Dialysis Should Be Started When Absolutely Necessary, Not Early and Incrementally.

To the Editor:

The commentary by Obi and Kalantar–Zadeh and clinical paper by Chin et al. on incremental dialysis deserves further scrutiny. The first issue is which patients, if any, are “optimal” for the incremental dialysis approach? Patients who are dialyzed early, at an estimated glomerular filtration rate (EGFR) >10 ml/min per 1.73 m2, may not have a better survival with incremental dialysis versus waiting to initiate standard dialysis at lower eGFR levels. Increasing urea clearance with 3 times per week dialysis, although used for assessing dialysis adequacy, has not been shown to have a survival benefit. Thus, the incremental increase in small molecule/urea clearance above endogenous renal clearance, does not justify starting dialysis in these patients with significant residual renal function, especially when considering the potential harms of dialysis. In contrast, maximizing diuretic therapy for patients with intractable fluid overload may be a good approach to delay dialysis initiation. Patients in whom diuretic management fails are not candidates for twice weekly incremental dialysis with limited weekly ultrafiltration, but may be appropriate candidates for conventional hemodialysis with an “early start” in some cases. Nephrologists who want to consider twice weekly hemodialysis for palliative care need to consider the high 3- and 6-month mortality rates in many of these high-risk patients. Twice weekly hemodialysis may be used in 25% of patients dialyzed in China, a country where few patients initiate dialysis early. This approach makes good sense in countries with limited resources for dialysis and may have the added benefit of preserving residual endogenous renal function with its potential survival benefit.