 4.4 grams (SD ± 2.6) and 5.9 grams (SD ± 2.7), in GA and GB, respectively) and azathioprine in the maintenance phase, with a rapid corticosteroid weaning scheme in both phases. Median follow-up time (months) of GA and GB was 11 (IQR [4.8-37.5]) and 58.5 (IQR [14.5-92]) months. Regarding renal prognosis, 9 of 20 patients in GA and 3 of 7 patients in GB who needed dialysis on admission, partial recover of renal function (p = 1.00). About 37% (n = 17) of patients of GB and 73% (n = 19) of patients of GA reached the composite renal outcome, with a median time of 0 (IQR 0 - 21.8) and 38.5 months (IQR 5.3-92), respectively. GA showed worse renal survival, with a mean of 30.9 months vs. 116.5 months in GB, p = 0.01. Renal survival at 2 years was 68.7% in GB and 24% at GA. Patients in both groups with admission serum creatinine ≥ 5.7 mg/dl had similar renal survival (GA 39.9 and GB 32 months, p = 0.714). Regarding the histological classification, 91.7% and 38.9% in GA and GB belonged to the crescentic class, (p = 0.009). Renal survival in the subgroup of patients with crescentic class was worse in GA (16.6±10.2 vs 83.3±16.1, p = 0.015). The analysis of the Renal Risk Score showed no statistically significant difference between the groups, and between patients with crescentic class in both groups, p = 0.274. Mortality and hospitalization rate due to infection were similar in both groups (p>0.05). Conclusion Patients undergoing PLF had a worse renal outcome, consistent with the higher proportion of patients with severe azotemia and requiring dialysis at admission, in a sample that was mostly ANCA MPO. As described in the literature, the experience of our centre did not demonstrate benefit in the addition of PLF, despite the limitations inherent to an observational, retrospective study and the modest sample size. A better knowledge of etiopathogenesis is essential for the optimization of the therapeutic strategy, together with the promotion of measures that allow timely referral, an early diagnosis and consequent improvement of renal prognosis.