Modelling structural zeros in compositional data

This analysis was stimulated by the real data analysis problem of householdexpenditure data. The full dataset contains expenditure data for a sample of 1224 households. The expenditure is broken down at 2 hierarchical levels: 9 major levels (e.g. housing, food, utilities etc.) and 92 minor levels. There are also 5 factors and 5 covariates at the household level. Not surprisingly, there are a small number of zeros at the major level, but many zeros at the minor level. The question is how best to model the zeros. Clearly, models that tryto add a small amount to the zero terms are not appropriate in general as at least some of the zeros are clearly structural, e.g. alcohol/tobacco for households that are teetotal. The key question then is how to build suitable conditional models. For example, is the sub-composition of spendingexcluding alcohol/tobacco similar for teetotal and non-teetotal households?In other words, we are looking for sub-compositional independence. Also, what determines whether a household is teetotal? Can we assume that it is independent of the composition? In general, whether teetotal will clearly depend on the household level variables, so we need to be able to model this dependence. The other tricky question is that with zeros on more than onecomponent, we need to be able to model dependence and independence of zeros on the different components. Lastly, while some zeros are structural, others may not be, for example, for expenditure on durables, it may be chance as to whether a particular household spends money on durableswithin the sample period. This would clearly be distinguishable if we had longitudinal data, but may still be distinguishable by looking at the distribution, on the assumption that random zeros will usually be for situations where any non-zero expenditure is not small.While this analysis is based on around economic data, the ideas carry over tomany other situations, including geological data, where minerals may be missing for structural reasons (similar to alcohol), or missing because they occur only in random regions which may be missed in a sample (similar to the durables)

Energy cost of glucose storage in human subjects during glucose-insulin infusions.

The effect of graded levels of hyperinsulinemia on energy expenditure, while euglycemia was maintained by glucose infusion, was examined in 22 healthy young male volunteers by using the euglycemic insulin clamp technique in combination with indirect calorimetry. Insulin was infused at five rates to achieve steady-state hyperinsulinemic plateaus of 62 +/- 4, 103 +/- 5, 170 +/- 10, 423 +/- 16, and 1,132 +/- 47 microU/ml. Total body glucose uptake during each of the five insulin clamp studies was 0.41, 0.50, 0.66, 0.74, and 0.77 g/min, respectively. Glucose storage (calculated from the difference between total body glucose uptake minus total glucose oxidation) was 0.25, 0.29, 0.43, 0.49, and 0.52 g/min for each group, respectively, and represented over 60-70% of total glucose uptake. The net increment in energy expenditure after intravenous glucose was 0.08, 0.10, 0.14, 0.17, and 0.23 kcal/min, respectively. Throughout the physiological and supraphysiological range of insulinemia, there was a significant relationship (r = 0.95, P less than 0.001) between the increment in energy expenditure and glucose storage, indicating an energy cost of 0.45 kcal/g glucose stored. However, at each level of hyperinsulinemia, the theoretical value for the energy cost of glucose storage (assuming that all of the glucose is stored in the form of glycogen) could account for only 45-63% of the actual increase in energy expenditure that was measured by indirect calorimetry. These results indicate that factors in addition to glucose storage as glycogen must be responsible for the increase in energy expenditure that accompanies glucose infusion.

On optimal computation of MPLS label binding for multipoint-to-point connections

Most network operators have considered reducing Label Switched Routers (LSR) label spaces (i.e. the number of labels that can be used) as a means of simplifying management of underlaying Virtual Private Networks (VPNs) and, hence, reducing operational expenditure (OPEX). This letter discusses the problem of reducing the label spaces in Multiprotocol Label Switched (MPLS) networks using label merging - better known as MultiPoint-to-Point (MP2P) connections. Because of its origins in IP, MP2P connections have been considered to have tree- shapes with Label Switched Paths (LSP) as branches. Due to this fact, previous works by many authors affirm that the problem of minimizing the label space using MP2P in MPLS - the Merging Problem - cannot be solved optimally with a polynomial algorithm (NP-complete), since it involves a hard- decision problem. However, in this letter, the Merging Problem is analyzed, from the perspective of MPLS, and it is deduced that tree-shapes in MP2P connections are irrelevant. By overriding this tree-shape consideration, it is possible to perform label merging in polynomial time. Based on how MPLS signaling works, this letter proposes an algorithm to compute the minimum number of labels using label merging: the Full Label Merging algorithm. As conclusion, we reclassify the Merging Problem as Polynomial-solvable, instead of NP-complete. In addition, simulation experiments confirm that without the tree-branch selection problem, more labels can be reduced

Label Space Reduction in MPLS Networks: How Much Can A Single Stacked Label Do?

