Initiation of dialysis should be timely: neither early nor late.

Over the last decade, individuals with end-stage renal disease (ESRD) in the United States are starting maintenance dialysis therapy at progressively higher estimated glomerular filtration rate (eGFR). Moreover, several observational studies have demonstrated an association of a higher risk of death with higher serum creatinine-based estimates of GFR at the time of initiation of dialysis. In contrast, studies in which renal function has been measured by timed urinary collection show either a lower risk of death or no significant association with higher GFR at the time of initiation of dialysis. There are numerous potential sources of bias in such observational studies, particularly in those that use serum creatinine-based eGFR. The only randomized controlled clinical trial to have examined this question did not demonstrate either benefit or harm with initiation of dialysis at higher level of renal function. Thus, the data to date suggest that eGFR should not be the sole consideration when assessing the need for initiating maintenance dialysis in patients with advanced chronic kidney disease. Given the high societal costs of starting renal replacement therapy earlier in the course of the disease, these considerations also suggest that dialysis can be safely be postponed in otherwise asymptomatic individuals with advanced chronic kidney disease. By the same token, dialysis should not be denied to individuals who could clearly benefit from renal replacement therapy simply because the GFR is too high (viz., volume overload, refractory hyperkalemia). Finally, there is a compelling need to reexamine the symptoms that could be attributed to uremia and clearly improve upon initiation of dialysis to better guide clinical decision-making.