Nonalcoholic fatty liver disease (NAFLD) is a complex metabolic-inflammatory disease associated with poor outcomes and decreased quality of life. NAFLD is overrepresented in patients with psychiatric disorders like depression, bipolar disorder, and schizophrenia; however, a comprehensive review on NAFLD and psychiatric disorders remains to be delineated. This review endeavors to investigate the association of NAFLD with psychiatric disorders, including shared pathogenesis and future clinical derivatives. Extant literature suggests that patients with psychiatric disorders (in particular, mood disorders) are more susceptible to the development of NAFLD due to multiple reasons, including but not limited to hypothalamic–pituitary–adrenal axis dysregulation, metabolic syndrome, and chronic perceived stress. Moreover, the clinical manifestations of mood disorders (e.g., anhedonia, psychomotor retardation, lifestyle modification, etc.), and potentially long-term treatment with weight-gaining agents, differentially affect these patients, making them more prone to NAFLD. Considering the increased morbidity associated with both mood disorders and NAFLD, our review recommends regular screenings for NAFLD in select patients with mood disorders exhibiting signs of increased risk (i.e., obesity, metabolic syndrome, diabetes, or family history of NAFLD) for better diagnosis and holistic care of both potentially interrelated conditions.

Paediatric fatty liver disease (FLD) can develop into steatohepatitis, cirrhosis and hepatocellular carcinoma in adulthood. We assessed if early life physical exercise reduced the effects of high-fat (HF) diet-induced steatosis. Male HF-fed rats with access to a running wheel from weaning until day (D)60 (early exercise) or from D67 to D120 (late exercise) were compared with control HF- or chow-fed groups with no wheel. At D63 and D120, liver histopathology (Kleiner score), type I collagen and plasma enzymes were assessed. At D63, early life activity significantly reduced histopathology scores (total, portal inflammation, steatosis, ballooning, but not lobular inflammation or fibrosis) and the number of rats affected. At D120, HF control scores were higher than in chow-fed controls, but the effect of activity was selective: early exercise reduced portal inflammation, steatosis, ballooning and fibrosis, but late activity affected only portal inflammation and ballooning. The chow-fed portal inflammation score was significantly less than all HF groups, but lobular inflammation was lower in the HF control group only. The fibrosis score in the HF early exercise and control chow group were lower than in the late exercise and sedentary HF groups, indicating that early life exercise was more effective than when activity was introduced later in life. Plasma biomarkers showed minor between-group differences. The retained effect on liver histopathology rat at D120 after only early life exposure activity suggests that timing of introduction of exercise is critical in reducing FLD scores and prevalence in children, young adults and possibly into adulthood.

Prospective cohort studies linking organ meat consumption and non-alcoholic fatty liver disease (NAFLD) are limited, especially in Asian populations. This study aimed to prospectively investigate the association between organ meat consumption and risk of NAFLD in a general Chinese adult population. This prospective cohort study included a total of 15 568 adults who were free of liver disease, CVD and cancer at baseline. Dietary information was collected at baseline using a validated FFQ. NAFLD was diagnosed by abdominal ultrasound after excluding other causes related to chronic liver disease. Cox proportional regression models were used to assess the association between organ meat consumption and risk of NAFLD. During a median of 4·2 years of follow-up, we identified 3604 incident NAFLD cases. After adjusting for demographic characteristics, lifestyle factors, vegetable, fruit, soft drink, seafood and red meat consumption, the multivariable hazard ratios (95 % CI) for incident NAFLD across consumption of organ meat were 1·00 (reference) for almost never, 1·04 (0·94, 1·15) for tertile 1, 1·08 (0·99, 1·19) for tertile 2 and 1·11 (1·01, 1·22) for tertile 3, respectively (P for trend < 0·05). Such association did not differ substantially in the sensitivity analysis. Our study indicates that organ meat consumption was related to a modestly higher risk of NAFLD among Chinese adults. Further investigations are needed to confirm this finding.

