The objective of the research was to determine the blood pressure (BP) lowering effects in older people of 50 g carbohydrate drinks with varying carbohydrate content using a randomised, cross-over study with ten (six females) healthy older subjects (mean age 72·20 (sem 1·50) years). BP, heart rate and glucometer-derived blood glucose levels were determined at baseline and following the ingestion of equal volumes (300 ml) of water and carbohydrate drinks with varying nutrient content (glucose, sucrose and fructose). A significant decline in BP over the first 60 min was seen following glucose (systolic BP (SBP) P<0·01, diastolic BP (DBP) P<0·01, mean arterial BP (MAP) P=0·03) and sucrose (SBP P<0·01, DBP P<0·01, MAP P<0·01) ingestion, although the decrease occurred earlier after glucose than sucrose ingestion (SBP 7·33 (sem 2·19) v. 21·00 (sem 4·30) min (P=0·03) and MAP 11·22 (sem 3·10) v. 17·00 (sem 3·78) min (P=0·03)). BP increased after water ingestion (SBP P=0·04, DBP P=0·18, MAP P=0·02) but did not change after fructose ingestion (SBP P=0·36, DBP P=0·81, MAP P=0·34). Post hoc analyses revealed that the BP (SBP, DBP and MAP) decrease following glucose and sucrose ingestion were similar but significantly greater than following fructose or water ingestion. Sucrose, which is used widely (table sugar), reduces BP as much as glucose. In contrast to this, fructose ingestion causes no change in BP. Further studies are required to determine if the substitution of glucose or sucrose with fructose may be beneficial in the medical management of older people with severe symptomatic postprandial hypotension.

High plasma level of cholesterol is a well-known risk factor for atherosclerotic diseases. Oxidized LDL induces cellular and nuclear damage that leads to apoptotic cell death. We tested the hypothesis that flavonoids may function as antioxidants with regard to LDL incubated with 5 μm-Cu2+ alone or in combination with human umbilical vein endothelial cells (HUVEC). Cytotoxicity and formation of thiobarbituric acid-reactive substances induced by Cu2+-oxidized LDL were examined in the presence of various subtypes of flavonoid. Flavanols, flavonols and flavanones at a non-toxic dose of 50 μm markedly inhibited LDL oxidation by inhibiting the formation of peroxidative products. In contrast, the flavones luteolin and apigenin had no such effect, with >30 % of cells killed after exposure to 0.1 mg LDL/ml. Protective flavonoids, especially (−)-epigallocatechin gallate, quercetin, rutin and hesperetin, inhibited HUVEC nuclear condensation and fragmentation induced by Cu2+-oxidized LDL. In addition, immunochemical staining and Western blot analysis revealed that anti-apoptotic Bcl-2 expression was enhanced following treatment with these protective flavonoids. However, Bax expression and caspase-3 cleavage stimulated by 18 h incubation with oxidized LDL were reduced following treatment with these protective flavonoids. The down-regulation of Bcl-2 and up-regulation of caspase-3 activation were reversed by the cytoprotective flavonoids, (−)-epigallocatechin gallate, quercetin and hesperetin, at ≥10 μm. These results suggest that flavonoids may differentially prevent Cu2+-oxidized LDL-induced apoptosis and promote cell survival as potent antioxidants. Survival potentials of certain flavonoids against cytotoxic oxidized LDL appeared to stem from their disparate chemical structure. Furthermore, dietary flavonoids may have therapeutic potential for protecting the endothelium from oxidative stress and oxidized LDL-triggered atherogenesis.

