Osmolality of elemental and semi‐elemental formulas supplemented with nonprotein energy supplements

Background Elemental and semi‐elemental formulas are used to feed infants with short bowel syndrome, who may not be able to tolerate feeds of more than 310 mOsm kg^?1. The present study aimed to measure the osmolality of elemental and semi‐elemental formulas at different concentrations, with and without the addition of nonprotein energy supplements. Methods The osmolality of one elemental and three semi‐elemental formulas was measured by the freezing point depression method at concentrations of 10, 12, 14 and 16 g per 100 mL, with and without 10% or 20% of additional calories, in the form of glucose polymers and medium chain triglycerides. Inter‐analysis and intra‐analysis coefficients of variation of the measurements were less than 3.9%. Results The mean osmolalities of formulas reconstituted up to 12 g per 100 mL did not exceed 305.3 mOsm kg^?1, even with added energy supplements. The mean osmolalities of formulas at 14 and 16 g per 100 mL, with or without added energy supplements varied between 205.8 and 421.6 mOsm kg^?1. Conclusions A comprehensive list of elemental and semi‐elemental formulas at different concentrations, enriched or not with calories, is made available. This will enable professionals to customize feeds with the optimum composition, without exceeding the osmolality suggested for infants with short bowel syndrome.     

Osmotic load from glucose polymers

Glucose polymer is a carbohydrate source with variable chain lengths of glucose units which may result in variable osmolality. The osmolality of two commercial glucose polymers was measured in reconstituted powder infant formulas, and the change in osmolality of infant milk formulas at the same increases in energy density (67 kcal/dL to 81 and 97 kcal/dL) from the use of additional milk powder or glucose polymers was compared. All samples were prepared from powders (to nearest 0.1 mg), and osmolality was measured by freezing point depression. For both glucose polymers the within-batch variability of the measured osmolality was less than 3.5%, and between-batch variability of the measured osmolality was less than 9.6%. The measured osmolality varies linearly with energy density (p less than 0.001) and was highest in infant formula reconstituted from milk powder alone. However, there exist significant differences in the measured osmolality between different glucose polymer preparations. At high energy densities (greater than or equal to 97 kcal/dL), infant milk formulas prepared with milk powder alone or with the addition of certain glucose polymer preparation may have high osmolality (greater than or equal to 450 mosm/kg) and theoretically predispose the infant to complications of hyperosmotic feeds.     

The Effect of Additives for Reflux or Dysphagia Management on Osmolality in Ready‐to‐Feed Preterm Formula: Practice Implications

Background: A common osmolality threshold for feedings is to stay <450 mOsm/kg for normal infants. Preterm formulas are frequently modified to improve growth, modify nutrition, and manage gastroesophageal reflux (GER) or dysphagia. Relationships between osmolality and additives to ready‐to‐feed preterm formulas are unclear. Our aims were to evaluate and compare the effects of caloric density, thickening agent recipes, and supplements to ready‐to‐feed preterm formula on osmolality. Methods: A freezing point osmometer was used to measure the osmolality of 47 preterm infant formula combinations with varying caloric densities (ready‐to‐feed 22 and 30 cal/oz), thickening agents (rice vs oatmeal cereal), thickener amounts (0.0, 0.5, 1.0, 1.5, 2.0, 2.5, and 3.0 tsp/oz), and combinations of supplements (saline, iron, vitamin D, or multivitamin). Ten samples per combination were tested using a 10‐μL pipette. Comparisons were made using analysis of variance and t‐tests for group and pair‐wise comparisons, respectively. Results: A total of 470 osmolality samples were analyzed: (1) raters had high agreement (r = 0.98; P < .001); (2) for every 0.5 tsp/oz of thickener, the osmolality increases by 30 mOsm/kg (P < .001); (3) osmolality was higher with the oatmeal (vs rice) thickening agent (P < .001); and (4) vitamin and electrolyte supplement combinations increase osmolality. Conclusions: Alteration of ready‐to‐feed preterm formulas may significantly increase osmolality and have unintended consequences. Caution and monitoring should be exercised with modifying ready‐to‐feed preterm formulas for regurgitation, rumination, GER, dysphagia, feeding intolerance, or emesis. This study supports the concept of achieving volume tolerance before further manipulation of additives.     