Dietary modification plays a vital role in the treatment of non-alcoholic liver diseases. We investigated the effects of the consumption of a different amount of dehulled adlay, which has hypolipidaemic and anti-inflammatory properties, on non-alcoholic fatty liver disease (NAFLD). We fed rats a high-fat-high-fructose liquid diet for 16 weeks to induce NAFLD. The rats were divided into three groups fed the NAFLD diet only (NN) or a diet containing 44·9 or 89·8 g/l of dehulled adlay (NA and NB groups, respectively). After 8 weeks, the NA and NB groups had lower C-reactive protein levels and improvement in insulin resistance. In addition, the NB group had lower liver weight and hepatic TAG and cholesterol concentrations than did the NN group. Compared with the NN group, the high-dose NB group had improved steatosis, lower hepatic TNF-α, IL-1β and IL-6 levels and lower adipose leptin levels. Our results suggest that a diet containing dehulled adlay can ameliorate NAFLD progression by decreasing of insulin resistance, steatosis and inflammation.

Non-alcoholic fatty liver disease (NAFLD) was defined in 1980 and has the same histological characteristics as alcoholic liver disease except for alcohol consumption. After 40 years, the understanding of this disease is still imperfect. Without specific drugs available for treatment, the number of patients with NAFLD is increasing rapidly, and NAFLD currently affects more than one-quarter of the global population. NAFLD is mostly caused by a sedentary lifestyle and excessive energy intake of fat and sugar. To ameliorate or avoid NAFLD, people commonly replace high-fat foods with high-carbohydrate foods (especially starchy carbohydrates) as a way to reduce caloric intake and reach satiety. However, there are few studies that concentrate on the effect of carbohydrate intake on liver metabolism in patients with NAFLD, much fewer than the studies on fat intake. Besides, most of these studies are not systematic, which has made identification of the mechanism difficult. In this review, we collected and analysed data from studies on human and animal models and, surprisingly, found that carbohydrates and liver steatosis could be linked by inflammation. This review not only describes the effects of carbohydrates on NAFLD and body lipid metabolism but also analyses and predicts possible molecular pathways of carbohydrates in liver lipid synthesis that involve inflammation. Furthermore, the limitations of recent research and possible targets for regulating inflammation and lipogenesis are discussed. This review describes the effects of starchy carbohydrates, a nutrient signal, on NAFLD from the perspective of inflammation.

Consumption of a high-fat diet increases fat accumulation and may further lead to inflammation and hepatic injuries. The aim of the study was to investigate the effects of Camellia oleifera seed extract (CSE) on non-alcoholic fatty liver disease (NAFLD). After a 16-week NAFLD-inducing period, rats were assigned to experimental groups fed an NAFLD diet with or without CSE. At the end of the study, we found that consuming CSE decreased the abdominal fat weight and hepatic fat accumulation and modulated circulating adipokine levels. We also found that CSE groups had lower hepatic cytochrome P450 2E1 and transforming growth factor (TGF)-β protein expressions. In addition, we found that CSE consumption may have affected the gut microbiota and reduced toll-like receptor (TLR)-4, myeloid differentiation primary response gene 88, toll/IL-1 receptor domain-containing adaptor-inducing interferon-β (TRIF) expression and proinflammatory cytokine concentrations in the liver. Our results suggest that CSE may alleviate the progression of NAFLD in rats with diet-induced steatosis through reducing fat accumulation and improving lipid metabolism and hepatic inflammation.

Some studies have shown that expression of peroxisome proliferator-activated receptor gamma (PPARG), a key regulator of adipogenesis, and of some adipocyte-specific genes or adipokines are expressed in hepatic steatosis, leading to the concept of ‘adipogenic hepatic steatosis’ or ‘hepatic adiposis.’ Most of these studies were conducted in genetic obese mouse models or after manipulation of gene expression. The relevance of this concept to other species and more physiological models was here addressed in ducks which are able to develop hepatic steatosis after overfeeding. The expression of PPARG and other adipocyte-specific genes was thus analyzed in the liver of ducks fed ad libitum or overfed and compared with those observed in adipose tissues. Pekin (Anas platyrhynchos) and Muscovy ducks (Cairina moschata) were analyzed, as metabolic responses to overfeeding differ according to these two species, Muscovy ducks having a greater ability to synthesize and store lipids in the liver than Pekin ducks. Our results indicate that adipocyte-specific genes are expressed in the liver of ducks, PPARG and fatty acid-binding protein 4 being upregulated and adiponectin and leptin receptor downregulated by overfeeding. However, these expression levels are much lower than those observed in adipose tissue suggesting that fatty liver cells are not transformed to adipocytes, although some hepato-specific functions are decreased in fatty liver when compared with normal liver.

