From malnutrition to refeeding during anorexia nervosa

Anorexia nervosa is one of the most common forms of malnutrition observed in Western society in individuals without physical diseases, with an average risk of mortality of 20% in a younger population aged between 15 and 25 years. It is characterised by an initial dramatic decrease in food intake that leads to profound depletion in muscle and fat mass. During the course of the disease, the resting energy expenditure decreases proportionally to the loss of lean body mass with a decrease in thyroid hormone secretion. The metabolic adaptation during anorexia nervosa is similar to that observed during starvation with a relative sparing of protein stores. After an initial weight loss, the total energy expenditure is similar to that in normal individuals, with a decrease in resting energy expenditure and an increased energy-related physical activity. At the end stage of wasting, however, physical activity dramatically decreases as well as energy intake. This metabolic adaptation of semi-starvation is impaired during refeeding with an increase in the thermic effect of food and a high risk of refeeding syndrome with severe hypophosphatemia.     

Ornithine alpha-ketoglutarate and glutamine supplementation during refeeding of food-deprived rats.

The aim of this study was to compare the efficiency of ornithine alpha-ketoglutarate (OKG) and glutamine supplementation in an experimental model of denutrition that provides well-characterized disturbances of amino acid patterns. Male Wistar rats (187 +/- 11 g; five in each group) were starved for 3 days and then refed for 7 days with an oral diet (192 kcal kg-1.day-1 and 2.25 g of nitrogen kg-1.day-1), supplemented with 0.19 g of nitrogen kg-1.day-1 in the form of OKG, glutamine, or casein (control group). Food deprivation induced a fall in most tissue amino acids, with the notable exception of muscle leucine and liver glutamate, which increased by 43% (p < .01), and 11% (p < .05), respectively. The main effect of OKG was seen in the viscera, with a normalization of most amino acid pools (including proline and branched-chain amino acids) in the small bowel and liver. The main effect of glutamine was observed in the muscle, with a normalization of the glutamine and leucine pools. We conclude that, in this model and with the doses used, OKG and glutamine act in different target tissues, ie, splanchnic areas and muscle, respectively.     

Influence of feeding conditions on slowing of gastric emptying by edible gums

The feeding of 4% guar or xanthan gum versus 4% cellulose in a final test meal was expected to slow the gastric emptying of accompanying nutrients, and the degree of this slowing was examined in rats trained to three feeding regimens. Gastric emptying of nutrient energy was calculated from the measured disappearance of fat and of either carbohydrate or total diet residue from the stomach. In experiment 1, rats were ad libitum fed a diet containing 4% cellulose for 6 1/2 days and, after 12 to 14 hours without food, were given a test meal containing either 4% guar or 4% cellulose. In experiment 2, a growth-restricting amount of 6% and 2% guar diets was fed as a single meal on each of 7 days, and the same diets were used for the final test meals. In experiment 3, rats ate 2-hour meals containing 4% and 0% guar gum, or 4% and 0% xanthan gum, on alternate days, and again the same diets were used for the test meal. The feeding of both xanthan and guar gums in a dry form did slow gastric emptying of nutrient energy, although the feeding of 6% versus 2% guar gum was without demonstrable effect (experiment 2). The slowing effect varied with past feeding conditions, being greater in experiment 1 than in experiment 3.     