Gastric emptying of two whey-based formulas of different energy density and its clinical implication in children with volume intolerance

OBJECTIVE: Whey-based formulas have faster gastric emptying than casein-based formulas. Isoenergetic, isovolumic, whey-based formulas of different osmolarity and fat content empty in a similar manner. Will the gastric emptying of high and low energy density whey-based formulas be similar? DESIGN: We studied the gastric emptying rate of equal volumes of two whey-based formulas of different energy density (4.18 kJ/ml and 6.27 kJ/ml) and osmolality (270 and 450 mOsm/kg, respectively) in 10 children (4.5-12 y) with volume intolerance and resultant inability to gain weight. RESULTS: The two formulas had comparable gastric emptying rates at 30, 60, 90 and 120 min. Over a one month clinical trial, substitution of the lower energy density whey-based formula (no weight gain over 2 months) with an equal volume of the high energy density formula produced a mean weight gain of 1.17+/-0.5 kg per patient without change in tolerance. CONCLUSION: The higher density whey-based formula can safely substitute an equal volume of a lower energy density formula to produce weight gain without affecting tolerance. IMPLICATION: This provides an important intervention for increasing energy intake in children with volume intolerance or fluid restriction.     

Iron deficiency and anemia in iron-fortified formula and human milk-fed preterm infants until 6 months post-term

Purpose

An iron intake of >2 mg/kg/d is recommended for preterm infants. We hypothesized that human milk (HM)-fed preterm infants require iron supplementation after discharge, whereas iron-fortified formulae (IFF; 0.8–1.0 mg iron/100 ml) may provide sufficient dietary iron until 6 months post-term.

Methods

At term age, 3 and 6 months post-term, ferritin (μg/l) was measured in 92 IFF-fed infants (gestational age (median (interquartile range)) 30.7 (1.4) weeks, birth weight 1,375 (338) gram) and 46 HM-fed infants (gestational age 30.0 (1.7) weeks, birth weight 1,400 (571) gram). Iron intake (mg/kg/d) between term age and 6 months post-term was calculated.

Results

Iron was supplemented to 71.7 % of HM-fed and 83.7 % of IFF-fed infants between term age and 3 months post-term and to 13 % of HM-fed and 0 % of IFF-fed infants between 3 and 6 months post-term. IFF-fed infants had an iron intake from supplements and formula of 2.66 (1.22) mg/kg/d between term age and 3 months post-term and 1.19 (0.32) mg/kg/d between 3 and 6 months post-term. At 3 and 6 months post-term, the incidence of ferritin <12 μg/l was higher in HM-fed compared to IFF-fed infants (23.8 vs. 7.8 % and 26.3 vs. 9.5 %, P < 0.02).

Conclusion

This observational study demonstrates that ferritin <12 μg/l is more prevalent in HM-fed infants until 6 months post-term. This may be due to early cessation of additional iron supplementation. We speculate that additional iron supplementation is not necessary in preterm infants fed IFF (0.8–1.0 mg iron/100 ml), as they achieve ferritin ≥12 μg/l without additional iron supplements between 3 and 6 months post-term.     

Effects of glucose and glucose polymers on calcium absorption in healthy subjects

Glucose polymers have been shown to enhance intestinal calcium absorption in normal subjects as well as in patients with gastrointestinal disease. Glucose polymers are widely used as an energy source in enteral nutritional supplements and infant formulas. In this study, eight normal subjects underwent 47Ca absorption tests to compare the effect on calcium absorption of carbohydrate given orally as a simple carbohydrate (glucose) or a more complex carbohydrate (glucose polymers). Oral coadministration of glucose with calcium increased the efficiency of intestinal calcium absorption by 20% and coadministration of glucose with glucose polymers increased calcium absorption by 27%. This suggests that glucose, derived from glucose polymers by hydrolysis, may be responsible for the positive effect of polymers on calcium absorption. The potential positive effect of coadministration of carbohydrate on calcium bioavailability deserves considerations in the design of calcium supplementation strategies.     

Effect of a lactase preparation on lactose content and osmolality of preterm and term infant formulas

Lactose intolerance due to lactase deficiency often follows acute gastroenteritis. In such situations, a lactose-free formula may be indicated for preterm infants. Therefore, the effect of addition of lactase on the lactose content and osmolality of preterm and term infant formulas was studied. Lactose content of formulas at room temperature was decreased by approximately 50% 1 hour after addition of lactase. Concentration of lactose was reduced by 70% or more after 2 hours in all formulas. Because of the higher initial lactose concentration in term formulas, it took 24 hours to reach the same absolute lactose concentration (10 g/kg formula) found in preterm formulas after 2 hours. There was a moderate increase in osmolality in preterm formulas. The increase was greater in term formulas because of the greater initial concentration of lactose. The addition of lactase appears to be a suitable method for reduction of lactose content of preterm and term formulas, although the increase in osmolality of term formulas may preclude their clinical use.     

