#1800 Cascade filtration in severe pulmonary-kidney syndrome: an one-center experience

Abstract

Abstract Background and Aims In some forms of rapidly progressive glomerulonephritis (RPGN), the effect of plasma exchange (PE) is not well understood. Cascade filtration (CF) differs from standard PE in that it uses a double filter system, where the first filter is used to separate plasma from whole blood and the second filter is used to filter high-molecular weight substances. In this study, we investigated the efficacy and safety of CF in patients with ANCA-associated vasculitis (AAV) and Goodpasture's syndrome (GPS). Method During one year we included 18 patients with RPGN whose received CF in the study. The eligibility criteria was eGFR ≤15 ml/min/1.73 m2 and/or diffuse alveolar hemorrhage (DAH), presence of antineutrophilic antibodies or anti-GBM-antibodies. CF was performed on a Plasauto Sigma using a Plasmaflo OP-05W plasma separator and a Cascadeflo EC-30W separator with a fresh frozen plasma or 25% albumin solution. All patients received immunosuppressive therapy with steroids and cyclophosphamide, one patient with AAV received rituximab due to anemia and thromobocytopenia. All data were obtained from one hospitalization with the median of 43 days. Results Eight patients with AAV and three patients with GPS met the eligibility criteria. The median age of the patients in the AAV group was 59.8 (48; 66) years, with a female to male ratio of 1:1. All patients had evidence of systemic disease: seven patients had lung damage, including DAH, two patients had arthralgia, one had purpura and one had severe sinonasal involvement. The median creatinine level was 588 µmol/l (382; 1352), dialysis was initiated in 6 patients, and the median Birmingham Vasculitis Activity Score (BVAS) was 17 (15; 20). After three procedures of CF, all patients showed positive changes on CT scan, and dialysis was stopped in three patients. The immunological response to CF was observed quite rapidly in seven patients, with decreasing ANCA titer after three exchanges and decreasing antibody levels after six exchanges in one patient. There was a significant decrease in inflammatory markers, median C-reactive protein at baseline 179 (147; 226) decreased to 23 (2.8; 30.9) mg/l (p = 0.041). The CF was well tolerated with no significant adverse events. One patient receiving rituximab died after 4 months due to a severe COVID 19 infection. Three patients with GPS, one male 23 years old and two females 56 and 60 years old, also received CF. Creatinine concentrations at baseline were 1285, 1299 and 661 µmol/l, BVAS scores were 18, 20 and 15, respectively. After three procedures of CF, there was an immunological response: decrease in antibody levels from 121 U/ml to 15 U/ml, from 88.2 to 18.4 U/ml, from > 200.0 to 85.4 U/ml, CRP from 252 to 0.7 mg/l, from 120 to 5 mg/l, from 81.9 to 12.58 mg/l. There were no deaths and all three patients remained on dialysis. Conclusion CF was an effective and well tolerated method of treating severe AAV and GPS in addition to immunosuppressive treatment. Possibly due to the different filtration principle and the absence of donor plasma, we did not observe any infection-related adverse events and a lower number of procedures of CF was required compared to standard plasma exchanges. Further larger studies are needed to evaluate the benefits and safety of CF in combination with immunosuppressive treatment in RPGN patients.