A rapid intravenous phosphate replacement protocol for critically ill patients.

OBJECTIVE There is a high incidence of hypophosphataemia in the critically ill. It is known that even moderate hypophosphataemia can produce a variety of adverse consequences. Many phosphate replacement regimens have been suggested, most are conservative and few validated in the critical care environment. We designed a study to determine the apparent volume of distribution of phosphate and quantify phosphate loss in critically ill patients following rapid phosphate infusion. With these data we sought to design a phosphate replacement protocol. METHODS A prospective clinical study of an infusion of 14.5 mmol of phosphate ions over 1 hour in seven patients with mild, moderate or severe hypophosphataemia was performed. Serum phosphate, total calcium, ionised calcium, urea, creatinine, urinary phosphate excretion and creatinine clearance were measured prior to phosphate administration. Urinary phosphate excretion and creatinine clearance and serum phosphate levels were followed for 24 hours. RESULTS The median phosphate increment was 0.65 mmol/L (0.17 - 0.85). The median initial volume of distribution was 0.45 L/kg (0.21 - 0.87). There was no reduction in ionised calcium or creatinine clearance. The renal phosphate threshold was reduced in all patients. CONCLUSIONS The apparent volume of distribution of phosphate in this group of critically ill patients is large. Urinary phosphate losses contributed to the development of hypophosphataemia in these patients. In this patient cohort, there were no adverse effects from phosphate infusion of 14.5 mmol/hr. A rational protocol based on the estimated volume of distribution and estimated urinary losses can be developed to achieve the rapid and sustained correction of hypophosphataemia in such patients.