The effect of lactase and formula reconstitution on milk osmolality

These experiments investigated the reaction rate of lactase on milk lactose by measuring milk osmolality; and explored the effect of formula reconstitution on milk osmolality. The investigations measured milk osmolality with the Fiske Os, freezing-point osmometer. Lactase (Lactaid) incubated with pure lactose solutions established the validity of the method. Lactase was incubated for 24 hours with four reconstituted milk formulas (Milumil, and Cow and Gate Nutrilon Plus, Farley's First Milk, SMA Gold). Milk osmolality increased most rapidly in the first 4 hours after the addition of lactase. The lactase enzyme completed over 90% of the reaction within 12 hours. The milk osmolalities ranged from 487 to 591 mosm/kg after 24 hours with 2-4 drops of lactase in 240 ml of formula.A clinical guideline osmolality of 400 mosm/kg was reached in 240ml of formula at 1 to 12 hours depending on the dose of lactase. High milk osmolalities due to prolonged enzyme incubation, or high lactase doses could be reduced to around 400 mosm/kg by dilution of 240 ml of formula with an extra 60 ml of water. The initial osmolality of formula after reconstitution by paediatric nurses varied widely and usually exceeded the manufacturer's quoted osmolality. This initial osmolality was a further influence on the final osmolality reached after the addition of lactase. It is concluded that the recommended incubation time for Lactaid of 24 hours is unnecessary as lactase exerts the majority of its effect in less than 12 hours. Adjustment of Lactaid dose and incubation times will maintain milk formula osmolality within standard guidelines. Dilution with extra water will correct inadvertent high enzyme doses and prolonged incubation times. The normal method of reconstituting milk formulas from powder may be unreliable as the manufacturer's quoted osmolality was not reproduced when milk formulas were reconstituted by paediatric nurses.     

The effect of lactase and formula reconstitution on milk osmolality

These experiments investigated the reaction rate of lactase on milk lactose by measuring milk osmolality; and explored the effect of formula reconstitution on milk osmolality. The investigations measured milk osmolality with the Fiske Os, freezing-point osmometer. Lactase (Lactaid) incubated with pure lactose solutions established the validity of the method. Lactase was incubated for 24 hours with four reconstituted milk formulas (Milumil, and Cow and Gate Nutrilon Plus, Farley's First Milk, SMA Gold). Milk osmolality increased most rapidly in the first 4 hours after the addition of lactase. The lactase enzyme completed over 90% of the reaction within 12 hours. The milk osmolalities ranged from 487 to 591 mosm/kg after 24 hours with 2-4 drops of lactase in 240 ml of formula. A clinical guideline osmolality of 400 mosm/kg was reached in 240 ml of formula at 1 to 12 hours depending on the dose of lactase. High milk osmolalities due to prolonged enzyme incubation, or high lactase doses could be reduced to around 400 mosm/kg by dilution of 240 ml of formula with an extra 60 ml of water. The initial osmolality of formula after reconstitution by paediatric nurses varied widely and usually exceeded the manufacturer's quoted osmolality. This initial osmolality was a further influence on the final osmolality reached after the addition of lactase. It is concluded that the recommended incubation time for Lactaid of 24 hours is unnecessary as lactase exerts the majority of its effect in less than 12 hours. Adjustment of Lactaid dose and incubation times will maintain milk formula osmolality within standard guidelines. Dilution with extra water will correct inadvertent high enzyme doses and prolonged incubation times. The normal method of reconstituting milk formulas from powder may be unreliable as the manufacturer's quoted osmolality was not reproduced when milk formulas were reconstituted by paediatric nurses.     