Most network operators have considered reducing LSR label spaces (number of labels used) as a way of simplifying management of underlaying virtual private networks (VPNs) and therefore reducing operational expenditure (OPEX). The IETF outlined the label merging feature in MPLS-allowing the configuration of multipoint-to-point connections (MP2P)-as a means of reducing label space in LSRs. We found two main drawbacks in this label space reduction a)it should be separately applied to a set of LSPs with the same egress LSR-which decreases the options for better reductions, and b)LSRs close to the edge of the network experience a greater label space reduction than those close to the core. The later implies that MP2P connections reduce the number of labels asymmetrically

Influences of body weight, body composition, and substrate oxidation rate on resting postabsorptive glucose production and gluconeogenesis.

OBJECTIVE: To determine the influence of body weight, fat mass, and fat distribution on resting endogenous glucose production in healthy lean and overweight individuals. DESIGN: measurements were performed in the resting postabsorptive state in individuals receiving an unrestricted diet. SETTING: Institute of Physiology of Lausanne University. MEASUREMENTS: resting post absorptive glucose production, glycogenolysis and gluconeogenesis; resting energy expenditure and net substrate oxidation. RESULTS: Endogenous glucose production was positively correlated with body weight, lean body mass, energy expenditure and carbohydrate oxidation. Gluconeogenesis was positively correlated with net lipid oxidation and energy expenditure, and negatively correlated with net carbohydrate oxidation. No correlation with body fat or fat distribution was observed. CONCLUSIONS: Gluconeogenesis shows a large interindividual variability. Net lipid oxidation and not body fat appears to be a major determinant of gluconeogenesis.

Guidelines for specialized nutritional and metabolic support in the critically-ill patient. Update. Consensus SEMICYUC-SENPE: Septic patient

Nutritional metabolic management, together with other treatment and support measures used, is one of the mainstays of the treatment of septic patients. Nutritional support should be started early, after initial life support measures, to avoid the consequences of malnutrition, to provide adequate nutritional intake and to prevent the development of secondary complications such as superinfection or multiorgan failure. As in other critically-ill patients, when the enteral route cannot be used to ensure calorie-protein requirements, the association of parenteral nutrition has been shown to be safe in this subgroup of patients. Studies evaluating the effect of specific pharmaconutrients in septic patients are scarce and are insufficient to allow recommendations to be made. To date, enteral diets with a mixture of substrates with distinct pharmaconutrient properties do not seem to be superior to standard diets in altering the course of sepsis, although equally there is no evidence that these diets are harmful. There is insufficient evidence to recommend the use of glutamine in septic patients receiving parenteral nutrition. However, given the good results and absence of glutamine-related adverse effects in the various studies performed in the general population of critically-ill patients, these patients could benefit from the use of this substance. Routine use of omega-3 fatty acids cannot be recommended until further evidence has been gathered, although the use of lipid emulsions with a high omega-6 fatty acid content should be avoided. Septic patients should receive an adequate supply of essential trace elements and vitamins. Further studies are required before the use of high-dose selenium can be recommended.

Effect of carbohydrate overfeeding on whole body and adipose tissue metabolism in humans.

