OBJECTIVE： To provide reference for rational use of parenteral nutrition.  METHODS： Combining with the patients’ nutritional status， clinical diagnosis and lab examination， 5 770 parenteral nutrition medical orders were collected from our hospital during May 2015-Apr. 2016， and then analyzed in respects of the stability of drug use， rationality of drug selection and  pharmacoeconomics. RESULTS： The qualified stability of nutrient solution in 5 770 parenteral nutrition medical orders accounted for 89.46％. The rates of reasonable drug selection in patients with abnormal liver function， renal function， hyperemesis gravidarum， pancreatitis， complete fasting， long term parenteral nutrition （＞14 d） and triacylglycerol ＞4.5 mmol/L were 76.19％， 63.16％， 75.00％， 78.95％， 65.95％， 64.06％ and 50.00％， respectively.  Four kinds of drugs were with same effect and dose， price difference between 4-34 times， the rate of inappropriate drug selection ranged 7.58％-61.64％. CONCLUSIONS： Parenteral nutrition design need to not only consider about nutritional status， clinical diagnosis and lab examination， but also take into account high efficiency， safety， economy and the stability of whole nutrient solution.