The extent to which maternal folate and vitamin B12 modulate infant birth weight is unclear. The present study investigated mothers in early gestation (mean 11·5 (sd 5·8) weeks) and neonates, at delivery. Erythrocyte (RBC) folate (mothers: n 683; neonates: n 614) and vitamin B12 (mothers: n 534; neonates: n 614) were measured. Data on smoking habits were available for 44 % of pregnancies (n 443). The relationship between vitamin levels and birth weight standardized for gender and gestational age was investigated, using linear regression and adjusting for possible confounding variables (maternal age, parity). Results are presented as standardized regression coefficients (b). Increasing maternal age was associated with elevated RBC folate (b 0·11 (95 % CI 0·08, 0·15), P<0·001; n 674) and smoking was associated with a decrease in maternal RBC folate (b −1·38 (95 % CI −1·92, −0·86), P=0·001; n 319). Neonatal RBC folate was predicted by maternal RBC folate (b 0·08 (95 % CI 0·04, 0·11), P=0·001; n 315) and maternal vitamin B12 (b 0·08 (95 % CI 0·01, 0·16), P=0·02; n 252). Smoking influenced maternal vitamin B12 status (b −0·88 (95 % CI −1·49, −0·27), P=0·005; n 231). Using univariate regression, smoking significantly influenced infant birth weight (b −2·15 (95 % CI −3·24, −1·04), P<0·001; n 437). However, the effect of smoking on birth weight was statistically non-significant when considered in a multivariate regression model, leaving maternal RBC folate as the only significant predictor of birth weight (b 0·25 (95 % CI 0·08, 0·42), P=0·005; n 145). These findings suggest that maternal folate status is an important determinant of infant birth weight. The combined effects of smoking and reduced RBC status on birth weight require further investigation.

One of the debates in infant nutrition concerns whether dietary 18: 3n-3 (linolenic acid) can provide for the accretion of 22: 6 n-3 (docosahexaenoic acid, DHA) in neonatal tissues. The objective of the present study was to determine whether low or high 18: 3 n-3 v. preformed 22: 6 n-3 in the maternal diet enabled a similar 22: 6 n-3 content in the phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylinositol (PI) and phosphatidylserine (PS) of glial cells from whole brain (cerebrum and cerebellum) of 2-week-old rat pups. At parturition, the dams were fed semi-purified diets containing either increasing amounts of 18: 3 n-3 (18: 2 n-6 to 18: 3 n-3 fatty acid ratio of 7·8: 1, 4·4: 1 or 1: 1), preformed DHA, or preformed 20: 4 n-6 (arachidonic acid)+DHA. During the first 2 weeks of life, the rat pups from the respective dams received only their dam's milk. The fatty acid composition of the pups' stomach contents (dam's milk) and phospholipids from glial cells were quantified. The 20: 4n-6 and 22: 6 n-3 content in the stomach from rat pups at 2 weeks of age reflected the fatty acid composition of the dam's diet. The 20: 4n-6 content of PE and PS in the glial cells was unaffected by maternal diet treatments. Preformed 22: 6 n-3 in the maternal diet increased the 22: 6 n-3 content of glial cell PE and PS compared with maternal diets providing an 18: 2n-6 to 18: 3 n-3 fatty acid ratio of 7·8: 1, 4·4: 1 or 1: 1 (P<0·0001). There was no significant difference in the 20: 4 n-6 and 22: 6 n-3 content of glial cell PC and PI among maternal diet treatments. It was concluded that maternal dietary 22: 6n-3 is more effective than low or high levels of maternal dietary 18: 3 n-3 at increasing the 22: 6 n-3 content in PE and PS of glial cells from the whole brain of rat pups at 2 weeks of age. The findings from the present study have important implications for human infants fed infant formulas that are devoid of 22: 6 n-3.

The current epidemic of diabetes likely reflects marked changes in environmental factors, although genetic susceptibility plays a powerful role in the occurrence of diabetes in certain populations. We investigated whether long-term sucrose-drinking causes hyperglycaemia in male Wistar-Imamichi littermates (n 32), which are not genetically susceptible to diabetes or obesity. Each litter was divided equivalently into two groups, the sucrose group and the control group. The sucrose group received 300 g/l sucrose water and the control group received regular water until 42 weeks of age. Rats were weighed every 1 or 2 weeks. Oral glucose tolerance tests were performed at 28 and 36 weeks of age. Plasma glucose and insulin concentrations were measured. Body weights were significantly greater in the sucrose group than in the control group in 18-week-old rats (P<0·05), and the difference between the two groups reached 163 g by the end of the study (P<0·01). The 120 min post-load plasma glucose concentration in the sucrose group was 11·4 (sd 2·8) mmol/l in 28-week-old rats and 12·7 (sd 2·2) mmol/l in 36-week-old rats, while that of the control group remained approximately 7·3–7·7 mmol/l. In the sucrose group, the plasma insulin peak occurred 30 min post-load at 28 weeks of age; but the peak disappeared and hyperinsulinaemia was prolonged at 36 weeks of age. In conclusion, long-term sucrose-drinking causes increased body weight and glucose intolerance in normal male rats.

