There are physical, mental, social and environmental changes which take place with ageing; for example, decreased physical activity, increase in body fat, decrease in lean body mass and consequently decreased energy intake may be associated with physiological functions that affect metabolism, nutrient intake, physical activity and risk of disease. There are now many studies which have found that undernutrition is prevalent and often unrecognized in patients admitted to hospitals and institutions. There is also evidence which links protein-energy undernutrition or its markers with clinical outcomes in acute and non-acute hospital settings and that nutritional supplements can improve outcomes in some of these settings. However, most clinically-available nutrition screening instruments lack sensitivity and specificity, and abnormal nutritional indicators may simply reflect effects of age, functional disability, or severe underlying disease. Thus, causal relationship cannot be assumed without a sufficiently powerful intervention study which adequately adjusts for the effects of non-nutritional factors, such as the number and severity of co-morbid conditions on clinical outcome.

Human milk was collected from women in their 10th–14th weeks of lactation, and was analysed for amino acids. Corrections were made for losses of amino acids which were presumed to occur during acid hydrolysis, using a non-linear mathematical model that describes the simultaneous processes of amino acid yield and decay. The mean amino acid composition of the human milk was found to be similar to previously reported estimates, although the cysteine content of the human milk in the present study was 20% higher than the average literature estimate. True (corrected for endogenous amino acid excretions) ileal amino acid digestibility of human milk was determined using the 3-week-old piglet as a model animal for the human infant. The piglets were given either human milk (n 6) or a protein-free diet (n 6) for a 6 d experimental period. Cr2O3 was added as an indigestible marker, to both the human milk and protein-free diet. At the end of the experimental period the piglets were anaesthetized and samples of digesta removed from the terminal ileum of each piglet. After sampling the piglets were killed. Endogenous ileal excretions of amino acids were determined in piglets fed on the protein-free diet. The true digestibilities of total N and amino acid N were 88% and 95% respectively. The true ileal digestibility of the non-amino acid N fraction in human milk, when calculated by difference was only 50%. The true digestibility of the amino acids in human milk ranged from 81–101% with threonine (86%) being the least digestible essential amino acid. When the true ileal digestibility values were used to correct the amino acid composition of human milk, the pattern of digestible amino acids in human milk was different compared with the currently recommended pattern of amino acid requirements for the infant.

The effects of gestational age at delivery (GA), postnatal age (PNA) and post-menstrual age (PMA=PNA+GA, an indicator of autonomous developmental processes not affected by the moment of birth) on macronutrient composition of very preterm milk were studied. Total N, fat, lactose and carbohydrate concentrations, energy density and 24 h volume were determined in 282 24 h milk samples collected at weekly intervals (days 7–55 of lactation) from seventy-nine women delivering their babies between 25 and 29 weeks of gestation. GA related differences were found for carbohydrate concentration only: carbohydrate concentration was lower with increasing GA. PNA was related to a decrease in total N and an increase in lactose concentration. PMA was not related to milk composition. Our data indicate that PNA strongly influences the development of the composition of very preterm human milk, while GA affects carbohydrate content with a negligible effect on the nutritional value of the milk. We conclude that in accordance with current opinion in paediatrics, human milk is the best source of nutrients even for very preterm (< 30 weeks GA) infants.

Sucrose polyester (SPE) is a tasteless, odourless substance which reduces plasma cholesterol concentrations and may therefore be valuable as a fat substitute in human foodstuffs. It has recently been approved for use in snack foods by the United States Federal Drug Administration. The current study was designed to investigate its effects on gastrointestinal physiology and nutrient absorption in human subjects. A 6-month (2×3-month periods) double-blind, lacebo-controlled, randomized, cross-over trial of SPE and control fat was performed in healthy free-living volunteers. Subjects consumed 20–40 g of SPE daily (mean 26·8 (se 6·8) g) which reduced the intake of total and saturated fat but had no effect on energy intake or body weight. Plasma cholesterol and triacylglycerols were reduced. The frequency of bowel movements and their urgency were increased and anal leakage occured in 7·2% of subjects. Abdominal pain was more frequent in subjects receiving SPE and was significantly greater than in the control group after 8 weeks feeding. The plasma concentrations of vitamin E and six carotenoids were significantly reduced. Routine haematology and biochemistry, other vitamins, intestinal biopsies, bile-salt retention, rectal prostaglandins, fractional Ca absorption and aminopyrine metabolism were unaffected. The ingestion of foods containing 20–40 g SPE daily provoked significant gastro-intestinal problems. This intake is greater than that to be expected from the use of SPE in savoury snack foods, for which it has been approved by the United States Federal Drug Administration. However, the favourable effects on lipid profiles must be balanced against the reduction in the concentrations of vitamins and carotenoids, as these compounds may have beneficial effects on health through protection from free-radical oxidative stress.