OBJECTIVE: To evaluate the effect of a 4-day carbohydrate overfeeding on whole body net de novo lipogenesis and on markers of de novo lipogenesis in subcutaneous adipose tissue of healthy lean humans. RESEARCH METHODS AND PROCEDURES: Nine healthy lean volunteers (five men and four women) were studied after 4 days of either isocaloric feeding or carbohydrate overfeeding. On each occasion, they underwent a metabolic study during which their energy expenditure and net substrate oxidation rates (indirect calorimetry), and the fractional activity of the pentose-phosphate pathway in subcutaneous adipose tissue (subcutaneous microdialysis with 1,6(13)C2,6,6(2)H2 glucose) were assessed before and after administration of glucose. Adipose tissue biopsies were obtained at the end of the experiments to monitor mRNAs of key lipogenic enzymes. RESULTS: Carbohydrate overfeeding increased basal and postglucose energy expenditure and net carbohydrate oxidation. Whole body net de novo lipogenesis after glucose loading was markedly increased at the expense of glycogen synthesis. Carbohydrate overfeeding also increased mRNA levels for the key lipogenic enzymes sterol regulatory element-binding protein-1c, acetyl-CoA carboxylase, and fatty acid synthase. The fractional activity of adipose tissue pentose-phosphate pathway was 17% to 22% and was not altered by carbohydrate overfeeding. DISCUSSION: Carbohydrate overfeeding markedly increased net de novo lipogenesis at the expense of glycogen synthesis. An increase in mRNAs coding for key lipogenic enzymes suggests that de novo lipogenesis occurred, at least in part, in adipose tissue. The pentose-phosphate pathway is active in adipose tissue of healthy humans, consistent with an active role of this tissue in de novo lipogenesis.

Disease identification based on ambulatory drugs dispensation and in-hospital ICD-10 diagnoses: a comparison.

BACKGROUND: Pharmacy-based case mix measures are an alternative source of information to the relatively scarce outpatient diagnoses data. But most published tools use national drug nomenclatures and offer no head-to-head comparisons between drugs-related and diagnoses-based categories. The objective of the study was to test the accuracy of drugs-based morbidity groups derived from the World Health Organization Anatomical Therapeutic Chemical Classification of drugs by checking them against diagnoses-based groups. METHODS: We compared drugs-based categories with their diagnoses-based analogues using anonymous data on 108,915 individuals insured with one of four companies. They were followed throughout 2005 and 2006 and hospitalized at least once during this period. The agreement between the two approaches was measured by weighted kappa coefficients. The reproducibility of the drugs-based morbidity measure over the 2 years was assessed for all enrollees. RESULTS: Eighty percent used a drug associated with at least one of the 60 morbidity categories derived from drugs dispensation. After accounting for inpatient under-coding, fifteen conditions agreed sufficiently with their diagnoses-based counterparts to be considered alternative strategies to diagnoses. In addition, they exhibited good reproducibility and allowed prevalence estimates in accordance with national estimates. For 22 conditions, drugs-based information identified accurately a subset of the population defined by diagnoses. CONCLUSIONS: Most categories provide insurers with health status information that could be exploited for healthcare expenditure prediction or ambulatory cost control, especially when ambulatory diagnoses are not available. However, due to insufficient concordance with their diagnoses-based analogues, their use for morbidity indicators is limited.

Rôle du transporteur de glucose GLUT2 dans les mécanismes centraux de glucodétection impliqués dans le contrôle de la sécrétion du glucagon et de la prise alimentaire