Inflammation, a risk factor for cardiovascular disease, is associated with low plasma levels of antioxidant vitamins. In addition to vitamins, other antioxidants modulate the synthesis of inflammatory markers in vitro and contribute to the total antioxidant capacity (TAC) of a diet. However, the relationship between dietary TAC and markers of inflammation has never been evaluated in vivo. We investigated the relationship between dietary TAC and markers of systemic (high-sensitivity C-reactive protein (hs-CRP), leucocytes) and vascular (soluble intercellular cell adhesion molecule-1) inflammation in 243 non-diabetic subjects. General Linear Model (GLM) analysis showed a significant (P=0·005) inverse relationship between hs-CRP and quartiles of energy-adjusted dietary TAC, even when recognized modulating factors of inflammation, namely alcohol, fibre, vitamin C, α-tocopherol, β-carotene, BMI, waist circumference, HDL-cholesterol, hypertension, insulin sensitivity and plasma β-carotene, were included in the model as covariates (P=0·004). The relationship was stronger for subjects with hypertension (P=0·013 v. P=0·109 for normotensive individuals). Among dietary factors, TAC was significantly higher (5·3 (sd 3·0) v. 4·9 (sd 2·7) mmol Trolox/d; P=0·026) in subjects with low plasma hs-CRP (range: 0·0–4·1 mg/l) than in subjects with high plasma hs-CRP (range: 4·2–27·8 mg/l). We conclude that dietary TAC is inversely and independently correlated with plasma concentrations of hs-CRP and this could be one of the mechanisms explaining the protective effects against CVD of antioxidant-rich foods such as fruits, whole cereals and red wine. This could be of particular significance for subjects with high blood pressure.

Previous studies of vitamin C absorption in man using stable isotope probes have given results which cannot easily be reconciled with those obtained using non-isotope measurement. In order to investigate some of the apparent paradoxes we have conducted a study using two consecutive doses of vitamin C, one labelled and one unlabelled, given 90 min apart. Compatibility of the experimental results with two feasible models was investigated. In Model 1, ingested vitamin C enters a pre-existing pool before absorption, which occurs only when a threshold is exceeded; in Model 2, ingested vitamin C is exchanged with a pre-existing flux before absorption. The key difference between these two models lies in the predicted profile of labelled material in plasma. Model 1 predicts that the second unlabelled dose will produce a secondary release of labelled vitamin C which will not be observed on the basis of Model 2. In all subjects Model 1 failed to predict the observed plasma concentration profiles for labelled and unlabelled vitamin C, but Model 2 fitted the experimental observations. We speculate on possible physiological explanations for this behaviour, but from the limited information available cannot unequivocally confirm the model structure by identifying the source of the supposed flux.

The present study was performed to investigate the effect of dietary fat and vitamin E on concentrations of cholesterol oxidation products (COP) in broiler muscle. A total of 144 1-d-old broiler chicks were fed diets with either palm oil, soyabean oil or linseed oil and vitamin E concentrations of 20, 40 or 200 mg/kg for 35 d. COP concentrations were analysed in raw, heat-processed (180°C, 20 min) and frozen-stored (−20°C, 6 months) breast and thigh muscles. COP concentrations were influenced by dietary vitamin E concentration, dietary fat, treatment and type of muscle (P<0·001). Increasing the dietary vitamin E concentration generally reduced the concentration of COP. This effect was strongest in broilers fed linseed oil and weakest in broilers fed palm oil; the effect of vitamin E was also stronger in heated muscles than in raw or frozen-stored muscles. Moreover, the concentration of COP in thigh muscle was more strongly influenced by dietary vitamin E than that in breast muscle. COP concentrations in muscles were on average highest in broilers fed linseed oil and lowest in broilers fed palm oil, but the effect of the dietary fat also depended on the vitamin E concentration, the treatment and the type of muscle. In conclusion, our study shows that dietary fat and vitamin E influence the concentrations of total COP in broiler muscle. However, the effects of these factors were not only influenced by interactions between each other, but also depended on the treatment of the muscle and the type of muscle.