The effects of high-energy feeding on energy balance and growth in infants with congenital heart disease and failure to thrive

Failure to thrive (FTT) in infants with congenital heart disease (CHD) can be attributed to their low energy intakes and high resting energy expenditures. Energy intake, energy expenditure and growth were studied in infants with CHD on normal formula feeds and then on feeds supplemented with glucose polymer to see whether supplementation improved energy retention and growth. Mean gross energy intakes increased by 31.7% on high-energy feeding and mean weight gain improved from 1.3 g/kg per d on control to 5.8 g/kg per d on high-energy feeding. Resting oxygen consumption (VO2 ml/kg per min) was not significantly different on the two feeding regimens, although respiratory quotient rose on high-energy feeding reflecting the increased carbohydrate intake. Estimated energy costs of growth on high-energy feeding fell within the previously described range for normal infants. It is recommended that infants with CHD known to be associated with FTT are fed on high-energy diets from the time of diagnosis in order to optimize growth.     

Effects of exogenous catecholamines on glucose and fat metabolism and on triglycerides in the rat liver during total parenteral nutrition

An experiment was undertaken to investigate the effects of a continuous infusion of catecholamines on glucose and fat metabolism as well as on nitrogen balance and the amount of triglycerides in the rat liver. The animals were nourished by total parenteral nutrition for 5 days and divided into six groups (n = 5 in each group) on the basis of nonprotein calories given with or without an infusion of catecholamines: group G received 100% of nonprotein calories with glucose, group F received 50% of nonprotein calories with glucose, and the remaining 50% with lipid emulsion, groups Epi-G and Epi-F received epinephrine (1 microgram/kg body weight/min) in addition to the same total parenteral nutrition solution as group G or F, and groups Nor-G and Nor-F received norepinephrine (1 microgram/kg/min) in a similar manner. Each group was administered the same number of total calories (252 cal/kg/day) and the same amount of nitrogen (1.49 g/kg/day). Nitrogen balance was better in group G than in group F. Under catecholamine infusion, there were no significant differences in nitrogen balance between groups Epi-G, Nor-G, Epi-F, and Nor-F, but this parameter improved significantly in group Nor-F in comparison to group F. Liver triglycerides was higher in groups Epi-G and Nor-G than in groups Epi-F and Nor-F. Liver triglycerides was directly related to the blood sugar level. These results indicate that nitrogen conservation is improved with lipid emulsion and that glucose rather than lipid plays a significant role in the genesis of fatty liver, when they are administered under catecholamine-induced stress.     

The influence of calorie source on water and sodium balances during intravenous refeeding of malnourished rabbits

34 New Zealand rabbits were starved until a mean weight loss of 15% to 20% was achieved and then were parenterally re-fed with four different total parenteral nutrition (TPN) formulae for 6 days. Regimens were isocaloric (85 kcal/kg/day) and isonitrogenous (0.52 gN/kg/day) with the following formulae: group S-GLU-70 (water 100 ml/kg/day, Na 7 mEq/day and 70 30 % of calories as glucose/lipids, n = 9), group S-GLU-50 (water 100 ml/kg/day, Na 7 mEq/day and 50 50 % of calories as glucose/lipids, n = 9), group L-GLU-30 (water 70 ml/kg/day, no sodium and 30 70 % of calories as glucose/lipids, n = 8) and group L-GLU-50 (water 70 ml/kg/day, no sodium and 50 50 % of calories as glucose/lipids, n = 8). The group S-GLU-70 showed the highest water and sodium balances (p = 0.001 respectively, ANOVA test). Group S-GLU-70 showed also a greater weight gain (p = 0.0001) and, although not statistically significant, the lowest albumin at the end of the TPN. Glucose load appears to be responsible for the rapid weight gain and positive water and sodium balances during TPN in depleted non-stressed animals.     

Net calcium absorption in premature infants: results of 103 metabolic balance studies.

Net calcium absorption was evaluated in 103 low-birth-weight preterm infants by a 72-h balance technique. At birth the infants had a mean (+/- SE) gestational age of 30.9 +/- 0.2 wk and weighed 1.43 +/- 0.03 kg. When tested 3 wk later, their net calcium absorption averaged 58 +/- 1% with an intake of 80 +/- 2 mg Ca.kg body wt-1.d-1. Of the 103 infants, 58 had been fed low-birth-weight formulas supplemented with vitamin D. The remainder received banked human milk, of whom 34 were supplemented with vitamin D and calcium; 11 infants received no supplementation. Calcium absorption in the four subgroups did not differ significantly, with neither vitamin D supplementation nor supplementation with vitamin D and calcium affecting percent absorption significantly. Net calcium absorption was a linear function of intake (40-130 mg Ca.kg body wt-1.d-1) with a zero intercept. Because vitamin D supplementation did not increase net calcium absorption, it is concluded that in preterm low-birth-weight infants calcium absorption proceeds by a nonsaturable route, with the transcellular, vitamin D-regulated mechanism not yet expressed.     