Résumé Rôle du transporteur de glucose GLUT2 dans les mécanismes centraux de glucodétection impliqués dans le contrôle de la sécrétion du glucagon et de la prise alimentaire. Les mécanismes centraux de glucodétection jouent un rôle majeur dans le contrôle de l'homéostasie glucidique. Ces senseurs régulent principalement la sécrétion des hormones contre-régulatrices, la prise alimentaire et la dépense énergétique. Cependant, la nature cellulaire et le fonctionnement moléculaire de ces mécanismes ne sont encore que partiellement élucidés. Dans cette étude, nous avons tout d'abord mis en évidence une suppression de la stimulation de la sécrétion du glucagon et de la prise alimentaire en réponse à une injection intracérébroventriculaire (i.c.v.) de 2-déoxy-D-glucose (2-DG) chez les souris de fond génétique mixte et déficientes pour le gène glut2 (souris RIPG1xglut2-/-). De plus, chez ces souris, l'injection de 2-DG n'augmente pas l'activation neuronale dans l'hypothalamus et le complexe vagal dorsal. Nous avons ensuite montré que la ré-expression de GLUT2 dans les neurones des souris RIPG1xg1ut2-/- ne restaure pas la sécrétion du glucagon et la prise alimentaire en réponse à une injection i.c.v. de 2-DG. En revanche, l'injection de 2-DG réalisée chez les souris RIPG1xg1ut2-/- ré-exprimant le GLUT2 dans leurs astrocytes, stimule la sécrétion du glucagon et l'activation neuronale dans le complexe vagal dorsal mais n'augmente pas la prise alimentaire ni l'activation neuronale dans l'hypothalamus. L'ensemble de ces résultats démontre l'existence de différents mécanismes centraux de glucodétection dépendants de GLUT2. Les mécanismes régulant la sécrétion du glucagon sont dépendants de GLUT2 astrocytaire et pourraient être localisés dans le complexe vagal dorsal. L'implication des astrocytes dans ces mécanismes suggère un couplage fonctionnel entre les astrocytes et les neurones adjacents « sensibles au glucose ». Lors de cette étude, nous avons remarqué chez les souris RIPG1xg1ut2-/- de fond génétique pur C57B1/6, que seul le déclenchement de la prise alimentaire en réponse à l'injection i.p. ou i.c.v. de 2-DG est aboli. Ces données mettent en évidence que suivant le fond génétique de la souris, les mécanismes centraux de glucodétection impliqués dans la régulation de la sécrétion peuvent être indépendants de GLUT2. Summary. Role of transporter GLUT2 in central glucose sensing involved in the control of glucagon secretion and food intake. Central glucose sensors play an important role in the control of glucose homeostasis. These sensors regulate general physiological functions, including food intake, energy expenditure and hormones secretion. So far the cellular and molecular basis of central glucose detection are poorly understood. Hypoglycemia, or cellular glucoprivation by intraperitoneal injection of 2-deoxy¬glucose (2-DG) injection, elicit multiple glucoregulatory responses, in particular glucagon secretion and stimulation of feeding. We previously demonstrated that the normal glucagon response to insulin-induced hypoglycemia was suppressed in mice lacking GLUT2. This indicated the existence of extra-pancreatic, GLUT2-dependent, glucose sensors controllling glucagon secretion. Here, we have demonstrated that the normal glucagon and food intake responses to central glucoprivation, by intracerebroventricular (i.c.v.) injections of 2-DG, were suppressed in mice lacking GLUT2 (RIPG1xglut2-/- mice) indicating that GLUT2 plays a role in central glucose sensing units controlling secretion of glucagon and food intake. Whereas it is etablished that glucose responsive neurons change their firing rate in response to variations of glucose concentrations, the exact mechanism of glucose detection is not established. In particular, it has been suggested that astrocytic cells may be the primary site of glucose detection and that a signal is subsequently transmitted to neurons. To evaluate the respective role of glial and neuronal expression of GLUT2 in central glucodetection, we studied hypoglycemic and glucoprivic responses following cellular glucoprivation in RIPG1xglut2-/- mice reexpressing the transgenic GLUT2 specifially in their astrocytes (pGFAPG2xRIPG1xglut2-/- mice) or their neurons (pSynG2xRIPG1xglut2-/- mice). The increase of food intake after i.p. injection of 2-DG in control mice was not observed in the pGFAPG2xRIPG1xglut2-/- mice. Whereas a strong increase of glucagon secretion was observed in control and pGFAPG2xRIPG1xglut2-/- mice, not glucagonemic response was induced in pSynG2xRIPG1xglut2-/- mice. Our results show that GLUT2 reexpression in glial cells but not in neurons restored glucagon secretion and thus present a strong evidence that glucose detection and the control of glucagon secretion require a coupling between glial cells and neurons. Furthermore, these results show the existence of differents glucose sensors in CNS. Résumé tout public. Rôle du transporteur de glucose GLUT2 dans les mécanismes centraux de glucodétection impliqués dans le contrôle de la sécrétion du glucagon et de la prise alimentaire. Chez les mammifères, en dépit des grandes variations dans l'apport et l'utilisation du glucose, la glycémie est maintenue à une valeur relativement constante d'environ 1 g/l. Cette régulation est principalement sous le contrôle de deux hormones produites par le pancréas l'insuline et le glucagon. A la suite d'un repas, la détection de l'élévation de la glycémie par le pancréas permet la libération pancréatique de l'insuline dans le sang. Cette hormone va alors permettre le stockage dans le foie du glucose sanguin en excès et diminuer ainsi la glycémie. Sans insuline, le glucose s'accumule dans le sang. On parle alors d'hyperglycémie chronique. Cette situation est caractéristique du diabète et augmente les risques de maladies cardiovasculaires. A l'inverse, lors d'un jeûne, la détection de la diminution de la glycémie par le cerveau permet le déclenchement de la prise alimentaire et stimule la sécrétion de glucagon par le pancréas. Le glucagon va alors permettre la libération dans le sang du glucose stocké par le foie. Les effets du glucagon et de la prise de nourriture augmentent ainsi les concentrations sanguines de glucose pour empêcher une diminution trop importante de la glycémie. Une hypoglycémie sévère peut entraîner un mauvais fonctionnement du cerveau allant jusqu'à des lésions cérébrales. Contrairement aux mécanismes pancréatiques de détection du glucose, les mécanismes de glucodétection du cerveau ne sont encore que partiellement élucidés. Dans le laboratoire, nous avons observé, chez les souris transgéniques n'exprimant plus le transporteur de glucose GLUT2, une suppression de la stimulation de la sécrétion du glucagon et du déclenchement de la prise alimentaire en réponse à une hypoglycémie, induite uniquement dans le cerveau. Dans le cerveau, le GLUT2 est principalement exprimé par les astrocytes, cellules gliales connues pour soutenir, nourrir et protéger les neurones. Nous avons alors ré-exprimé spécifiquement le GLUT2 dans les astrocytes des souris transgéniques et nous avons observé que seule la stimulation de la sécrétion du glucagon en réponse à l'hypoglycémie est restaurée. Ces résultats mettent en évidence que la sécrétion du glucagon et la prise alimentaire sont contrôlées par différents mécanismes centraux de glucodétection dépendants de GLUT2.