Diet is known to modulate the development of diabetes in diabetes-prone BioBreeding (BBdp) rats. The objective of the present study was to determine the effect of fermentable fibre (FF) on immune function in BBdp and diabetes-resistant BioBreeding (BBdr) rats after weaning. Weanling BBdp (thirty-six to thirty-eight per diet) and BBdr rats (thirty to thirty-two per diet) were fed a nutritionally complete, semi-purified, casein-based diet containing either cellulose (control diet, 8 % w/w) or FF (3·2 % cellulose+4·8 % w/w inulin). At 35 d, the small intestine was excised and lymphocytes isolated from spleen, mesenteric lymph nodes and Peyer's patches. Feeding FF to both BBdr and BBdp rats affected the production of anti-inflammatory cytokines (P=0·02). In BBdr rats, feeding FF compared with cellulose resulted in an increased small intestinal length (P=0·0031), higher proliferative (stimulation) index from both splenocytes (P=0·001) and mesenteric lymph nodes (P=0·04), and an increased proportion of CD8+ T-cells in the Peyer's patches (P=0·003). We did not observe an effect of diet on the number of IgA-bearing cells in the jejunum from BBdr rats. Feeding FF to BBdp rats did not affect the same parameters. BBdp rats had both a higher proportion of B-cells in the Peyer's patches (P=0·01) and a higher number of IgA+ cells in the jejunum (P=0·0036) when fed a diet containing FF, a response not observed in BBdr rats. We demonstrate that several aspects of the BBdp immune system respond differently than that of BBdr rats when challenged at weaning with FF.

The purpose of the present study was to investigate the effect of vitamin E supplementation on autoimmune disease in New Zealand black×New Zealand white F1 (NZB/W F1) female mice fed an oxidised oil diet. First, 5-month-old mice were fed an AIN-76 diet containing either 150 g fresh soyabean oil/kg (15S), 50 g fresh soyabean oil/kg + 100 g oxidised frying oil/kg (5S10F) or 5S10F supplemented with all-rac-α-tocopheryl acetate at 275 mg/kg diet level (5S10F5E) or 550 mg/kg (5S10F10E), respectively, in experiment 1. The results showed that mice fed the 5S10F10E diet had a lower anti-double-stranded DNA IgG antibody level and a longer lifespan than those fed the 15S and 5S10F diets. Therefore, the 5S10F and 5S10F10E treatments were repeated in experiment 2 for further analysis. The results showed that vitamin E supplementation in the oxidised oil significantly decreased thiobarbituric acid-reactive substance values in the kidney and spleen of NZB/W F1 mice. Interferon-γ and IL-6 production by mitogen-stimulated splenocytes decreased in mice fed the 5S10F10E diet, whereas the secretion of IL-2 and IL-10 was not affected. The percentage of T-cells was significantly higher and that of MHC class II-bearing cells was lower in the spleens of the 5S10F10E group. The 5S10F10E group had a significantly higher linoleic acid (18: 2n-6) composition than the 5S10F diet group. Therefore, vitamin E supplementation in oxidised oil might decrease oxidative stress, anti-double-stranded DNA IgG antibody, regulate cytokines and lymphocyte subsets, and subsequently alleviate the severity of autoimmune disease such as systemic lupus erythematosus under oxidative stress.