The effect and the role played by dietary α-lactalbumin (α-LAC) on hepatic fat metabolism are yet to be fully elucidated. We reported previously that α-LAC intake induced atherogenic dyslipidaemia in Balb/c mice. The aim of the present study was to investigate if this atherogenic effect could be due to a possible α-LAC-induced hepatic steatosis. We examine the ability of dietary α-LAC to induce liver steatosis, identifying the molecular mechanisms underlying hepatic lipid metabolism in association with the lipid profile, peripheral insulin resistance (IR) and changes in the hepatic oxidative environment. Male Balb/c mice (n 6) were fed with diets containing either chow or 14 % α-LAC for 4 weeks. The α-LAC-fed mice developed abdominal adiposity and IR. Moderate liver steatosis with increased TAG and NEFA contents was correlated with atherogenic dyslipidaemia. There was increased nuclear expression of liver X receptor αβ (LXRαβ), sterol regulatory element-binding protein-1c (SREBP-1c) and PPARγ transcription factors and of the cytosolic enzymes acetyl-CoA carboxylase 1 (ACC1) and fatty acid synthase involved in the hepatic de novo lipogenesis. The opposite was found for the nuclear receptor PPARα and the mitochondrial enzyme carnitine palmitoyltransferase-1 (CPT-1), leading to reduced fatty acid β-oxidation (FAO). These changes were associated with a significant decrease in both p-Thr172-AMP-activated protein kinase α (AMPKα) (inactivation) and p-Ser79-ACC1 (activation) and with a more oxidative liver environment increasing lipid peroxidation and protein oxidation and reducing GSH:GSSG ratio in the α-LAC-fed mice. In conclusion, 4 weeks of 14 % α-LAC feeding induced liver steatosis associated with atherogenic dyslipidaemia, IR and oxidative stress by enhancing nuclear LXRαβ/SREBP-1c/PPARγ expression and diminishing PPARα/CPT-1 expression and AMPKα phosphorylation shifting the hepatic FAO toward fatty acid synthesis in Balb/c mice.

Although non-alcoholic fatty liver disease (NAFLD) is the leading aetiology of liver disorders in the world, there is no proven treatment for NAFLD patients with normal or low BMI. The aim of this study was to evaluate the efficacy of synbiotics supplementation in NAFLD patients with normal or low BMI. In this randomised, double-blind, placebo-controlled, clinical trial, fifty patients with NAFLD were assigned to take either a synbiotic supplement or a placebo capsule for 28 weeks. Both groups were advised to follow a healthy lifestyle. At the end of the study, hepatic steatosis and fibrosis reduced in both groups; however, the mean reduction was significantly greater in the synbiotic group rather than in the placebo group (P<0·001). Furthermore, serum levels of fasting blood sugar, TAG and most of the inflammatory mediators reduced in the synbiotic group significantly compared with the placebo group (P<0·05). Our results provide evidence that synbiotic supplementation improves the main features of NAFLD in patients with normal and low BMI, at least partially through reduction in inflammatory indices. Further studies are needed to address the exact mechanism of action of these effects.

We previously reported that rice endosperm protein (REP) has renoprotective effects in Goto–Kakizaki rats, a non-obese diabetic model. However, whether these effects occur in obese diabetes remains unclear. This study aimed to clarify the effects of REP on obese diabetes, especially on fatty liver and diabetic nephropathy, using the obese diabetic model Zucker diabetic fatty (ZDF) rats. In total, 7-week-old male ZDF rats were fed diets containing 20 % REP or casein (C) for 8 weeks. Changes in fasting blood glucose levels and urinary markers were monitored during the experimental period. Hepatic lipids and metabolites were measured and renal glomeruli were observed morphologically. HbA1c levels were significantly lower in rats fed REP, compared with C (P<0·05). Compared with C in the liver, REP prevented lipid accumulation (total lipid, TAG and total cholesterol, P<0·01). Liver metabolome analysis indicated that levels of metabolites associated with glycolysis, the pentose phosphate pathway and carnitine metabolism were significantly greater in the REP group than in the C group (P<0·05), suggesting activation of both glucose catabolism and fatty acid oxidation. The metabolite increases promoted by REP may contribute to suppression of liver lipid accumulation. Urinary excretion of albumin and N-acetyl-β-d-glucosaminidase was significantly reduced in rats fed REP for 8 weeks (P<0·01). In addition, there was a distinct suppression of mesangial matrix expansion and glomerular hypertrophy in response to REP (P<0·01). Thus, REP had preventive effects on obese diabetes, fatty liver and diabetic nephropathy.