Metabolic responses to prolonged fasting and subsequent refeeding in the pig

Metabolic responses associated with prolonged fasting and subsequent refeeding of pigs were investigated. Fasting for 14 or 28 days produced significant increases in serum levels of alanine, aspartic and glutamic acid in the three branched-chain amino acids. Glycine, serine and lysine levels were elevated after 28 days of fasting while the levels of histidine, methionine, threonine and phenylalanine were reduced. Fasting markedly stimulated hepatic and renal gluconeogenesis and the activity of the urea cycle enzymes. Fatty acid synthesis and glucose oxidation were virtually abolished in hepatic and adipose tissue in pigs subjected to a 14- or 28-day fast. After the first day of refeeding, the levels of amino acids returned to the control values. The activity of the hepatic urea cycle enzymes, fructose-1,6-diphosphatase and phosphoenolpyruvate carboxykinase remained elevated after the first day of refeeding but returned to the control levels thereafter. The activity of hepatic glucose-6-phosphate dehydrogenase, malic dehydrogenase and acetyl CoA carboxylase were slightly enhanced in pigs refed for 4 and 8 days. The activity of these enzymes in adipose tissue was enhanced 8 days after refeeding. Hepatic synthesis of fatty acids from glucose was slightly stimulated in refed pigs on days 4 and 8 but returned to control values on day 16. Refeeding did not enhance glucose incorporation into fatty acids in adipose tissue above the values observed in fed controls.     

Hepatic glutamine metabolism during endotoxemia in neonatal rats

OBJECTIVES: The liver plays a central role during endotoxemia. We investigated the biochemical changes that occur in neonatal liver during early stages of endotoxemia. METHODS: Twenty neonatal rats (10 to 15 d; n = 10/group) were studied. Endotoxemic rats received intraperitoneal injections of 300 microg/kg of 12.5 mg/L of lipopolysaccharide and control rats received isovolemic normal saline. Two hours after injection, all lipopolysaccharide-injected animals exhibited signs of endotoxemia. Livers were removed and extracted into 12% perchloric acid. 1H and 31P magnetic resonance spectroscopy measured hepatic levels of glutamine, glutamate, alanine, lactate, glucose, beta-hydroxybutyrate, adenosine triphosphate, and adenosine diphosphate. Unpaired t test compared groups. RESULTS: No mortality occurred during the first 2 h after injection. Endotoxemia significantly decreased hepatic levels of glutamine (P < 0.001), glucose (P = 0.047), and beta-hydroxybutyrate (P < 0.001). There was no difference in hepatic levels of glutamate (P = 0.050), alanine (P = 0.165), lactate (P = 0.478), adenosine triphosphate (P = 0.165), and adenosine diphosphate (P = 0.136) between groups. CONCLUSIONS: Early endotoxemia caused significant changes in the hepatic metabolism of glutamine, glucose, and beta-hydroxybutyrate. These findings increase our understanding of the pathophysiology of neonatal endotoxemia.     

Hepatic steatosis and serum very low density lipoproteins during two types of protein malnutrition followed by balanced refeeding

The relation of serum very low density lipoproteins (VLDL) to hepatic steatosis was studied during protein malnutrition followed by refeeding of a balanced diet in growing rats. A control group was fed a balanced diet containing 15% casein for 42 days. Two depleted groups were fed low protein diets containing 2% casein (group C) or 5% gluten (group GI) (protein malnutrition phase) for 28 days and then were fed the balanced diet for 14 days (refeeding phase). The concentrations of phospholipids and proteins in both liver and serum VLDL were decreased during protein malnutrition, whereas triacylglycerols, unesterified cholesterol, and cholesteryl esters were higher in the liver and lower in the serum VLDL in the C and GI groups compared with the control group. There was a significant inverse relation between serum VLDL apolipoproteins and liver triacylglycerols on the one hand and between serum VLDL triacylglycerols and liver triacylglycerols on the other hand, in both depleted groups, although this relation was less important in the GI group. The major fatty acid levels of liver triacylglycerols were negatively correlated with those of serum VLDL during protein malnutrition. Our results show that in spite of a similar fatty acid intake, protein malnutrition involved an important decrease in essential fatty acids in VLDL triacylglycerols and phospholipids. Moreover, triacylglycerol accumulation was accompanied by increases in unesterfied cholesterol and cholesteryl esters in the liver of rats fed low protein diets, especially with 5% gluten. Hence, the hepatic steatosis was not entirely attributable to impaired transport of triacylglycerols by VLDL.     