In children, inappropriate eating habits can induce a disease known as nutritional dwarfing (ND). Due to the link between nutritional condition and bone growth, the effects induced by a 20 % reduction of food intake on bone competence were assessed in an animal model of ND. Bone status during catch-up growth was also analysed. Male Wistar rats were divided into control (C) and ND groups. C rats were fed ad libitum. ND received 80 % of the diet consumed by C for 4 weeks (T4); thereafter, they were fed ad libitum for 8 weeks. Results, expressed as mean (sem) for ND v. C, were as follows. At T4, body weight (g) and length (cm) and femur weight (g) and length (mm) were 97·35 (sem 5·89) v. 199·07 (sem 9·24), 16·91 (sem 0·41) v. 20·26 (sem 0·31), 0·30 (sem 0·01) v. 0·46 (sem 0·01) and 23·09 (sem 0·29) v. 26·98 (sem 0·26), respectively (P<0·001); bone mineral content (g) and density (g/cm2) were 0·014 (sem 0·002) v. 0·030 (sem 0·002) and 0·061 (sem 0·004) v. 0·080 (sem 0·003), respectively (P<0·001); load-bearing capacity (N), yielding load (N) and elastic stiffness (N/mm) were 25·06 (sem 1·24) v. 50·34 (sem 2·94), 23·72 (sem 1·02) v. 46·97 (sem 1·75) and 65·98 (sem 4·42) v. 115·07 (sem 3·85), respectively (P<0·001); cross-sectional area (mm2) and moment of inertia (mm4) were 2·86 (sem 0·19) v. 4·54 (sem 0·17) and 1·27 (sem 0·08) v. 3·03 (sem 0·16), respectively (P<0·001). Significant effects were not evident in material properties. Parameters assessed normalized during re-feeding. These results suggest that the impaired mechanical femur competence in ND rats could be due to an altered bone mass and architectural distribution rather than to intrinsic quality. Re-feeding caused a reversal of the effects of food restriction on growth and bone parameters in ND rats.

Accurate measurement of energy intake (EI) is essential in studies of energy balance in all age groups. Reported values for EI can be validated against total energy expenditure (TEE) measured using doubly labelled water (DLW). Our previous work has indicated that the use of the standardized 24 h multiple pass recall (24 h MPR) method produces slight overestimates of EI in pre-school children which are inaccurate at individual level but acceptable at group level. To extend this work, the current study validated EI by 24 h MPR against TEE by DLW in sixty-three (thirty-two boys) school-aged children (median age 6 years). In both boys and girls, reported EI was higher than TEE, although this difference was only significant in the girls (median difference 420 kJ/d, P=0·05). On analysis of agreement between TEE and EI, the group bias was an overestimation of EI by 250 kJ/d with wide limits of agreement (−2880, 2380 kJ/d). EI was over-reported relative to TEE by 7 % and 0·9 % in girls and boys, respectively. The bias in the current study was lower than in our previous study of pre-school children, suggesting that estimates of EI become less inaccurate as children age. However, the current study suggests that the 24 h MPR is inaccurate at the individual level.

We investigated the effect of Brussels sprouts, inulin and a fermented milk on the faecal microbiota diversity of human microbiota-associated (HMA) rats by PCR-temporal temperature gradient gel electrophoresis (PCR-TTGE) using universal and group-specific 16S rRNA gene primers. The HMA rats were submitted to a control diet for 10 d (initial time), then switched to the experimental diets for 4 weeks (final time). Using universal primers, the mean degree of similarity between all faecal samples at initial time was 80·8 %. In the group consuming the control diet throughout the experiment, the mean degree of similarity between the PCR-TTGE profiles at initial v. final time was 76·8 %, reflecting a spontaneous temporal variation. The mean degree of similarity between control and experimental groups at final time was lower, 72·4 %, 74·4 % and 75·6 % for inulin, Brussels sprouts and fermented milk, respectively, indicating a dietary effect on the predominant populations. Using specific primers, bifidobacteria could be detected only in those rats that had consumed inulin, showing a specific increasing effect of this dietary compound. The Lactobacillus population was very heterogeneous at initial time but tended to homogenize within each dietary group. At final time, caecal contents were collected for analysis of SCFA and β-glucuronidase activity. Inulin and Brussels sprouts increased the butyrate and acetate proportion, respectively, while the fermented milk did not modify the caecal biochemistry. This experiment shows for the first time that cruciferous vegetables are able to alter the diversity and the metabolic activities of the digestive microbiota in HMA rats.