Ghrelin-induced orexigenic effect in rats depends on the metabolic status and is counteracted by peripheral CB1 receptor antagonism.

Ghrelin is an endogenous regulator of energy homeostasis synthesized by the stomach to stimulate appetite and positive energy balance. Similarly, the endocannabinoid system is part of our internal machinery controlling food intake and energy expenditure. Both peripheral and central mechanisms regulate CB1-mediated control of food intake and a functional relationship between hypothalamic ghrelin and cannabinoid CB1 receptor has been proposed. First of all, we investigated brain ghrelin actions on food intake in rats with different metabolic status (negative or equilibrate energy balance). Secondly, we tested a sub-anxiogenic ultra-low dose of the CB1 antagonist SR141716A (Rimonabant) and the peripheral-acting CB1 antagonist LH-21 on ghrelin orexigenic actions. We found that: 1) central administration of ghrelin promotes food intake in free feeding animals but not in 24 h food-deprived or chronically food-restricted animals; 2) an ultra-low dose of SR141716A (a subthreshold dose 75 folds lower than the EC50 for induction of anxiety) completely counteracts the orexigenic actions of central ghrelin in free feeding animals; 3) the peripheral-restricted CB1 antagonist LH-21 blocks ghrelin-induced hyperphagia in free feeding animals. Our study highlights the importance of the animaĺs metabolic status for the effectiveness of ghrelin in promoting feeding, and suggests that the peripheral endocannabinoid system may interact with ghrelińs signal in the control of food intake under equilibrate energy balance conditions.

Rationale and design of a randomized, double-blind, placebo controlled multicenter trial to study efficacy, security, and long term effects of intermittent repeated levosimendan administration in patients with advanced heart failure: LAICA study.

BACKGROUND Advanced heart failure (HF) is associated with high morbidity and mortality; it represents a major burden for the health system. Episodes of acute decompensation requiring frequent and prolonged hospitalizations account for most HF-related expenditure. Inotropic drugs are frequently used during hospitalization, but rarely in out-patients. The LAICA clinical trial aims to evaluate the effectiveness and safety of monthly levosimendan infusion in patients with advanced HF to reduce the incidence of hospital admissions for acute HF decompensation. METHODS The LAICA study is a multicenter, prospective, randomized, double-blind, placebo-controlled, parallel group trial. It aims to recruit 213 out-patients, randomized to receive either a 24-h infusion of levosimendan at 0.1 μg/kg/min dose, without a loading dose, every 30 days, or placebo. RESULTS The main objective is to assess the incidence of admission for acute HF worsening during 12 months. Secondarily, the trial will assess the effect of intermittent levosimendan on other variables, including the time in days from randomization to first admission for acute HF worsening, mortality and serious adverse events. CONCLUSIONS The LAICA trial results could allow confirmation of the usefulness of intermittent levosimendan infusion in reducing the rate of hospitalization for HF worsening in advanced HF outpatients.