The demand for glycine to satisfy normal growth during early life is considerable and most has to be made endogenously. The extent to which adequate glycine is available can be assessed by measuring the urinary excretion of 5-L-oxoproline. The excretion of 5-L-oxoproline at 6 weeks of age for infants in Trinidad of African, Indian or mixed parentage (398 μmol/mmol creatinine) was significantly greater than for infants born in England of Caucasian parentage (194 μmol/mmol creatinine). There was no relationship between 5-L-oxoproline excretion and either sex or pattern of feeding. There were significant inverse relationships between 5-L-oxoproline/creatinine and birth weight, and head circumference either at birth or 6 weeks of age, suggesting that limited availability of glycine is associated with poorer growth before and after birth. For a group of infants born in England of Indian parentage, excretion of 5-L-oxoproline (155 μmol/mmol creatinine) was not different to infants of Caucasian parentage, but significantly less than infants born in Trinidad. The demonstration that 5-L-oxoproline/creatinine was similar in infants born in England, regardless of parentage, shows that the differences between England and Trinidad are related to environment and are unlikely to be accounted for by genetic differences or ethnicity.

The present study examined the influence of training, followed by a short period of detraining, on postprandial lipaemia. Fourteen normolipidaemic, recreationally active young adults aged 18–31 years participated, in two self-selected groups: three men and five women (BMI 21·7–27·6 kg/m2) completed 13 weeks of running training, after which they refrained from exercise for 9 d; three men and three women (BMI 21·5–25·6 kg/m2) maintained their usual lifestyle. Oral fat tolerance tests were conducted at baseline and again 15 h, 60 h and 9 d after the runners' last training session. Blood samples were drawn after an overnight fast and at intervals for 6 h after consumption of a high-fat meal (1·2 g fat, 1·4 g carbohydrate, 70·6 kJ energy/kg body mass). Heparin was then administered (100 IU/kg) and a further blood sample was drawn for measurement of plasma lipoprotein lipase (EC 3.1.1.34; LPL) activity. Endurance fitness improved in runners, relative to controls (maximal O2 uptake +3·2 (SE 1·1) ml/kg per min v. − 1·3 (SE 1·2) ml/kg per min; P < 0·05). In the absence of the acute effect of exercise, i.e. 60 h after the last training session, there was no effect of training on either postprandial lipaemia or on post-heparin LPL activity. However, changes during 9 d of detraining in both these variables differed significantly between groups; after 2 d without exercise (60 h test), the runners' lipaemic response was 37% higher than it was the morning after their last training session (15 h test; runners v. controls P < 0·05), with a reciprocal decrease in post-heparin LPL activity (P < 0·01). These findings suggest that improved fitness does not necessarily confer an effect on postprandial lipaemia above that attributable to a single session of exercise.

During pregnancy, maternal plasma concentrations of the peroxidation-susceptible polyunsaturated fatty acids (polyenes) increase. In addition, the proportion of polyenes is higher in neonatal plasma than in maternal plasma. To study whether these increased amounts of polyenes affect antioxidant levels, we measured lipid-soluble antioxidants in maternal and neonatal plasmas obtained during thirty-five normal pregnancies. These values were then related to the degree of phospholipid-fatty acid unsaturation. Maternal plasma levels of tocopherols and lutein increased during pregnancy, as assessed at 14, 22, and 32 weeks of gestation. However, β-carotene levels decreased, and levels of other carotenoids remained unchanged. Retinol levels were only decreased at 32 weeks of gestation. The value for α-tocopherol: phospholipid-polyene unsaturation index (UI) also increased during pregnancy, despite the observed increase in UI. Corresponding ratios for several carotenoids and retinol, however, decreased during pregnancy. After delivery, maternal plasma levels of δ-tocopherol and β + γ-tocopherol, as well as β + γ-tocopherol: UI values, were lower than values at 32 weeks of gestation. Umbilical-cord plasma antioxidant levels and antioxidant: UI values, except retinol: UI, were significantly lower than maternal values. Significant and consistent cord ν. maternal correlations were observed for plasma levels of β + γ-tocopherol, lutein and β-carotene, but not for δ-tocopherol, α-tocopherol, lycopene, α-carotene, and retinol. In conclusion, although during pregnancy maternal plasma tocopherol levels increased concurrently with, or more than, fatty acid unsaturation in plasma phospholipids, the decrease in carotenoid: UI values during gestation, the decrease in maternal plasma levels of δ-tocopherol and β + γ-tocopherol after delivery, and the low neonatal antioxidant levels merit further investigation.