High-producing dairy cows enter a period of negative energy balance during the first weeks of lactation. Energy intake is usually sufficient to cover the increase in energy requirements for fetal growth during the period before calving, but meeting the demand for energy is often difficult during the early stages of lactation. A catabolic state predominates during the transition period, leading to the mobilisation of energy reserves (NEFA and amino acids) that are utilised mainly by the liver and muscle. Increased uptake of mobilised NEFA by the liver, combined with the limited capacity of hepatocytes to either oxidise fatty acids for energy or to incorporate esterified fatty acids into VLDL results in fatty liver syndrome and ketosis. This metabolic disturbance can affect the general health, and it causes economic losses. Different nutritional strategies have been used to restrict negative effects associated with the energy challenge in transition cows. The provision of choline in the form of rumen-protected choline (RPC) can potentially improve liver function by increasing VLDL exportation from the liver. RPC increases gene expression of microsomal TAG transfer protein and APOB100 that are required for VLDL synthesis and secretion. Studies with RPC have looked at gene expression, metabolic hormones, metabolite profiles, milk production and postpartum reproduction. A reduction in liver fat and enhanced milk production has been observed with RPC supplementation. However, the effects of RPC on health and reproduction are equivocal, which could reflect the lack of sufficient dose–response studies.

Non-alcoholic fatty liver disease (NAFLD) is usually associated with insulin resistance, central obesity, reduced glucose tolerance, type 2 diabetes mellitus and hypertriacylglycerolaemia. The beneficial effects of resveratrol on metabolic disorders have been shown previously. The aim of this study was to evaluate the effects of resveratrol supplementation on cardiovascular risk factors in patients with NAFLD. In this randomised double-blinded placebo-controlled clinical trial, fifty NAFLD patients were supplemented with either a 500-mg resveratrol capsule or a placebo capsule for 12 weeks. Both groups were advised to follow an energy-balanced diet and physical activity recommendations. resveratrol supplementation reduced alanine aminotransferase (ALT) and hepatic steatosis significantly more than placebo (P<0·05). BMI, waist circumference, serum aspartate aminotransferase, bilirubin, HDL-cholesterol and apo a1 were reduced significantly in both groups (P<0·05); however, there were no significant differences between the two groups (P>0·05). There were no significant changes in blood pressure, insulin resistance markers and TAG in either group (P>0·05). Our data have shown that 12-week supplementation of 500 mg resveratrol does not have any beneficial effect on anthropometric measurements, insulin resistance markers, lipid profile and blood pressure; however, it reduced ALT and hepatic steatosis in patients with NAFLD.

In rats, maternal exposure to restraint stress during pregnancy can induce abnormalities in the cardiovascular and central nervous systems of the offspring. These effects are mediated by long-lasting hyperactivation of the hypothalamic–pituitary–adrenal axis. However, little is known about the potential effects of stress during pregnancy on metabolic systems. We examined the effect of restraint stress in pregnant mice on the liver function of their offspring. The offspring of stressed mothers showed significantly higher lipid accumulation in the liver after weaning than did the controls; this accumulation was associated with increased expression of lipid metabolism-related proteins such as alanine aminotransferase 2 diglyceride acyltransferase 1, peroxisome proliferator-activated receptor gamma and glucocorticoid receptor. Additionally, we observed increased levels of 11β-hydroxysteroid dehydrogenase type 1, an intercellular mediator that converts glucocorticoid from the inactive to the active form, in the foetal and postnatal periods. These results indicate that restraint stress in pregnancy in mice induces metabolic abnormalities via 11β-hydroxysteroid dehydrogenase type 1-related pathways in the foetal liver. It is therefore possible that exposure to stress in pregnant women may be a risk factor for metabolic syndromes (e.g. fatty liver) in children.