Response of adult lean and obese female Zucker rats to food restriction and refeeding

Lean and obese female Zucker rats were either fed ad libitum (ad libitum-fed lean and ad libitum-fed obese), food-restricted (restricted lean and restricted obese) at 50% of ad libitum intake for 3 weeks from 19–20 weeks of age, or food-restricted and then refed ad libitum for 1 week (restricted-refed lean and restricted-refed obese). Following food restriction, body weights of restricted rats were significantly lower than ad libitum-fed rats within genotype. Body weights of restricted-refed rats were not different from either ad libitum-fed or restricted rats within genotype. Restricted-refed lean rats returned to their previous ad libitum food intake, whereas restricted-refed obese rats ate significantly more food. Both restricted and restricted-refed lean rats had lowered serum insulin levels compared to ad libitum-fed lean rats, whereas there was no effect on serum insulin levels in obese rats. Liver weights and hepatocyte conversion of glucose to fatty acids, glyceride-glycerol and CO₂ were not affected by food restriction or refeeding in lean rats. Among obese rats, restricted obese rats had the smallest liver weights, and restricted-refed obese rats had the highest. Restricted-refed obese rats had greater rates of hepatic glucose metabolism compared to all other groups. Ad libitum-fed lean and restricted-refed lean rats had similar fat pad weights that were significantly greater than those of restricted lean rats. Dietary intervention had no effect on fat pad weight in obese rats. There was no effect of food restriction or refeeding on adipocyte glucose metabolism except for higher glucose conversion to CO₂ for restricted lean rats in comparison to values for all other groups. These results demonstrate that body weight changes of adult female obese rats in response to food restriction and refeeding are similar to those of lean rats; but the effects of liver and adipose tissue weights are different.     

Management of patients during hunger strike and refeeding phase

ObjectiveHunger strikers resuming nutritional intake may develop a life-threatening refeeding syndrome (RFS). Consequently, hunger strikers represent a core challenge for the medical staff. The objective of the study was to test the effectiveness and safety of evidence-based recommendations for prevention and management of RFS during the refeeding phase.MethodsThis was a retrospective, observational data analysis of 37 consecutive, unselected cases of prisoners on a hunger strike during a 5-y period. The sample consisted of 37 cases representing 33 individual patients.ResultsIn seven cases (18.9%), the hunger strike was continued during the hospital stay, in 16 episodes (43.2%) cessation of the hunger strike occurred immediately after admission to the security ward, and in 14 episodes (37.9%) during hospital stay. In the refeed cases (n = 30), nutritional replenishment occurred orally, and in 25 (83.3%) micronutrients substitutions were made based on the recommendations. The gradual refeeding with fluid restriction occurred over 10 d. Uncomplicated dyselectrolytemia was documented in 12 cases (40%) within the refeeding phase. One case (3.3%) presented bilateral ankle edemas as a clinical manifestation of moderate RFS. Intensive medical treatment was not necessary and none of the patients died. Seven episodes of continued hunger strike were observed during the entire hospital stay without medical complications.ConclusionsOur data suggested that seriousness and rate of medical complications during the refeeding phase can be kept at a minimum in a hunger strike population. This study supported use of recommendations to optimize risk management and to improve treatment quality and patient safety in this vulnerable population.     

Restoration of body energy reserves during refeeding in rats is dependent on both the intensity of energy restriction and the metabolic status at the onset of refeeding [corrected

