Metabolic complications of total parenteral nutrition: Effects of a nutrition support service

Background: The effectiveness of Nutrition Support Services in optimizing parenteral nutrition has not been evaluated since the 1980s. Methods: We prospectively monitored medical and surgical patients on total parenteral nutrition (TPN) in a university hospital who did not receive Nutrition Support Service recommendations to compare the incidence of metabolic complications in 1979 (group 1, n = 100) with that in 1992 (group 2, n = 106). The Service provided automatic recommendations on a subsequent group of medical service patients (group 3, n = 128) and compared them with the patients in group 2 who were on the medical service (group 2B, n = 29). Results: Statistically significant changes between 1979 and 1992 included a decline in the incidence of hyperglycemia from 47% to 22% and in hypokalemia from 12% to 3% of surgical patients and an increase in hypomagnesemia from 0% to 23% of surgical patients and from 2% to 14% of medical patients. The incidence of hypophosphatemia remained >20% in both medical and surgical patients. Within 1992, the addition of automatic recommendations had little impact on metabolic abnormalities and was associated with slightly but insignificantly lower TPN costs (not counting Service personnel costs). Conclusions: Factors such as the general integration of parenteral nutrition into tertiary medical care, standard protocols and order forms, automatic Nutrition Support Service consultations in an affiliated hospital, and nutrition curricula may be responsible for the improvements seen since 1979. However, the addition of automatic Service consultation in 1992 had only a marginal effect on metabolic complications and costs of parenteral nutrition.