Excessive interferon-gamma (IFN-γ) production appears to be a primary immunological lesion in vitamin A-deficient experimental animals but comparable data from humans is lacking. We investigated IFN-γ production in South African children by measurement of urinary excretion of neopterin, a product of IFN-γ-activated monocytes or macrophages. Preschool children were examined during an acute inflammatory illness resulting from accidental ingestion of kerosene and they and a neighbourhood control child were examined 3 months later when well. Vitamin A status was assessed by the modified relative dose response (MRDR) test at 3 months and serum retinol and acute phase proteins were measured at both time points. Urinary neopterin was measured for forty cases in hospital, forty-six cases after recovery, and forty-one controls. Significantly increased neopterin excretion was seen following kerosene ingestion and in association with raised serum acute phase protein concentrations. There was no relationship between neopterin excretion at either time point and vitamin A status as assessed by MRDR test. Urinary neopterin was negatively correlated with serum retinol but no significant relationship was observed when acute phase protein concentrations were included in a multiple regression, suggesting the correlation was secondary to illness-induced changes in serum retinol. The results indicate that, contrary to what is observed in rodents under experimental conditions, poor vitamin A status is not associated with altered regulation of IFN-γ production in children.

We evaluated the accuracy of self-reported home-assessed and self-measured waist circumference in 101 men and eighty-three women aged 28–67 years. The main outcome measures were subjects' self-reported and self-measured waist circumference, and self-classification according to the previously defined waist action level 1 (940 mm in men, 800 mm in women) and action level 2 (1020 mm in men, 880 mm in women), and waist circumference measured by the investigator using the ‘Waist Watcher’ tape-measure, as the reference method. The mean errors (95% CI limits of agreement) for subjects' self-reported waist circumference (self-reported minus reference; mm) were −67 (95% CI −210, 77) in men and −43 (95% CI −211, 123) in women, and for self-measured waist circumference (mm) using the ‘Waist Watcher’ (self-measured minus reference) were −5 (95% CI −62, 52) in men and −4 (95% CI −50, 42) in women. The proportions of subjects classified into waist action level 1 or action level 2 by the investigator were used as the reference method. Self-reported waist circumference of men and women respectively would be classified correctly in different categories based on action level 1 with sensitivities of 58·3 and 78·7%, and specificities of 98·5 and 98·7%, and action level 2 with sensitivities of 38·3 and 48·9%, and specificities of 98·5 and 98·7%. Using the ‘Waist Watcher’ with different colour bands based on the action levels, male and female subjects respectively classified themselves into correct categories according to action level 1 with sensitivities of 100 and 98·7%, and specificities of 98·1 and 98·2%, and according to action level 2 with sensitivities of 98·1 and 100%, and specificities of 100% for both sexes. Only 2% of the sample misclassified themselves into the wrong categories according to waist circumference action levels. In conclusion, people tend to underestimate their waist circumference, but the ‘Waist Watcher’ tape-measure offers advantages over self-reported home-assessed measurement, and may be used as a screening tool for self-classifying the risk of ill health through intra-abdominal fat accumulation.

Increasing the energy value of diets with dietary fat, particularly fats rich in saturated fatty acids, can result in the elevation of plasma total and lipoprotein cholesterol. In the present study, experimental diets were designed to examine the effects of increasing the energy content of diets with a saturated fat source and cholesterol in a non-purified diet on hyperlipoproteinaemia and aortic plaque composition in the atherosclerosis-susceptible Japanese quail (Coturnix japonica) model of human atherosclerosis. Commercial poultry diets containing two levels (i.e. 60 or 120 g/kg) of beef tallow as the primary source of saturated fat were balanced for endogenous cholesterol or supplemented with cholesterol (i.e. 0·5 or 5·0 g/kg) and fed to quail for 9 weeks to examine the effects on whole plasma, lipoprotein and aortic plaque lipid composition in relation to aortic plaque formation. Hypercholesterolaemia (P<0·001) was confirmed in birds fed on high-cholesterol (HC) diets only. An interaction (P=0·05) between dietary cholesterol and fat intake level was observed for plasma triacylglycerols (TG) and was specific to changes observed in VLDL composition. Diet-induced changes in lipoprotein total cholesterol, TG and phospholipid composition were greatest in the portomicron and VLDL fractions in birds fed on atherogenic diets. Hyperlipoproteinaemia induced by the 60 g/kg added beef tallow–HC diet resulted in significant (P<0·001) aortic plaque deposition, which was further enhanced in birds fed on the 120 g/kg beef tallow–HC diet. Quail fed on 120 g/kg beef tallow-HC diets exhibited the most severe aortic plaque formation, with marked increases in aortic tissue cholesterol content and quantifiable amounts of several cholesterol oxides (5,6α-epoxy-5α-cholesterol, 7β-hydroxycholesterol, cholestanetriol, 7-ketocholesterol and 25-hydroxycholesterol). In summary, hyperlipoproteinaemia associated with HC diets with a greater proportion of energy from saturated fat produced a combined effect in altering plasma and lipoprotein lipid composition as well as aortic tissue cholesterol and cholesterol oxide content in the Japanese quail.

