A Randomised Controlled Trail of Plasma Exchange in Rapidly Progressive Renal Failure of Myeloma.

Abstract In myeloma, plasma exchange (PE) has been suggested to prevent rapidly progressive kidney failure by reducing exposure to nephrotoxic light chains. We carried out a randomized controlled multi-centre trial comparing PE or no PE in 104 patients of whom 101 met the inclusion, exclusion criteria and 4 were lost to follow-up. We compared baseline characteristics as well as renal outcomes and performed a futility analysis to determine the sample size necessary for potential statistical significance for the changes noted. Thirty-nine patients were randomized to the control group and 58 to the PE group with a 6-month follow-up. The baseline characteristics of these 2 groups were similar including serum creatinine, dialysis dependence, age, gender, serum calcium, serum albumin, 24 -hour urine for protein levels and Durie-Salmon myeloma staging. Thirteen (33.3%) of the control group and 19 (33.3%) of the PE group died within 6 months of follow up. Ten patients (31%) in the control and 10 patients (21%) in the PE arm were dialysis dependent at 6 months. Seven patients (47%) came off dialysis in the control and 13 patients (59%) in the PE arm with the mean number of dialysis days from 0–6 months being 45.7±67.6 in the control versus 29.2±56.1 in the PE arm at 6 months. The mean serum creatinine in the control group was 314.6±256.1 μmol/L versus 215.4±215.3 μmol/L in the PE group and the composite end point of death, dialysis or serum creatinine >254 μmol/L occurred in 12 (30.8%) in the control and 11 (19.3%) in the PE arm. The futility analysis to indicate the per group sample size necessary to achieve statistical significance at 6 months for the difference we observed was infinite for cumulative mortality, 805 for dialysis dependence, 2418 for coming off dialysis, 321 for number of dialysis days, 132 for creatinine difference of 100 μmol/L and for the composite outcome of death, dialysis or creatinine>354 μmol/L, 737. We did not observe a statistically significant difference in mortality or renal morbidity for PE versus no PE in patients with myeloma and rapidly progressive kidney failure.