Methyl donor supplementation has been reported to prevent obesity-induced liver fat accumulation in adult rats. We hypothesized that this protection could be mediated by perinatal nutrition. For this purpose, we assessed the response to an obesogenic diet (high-fat-sucrose, HFS) during adulthood depending on maternal diet during lactation. Female Wistar rats fed control diet during pregnancy were assigned to four postpartum dietary groups: control, control supplemented with methyl donors (choline, betaine, folic acid, vitamin B12), HFS and HFS supplemented with methyl donors. At weaning, the male offspring was transferred to a chow diet and at week 12th assigned to a control or a HFS diet during 8 weeks. The offspring whose mothers were fed HFS during lactation showed increased adiposity (19%, P<0.001). When fed the HFS diet as adults, offspring whose mothers were HFS supplemented had more body fat (23%, P<0.001) than those from HFS non-supplemented. However, they showed lower liver fat accumulation (−18%, P<0.001). Srebf1, Dnmt1 and Lepr liver mRNA levels increased after adulthood HFS feeding. In those animals HFS fed during adulthood, previous maternal HFS decreased Lepr and Dnmt1 expression levels when compared with c-HFS offspring, while the supplementation of control and HFS-fed dams, respectively, induced higher hepatic Mme and Lepr mRNA levels after adult HFS intake compared with hfs-HFS offspring. In conclusion, maternal HFS diet during lactation influenced the response to an obesogenic diet in the adult progeny. Interestingly, dietary methyl donor supplementation in lactating mothers fed an obesogenic diet reduced liver fat accumulation, but increased adipose tissue storage in adult HFS-fed offspring.

Various human trials and pre-clinical studies have suggested that dietary plant sterols possess hypotriacylglycerolaemic properties apart from their cholesterol-lowering properties. We hypothesised that phytosterols (PS) might attenuate triacylglycerolaemia by interfering with the deleterious effects of cholesterol overload in the liver. In the present study, twenty hamsters (Mesocricetus auratus) with diet-induced combined hyperlipidaemia were fed a high-fat diet (HFD, n 10) or a HFD supplemented with soyabean PS (n 10) for 40 d. In parallel, a healthy group was fed a standard diet (n 10). PS normalised fasting plasma cholesterol concentrations completely after 20 d and were also able to normalise serum TAG and NEFA concentrations after 40 d. HFD feeding caused microvesicular steatosis and impaired the expression of key genes related to fatty acid oxidation such as PPARA, carnitine palmitoyltransferase-Iα (CPT1A) and phosphoenolpyruvate carboxykinase 1 (PCK1) in the liver. PS treatment completely protected against HFD-induced steatosis and resulted in a normalised hepatic gene expression profile. The protection of the hepatic function by PS was paralleled by increased faecal cholesterol excretion along with a 2-fold increase in the biliary bile acid (BA):cholesterol ratio. The present study supports the conclusion that long-term consumption of PS can reduce serum TAG and NEFA concentrations and can protect against the development of fatty liver via different mechanisms, including the enhancement of BA synthesis. The results of the present study place these compounds as promising hepatoprotective agents against fatty liver and its derived pathologies.

The aim of this study was to analyse the effects of species (Muscovy and Pekin ducks) and age at the beginning of the overfeeding period on fatty liver production, carcass composition and lipid and moisture content of the liver and breast muscle. We reared four groups of 40 ducks per species for the study, starting at 2-week intervals in order to have four different ages together at the beginning of the overfeeding period (10, 12, 14 and 16 weeks). At the end of the overfeeding period, all ducks were slaughtered. Our results confirmed the high levels of difference in carcass composition and lipid content in the plasma, liver and breast muscle between Muscovy and Pekin ducks at all ages. Pekin ducks were not able to develop a high degree of hepatic steatosis, but had increased lipid storage in peripheral adipose and muscle tissues than Muscovy ducks. However, the fatty liver weight of Pekin ducks increased with age, with lipid deposition in the liver and peripheral tissues. The ability of Muscovy ducks to produce fatty livers remained unchanged with age in line, with lipid deposition in the liver and peripheral tissues. The sites of lipid deposition thus depend on species and not on the physiological maturity of ducks.