Acetylation is a potential method for protecting dietary peptides from degradation by rumen micro-organisms. As a first step in determining the nutritive value of acetylated peptides, their disappearance in the small intestine of sheep and their ability to support growth in a rat bioassay were measured. 15N-labelled peptides were prepared from lucerne which had been grown with 15N-labelled (NH4)2 SO4 in the absence of Rhizobium. Peptides were prepared by enzymic hydrolysis of the extracted protein. Two peptide preparations were made using different proteinase mixtures. These mixtures contained peptides with an average molecular weight of 559 and 522 Da. They were treated with acetic anhydride, which resulted in 85 and 88% modification respectively, and their uptake from the small intestine was determined by injecting 1 g of untreated or acetylated peptides in a Cr-EDTA solution into the jejunum of two sheep fitted with jejunal catheters and ileal cannulas. Ileal digesta were collected and analysed for Cr and 15N. The uptake of dialanine (Ala2) and N-acetyl-Ala2 were compared in a similar way. The disappearance of 15N from lucerne peptides was high (88 and 93% respectively) and this was not affected significantly by acetylation (86 and 87%). Corresponding values for Ala2 and N-acetyl-Ala2 were both 96%, as measured by HPLC. It was therefore concluded that acetylation did not affect the uptake of peptides from the small intestine in sheep. Two feeding trials were carried out with rats. The first trial was carried out with a protein-free diet to which was added 10% lactalbumin or 5% lactalbumin and then a mixture of methionine-free amino acids, either alone or supplemented with Met, Gly-Met or acetylated Gly-Met. The rats grew equally well on all sources of Met, but failed to grow significantly on the mixture of Met-free amino acids. In the second trial the diet contained casein as 5·9% of the basal diet. Additional casein, pancreatic casein hydrolysate (peptides) and acetylated pancreatic casein hydrolysate (acetylated peptides) were compared as sources of amino acids, at inclusion rates of 100 g/kg final diet. Feed intake was similar with casein and peptides treatments, but was depressed by 23% with acetylated peptides. Live weight gain was 15 and 75% lower with the peptides and acetylated peptides diets respectively. Addition of lysine, arginine or histidine did not restore feed intake or weight gain of rats receiving acetylated peptides, but feed intake was restored immediately when peptides replaced acetylated peptides. When intake was restricted to 9 g/d and acetylated casein hydrolysate replaced half of the protein in the diet, rats gained weight less rapidly (1·44 v: 1·09 g/d) and retained less N, such that only 0·36 of the acetylated peptide-N was calculated to remain available to the animal. This N retention compared with 0·70 for unmodified casein. Thus, the rat bioassay indicated that certain specific peptides may well be of high nutritive value following acetylation, but that there may be problems of inappetance and inefficient utilization with acetylated peptide mixtures.

Agrimony eupatoria (agrimony) has been documented as a traditional treatment of diabetes. Here, the effects of dietary administration of agrimony on streptozotocin (STZ)-diabetic mice and on in vitro glucose uptake and glucose metabolism, and on insulin secretion by BRIN-BD11 cells were investigated. Agrimony incorporated into the diet (62·5 g/kg) and drinking water (2·5 g/l) countered the weight loss, polydipsia, hyperphagia and hyperglycaemia of STZ-diabetic mice. Aqueous extract of agrimony (1 mg/ml) stimulated 2-deoxy-glucose transport (1·4-fold), glucose oxidation (1·4-fold) and incorporation of glucose into glycogen (2·0-fold) in mouse abdominal muscle comparable with 0·1 μM-insulin. In acute 20 min tests, 0·25-1 mg/ml aqueous extract of agrimony evoked a stepwise 1·9–3·8-fold stimulation of insulin secretion from the BRIN-BD11 pancreatic B-cell line. This effect was abolished by 0·5mM-diazoxide and previous exposure to extract did not adversely affect subsequent stimulation of insulin secretion by 10 mM-L-alanine, thereby indicating that there was no detrimental effect of the extract on cell viability. The effect of extract was glucose-independent and was not evident in BRIN-BD11 cells exposed to a depolarizing concentration of KCl. The ability of agrimony extract to enhance insulin secretion was dependent on use of heat during extract preparation. These results demonstrate the presence of antihyperglycaemic, insulin-releasing and insulin-like activity in Agrimony eupatoria.