A simple equation to estimate the osmolarity of neonatal parenteral nutrition solutions

A predictive equation of osmolarity that correlates closely with the measured osmolality was determined. Taking into account that an osmometer is not available in most clinical settings, the proposed equation appears to provide a quick and simple osmolarity calculation of neonatal parenteral nutrition solutions. OBJECTIVE: We measured the osmolalities of neonatal parenteral nutrition (PN) solutions to determine if these values may be predicted by a simple equation for calculation of their osmolarity values. METHODS: The osmolalities of 101 consecutive different final PN admixtures, prepared for 36 neonates, were measured by the freezing point depression method. The respective intra-assay and interassay coefficients of variation were always <2.1%. Linear multivariate regression analysis was used to determine a predictive equation of osmolarity that correlates closely with the value of measured osmolality. RESULTS: The mean (SD) osmolality of the final PN admixtures was 749.7 (165.4) mOsm/kg. The best-fitted equation, with a coefficient of discrimination R2 = .95 (R2 = .90 for samples between 500 and 1000 mOsm/L) is osmolarity (mOsm/L) = (nitrogen x 0.8) + (glucose x 1.235) + (sodium x 2.25) + (phosphorus x 5.43)-50, with the concentration of components in mmol/L. Adapting the equation in our daily practice, using g/L for glucose and amino acids, mg/L for phosphorus, and mEq/L for sodium, the equation is osmolarity (mOsm/L) = (amino acid x 8) + (glucose x 7) + (sodium x 2) + (phosphorus x 0.2) - 50, with a similar R2. CONCLUSIONS: Taking into account that an osmometer is not available in most clinical settings, the proposed equation appears to provide a quick and simple osmolarity calculation of neonatal PN solutions, thus allowing more accurate decisions to be taken regarding the choice of route and rate of administration of PN solutions.     

Increased concentration of vasopressin in plasma of essential fatty acid-deficient rats

The effect of essential fatty acid deficiency (EFA-D) on the plasma concentration of arginine-vasopressin (AVP) and the urinary AVP excretion was investigated. Weanling rats were fed a fat-free diet (FF-rats). Control rats received the same diet in which 6% by wt. of sucrose was replaced by arachis oil. After 4–6 weeks of feeding, urine and plasma were analysed for AVP, osmolality, sodium and potassium. When compared to control rats FF-rats had decreased urine volume (6.0±1.6 ml/24 hr versus 11.7±3.2 ml/24 hr), increased urine osmolality (2409±691 mOsm/kg versus 1260±434 mOsm/kg), increased urinary AVP excretion (5.52±1.62 ng/24 hr versus 2.19±1.20 ng/hr and increased plasma AVP (2.55±0.45 pg/ml versus 1.95±0.25 pg/ml. There was no difference between the groups in plasma osmolality. The triene/tetraene ration was 0.78+0.15 (n=12) in total kidney lipids of the FF-rats. The triene/tetraene ratio was positively correlated with urinary AVP excretion. It is concluded that FF-rats had elevated plasma AVP concentration and disturbed water balance, both of which probably were provoked by an increased transepidermal water loss.     

The Henry M. Vars Award. The effect of lipid emulsions on reticuloendothelial system function in the injured animal

Use of intravenous lipid emulsions in trauma and sepsis still remains controversial. In order to examine the impact lipid emulsions have on host defense against bacterial infection during total parenteral nutrition (TPN), 56 male Sprague-Dawley rats underwent jugular cannulation and were randomly divided into three groups, each receiving one of three TPN regimens. All regimens delivered approximately 250 kcal/kg X body weight/day, of which 12.5 g were as amino acids. Group 1 received 100% of the nonprotein calories as glucose (AA + G). Group 2 was given 50% of the nonprotein calories as a longchain triglyceride emulsion (100% LCT). Group 3 received 50% of nonprotein calories as a mixed lipid system, composed of medium- and long-chain triglycerides (75% MCT/25% LCT). After 24 hr on intravenous nutrition, all animals received bilateral septic femur fractures and were continued on TPN for 3 days. On the last day, the level of bacteremia and the in vivo response to an intravenous challenge of 59Fe-labeled Escherichia coli were examined. Three days following the septic injury, animals given MCT as part of their lipid calories were not bacteremic, whereas the other groups had greater than 10(2) cfu/ml of blood. Animals receiving TPN with MCT sequestered a greater percentage of exogenously administered bacteria in the liver and sequestered less in the lung compared to animals given 100% LCT (p less than 0.05). From these data, we conclude that parenteral nutrition formulas where LCT has been partially replaced with MCT may better support host bactericidal capacity than similar regimens comprised of LCT as the sole lipid source.     

