Latest US KDOQI Anaemia Guidelines update--what are the implications for Europe?

Keywords: anaemia; clinical guidelines; EBPG;KDIGO; KDOQI; CREATE; CHOIR; TREATIn medicine, large randomized controlled trials arepowerful entities; not only do they influence clinicalpractice and impact on clinical practice guidelines, butthey also have the ability to confuse the medical andnon-medical community by producing surprising oralarming results that question current concepts ofclinical management. Examples of this in the non-nephrological literature include the Women’s HealthInitiative trial of hormone replacement therapy forcardiovascular protection in post-menopausal women[1]. This very large placebo-controlled randomized trialin over 16000 post-menopausal women showedbeyond doubt that combined oestrogen and progester-one replacement did not reduce cardiovascular risk,but in fact increased it. Similarly, in the CAST trial(Cardiac Arrhythmia Suppression Trial), the hypoth-esis was that anti-arrhythmic drugs such as flecainideand encainide would reduce the incidence of ventri-cular arrhythmias post-myocardial infarction [2].In fact, mortality was increased with the use of suchagents. Closer to home, in nephrological practice, theresults of the HEMO study countermanded theprevious belief that more dialysis was better, and thathigh-flux dialysis was better than low-flux dialysis [3].Similarly, the 4D (Die Deutsche Diabetes Dialyse)study of atorvastatin in diabetic haemodialysispatients confused the issue regarding the cardiovascu-lar protective role of statin therapy in this patientpopulation [4].In the renal anaemia field, the US NormalHematocrit Trial by Besarab and colleagues [5] beganto discredit the hypothesis that aiming fornormalization of haemoglobin in dialysis patientswith symptomatic cardiovascular disease was betterthan aiming for incomplete correction of anaemia byepoetin therapy. Although this addressed patients withsevere cardiac disease only [6], a subsequent rando-mized double-blind trial in dialysis patients withoutsymptomatic cardiac disease, reported by Parfrey et al.[7], also showed no benefit of a haemoglobin targetwithin the normal range, in comparison with partialanaemia correction, on left ventricular hypertrophy.More recently, we have seen the publication ofthe CREATE and CHOIR studies in the sameissue of the New England Journal of Medicine [8,9].Both of these studies yielded useful data addressing theissue of optimal target haemoglobin ranges in chronickidney disease patients not on dialysis, an area thatpreviously lacked high-quality evidence to supportclinical practice recommendations. The CREATEstudy suggested that early and complete correction ofanaemia does not improve cardiovascular outcomescompared with later and incomplete correction ofanaemia [8]. Moreover, the number of patients startingchronic dialysis was higher in the higher haemoglobinarm. The main limitation of the CREATE study wasthat it ended up under-powered, due to a lower annualevent rate incidence of 6% compared with the 15%used for the sample size calculation.Of even greater concern, however, were the datafrom the CHOIR study [9]. This large prospectiverandomized trial in the United States suggestedthat aiming for a target haemoglobin of 13.5g/dl isharmful compared to targeting a haemoglobin of11.3g/dl, in terms of the composite end-point ofdeath, myocardial infarction, stroke and hospitaliza-tion for cardiac failure. Although the CHOIR data areimportant and informative, one should be aware of thefeatures of this study, which may limit the general-izability and the strength of its results. These have beenextensively discussed elsewhere [10,11], but in short,these include: 50% of the patient population, diabetic,imbalances in the patient groups at baseline [the higher