This study compared plasma and urinary carnitine concentrations in pregnant and non-pregnant Korean women. The subjects were fifty pregnant women and thirty non-pregnant women aged 24–28 years. During the first trimester, dietary carnitine intakes in the pregnant women were much lower than in non-pregnant women (70·00 (sd 29·22) μmol/d), but over the course of pregnancy carnitine intake increased from 44·64 (sd 24·84) μmol/d during the first trimester to 96·11 (sd 36·56) μmol/d during the third trimester. Pregnant women had a significantly lower plasma carnitine concentration than non-pregnant women. Plasma concentrations of non-esterified carnitine, acid-soluble acylcarnitine and total carnitine were significantly lower during the second and third trimesters than the first. Plasma acid-insoluble acylcarnitine levels, which tended to be higher in the non-pregnant women compared with the pregnant women, increased significantly as gestation proceeded. The urinary excretion of non-esterified carnitine, acid-soluble acylcarnitine and total carnitine was significantly higher in the pregnant women during the first and second trimesters than in non-pregnant women and decreased significantly as gestation proceeded. We found that there was a significant decrease in plasma carnitine level even though dietary carnitine intake increased as gestation proceeded. The low urinary excretion of carnitine in late pregnancy may be caused by an increased demand during pregnancy.

Our aim was to determine whether meal fatty acids influence insulin and glucose responses to mixed meals and whether these effects can be explained by variations in postprandial NEFA and Apo, which regulate the metabolism of triacylglycerol-rich lipoproteins (Apo C and E). A single-blind crossover study examined the effects of single meals enriched in saturated fatty acids SFA), n-6 PUFA and MUFA on plasma metabolite and insulin responses. The triacylglycerol response following the PUFA meal showed a lower net incremental area under the curve than following the SFA and MUFA meals (P<0·007). Compared with the SFA meal, the PUFA meal showed a lower net incremental area under the curve for the NEFA response from initial suppression to the end of the postprandial period (180–480 min; P<0·02), and both PUFA and MUFA showed a lower net incremental glucose response (P<0·02), although insulin concentrations were similar between meals. The pattern of the Apo E response was also different following the SFA meal (P<0·02). There was a significant association between the net incremental NEFA (180–480 min) and glucose response (rs=0·409, P=0·025), and in multiple regression analysis the NEFA response accounted for 24 % of the variation in glucose response. Meal SFA have adverse effects on the postprandial glucose response that may be due to greater elevations in NEFA arising from differences in the metabolism of SFA- v. PUFA- and MUFA-rich lipoproteins. Elevated Apo E responses to high-SFA meals may have important implications for the hepatic metabolism of triacylglycerol-rich lipoproteins.

Commercial hydrogenation results in the formation of trans fatty acids. An unintended consequence of the hydrogenation process is conversion of phylloquinone (vitamin K1) to dihydrophylloquinone. Plasma dihydrophylloquinone concentrations have yet to be characterized in population-based studies. Dietary determinants of plasma dihydrophylloquinone were estimated using a semi-quantitative food frequency questionnaire in 803 men and 913 women in the Framingham Offspring Study. Geometric mean dihydrophylloquinone intake was 21·3 (95 % CI 20·4, 22·3) μg/d in men and 19·4 (95 % CI 18·5, 20·2) μg/d in women. Detectable (>0·05 nmol/l) plasma dihydrophylloquinone concentrations were measured in 41 % and 30 % of men and women, respectively. The multivariate odds ratio (OR) of detectable plasma dihydrophylloquinone from the lowest to the highest quartile category of dihydrophylloquinone intake were 1 (referent), 1·13 (95 % CI 0·83, 1·53), 1·66 (95 % CI 1·21, 2·26) and 1·84 (95 % CI 1·31, 2·58), P for trend <0·001, adjusted for sex, age, body mass index, triacylglycerols, season and energy intake. Higher trans fatty acid intake was associated with higher multivariate OR for detectable plasma dihydrophylloquinone (OR comparing extreme quartiles 2·41 (95 % CI 1·59, 3·64), P for trend <0·001). There were limitations in the use of plasma dihydrophylloquinone, evident in the high proportion of the population that had non-detectable dihydrophylloquinone concentrations. Despite this caveat, higher plasma dihydrophylloquinone was associated with higher dihydrophylloquinone intake and higher trans fatty acid intake.