During starvation, after a short dynamic period of adaptation (phase I), a metabolic steady state is reached in which proteins are spared and lipids provide most of the energy expended [phase II (P2)]. However, protein breakdown increases dramatically once a lower threshold of body lipids is reached [phase III (P3)]. Body composition, energy intake, energy expenditure, and energy efficiency were determined in 8 groups of rats (fed, food-deprived up to P2 or P3 of starvation and refed for 3 d, 7 d, or until body mass restoration) to determine whether the kinetics of lipid and/or protein reserve recovery may be slowed down when refeeding occurs after the lipid threshold has been reached. Despite larger losses, P3 refed rats restored their body reserves as efficiently as those refed in P2. Whatever the nutritional status at the onset of refeeding, rehydration occurred first and hyperphagia played a more important role than hypometabolism in the restoration of the lost reserves. However, the pattern of body component gains was different during early refeeding. In P3 refed rats, body lipids were restored preferentially by significant contribution from endogenous lipid production. Thus, the extent of lipid depletion has important consequences for the restoration pattern of the body reserves. It depends not only on the intensity of the energy restriction (partial or total) as already demonstrated but also on the metabolic status at the onset of refeeding. These results may have significant implications on the way refeeding should be conducted after severe energy depletion.     

Response of adult lean and obese female Zucker rats to intermittent food restriction/refeeding

Previously, it was found that lean and obese Zucker rats (9-15 wk of age) responded differently to the first of four cycles of food restriction/refeeding. In later cycles, they responded similarly. The present study was undertaken to determine if this finding was due to age, adaptation to the intervention or the obesity. Adult (35-wk-old) lean and obese rats were classified into four groups, ad libitum-fed lean and obese and food-restricted lean and obese. Food-restricted rats underwent four 3-wk periods when they were fed 50% of their ad libitum intake, each followed by a 3-wk period of ad libitum refeeding. Food-restricted rats lost and regained sufficient weight in each cycle to weigh a similar amount as their ad libitum-fed groups by the end of each refeeding period. In lean rats, there were no permanent effects of this intervention except for a 25% reduction in carbohydrate intake. Similar results were found in obese rats, although they did have significantly lower retroperitoneal fat pad weight and serum triacylglycerol levels than ad libitum-fed obese rats at the end of the experiment. These results indicate that lean and obese adult rats respond to each food restriction/refeeding cycle in a similar manner. Results in the earlier experiment would appear to be due both to age and genotype.     

Cocaine-induced cardiovascular responses in rats during acute ethanol withdrawal.

The effects of cocaine administration during acute ethanol withdrawal on both the cardiovascular system and cocaine pharmacokinetics are unclear. This study demonstrated differences in the cardiovascular effects of i.v.-administered cocaine during acute ethanol withdrawal in awake, freely moving rats. The altered responses to cocaine while in acute ethanol withdrawal compared to control animals included: enhanced increases in mean arterial pressure and systemic vascular resistance, attenuated heart rate decreases, and enhanced cardiac index and stroke volume decreases. These results may suggest that acute ethanol withdrawal disrupts myocardial contractility when the myocardium is subjected to a large increase in blood pressure. Serial arterial blood sampling in additional groups of rats were done to assess plasma cocaine concentrations and to confirm the absence of ethanol in the blood. Plasma cocaine concentrations were not effected by acute ethanol withdrawal. These results indicate that the altered cardiovascular responses to cocaine during acute ethanol withdrawal were not a result of differences in cocaine plasma concentrations.     

Subcellular location of phosphoenolpyruvate carboxykinase in hepatocytes from fed and starved rats

To evaluate published indications that about 25% of the gluconeogenic enzyme, phosphoenolpyruvate carboxykinase (PEPCK), is located in mitochondria of adult rat liver, cell fractionations were conducted with hepatocytes isolated from rats that were fed ad libitum or starved for 2 days. Hepatocytes were exposed to digitonin for 10 s, and the released materials were separated from residual cell structures by centrifugation through a layer of brominated hydrocarbon. In addition to PEPCK, activities of 9 other enzymes were measured in the untreated cells and with good recovery in the two fractions obtained with digitonin treatment. By comparison with the release of marker enzymes for the cytosol and mitochondria, the subcellular distribution of PEPCK was determined. With cells from either fed or 2-day-starved rats, this enzyme was released exactly like lactate dehydrogenase and within 2-3% of phosphoglycerate kinase and pyruvate kinase. These results indicate that, even after induction by starvation, at least 97% of PEPCK activity is located in the cytosol of rat liver.