Effect of prolonged bicarbonate administration on plasma potassium in terminal renal failure

Abstract

In hemodialysis patients with hyperkalemia, i.v. sodium bicarbonate has recently been found to be ineffective in lowering plasma potassium within one hour. In the present study the effect of a prolonged bicarbonate infusion on plasma potassium was investigated. Twelve patients with terminal renal failure who were on hemodialysis were infused i.v. with 8.4% sodium bicarbonate (4 mmol/min) for one hour and with 1.4% (0.5 mmol/min) for five hours (total amount 390 mmol). Plasma bicarbonate rose from 17.5 at baseline to 28.4 and 29.6 mmol/liter, and blood pH from 7.32 to 7.46 and 7.48 at one and six hours, respectively. Plasma potassium did not change significantly after one and two hours (6.04 at baseline, 5.91 and 5.77 mmol/liter, respectively). Only at four and six hours did a moderate decline to 5.44 (P < 0.05) and to 5.30 (P < 0.01) occur, of which approximately half was calculated to be due to ECF volume expansion. However, no change or a very moderate decrease was observed in three patients even after six hours (+0.19, -0.32, -0.33 mmol/liter). Five patients with higher baseline plasma potassium (6.15 to 8.15 mmol/liter) behaved like seven with lower levels (5.25 to 5.87 mmol/liter). Tented T-waves in the ECG of seven patients disappeared after one hour only in one patient. Plasma aldosterone, norepinephrine and epinephrine were normal to elevated before and tended to fall during i.v. bicarbonate. Plasma dopamine and insulin were in the normal range. It is concluded that in patients with terminal renal failure maintained on hemodialysis and with sufficient levels of potassium regulating hormones, i.v. bicarbonate was ineffective as a treatment of hyperkalemia for up to two hours; after four and six hours it was accompanied by a modest fall in plasma potassium which, however, was not quite reliable and was obtained at the cost of a considerable sodium-volume load.