Dietary patterns are comprehensive variables of dietary intake appropriate to model the complex exposure in nutritional research. The objectives of this study were to identify dietary patterns by applying two statistical methods, principal component analysis (PCA) and reduced rank regression (RRR), and to assess their ability to predict all-cause mortality. Motivated by previous studies we chose percentages of energy from different macronutrients as response variables in the RRR analysis. We used data from 9356 German elderly subject enrolled in the European Prospective Investigation into Cancer and Nutrition study. The first RRR pattern, subjects which explained 30·8 % of variation in energy sources and especially much variation in intake of saturated fat, monounsaturated fat and carbohydrates was a significant predictor of all-cause mortality. The pattern score had high positive loadings in all types of meat, butter, sauces and eggs, and was inversely associated with bread and fruits. After adjustment for other known risk factors, the relative risks from the lowest to highest quintiles of the first RRR pattern score were 1·0, 1·01, 0·96, 1·32, 1·61 (P for trend: 0·0004). In contrast, the first two PCA patterns explaining 19·7 % of food intake variation but only 7·0 % of variation in energy sources were not related to mortality. These results suggest that variation in macronutrients is meaningful for mortality and that the RRR method is more appropriate than the classic PCA method to identify dietary patterns relevant to mortality.

The present study investigated the effect of l-carnitine supplementation during pregnancy (125 mg/d) and lactation (250 mg/d) on milk production, litter gains and back-fat thickness in sows fed a low-energy and low-protein diet during lactation. Sows supplemented with l-carnitine produced more milk on days 11 and 18 of lactation (+18 %; P<0·05) and had higher litter gains during suckling (+20 %; P<0·01) than control sows. Loss of body weight during lactation was similar in both groups, but sows supplemented with l-carnitine had a greater reduction of back-fat thickness (+45 %; P<0·05) during lactation than control sows. In conclusion, this study shows that l-carnitine increases milk production and litter gains in sows in a strongly negative energy and N balance, and enhances body fat mobilisation.

Four trials were carried out to determine the optimal level of dietary pyridoxine (vitamin B6) and its interaction with riboflavin (vitamin B2) in early-weaned piglets. In Trial 1, twelve piglets were tube-fed graded supplements of B6, 0, 10, 50 or 100 mg/kg. The level of 50 mg/kg maximized B6 in red blood cells (P<0·05). In Trial 2, thirty-six piglets were tube-fed with four combinations of B6 (0 v. 50 mg/kg) and B2 (0 v. 25 mg/kg). The B6 supplement increased (P<0·01) B6 in red blood cells. C-peptide and insulin responses to intravenous glucose tended (P<0·08) to or decreased (P<0·03) with B2 while no effect was observed on glucose. After gastro-enteral glucose, dietary B2 depressed C-peptide and insulin responses in B6-unsupplemented piglets and increased them in B6-supplemented piglets (P<0·03). The glucose response tended to be higher in B6-supplemented piglets (P<0·06). Trials 3 and 4 were carried out in commercial conditions using either B6 and/or B2 supplements given during 2 weeks after weaning (Trial 3) or a B6 supplement alone (50 mg/kg) given between 2 (weaning) and 10 weeks of age. Despite a marked and persistent increase (P<0·01) of B6 in red blood cells in B6-supplemented piglets, the effect on growth performance was either none (P>0·39; Trial 3) or marginally lower (<−2 %; P<0·03; Trial 4). In conclusion, it appears that a dietary supplement of 50 mg/kg B6 saturated the red blood cell pool in B6 and influenced, along with B2, the glucose homeostasis through the entero-insular axis. Nevertheless, such metabolic effects are not reflected on growth performance.