Conjugated linoleic acid (CLA) has anti-obesity effects, but induces fatty liver in mice. The present study investigated whether co-administration of arachidonic acid (ARA) attenuates fat accumulation in the mouse liver induced by CLA. Male mice (8 weeks old) were given diets with either no addition of dietary fat (control), 3 % linoleic acid (LA), 3 % CLA, 3 % CLA+1 % ARA, or 3 % CLA+2 % ARA for 4 weeks. The perirenal fat weight in ARA-treated groups decreased similarly as with CLA alone, when compared to control or LA. Plasma TAG concentration was significantly higher in the CLA group than in either CLA+ARA group, while plasma cholesterol and NEFA concentrations did not vary among the groups. In contrast to visceral fat, liver weight was significantly higher in the CLA group than in the control or LA groups, and the effects of CLA were attenuated by ARA. TAG and cholesterol were markedly accumulated in the liver with dietary CLA, whereas co-administration with ARA, at either concentration, suppressed CLA-induced lipid accumulation. Liver PGE2 was enhanced by a combination of CLA and ARA when compared with CLA alone, but PGE1 level was not significantly different among groups. In conclusion, fatty liver induced by CLA was attenuated by co-administration with ARA, furthermore, a combination of these fatty acids maintained the anti-obese effect of CLA.

During early lactation, many dairy cows develop fatty liver, which is associated with decreased health and reproductive performance. Currently, fatty liver can be detected reliably only by using liver biopsy followed by chemical or histological analysis, which is not practical in most on-farm situations. We tested whether digital analyses of hepatic ultrasonograms can be used to detect non-invasively fatty liver and estimate liver triacylglycerol content. A total of 49 liver biopsies and ultrasonograms were taken from 29 dairy cows within 2 weeks postpartum. The usefulness of 17 first- or second-order parameters from digital analysis of B-mode ultrasonograms were evaluated by discriminant, correlation, and regression analyses. A group of linear combinations of the 17 parameters correctly classified 40 of 49 samples into normal liver as well as mild, moderate and severe fatty liver when cut-off values were 1%, 5% and 10% and correctly classified 45 of 49 samples when cut-off values were 5% and 10% triacylglycerol of wet weight. A linear combination of 16 image parameters estimated triacylglycerol concentrations of 38 of the 39 liver samples below the cut-off value of 10% within 2·5% of liver wet weight, and a linear combination of 3 parameters estimated triacylglycerol concentrations of the 10 liver samples above the cut-off value of 10% within 2% of liver wet weight. Therefore, ultrasound imaging followed by digital analysis of sonograms has potential to non-invasively detect fatty liver and estimate liver triacylglycerol content.

The aim of the study was to characterize the time course of the development of high-fat diet-induced hepatic steatosis and its relation to body fat accretion and changes in plasma lipid profile. Female Sprague–Dawley rats were high-fat fed (HF; 42%, kJ) for 1, 2, 4, 6, 12 and 16 weeks and compared to standard fed rats (SD). Data obtained from HF rats were further analysed by classifying the animals into obesity-prone and obesity-resistant. In HF rats, liver lipid content increased rapidly by approximately 200% during the first 2 weeks, decreased almost to baseline levels between weeks 2 and 6, and re-increased by 17% between weeks 6 and 16 (P<0·05). Body weight, body fat accretion, plasma leptin, NEFA and glycerol concentrations were higher in HF than in SD rats (P<0·05). These higher values were established in 2 weeks and the differences between the groups did not further enlarge from weeks 2 to 16. Obesity-prone rats depicted higher body weight and body fat accretion than obesity-resistant and SD rats. Surprisingly, however, liver lipid content was the same in obesity-prone as in obesity-resistant rats as they were both higher than in SD rats (weeks 2 and 16; P<0·05). Our data support the hypothesis that the liver acts as a systemic buffer, largely increasing its lipid content in the early stage of high-fat feeding. Our results also suggest that the development of non-alcoholic hepatic steatosis is more linked to dietary fat ingestion than to body weight gain.

Fatty acids are the major source of energy for most tissues during periods of negative energy balance; however, fatty acids can, in some circumstances, have pathological effects. Fatty acids are stored as triacylglycerols (TAG), mostly in the various adipose tissue depots of the body. However, if blood unesterified fatty acid (NEFA) levels are elevated for prolonged periods, as may occur during lactation or obesity, TAG can accumulate in other tissues including liver and muscle cells (myocytes), and this can have pathological consequences such as the development of ketosis (Grummer, 1993; Drackley et al. 2001) or type 2 diabetes (Boden & Shulman, 2002; McGarry, 2002).