Whenever possible, intravenous dextrose infusions were avoided. Insulin was infused to maintain a target blood glucose level of 80�C110mg/dL. All subjects received enteral and/or parenteral feeding, as part of their standard treatment. Feeding inhibitor order us was provided according to a dedicated nutritional care plan. The used protocol closely reflected current evidence-based, easy-to-use feeding algorithms indicating amount, composition and route of delivery.Gender, age, weight, height, and type of pathology were recorded at study entry. For obese patients, optimal caloric intake was calculated for a theoretical weight corresponding to a body mass index of 30kg/m2 [13]. Total caloric need (N, kcal/d) was assessed by multiplying basal energy expenditure, calculated with a modified Harris-Benedict equation as follows: male: resting energy expenditure (REE) (kcal/d) = 66.
47 + 13.75 (wt) + 5.003 (ht) ? 6.755 (a) (years); female: REE (kcal/d) = 655.1 + 9.563 (wt) + 1.850 (ht) ? 4.676 (a) (years), where wt denotes weight, ht represents height and a refers to age [14], which was, daily adjusted for weight and stress. For uncomplicated and complicated surgery, a stress factor of, respectively, 1.1 and 1.3 was used. Fractures and polytrauma were given, respectively, 1.1 and 1.3. Patients with an uncomplicated infection received a correction factor of 1.1, but sepsis was attributed 1.3.Attending ICU physicians, unaware of the study, based their daily caloric prescriptions on the expert-recommended 25kcal/kg/d regimen [15]. For seven consecutive days, prescriptions (P, kcal/d) and effective intake (I, kcal/d) were recorded.
A dedicated nutrition team measured caloric requirements with the “stress-adapted” Harris-Benedict formulas and estimated correctness of the prescription by calculating the P/N ratio. The accuracy of translating a prescription into really administered feeding was assessed by the I/P ratio. Finally, the I/N ratio was calculated to compare the amount of delivered calories with the theoretical caloric need. The latter was set at 100%, and both prescriptions and actual feeding were expressed as proportional to this percentage. A prescription was considered to be adequate when it covered 90 to 110% of total caloric need. Prescriptions not attaining 90% were considered to be ��underestimated�� whilst those exceeding 110% were said to be ��overestimated.��Statistical analysis used SPSS 12.0 for Windows (Chicago, IL, USA). Results were expressed as means �� standard deviation and medians (range). Means between groups were compared with the Student’s t-test. Statistical significance was accepted at a P value < 0.05.3. Results579 patients were admitted to the ICU during the study Carfilzomib period. 231 subjects were mechanically ventilated.