A KANG1, R ARNOLD2, M GALLAGHER1,3, P SNELLING4, J GREEN5, M FERNANDO2,6, M KIERNAN4,7, S HAND3, K GRIMLEY4, J BURMAN3, A HEATH5, K ROGERS1,8, A BHATTACHARYA1, B SMYTH1,9, T BRADBURY1, C HAWLEY10, V PERKOVIC1,2, A V KRISHNAN2,6, M JARDINE1,3
1The George Institute for Global Health, UNSW Sydney, 2UNSW Sydney, 3Concord Hospital, 4Royal Prince Alfred Hospital, 5Sydney Adventist Hospital, 6Prince of Wales Hospital, 7Brain and Mind Centre, University of Sydney, 8University of Technology Sydney, 9St George Hospital, 10University of Queensland
Aim: We aimed to determine whether haemodiafiltration reduces the progression of uraemic neuropathy compared to standard haemodialysis.
Background: Neuropathy is a common complication of kidney disease without proven disease-modifying treatments. Haemodiafiltration provides better clearance of uraemic toxins and is associated with improved nerve function over haemodialysis.
Methods: FINESSE was an open-label, blinded endpoint assessment, randomised controlled trial that allocated maintenance dialysis recipients to haemodiafiltration or high flux haemodialysis for 48 months, or until death or cessation of dialysis at study centres. The primary outcome was the change in the modified Total Neuropathy Score (mTNS) from baseline. Secondary outcomes included a rank composite score combining survival with change in mTNS, survival and adverse events.
Results: A total of 124 participants were randomised and followed for a mean of 41.2 months. At baseline the mean (standard deviation (SD)) mTNS was 6.0 /28. mTNS (SE) worsened by 1.7 (0.4) and 1.2 (0.4) in the haemodiafiltration and haemodialysis groups respectively, giving a mean difference of 0.54 [95% confidence interval 0.65 to 1.73], p=0.37. There was no difference in participant outcome in the ranked composite analysis (log rank p=0.36) or survival (log rank p=0.55). Similarly there was no difference in any of the pre-specified adverse events (cardiovascular composite, access adverse events or time to access failure). There was no difference in the total number of adverse events (Rate ratio 0.88 (95% CI 0.64 to 1.22), p=0.45). The median (interquartile range) convection volume was 24.7 (22.4-26.5)L.
Conclusions: In haemodialysis recipients, haemodiafiltration, compared with high flux haemodialysis, did not affect neuropathy, which remains prevalent in the modern era and continues to be a condition with no known disease-altering therapies.
Trial registration: ACTRN12609000615280
Amy Kang is a nephrologist and PhD student.