Comparison Between Premixed and Compounded Parenteral Nutrition Solutions in Hospitalized Patients Requiring Parenteral Nutrition

Rationale: Parenteral nutrition (PN) may be provided through compounded or premixed solutions. Objective: To determine the proportion of stable custom‐compounded PN prescriptions that would fit within a 20% deviance of an existing premixed PN solution. Methods: A retrospective study design was used. Inpatients who received PN in non–critical care units in the preceding year were screened for eligibility. Results are reported descriptively as means (95% confidence intervals) and proportions. Results: We reviewed 97 PN prescriptions that met inclusion criteria. Stable hospital PN prescriptions compared with the reference premixed prescription provided 1838 (1777–1898) vs 1843 (1781–1905) kcal/d, P = .43; dextrose, 266 (254–277) vs 225 (216–234) g/d, P < .001; amino acids, 100 (95.9–104) vs 95.2 (91.7–98.7) g/d, P < .001; and lipids, 53.2 (51.3–55.1) vs 76.5 (73.8–79.2) g/d, P < .001. Fifty‐eight of 97 (59.8%) matched for 2 of 3 macronutrients. Hospital compared with premixed lipid was lower 53.6 (43–64.2) g/d vs 75.5 (60.5–90.5) g/d, P < .001. Electrolytes differed between hospital and premixed solutions: sodium, 98.6 (95.0–102) vs 66.9 (64.6–69.9) mmol/L, P < .001; potassium, 93.7 (89.0–98.3) vs 57.4 (55.4–59.4) mmol/L, P < .001; and magnesium, 5.4 (4.8–5.4) vs 7.6 (7.4–7.9) mmol/L. Conclusions: Calories and protein were remarkably similar, but dextrose, lipid, and electrolytes differed between hospital PN and the reference premixed prescription. We believe that there may be a role for premixed solutions in quaternary centers in stable non–critically ill patients.