Plasma exchange or immunoadsorption in patients with rapidly progressive crescentic glomerulonephritis. A Swedish...

A therapeutic removal of antibodies may be achieved by immunoadsorption (IA) or by plasma exchange (PE). The aim of this prospective randomised study was to compare the efficacy of these different techniques with regard to treatment of patients with rapidly progressive glomerulonephritis (RPG) having at least 50% crescents. Forty-four patients with a RPG were included for treatment either by IA or PE (with albumin as substitution for removed plasma). All patients were additionally treated with immunosuppression. A median of 6 sessions of PEs were performed in 23 patients compared with 6 IAs in 21 patients. Goodpasture's syndrome (GP) was present in 6 patients (PE 3, IA 3). All of them started and ended in dialysis, two died. Among the remaining 38 patients (26 men, 12 women) 87% had antibodies to ANCA. Creatinine clearance for PE versus IA were at a median at start 17.1 and 19.8 ml/min, and at 6 months 49 and 49 ml/min, respectively. At 6 months 7 of 10 patients did not need dialysis (remaining: IA 0/5 and PE 2/5, n.s.). The extent of improvement did not differ between the groups. Three patients died during the observation period of 6 months (IA 2; PE 1, on HD). Although no difference was found between the IA or the PE group this study shows that the protocol used was associated with an improved renal function in most patients (except for Goodpasture's syndrome) whereas 70% of them could leave the dialysis program.