Factors contributing to the development of hypophosphataemia when refeeding using parenteral nutrition

To identify individual attributes or risk factors which predispose to the development of refeeding hypophosphataemia in patients on parenteral nutrition (PN).

The Royal Surrey County Hospital (RSCH) a 530 bed, non-teaching Trust with a cancer centre, medical and surgical inpatients and intensive care unit (ICU). Subjects were recruited prospectively from all adult inpatients referred for initiation of PN.

Seventy patients (cases) with refeeding hypophosphataemia were matched with controls who had not experienced a fall in phosphate levels when commenced on PN. Their nutritional requirements, nutrition intake, and biochemical test results were compared and statistical analyses performed to show if any differences between cases and controls were due to chance.

Independent risk factors for developing refeeding hypophosphataemia were: significant malnutrition measured as a Nutrition Risk Screening (NRS) score of three or more; less than 12 mmols total phosphate in the first day's PN regimen; and an initial rate of infusion of PN of more than 70% of calculated requirements. In addition increasing amounts of non-lipid phosphate in the first day's PN regimen were found to be protective. Hypomagnesaemia prior to starting PN was non-significantly associated with refeeding hypophosphataemia. Other biochemical markers included in the study: albumin, calcium, C-reactive protein, glucose and urea, did not show an association. ICU, cancer and postoperative patients were not found to be more at risk.

Patients with a high NRS score prior to commencing nutrition support may be more at risk than others of refeeding hypophosphataemia. The first 24 h PN regimen should be run slowly providing less than 70% of calculated protein and calorie requirements but containing more than 12 mmol phosphate.