Dialysis therapy: ‘think differently’

The first successful dialysis treatment was performed, as almost every nephrologist would know, by Dr Willem Kolff in 1945. When Dr Kolff designed the first artificial kidney, he had to determine the size of the dialysis membrane that he would be using. To solve the problem of how to effect a dialysis treatment, he took a piece of artificial sausage skin (it was cellophane material), about 40 cm in length, filled it with 25ml of blood and added 400mg/100ml of urea, shook it up and down in a bath with saline and found, to his surprise, that within 5min, almost all of the urea had been removed by dialysis. Multiplying the length of the tubing by 20 gave him the length of tubing that he would need for an effective artificial kidney. It is really surprising that 62 years on, we are still using urea as the only marker of the adequacy of dialysis. Everybody knows about the urea reduction ratio (URR), percentage reduction in urea (PRU), Kt/V and eKt/V, when we also know that there is no study that has demonstrated any real biological impact of urea as a toxin. In nature, some fishes build up very high levels of urea in their blood. Sharks, for instance, may have a blood urea level of 2.5%, in contrast to the level of 0.01–0.03% in other vertebrates; such high levels allow them to be in osmotic equilibrium with the sea environment. In humans, the addition of urea to dialysate at even higher concentrations than those usually encountered in dialysis patients failed to demonstrate any impact on the symptoms related to the uraemic syndrome. In the HEMO Study, no difference in survival was found between patients treated at standard Kt/V (1.3) vs high Kt/V (1.7) [1]. The French Study Group for Nutrition in Dialysis showed that the survival of dialysis patients was not influenced by the Kt/V value [2]. Patients on haemodialysis (HD) continue to have high mortality rates, a high incidence of cardiovascular morbidity and mortality, high prevalence of hypertension, inflammation and malnutrition, and we are still using urea Kt/V as a marker of the adequacy of dialysis, when it might only be a marker of the adequacy of the diffusive process during a dialysis session. Moreover, urea has a kinetic behaviour which is not representative of many other known uraemic toxins, even low-molecular weight water-soluble compounds. So, maybe the time has come for us to accept that urea is definitely not a uraemic toxin. When Babb and Scribner proposed the middle molecule hypothesis in the early 70s, they had no means of identifying the culprits. In recent years, however, considerable work has been done by the EUTox Group towards facilitating an understanding of what might be the true uraemic toxins [3,4]. Among these, the protein-bound uraemic toxins (PBUTs) have attracted great interest: it is essential to increase our knowledge about these compounds, in order to better understand the complexity of the uraemic syndrome and to improve the outcomes of therapy for end-stage renal disease (ESRD). PBUTs certainly play a key role in the morbidity associated with ESRD. They are now known to be more efficiently removed by the use of high-permeability membranes and convective processes. It is not surprising, therefore, that the mortality associated with high-efficiency haemodiafiltration has been reported to be significantly lower as compared with that associated with other dialysis strategies [5]. An interesting alternative, however, to convective removal of PBUT could be the use of the high adsorptive capacity of some dialysis membranes, which deserves thorough investigation [6]. Most of the papers included in this issue are based upon the ’06 ERA-EDTA Congress Glasgow Industry Symposium: ‘The multiple aspects of middle-large molecular weight uremic toxins: state of the art’. The first three papers are related to uraemic pruritus taken here as an example of how an old and still unresolved clinical problem may be considered from a new point Correspondence and offprint requests to: Bruno Perrone, MD, Centre Hospitalier de Saintes, F-17100 Saintes, France. Email: perrone.bruno@wanadoo.fr Nephrol Dial Transplant (2007) 22 [Suppl 5]: v1–v2