Adequacy and prescription

The older KDOQI formulation gave a single-pool Kt/V above 1.2, or a urea reduction ratio of at least 65%, as the minimum adequate dose for thrice-weekly haemodialysis, with target values of a single-pool Kt/V of 1.4 and a URR of around 70% (KDOQI Hemodialysis Adequacy, 2006).

The two are not interchangeable, and treating them as though they are is a genuine error rather than a pedantic one. URR takes no account of the effect of ultrafiltration on solute removal, and none of variable treatment time. Two patients with an identical URR can have had materially different sessions. URR survives largely because it is trivially easy to calculate and therefore makes a serviceable quality metric, and it remains a national audit measure in some systems, including UK Renal Registry reporting.

Use URR as an audit signal. Do not use it as a substitute for Kt/V, and do not convert between them as though the mapping were clean.

The two Frequent Hemodialysis Network trials are the cleanest evidence on frequency, and they do not point the same way, which is exactly why both should be quoted.

The FHN Daily Trial randomised 245 patients to six versus three in-centre sessions per week for 12 months. It found benefit on both co-primary composite outcomes: death or increase in left ventricular mass (HR 0.61, 95% CI 0.46 to 0.82), and death or decrease in the physical-health composite score (HR 0.70, 95% CI 0.53 to 0.92), with better blood pressure and phosphate control. It also found an increase in vascular access procedures in the frequent-dialysis arm (FHN Daily Trial, New England Journal of Medicine, 2010).

The FHN Nocturnal Trial, with 87 patients, found no significant effect on either co-primary outcome, although blood pressure and phosphate control again improved (FHN Nocturnal Trial, Kidney International, 2011).

The honest reading: more frequent dialysis is probably better for surrogate outcomes, symptoms, blood pressure and phosphate, the mortality evidence is not conclusive, and the access takes more punishment. That is a trade-off to present to a patient, not a recommendation to impose on one.

What follows is the range that unit formularies generally work within, not a prescription. Dialysate composition is individualised, formularies differ, and the decisions below belong to the treating team.

The potassium bath deserves its own paragraph because it is the dialysate decision most likely to kill someone, and because practice varies widely without much acknowledgement that it does.

Low-potassium baths, at 1.0 to 2.0 mmol/L, increase the intradialytic potassium gradient, and have been associated with arrhythmia and sudden cardiac death, particularly in patients who were not hyperkalaemic before the session. Conversely, in patients with a pre-dialysis potassium of 5.0 mEq/L or above, a low-potassium dialysate has been associated with longer survival (published cohort analyses, including PLOS One, 2015).

The Monitoring in Dialysis study, which used implantable loop recorders, found a reduction in clinically significant arrhythmias, including bradycardia, ventricular tachycardia and asystole, with a 3.0 mmol/L bath compared with a 2.0 mmol/L bath (Monitoring in Dialysis study, Kidney International, 2018).

This remains contested and the cohort data are observational. The direction of travel is towards avoiding very low potassium baths in patients who are not hyperkalaemic before dialysis. It is not a settled question, and the decision belongs to the treating team and the unit protocol.