Practice-based validation of calcium and phosphorus solubility limits for pediatric parenteral nutrition solutions

In an effort to maximize the precipitation-free delivery of calcium and phosphorus to neonates, Fitzgerald and MacKay published in 1986 the results of empirical determination of calcium-phosphate saturation curves for a number of parenteral nutrition (PN) solutions. The saturation curves generated from these investigations have been used to formulate thousands of PN solutions. The curves were developed testing only calcium and phosphate without other components added to PN solutions. The authors reviewed 38,019 PN orders from 2007-2010 and plotted the calcium and phosphate concentrations for each solution in relation to the published curves to assess the practical validity of the curves. The solutions reviewed were compounded using standard weight ranges for electrolytes, trace minerals, and vitamins. The solutions were evaluated for precipitation using standards for visual compatibility against a black and white background. There were no visual precipitates found in the 38,019 PN solutions. All calcium and phosphorus concentrations plotted below the precipitation limits predicted by the published curves despite a large range of concentrations of electrolytes and minerals. There has always been concern about extrapolating data from solubility curves that were developed empirically from a limited number of test solutions based on the few variables of calcium, phosphorus, amino acid concentration, and presence of cysteine HCl and/or fat emulsion. This experience validates the calcium and phosphorus solubility limits represented by published curves. Moreover, the findings support the concept that principal variables governing calcium and phosphorus precipitation in PN solutions are calcium, phosphorus, amino acid concentrations, temperature, and pH.