Haemoglobin targets: we were wrong, time to move on

BACKGROUND: The current 3rd edition of the Italian Society of Nephrology guidelines has been drawn up to summarize evidence of key intervention issues on the basis of systematic reviews (SR) of randomized trials (RCT) or RCT data only. In the present guideline, evidence of optimal haemoglobin (Hb) target levels in chronic kidney disease (CKD), either for pre-dialysis, dialysis or renal transplanted patients, is presented. METHODS: SR of RCT and RCT on different Hb target levels in patients with CKD were identified, referring to a Cochrane Library and Renal Health Library search (2005 update). Quality of SR and RCT was assessed according to current methodological standards. RESULTS: Four SR (19 RCT) were found addressing the point. Methodological quality of available trials was suboptimal. In CKD patients (non-dialysis patients) Hb targets of 11.3 g/dL should be preferred to Hb >13.5 g/dL (evidence from RCT). A Hb target of 11.0-11.5 g/dL should be preferred in CKD patients receiving dialysis treatment without significant cardiac disease, since no survival benefits has been showed with Hb >14 g/dL (evidence from RCT). The optimal Hb target in haemodialysis patients with severe cardiac disease should be 10.0-10.5 g/dL (evidence from SR). Increases in Hb target lev-els are associated with improved quality of life, although this was mainly noticed in observational studies and in few RCT often relying on unvalidated quality of life assessment scales. CONCLUSION: In CKD patients current available evidence supports the hypothesis that optimal Hb targets should be low to subnormal.