Motor overactivity and loss of motor circadian rhythm in fatal familial insomnia: an actigraphic study.

The 24-hour rest-activity pattern and the amount of motor activity was studied in a patient with fatal familial insomnia (FFI) by means of wrist actigraphy. During the study, the patient underwent indirect calorimetry. The 52-day recording showed severe disruption of the 24-hour rest-activity pattern with increased motor activity up to 80%. The 24-hour energy expenditure, assayed in a respiration chamber, was strikingly elevated by 60%. Chronic motor overactivity and loss of circadian rest-activity rhythm may play a role in the progressive metabolic exhaustion leading to death in FFI patients.

Lack of thermogenic response to glucose/insulin infusion in diabetic obese subjects.

The change in energy expenditure consecutive to the infusion of glucose/insulin was examined in 17 non-obese (ten young, seven middle-aged) and 27 diabetic and non-diabetic obese subjects by employing the euglycemic insulin clamp technique in conjunction with continuous indirect calorimetry. The obese subjects were divided into four groups according to their response to a 100-g oral glucose test: group A, normal glucose tolerance; group B, impaired glucose tolerance; group C, diabetes with increased insulin response; group D, diabetes with reduced insulin response. The glucose/insulin infusion provoked an increase in energy expenditure in both young and middle-aged controls (+8.2 +/- 1.3 percent and +5.9 +/- 0.5 percent over the preinfusion baseline respectively), but a lower increase in the non-diabetic obese groups A and B (+4.0 +/- 0.7 percent and +2.0 +/- 1.0 percent over the preinfusion baseline respectively, P less than 0.05 and P less than 0.01 vs young controls). However, in the diabetic obese groups C and D, energy expenditure failed to increase in response to the glucose/insulin infusion (mean change: +0.1 +/- 1.0 percent and -2.0 +/- 1.9 percent (P less than 0.01, vs middle-aged) over the preinfusion baseline respectively). When the glucose-induced thermogenesis (GIT) was related to the glucose uptake--taking into account the hepatic glucose production--the GIT was found to be similarly reduced in the diabetics groups (C and D). The net change in the rate of energy expenditure was found to be significantly correlated with the rate of glucose uptake (r = +0.647, n = 44, P less than 0.001) when all the individuals were pooled. In conclusion, this study shows that the low glucose-induced thermogenesis in obese diabetics during glucose insulin infusion is mainly related to a reduced rate of glucose uptake; in addition, inhibition of gluconeogenesis by the glucose/insulin infusion may also contribute to decrease the thermogenic response.

The forgotten face of regular physical exercise: a 'natural' anti-atherogenic activity.

Humans are not programmed to be inactive. The combination of both accelerated sedentary lifestyle and constant food availability disturbs ancient metabolic processes leading to excessive storage of energy in tissue, dyslipidaemia and insulin resistance. As a consequence, the prevalence of Type 2 diabetes, obesity and the metabolic syndrome has increased significantly over the last 30 years. A low level of physical activity and decreased daily energy expenditure contribute to the increased risk of cardiovascular morbidity and mortality following atherosclerotic vascular damage. Physical inactivity leads to the accumulation of visceral fat and consequently the activation of the oxidative stress/inflammation cascade, which promotes the development of atherosclerosis. Considering physical activity as a 'natural' programmed state, it is assumed that it possesses atheroprotective properties. Exercise prevents plaque development and induces the regression of coronary stenosis. Furthermore, experimental studies have revealed that exercise prevents the conversion of plaques into a vulnerable phenotype, thus preventing the appearance of fatal lesions. Exercise promotes atheroprotection possibly by reducing or preventing oxidative stress and inflammation through at least two distinct pathways. Exercise, through laminar shear stress activation, down-regulates endothelial AT1R (angiotensin II type 1 receptor) expression, leading to decreases in NADPH oxidase activity and superoxide anion production, which in turn decreases ROS (reactive oxygen species) generation, and preserves endothelial NO bioavailability and its protective anti-atherogenic effects. Contracting skeletal muscle now emerges as a new organ that releases anti-inflammatory cytokines, such as IL-6 (interleukin-6). IL-6 inhibits TNF-α (tumour necrosis factor-α) production in adipose tissue and macrophages. The down-regulation of TNF-α induced by skeletal-muscle-derived IL-6 may also participate in mediating the atheroprotective effect of physical activity.