Effect of zinc supplementation on the dietary intake and weight gain of Bangladeshi children recovering from protein-energy malnutrition

Sixty severely malnourished children aged between 5 and 60 months were studied during nutritional rehabilitation. They all received a rice-based diet ad libitum plus vitamins and iron supplementation. Thirty children received zinc supplements (10 mg/kg/d for those weighing less than 6 kg and 50 mg daily for those over 6 kg) on a random basis. Zinc was started from the 15th hospital day when they were free of infection and continued for a period of 3 weeks. Both groups had a mean energy intake of 200 kcal/kg/d, but the majority of the supplemented children had a better rate of weight gain: 66 per cent of the supplemented compared with 33 per cent of the controls gained more than 10 g/kg body weight/d. Moreover, 76 per cent of the supplemented children compared with 23 per cent of the controls were over 90 per cent of Harvard median weight for height on discharge. It appears from this study that zinc supplementation promotes growth and enhances the rate of clinical recovery from severe PEM.     

Linseed and cod liver oil induce rapid growth in a 7-year-old girl with N-3- fatty acid deficiency

N-3 fatty acid deficiency is described in a 7-year-old girl who had been fed solely by gastric tube from the age of 3 years. She had received the same nutrient during the last 22 months, supplying 16.2% of calories from linoleate and 0.07% from alpha-linolenate. She weighed 9.5 kg at the age of 3, and 10.3 kg at the start of the study, the weight being constant the last 15 months. She was now supplemented daily with 1.2 ml of a linseed- and cod liver-oil mixture (5:1, v/v), thereby increasing her alpha-linolenate intake to 0.71% and total N-3 fatty acids to 0.74% of total energy. Two months later, her rate of weight gain was 0.43 kg/month. After 5 months, the fatty acid supplement was changed to 7.5 ml of cod liver oil daily, thereby changing alpha-linolenate and total N-3 acid intake to 0.10% and 1.3% of total energy, respectively. Her rate of weight gain now increased to 0.64 kg/month, and her length increased from 117 to 122 cm in 5 months. The results indicate that the optimal dietary requirement for N-3 fatty acids in this child is higher than 0.74%, and extrapolation estimated the optimal requirement to be 1.1%-1.2% of total calories. The results suggest that N-3 fatty acids are required for normal growth in man.     

Use of a mixture of medium-chain triglycerides and longchain triglycerides versus long-chain triglycerides in critically ill surgical patients: a randomized prospective double-blind study

Twenty critically-ill surgical patients who needed total parenteral nutrition were randomly enrolled in a double-blind study comparing two intravenous fat emulsions: one containing a mixture of 50% medium-chain triglycerides and 50% long-chain triglycerides and another containing 100% longchain triglycerides. The purpose of this study was to investigate metabolic and biochemical differences between both emulsions with special reference to liver enzymes. After a baseline period of 24 h with only glucose and NaCl infusion, the lipid emulsion was added continuously during 24 h over 5 days. The parenteral nutrition was administered in mixture bags containing amino-acids, glucose and lipids together. Two-thirds of the non-protein calories were administered as glucose 40% and one third as either long-chain triglycerides or a mixture of medium-chain triglycerides and long-chain triglycerides. The total amount of non-protein calories received was the measured energy expenditure during the baseline period plus 10% and was fixed during the study. Plasma substrate concentrations, energy expenditure, and nitrogen balance were determined and arterial blood samples were taken. No toxic effects or complications attributable to one of the two emulsions were observed. There was no significant difference in energy expenditure, nitrogen balance, liver function tests, carnitine, transferrin, pre-albumin, albumin, cholesterol, triglycerides and free fatty acids. The only parameter that showed a different pattern of reaction between the two emulsions was serum bilirubin concentration. In this study no evidence of any advantageous effect of a mixture of medium-chain triglycerides and long-chain triglycerides was seen.