840; one-way ANOVA). Data are mean Crizotinib c-Met and SD from four experiments. *P < 0.05 versus saline (Holm-Sidak method). ANOVA, analysis of variance; SD, standard deviation.Click here for file(23K, PDF)Additional File 4:Effects of a highly toxic dose of metformin on red blood cell lactate production. Red blood cells from healthy donors were incubated with either saline (white bar) or metformin diluted in saline (16,600 mg/L) (black bars). Lactate levels were measured every 24 hours, up to 72 hours (P = 0.927; two-way repeated measures ANOVA on ranks). Data are mean and SD, from three experiments. ANOVA, analysis of variance; SD, standard deviation.Click here for file(14K, PDF)NotesSee related commentary by Orban et al., http://ccforum.

com/content/16/5/164AcknowledgementsPreliminary results have been presented in abstract form at the 21st SMART Symposium (Italy, 2010) and 31st ISICEM Symposium (Belgium, 2011).This study was supported in part by an Italian grant provided by Fondazione Fiera di Milano for Translational and Competitive Research (2007, Luciano Gattinoni).
Marked malnutrition in the critically ill is associated with increased morbidity [1,2]. While feeding by the nasogastric route is preferred, approximately 50% of critically ill patients fail to meet their caloric needs using this approach [3,4]. When this occurs, nutrient is frequently delivered directly into the small intestine, bypassing the stomach [5], in the belief that this will increase caloric delivery and thereby optimise nutritional therapy, leading to improved outcomes [6].

However, data that have evaluated caloric intake during either intragastric or post-pyloric delivery are inconsistent, which may relate to the time taken in placing post-pyloric feeding [7-11]. Moreover, the premise that small intestinal feeding will increase absorption has not been tested.We have reported that glucose absorption is markedly impaired in the critically ill, both following gastric and small intestinal nutrient administration [12,13]. Glucose is not absorbed within the stomach, and so absorption will be limited both by the rate of gastric emptying. However, we recently observed that even when administered directly into the small intestine, glucose absorption is markedly diminished in critical illness [12].

This latter finding leads to the hypotheses that factors distal to the pylorus impair glucose absorption in the critically ill, so that post-pyloric Brefeldin_A nutrient delivery may not increase nutrient absorption when compared to intragastric delivery. If post-pyloric feeding does not increase nutrient absorption, it is unlikely to improve clinical outcomes.Hyperglycaemia occurs frequently in critically ill patients and is detrimental to patient outcomes [14]. Furthermore, variability in glycaemia may be just as, or even more, harmful than elevated mean glucose concentrations [15].