A need for a paradigm shift in focus: From Kt/V urea to appropriate removal of sodium (the ignored uremic toxin).

Hemodialysis for chronic renal failure was introduced and developed in Seattle, WA, in the 1960s. Using Kiil dialyzers, weekly dialysis time and frequency were established to be about 30 hours on 3 time weekly dialysis. This dialysis time and frequency was associated with 10% yearly mortality in the United States in 1970s. Later in 1970s, newer and more efficient dialyzers were developed and it was felt that dialysis time could be shortened. An additional incentive to shorten dialysis was felt to be lower cost and higher convenience. Additional support for shortening dialysis time was provided by a randomized prospective trial performed by National Cooperative Dialysis Study (NCDS). This study committed a Type II statistical error rejecting the time of dialysis as an important factor in determining the quality of dialysis. This study also provided the basis for the establishment of the Kt/Vurea index as a measure of dialysis adequacy. This index having been established in a sacrosanct randomized controlled trial (RCT), was readily accepted by the HD community, and led to shorter dialysis, and higher mortality in the United States. Kt/Vurea is a poor measure of dialysis quality because it combines three unrelated variables into a single formula. These variables influence the clinical status of the patient independent of each other. It is impossible to compensate short dialysis duration (t) with the increased clearance of urea (K), because the tolerance of ultrafiltration depends on the plasma-refilling rate, which has nothing in common with urea clearance. Later, another RCT (the HEMO study) committed a Type III statistical error by asking the wrong research question, thus not yielding any valuable results. Fortunately, it did not lead to deterioration of dialysis outcomes in the United States. The third RCT in this field ("in-center hemodialysis 6 times per week versus 3 times per week") did not bring forth any valuable results, but at least confirmed what was already known. The fourth such trial ("The effects of frequent nocturnal home hemodialysis") too did not show any positive results primarily due to significant subject recruitment issues leading to inappropriate selection of patients. Comparison of the value of peritoneal dialysis and HD in RCTs could not be completed because of recruitment problems. Randomized controlled trials have therefore failed to yield any meaningful information in the area of dose and or frequency of hemodialysis.